University of Veterinary Medicine THESIS

Transcription

!
University of Veterinary Medicine
Institute of Zoology
Auditory Spatial Cognition in Different Vertebrate Taxa
THESIS
submitted in partial fulfilment of the requirements for the
degree
- Doctor rerum naturalium (Dr. rer. nat)
by
Daniel Schmidtke
Hannover, Germany
Hannover 2013
!
Main Supervisor:
PD Dr. Karl-Heinz Esser
Co-Supervisors:
Prof. Dr. Eckart Altenmüller
Apl. Prof. Dr. Manuela Gernert
1st Evaluation:
Institute of Zoology
University of Veterinary Medicine, Hannover
Prof. Dr. Eckart Altenmüller
Institute of Music Physiology and Musicians’ Medicine
Hannover University of Music, Drama and Media
Apl. Prof. Dr. Manuela Gernert
Institute for Pharmacology, Toxicology and Pharmacy
University of Veterinary Medicine, Hannover
2nd Evaluation:
Prof. Jagmeet S. Kanwal, Ph.D.
Department of Physiology and Biophysics
Georgetown University, Washington
Date of final exam:
13.04.2013
Sponsorship:
Funding for this work was provided by the Studienstiftung des
Deutschen Volkes in form of a Ph.D. scholarship and a
scholarship for studies abroad to Daniel Schmidtke.
!
Parts of this thesis were presented to a scientific audience on the following occasions:
2008:
Invited talk at the Max-Planck-Institute for Ornithology (Seewiesen, Germany): ''Spatial
cognition in mammals: new insights from a bat model.''
Talk at the Doctoral Student Forum of the Studienstiftung des Deutschen Volkes (Cologne,
Germany): ''Neuroethological studies on three-dimensional spatial cognition in mammals.''
2009:
Talk at the 2009 Meeting of the German Bat Research Community (Frauenwörth, Germany):
''Gender-specific orientation strategies in the lesser spear-nosed bat Phyllostomus discolor.''
Talk at the 2nd Braunlage Meeting of the Center for Systems Neuroscience (Braunlage,
Germany): ''Neuroethological studies on three-dimensional allocentric spatial cognition in
mammals.''
2010:
Invited talk at the Center for Neurogenomics and Cognitive Research (Amsterdam,
Netherlands): ''Spatial cognition in mammals: a story of mice, men, and microbats.''
Poster presentation at the 7th FENS Forum of European Neuroscience (Amsterdam,
Netherlands): ''Manufacturing short Elgiloy (stainless steel) microelectrodes for telemetric
neuronal recordings in freely behaving small mammals.'' – Abstracts: 7th FENS Forum of
European Neuroscience (2010), 207.1
Poster presentation at the 2010 Meeting of the Center for Systems Neuroscience (Hannover,
Germany): ''Studies on allocentric spatial cognition in mammals.''
2011:
Talk at the 2011 Meeting of the German Bat Research Community (Frauenwörth, Germany):
''Towards telemetric electrophysiological recordings in freely behaving bats.''
Poster presentation at the 9th Göttingen Meeting of the German Neuroscience Society
(Göttingen, Germany): ''The utilisation of acoustic landmarks for orientation in humans
without vision.'' – Proceedings: 9th Göttingen Meeting of the German Neuroscience Society
(2011), T24-10C
2013:
Poster presentation at the 10th Göttingen Meeting of the German Neuroscience Society
(Göttingen, Germany): ''Acoustically guided wayfinding in humans: the role of gender and
sightedness.'' – Proceedings: 10th Göttingen Meeting of the German Neuroscience Society
(2013), T25-4B
!
Chapters 3 and 4 of this thesis have been submitted for publication in peer-reviewed,
scientific journals. In terms of content, chapter 3 of this thesis is identical to the respective,
submitted manuscript. However, some minor modifications have been made to the text for
consistency in writing style and layout.
Chapter 3:
Title:
Acoustic Landmarks in Human Wayfinding: the Role of Gender
and Sightedness.
Authors:
Daniel Schmidtke
Sarah Galinski
Authors’ contributions:
DS, SG and K-HE conceived and designed the experiments, DS
and SG performed the experiments, DS analysed the data, K-HE
contributed materials and analysis tools, DS wrote the
manuscript.
Journal:
The manuscript has been submitted to PLOS ONE
(status: under peer review)!
Chapter 4:
Title:
Structure and Possible Functions of Constant-Frequency Calls in
Ariopsis seemanni (Osteichthyes, Ariidae).
Authors:
Daniel Schmidtke
Dr. Jochen Schulz
Prof. Dr. Dr. h.c. Jörg Hartung
DS and K-HE conceived and designed the experiments, DS
performed the experiments, DS analysed the data, JS and JH and
K-HE contributed materials and analysis tools, DS wrote the
manuscript.
Journal:
(status: published)!
!
!
Index
Summary!.............................................................................................................................................!1!
Zusammenfassung!.............................................................................................................................!3!
Chapter 1: General Introduction!...................................................................................................!7!
Space!..............................................................................................................................................................!7!
Egocentric Representations of Space!......................................................................................................!8!
Allocentric Representations of Space!...................................................................................................!10!
The Firing Properties of Place and Grid Cells!...................................................................................!11!
Gender Differences in Orientation Strategies!.....................................................................................!16!
Aims of This Thesis!...................................................................................................................................!17!
Chapter 2: Towards Telemetric Recordings of Allocentric Spatial Representations in
Freely Behaving Small Mammals!...............................................................................................!19!
Characterising the Firing Properties of Place and Grid Cells!........................................................!19!
Telemetric Devices!....................................................................................................................................!21!
Manufacturing Short Elgiloy (Stainless Steel) Microelectrodes for Telemetric Neuronal
Recordings in Freely Behaving Small Mammals!...............................................................................!23!
Surgical Protocols for the Chronic Implantation of the Electrodes in Mice (Mus musculus)
and Microbats (Phyllostomus discolor)!.................................................................................................!27!
Results and Discussion!.............................................................................................................................!30!
Chapter 3: Acoustic Landmarks in Human Wayfinding: the Role of Gender and
Sightedness!.......................................................................................................................................!38!
Abstract!.......................................................................................................................................................!38!
Introduction!................................................................................................................................................!39!
Materials and Methods!............................................................................................................................!41!
Ethics statement!.......................................................................................................................................................!41!
Test subjects!..............................................................................................................................................................!42!
Experimental environment!...................................................................................................................................!42!
Acoustic landmarks!.................................................................................................................................................!44!
Videography!..............................................................................................................................................................!46!
Experimental procedure!........................................................................................................................................!46!
Personality measures!..............................................................................................................................................!47!
Data acquisition!........................................................................................................................................................!49!
Data preparation!.......................................................................................................................................................!51!
Statistics!......................................................................................................................................................................!51!
Results!..........................................................................................................................................................!55!
Comparison of the training performances between the groups!................................................................!55!
Individual reactions to the modifications of the acoustic landmarks!.....................................................!58!
Group-level reactions to the modifications of the acoustic landmarks!..................................................!60!
Comparison of the plasticine models!................................................................................................................!62!
Bivariate correlations!.............................................................................................................................................!62!
Discussion!....................................................................................................................................................!65!
Landmark qualities I!...............................................................................................................................................!65!
Landmark qualities II!.............................................................................................................................................!66!
Gender differences!..................................................................................................................................................!67!
!
Differences between visually impaired and sighted subjects!....................................................................!74!
On the development of electronic travel and orientation aids!..................................................................!85!
Conclusions/Summary!.............................................................................................................................!87!
Chapter 4: Structure and Possible Functions of Constant-Frequency Calls in Ariopsis
seemanni (Osteichthyes, Ariidae)!................................................................................................!92!
Abstract!.......................................................................................................................................................!92!
Chapter 5: General Discussion!....................................................................................................!93!
The Dimensionality of Mammalian Allocentric Representations of Space!..................................!93!
Gender Differences in Mammalian Orientation and the Modality of Allocentric Spatial
Representations!.........................................................................................................................................!95!
Core Systems of Spatial Cognition in Vertebrates!.............................................................................!98!
Final Remark!..........................................................................................................................................!101!
References!.....................................................................................................................................!103!
Acknowledgements!......................................................................................................................!123!
Supporting Material!....................................................................................................................!125!
!
Abbreviations
3D
BS
CA1
cf
CF
DI
DIS
ES
ETA
ETO
FM
fMRI
fps
G1 - G3
GPS
h0
h1, h2, …
IC
IR
Jpeg
LM
O&M
PC
PF
RC
SARA
SNR
SPL
SSS-V
TAS
TL
WAV
three-dimensional
boredom susceptibility
cornu ammonis field 1
carrier frequency
centre frequency
distortion index
disinhibition
experience seeking
electronic travel aid
electronic orientation aid
frequency modulated
functional magnetic resonance imaging
frames per second
transmitter generation 1 - 3
global positioning system
fundamental
harmonic 1, 2, …
integrated circuit
infrared
“Joint Photographic Expert Group”
landmark modification
orientation and mobility training
personal computer
peak frequency
resistor-capacitor
spatialised augmented reality audio
signal-to-noise ratio
sound pressure level
Sensation Seeking Scale, Form V
thrill and adventure seeking
total body length
waveform audio file format
!
List of Figures
Fig. 1.1: Idealised firing-rate maps of place and grid cells ...................................................... 12
Fig. 1.2: Hypothetical firing fields of place cells in 3D environments .................................... 14
Fig. 2.1: Telemetric devices (G1 - G3) .................................................................................... 22
Fig. 2.2: Electrode manufacturing ............................................................................................ 25
Fig. 2.3: Telemetric recordings from the inferior colliculus (Mus musculus).......................... 31
Fig. 2.4: Telemetric recordings from the hippocampus (Mus musculus) ................................. 33
Fig. 2.5: Photomicrographs from the brain of Mus musculus .................................................. 35
Fig. 3.1: Schematic drawing of the maze (top-view) ............................................................... 43
Fig. 3.2: Time-varying spectral composition of the acoustic landmarks ................................. 45
Fig. 3.3: Calculated spatial benchmarks and reconstructed walking paths .............................. 52
Fig. 3.4: Learning performance ................................................................................................ 57
Fig. 3.5: Utilisation of the acoustic landmarks......................................................................... 59
Fig. 3.6: Plasticine-model quality ............................................................................................ 63
Supporting Fig. 1: Ariopsis seemanni .................................................................................... 125
List of Tables
Table 3.1: Scheme for the pseudo-randomisation of the landmark modifications................... 48
Table 3.2: Parameters used to evaluate individual maze performances ................................... 50
Table 3.3: Hypothesis matrices for the linear-hypothesis tests ................................................ 56
Table 3.4: Summary of the results of the linear-hypothesis tests............................................. 61
Table 3.5: Correlations between individual training performance and SSS-V scores ............. 64
List of Equations
Equation 3.1: Perspective correction of the auxiliary grids (x-coordinate) ............................. 49
Equation 3.2: Perspective correction of the auxiliary grids (y-coordinate) ............................. 49
Equation 3.3: Spatial benchmark (room 1) .............................................................................. 51
!
!
1!
Summary
Daniel Schmidtke
Auditory spatial cognition in different vertebrate taxa
The thesis at hand presents three studies in the area of auditory spatial cognition that were
conducted in different vertebrate species. Owing to intense research in the field of allocentric
spatial cognition in rodents, a large number of important discoveries about the neuronal
substrates of inner, spatial representations of the outer world have been made during the last
decades. The utilisation of well-defined animal models has several advantages
(standardisation of methods, high comparability between the results from different studies,
etc.). Yet, in the research on spatial cognition, it becomes increasingly apparent that the
reliance on only a few model species also bears some clear disadvantages. For example, a
species-based approach naturally does not allow for statements about the phylogeny and
generalisability of an investigated phenomenon. Due to the standardised utilisation of twodimensional experimental arenas and the provision of mainly visual cues for orientation in
past experiments, we further lack knowledge of the dimensionality and modality dependence
of allocentric spatial representations. This knowledge, however, can only be gained through a
change of the experimental species and by using complex, three-dimensional experimental
environments.
The first study presented here, therefore, deals with the question, in how far a
miniaturised telemetric device that has recently been developed by colleagues from the MaxPlanck-Institute for Ornithology (Seewiesen, Germany) and tested in zebra finches allows for
the characterisation of the receptive fields of neuronal, allocentric spatial representations in
the medial temporal lobe of small mammals. To answer this question, it initially was
necessary to develop a protocol for the manufacturing of very short stainless-steel
microelectrodes (which was done in cooperation with the Laser Zentrum Hannover e.V.). My
recording experiments in awake, freely behaving mice (Mus musculus) and microbats
(Phyllostomus discolor) show that the self-made microelectrodes in combination with the
telemetric device readily allow for neuronal multi-unit recordings in awake rodents. The
possibility to achieve high-quality single-unit recordings, however, seems to be highly
dependent on the neuroanatomical and neurophysiological properties of the target area. The
!
2!
here presented method does not allow the recording of neuronal activity from the brain of
flying microbats, which is mainly due to the high muscular activity during flight.
The second study deals with the utilisation of acoustic landmarks during mazenavigation in visually impaired/blind and sighted but blindfolded human subjects. The maze
consisted of three adjacent, interconnected rooms. In short, the study revealed gender-specific
differences in some (but not all) of the investigated aspects. The found gender differences are
in accordance with previous studies in humans and other mammals in which gender
differences in visually-guided orientation have been described. However, the here presented
results also indicate that non-spatial internal mediators such as “security seeking” might have
amplified these differences. The study further revealed that the visually impaired and blind
subjects used the provided acoustic landmarks to a lesser extent than the fully sighted but
blindfolded control group. These results might appear to be unexpected at first glance. I
concluded that the visually impaired and blind subjects had an advantage over the sighted but
blindfolded control in that they were more experienced in the utilisation of passive (and
sometimes even active) echoacoustic spatial cues for orientation. They, thus, might have used
this additional source of spatial information instead of the landmarks. The fact that most
subjects were able to produce accurate, bird-view (i.e. allocentric) plasticine models of the
way through the maze after they walked it several times in the complete absence of any visual
information indicates that allocentric spatial representations are a- or supramodal.
The third study provides a first scientific, physical description of the vocalisations of
the bony-fish species Ariopsis seemanni and investigates, whether the spatial dimensions of a
test environment (aquarium vs. pool) can critically influence the measuring of such calls. In
three subsequent behavioural experiments, it was examined, whether the vocalisations are
used for a coarse from of echolocation as postulated in the 1970s. The first two behavioural
experiments did not provide supportive evidence for this function of the calls in A. seemanni.
The results of the third behavioural experiment, in which some of the experimental animals
showed a positive phonotaxic response to the calls of conspecifics, rather point towards a
possible role of the calls in school formation and preservation.
!
3!
Zusammenfassung
Daniel Schmidtke
Auditorische Raumkognition bei verschiedenen Wirbeltiertaxa
In der vorliegenden These werden drei Studien im Bereich der auditorischen
Raumwahrnehmung vorgestellt, welche an verschiedenen Wirbeltierarten durchgeführt
wurden. Dank intensiver Forschung im Bereich der allozentrischen Raumwahrnehmung bei
Nagetieren, konnte in den letzten Jahrzehnten eine Vielzahl an wichtigen Erkenntnissen über
die neuronalen Substrate innerer, räumlicher Repräsentationen der Außenwelt gewonnen
werden. Während die Verwendung definierter Tiermodelle klare Vorteile mit sich bringt
(Standardisierbarkeit von Methoden, hohe Vergleichbarkeit von Versuchsergebnissen, etc.),
zeigen sich im Bereich der Raumkognitions-Forschung zunehmend auch Nachteile, die aus
der Beschränkung auf eine kleine Anzahl von Arten hervorgehen. So lassen sich durch den
Einsatz einiger weniger, nahe verwandter Spezies nur geringe Aussagen über die Phylogenie
und die Generalisierbarkeit eines betrachteten Phänomens treffen. Durch die standardisierte
Verwendung
von
Orientierungsreizen
zweidimensionalen
fehlen
uns
ferner
Versuchsarenen
Erkenntnisse
und
über
hauptsächlich
die
visuellen
Mehrdimensionalität
allozentrischer Raumrepräsentationen und deren Abhängigkeit von verschiedenen Sinnesmodalitäten. Diese werden jedoch nur über einen Wechsel des Tiermodells und unter der
Verwendung komplexer, dreidimensionaler Versuchsumgebungen gewonnen werden können.
Die erste, hier vorgestellte Studie beschäftigt sich daher mit der Frage, inwieweit sich
ein kürzlich am Max-Planck-Institut für Ornithologie entwickeltes und an Zebrafinken
getestetes, miniaturisiertes Telemetrie-System für die Charakterisierung der rezeptiven Felder
neuronaler,
allozentrischer
Repräsentationen
in
medial-temporalen
Bereichen
des
Kleinsäuger-Hirnes eignet. Um dies testen zu können, war es zunächst notwendig, in
Kooperation mit dem Laser Zentrum Hannover e.V. ein Protokoll für die Herstellung von sehr
kurzen Metall-Mikroelektroden zu entwickeln. Die hier präsentierten Ableitungsexperimente
an wachen, sich frei bewegenden Mäusen (Mus musculus) und Fledermäusen (Phyllostomus
discolor) zeigen, dass sich die Kombination aus selbsthergestellten Mikroelektroden und
Telemetrie-System bestens eignet, um bei wachen Kleinnagern neuronale Antworten auf
Mehrzellebene abzuleiten. Die Möglichkeit, qualitativ hochwertige Einzelzellableitungen zu
!
4!
erhalten, scheint dagegen stark von den neuroanatomischen und -physiologischen
Eigenschaften des Zielareals abhängig zu sein. Das Ableiten neuronaler Aktivität aus dem
Gehirn fliegender Fledermäuse war aufgrund der hohen Muskelaktivität während des Fluges
mit der verwendeten Methode nicht möglich.
Die zweite Studie beschäftigt sich mit der Nutzung akustischer Landmarken bei der
Orientierung von stark sehbehinderten/blinden und temporär visuell deprivierten Menschen in
einem aus drei aneinandergrenzenden Räumen bestehenden Labyrinth. Zusammengefasst
zeigt die Studie, dass in einigen (aber nicht allen) der untersuchten Teilaspekte
Geschlechtsunterschiede bestehen, die dem entsprechen, was zuvor bei Menschen und
anderen Säugetieren für die Verwendung visueller Landmarken beschrieben wurde. Die hier
präsentierten Ergebnisse weisen jedoch auch darauf hin, dass diese Unterschiede
möglicherweise durch nicht-räumliche, interne Mediatoren (z.B. „security seeking“) verstärkt
werden. Des Weiteren ergab die Studie, dass sehbehinderte und blinde Probanden weniger
stark von den als Orientierungshilfen dargebotenen akustischen Landmarken Gebrauch
machten, als die temporär visuell deprivierte Kontrollgruppe. Ich komme zu der
Schlussfolgerung, dass den sehbehinderten und blinden Probanden zusätzlich zu den
akustischen Landmarken passive (und teils auch aktive) echoakustische räumliche Hinweise
als Informationsquelle zur Verfügung standen, deren Nutzung sie sich in Folge ihrer
Sehbehinderung zuvor allmählich erschlossen haben. Der temporär visuell deprivierten
Kontrollgruppe fehlte diese Informationsquelle aufgrund mangelnder Erfahrung. Die
Tatsache, dass die meisten der Probanden trotz der vollständigen Abwesenheit visueller
Informationen in der Lage waren, nach Abschluss der Labyrinthversuche ein akkurates,
allozentrisches Modell des Weges durch das Labyrinth aus der Vogelperspektive zu kneten,
deutet darauf hin, dass allozentrische Raumrepräsentationen a- oder supramodal sind.
Die dritte Studie liefert eine erste physikalische Beschreibung der Laute der
Knochenfischart Ariopsis seemanni und untersucht, ob die räumlichen Dimensionen des
Versuchsbeckens (Aquarium vs. Pool) einen wesentlichen Einfluss auf die naturgetreue
Messung derartiger Laute haben. Ferner wurde in drei Verhaltensversuchen überprüft, ob die
Laute, wie in den 70er Jahren postuliert, für eine grobe Form der Echoortung genutzt werden.
Die ersten beiden dieser Versuche lieferten keine Evidenzen für eine solche Funktion der
Laute. Die Ergebnisse des dritten Verhaltensversuchs, in dem einige der Fische Anzeichen für
!
5!
eine positive phonotaktische Reaktion auf die Laute von Artgenossen zeigten, deuten auf eine
mögliche Rolle der Laute für die Schwarmbildung bzw. den Schwarmzusammenhalt hin.
!
6!
!
7!
Chapter 1: General Introduction
Space
“Space is a form abstracted from matter and exists only in consciousness”
Ikhwan al-Safa (group of Muslim penseum libres ca. 900 A.D.)
The question on the nature of space has bothered the minds of mankind’s greatest thinkers
from the Geometers and Atomists of the classical antiquity to the philosophers and natural
scientists of today. As a consequence of this extensive, more than two and a half millennia
lasting debate, a large amount of partially competing concepts of space were established, with
our current physical understanding of space being the result of a continuous process of
abstractions from the mere cognitive experience of space. Since it would fill too many pages,
a review of the philosophical considerations that led to the latest conceptions of space can,
unfortunately, not be given here, even though I would regard it as highly beneficial. As
Einstein (1954) notes in the foreword of Jammers monograph on the history of theories of
space in physics, a scientist, while focusing only on observable phenomena, makes use of a
vast amount of concepts, treating them as “something obviously” and “immutably given”
without “being aware of the eternally problematic character of his concepts”. Einstein
considers it “necessary over and over again to engage in the critique of these fundamental
concepts, in order that we may not unconsciously be ruled by them”. This might especially be
true for researchers working in the field of spatial cognition. For the interested reader I,
therefore, recommend the third, enlarged edition of Jammers monograph (1993) to get an
insight into the physical and philosophical conceptions of space and their historical dynamics.
In the online edition of the Oxford Dictionary (accessed in January 2013, url:
http://oxforddictionaries.com/definition/english/space), “space” is defined as “the dimensions
of height, depth, and width within which all things exist and move”. While most nonphysicists intuitively would agree to this definition of space, it is, in my eyes, noteworthy that
the definition is neither in accordance with our current physical concepts of space, nor with
how we, as humans, directly perceive space (as it includes several levels of abstraction). With
regard to the former, the Oxford Dictionary fails to mention the mathematical unity of space
!
8!
and time (Minkowski 1909), the relativity of this four-dimensional space-time continuum in
terms of geometry (Einstein 1917), and considerations on an even higher dimensionality of
space-time as made by some of the more recent, mathematical models of the universe (e.g.
Kaluza–Klein Theory: Kaluza 1921, Klein 1926; M-Theory: Townsend 1995, Witten 1995).
Instead, it intuitively suggests a concept of space having a three-dimensional Euclidean
geometry and, hence, complying with the laws of Galilean-Newtonian physics. Spatial
cognition, however, is based on the processing and storage of spatial information by several
parallel systems (Burgess 2008), of which at least some are inherently non-Euclidean
(Fernandez and Farell 2009, Spelke and Lee 2012). These parallel systems mutually interact
with each other to support behaviour under varying circumstances.
The study at hand mainly focuses on the long-term aspect of spatial cognition in
mammals. The two succeeding subchapters, therefore, will give a brief introduction to what is
currently known about egocentric and allocentric spatial representations and their roles in
spatial memory, orientation, and navigation, followed by a description of two spatially tuned
cell types (place and grid cells) constituting a part of the neuronal network for allocentric
spatial cognition. The lack in our knowledge about the three-dimensional firing properties of
these neurons will be discussed, as it explains the motivation behind the two studies presented
in chapter 2 and chapter 3 of this thesis. The last two subchapters of chapter 1 introduce the
phenomenon of gender differences in spatial learning and navigation in mammals and, finally,
integrate the studies of chapter 2 to chapter 4 into the general theoretical framework of this
thesis.
Egocentric Representations of Space
In egocentric representations of space, as the name implies, spatial information of the
environment and idiothetic spatial information (e.g. vestibular or proprioceptive cues) are
represented in terms of their relation to the subject, i.e. in body-centred coordinates (Hartley
and Burgess 2003). Since sensory information generally is acquired in different body-centred
coordinates (e.g. head-centred auditory information or retinotopically organised visual
information), the information of, for example, the location of a multimodally perceived object
has to be translated between various egocentric reference frames. Further such translations are
!
9!
needed for sensorimotor integration, allowing for guided interactions with the perceived
environment. While evidence for different egocentric representations can be found in the
firing characteristics of single neurons in the motor and sensory cortices of mammals, a
candidate substrate for the translation processes is the posterior parietal area 7a (e.g. Burgess
2008, Deneve et al. 2001, Zipser and Andersen 1988). In rhesus monkeys, neurons from this
area show so-called “gain-field” responses: They exhibit their peak firing rate if the visual
information of an object stimulates a specific retinotopic location. However, the firing rate of
these neurons has been shown to be additionally modulated by (i) the orientation of the
animal within its environment (Burgess 2008), (ii) the direction of the head relative to the
body (Snyder et al. 1998), or (iii) the position of the eyes relative to the head (Andersen et al.
1987). It is noteworthy that, while the last two modulating factors are perceptual and
egocentric in quality, the orientation of the monkey within the environment is an allocentric
variable. Hence, some neurons in area 7a potentially allow for the translation of information
between
allocentric
and
egocentric
coordinate
systems,
most
probably
through
interconnections with the medial temporal lobe (Burgess 2008; see below).
In terms of scale, parietal, egocentric processes are mainly concerned with the
immediate surrounding of a subject and take place on relatively short time scales. In order to
retain validity over longer periods of time, egocentric representations have to be constantly
updated with new sensory information to correct for changes in, for example, the location of a
subject within its environment (Hartley and Burgess 2003). The process of continuously
updating the representation of one’s own position relative to an object or a starting point is
called “path integration” and was first postulated by Darwin (1873). It is based on self-motion
cues (including visual, vestibular, and proprioceptive cues) and possibly on short-term copies
of locomotor commands, so-called “efference copies” (Etienne and Jeffery 2004). Whether
path integration is considered an egocentric or an allocentric process solely depends on the
point of view. It can be regarded as the updating of the position of an object or a starting point
relative to one’s own position (egocentric) or vice versa (allocentric). Path integration,
however, is prone to cumulative errors that can rapidly (i.e. after short times/traveling
distances) exceed a critical level (e.g. Etienne et al. 1996). When locations have to be
revisited over longer periods of time or after a high amount of self-motion, more enduring
representations are needed for successful navigation (Burgess 2006).
!
10!
Allocentric Representations of Space
In allocentric spatial representations, locations are related to the position of prominent,
environmental landmarks or to each other. This means that spatial information is stored and
processed in world-centred coordinate systems that are view-point independent (Hartley and
Burgess 2003), as opposed to the body-centred reference frames of egocentric representations.
Since the environment, in which a subject lives, under normal conditions is relatively stable,
allocentric representations do not have to be continuously updated and, consequently, are
much less vulnerable to cumulative errors. Allocentric spatial representations, therefore, can
operate over substantially longer periods of time (up to the life-span of a subject) and over
large distances.
The first to postulate the existence of allocentric spatial representations was Tolman in
1948. In a maze study with rats, he detected a decreasing error rate in maze-solving the more
often the task was performed and reasoned that this finding can be explained best under the
assumption of the formation of some form of mental map of the maze (Tolman 1948). Thirty
years later, O’Keefe and Nadel (1978) proposed that the hippocampus is the neuronal
substrate for this mental map. Their most valid argument supporting this idea was the
discovery of pyramidal neurons in the hippocampus proper of rats that show an increased
firing activity whenever a subject is at a certain location relative to its environment (“place
cells”; O’Keefe and Dostrovsky 1971). Three further allocentric spatial representations, one
of horizontal orientation (“head-direction cells”; Taube et al. 1990), one with a hexagonal,
grid-like receptive field possibly being involved in path-integration processes (“grid cells”,
Hafting et al. 2005), and one with a receptive field that is responsive to the presence of
environmental boundaries (“border cells”; Lever et al. 2009, Solstad et al. 2008), were
discovered later.
Today, it is known that place cells also occur in neighbouring brain areas of the
hippocampus proper, such as the dentate gyrus (Jung and McNaughton 1993) and the medial
entorhinal cortex (Hafting et al. 2005). Head-direction cells are even more widely distributed
and can, for example, be found in the thalamus, the presubiculum, the lateral mammillary
bodies (Doeller et al. 2012), the striatum (Wiener 1993), and the entorhinal cortex (Sargolini
!
11!
et al. 2006). Border cells have so far only been found in the entorhinal cortex (Solstad et al.
2008) and the subiculum (Lever et al. 2009), while grid cells seem to be a type of neurons that
is unique to the entorhinal cortex. Together, these mainly medial temporal structures form a
network of functionally related brain areas, bearing different types of allocentric
representations that guide rodent behaviour in situations where egocentric representations are
insufficient (Burgess 2008).
The Firing Properties of Place and Grid Cells
As the list of studies briefly summarised in the preceding subchapter implies, the bulk of our
knowledge about allocentric spatial representations stems from single-cell recording
experiments in rodents. Consequently, our current models of spatial cognition (for review see
Hartley and Burgess 2003) are mainly based on the rodent literature. Only in the more recent
past, neurons with spatially modulated firing similar to that of place cells in rats have also
been found in other mammals (e.g. humans: Ekstrom et al. 2003; non-human primates:
Ludvig et al. 2004; mice: Kentros et al. 2004; microbats: Ulanovsky and Moss 2007).
While grid cells could be proven to exist in approximately the same spectrum of
mammalian species (non-human primates: Killian et al. 2012; mice: Fyhn et al. 2008;
macrobats: Yartsev et al. 2011), the grid network described in the rhesus monkey (Killian
et al. 2012) is laid across visual space rather than physical space (i.e. different from those in
rodents and macrobats). Evidence for grid cells in humans has only been found indirectly
(Doeller et al. 2010, Doeller et al. 2012), yet. In general, it remains unclear and is highly
controversially discussed, in how far models of allocentric spatial cognition can be
generalised for different mammalian (super-) orders (Bingman 1990, Jacobs and Schenk
2003).
A second, less obvious consequence of the fact that mainly rodents have been used to
study the neuronal basis of allocentric spatial representations is that very little is known about
how a subject’s elevation relative to the environment (i.e. the third axis of 3D-space) is
represented in the brain. The standard layout of the experimental environments that have been
used in most of the studies investigating the receptive fields of the different neuronal
types of allocentric spatial representations (compare Fig. 1.1) was that of a horizontal,
!
12!
Fig. 1.1. Exemplary depiction of idealised firing-rate maps of a place and grid cell as measured in rats
that freely moved along horizontal surfaces. The colour scheme is linearly scaled from the minimum
firing rate (dark blue) to the maximum firing rate (dark red). During exploration of a two-dimensional
surface, place cells in the hippocampus of rats exhibit approximately circular, isotropic receptive fields
(left). The same is true for grid cells (right). Each grid cell, however, has multiple receptive fields that
are organised in a regular, hexagonal pattern (indicated by the dashed white line).
!
13!
two-dimensional plane on which the subjects could freely move and that was limited by
vertical boundaries. Strictly spoken, our current knowledge of the neural representations of
space supporting large-scale navigation, therefore, merely allows for statements about the
neuronal networks underlying navigation in horizontal, two-dimensional planes. The
existence of three-dimensional, allocentric representations and their possible forms can only
be speculated upon (Fig. 1.2).
I first encountered this problem when working on a study on gender-differences in
orientation strategies in Phyllostomus discolor, a laryngeally echolocating microbat species
(Schmidtke and Esser 2011; vide infra). In their everyday life, during flight, microbats
routinely move actively and freely along all axis of three-dimensional space. Consequently, it
has to be assumed that allocentric spatial representations in microbats have to account for an
animal’s elevation relative to the environment in order to allow for a successful revisiting of
known places by flight. This, however, does not necessarily have to be a feature that has
evolved exclusively in microbats and other flying or diving mammals (such as megabats,
sirenians, and cetaceans). Since the natural habitats of most mammals are more complex than
simple, two-dimensional planes, it is more likely that three-dimensional allocentric
representations generally exist in the mammalian brain.
Previous to the time when I started working on this thesis, to my knowledge, only four
studies had been published that investigated whether location-sensitive neurons, such as place
and grid cells, show three-dimensional receptive fields. The first one (Knierim et al. 2000)
was conducted on a space-shuttle mission and examined the spatial selectivity of hippocampal
neurons in rats under micro-gravity conditions. The authors could show that hippocampal
place cells can form reliable representations of locations on the three orthogonal surfaces of a
modified Escher/Penrose staircase in microgravity. However, the collected data was
insufficient to determine whether the staircase track was represented fully threedimensionally, or whether independent two-dimensional representations were developed for
each orthogonal surface. In a follow-up study (Knierim and McNaughton 2001), the firing
properties of hippocampal place cells were analysed in rats moving along a rectangular
surface that could be tilted by 45° relative to three-dimensional cues of the environment. In
response to this tilting, some of the place cells that were recorded from maintained their place
!
14!
Fig. 1.2. Hypothetical firing fields of place cells as they might be recorded during unrestricted
movement of the experimental animal (e.g. a microbat) along all the axes of three-dimensional space
and in a three-dimensional environment with symmetrically distributed spatial cues (Xs and Os). The
hypothetical firing fields are grouped into three classes: two-dimensional isotropic (left), threedimensional isotropic (centre), and three-dimensional anisotropic (right). Hypothesis I (green, left):
Place cells have isotropic (circular) firing fields that are two-dimensional and restricted to the
horizontal plane. Hypothesis II (purple, left): Place cells have two-dimensional, isotropic (circular)
firing fields that can also code for locations on vertical or tilted planes (compare Knierim et al. 2000,
Knierim and McNaughton 2001, Ulanovsky and Moss 2007). Hypothesis III (green, centre): Place
cells have three-dimensional, isotropic (spherical) firing fields and can code for locations everywhere
in the environment. Hypothesis IV (purple, centre): Place cells have three-dimensional, isotropic
(quasi-spherical) firing fields that code only for locations in the vicinity of bounding surfaces.
Hypothesis V (green, right): Place cells have three-dimensional, anisotropic (columnar) firing fields
that code for locations on the horizontal plane irrespective of the animals elevation. Hypothesis VI
(purple, right): Place cells have three-dimensional, anisotropic (elongated) firing fields that code for
locations in all spatial dimensions but with a lower sensitivity for changes in elevation than for
changes in the horizontal plane (compare Hayman et al. 2011). All hypotheses presented above can
also readily be applied to the multiple firing fields of grid cells. However, if the receptive fields of grid
cells should, indeed, be three-dimensional isotropic (Hypothesis III), the fascinating question arises,
whether the regular pattern of the receptive fields of a single grid cell (compare Fig. 1.1) remains
planar or if it extends into a “crystalline”, three-dimensional structure.
!
15!
fields, while others completely changed their receptive fields (i.e. they “remapped”). The
phenomenon of remapping can often be observed in place cells after (even minor) changes in
inputs to the hippocampus (for review see Colgin et al. 2008). While these results could show
that hippocampal place cells are sensitive to the manipulation of three-dimensional,
environmental cues, they could not prove that the receptive fields of place cells are inherently
three-dimensional.
The experiments of the third (Jeffery et al. 2006) and the fourth (Ulanovsky and Moss
2007) study took place in comparable, three-dimensional environments, as the experimental
animals again moved on tilted surfaces (Jeffery and colleagues: 30°; Ulanovsky and Moss:
almost vertical) while place-cell responses were recorded. In both studies, the receptive fields
of the place cells were approximately circular and isotropic, thereby closely resembling the
place fields that had previously been described in horizontal environments. The key finding of
the study performed by Jeffery and his co-workers was that place cells in the hippocampus of
rats can use the slope of an environment as a setting cue. The importance of the study of
Ulanovsky and Moss lies in the choice of their experimental animal, the big brown bat
(Eptesicus fuscus), and the fact that they were the first to describe hippocampal place cells in
a chiropteran species. Following Krogh’s principle (Krogh 1929), bats in general provide an
excellent opportunity to conduct research into the dimensionality of allocentric spatial
representations in mammals, as their primary mode of locomotion is aerial. Accordingly,
freely behaving bats can unrestrictedly explore three-dimensional environments.
To date, three studies on allocentric representations (here: place- and grid-cell firing)
in bats have been conducted: (i) the initial study of Ulanovsky and Moss (2007), (ii) a followup study on place cells using the same species (E. fuscus; Ulanovsky and Moss 2011), and
(iii) a study investigating place and grid cells in a macrobat (Rousettus aegyptiacus; Yartsev
et al. 2011). All of these studies, however, investigated the receptive fields of place and grid
cells in crawling instead of in flying bats. The reason behind this approach can be found in
technical restrictions of the recording setups used in these studies: In order to profit from the
benefits of a four-tetrode microdrive (i.e. an increased chance of finding appropriate recording
loci and the possibility to post-surgically correct these loci as well as advanced spike-sorting
possibilities in multitrode recordings; compare chapter 2), the neuronal signals were passed
on from a unity-gain preamplifier that was mounted on the microdrive to the downstream
!
16!
recording setup via counterbalanced wire connections. Such a system of counterbalanced
wires has the clear disadvantage that it does not allow for an unrestricted flight of the animals
over considerably long durations. Thus, the development of suitable miniaturised telemetric
recording systems would be a great leap forward to investigations of allocentric spatial
representations of complex, three-dimensional environments, be it in rodents (Knierim 2001)
or in bats (Ulanovsky 2011).
Gender Differences in Orientation Strategies
As stated above, it is unclear in how far the medial temporal networks supporting allocentric
spatial cognition in different mammalian species can be considered as being functionally
homologous. One approach to indirectly answer this question, is to search for a cognitive
phenomenon that emerges from complex interactions between different allocentric
representations and to test whether this phenomenon can consistently be found across taxa. A
possible candidate phenomenon for such an approach is that of gender differences in
orientation.
Gender differences in orientation have intensively been studied in humans (e.g. Beatty
2002, Galea and Kimura 1993, Sandstrom et al. 1998, Ward et al. 1986). It was found that
women and men differ in their relative use of spatial cues for orientation, with men relying
preferentially on Euclidean/geometric information, whereas women prefer to use landmark
cues (Sandstrom et al. 1998; a more detailed definition of the cue types mentioned here will
be given in chapter 3). On average, men outperform women in route-learning tasks and
direction judgments (Beatty 2002, Galea and Kimura 1993, Ward et al. 1986). Comparable
results have also been found in other mammalian species from the superorder
Euarchontoglires, such as rats (Roof and Stein 1999, Williams and Meck 1991), meadow
voles (Jacobs et al. 1990), or rhesus macaques (Lacreuse et al. 2005).
A study I conducted in the laryngeally echolocating microbat species P. discolor
recently provided evidence for the existence of comparable gender differences in orientation
in a laurasiatherian species (Schmidtke and Esser 2011). In this study, individuals of
P. discolor were trained to perform a simple orientation task (finding their way from a fixed
starting position to a landing platform at the opposite end of a flight tunnel), first in the
!
17!
absence and later in the presence of obstacles/echo-acoustic landmarks. Once the
experimental animals had established an individual, stereotyped route through the flight
tunnel, the landmark configuration was modified in order to investigate, whether this
modification leads to gender-specific behavioural responses. I found that, in the absence (but
not in the presence) of echo-acoustic landmarks, female individuals of P. discolor showed
lower learning rates than male conspecifics. In addition, the modification of the landmark
configuration led to a reduced landing-probability and inter-individually more variable
deviations from the learned route in female individuals as compared to males. These findings
are in line with what has previously been described in euarchontogliran species for visuallyguided orientation and, therefore, indicate that the phenomenon of gender-specific orientation
is widely distributed among mammals and independent of the modality that is primarily used
to sense the environment during orientation.
The only theoretical model on the role of the medial temporal structures in spatial
memory formation that incorporates the just described gender differences in orientation is the
“parallel map theory” (Jacobs and Schenk 2003). In short, this theory postulates two parallel
mapping systems encoding space in the hippocampal formation, the so-called “bearing map”
(constructed primarily in the dentate gyrus from directional cues) and the “sketch map”
(constructed within the hippocampus proper from positional cues). Both maps are
independent and complementary, but finally combine to form an integrated mental
representation of the environment that allows for successful navigation. Regarding genderspecific differences in orientation, the authors suggest that males and females rely to different
extents on the bearing map and on the sketch map, with females relying more on the latter.
Aims of This Thesis
The dissertational thesis at hand has two topical foci, the dimensionality of allocentric spatial
representations in mammals and acoustic orientation through echolocation. In chapter two and
three, these two topics go hand in hand, due to the choice of experimental animals or test
subjects that have been used: In chapter two, a methodological study is presented that sets out
to investigate in how far a recently designed, miniaturised micro-transmitter (Schregardus
et al. 2006) and two of its successors are suited for the utilisation in studies on the three-
!
18!
dimensional firing properties of spatially tuned cell types in the hippocampal formation of
freely behaving individuals of P. discolor. As mentioned above, the development and testing
of appropriate telemetric devices is a challenging and necessary prerequisite for the enterprise
to fill the gap in our knowledge of the dimensionality of allocentric spatial representations.
Chapter three describes a study investigating, whether the gender differences in the
utilisation of acoustic landmarks found in P. discolor (Schmidtke and Esser 2011) also exist
in blind and blindfolded humans. A positive finding in this matter would corroborate the ideas
that the brain areas belonging to the hippocampal formation constitute a neuronal substrate for
allocentric spatial representations that are (i) shared by and functionally equivalent in all
mammals and (ii) a- or supramodal, i.e. independent of the sensory quality of the spatial
information that is processed. A validation of both postulations would lend further support to
the establishment of bats as a new model for three-dimensional allocentric spatial
representations in mammals.
Since acoustic orientation in both, microbats and humans, is based on the processing
of echoacoustic cues (albeit to varying degrees), the third study performed in this thesis
(chapter 4) investigates the possible role of call production in Ariopsis seemanni, a vocal fish
species, in echolocation. Even though it could not be expected to find gender differences in
the use of landmarks in this species, an easily accessible and maintainable echolocating fish
species would provide a valuable tool for future studies on active acoustic orientation at the
lower end of the vertebrate spectrum and, consequently, for comparative studies on (echo-)
acoustic forms of spatial cognition in vertebrates.
!
!
19!
Chapter 2: Towards Telemetric Recordings of Allocentric Spatial
Representations in Freely Behaving Small Mammals
Characterising the Firing Properties of Place and Grid Cells
In order to characterise the receptive fields of cellular allocentric spatial representations, such
as place and grid cells, it is necessary to record action potentials of individual neurons in a
freely behaving subject. In the recent literature on place and grid cells, this is routinely done
using multitrodes (e.g. stereotrodes: McNaughton et al. 1983; tetrodes: Recce and O’Keefe
1989, Hafting et al. 2005, Ulanovsky and Moss 2007) that are chronically implanted in the
brain in close proximity to cell bodies of the target neurons. One or several of these
multitrodes are then attached to a microdrive allowing for further, post-surgical
micromanipulations of the multitrode position when searching for spatially modulated
neurons. The microdrive, in turn, is mounted to the headstage of the experimental animal. The
electric signals picked up by the multitrodes are then passed on to the downstream recording
setup via cable-connections.
The utilisation of multitrodes in experiments that aim to record the activity of single
neurons in a certain target area has two decisive advantages over the application of
conventional single-wire electrodes. The most obvious advantage is that the utilisation of
more than one recording electrode increases the chance of finding the desired target cells in
their respective brain area. The usage of microdrives even potentiates this advantage, as
microdrives allow for post-surgical corrections of the recording loci. The second advantage is
the increased discriminative power of multitrode recordings regarding different units recorded
with the same multitrode. The technique that is normally used for isolating single-cell
responses from multiunit recordings (for single-wire electrodes and multitrodes) is cluster
cutting, where different waveform parameters of each spike are plotted against each other.
The underlying assumption is that spikes being recorded from different cells show a higher
variation in these parameter combinations than action potentials from the same neuron. In the
ideal case, the plotted waveform parameters, therefore, form clusters that can be used to
discriminate different cells. Since in multitrode recordings, the spikes produced by a certain
neuron are usually picked-up by more then just one of the wires of the multitrode, the
!
20!
information from neighbouring wires can be combined to calculate new parameters that can
be used for cluster cutting. For example, if the spikes of a neuron are picked-up by the four
wires of a tetrode, the peak amplitude of the spikes will most likely vary between the
recordings from the different channels, since each wire has a different distance to the cell
body of the spike-producing neuron. The variance in peak-amplitudes between the four
recording-channels can, thus, be used to calculate the approximate 3D location of the neuron
relative to the individual wires of the tetrode, a parameter that has a high discriminative power
in cluster-cutting analyses (Gray et al. 1995, McNaughton et al. 1983, Recce and O'Keefe
1989).
Regarding the endeavour to three-dimensionally characterise the receptive fields of
allocentric spatial representations in freely behaving mammals (especially microbats), the
multitrode technology also has a clear disadvantage: The multitrode setups that are currently
used in studies on place and grid cells in mammals are tethered, i.e. they conduct the neuronal
signals via cable connection to the subsequent recording/data-analysis devices (e.g.
Ulanovsky and Moss 2011, Yartsev et al. 2011). While movement in two-dimensions in small
arenas might be considered as being unrestricted by such a tethered setup, it is obvious that in
complex, three-dimensional environments, telemetric solutions have to be used in order to
allow for an unimpeded movement of the experimental animal over larger distances and
longer periods of time.
A suitable transmitter for neuronal signals from freely behaving animals has to fulfil at
least three requirements (Fan et al. 2011): Firstly, is has to be compact and as light in weight
as possible to limit the impact of the additional weight on the animal’s behaviour. In order to
avoid detrimental effects to behaviour, as a rule of thumb, an experimental animal should not
be loaded with more than 5% of its bodyweight (Cochran 1980). Secondly, wireless
transmission should not lead to a significant decrease in signal quality as compared to a
tethered setup. Finally, signal transmission has to be guaranteed over a purposeful range, so
that the animal can move away from the receiver without reductions in recording quality.
When I started working on the methodological study presented in this chapter in 2008,
telemetric multi-channel devices that fulfil all these criteria with respect to the designated
experimental animal (P. discolor) were not commercially available. Adult specimens of
P. discolor weigh approximately 40 g (Kwiecinsky 2006), leaving a maximum of 2 g for a
!
21!
telemetric device (including batteries, electrodes and headstage) to comply with Cochran’s
rule. However, a miniaturised (two-channel) telemetric device developed by Schregardus and
his co-workers (2006) for recording neuronal activity in freely behaving zebra finches
satisfied all the above-mentioned criteria. I, therefore, decided to investigate whether this
telemetric device or one of two follow-up models allow for single-unit recordings in the
hippocampus of freely behaving small mammals.
In the subsequent subchapter, I will briefly summarise the characteristics of the
telemetric device and its successors that were mainly developed at the Max-Planck-Institute
for Ornithology in Seewiesen (by the working group of Prof. Dr. Manfred Gahr), followed by
two subchapters introducing a protocol for the manufacturing of stainless-steel
microelectrodes that I developed in collaboration with the Laser Zentrum Hannover e.V. and
a surgical protocol that describes how to chronically implant these electrodes in mice and
microbats in order to work with the transmitter devices. In the last subchapter, I will briefly
summarise the results of my methodological work and discuss the limitations and capabilities
of the here described methods.
Telemetric Devices
The first generation of transmitters used for the methodological study presented here (G1)
was developed by colleagues from the Max-Planck-Institute for Ornithology in Seewiesen
and the Center for Neurogenomics and Cognitive Research (University of Amsterdam;
Schregardus et al. 2006). The device has a size of approx. 12 x 5 x 8 mm and a weight
ranging between 0.45 and 0.47 g (Fig. 2.1). A high input impedance (> 1 TΩ) guarantees
independence of the system from electrode impedance. Technical components of the device
are a differential amplifier (11x gain, ac-coupled), amplifying the voltage difference between
a recording and a reference electrode and providing a mono signal, an RC filter (bandpass:
100 - 15000 Hz; vertebrate neuronal spikes have a frequency content of 200 - 10000 Hz), and
a FM-transmitter (carrier frequency: 450 MHz < cf < 500 MHz). The range of the transmitter
in which a high quality of the signal can be guaranteed is 4 m, allowing for transmission in a
room of approx. 8 x 8 x 4 m (width x depth x height) when a single, centrally installed
antenna is used. Two female IC-connectors are used to support the device and to connect it
!
22!
Fig 2.1. Depiction of the three generations of transmitters used in the here presented study and size
comparison. a The three generations of transmitters (G1 - G3) were developed at the Max-PlanckInstitute for Ornithology in Seewiesen, Germany (working group of Prof. Dr. Manfred Gahr) in
collaboration with colleagues from the Center for Neurogenomics and Cognitive Research at the
University Amsterdam, Netherlands (working group of Prof. Dr. Oliver Stiedl). b The dimensions of
the transmitters are approx. 12 x 5 x 8 mm. For a size comparison, a transmitter of the first generation
(G1) with a connected microelectrode is depicted next to the skull of a specimen of P. discolor and a
European two-cent coin. The transmitters weigh between approx. 0.45 (G1 and G2) and 0.38 g (G3).
!
23!
to a recording and a reference electrode (Fig. 2.1b). The transmitter operates on two standard
hearing-aid batteries (zinc-air, ZA10, 1.4 V, 90 mAh) that add an extra 0.6 g of weight. The
recording electrodes I used (see below) weighed 0.05 g each and were attached to the skull of
the experimental animal using approx. 0.2 g of dental cement (see below). In total, the
G1-transmitter including batteries and the chronically implanted electrodes adds a weight of
1.3 to 1.4 g to the head of an experimental animal.
The telemetric device, as described above, was successfully tested in freely behaving
zebra finches using chronically implanted, sharp tungsten electrodes to record from field L
units in the bird brain (Schregardus et al. 2006). Zebra finches weigh between 12 to 17 g.
Even though the weight of the transmitter clearly exceeds 5% of the body weight of the zebra
finches, Schregardus et al. (2006) reported that no behavioural abnormalities in the
experimental animals could be observed in terms of song rate and perch hopping activity. The
two successors of the transmitter (G2 and G3, Fig. 2.1a), which I used later in my work,
differed only slightly from the G1 model: The final G3 model featured an even lower weight
(0.37 to 0.39 g), longer battery durability (up to 5 days), and improved transmission stability.
In addition, the gain of the differential amplifier was reduced from 11x (G1) to 10x (G2 and
G3).
Manufacturing Short Elgiloy (Stainless Steel) Microelectrodes for
Telemetric Neuronal Recordings in Freely Behaving Small Mammals
For the first experiments conducted as a part of this methodological study, I used
commercially available (FHC), custom-made, sharp Elgiloy (stainless steel) electrodes for
recording. However, in the long term, the purchase/ordering of electrodes proved to be too
inflexible, due to the long production and delivery times of custom-made electrodes (up to 10
weeks for each new set of electrodes). Since I needed electrodes with constantly varying
parameters in terms of length and impedance for my experiments, I found it more feasible to
develop a new manufacturing protocol for Elgiloy electrodes that allows for the fabrication of
customised electrodes on-site. I developed this protocol in cooperation with the working
group of Prof. Dr. Holger Lubatschowski (Laser Zentrum Hannover e.V.), who provided me
!
24!
with an exciplex laser and gave me access to a scanning electron microscope. The next
paragraphs will briefly describe the single steps of the protocol.
As raw materials for electrode production, stainless steel wire (Elgiloy®, green, semiresilient, Ø = 0.25 mm, RMO) was used for the electrode shaft and standard IC-pins (AL-ST,
Assmann) for the electrode termination. The IC-pins were cut in half and a hole (approx.
0.2 mm in diameter and 2 mm deep) was drilled into the thin half of the pin. In a first step
(“Capping”, Fig. 2.2a), one end of the wire had to be electrolytically etched in sulfuric acid
(t = 4 s; UDC = 12 V, I = 0.5 A) in order to allow for the assembly of the stainless-steel wire
and the male connector. If done correctly, the friction between the two parts was sufficient to
keep them persistently connected to each other.
In a second step (“Etching of the shaft”, Fig. 2.2b), the Elgiloy part of the assembled
workpiece was repeatedly dipped into and pulled out of the sulfuric acid to reduce the shaft
diameter and to produce a sharp electrode tip through electrolytical etching (t = 7 - 10 min,
depending on the desired shaft length; UDC = 12 V, I = 0.2 A). The dipping motion during this
process followed a sinusoidal acceleration curve (s = 2 cm; f = 0.36 Hz), leading to a coneshaped shaft with a steadily decreasing diameter and a nano-scale sized tip.
After the electrode was etched to the desired length, the whole shaft including a small
part of the pin was immersed into a cathodic electropaint solution (Clearclad, HSR; step 3:
“Galvanic painting”, Fig. 2.2c). Electrodeposition of the paint (t = 2 min; UDC = 20 V)
followed by baking of the electrode (t = 25 min; T = 200°C) led to a uniform, insulating coat
(R ≥ 20 MΩ).
When painting the electrode with cathodic paint as described above, the whole
electrode shaft (i.e. including the tip) and the base of the male connector pin became
electrically insulated. In order to focally restore conductivity at the electrode tip, an exciplex
laser (KrF, λ = 248 nm) was used to fire short pulses of ultraviolet laser light at the tip
(n = 20, Epulse = 9.42 µJ, repetition rate = 100 Hz; step 4: “Laser ablation of the electrode tip”,
Fig. 2.2d). This procedure basically "cut off" the very tip of the electrode shaft, leaving
behind a small area of exposed stainless steel. Electrode impedance could be regulated by
varying the size of this exposed area.
In summary, the protocol I described here allows for the production of leak-free
stainless-steal electrodes of any desired length (or shortness). The electrodes terminate in a
!
25!
Fig. 2.2 (part 1). Depiction of the first two working steps necessary for the manufacturing of the here
described Elgiloy electrodes (a, b). The protocol was developed by me in collaboration with the Laser
Zentrum Hannover e.V., Hannover, Germany (working group of Prof. Dr. Holger Lubatschowski) that
provided me with an exciplex laser and gave me access to a scanning electron microscope. Details on
the protocol are described in the main text. a “Capping”. b “Etching of the shaft”.
!
26!
Fig. 2.2 (part 2). Depiction of the last two working steps necessary for the manufacturing of the here
described Elgiloy electrodes (c, d) and the resulting electrode tip after laser ablation (e). c “Galvanic
painting”. d “Laser ablation of the electrode tip”. e Scanning electrode micrographs (magnification:
5000 x) of the tips of two finished electrodes.
!
27!
standard IC-pin, so that they can easily be connected to downstream (telemetry) devices.
Electrode impedance can also effortlessly be controlled. For example, if the electrode is “cut”
at a distance of 50 µm from the tip, an active surface of approx. 78 µm2 (Fig. 2.2e) with a
corresponding impedance of about 8 MΩ is produced. For the experiments described in the
following subchapter, I usually used electrodes with impedances between 3 and 5 MΩ.
Surgical Protocols for the Chronic Implantation of the Electrodes in Mice
(Mus musculus) and Microbats (Phyllostomus discolor)
As a basis for the surgical protocol described in this subchapter, I used the methods
established by the colleagues who developed and successfully tested the telemetric device in
zebra finches at the Max-Planck-Institute for Ornithology (compare Schregardus et al. 2006
for details). I learned the protocol during two research stays at the Max-Planck-Institute for
Ornithology in 2008. Due to considerable differences between the anatomy of birds and
mammals and the different target area(s), however, several major adaptations had to be made
in order to allow for the utilisation of the telemetric devices in mice (and microbats).
For an initial testing of the telemetric devices, I decided to record from neurons in the
inferior colliculus of the house mouse (M. musculus). The inferior colliculus is the main
auditory midbrain nucleus in mice (Portfors et al. 2011) and highly tonotopically organised
(Stiebler and Ehret 1985). Auditory stimulation of an experimental animal using sinusoidal
(pure tone) acoustic stimuli reliably leads to highly specific neuronal activities in this brain
area. Hence, the inferior colliculus was an appropriate target area for a first investigation of
the suitability of the transmitters for electrophysiological studies in freely behaving, awake
mice.
For the surgery, the experimental animals were anesthetised by means of halothane
(Fluothane, ICI Pharma) inhalation (0.5 - 2.0% at 0.15 l O2/min) and placed on a heating pad
to keep their body temperature constant at 37°C. In order to get access to the cranial bones, a
craniocaudal incision was made from the nasal to the interparietal bone and the skin covering
the skull above the inferior colliculus was carefully reflected. Subsequently, a small, L-shaped
metal pin was attached to the skull (on the border between the nasal and the frontal bones)
using dental cement in a three-step process: Firstly, the bone-tissue was conditioned using a
!
28!
20% phosphoric-acid gel (Etch 20 Gel, GLUMA). Secondly, an adhesive system (Comfort
Bond, GLUMA) was applied to the conditioned location. Lastly, the metal pin was brought in
position and fixated using a light-curing composite (CHARISMA, Heraeus). The pin was then
clamped to a headholder that allowed for a three-dimensional manipulation of the position of
the skull. This stereotaxic solution had to be used, since most of the commercially available
stereotaxic headholders use ear bars to rigidly fix the experimental animal. It is a known
problem of such ear bars that they can occasionally cause severe damage to the auditory
apparatus of the experimental animal. Thus, ear bars are neither appropriate for the fixation of
mice in auditory studies, nor for studies in microbats in general, since microbats are reliant on
a functioning auditory apparatus for echolocation purposes.
After the skull of the experimental animal had been brought into a standard horizontal
position, the bone tissue surrounding the designated electrode locations had to be thoroughly
prepared using the same conditioning gel and adhesive system as described above (Etch 20
Gel and Comfort Bond, GLUMA). Subsequently, a dentist drill (F680, Faro) was used to
remove the bone tissue covering the inferior colliculus of the right hemisphere (the radius of
the craniotomy was approx. 0.7 mm). A second, smaller craniotomy was performed on the
left side of the skull to allow for the positioning of the reference electrode (flexible stainless
steel wire, Ø = 0.1 mm). The tip of the reference electrode was inserted between the skull and
the dura mater. Once in place, the remaining stainless steal wire protruding from the skull was
fixated and the craniotomy was closed using a small amount of light-curing dental cement
(CHARISMA, Heraeus). Afterwards, the dura mater beneath the right craniotomy was
removed and the recording electrode was brought into position using the λ-point of the skull
as a stereotaxic reference.
At this point, the reference electrode and the recording electrode were connected to a
differential amplifier (DAM 80, World Precision Instruments) via cables. The amplified
signal was then fed into a band-pass filter (Heinecke, Germany; bandwidth: 0.3 - 10 kHz),
followed by an oscillograph and an audio monitor. In addition, the filtered signal was
digitised for data storing and off-line analysis. The acoustic stimuli (pure tones) were
produced using a sine-wave generator (model 311B, OR-X) and subsequently pulse-shaped
(self-built pulse shaper, University of Ulm, Germany), attenuated (RA-920A, Kenwood), if
!
29!
necessary, amplified by an audio-amplifier (PMA-1055R, DENON), and played back through
a dynamic loudspeaker (D28/2, Dynaudio).
With the acoustic stimuli of known frequency continuously being played back as short
(depending on the recording session between 50 and 100 ms) pulses with inter-pulse intervals
between 0.5 and 1 s (the duration of the inter-pulse intervals was kept stable within the same
recording session but could vary between different sessions), the recording electrode was
advanced into the inferior colliculus by a remote-controlled nano-stepper (model NS 1010,
Scientific-Precision-Instruments). Once a robust neuronal response to the acoustic stimuli was
found, the stimulus frequency was adjusted (in a range from approx. 1 to 25 kHz) to the best
frequency of the recorded neurons and the electrode position was corrected on the micronscale in order to get as close to the cell bodies of the spiking neurons as possible. When a
stable recording site was found this way, the bases of the male IC-pins of both recording and
reference electrode were durably affixed to the previously conditioned bone-tissue using the
same light-curing dental cement as before (CHARISMA, Heraeus).
In a final step, the cable connections were carefully removed from the male connectors
of the electrodes and replaced by the telemetric device. A passive antenna was used for
reception of the transmitted, frequency-modulated radio signal. For demodulation, I used a
wide-band receiver (AR5000, AOR) connected to a spectrum scope (SDU5600, AOR) that
allowed for an online monitoring of the quality of the radio signal as well as for automated
tracking of the radio signal to compensate for minor fluctuations in the carrier frequency of
the transmitter. The demodulated signal was then fed into the same setup as described above
for the tethered recordings.
For a second testing of the telemetric devices, I recorded from neurons in the
hippocampus of M. musculus. The experiments were conducted during a research stay at the
Center for Neurogenomics and Cognitive Research (University of Amsterdam) in the working
group of Prof. Dr. Oliver Stiedl that has an expertise in telemetric multi-unit recordings from
the mouse hippocampus in the context of fear conditioning (Jansen et al. 2010, Stiedl et al.
2010). The protocol for the chronic implantation of the electrodes was similar to the one
described above, except for adapted stereotaxic coordinates and the fact that no acoustic
!
30!
stimulation was needed. Instead, the recording electrode was advanced into the hippocampus
until a (presumably) single neuron showing clear, spontaneous spike activity was found.
In the last testing stage, electrodes were chronically implanted in the auditory cortex
of the microbat species P. discolor. In order to get access to the skull of an anaesthetised
individual of P. discolor, different than in mice, the temporalis muscles had to be separated
from the weakly developed, sagittal crest. Afterwards, the muscles were reflected and the
surface of the dorsal skull was thoroughly cleaned from tissue debris. The rest of the surgical
protocol corresponded to the one described above for recording from the inferior colliculus in
mice. Using the appropriate stereotaxic coordinates, the recording electrode was advanced
into the auditory cortex of P. discolor. Pulses of pure-tone stimuli in the ultra-sonic range
were presented via a condenser loudspeaker (self-built condenser loudspeaker, University of
Ulm).
Results and Discussion
With the telemetric devices, the self-manufactured microelectrodes, and the surgical protocol
described in the preceding subchapters, it was possible to record the expected, stimulusfrequency dependent neuronal activity in the inferior colliculus of mice (Stiebler and Ehret
1985) that were freely exploring a makrolon cage (length x depth x height: 0.6 x 0.38 x
0.20 m). Figure 2.3 shows two examples of such recordings from two different individuals.
Both examples were recorded approximately three hours after surgery. They were chosen to
illustrate the variability in recording quality that can be achieved with the here-described
method. During surgery, before the electrodes were ultimately fixed to the skull, the position
of the recording electrode was adjusted to obtain a good signal-to-noise ratio (SNR) of 3:1 or
above, with clearly identifiable, individual spikes and a maximum response of the recorded
neurons to the presented, auditory stimuli. In some of the experiments, the recording quality
of the neuronal signals remained stable even after an animal had fully recovered from
anaesthesia (Fig. 2.3b). The occurrence of only a few isolated spikes in response to each
stimulus suggests that, in these cases, a single neuron or a very small number of cells
contributed to the recorded neuronal activity. In other individuals, however, the recorded
!
31!
)
Fig. 2.3. Exemplary telemetric recordings from the inferior colliculus in two individuals (a, b) of
M. musculus that were freely exploring a makrolon cage (length x depth x height: 0.6 x 0.38 x
0.20 m). The upper track of each sub-figure shows the oscillogram of the respective neuronal activity,
whereas the lower track shows the oscillogram of the presented, auditory stimuli. The timescale is
identical for both sub-figures. a Multi-unit recording. b Single- to few-unit recording.
!
32!
spike activity drastically increased during recovery from anaesthesia and remained at a high
level thereafter, so that the recorded neuronal signals were clearly caused by multiple neurons
responding to the same stimulus. In some of the latter cases, this increase in spike activity was
accompanied by a decrease of the SNR (Fig. 2.3a).
There are two possible explanations for the observed changes in recording quality in
some of the experiments: (i) An increase in spike activity without a simultaneous decrease of
the SNR indicates that the electrode position relative to the recorded neurons remained stable
and that the increase in spike number was caused by the reactivation of neighbouring neurons
that had previously been silenced by effects of the anaesthesia. (ii) In those cases where the
increase in spike activity was accompanied by a decrease of the SNR, it is likely that the
recording loci had changed, too. Such shifts in recording position can either be caused by
respiration/blood-pressure induced changes in brain volume or by delayed relaxing
movements of the brain tissue after the electrode has been advanced (Jog et al. 2002).
The recordings from the hippocampus in freely behaving mice conducted at the Center
for Neurogenomics and Cognitive Research in Amsterdam (Netherlands) yielded comparable
results as the recordings from the inferior colliculus. Figure 2.4 exemplary illustrates
recordings from the hippocampus of an individual of M. musculus under anaesthesia
(isoflurane; Fig. 2.4a) and while exploring one side of an automated DualCage (length x
width: 0.3 x 0.24 m, Stiedel et al. 2010; Fig. 2.4b). During anaesthesia, spontaneous neuronal
activity from probably one or two hippocampal neurons could be recorded with high quality
in terms of SNR (Fig. 2.4a). After the animal had fully recovered from anaesthesia (approx.
three hours after surgery), the signal quality of the radio-transmitted recordings remained at a
high level (Fig. 2.4b), but the general neuronal activity in the target area significantly
increased. The identification of individual neurons using standard cluster-cutting algorithms
was not possible under these conditions. The increase of neuronal activity most likely was,
again, caused by the reactivation of neurons in the direct vicinity of the recording electrode
that had previously been silenced by effects of the anaesthesia. However, in the case of
hippocampal recordings, especially if aimed at the characterisation of the receptive fields of
place cells, this effect becomes eminently problematic for two reasons: (i) In terms of
neuroanatomy, the neurons in the hippocampus are characterised by small cell bodies, that are
!
33!
Fig. 2.4. Exemplary telemetric recordings from the hippocampus of an individual of M. musculus. a
Telemetric recording of neuronal activity from the hippocampus during anaesthesia. b Telemetric
recording of neuronal activity from the same individual while awake and freely exploring an operant
chamber (same recording/electrode position as in a).
!
34!
more densely packed than in many other (for example cortical) brain regions (compare
Fig. 2.5). This implies that, at a given hippocampal recording site, a high number of
individual neurons can potentially contribute to the recorded signal. (ii) In addition, the
designated target neurons in the hippocampus (i.e. place cells) show a special form of
discharge pattern that has been termed “complex spike” (Ranck 1973) or “complex burst”.
These complex bursts are characterised by a rapid sequence of discharges (with inter-pulse
intervals of 1.5 - 6 ms) with constantly decreasing spike amplitudes. The complex spike
characteristic of place cells, therefore, impairs the discriminative power of spike amplitude as
a parameter for single-unit identification through cluster-cutting analyses.
The experiments conducted in P. discolor showed that the recording setup as
described above is not suited for recordings in freely behaving (i.e. flying) microbats. The
quality of the recordings from the auditory cortex in anaesthetised animals was comparable to
that from the inferior colliculus in mice. When recordings were performed in individuals that
freely explored a flight tunnel (length x width x height: 6 x 1 x 1 m) one day after surgery,
however, an increase in noise level occurred that did not allow for a discrimination of the
neuronal activity from the background. Since the transmission of the radio signal remained
stable during these experiments, even though the animals rapidly changed their position
relative to the receiving antenna during flight, it obviously were muscle artefacts that
continuously disturbed the neuronal recordings. It is reasonable to assume that the flightmuscles, the muscles that control the movement of the outer ears, and the above-mentioned
temporalis muscles all have contributed to the high background-noise level. In freely
behaving mice, in contrast, disruptive muscle artefacts only occurred when the animals were
chewing on food pellets, but not during locomotion.
In summary, the here presented results show that the telemetric devices in combination
with the self-manufactured microelectrodes readily allow for multi-unit recordings from
different brain areas in freely behaving mice. It is likely that single-unit recordings can also
routinely be achieved in brain areas that feature a low density of target cells in combination
with a low general activity of the surrounding neurons. The hippocampus, as the brain area
where place cells have first been discovered, however, has the opposite characteristics (i.e.
densely packed, small neuronal cell bodies). One of the features of place cells, their complex
discharge patterns, additionally complicates the matter of single-unit isolation in this area.
!
35!
Fig 2.5. Photomicrographs from the brain of M. musculus (coronal sections, Nissl staining). Source:
brainmaps.org. As compared to neurons in cortical areas (e.g. primary somatosensory cortex; left) or
in the inferior colliculus (centre), the cell bodies of hippocampal neurons (here: cornu ammonis
field 1, CA1; right) are relatively small and more densely packed. The dashed black lines encircle the
region of CA1 where place cells can potentially be found.
!
36!
Thus, the development of miniaturised multi-channel transmitters that can be combined with
lightweight microdrives that allow for post-surgical manipulation of the multitrode locus is
the next, necessary step towards the characterisation of neuronal receptive fields in allocentric
representations of three-dimensional space (compare chapter 5). Multitrode recordings do not
only allow for better spike sorting/cluster cutting, but also improved algorithms for the
elimination of muscle-generated noise can be applied on multitrode data (Jog et al. 2002). The
disruptive effects that such muscle artefacts had on the recordings I conducted in freely
behaving individuals of P. discolor might, thus, also be eliminable using multitrode
technology.
!
37!
Acoustic Landmarks in Human Wayfinding:
the Role of Gender and Sightedness
Authors:
Daniel Schmidtke1,2
Sarah Galinski1
PD Dr. Karl-Heinz Esser1,2
Authors’ affiliations:
1
DS, SG and K-HE conceived and designed the experiments, DS
and SG performed the experiments, DS analysed the data, K-HE
contributed materials and analysis tools, DS wrote the
manuscript.
Journal:
(status: under revision)!
Institute of Zoology, University of Veterinary Medicine,
Hannover, Germany
2
Center for Systems Neuroscience, University of Veterinary
Medicine, Hannover, Germany
!
38!
Chapter 3: Acoustic Landmarks in Human Wayfinding: the Role
of Gender and Sightedness
!
Abstract
!
The research reported here examined the role of sound-emitting objects as acoustic landmarks
in a medium-scale, indoor wayfinding task in completely blind, visually impaired, and fully
sighted human subjects (all wearing blindfolds to exclude any visual input). The participants
first had to learn to find their way through a walkable maze in which three sound-emitting,
acoustic landmarks indicated important waypoints en-route from a fixed starting position to
the goal (walking distance approx. 28.5 m). During the subsequent, critical experiment, we
modified the acoustic landmarks in order to provoke a behavioural reaction to these
modifications in those subjects, who used the landmark information during wayfinding. The
goal of the study was to investigate to which degree humans make use of sound-emitting,
acoustic objects during navigation in a non-visual environment and to test for differences
between the two groups visually impaired/blind and sighted as well as between male and
female participants. The results demonstrated that, during the initial training phase, gender
and sightedness had a significant effect on the number of test trials the subjects completed
before they felt safe enough to proceed to the critical phase of the experiment, with the
sighted and female subjects needing more of these test trials than their respective counterpart.
During the critical experiment, a substantial amount of individuals from all groups responded
to at least one of the landmark modifications. On this individual level, we did not find any
significant differences in the number of responding subjects between the male and the female
group, but between sighted and visually impaired/blind subjects, with significantly less of the
latter reacting to at least one of the landmark modifications. We also used mixed linear
models and linear-hypothesis testing to compare the responses to the landmark modifications
on the group level. The results from these models mainly reflect the outcomes of the
comparisons of the individual reactions to the landmark modifications.
In a third, post-experimental phase, each subject had to model the “ideal path” through
the maze using plasticine. Top views of the plasticine models were digitised and analysed in
order to estimate, in how far the subjects were able to establish a Euclidean inner
!
39!
representation of the maze. Here, we found significant differences between visually
impaired/blind and fully sighted subjects (the sighted subjects produced more accurate
models), but no significant differences between the sexes. The group of the sighted subjects
additionally answered the German version of the Sensation Seeking Scale - Form V (SSS-V).
The individual SSS-V scores were used to check for relations between training performance
and the internal mediators parameterised by the scale. Significant correlations were found
between training performance and the disinhibition sub-scale of the SSS-V.
The results will be discussed in the context of the current literature on both gender
differences in mammalian navigation and spatial cognition in the visually impaired and blind.
Introduction
!
One of the most important and fundamental tasks in the everyday life of most animals,
including humans, is to orientate themselves in their environment in order to find food, to
rest, or to socially interact with other species members. This includes the exploration of
unknown territory as well as the repeated visiting of known places. To successfully do the
latter, an individual has to navigate, which can be defined as “determining and maintaining a
course or trajectory from place to place” (Gallistel 1989). In mammals, several behavioural
strategies exist for navigation that can be classified on the basis of hierarchical navigation
competences (Franz and Mallot 2000, Trullier et al. 1997). In this classification system, there
are two main groups of navigation behaviours, being referred to as local and wayfinding
behaviours. Local navigation behaviours include search, direction-following, aiming, and
guidance, with an increasing complexity of the problem that can be solved by the respective
behaviour from search to guidance. Wayfinding navigation behaviours include recognitiontriggered response, topological and survey navigation (Franz and Mallot 2000).
Under normal conditions, the sensory system that humans predominantly use to
acquire spatial information is the visual system, which seems to be better suited for this
purpose than other modalities, as it provides almost simultaneous perception of a large spatial
field including both near and far space information (Foulke and Hatlen 1992, Strelow 1985,
Thinus-Blanc and Gaunet 1997, Ungar 2000). However, there are situations or medical
conditions in which visual information is severely reduced or completely absent. Under these
!
40!
conditions, other sensory modalities have to be used to gather spatial information in order to
compensate for the loss of the visual input. In the visually impaired and blind, for example,
haptics and hearing can become valuable compensatory sources for spatial information. The
former is suitable to provide an individual with information about the near or manipulatory
space, i.e. small-scale spatial information that concerns the immediate, reachable surrounding
of an individual (Ungar 2000). The perceptual range of the haptic senses can to some degree
be artificially extended through the utilisation of canes. The auditory sense, on the other hand,
allows for the acquisition of near- as well as medium-scale far-space information.
The study presented here investigates human navigation on the level of wayfinding
behaviours, focusing on the role that sound-emitting, acoustic landmarks play in a mediumscale wayfinding task in visually impaired and blind as well as in fully sighted but blindfolded
human subjects. A landmark can be defined as any prominent feature of an environment that
directly or indirectly indicates the position of an important waypoint or goal, thereby being a
relevant source of spatial information for wayfinding, but also for some local navigation
behaviours (Franz and Mallot 2000; for a more detailed definition of landmarks see
“Discussion”, this chapter). The aim of the study is twofold:
Firstly, it sets out to determine, to which degree blind and visually impaired persons
rely on landmark information from sound-emitting objects during middle-scale wayfinding
when haptic and visual information is additionally reduced and whether there are differences
in this degree of acoustic landmark utilisation between visually impaired/blind subjects and a
fully sighted but blindfolded control group. It has previously been shown that, under certain
conditions, blind subjects have improved sound-localisation abilities as compared to sighted
controls (e.g. Ashmead et al. 1998, Lessard et al. 1998, Röder et al. 1999). One could,
therefore, expect the former to rely more on stationary, sound emitting landmarks during
wayfinding, as they should be able to extract more accurate spatial information from these
cues than sighted subjects. However, blind humans have also been found to process and use
echo cues more efficiently than sighted subjects, leading to higher abilities in echoacoustic
object detection and localisation (Dufour et al. 2005, Kellogg 1962, Rice 1969, Rice et al.
1965) as well as navigation (Schenkman 1986, Strelow and Brabin 1982). With these
additional acoustic tools at hand, the utilisation of sound emitting objects as landmarks for
!
41!
navigation might be of reduced importance for the visually impaired and blind in mediumscale wayfinding tasks.
Secondly, the present study investigates possible gender differences in the utilisation
of acoustic landmarks in humans. Previous studies in several polygamous mammalian species
have shown that male and female individuals differ in the relative use of two kinds of spatial
cues during visually-guided navigation, with female individuals preferentially using visual
landmark information, whereas male individuals preferentially use so-called Euclidean cues
(e.g. Chai and Jacobs 2010, Sandstrom et al. 1998). In a recently published study, two of us
presented experimental data indicating that the same is true for the laryngeally echolocating
bat Phyllostomus discolor when acoustic (in this case echo-reflective surfaces) instead of
visual landmarks are provided (Schmidtke and Esser 2011). From these findings, we
hypothesised further that the phenomenon of a female preference for landmark-based
navigation is independent of the sensory system that is used to acquire the landmark
information and that a female preference for the utilisation of acoustic landmarks, therefore,
should also be present in other mammalian species, including humans.
Materials and Methods
!
Ethics statement
!
The research protocol was approved by the ethics committee of the Hannover Medical School
(MHH). Prior to participating in the study, each test person had to sign a declaration of
consent. In those cases where a participant was under age, the declaration had to be signed by
at least one of the parents or legal guardians. For participating in the study, the subjects
received an invitation to a barbecue at which they later learned about the goal of the study
and its key findings. Between successfully completed phases of the study itself, all
participants were provided with beverages and snacks.
!
!
!
!
!
42!
Test subjects
!
For our study, we compared the spatial performance of two groups of human subjects, one
consisting of visually impaired subjects with low residual vision (n = 14) and fully blind
individuals (n = 8), the other one consisting of fully sighted individuals. Within the two
groups, the genders were about equally distributed, so that we could further test for genderspecific differences in the utilisation of acoustic landmarks.
The group that consisted of visually impaired and blind subjects comprised a total of
22 individuals (nfemale = 11; nmale = 11) that were recruited from a local, state blind school
(Landesbildungszentrum für Blinde Hannover; Blindenverband Niedersachsen), teaching
visually impaired and blind children as well as adults, the latter for vocational education. The
medium age of the subjects was 25 years with the youngest subject having been 17 years old
and the oldest having been 55 years old. Despite their handicap, all of the subjects from this
group showed a high degree of independence and mobility in everyday life and reached the
site of the experiments alone, using public transportation. For readability, the individuals from
this group will from now on be referred to as the visually impaired. The sighted group
comprised a total of 25 fully sighted individuals (nfemale = 13; nmale = 12) with a medium age
of 23 years and individual ages reaching from 19 to 28 years. All subjects were preselected
for a normal IQ and did not have any history of neurological, cognitive or sensory-motor
deficits other than the visual impairment in the according group.
Experimental environment
!
The experiments took place in a maze that consisted of three uniformly illuminated,
interconnected rooms (room 1 - 3; Fig. 3.1) that were located at the otherwise unused top
floor of a building on the campus of the University of Veterinary Medicine Hannover
(Germany). Within these rooms, tilted tables provided additional boundaries, so that the walls
of the rooms did not necessarily corresponded to the border of the maze. The starting point of
the maze was located in room 1, the goal was located two rooms ahead in room 3. At the goal,
a small bell was installed that the subjects had to ring as soon as they reached the end of the
maze. Along the way from the starting point to the goal, we installed a set of four
!
43!
Fig. 3.1. Schematic drawing of the maze (top view). A, B, B', and C indicate the positions of the
loudspeakers from which the three acoustic stimuli (landmarks) were played back. The orange line
corresponds to the “ideal path” along which the subjects have initially been walked by the
experimenter (phase 1) and which they later had to model using plasticine (phase 3). The chessboard
patterns span the areas on which the movement of the subjects was reconstructed using frame-to-frame
video analysis (compare Fig. 3.3).
!
44!
loudspeakers, corresponding to three different acoustic landmarks (A, B, B', C, Fig. 3.1) that
indicated important waypoints en-route to the goal. The “ideal path” (Fig. 3.1; see also below:
“Experimental procedure”) through the maze had a length of 28.5 meters.
Acoustic landmarks
!
As an orientational aid during wayfinding, we provided three acoustic landmarks (A - C,
Fig.!3.1; Fig. 3.2) that differed from each other in the object they represented as well as in
their spatial information content (see “Discussion”, this chapter). Landmark A was the sound
of a ticking clock and was installed above the exit of room 1. It, therefore, directly marked the
narrowing of the room that led over to room 2. Landmark B played back the continuous sound
of an aquarium with external filter and was attached to the (in movement direction) left wall
of room 2 at a height of 1.6 meters. It approximately indicated the y-coordinate at which the
subjects had to take a 90° turn to the right to reach the exit of room 2. A second, identical
loudspeaker was installed at the same wall and height but 1.25 meters in front of landmark B
(i.e. closer to the entrance of room 2). This speaker (landmark B') was only active during
those trials, in which landmark B was modified (see below: “Experimental procedure”).
Landmark C was installed in room 3 at a height of 2 meters and behind the distal border of the
maze. This loudspeaker continuously emitted the sound of a rotating fan. Since landmark C
was located in movement direction and out of reach for the subjects, its spatially dependent
sound-pressure level (SPL) had to be used as a distance information in order to find the ycoordinate at which a 90° turn to the left was necessary to reach the goal.
The emitted sounds were all recorded in an anechoic chamber using the corresponding
real-life objects and a condenser microphone (4191, Brüel & Kjær; 3 Hz - 40 kHz) that was
connected via a pre-amplifier (2660, Brüel & Kjær) to a measuring amplifier (2636, Brüel &
Kjær). To further improve signal quality, the signals were band-pass filtered (high-pass:
65 Hz, low-pass: 48 kHz; Benchmaster VBF8, Kemo) and subsequently digitised with a
sampling rate of 96 kHz (Audacity 1.3.6 under OsX). The spectral composition of the sounds,
as compared to synthetic white noise, is shown in figure 3.2.
!
45!
Fig. 3.2. Time-varying spectral composition of the acoustic landmarks. Spectrograms (frequency vs.
time vs. intensity) of the used acoustic stimuli (“clock”, “aquarium”, “fan”), as compared to synthetic
white noise (lower right). Dynamic range: 0 (black) to -70 dB (white).
!
46!
Videography
In order to monitor the subjects' behaviour within the maze and to quantify their performance,
we installed three cameras (C500, Logitech) that constantly recorded the movement of the
subjects in important areas of the maze. Using a webcasting software (Wirecast 3.5.8,
Telestream Inc.), the information from these cameras was combined to a single video stream
(MPEG-4, 1280 x 720 pixel, 12 fps, 3500 kbit/s) and stored on a hard disk for off-line
analysis. To facilitate this analysis, we also added semi-transparent auxiliary grids to the
video images. These grids were spatially transformed to match the perspective of the
individual cameras and spanned different areas of the respective rooms (grid sizeroom 1:
2 x 4.4 m; grid sizeroom 2: 3.4 x 4 m; grid sizeroom 3: 3 x 4.20 m; compare Fig. 3.1). Each cell of
the grids had an area of 0.2 x 0.2 meters.
Experimental procedure
!
Before an individual subject was allowed to enter the experimental environment, the acoustic
landmarks were activated and a short information sheet was read to the subject by the
experimenter. Afterwards, each subject (sighted and visually impaired/blind individuals alike)
was instructed to close its eyes and to put on blackened glasses that were additionally rubbersealed at the rim, to assure that no visual information could reach the eyes. After the
experimenter had controlled the correct positioning of the blindfold, the subject was allowed
to enter room 1 of the maze and was led to the starting position (compare Fig. 3.1). The
following experiment was then divided into three phases, an initial training phase, a
subsequent critical phase, and a post-experimental phase.
At the beginning of the training phase, each individual was walked once through the
whole maze by the experimenter and along the aforementioned “ideal path” (Fig. 3.1).
Afterwards, each subject was allowed to explore the maze by itself and to walk through the
maze as often as desired. When an individual subject felt trained enough to proceed, the
second, critical phase started.
During the critical/experimental phase, each subject was asked to find its way through
the maze for 27 times and to complete each trial without pauses and without touching the
boundary of the maze. The respective subject had to start as soon as the experimenter gave the
!
47!
appropriate signal and the trial ended when the subject rang the bell at the goal of the maze. In
three arbitrarily interspersed trials per subject and landmark, we then modified one of the
acoustic landmarks in order to provoke a reaction in those subjects that relied on the acoustic
landmarks during wayfinding (for the modification scheme see Table 3.1). The modifications
of the individual acoustic landmarks were done by switching off landmark A, by switching
off landmark B while landmark B' was coincidentally activated, or by lowering the sound
pressure level of landmark C by 15 dB SPL.
In the post-experimental phase, all subjects were asked to model the “ideal path”,
along which they had initially been walked by the experimenter, using plasticine. The purpose
of these models was to assess in how far the subjects were able to establish a map-like mental
representation of the way through the maze during the first two phases of the study. An
individual session (training, experiment, and plasticine modelling) lasted between 1 and 2
hours, depending on the performance of the respective subject.
Personality measures
!
After producing the plasticine models, the sighted subjects additionally were asked to answer
the German version of the Sensation Seeking Scale, Form V (SSS-V; Beauducel et al. 1999,
Beauducel et al. 2003, Zuckerman et al. 1978), so that we could later test, whether possible
individual differences in the training performance relate to individual differences in sensation
seeking or to some of its sub-dimensions (for details see “Discussion”, this chapter). The
SSS-V is a self-report instrument consisting of 40 forced-choice items, 10 for each of the four
sub-scales the operational measure of the general factor of sensation seeking is composed of
(thrill and adventure seeking – TAS, experience seeking – ES, disinhibition – DIS, boredom
susceptibility – BS). The German version of the scale used in the present study was developed
by Beauducel and colleagues (1999) and is a close translation of the English original
(Zuckerman et al. 1994). Due to the low internal consistencies of the SSS-V (Beauducel et al.
1999), Beauducel and co-workers later (2003) provided B-weights for the calculation of
factor scores that have also been used here. Since, on the one hand, personal identification
was required, but, on the other hand, the nature of the demanded answers to the questions of
!
48!
Table 3.1. Scheme for the pseudo-randomisation of the landmark modifications.
Trial #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Modified landmark
A
C
B
B
-
Trial #
15
16
17
18
19
20
21
22
23
24
25
26
27
Modified landmark
C
A
B
C
A
!
49!
the SSS-V is highly confidential, we used code numbers instead of the subjects' real names on
the questionnaires. The counting of individual test scores could, thus, be performed by a
person unaware of the subjects' identities.
Data acquisition
!
To compare the spatial performances between the groups (visually impaired vs. sighted and
male vs. female), we measured sets of quantitative, temporal, and spatial data for each subject
(Table 3.2). The spatial and temporal data were acquired by frame-to-frame analyses of the
recorded videos. In order to reconstruct the paths the individuals took through the maze, the
video files of each trial were exported as stacks of Jpeg-files consisting of the exact frames in
which an individual entered or left one of the grids as well as sequences of pictures from all
movements that took place on the grids (1 fps). Within these exported image stacks, we could
then measure the pixel coordinates (x/y) of the subjects at a given point in time using ImageJ
(ImageJ 1.43, Wayne Rasband, National Institute of Health) for the frame-to-frame video
analyses. As a reference point for the respective position of a subject, we used the projection
of the mass centre of a given subject to the floor. Since the pictures from the cameras were
perspectively distorted, we corrected this distortion using the respective 3x3 transformation
matrices and the following equations:
Xcorrected =
(a * X pixel + b *Ypixel + c)
(g * X pixel + h *Ypixel + i)
(3.1)
Ycorrected =
(d * X pixel + e *Ypixel + f )
(g * X pixel + h *Ypixel + i)
(3.2)
with
a b
d e
c
f
g h
i
being the transformation matrix in its general form.
!
50!
Table 3.2. Listing of the parameters used to evaluate individual maze-performances during the first
two phases of the study.
!
!
Phase
Quantitative
parameters
Temporal
parameters
Average trial
duration
Duration of the first
critical trial
Time required to
pass room 1
Spatial
parameters
Calculated spatial
parameters
1
# of test trials
-
-
2
-
-
-
2
-
2
-
-
Path taken on grid 2
2
-
-
“Direction” taken on
grid 1
y-coordinate of the “right
turn” on grid 2
y-coordinate of the “left
turn” on grid 3
!
51!
The received data then allowed for the reconstruction of the undistorted individual paths taken
on each grid (Fig. 3.3) as well as for the calculation of the temporal parameters from the
critical phase (temporal resolution: 0.08 s).
Data preparation
While the quantitative and temporal parameters could directly be compared between the
groups and experimental conditions, we had to find and calculate suitable benchmarks to
compare the spatial parameters. For room 1, we used the maximum absolute value of the ratio
of the shift in x-direction to the shift in y-direction over 5 successive sample points as a
benchmark representing the directness with which the exit of room 1 was approached
(Fig. 3.3):
“direction” taken on grid 1
= Maxi=3,...,n−2 ( xi−2 − xi+2 : yi−2 − yi+2 )
(3.3)
For rooms 2 and 3, we calculated the y-coordinate at which the subjects turned to the right or
left, respectively, as the y-value at which the corresponding change in x-direction over five
sample points was maximal in relation to the associated absolute value of the change in
y-direction (Fig. 3.3):
y-coordinate of the “right-turn” on grid 2
= Maxi=3,...,n−2 ((xi+2 − xi−2 ) : yi−2 − yi+2 )
(3.4)
y-coordinate of the “left-turn” on grid 3
= Maxi=3,...,n−2 ((xi−2 − xi+2 ) : yi−2 − yi+2 )
(3.5)
Statistics
!
Missing Data. Due to initial technical problems with the video surveillance in room 3, parts
of the respective spatial data from the first four subjects (2 male and 2 female subjects, all
visually impaired) are missing. We, therefore, excluded these subjects from the statistical
analysis of the spatial parameters of room 3. The temporal data was not affected. Of the
!
52!
Fig. 3.3. Depiction of the reconstructed walking paths (n = 27) of three subjects (one for each of the
rooms 1 to 3). Straight lines represent the benchmarks that have been calculated in order to
parameterise the spatial behaviour. The darker the shading of the lines/paths, the higher the trial
number. Trials under the condition of a modified landmark in the respective room are colour-coded in
red. The subjects have exemplarily been chosen as they were statistically identified as landmark users
in the corresponding room. It can be seen that the manipulation of the landmarks had the expected
effect on the walking behaviour of the respective subjects (compare below).
!
53!
25 questionnaires (SSS-V), five have been answered incompletely and, thus, have not been
used for further analysis.
One-tailed permutation tests (individual level). To find out, which of the subjects responded
to the different landmark modifications, we compared the individual, calculated spatial
parameters (from grids 1 - 3, Table 3.2) as well as the individual times required to pass room
1 under the condition of an unmodified landmark with those under the respective modified
condition. We used one-tailed permutation tests, since we had the following, reasoned
expectations concerning the effects that the different manipulations should have had on
subjects that used the acoustic landmarks as orientational aids (compare Fig. 3.3): (A) The
modification of landmark A conformed to the total loss of an acoustic signal that directly
indicated the position of the exit of room 1. The manipulation should, therefore, have led to
an increase in the time that was needed to cross the first room and/or to a less direct approach
towards the exit of room 1. (B) When landmark B was modified, it was virtually shifted 1.25
meters closer to the entrance of room 2. Since this landmark indicated the y-coordinate at
which the subjects had to turn to the right in order to reach the exit of room 2, this “rightturn” should have occurred earlier (i.e. at a lower y-value) under the modified condition. (C)
For the modification of landmark C, the opposite effect was expected. The decrease in the
SPL of the landmark corresponded to a greater distance of the landmark from the entrance of
room 3 as compared to the unmodified condition. Since this landmark was provided as an
orientational aid that allowed to estimate the y-coordinate at which it was necessary to turn to
the left to reach the goal of the maze, this “left-turn” should have occurred later (i.e. at a
higher y-value) under the modified condition.
Fisher's exact tests. The results of the permutation tests (see above) were used to compare the
numbers of subjects that responded to the individual landmark modifications between the
different subgroups (male vs. female; fully sighted vs. visually impaired) in 2x2 contingency
tables. Since the expected values in some of the tables we wished to analyse were relatively
small (< 5), the assumptions that Pearson's χ2 test makes have not always been met.
Therefore, we uniformly used the more conservative, two-tailed Fisher's exact tests to check
for significant differences between the aforementioned groups (α = 0.05).
!
54!
Linear models and hierarchical linear mixed models. In order to test for possible effects of
age, visual impairment/sightedness, and gender on the number of needed test trials, the
average trial-duration during the training phase, and the duration of the first trial of the
experimental phase (phase 2), we modelled the dependent variables as a function of the
combinations of the aforementioned predictors. The proposed models (Wilkinson-Rogers
notation: number_of_test_trials ~ age + visual_impairment + gender; average_trial_duration ~
age + visual_impairment + gender; first_trial_duration ~ age + visual_impairment + gender)
were fit and examined using the lm function in R (R Development Core Team 2010).
To double-check the results of the permutation tests and the subsequent Fisher's exact
tests on the group level, we fitted linear mixed models (Pinheiro and Bates 2000, Pinheiro
et al. 2009) including the following terms into the fixed-effects parts of the model formulas:
trial_number (1-27, since we expected a learning effect over time), age (age of the individual
subjects in years), visual_impairment (dichotomous variable: fully sighted/visually impaired),
gender (dichotomous variable: male/female), and manipulation (dichotomous variable:
respective landmark modified/not modified). In addition, we modelled interactions between
gender and manipulation and visual_impairment and manipulation, respectively, as it was of
interest whether subjects from the different groups react differently to the individual landmark
modifications (general form of the argument in Wilkinson-Rogers notation: response_variable
~ trial_number + age + gender*manipulation + visual_impairment*manipulation). In the
random-effects parts of the models, we specified that random intercepts and slopes for the
manipulation effects are to be predicted for each individual subject (random = ~ 1 +
manipulation | subject_id). Assuming that the residuals associated with observations next to
each other in a sequence of longitudinal observations, for a given subject, are not independent
of each other, we used a first-order autoregressive covariance structure for the Ri matrices of
our models (form = ~ trial_number | subject_id). To estimate the fixed-effect parameters, β,
and the covariance parameters, θ, we applied the method of maximum likelihood estimation.
Since the simultaneous testing of several null-hypotheses increases the risk of
erroneous rejections of single hypotheses at a given alpha-level, we used a special test
environment (the glht function in R; Hothorn et al. 2008) that allows for multiple comparisons
in linear mixed effect models as it characterises each individual null-hypothesis by a linear
!
55!
combination of all model-parameters. All tests of significance were performed at an adjusted
alpha-level of 0.05 (two-tailed, single-step method). A summary of the linear hypotheses that
have been tested in the linear models and in the hierarchical linear mixed models is given in
Table 3.3a and Table 3.3b, respectively.
Euclidean bi-dimensional regression and Mann-Whitney-U tests. To analyse the plasticine
models built by the subjects during the post-experimental phase of the study, we first used
Euclidean bi-dimensional regression to assess in how far the plasticine models resemble the
“ideal path” along which the subjects were led through the maze by the experimenter during
the learning phase (compare Fig. 3.1). Euclidean bi-dimensional regression allows for a
comparison of configurations of coordinates in a two-dimensional space by postulating a
regression-like relationship between the pairs of coordinates (Friedman and Kohler 2003,
Tobler 1994). The analysis was done using R and the formulas provided by Friedman and
Kohler (2003). The resulting distortion indices (DI, accounting for the proportion of
unexplained variance within the data; Friedman and Kohler 2003, Waterman and Gordon
1984) were used as a measure for the quality of the plasticine models, which could then be
compared between the groups (visually impaired vs. fully sighted and male vs. female) using
Mann-Whitney-U tests (two-tailed, α = 0.05).
Results
!
Comparison of the training performances between the groups
During the learning phase, male subjects needed a lower (linear model, general linear
hypothesis, two-tailed, α = 0.05, estimated β = 2.40399, adjusted p = 0.01186) number of
training trials (median = Q2 = 5, lower hinge = Q1 = 4, upper hinge = Q3 = 6.5, inter-quartilerange = IQR = 2.5; n = 23, one outlier: 13) as compared to the female subjects (Q2 = 7,
Q1!= 5, Q3 = 10, IQR = 5, n = 24, no outlier; Fig. 3.4a). A comparable result was found
between sighted (Q2 = 7, Q1 = 5, Q3 = 9, IQR = 4, n = 25; no outlier) and visually impaired
subjects (Q2 = 4, Q1 = 3, Q3 = 7, IQR = 4, n = 22, no outlier; Fig. 3.4b), with the latter
!
56!
Table 3.3. Hypothesis matrices for the linear-hypothesis tests (compare Table 3.4).
a Linear hypothesis (I – III)
Intercept
Age
Visual impairment
Gender
b Linear hypothesis (I – X)
Intercept
Trial #
Age
Gender
Landmark manipulation (LM)
Visual impairment
Gender:LM
Visual impairment:LM
I
0
1
0
0
0
0
0
0
II
0
0
1
0
0
0
0
0
III
0
0
0
1
0
0
0
0
IV
0
0
0
0
0
1
0
0
I
0
1
0
0
V
0
0
0
0
1
0
.5
0
II
0
0
1
0
III
0
0
0
1
VI
0
0
0
0
1
0
.5
1
VII
0
0
0
0
0
0
0
1
VIII
0
0
0
0
1
0
0
.5
IX
0
0
0
0
1
0
1
.5
X
0
0
0
0
0
0
1
0
!
57!
Fig. 3.4. Comparisons of the individual number of training trials needed to learn the way through the
maze during phase 1 (left), the averaged individual trial duration in phase 1 (centre), and the duration
of the very first trial in phase 2 (right) between the sub-groups (a: male vs. female; b: visually
impaired vs. fully sighted) using General Linear Models (two-tailed, adjusted p-values: ● p < 0.1,
* p < 0.05, ** p < 0.01).
!
58!
needing less training trials (linear model, general linear hypothesis, two-tailed, α = 0.05,
estimated β = -3.09188, adjusted p = 0.00125). No significant effects were found when the
averaged training-trial duration was compared between the subgroups (male: Q2 = 3 min,
Q1 = 2.57 min, Q3 = 3.63 min, IQR = 1.06 min, n = 23, two outliers: 6.67 min & 5.44 min;
female: Q2 = 2.63 min, Q1 = 2.21 min, Q3 = 3.07 min, IQR = 0.86 min, n = 24, one outlier:
5.00 min; male vs. female: estimated β = -0.379, adjusted p = 0.394; Fig. 3.4a; sighted:
Q2!=!2.63 min, Q1 = 2.33 min, Q3 = 3.00 min, IQR = 0.67 min, n = 25, one outlier: 5.44 min;
visually impaired: Q2 = 3.00 min, Q1 = 2.25 min, Q3 = 3.75 min, IQR = 1.5 min, n = 22, one
outlier: 6.67 min; visually impaired vs. sighted: estimated β = 0.586, adjusted p = 0.0998;
Fig. 3.4b). The same is true for the duration of the first experimental trial in the second phase
of the study (male: Q2 = 66.75 s, Q1 = 58.46 s, Q3 = 102.08 s, IQR = 43.62 s, n = 23, no
outlier; female: Q2 = 72.00 s, Q1 = 59.96 s, Q3 = 91.83 s, IQR = 31.87 s, n = 24, two outliers:
160.8 s & 356.6 s; male vs. female: estimated β = 11.953, adjusted p = 0.789; Fig. 3.4a;
sighted: Q2 = 71.17 s, Q1 = 60.83 s, Q3 = 93.5 s, IQR = 32.67 s, n = 25, one outlier: 356.6 s;
visually impaired: Q2 = 70.96 s, Q1 = 57.5 s, Q3 = 100.17 s, IQR = 42.67 s, n = 22, no outlier;
visually impaired vs. sighted: estimated β = 4.764, adjusted p = 0.983; Fig. 3.4b).
Individual reactions to the modifications of the acoustic landmarks
The results of the one-tailed permutation tests (α = 0.05) revealed that a total of 18 subjects
showed a response to the modification of landmark A (nmale = 9, nfemale = 9; nvisually impaired = 6,
nsighted = 12) by needing significantly more time to cross room 1, or by approaching the exit of
room 1 significantly less directly in the trials where landmark A was switched off as
compared to the unmodified condition, or both. In room 2, 18 subjects reacted to the
modification of landmark B (nmale = 8, nfemale = 10; nvisually impaired = 6, nsighted = 12) by turning
significantly earlier to the right in the trials where landmark B' was active than in the trials
where landmark B was active. In room 3, only 3 subjects reacted to the modification of
landmark C (nmale = 3, nfemale = 0; nvisually impaired = 2, nsighted = 1) by turning significantly later
to the left in the trials where the SPL of landmark C was reduced as compared to the trials
under the unmodified condition (Fig. 3.5).
!
59!
Fig. 3.5. Utilisation of the acoustic landmarks. A, B, C Comparison of the number of subjects [%] that
responded to the modification of the respective landmark (compare Fig. 3.3) between the sub-groups
(a: male vs. female; b: visually impaired vs. fully sighted) using Fisher's Exact Tests (two-tailed:
● p = 0.1004). A - C Comparison of the number of subjects [%] that responded to at least one of the
landmark manipulations between the sub-groups (a: male vs. female; b: visually impaired vs. fully
sighted) using Fisher's Exact Tests (two-tailed: * p < 0.05).
!
60!
In total, 14 male and 13 female subjects responded to at least one of the landmark
modifications, 18 of them being sighted and 9 of them being visually impaired or blind
(Fig. 3.5). Comparing the number of responding subjects between the subgroups male and
female using Fisher's exact tests (two-tailed, α = 0.05) shows that there are no significant
differences between men and women in the number of subjects that responded to the
modification of landmark A (p = 1) and B (p = 0.7661). The same is true if the total number
of subjects that responded to at least one of the landmark modifications is compared between
the sexes (p = 0.7702). In room 3, only male subjects responded to the modification of
landmark C, but still the difference between the groups is not significant at the 0.05 alphalevel (p = 0.1004; Fig. 3.5a).
Comparing the number of responding subjects between the groups visually impaired
and sighted using Fisher's exact tests reveals that, for the individual landmarks A, B, and C,
there are no significant differences (pA = 0.2293; pB = 0.2293; pC = 0.5624). However,
comparing the total number of responding subjects shows that significantly more sighted than
visually impaired or blind subjects responded to at least one of the landmark modifications
(p = 0.0419; Fig. 3.5b).
Group-level reactions to the modifications of the acoustic landmarks
The results from the group-level analysis of the data (mixed linear models) basically reflect
the results from the Fisher's-Exact-Tests described above. A summary of the p-values for the
linear hypotheses that the interaction between the particular landmark modifications and the
affiliation to one of the sub-groups (visually impaired, fully sighted, male, female; compare
Table 3.3b) has a significant effect on the considered parameters is given in Table 3.4. For the
modification of landmark C, no significant effects of the modification could be found in any
of the groups. These results correspond to the low number (3 out of 47) of subjects that
showed a reaction to the modification of this particular landmark. When comparing the effects
of the landmark modifications between the visually impaired and the fully sighted subjects, it
is the latter in which the modification of the landmarks has significant effects on more of the
measured variables. For the gender comparison, no differences in the number of effects of the
!
61!
Table 3.4. Summary of the results (p-values) of the linear-hypothesis tests highlighted in Table 3.3b
(V, VI, VIII, IX).
!
Dependent
variable
log(time required to
pass room 1)
log(benchmark 1)
log(benchmark 2)
log(benchmark 3)
!
Fully sighted
Visually impaired
Male
Female
p < 0.001 ***
p = 0.057 •
p < 0.001 ***
p = 0.042 *
p < 0.001 ***
p < 0.001 ***
p = 0.937
p = 0.817
p = 0.012 *
p=1
p = 0.014 *
p < 0.001 ***
p = 0.499
p = 0.027 *
p < 0.001 ***
p = 0.999
!
62!
landmark modifications on the measured variables could be found at the alpha-level of 0.05
(Table 3.4).
Comparison of the plasticine models
Of the 47 plasticine models built by the subjects in the post-experimental phase, a total of 9
models could not be analysed using Euclidean bi-dimensional regression, as they included
errors in the number and/or direction of turns (number of erroneous models: nmale = 2,
nfemale = 7; nvisually impaired = 7, nfully sighted = 2). A comparison of the DIs of the remaining 38
models between the subgroups revealed that the fully sighted subjects produced significantly
more accurate plasticine models than the visually impaired (M-W-U Test, two-tailed,
α = 0.05, p = 0.0055; Fig. 3.6b). Significant gender differences in model quality could not be
found (M-W-U Test, two-tailed, α = 0.05, p = 0.8487; Fig. 3.6a).
Bivariate correlations
!
Table 3.5 shows the bivariate correlations (r-values) between the sighted subjects' individual
SSS-V scores (unweighted and factor scores) and their training performances. Most of the
correlations are only weak (< 0.4 or > -0.4) and statistically negligible. Two moderate (> 0.4
< 0.7 or > -0.7 < -0.4) and significant (alpha = 0.05) correlations have been found between the
individual training performances and the disinhibition sub-scale scores in the female subpopulation (non-factored: p = 0.0498; factored: p = 0.0364) and in the whole population, i.e.
if the data of both sexes are pooled (non-factored: p = 0.0038; factored: p = 0.0036). In both
cases, the variables are negatively correlated. In the male sub-population, the correlation
between the training performance and DIS also is negative and moderate, but does not reach
statistical significance (non-factored: p = 0.2534; factored: p = 0.2789).
!
63!
Fig. 3.6. Plasticine-model quality. Comparison of the number of erroneous plasticine models (Fisher's
Exact Tests, two-tailed: ● p < 0.1) and model quality (benchmark: distortion index, DI; the higher the
DI, the lower the model quality; Mann-Whitney-U Tests, two-tailed: ** p < 0.01) between the
sub-groups (a: male vs. female; b: visually impaired vs. fully sighted).
!
64!
Table 3.5. Results of the correlation analyses (r-values) between the individual training performances
(number of training trials) and the individual SSS-V scores (main scale and sub-scales, factored and
non-factored).
Sample
SSS-V dimension
Both genders (n = 20)
Total
TAS
DIS
ES
BS
Total
TAS
DIS
ES
BS
Total
TAS
DIS
ES
BS
Women (n = 11)
Men (n = 9)
Significance code: * p < 0.05; ** p < 0.01.
r-values (non-factored
SSS-V scores)
- 0.33
- 0.18
** - 0.62
0.12
- 0.12
- 0.34
- 0.24
* - 0.60
0.07
0.16
- 0.10
0.24
- 0.43
- 0.31
- 0.12
r-values (factored
SSS-V scores)
- 0.35
- 0.10
** - 0.62
0.25
0.34
- 0.32
- 0.27
* - 0.63
0.39
0.43
- 0.22
0.19
- 0.41
- 0.25
0.10
!
65!
Discussion
!
Landmark qualities I
Depending on the scientific context in which the term “landmark” is used (ranging from the
field of geography over psychology to robotics), the definitions of what a landmark is may
differ in their details. In addition, different types of landmarks can be distinguished according
to their position within an environment and their information content for navigation. We,
therefore, start our discussion with a definition of how the term “landmark” is used in this
study and with assigning the acoustic objects from our experiments to the appropriate subcategories.
The urban planner K. A. Lynch (1960) defines landmarks as external references that
are unique and memorable within the environment on which its characterisation as a
landmark, therefore, depends. A distinguishable, environmental feature like that can be of
different sensory nature (i.e. visible, audible, touchable, etc.) or multi-modal and plays a role
in spatial decision-making. In addition, Lynch distinguishes between local and global
landmarks. Local landmarks can only be perceived in restricted localities, from short
distances, and sometimes only from certain approaches. They often are associated with routerelevant decision-making (Steck and Mallot 2000). Global landmarks, in contrast, can be
perceived from many angles and distances (Lynch 1960). As long as only short distances are
travelled, they provide stable, world-centred directional information or global reference
frames to an observer (Steck and Mallot 2000). Objects can serve the purpose of both local
and global landmarks, depending on the position of the observer relative to the object. There
is some discordance in the literature on whether a landmark mandatorily is stationary (e.g.
Werner et al. 1997) or if slowly moving objects, such as the sun, can also be considered as
being global landmarks (e.g. Lynch 1960).
All acoustic objects used in the study presented here were stationary. According to the
definitions given above, Landmark A and B are best described as local landmarks: Landmark
A directly indicated the passage from room 1 to room 2, above which it was installed, while
landmark B indicated the y-coordinate in room 2 at which the subjects had to turn right in
order to reach the passage from room 2 to room 3. The quality of landmark C, however, is
!
66!
less easily defined. It was installed behind the boundary of the maze, i.e. outside the
explorable space of the subjects, and its exact position could only roughly be assessed. In
addition, knowledge of the location of landmark C alone did not provide any valuable
information concerning the route to the goal. Instead, it was the information embedded in the
intensity gradient of the emitted sound, which means an acoustic distance estimate (see
“Landmark qualities II”, below), that additionally had to be used to identify the y-coordinate
at which it was necessary to turn to the left in order to find the goal. Thus, while sharing many
characteristics with a local landmark as defined above, landmark C also had some
characteristics of a more directional, gradient cue as described by Jacobs and Schenk (2003),
who in principle labelled all acoustic cues as gradient cues (but see below).
Landmark qualities II
Acoustic landmarks can additionally be distinguished according to their physical properties
(amplitude, temporal characteristics and spectral composition) and origin (natural vs.
synthetic sounds). Under the conditions of an indoor (highly reverberant) environment,
different acoustic cues are used for sound-source localisation in the far-field: Azimuth and
elevation of a sound source are determined from interaural time and intensity differences or
spectral cues, respectively (for review see Middlebrooks and Green 1991). For acoustic
distance perception, the level of the emitted signal and the ratio of the energies of direct and
reflected sound seem to be the principle cues (e.g. Bronkhorst and Houtgast 1999, Mershon
and Bowers 1979, Mershon and King 1975). While the latter generally provides absolute
distance information, the direct sound level alone can only provide absolute information if the
emitted sound and its level are familiar to the listener (Coleman 1962, Mershon and Bowers
1979, Mershon and King 1975, Gardner 1969). Hence, we only used natural sounds of
everyday objects (clock, aquarium, and fan) that were played back at authentic sound pressure
levels in order to facilitate the utilisation of the sounds as acoustic landmarks. In addition, the
sounds we used as potential landmarks can all be characterised as broadband signals
(Fig. 3.2), which contain more information about the position of a sound source than
narrowband signals (Blauert 1996).
!
67!
Gender differences
!
From the findings of our recent study on the laryngeally echolocating bat Phyllostomus
discolor (Schmidtke and Esser 2011), in which we described sex-specific differences in
echoacoustic orientation that closely resembled the gender differences found in other
mammalian species in visually-guided orientation, we hypothesised that sex differences in
spatial cognition are independent of the sensory modality that provides the spatial
information. In the following paragraphs, we will discuss our findings (present study) in the
light of this hypothesis as well as in the light of the current literature on sex differences in
spatial orientation. The discussion will be divided according to the three phases of the study,
since each of them covered a different aspect of spatial cognition.
Phase 1 - Sex differences in spatial learning. Within the literature on sex differences in spatial
learning, several studies can be found that report differences in the speed of spatial learning
between male and female individuals in a variety of mammalian species, with male
individuals typically acquiring spatial knowledge faster than female conspecifics (e.g. rats:
Forcano et al. 2009; laryngeally echolocating bats: Schmidtke and Esser 2011; humans: Galea
and Kimura 1993, Moffat et al. 1998, Schmitzer-Torbert 2007). A similar result was found in
the present study, with women needing more training trials than men to learn their way
through the maze before proceeding to the second phase of the study. While our findings,
therefore, are consistent with the above-mentioned studies, a noteworthy and distinctive
feature of the training procedure in the study at hand is that the subjects decided by
themselves when the training goal was achieved. The only decision-guidelines given to them
were that they should end the training phase and proceed to the second phase as soon as they
feel able to solve the maze without interruption and/or touching the boundary of the maze (i.e.
without using a thigmotaxic strategy to find their way through the maze). At first glance, this
approach might seem to be highly subjective. However, a comparison of the individual times
needed to finish the first experimental trial during phase II shows that both groups (males and
females) entered the main experiment at similar levels of maze performance, so that the lower
number of training trials in men in combination with the fact that the average training-trial
durations did not differ between the sexes, indeed, indicates sex differences in the speed of
spatial learning under the given experimental conditions.
!
68!
The idea of a general male superiority at spatial learning and ability has long time
been considered as being evident in both the psychological, human literature (Harris 1978,
Linn and Peterson 1986, Maccoby and Jacklin 1974, McGee 1979) as well as comparative
studies in non-primate species (Beatty 1979, Gaulin and Fitzgerald 1986, Williams et al.
1990, Williams and Meck 1991). It has been substantiated through reports of faster learning
rates (as they were found here) or higher accuracy scores in male individuals as compared to
females. The more recent literature on gender differences in mammalian spatial abilities,
however, has come to the conclusion that the supposed advantage of male individuals over
female conspecifics, on the one hand, is dependent on how the spatial performance is
measured (e.g. Forcano et al. 2009). On the other hand, it highly depends on specific task
parameters such as the quality of the cues provided for solving the spatial tasks (e.g. Barkley
and Gabriel 2007, Roof and Stein 1999, Rodríguez et al. 2010). It has been shown that, if
either one or both are appropriately altered, mammalian sex differences in spatial learning and
ability can disappear or even reverse (e.g. Barkley and Gabriel 2007, Forcano et al. 2009,
Roof and Stein 1999, Schmitzer-Torbert 2007).
The emerging consensus view, therefore, is that sex differences in mammalian spatial
learning and ability as well as in orientation do not reflect a general disadvantage of female
individuals in solving spatial tasks, but are the result of sex-specific preferences for the
utilisation of different sets of cues: A male superiority in spatial navigation and route-finding
ability can robustly be found in tasks favouring the utilisation of Euclidean and geometric
cues (cardinal directions, exact distances, room geometry) for task solving (humans: Barkley
and Gabriel 2007, Sandstrom et al. 1998, Saucier et al. 2002; rodents: Rodríguez et al. 2010,
Roof and Stein 1999), whereas in tasks that favour the utilisation of configurational landmark
and egocentric directional information, females perform better or at least just as good as
males (humans: Levy et al. 2005, O'Laughlin and Brubaker 1998, Saucier et al. 2002,
Silverman and Eals 1992; rodents: Roof and Stein 1999). This does not imply that males and
females exclusively rely on either the one or the other set of cues. Instead, both sexes have
been shown to be able to successfully use their respectively less preferred source of
information, if necessary (e.g. Barkley and Gabriel 2007, Rodríguez et al. 2010), and it is
reasonable to assume that both sexes use all types of available cues when navigating in
complex, open environments (Chai and Jacobs 2009).
!
69!
The fact that, in the present study, we found a male advantage in spatial learning in a
maze-solving task being designated for testing the role of acoustic, local landmarks in human
wayfinding, might seem to contradict the above given statements. Indeed, the acoustic objects
were installed in order to provide the subjects with local, topographical information that helps
them in route-relevant decision-making in a way, local landmarks do. However, two features
of the experimental design of the study have to be taken into consideration when deciding
upon this matter: (i) Apart from the acoustic objects, the subjects had several further sources
of spatial information available to learn their way through the maze, all of them being of a
more geometric or Euclidean nature and, therefore, potentially advantageous for male
subjects. Among these sources of spatial information was the possibility to count steps in
order to get estimates of the distances that have been walked. In addition, subjects could use
path integration, which is based on dynamic sensory flow (under the given experimental
conditions mainly proprioceptive, tactile, and acoustic) and the efference copy of the intended
action. Path integration provides direction and distance information relative to a starting
point. A possible third source of spatial information (in this case on room geometry), socalled human echolocation or the extended concept of spatial hearing, will be introduced in
detail in the second part of the discussion. (ii) The acoustic objects initially might have
provided more directional instead of positional information. In a natural, feature-rich scene,
the position of objects that are perceived visually can literally be determined “on first sight”,
since humans are reasonably accurate in estimating the distance, the direction, and the size of
objects based on visual cues. Crucial information for these estimations is in part provided by
monocular, static cues (such as relative visual-field height, relative and familiar size, and
occlusion) and grounded on assumptions and knowledge about the world, gathered through
extensive past experience (Dijkerman and Milner 1998, Gregory 1997, Gregory 2005, Milner
1997). The human knowledge base for the acoustic world, however, is less rich in substance.
Therefore, the acoustic perception of the location of novel objects, especially of their distance
based on volume information, is less straightforward (Coleman 1962, Gardner 1969, Mershon
and King 1975). Even though we tried to facilitate the utilisation of the presented sounds as
local landmarks by playing back recordings of everyday objects (compare above: “Landmark
qualities II”), it is likely that, during the very first training trials, the acoustic objects mainly
!
70!
served as gradient cues that polarised the environment and, thereby, provided directional
information (Jacobs and Schenk 2003).
Some authors justifiably have argued that, if sex differences in the acquisition of
spatial knowledge are in part caused by gender-specific differences in the preference for
environmental strategies, not only external, but also internal mediators are likely to influence
spatial learning (e.g. Schmitz 1999). Candidate concepts are spatial anxiety and self-reported
spatial competences, on the one hand, and curiosity, novelty seeking, sensation seeking, risk
taking, and the like, on the other hand. Here, we explored the relationship between training
performance (individual number of training trials) and sensation seeking (SSS-V; Beauducel
et al. 1999, Zuckerman 1994) in the sighted subjects (due to the item quality, the scale is not
suited for the quantification of sensation seeking in visually impaired subjects). In its most
recent form, the construct “sensation seeking is a trait defined by the seeking of varied, novel,
complex, and intense sensations and experiences, and the willingness to take physical, social,
legal, and financial risks for the sake of such experience” (Zuckerman 1994). Sensation
seeking differs from curiosity and novelty concepts in a less pronounced cognitive component
(Zuckerman 1979) and an emphasis on danger and risk taking, with which sensation seeking
is highly correlated (though not identical; Horvath and Zuckerman 1993).
Since the subjects in our study could decide by themselves when to end the training
phase, we considered it possible that the parameterisation of the training performance through
the number of needed training trials would correlate negatively with boredom susceptibility
and/or thrill and adventure seeking (two of the subscales of the SSS-V) for the following
reasons: (i) Each subject knew that it would have to walk through the maze for 27 times
during the experimental phase. Thus, the decision of when to end the training phase and to
start the experimental phase was prone to be influenced by the subject's individual boredom
susceptibility. (ii) Starting the experimental phase early without a correct internalisation of the
maze and its dimensions would have led to an increased risk to collide with the boundary of
the maze under the condition of visual deprivation. Therefore, one could expect those subjects
that completed only a low number of training trials to show higher scores in thrill and
adventure seeking, which has been found to be highly related to a lack of fear of physical
danger (Zuckerman, 1978).
!
71!
However, the analyses of correlation between individual SSS-V scores and training
performances showed that the internal mediators considered here did not have a major
influence on the individual number of training trials, since none of the expected negative
correlations was significant. Instead, we found significant, moderate, and negative
correlations between the number of training trials and the disinhibition sub-scale for both the
whole group of subjects that answered the questionnaires as well as the belonging female subpopulation. Zuckerman et al. (1978) defined disinhibition as “the desire for social and sexual
disinhibition as expressed in social drinking, partying, and variety in sexual partners”.
Accordingly, recent studies found positive correlations between disinhibition and
alcohol/substance abuse (e.g. Hittner and Swickert 2006), pathological gambling (e.g. Fortune
and Goodie 2010), and risky sexual and online behaviour (e.g. Manner et al. 2011). The
theoretical link between disinhibition and the training performance of our subjects, however,
is less obvious and can only be speculated upon. For example, disinhibition was shown to be
the best predictor for safety seeking in Norwegian school children (Thuen et al. 1992). Safety
seeking includes behaviours that are directed towards a reduction of the risk of mortality or
injuries. It is negatively correlated with risk seeking, but, instead of being the mere opposite
of risk seeking, seems to constitute a separate dimension (Thuen et al. 1992). Unfortunately, it
has not been investigated yet, if disinhibition has the same predictive power for safety seeking
in age groups other than adolescents. If so, this could explain the negative relation between
training performance and disinhibition found here, using a line of arguments comparable to
the one justifying our expectation for a negative relation between thrill and adventure seeking
and training performance (vide supra, (ii) ). The only difference would be that it is not the
lack of fear of physical danger, but low safety seeking that might have influenced the subjects'
decisions about when to start the experimental phase.
Phase 2 – Sex differences in the utilisation of acoustic landmarks. The hypothesis that males
and females to different degrees make use of two distinct sets of cues during navigation has
recently gained further support through fMRI studies reporting sex-specific patterns of brain
activation during virtual navigation in humans (Grön et al. 2000, Nowak et al. 2011) and
through behavioural studies showing that the modification of Euclidean and geometric cues in
a learned/familiar task has a more negative impact on male performance, whereas the
!
72!
modification of local landmarks mainly affects female performance (e.g. Chai and Jacobs
2010, Sandstrom et al. 1998). If the female preference for the utilisation of local landmarks as
navigational aids also applies for acoustically gathered landmark information, as hypothesised
(Schmidtke and Esser 2011), the modifications to the acoustic landmarks in the present study
should have led to more significant responses in the female than in the male subjects.
What the results of our study show, is that subjects from both sexes used at least some
of the acoustic objects for route-relevant decision-making. If they did so, the modification of
the respective landmarks led to reasonably predictable, directed reactions. However, no
significant differences between the sexes were found in the number of subjects that responded
to the individual landmark modifications. A noteworthy exception is landmark C. Only male
subjects responded to the modification of this landmark, so that, here, a weak statistical trend
(p = 0.1004) towards an effect of sex on cue utilisation exists. Due to the special quality of
landmark C as compared to the other two landmarks (see above: “Landmark qualities I”), its
utilisation as an orientational aid required the interpretation of the volume gradient emanating
from the sound source in order to estimate the distance between landmark and perceiver.
Based on current theories on sex differences in spatial cognition (e.g. Jacobs and Schenk
2003), such cues should preferentially be used by males.
When it comes to human depth-perception, it has been shown for the visual domain (in
a virtual reality application) that, even though both sexes systematically underestimate
distances in depth, men have significantly higher accuracy scores than women (Armbrüster
et al. 2005). Comparable results have also been found in an acoustic study on distance
perception of looming (but not receding) sound sources that approach the listener from the
side (Neuhoff et al. 2009). However, since Neuhoff and colleagues used virtual sound sources
that approached the subjects at relatively high velocities (between 15 and 25 ms-1), it is not
clear in how far a comparable sex difference in the accuracy of acoustic depth-perception
could have been expected for the study at hand, since, here, the decrease in distance between
listener and sound source occurred much slower and was caused by active movements of the
subject instead of by a moving sound source.
On the whole, the finding that only male individuals responded to the modification of
landmark C is consistent with the literature. However, the small overall percentage of subjects
who responded to the modification of landmark C (6.98%) as compared to those that reacted
!
73!
when landmark A or B were modified (in both cases 38.3%) indicates that the utilisation of
this cue must have been very unattractive to both sexes. It is possible that, under the given
experimental conditions, task difficulty obliterated more obvious sex differences in the
utilisation of landmark C. For example, the relatively small distance from the entrance of
room three to the goal might have encouraged the members of both sexes to apply an
idiothetic strategy instead of orientating based on the acoustic cue.
The considerations given so far, however, do not explain why the theoretically
expected gender differences in the utilisation of landmark A and B have not been found.
Williams and Meck (1993) stated that the gender differences in (visio-) spatial ability that
have been described in the literature are small as compared to other sexually dimorphic
behaviours and, consequently, are difficult to detect. Besides, both genders readily use their
less preferred source of information when other cues are absent or when task complexity
increases (Barkley and Gabriel 2007, Chai and Jacobs 2009, Rodríguez et al. 2010) and we
think that the same might be the case under situations of sensory deprivation (present study).
Hence, if subjects lose their most important source of spatial information, as in the case of
visual deprivation, it is likely that they use any additional spatial information that is available
in order to compensate for the loss of visual input. This, in turn, could then mask the genderspecific selectiveness for different cue types as it occurs under non-deprived conditions.
Phase 3 – Sex differences in spatial representation. Given that men and women prefer
different sets of spatial cues and consequently show differences in the information content
that is available for the encoding aspect of spatial-memory formation, it has to be assumed
that spatial representation, as the resulting product of spatial-memory formation, also differs
between the sexes. Recent models on the contribution of the medial temporal lobe to spatialmemory formation, the “model of neurocognitive mapping” (Roche et al. 2005) and the
“parallel map theory” (Jacobs and Schenk 2003), suggest that the formation of a spatial
representation is achieved by complementary information-gathering processes, each of them
using a different set of spatial cues. These processes, taking place during traveling, lead to
mental models of the world (Roche et al. 2005: egocentric, functional, and allocentric “spatial
constructs”; Jacobs and Schenk 2003: “bearing map” and “sketch map”) that are finally
combined in the mental representation of the environment (Roche et al. 2005: “neurocognitive
!
74!
map”; Jacobs and Schenk 2003: “integrated map”), thereby allowing for successful
navigation. While the “model of neurocognitive mapping” is purely anthropological and does
not explicitly account for gender differences in spatial representation, the “parallel map
theory” is based on a broader comparative and evolutionary approach that also takes the
literature on sex differences in mammalian spatial cognition into consideration. The authors
suggest that males and females rely to different extents on the bearing map (based on selfmovement and directional cues) and on the sketch map (based on unique local landmarks)
during orientation, with the males relying more heavily on the former and the females relying
more heavily on the latter.
The quality of the plasticine models of the “ideal path” through the maze the subjects
had to build in our study was parameterised by Euclidean bi-dimensional regression. In order
to achieve a high quality score (low distortion index), the subjects had to correctly reproduce
the positions of necessary right and left turns within the allocentric framework of the maze as
well as the relative distances between these turns. In the light of the “parallel map theory”,
male subjects should, therefore, have been at an advantage during this part of the study, since
the most valuable information for solving this modelling task can be derived from the bearing
map and its contribution to the integrated map (see also Iachini et al. 2005). What we found
was, indeed, that the bulk of the erroneous plasticine models (77.8%) was produced by female
subjects, but due to the relatively low overall occurrence of these erroneous models (9 out of
47), the difference in their amount between the sexes is not statistically significant
(p = 0.1365). We consider it possible that, with a bigger sample size, a small but significant
effect could be detected here.
Differences between visually impaired and sighted subjects
!
As stated in the introduction (this chapter), blind subjects have been shown to possess
advanced sound localisation capabilities, when single sound sources in the horizontal plane
(especially in the periphery) have to be localised (Ashmead et al. 1998, Doucet et al. 2005,
Fieger et al. 2006, Lessard et al. 1998, Röder et al. 1999). One of the current views is that
cross-modal plasticity and sharpening of the remaining senses can compensate or even
hypercompensate for the loss of vision in both the early- and the late-blind (e.g. Fieger et al.
!
75!
2006, Kujala et al. 1997). This idea is substantiated by several electrophysiological and
neuroimaging studies, in which an increased, acoustically evoked activity has been found in
the occipital and parietal cortex of blind subjects with improved spatial hearing abilities
(Gougoux et al. 2005, Kujala et al. 1992, Kujala et al. 1997, Leclerc et al. 2000, Weeks et al.
2000). However, the described hyper-compensation does not extend to the localisation of
frontal sound sources or to the vertical plane. Instead, when localising sound sources in the
vertical plane, early-blind subjects have been shown to perform worse than sighted controls
(Lewald 2002, Zwiers et al. 2001). It seems, that the spectral pinna cues that are used for
acoustic elevation perception (Middlebrooks and Green 1991; compare “Landmark qualities
II”) need to be visually calibrated in order to provide proper spatial information (Lewald
2002, Zwiers et al. 2001).
The subjects in our study were allowed to move freely within the maze (i.e. to change
their position/orientation relative to the sound sources) and most of the relevant information
of the acoustic landmarks presented here was linked to their position in the horizontal plane.
One could, therefore, have expected that the visually impaired subjects would more
intensively make use of the landmarks than the controls, as the visually impaired subjects
should be able to extract more precise spatial information from these cues than the latter. For
example, a study by Després and colleagues (2005) demonstrated improved auditory selflocalisation abilities in early-blind subjects in both reduced, unechoic and complex,
reverberant acoustic environments. While most of the above-mentioned studies on hypercompensation and cross-modal compensation focused on the occurrence of these phenomena
in congenitally or early-blind subjects, the study at hand investigated the utilisation of sound
cues for orientation in a group of visually impaired subjects that was much more
heterogeneous, as it comprised totally blind subjects with both early and late blindness-onset
as well as visually impaired subjects with some residual vision. This comparative approach
(as opposed to a differential approach; see Warren 1994) was taken deliberately, as we were
interested in the question if and to which amount the whole population of the visually
impaired uses acoustic landmarks for medium-scale wayfinding. Consequently, the subjects
were chosen to provide a good cross-section of this population, allowing for a broad
generalisation of the results. We would like to note that focusing on totally blind subjects
exclusively, would not necessarily have reduced heterogeneity of the sample, since many
!
76!
different medical conditions can lead to total blindness, many of them affecting different parts
of the visual pathway, i.e. different regions of the nervous system. Thus, subject-selection for
studies in blind and visually impaired individuals always is a trade-off between homogeneity
of the sample and generalisability of the results. However, this means that we have to be
cautious when making assumptions based on results from more differential studies, as done
above. For example, visually impaired subjects with residual peripheral vision have been
shown to localise horizontal sound sources with less precision than totally blind and sighted
subjects (Lessard et al. 1998). Therefore, the postulation of a more intense utilisation of
acoustic landmarks among visually impaired subjects as compared to the sighted control
might have been precipitately, as we will discuss based on our actual findings. Like before
and in accordance with the three phases of the study, the following discussion will be divided
into three parts.
Phase 1 - Differences in spatial learning. Concerning the training phase, we made the
straightforward and judicious assumption that the visually impaired and the fully sighted
subjects would differ in their number of needed training trials, since the former were more
familiar with navigating under the condition of visual deprivation. We, therefore, expected the
visually impaired to need significantly less training trials than the group of the sighted
subjects, be it due to faster maze learning under the given conditions or to less timidity when
locomoting in the absence of visual information. What we found was, indeed, a highly
significant effect of visual impairment on the number of training trials, with the visually
impaired subjects on average needing about three training trials less than the sighted subjects.
Different from what we found when comparing the average trial durations between the sexes,
where there have been no differences, there was at least a statistical trend towards longer
average trial durations in visually impaired subjects (p < 0.1). The durations of the very first
trial in phase 2, however, did not differ between the groups. This might indicate that the
sighted subjects completed further training trials even though they had already reached a
performance level comparable to that at which the visually impaired subjects immediately
decided to enter the next phase. The significant effect of the visual impairment on the number
of the training trials, therefore, does not necessarily mean that the visual impaired learned
their way through the maze faster than the fully sighted subjects. Instead, it is very likely that
!
77!
at least part of the difference between the groups is due to a higher level of insecurity in the
sighted subjects under the given, non-visual conditions.
Phase 2 – Differences in the utilisation of acoustic landmarks. Our results concerning the
individual reactions to the landmark modifications during the second phase undoubtedly show
that our postulation of an increased utilisation of the acoustic landmarks for orientation in the
visually impaired as compared to the fully sighted subjects does not hold. Substantially more
than half of the visually impaired subjects (59.1%) did not show a significant response to any
of the landmark modifications (A, B, or C). In contrast, only about one quarter (28.0%) of the
fully sighted subjects did not make use of the provided landmark information, while 72% of
them reacted to at least one of the three modifications. This difference between the two
groups is statistically significant and indicates that it is, indeed, the fully sighted and not the
visually impaired subjects that make more intense use of the acoustic objects during maze
navigation. It would be reasonable to explain this finding with the heterogeneity of our
sample (the visually impaired) and the aforementioned, inferior sound localisation abilities of
visually impaired subjects with residual sightedness (Lessard et al. 1998). However, our data
shows that the ratio of visually impaired subjects with residual vision and fully blind subjects
among the non-responders is about the same as it is in the whole sample. Therefore, the
heterogeneity of the sample clearly does not explain the finding. We think that the higher
utilisation of acoustic landmarks for medium-scale wayfinding in the fully sighted, as it was
found in the study at hand, has a different reason: the experience-based advantage of the
visually impaired and blind individuals over sighted subjects in human echolocation and
spatial hearing tasks.
Here, human echolocation refers to hearing in which the information from sound
reverberations is used to gather information about the location and/or structure of objects,
including the self, as well as the environment containing them. The sounds used for
echolocation are actively generated by the echolocating individual, using either its vocal tract
or its extremities (e.g. long-cane tapping sounds or footfalls) for the production of brief clicks.
This active mode of echolocation resembles echolocation as it has been described in other,
non-primate mammalian species, such as laryngeally echolocating bats and flying foxes of the
genus Rousettus (Chiroptera) or toothed whales (Odontoceti), in which the perceiving listener
!
78!
has rich information about both the produced signal and the returning echo (e.g. Evans 1973,
Griffin 1958, Möhres and Kulzer 1956, Norris 1969). However, humans can also use spatial
information provided by the echoes of externally generated sounds, i.e. passive echolocation
(e.g. Rice 1967, Strelow and Brabyn 1982), or by fluctuations in the ambient sound field, like
“sound shadows”, “bright spots”, or fluctuations caused by open spaces, such as doorways
(e.g. Ashmead and Wall 1999, Simon et al. 2012). Since the latter mode of acoustic object
detection does not include the utilisation of discrete, reflected sounds, it is better referred to as
spatial hearing rather than echolocation (Ashmead and Wall 1999, Schenkman and Nilsson
2010).
It is assumed that both sighted and visually impaired humans can process information
provided by echolocation and spatial hearing and that this information is subliminally used by
all humans during everyday life (Schwitzgebel and Gordon 2000, Stoffregen and Pittenger
1995). While most people normally are not consciously aware of their ability to use this kind
of acoustic spatial information, many visually impaired and blind people actively use
echolocation and spatial hearing during wayfinding and their audio-spatial abilities are
generally considered superior as compared to those of the sighted (Schwitzgebel and Gordon
2000; recent studies supportive of this view: Dufour et al. 2005, Schenkman and Nilsson
2010, Schenkman and Nilsson 2011). However, it has also been shown in numerous studies
that sighted, blindfolded individuals effortlessly can learn to echolocate and partially reach
performance levels comparable to those of visually impaired and blind subjects (e.g. Hausfeld
et al. 1982, Rosenblum et al. 2000, Supa et al. 1944, Teng and Whitney 2011).
For our study, it should be legitimate to consider the group of the sighted subjects as
having been naïve in terms of conscious echolocation and spatial hearing. Hence, the group of
visually impaired and blind subjects should have been at an advantage when solving the maze
during the second phase of the study, since they had an additional orientational tool at hand,
as compared to the group of sighted subjects. This assumption could partially be substantiated
by observations of the experimenter and statements from the subjects during the debriefing.
For example, even though the maze was not under audio-surveillance, the experimenter
clearly heard that at least one of the blind subjects was frequently producing clicks with
his/her vocal tract while locomoting. Other blind and visually impaired subjects reported
during the debriefing session that they could hear or “feel” (Schwitzgebel and Gordon 2000,
!
79!
compare Supa et al. 1944) the open doorway leading from room 2 to room 3. Consequently,
they did not have to use landmark B as a spatial cue, but could use the open doorway itself as
a reference point for navigation. Another group of visually impaired and blind subjects
complained about “the fan” (landmark C) in room 3, as a source of disturbance. We think
that, due to its noisy character (see Fig. 3.2), landmark C might, indeed, have interfered with
echolocation and other forms of spatial hearing, by masking important acoustic information.
The fully sighted subjects made no comparable statements during debriefing. Instead, they
mainly named the utilisation of the acoustic objects/landmarks and counting steps as
strategies to solve the maze. We, therefore, think that the differences in the number of
“landmark users” between the two groups are due to the superior/better-trained echolocation
and spatial hearing abilities of the visually impaired and blind as compared to the blindfolded
sighted.
Phase 3 – Differences in spatial representation. The visually impaired subjects in our study
produced plasticine models of the path through the maze that were significantly less accurate
than those of the sighted group, so that the question arises, whether the spatial representations
that the subjects established during the first two phases of the experiments differed between
the groups. In fact, there is a long-standing and on-going debate in the scientific literature on
whether internal spatial representations formed by visually impaired individuals (especially
within the congenitally blind) are qualitatively equivalent to that of normally sighted
individuals or whether there are differences between these two groups in terms of
representation quality and/or accuracy. Important early theories that arose from a more
general discussion on the mental representation of knowledge are the “propositional theory”
(Pylyshyn 1973) and the “imagery theory” (Kosslyn and Pomerantz 1977). In the former, a
proposition, as a conceptual structure, represents knowledge (the relationship between
objects) in an abstract, factual (true/false) format, without encoding the actual physical
properties of the object that is represented. According to the imagery theory, sensory
information is stored in an organised, simplified format that, for the example of visually
gathered information, can be used as a template for the formation of pseudo-pictorial mental
images during memory retrieval.
!
80!
We would like to stress, that the discussion between Kosslyn, Pylyshyn, and others
was not led as an “either/or”-discussion. Kosslyn did not dispute the existence of an abstract,
propositional code and agreed with Pylyshyn (1973) that such a code exists (Kosslyn and
Pomerantz 1977). The debate was rather driven by the two questions whether propositional
representations and imagery are structurally and functionally alike (i.e. propositional) or
whether mental images have properties that cannot be derived directly from propositional
representations. Such distinctive properties could, for example, be a non-factuality (as in
mental imagery a truth value is assigned only when the representation is under a particular
description) and analogue coding. It should be easy to see, how the discussion relates to the
question whether spatial representations in the visually impaired differ from spatial
representations in fully sighted humans: if (i) imagery does exist as an distinctive form of
internal representation and (ii) has the properties suggested by Kosslyn and Pomerantz
(1977), then it is reasonable to assume that internal spatial representations in fully sighted
humans, whose predominant spatial sense is vision, are at least partially based on quasipictorial, spatial entities, bearing the advantage that geometric information can more readily
be deduced from them. Spatial representations in blind and/or visually impaired humans could
then intuitively be expected to be qualitatively different from those of fully sighted persons
and, for example, purely propositional.
During the last decades, evidence accumulated that visio-spatial information can
indeed be coded in analogue representations, i.e. representations that preserve some of the
physical properties of the world (e.g. mental scanning: Kosslyn 1973, Kosslyn et al. 1978,
Kosslyn 1980; mental rotation: Shepard and Cooper 1982, Shepard and Metzler 1971;
extrapolation of unmentioned physical attributes: Thompson et al. 2008). More recently,
several neuroimaging studies provided evidence that mental imagery and perception are
accompanied by the activation of partially overlapping networks within the corresponding
sensory cortices (Dechent et al. 2004, Goyal et al. 2006, Jeannerod et al. 1995, Kobayashi et
al. 2004, Kosslyn et al. 2001, Stevenson and Case 2005, Yoo et al. 2003, Yoo et al. 2001).
This indicates that the percept-like experience of vivid mental images is not, as suggested by
contestants of the imagery theory (e.g. Pylyshyn 1973, Pylyshyn 2003), epiphenomenal.
However, this does not necessarily confirm the speculation that vision is a prerequisite for
analogous, spatial representations of the world. Gardini and co-workers (2005, 2009)
!
81!
suggested that a differentiation has to be made between what they call general and specific
images. The former provide skeletal, basic, and prototypical information about objects and are
equated with spatial images by some authors (e.g. Cattaneo and Vecchi 2011, p. 73), whereas
the latter are detailed, visual representations of particular exemplars of objects (Gardini et al.
2009). Evidence for the validity of this classification was gathered from fMRI studies
indicating that the generation of each of these two classes of mental representation is
supported by a distinct neuronal pathway (Gardini et al. 2005, Gardini et al. 2009).
Numerous studies on mental imagery in the congenitally blind have shown that
imagery-like spatial representations in this group share many similarities to those of normally
sighted humans and that sensory information can, indeed, be represented in an analogical
format in the blind (e.g. Afonso et al. 2010, Aleman et al. 2001, Arditi et al. 1988, Bértolo
et al. 2003, Carreiras and Codina 1992, Cattaneo et al. 2010, Kaski 2002, Kerr 1983, Klatzky
et al. 1990, Thinus-Blanc and Gaunet 1997, Tinti et al. 2006, Vanlierde and Wanet-Defalque
2004, Vecchi et al. 2004,). The currently emerging, consensus view, therefore, is that spatial
representations are supramodal (for review see Struiksma et al. 2009). This means, that
different input channels, depending on their availability, can contribute to the generation of
spatial representations. While retaining a modality-specific component on the one hand, these
spatial images, on the other hand, exceed the input from different sensory channels (Barsalou
1999, Pietrini et al. 2004) by activating brain areas that process spatial information
independently of the input modality (Struiksma et al. 2009), allowing the blind to generate
spatial representations in the lack of vision by using compensatory spatial information from
other sensory channels. Spatial images of the visually impaired/blind and sighted people,
accordingly, would only differ in the “degree to which specific visual knowledge and
processes are involved” in representation generation (Cattaneo and Vecchi 2011), but not in
quality.
So why is it then, that the visually impaired subjects in our study performed
significantly worse than the sighted but blindfolded group when building a plasticine model
of the way through the maze from their memory? It is reasonable to assume that the
acquisition of spatial information through non-visual channels (haptic, audition,
locomotion/proprioception, or semantics) is more demanding than through vision: While
vision allows for the simultaneous perception of large sets of objects, tactile, auditory, and
!
82!
locomotory exploration of an environment is mainly sequential in nature. Accordingly, blind
individuals have been shown to preferentially use egocentric, route-like representations for
navigation (e.g. Corazzini et al. 2010, Noordzij et al. 2006), while normally sighted people
more routinely also make use of allocentric, survey-like representations. In order to build and
validate the plasticine models, however, the subjects in our study had to create a mental
survey representation of the route they walked through the maze, maintain it in memory, and
compare it with their perceptual experience of the plasticine model they formed. The
significant difference in model quality between the two groups of our study, thus, might either
be due to the utilisation of different preferred navigation strategies between the visually
impaired and sighted subjects (route vs. survey navigation, respectively) or to a disadvantage
of the visually impaired group in model validation, i.e. matching the perceptual information
from the plasticine model with its mental representation.
We do not think that the latter possibility alone can satisfactorily explain the
differences we found. Of course, one could argue that the sighted subjects were at an
advantage as they could use vision for the validation of their models. One possible advantage
vision as a perceptive tool for model validation bears is the difference in spatial acuity
between vision (approximately 1 min of arc or 0.15 mm at a distance of 0.5 m between object
and eye; Westheimer 1965) and proprioception at the fingertips (approximately 1 - 1.5 mm;
Legge et al. 2008, Wong et al. 2011). However, compared to the size of the models the
subjects built (spanning areas of at least 10 x 15 cm), these differences in spatial resolution
are subtle and cannot account for the found differences in model quality. The second
advantage vision has over proprioception in this context is, again, that it allows for a
simultaneous perception of the whole model, whereas tactile exploration of the model has to
be done sequentially. For this reason, model validation in the visually impaired included at
least one additional transfer from route-like to survey-based representations. However, if the
step of matching the perceptive information from the built model with the inner representation
alone should be responsible for the found differences in model quality, one would expect a
negative correlation between the distortion index and the residual sightedness of our visually
impaired subjects. Such a correlation has not been found (r = 0.1284, p = 0.5691). We,
therefore, consider it more likely that it is, indeed, a difference in preferred navigation
strategies between the groups that explains the found divergence in modal accuracy: When
!
83!
exploring and solving the maze during the first two phases, all subjects of our study were
unaware of the fact that they would later have to reconstruct the path they walked.
Accordingly, it can be ruled out that they deliberately tried to generate a survey-like
representation. Instead, each subject should have used its naturally preferred navigation
strategy to solve the maze. Even though all subjects exclusively had to rely on sequentially
gathered spatial information (acoustic and proprioceptive, mainly idiothetic, gained through
locomotion), it has to be assumed that the sighted subjects more readily transferred this
sequential information into an allocentric, survey-like representation (Corazzini 2010).
Therefore, the visually impaired presumably had to perform more changes in reference frame
for both the generation of a mental image of the model and model validation, leading to
higher demands on information processing and working memory and, possibly, to the
decrease in model accuracy that has been found on the group level (compare also Rieser et al.
1980).
Sample composition. Many of the studies on spatial cognition in the visually impaired and
blind cited throughout the preceding discussion made a clear distinction between different
subpopulations within the visually impaired. For example, differences have been made and
found between congenitally, early, and late blind individuals and are assumed to be associated
with cortical plasticity phenomena (for an excellent and extensive review on the importance
of blindness onset see Cattaneo and Vecchi 2011). Another distinction that has often been
found to be necessary is that between fully blind and visually impaired subjects with low
residual vision. A corresponding example has already been mentioned earlier, which was the
localisation of horizontal sound sources, where subjects with low residual vision have been
found to be less precise than totally blind and sighted subjects (Lessard et al. 1998). A
comparable finding has been published by Lewald (2002), who described an impaired
performance in subjects with low residual vision in a vertical sound localisation task. In
general, low residual vision seems to be disadvantageous in situations where the distorted
visual information is in conflict with non-visual cues for the recalibration of the auditory
input (Cattaneo and Vecchi 2011).
In the study at hand, we pooled the data of our visually impaired subjects, resulting in
a sample that was composed of both visually impaired subjects with low residual vision
!
84!
(n = 14) as well as fully blind individuals (n = 8). As we mentioned earlier, this approach was
taken deliberately to achieve a higher generalisability of our results that allows for an
evaluation of current trends in the development of universal electronic travel and orientation
aids (vide infra). We decided that, for our study, the aforementioned differences between the
possible subpopulations of the visually impaired could mostly be disregarded for the
following reasons: (i) Navigational tasks, like the one set here, heavily rely on information
processing in the hippocampal formation (compare above: “Gender differences”). The
hippocampus proper and adjacent, functionally related brain structures receive highly
multimodal input and store spatial information in an abstract, supramodal (i.e. modality
invariant) format (Wiener 1996). Accordingly, animal studies in rodents have shown that the
hippocampal formation retains its functionality under conditions of artificial visual
deprivation (Hill and Best 1981) as well as early blindness (Save et al. 1998). In addition, the
hippocampal formation is known for its exceptional ability to generate neurons even during
adulthood (Eriksson 1998). It, therefore, has a high potential for adaptability and plasticity
throughout life and the onset of blindness should have a much lower impact on the course of
plasticity processes than it has in neocortical brain areas. In line with this speculation, a recent
study on the effect of blindness on structural changes in the human hippocampus revealed an
increased volume of the anterior portions of the right hippocampus in both early and late blind
subjects, as compared to sighted controls (Fortin et al. 2008). While similar structural changes
in the hippocampus of the blind have also been found in other studies (e.g. Lepore et al. 2009,
Chebat et al. 2007), the value of the work of Fortin and colleagues (2008) for our reasoning is
that they explicitly compared hippocampus size and performance in a navigational task
between early-blind, late-blind, and sighted subjects and that they found significant
differences between blind and sighted, but not between early- and late-blind individuals.
Hence, the onset of blindness seems to play a negligible role in navigational tasks and
blindness-related structural plasticity in the hippocampus and we assume that such structural
changes can also be found in the hippocampus of visually impaired individuals, as the
structures probably start to change as soon as vision looses its function as primary and reliable
source of spatial information. (ii) Many authors recently emphasised the role of individual
differences based on life history, degree of independence, and orientation and mobility
training (O&M) on spatial performance in the visually impaired and blind (e.g. Cattaneo and
!
85!
Vecchi 2011, Loomis et al. 1993). It has been discussed, whether the pronounced interindividual variability in the whole population of the visually impaired and blind (ThinusBlanc and Gaunet 1997) accounts for some of the performance differences between fully
sighted and blind individuals or between different subpopulations of the visually impaired
(e.g. Ungar 2000). For the present study, all visually impaired subjects (the fully blind and
those with low residual vision) were recruited from a state blind school at which O&M is an
integral part of the curriculum. Accordingly, all our subjects showed a high degree of
independence when travelling and reached the site of our study alone and via public
transportation. This, in addition to the considerations given under (i), might have further
increased the homogeneity of our sample.
On the development of electronic travel and orientation aids
!
An effort has been made during the last 50 years to develop electronic travel (ETAs) and
orientation (ETOs) aids in order to provide visually impaired and blind people with spatial
information for micro-navigation and navigational instructions that allow for independent
wayfinding, respectively. In their review article, Roentgen et al. (2008) classified 146 of these
electronic mobility aids, with only 21 of them being commercially available and the rest being
either prototypes or not on the market anymore. Those systems that are commercially
available are barely used, as they often do not fully meet the individual needs of the visually
impaired users, are imprecise in terms of global positioning (in the case of ETAs), and
insufficient in terms of output interaction (Kammoun et al. 2012). However, recent
developments in mobile technology, global positioning, and non-visual user interfaces offer
hope for the future of electronic mobility aids. Important advances have been made by
Loomis and co-workers, who introduced user-guidance interfaces based on virtual 3D sound
(e.g. Loomis et al. 1998). To improve the outdoor positioning performance of future mobility
aids, the same authors suggested the utilisation of differential GPS (Loomis et al. 1994;
Personal Guidance System), while others suggested the combination of GPS tracking with
dead-reckoning techniques (Helal et al. 2001; Drishti System) or with inertial motion tracking
and artificial vision (Kammoun et al. 2012; NAVIG System). What all recent, prototypical
electronic mobility aids have in common, is the delivery of environmental information via the
!
86!
haptic and/or the auditory channel (Fusiello 2002), with an observable trend towards spatially
augmented reality audio (SARA), where sounds from virtual sources are processed and
presented in a way that simulates spatial directionality (Martin 2009). A very promising and
conscientious project following this trend is the aforementioned NAVIG project (Kammoun
et al. 2012, Katz et al. 2012, Katz et al. 2010). The distinctiveness of this project lies in its
long-term user-centred design approach in collaboration with a panel of 21 visually impaired
consultants being involved in all the design steps of the NAVIG system. The major goal of
the project is nothing less than the development of a universal electronic mobility aid being
adaptable to individual needs and applicable for both indoor and outdoor navigation
(Kammoun et al. 2012).
In the light of our results, an important issue with the conveyance of auditory guidance
information and SARA can easily be identified: To begin with, every additionally provided
auditory information interferes with normal spatial hearing and can be highly disturbing for
individuals that rely on natural sound cues during navigation. It, consequently, does not come
as a surprise that the visually impaired consultants of the NAVIG project asked for the
provided auditory information to be “minimally intrusive” and “without excess” (Kammoun
et al. 2012). However, much more problematic than balancing out the minimal but sufficient
amount of augmented auditory information, is the task to find suitable means for delivering
this information. In order to spatialise virtual sound sources, there is no way around the
utilisation of stereo earpieces, which normally have to be inserted into the ear canal, thereby
not only interfering occasionally, but permanently with the user's normal free-field hearing.
Kammoun et al. (2012) suggested the utilisation of either bone-conduction headsets or
acoustically transparent/open earpieces to overcome this problem. A third alternative that can
be found in the literature are headphone-microphone combinations, through which the natural
acoustic environment is picked up and directly played back into the ear (for review see Härmä
et al. 2004). Since the early virtual 3D auditory displays were developed for the use with airconduction headphones, the same pertains to the signal-processing techniques that spatialise
the presented sounds. Therefore, the application of bone-conduction headsets for the delivery
of SARA is difficult at present and intense re-evaluation of the design-guidelines for 3D
auditory displays will be necessary, to allow their use with bone-conduction headphones
(Stanley 2006). Microphone-headphone combinations also have a decisive disadvantage.
!
87!
Their earpieces usually block the ear canal, leading to an amplification of low-frequency
sounds within the head, a phenomenon known as the “occlusion effect” (Martin et al. 2009).
Thus, acoustically transparent earpieces currently seem to be the best suited option for the
delivery of virtual 3D auditory information.
In a psychoacoustic study, Martin et al. (2009) evaluated the applicability of two
commercially available, acoustically transparent earpieces for SARA. They could show that
one of them (Surefire CommEarTMComfort EP1 earpieces on Etymotic Research ER4P
MircoPro earphones) allows for high quality SARA presentation while showing only small
(but significant) negative effects on free-field directional hearing and no occlusion effect.
Martin and colleagues, therefore, proposed this headphone model as a candidate for
application in mobility aids such as NAVIG. However, since our study showed that visually
impaired subjects only moderately make use of acoustic landmarks during navigation in an
artificial indoor environment, even if these landmarks are the most prominent acoustic
features of an otherwise virtually noise-free experimental setting, the question should be
raised, whether it is enough to show that an earpiece allows for good spatial hearing of
artificial broadband stimuli in an anechoic laboratory setup. The results of our study indicate
that more delicate acoustic information, such as echo-acoustic cues and fluctuations in the
ambient sound field, might play a vital role for wayfinding in the visually impaired, especially
in indoor environments. If the trend towards the utilisation of SARA as an important channel
of communication between mobility aid and user will be followed up on, the development of
better acoustically transparent earpieces or the adaptation of bone-conducting headphones for
spatial audio presentation will be challenging but inevitable tasks on the way towards an
universal electronic mobility aid for in- and outdoor navigation.
Conclusions/Summary
The study at hand set out to investigate, whether the gender differences in spatial learning and
memory that have been described for several mammalian species in vision-based experiments
can also be found in humans in tasks where the most prominent spatial information is
delivered via the auditory channel (i.e. whether the phenomenon is a- or supramodal). We
found that, during spatial learning under the described experimental conditions, male
!
88!
individuals needed less training-trials than the female subjects, which indicates differences in
the acquisition of spatial knowledge between the sexes. However, the found differences seem
to be partially influenced by internal mediators, such as higher security seeking in women.
The expected gender-differences in the utilisation of acoustic landmark information have not
convincingly been found. We assume that the loss of vision, as the most important source of
spatial information in normally sighted humans, has an effect comparable to that described for
increasing task complexity, i.e. masking of the gender differences that can be found in simple
orientation tasks in highly controlled laboratory settings by forcing the subjects to use all
available sources of spatial information, including the less preferred ones. Therefore, we hold
it possible that subtle gender-differences in acoustic landmark utilisation would have been
detectable under less complex laboratory conditions, but do not think that these differences
are of vital importance during everyday navigation, neither in normally sighted nor in visually
impaired humans.
The second goal of the study was to investigate, whether visually impaired and blind
subjects differ from a fully sighted but blindfolded control in the degree of sound-emitting
(i.e. acoustic) landmark utilisation during wayfinding. We found a significant difference in
the training performance between the two sub-groups (visually impaired vs. fully sighted) and
concluded that the earlier completion of the training phase in the visually impaired subjects
can probably be ascribed to an experience-based advantage in non-visual navigation these
subjects had over the fully sighted but blindfolded individuals and does not necessarily
indicate a higher learning rate. We also found that, among the visually impaired and blind
subjects, a significantly higher percentage, as compared to the sighted subjects, did not use
the provided acoustic landmarks for maze-navigation. From our observations during the
experiments and from statements the participants made during debriefing, we concluded that
some of the visually impaired subjects, instead, used a technique that is referred to as human
echolocation as well as other forms of spatial hearing to gather information about the spatial
layout of the maze during navigation. Consequently, it has to be assumed that spatial hearing
plays a non-negligible role in navigation in visually impaired and blind humans, especially
indoors. Since the acoustic cues that allow for spatial hearing are very delicate, one has to be
cautious when developing audio-based electronic travel and orientation aids: The
disadvantages that occur when audio presentation via headphones interferes with normal free-
!
89!
field hearing might easily outweigh the benefits that spatial augmented reality audio is
supposed to provide for navigation. The refinement of spatial audio presentation via boneconduction headphones, in our opinion, would be the most promising approach to circumvent
this trade-off.
Lastly, we found that the visually impaired subjects in our study performed
significantly worse than the fully sighted control when they had to generate plasticine models
of the “ideal path” through the maze from memory (i.e. they produced significantly less
accurate plasticine models). We concluded that these differences are due to a preference of
the visually impaired subjects for route over survey navigation: Survey representations of the
maze directly allow for the generation and validation of the requested plasticine model,
whereas model generation from route-like representations requires additional transformations
from somatotopic to allocentric spatial coordinates. While visually impaired subjects can
undoubtedly perform this task with good precision, as our results confirm, each change of
reference frame from egocentric to allocentric is a potential source of error that could explain
the decrease in model quality in the visually impaired as compared to the fully sighted
subjects, who more readily use survey-like, allocentric representations for navigation.
The fact that our “visually impaired” subjects behaved like a homogeneous group
throughout the whole study, even though the group was composed of congenitally blind and
late blind as well as visually impaired subjects with low residual vision, is probably based on
both the prominent role of the hippocampal formation for navigational tasks as well as the
uniform mobility training our subjects underwent during their education at the
“Landesbildungszentrum für Blinde Hannover”.
!
!
90!
!
!
91!
Structure and Possible Functions of Constant-Frequency Calls
in Ariopsis seemanni (Osteichthyes, Ariidae)
Authors:
Daniel Schmidtke1,2
Dr. Jochen Schulz3
Prof. Dr. Dr. h.c. Jörg Hartung3
PD Dr. Karl-Heinz Esser1,2
Authors’ affiliations:
1
Institute of Zoology, University of Veterinary
Hannover, Germany
2
Center for Systems Neuroscience, University of
Medicine, Hannover, Germany
3
Institute for Animal Hygiene, Animal Welfare,
Animal Behaviour, University of Veterinary
Hannover, Germany
Medicine,
Veterinary
and Farm
Medicine,
DS, JS, JH, and K-HE conceived and designed the experiments,
DS performed the experiments, DS analysed the data, JS and JH
and K-HE contributed materials and analysis tools, DS wrote the
manuscript.
Journal:
(status: published - doi:10.1371/journal.pone.0064864)
!
92!
Chapter 4: Structure and Possible Functions of ConstantFrequency Calls in Ariopsis seemanni (Osteichthyes, Ariidae)
Abstract
In the 1970s, Tavolga conducted a series of experiments in which he found behavioral
evidence that the vocalizations of the catfish species Ariopsis felis may play a role in a coarse
form of echolocation. Based on his findings, he postulated a similar function for the calls of
closely related catfish species. Here, we describe the physical characteristics of the
predominant call- type of Ariopsis seemanni. In two behavioral experiments, we further
explore whether A. seemanni uses these calls for acoustic obstacle detection by testing the
hypothesis that the call-emission rate of individual fish should increase when subjects are
confronted with novel objects, as it is known from other vertebrate species that use pulse-type
signals to actively probe the environment. Audio-video monitoring of the fish under different
obstacle conditions did not reveal a systematic increase in the number of emitted calls in the
presence of novel objects or in dependence on the proximity between individual fish and
different objects. These negative findings in combination with our current understanding of
directional hearing in fishes (which is a prerequisite for acoustic obstacle detection) make it
highly unlikely that A. seemanni uses its calls for acoustic obstacle detection. We argue that
the calls are more likely to play a role in intra- or interspecific communication (e.g. in school
formation or predator deterrence) and present results from a preliminary Y-maze experiment
that are indicative for a positive phonotaxis of A. seemanni towards the calls of conspecifics.
The online-version of the article can be found here:
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0064864
Digital Object Identifier:
doi:10.1371/journal.pone.0064864
!
93!
Chapter 5: General Discussion
The Dimensionality of Mammalian Allocentric Representations of Space
As discussed in chapter 2 of this thesis, the two-channel telemetric devices I tested in several
experiments and the protocols I developed for the manufacturing of Elgiloy micro-electrodes
and their chronic implantation in the brain of small mammals are very well suited for
neuronal recordings of multi-units in freely behaving mice. Examples for such multi-unit
recordings from the inferior colliculus and the hippocampus were presented in chapter 2
(Figs. 2.3a and 2.4b) and it is likely that even single-unit recordings in freely behaving
animals are possible in target areas that show low general activities and moderate or high
spacing between the target neurons (compare Fig. 2.3b). The hippocampus, however,
combines a relatively high general activity with small neuronal cell bodies that are densely
packed, as figures 2.4b and 2.5 show. These properties of the hippocampus, together with the
complex discharge characteristics of the target cells (Ranck 1973), notably complicate the
recording of neuronal activity from single cells in this brain area. While the third generation
of the transmitters tested here is routinely used by the working group of Prof. Dr. Oliver
Stiedl (Center for Neurogenomics and Cognitive Research, University of Amsterdam) for
hippocampal multi-unit recordings in the context of fear conditioning (Jansen et al. 2010),
current spike-sorting algorithms did not suffice to reliably isolate single units from the signals
I recorded from the hippocampus. In freely behaving (i.e. flying) microbats, the matter
becomes even more complicated, since muscle artefacts continuously mask the neuronal
signals during flight. Improving some of the methods described here, for example lowering
the impedance of the used microelectrodes through gold-plating to allow for smaller electrode
tips or refining the available spike-sorting algorithms, might solve a few of the encountered
problems. However, the development of multichannel transmitters has recently made
significant progress and will soon reach a stage, where the utilisation of multitrodes and
compactness of the telemetric devices will no longer be mutually exclusive. Thus, using
tetrodes in combination with multichannel telemetry will, in the long term, probably be the
more feasible approach to the special case of medial temporal single-cell recordings in freely
behaving small mammals.
!
94!
The test results of two very promising, telemetric multichannel neural headstages were
published in 2011 (Fan et al.) and 2012 (Roy and Wang). The transmitter developed by Fan
and colleagues weighs 4.5 g (including headstage and battery) and uses multiplexing in order
to combine 16 channels into a single, analogue signal that can later be de-multiplexed. The
transmitter was tested in mice and rats using chronically implanted, non-movable silicon
probes for single-unit recordings, which partially accounts for the overall low weight of the
device. Using movable tetrodes, instead, would have increased the weight of the setup, due to
the necessity of an additional microdrive. The second device (Roy and Wang 2012) has a total
weight of 12 g (including headstage and battery) and was tested in freely behaving
marmosets. Here, an array of 16 sharp metal electrodes was used for single-unit recordings.
The electrodes could individually be moved in the forward direction, but not backwards. The
implementation of a bidirectional microdrive would, again, have caused an increase in weight.
Two additional multichannel transmitters that are currently under development at the MaxPlanck-Institute for Ornithology in Seewiesen, Germany (Manfred Gahr, personal
communication) and at the Weizmann Institute of Science in Rehovot, Israel (Ulanovsky
2011) are directly designed for the utilisation with tetrodes, but prototypes have not been
published, yet.
Further support for the idea that the receptive fields of medial temporal allocentric
representations can only satisfactorily be characterised when investigated in flying (or diving)
mammalian species comes from a study in rats (Hayman et al. 2011), where place- and gridcell firing was recorded while animals explored a vertical climbing wall or a helical track.
Under these conditions, place and grid cells showed vertically elongated, i.e. anisotropic
receptive fields (compare Fig. 1.2). These findings are in discordance with the results of
earlier studies on three-dimensional place- and grid-cell firing (Jeffery et al. 2006, Knierim et
al. 2000, Knierim and McNaughton 2001, Ulanovsky and Moss 2007; summarised in chapter
1 of this thesis), where more isotropic receptive fields have been found. Based on their
results, Hayman and his colleagues (2011) concluded that the encoding of three-dimensional
space in mammals is less accurate or impaired in the vertical dimension. However, this is not
the only possible explanation for the vertically elongated receptive fields. Firstly, two other
studies have recently shown that in highly repetitive environments, such as multiple-U mazes
(Singer et al. 2010) or hairpin mazes (Derdikman et al. 2009), place and grid cells fire in
!
95!
multiple locations, but in stable relations to the geometry of the recurring elements (termed
“path equivalence” by Singer and colleagues). In place cells, the phenomenon even occurred
when animals behaviourally distinguished between the repeating elements or when the
recurring elements had distinct visual and olfactory cues (Singer et al. 2010; the same is
possibly true for grid cells, but Singer and co-workers only recorded from place cells). The
two experimental environments used by Hayman et al. (2011) also were highly repetitive. For
example, the rats might have experienced each complete winding of the helix as a recurring,
ring-shaped element of the environment. Path equivalence alone could then account for the
elongated firing fields that have been reported. Secondly, the movement patterns of the
experimental animals as well as their visual inputs were highly asymmetrically: Movements
occurred mainly in the horizontal plane and most of the visual cues (e.g. shelves, doors,
equipment) were restricted to the walls of the experimental room (Hayman et al. 2011,
supplementary materials). Thus, the anisotropic receptive fields of the place and grid cells
observed by Hayman and colleagues might also have been caused by asymmetrically
distributed visual cues and anisotropic movement patterns of the experimental animals
(Ulanovsky 2011). The only possibility to control for these factors would be to characterise
the firing fields of place and grid cells in an experimental animal that moves isotropically (i.e.
unrestrictedly along all axes of three-dimensional space) through an experimental
environment with symmetrically distributed orientation cues. The further development and/or
improvement of multichannel telemetric devices that allow for such recordings in bats might,
therefore, be a challenging but also a necessary step to answer the question of the
dimensionality of allocentric spatial representations in mammals.
Gender Differences in Mammalian Orientation and the Modality of
Allocentric Spatial Representations
As stated earlier in chapter 1, in order to establish echolocating bats as a model for research
into the dimensionality of allocentric spatial representations in mammals, it should be ensured
that the allocentric spatial representations of interest are (i) homologous rather than analogous
traits and (ii) that no major functional diversification of these representations occurred in
parallel to the adaptive radiation of mammals. For example, allocentric spatial representations
!
96!
in laryngeally echolocating microbats might have adapted to the aerial life style of this suborder and to the utilisation of echolocation as a primary source of environmental spatial
information and, therefore, be non-representative for other mammals.
I provided indirect evidence that this is not the case in a study on gender differences in
orientation in the laryngeally echolocating microbat species P. discolor (Schmidtke and Esser
2011), where female individuals were at a disadvantage (in terms of learning rate) in a routelearning task performed in the absence of landmarks. In a second experiment, navigation in
female bats was more affected by the modification of a previously learned configuration of
echo-acoustic landmarks than in male conspecifics. Both findings are in line with the results
of previous studies in species (including humans) from a different mammalian superorder
(Euarchontoglires), where visual instead of (echo-) acoustic landmarks have been used as
spatial cues. Since the phenomenon of gender differences in mammalian orientation in
euarchontogliran species is believed to emerge from a complex interplay of medial temporal
brain areas and the spatial representations they bear (Jacobs and Schenk 2003), my finding of
similar gender differences in a laurasiatherian species indicates that the underlying spatial
representations are rather functionally conservative within the mammalian class. I further
hypothesised from my results (Schmidtke and Esser 2011) that allocentric spatial
representations are largely independent of the sensory modality that is primarily used to
gather spatial information for orientation purposes.
One of the aims of the study presented in chapter 3 was to verify or disprove this
hypothesis by investigating the role of acoustic landmarks and possible gender-differences in
their utilisation in a way-finding task using visually impaired and blindfolded human subjects.
The results of the study show that significant differences in the rate of route learning between
the sexes (with lower learning rates in female individuals) can also be found in humans when
learning their way through a maze in the presence of acoustic landmarks, but in the absence of
any visual information. However, the SSS-V scores of the sighted but blindfolded subjects
indicate that internal mediators, such as higher security seeking in women, might also have
contributed to these differences.
An increased reliance on acoustic landmark information in women as compared to
men could not conclusively be found in the study. While this might indicate that gender
differences in cue-preferences as described for humans and other mammals in visually-guided
!
97!
orientation do not exist when landmark information is provided acoustically, a second
possibility is that task complexity (Barkley and Gabriel 2007, Chai and Jacobs 2009,
Rodríguez 2010) and/or the condition of visual deprivation masked gender differences in the
utilisation of acoustic landmarks that otherwise would have been detectable (compare chapter
3, “Discussion”). Further, the subjects might have used more abstract spatial representations
that are uniquely human to compensate for the loss of vision. Two possibilities for such
representations are linguistic representations of the pathway or the representation of distances
by counting discrete numbers of steps.
In the last part of the study, where the subjects had to generate plasticine models of the
“ideal path” through the maze from a “bird-view” (i.e. allocentric) perspective, most of the
erroneous models that had to be excluded from the subsequent analysis were produced by
female subjects. While this finding corresponds to what could have been expected in the light
of the literature on gender differences in spatial cognition in mammals (since it was mainly
Euclidean information that had to be incorporated into the models), the overall occurrence of
erroneous models was too low for the gender effect to reach the significance criterion of
α < 0.05.
In total, the results of the investigation of gender effects were ambiguous with respect
to my working hypothesis. While some of the findings fulfilled my expectations, others did
not and further studies under less complex conditions will be necessary to clarify the question,
whether gender differences in the utilisation of acoustic landmarks exist in humans or not. For
example, one could imagine a study where a modified version of the last room of the maze
(room 3/landmark C; compare Fig. 3.1) is used. Landmark C differed qualitatively from
landmark A and B (for details see chapter 3, “Discussion”) and its spatial information content
should have preferentially been used by male subjects. Accordingly, I found that only male
subjects showed a significant reaction to the modification of landmark C, resulting in a weak
statistical trend (p = 0.1004) towards a gender effect. Hence, conducting a study with a higher
number of subjects and an orientation task comparable to the situation in room 3 of the maze
is likely to reveal the expected gender differences.
Even though the analysis of gender effects on acoustic landmark utilisation in humans
presented in chapter 3 could not conclusively verify or disprove my hypothesis that
allocentric spatial representations are largely independent of the sensory modality that is
!
98!
primarily used to gather spatial information for orientation purposes, a different aspect of the
same study did provide evidence in favour for this hypothesis: All subjects of the study both
fully sighted and visually impaired ones were blindfolded while exploring and solving the
maze. Thus, the utilisation of visual spatial information for the formation of internal
representations of the maze by the subjects could per se be ruled out. However, the majority
of the subjects still produced accurate allocentric plasticine models of the “ideal path” through
the maze, showing that they were able to establish this allocentric representation solely on the
basis of idiothetic (i.e. internally generated) information and external, acoustic spatial cues.
Hence, the formation of allocentric spatial representations, indeed, seems to be independent of
the sensory modality that is primarily used by an individual for spatial-information gathering
(i.e. vision in normally sighted humans). The resulting spatial representations themselves,
consequently, are a- or supramodal.
Core Systems of Spatial Cognition in Vertebrates
In the very beginning of this thesis, I stated that our (adult) intuitive understanding of space is
that of an object space with a three-dimensional, Euclidean geometry. I pointed to the fact that
this intuition is not fully in accordance with our perception of space as it is based on abstract
knowledge that (in contrast to perceptual systems) is influenced by culture-specific
achievements and expressed through systems of symbols that are uniquely human (Spelke and
Lee 2012). The intrinsic geometry of perceptual space, instead, seems to be non-Euclidean as
accumulating evidence suggests (Fernandez and Farell 2009).
Currently, two fundamentally different schools of thought on how the intrinsic
geometry of perceptual space might differ from Euclidean geometry exist (Fernandez and
Farell 2009): The first one is founded on the works of Luneburg (1947) and Blank (1958,
1978) who reasoned that visual space can best be described as a Riemannian space of
constant curvature (Euclidean space is a special case of Riemannian space with zero
curvature). In a more recent experiment (Koenderink and van Doorn 2000) it was found that,
while visual space can, indeed, be described in terms of Riemannian geometry, the
assumption of constant curvature does not hold. Instead, the curvature of visual space seems
to vary between parabolic (at very large distances), hyperbolic (in far space), and elliptic (in
!
99!
near space). The second school of thought is based on the idea that the geometries of
perceptual space are non-Euclidean in that they are lacking an internal metric structure (e.g.
Gibson 1979). It could be shown in the past that human subjects can accurately assess the
topological, ordinal, and affine properties of an object, but accuracy decreases when metric
parameters, such as length and angles, have to be estimated (e.g. Domini and Braunstein
1998, Domini and Caudek 2003, Domini et al. 1998, Norman and Todd 1992, Tittle et al.
1995, Todd and Bressan 1990, Todd and Norman 1991). These findings are in support of the
view that metric structure might be less relevant to human perception than intuitively assumed
and that the basis of our conceptual understanding of Euclidean metric structure might rather
be cognitive than perceptual (Fernandez and Farell 2009).
Therefore, great care has to be taken when interpreting the results of studies in spatial
perception and cognition in humans and even more so in animals, since our intuitive concept
of Euclidean space is likely to distort these interpretations. A promising approach to
circumvent such problems is to use animal studies to identify cognitive “core systems” that
humans share with other animals and upon which uniquely human cognitive achievements
build (Spelke and Lee 2012). An example for the successful identification of such core
systems is the discovery of two non-verbal systems for quantity discrimination in humans and
other vertebrates. The first of these systems represents small numbers of objects (up to four)
in parallel and is ratio-independent, the other one has no upper limit in terms of represented
quantity but its precision is ratio-limited (i.e. the system serves the purpose of approximate
representation of large numerical magnitudes). Experimental evidence for the existence of
either one or both of these systems has been found in numerous vertebrate species as divers as
humans (e.g. Feigenson et al. 2004), different non-human primates (e.g. Brannon and Terrace
1998, Hauser et al. 2000, Merritt et al. 2011), black bears (Vonk and Beran 2012), pigeons
(e.g. Scarf et al. 2011), and guppies (Agrillo et al. 2012), to name only a few.
The allocentric spatial representations in the medial temporal brain areas introduced in
this thesis are likely to constitute such core systems of mammalian spatial cognition (Spelke
and Lee 2012): Place cells and border cells represent locations with respect to distance and
direction from extended surfaces in the environment (Lever et al. 2009, O’Keefe and Burgess
1996, Solstad et al. 2008), whereas grid cells bear the possibility to provide the human and
animal mind with the internal metric (Hafting et al. 2005, Jeffery and Burgess 2006) that
!
100!
seems to be lacking or underrepresented in perceptual space (see above). However, little is
known about the phylogeny of these systems in the vertebrate taxon. The study presented in
the fourth chapter of this thesis, therefore, investigated in how far the osteichthyan species A.
seemanni can be used for comparative studies in (auditory) vertebrate spatial cognition.
Is has been shown, in a row of laboratory-based spatial tasks, that different fish
species can use visual landmark information as direct cues for spatial problem solving
(Braithwaite et al. 1996, Girvan and Braithwaite 1998, Hughes and Blight 2000, Huntingford
and Wright 1989, Lopez et al. 1999, Lopez et al. 2000, Odling-Smee and Braithwaite 2003a,
Salas et al. 1996, Warburton 1990). Studies in the field revealed that natural migratory routes
are disturbed when landmarks are shifted or removed (Mazeroll and Montgomery 1998,
Reese 1989; for review see Odling-Smee and Braithwaite 2003b). In addition, fishes seem to
be able to use visual landmarks as indirect reference points (Warburton 1990) and to
recognize them from novel directions (Ingle and Sahagian 1973, Rodríguez et al. 1994).
Hence, the latter studies suggest that fishes are able to form some sort of a map-like,
allocentric representation of the environment.
When visual information is absent or scarce, fishes can also use landmark information
acquired through other sensory modalities: Gnathonemus petersii, a nocturnal momyrid,
recognises three-dimensional landmark configurations through electrolocation (Graff et al.
2004). The blind cavefish Astyanax fasciatus, instead, uses its lateral line system to learn
about landmark configurations by detecting small changes in water-flow patterns when
approaching the vicinity of objects (Burt de Perera 2004a,b, Burt de Perera et al. 2005).
Further possible, non-visual sources of spatial information are olfactory and acoustic
landmarks and gradients.
As mentioned in chapter 4 of this thesis, Tavolga (1971b, 1976, 1977) found evidence
for a coarse form of echolocation in the Hardhead sea catfish (Ariopsis felis). From his
findings, he hypothesised that the calls produced by other, closely related vocal fishes (such
as A. seemanni) serve the same purpose. Since then, this discovery of active acoustic
orientation in A. felis and the aforementioned hypothesis have been cited in numerous
scientific publications and textbooks, while the universal validity of the hypothesis has not
empirically been tested, yet. In the context of my thesis, it is easy to see, how an echolocating
fish species would provide a valuable tool for the investigation of the core systems of spatial
!
101!
cognition at the lower end of the vertebrate spectrum in general, but also in the special case of
(active) acoustic orientation that already connected chapter 2 and 3 of this thesis.
While the study presented in chapter 4 provides a first scientific description of the
physical characteristics of the calls emitted by A. seemanni and contributes to a long-lasting
debate on whether the spatial properties of a recording environment critically influence the
measurement of such physical parameters (e.g. Kaatz and Lobel 2001, Lugli et al. 1995,
Parvulescu 1967), no behavioural evidence (experiments 1 and 2, chapter 4) for a role of the
vocalisations of A. seemanni in echolocation was found. Instead, the results of the third
behavioural experiment presented in chapter 4 are indicative of a phonotaxic behaviour in
A. seemanni and a possible role of the calls in school formation (i.e. intraspecific
communication). Whether this phonotaxic behaviour can robustly be found in A. seemanni
and if it is mainly perception-based and egocentric (i.e. the fish orientate themselves towards
a sound source in egocentric, auditory space) or also allows for the investigation of the
utilisation of acoustic landmarks for orientation in this species, will be the subject of future
studies.
Final Remark
The research on spatial cognition has made enormous progress since Tolman (1948)
postulated the existence of a cognitive map. Studies in humans and non-human primates
elucidated many of the aspects of egocentric spatial representations and how information can
be exchanged between different body-centred reference frames or even between body- and
world-centred coordinate systems. Extensive research primarily conducted in rodents, on the
other hand, has yielded insight into the nature of allocentric spatial representations and their
neuronal correlates. However, the more knowledge we accumulated from these studies, the
more obvious it became that our reliance on a small set of model organisms in spatial,
cognitive research sets clear limits to our future gain of knowledge in this field. A more
comparative and highly collaborative approach, combining scientific fields as divers as
mathematics, engineering, geography, psychology, behavioural sciences and neurosciences
will, thus, be necessary to solve the puzzle of spatial cognition in mammals (and vertebrates)
to which this thesis provides a small piece.
!
102!
!
103!
References
Afonso A, Blum A, Katz B, Tarroux P, Borst G, Denis M (2010) Structural properties of
spatial representations in blind people: scanning images constructed from haptic
exploration or from locomotion in a 3-D audio virtual environment, Mem Cognit 38:
591-604.
Agrillo C, Piffer L, Bisazza A, Butterworth B (2012) Evidence for two numerical systems that
are similar in humans and guppies. PLOS ONE 7(2): e31923.
Aleman A, van Lee L, Mantione MH, Verkoijen IG, de Haan EH (2001) Visual imagery
without visual experience: evidence from congenitally totally blind people.
Neuroreport 12: 2601-2604.
Andersen RA, Essick GK, Siegel RM (1987) Neurons of area 7 activated by both visual
stimuli and oculomotor behaviour. Exp Brain Res 67(2): 316-322.
Arditi A, Holtzman JD, Kosslyn SM (1988) Mental imagery and sensory experience in
congenital blindness. Neuropsychologia 26: 1-12.
Armbrüster C, Heber I-A, Valvoda JT, Kuhlen T, Fimm B, Spijkers W (2005) Distance
estimations in a VR application: interindividual differences and interindividual
stabilities from a psychological point of view. In: Proceedings of the 11th International
Conference on Human-Computer Interaction (HCI Intl.). Mahwah: Lawrence Erlbaum
Associates.
Ashmead DH, Wall RS (1999) Auditory perception of walls via spectral variations in the
ambient sound field. J Rehabil Res Dev 36: 313-322.
Ashmead DH, Wall RS, Ebinger KA, Eaton SB, Snook-Hill MM, Yang X (1998) Spatial
hearing in children with visual disabilities. Perception 27: 105-122.
Au WWL, Hastings MC (2008) Principles of marine bioacoustics. New York: Springer
Science + Business Media.
Barkley CL, Gabriel KI (2007) Sex differences in cue perception in a visual scene:
investigation of cue type. Behav Neurosci 121: 291-300.
Barsalou LW (1999) Perceptual symbol systems. Behav Brain Sci 22: 610-60.
Bass AH, Baker R (1991) Evolution of homologous vocal control traits. Brain Behav Evol 38:
240-254.
Bass AH, Gilland EH, Baker R (2008) Evolutionary origins for social vocalization in a
vertebrate hindbrain-spinal compartment. Science 321: 417-421.
Beatty WW (1979) Gonadal hormones and sex differences in nonreproductive behaviors in
rodents: organizational and activational influences. Horm Behav 12: 112-163.
Beatty WW (2002) Sex difference in geographical knowledge: driving experience is not
essential. J Int Neuropsychol Soc 8: 804-810.
Beauducel A, Brocke B, Strobel A, Strobel A (1999) Construct validity of Sensation Seeking:
A psychometric investigation. Z Diff Diagn Psychol 20: 155-171.
!
104!
Beauducel A, Strobel A, Brocke B (2003) Psychometrische Eigenschaften und Normen einer
deutschsprachigen Fassung der Sensation Seeking-Skalen, Form V. Diagnostica 49:
61-72.
Bértolo H, Paiva T, Pessoa L, Mestre T, Marques R, Santos R (2003) Visual dream content,
graphical representation and EEG alpha activity in congenitally blind subjects. Brain
Res Cogn Brain Res 15: 277-284.
Betancur-R R (2009) Molecular phylogenetics and evolutionary history of ariid catfishes
revisited: a comprehensive sampling. BMC Evol Biol 9: 175-192.
Bingman VP (1990) Spatial navigation in birds. In: Kesner RP, Olton DS, editors.
Neurobiology of comparative cognition. Hillsdale: Lawrence Erlbaum. pp. 423-447.
Blank AA (1958) Analysis of experiments in binocular space perception. J Opt Soc Am 48:
911-925.
Blank AA (1978) Metric geometry in human binocular perception: theory and fact. In:
Leeuwenberg ELJ, Buffart HFJM, editors. Formal theories of visual perception. New
York: Wiley. pp. 83-102.
Blauert J (1996) Spatial hearing - revised edition: the psychophysics of human sound
localization. Cambridge: MIT Press.
Braithwaite VA, Armstrong JD, McAdam HM, Huntingford FA (1996) Can juvenile Atlantic
salmon use multiple cue systems in spatial learning? Anim Behav 51: 1409-1415.
Brannon EM, Terrace HS (1998) Ordering of the numerosities 1 to 9 by monkeys. Science
282: 746-749.
Bronkhorst AW, Houtgast T (1999) Auditory distance perception in rooms. Nature 397: 517520.
Burgess N (2006) Spatial memory: how egocentric and allocentric combine. Trends Cogn Sci
10(12): 551-557.
Burgess N (2008) Spatial cognition and the brain. Ann NY Acad Sci 1124: 77-97.
Burt de Perera T (2004a) Spatial parameters encoded in the spatial map of blind Mexican
cave fish, Astynax fasciatus. Anim Behav 68: 291-295.
Burt de Perera T (2004b) Fish can encode order in their spatial map. Proc Roy Soc Lond B
271: 2131-2134.
Burt de Perera T, de Vos A, Guilford T (2005) The vertical component of a fish’s spatial map.
Anim Behav 70: 405-409.
Campenhausen CV, Riess I, Weissert R (1981) Detection of stationary objects in the blind
cave fish Anoptichthys jordani (Characidae). J Comp Physiol A 143: 369-374.
Carreiras M, Codina B (1992) Spatial cognition of the blind and sighted: visual and amodal
hypotheses. Cah Psychol Cogn 12: 51-78.
Cattaneo Z, Fantino M, Silvanto J, Tinti C, Pascual-Leone A, Vecchi T (2010) Symmetry
perception in the blind. Acta Psychol (Amst) 134: 398-402.
!
105!
Cattaneo Z, Vecchi T (2011). Blind vision: the neuroscience of visual impairment.
Cambridge: MIT Press.
Chai XJ, Jacobs LF (2009) Sex differences in directional cue use in a virtual landscape.
Behav Neurosci 123: 276-283.
Chai XJ, Jacobs LF (2010) Effects of cue types on sex differences in human spatial memory.
Behav Brain Res 208: 336-342.
Chebat DR, Chen JK, Schneider F, Ptito A, Kupers R, Ptito M (2007) Alterations in right
posterior hippocampus in early blind individuals. Neuroreport 18(4): 329-33.
Clark CW, Ellison WT (2004) Potential use of low-frequency sounds by baleen whales for
probing the environment: evidence from models and empirical measurements. In:
Thomas JA, Moss CF, Vater M, editors. Echolocation in bats and dolphins. Cicago:
University Of Chicago Press. pp. 564-582.
Cochran W (1980) Wildlife telemetry. In: Schemnitz S, editor. Wildlife management
techniques manual (4th edition). Washington: The Wildlife Society.
Coleman PD (1962) Failure to localize the source distance of an unfamiliar sound. J Acoust
Soc Am 34: 345-346.
Coles RB, Konishi M, Pettigrew JD (1987) Hearing and echolocation in the Australian grey
swiftlet, Collocalia Spodiopygia. J Exp Biol 129: 365-371.
Colgin LL, Moser EI, Moser MB (2008) Understanding memory through hippocampal
remapping. Trends Neurosci 31(9): 469-477.
Corazzini LL, Tinti C, Schmidt S, Mirandola C, Cornoldi C (2010) Developing spatial
knowledge in the absence of vision: allocentric and egocentric representations
generated by blind people when supported by auditory cues, Psychol Belg 50: 327334.
Darwin C (1873) Origin of certain instincts. Nature 7: 417-418.
Dechent P, Merboldt K-D, Frahm J (2004) Is the human primary motor cortex involved in
motor imagery? Brain Res Cogn Brain Res 19: 138-144.
Deneve S, Latham PE, Pouget A (2001) Efficient computation and cue integration with noisy
population codes. Nat Neurosci 4(8): 826-831.
Derdikman D, Whitlock JR, Tsao A, Fyhn M, Hafting T, Moser MB, Moser EI (2009)
Fragmentation of grid cell maps in a multicompartment environment. Nat Neurosci
12: 1325-1332.
Després O, Boudard D, Candas V, Dufour A (2005) Enhanced self-localization by auditory
cues in blind humans. Disabil Rehabil 27: 753-759.
Dijkerman HC, Milner AD (1998) The perception and prehension of objects oriented in the
depth plane. II. Dissociated orientation functions in normal subjects. Exp Brain Res
118: 408-414.
Doeller CF, Barry C, Burgess N (2010). Evidence for grid cells in a human memory network.
Nature 463(7281): 657-661.
!
106!
Doeller CF, Barry C, Burgess N (2012) From cells to systems: grids and boundaries in spatial
memory. Neuroscientist 18(6): 556-566.
Domini F, Braunstein ML (1998) Recovery of 3-D structure from motion is neither Euclidean
nor affine. J Exp Psychol Human 24(4): 1273-1295.
Domini F, Caudek C, Richman S (1998) Distortions of depth-order relations and parallelism
in structure from motion. Percept Psychophys 60: 1164-1174.
Domini F, Caudek C. (2003) 3-D structure perceived from dynamic information: a new
theory. Trends Cogn Sci 10: 444-449.
Doucet M-E, Guillemot J-P, Lassonde M, Gagné J-P, Leclerc C, Lepore F (2005) Blind
subjects process auditory spectral cues more efficiently than sighted individuals. Exp
Brain Res 160: 194-202.
Dufour A, Després O, Candas V (2005) Enhanced sensitivity to echo cues in blind subjects.
Exp Brain Res 165: 515-519.
Einstein A (1917) Über die Spezielle und die Allgemeine Relativitätstheorie. Braunschweig:
Vieweg-Verlag.
Einstein A (1954) Foreword. In: Jammer M, author. Concepts of space. The history of
theories of space in physics. Cambridge: Harvard University Press. pp. XII-XVII.
Ekstrom AD, Kahana MJ, Caplan JB, Fields TA, Isham EA, Newman EL, Fried I (2003)
Cellular networks underlying human spatial navigation. Nature 425: 184-187.
Eriksson PS Perfilieva E, Björk-Eriksson T, Alborn AM, Nordborg C, Peterson DA, Gage FH
(1998) Neurogenesis in the adult human hippocampus. Nat Med 4: 1313-1317.
Etienne AS, Jeffery KJ (2004) Path integration in mammals. Hippocampus 14: 180-192.
Etienne AS, Maurer R, Seguinot V (1996) Path integration in mammals and its integration
with visual landmarks. J Exp Biol 199(1): 201-209.
Evans WE (1973) Echolocation by marine delphinids and one species of fresh-water dolphin,
J Acoust Soc Am 54: 191-199.
Fan D, Rich D, Holtzman T, Ruther P, Dalley JW, Lopez A, Rossi MA, Barter JW, SalasMeza D, Herwik S, Holzhammer T, Morizio J, Yin HH (2011) A wireless multichannel recording system for freely behaving mice and rats. PLOS ONE 6(7): e22033.
Fay RR, Edds-Walton PL (1997) Directional response properties of saccular afferents of the
toadfish, Opsanus tau. Hear Res 111 :1-21.
Fay RR, Popper AN, Webb JF (2008) Introduction to fish bioacoustics. In: Webb JF, Popper
AN, Fay RR, editors. Fish bioacoustics. New York: Springer Science + Business
Media. pp. 1-15.
Feigenson L, Dehaene S, Spelke E (2004) Core systems of number. Trends Cogn Sci 8(7):
307-314.
Fernandez JM, Farell B (2009) Is perceptual space inherently non-Euclidean? J Math Psychol
53(2): 86-91.
!
107!
Fieger A, Röder B, Teder-Sälejärvi W, Hillyard SA, Neville HJ (2006) Auditory spatial
tuning in late-onset blindness in humans. J Cogn Neurosci 18: 149-157.
Flock A (1964) Structure of the macula utriculi with special reference to directional interplay
of sensory responses as revealed by morphological polarization. J Cell Biol 22: 413431.
Forcano L, Santamaría J, Mackintosh NJ, Chamizo VD (2009) Single landmark learning in
rats: sex differences in a navigation task. Learn Motiv 40: 46-61.
Fortin M, Voss P, Lord C, Lassonde M, Pruessner J, Saint-Amour D, Rainville C, Lepore F
(2008) Wayfinding in the blind: larger hippocampal volume and supranormal spatial
navigation. Brain 131: 2995-3005.
Fortune EE, Goodie AS (2010) The relationship between pathological gambling and sensation
seeking: the role of subscale scores. J Gambl Stud 26: 331-346.
Foulke E, Hatlen PH (1992) A collaboration of two technologies. Part 1: perceptual and
cognitive processes: their implications for visually impaired persons. Br J Vis Impair
10: 43-46.
Franz MO, Mallot HA (2000) Biomimetic robot navigation. Robot Auton Syst 30: 133-153.
Friedman A, Kohler B (2003) Bidimensional regression: assessing the configural similarity
and accuracy of cognitive maps and other two-dimensional data sets. Psychol Methods
8: 468-491.
Fusiello A, Panuccio A, Murino V, Fontana F, Rocchesso D (2002) A multimodal electronic
travel aid device. In: Proceedings of the 4th IEEE international conference on
multimodal interfaces (ICMI'02). Washington: IEEE Computer Society. pp. 39.
Fyhn M, Hafting T, Witter MP, Moser EI, Moser MB (2008). Grid cells in mice.
Hippocampus 18(12): 1230-1238.
Galea LAM, Kimura D (1993) Sex differences in route-learning. Pers Individ Dif 14: 53-65.
Gallistel CR (1989) Animal cognition: the representation of space, time and number. Annu
Rev Psychol 40: 155-189.
Gardini S, Cornoldi C, De Beni R, Venneri A (2009) Cognitive and neuronal processes
involved in sequential generation of general and specific mental images. Psychol Res
73: 633-643.
Gardini S, De Beni R, Cornoldi C, Bromiley A, Venneri A (2005) Different neuronal
pathways support the generation of general and specific mental images. Neuroimage
27: 544-552.
Gardner MB (1969) Distance estimation of 0° degrees or apparent 0° degree-oriented speech
signals in anechnoic space. J Acoust Soc Am 45: 47-53.
Garilao CV, Bailly N (2013) Ariopsis seemanni. In: Froese R, Pauly D, editors. FishBase
(version 01/2013). World Wide Web electronic publication.
www.fishbase.org/summary/Ariopsis-seemanni.html. Accessed 26 January 2013.
Gaulin SJC, FitzGerald RW (1986) Sex differences in spatial ability: an evolutionary
hypothesis and test. Am Nat 127: 74-88.
!
108!
Gibson JJ (1979) The ecological approach to visual perception. Boston: Haughton Mifflin.
Girvan JR, Braithwaite VA (1998) Population differences in spatial learning in three-spined
sticklebacks. Proc R Soc Lond B 265: 913-919.
Gougoux F, Zatorre RJ, Lassonde M, Voss P, Lepore F (2005) A functional neuroimaging
study of sound localization: visual cortex activity predicts performance in early-blind
individuals. PLOS Biol 3: e27.
Goyal MS, Hansen PJ, Blakemore CB (2006) Tactile perception recruits functionally related
visual areas in the late-blind. Neuroreport 17: 1381-1384.
Graff C, Kaminski G, Gresty M, Ohlmann T (2004) Fish perform spatial pattern recognition
and abstraction by exclusive use of active electrolocation. Cur Biol 14: 818-823.
Gray CM, Maldonado PE, Wilson M, McNaughton B (1995). Tetrodes markedly improve the
reliability and yield of multiple single-unit isolation from multi-unit recordings in cat
striate cortex. J Neurosci Meth 63: 43-54.
Gregory RL (1997) Knowledge in perception and illusion. Philos Trans R Soc Lond B Biol
Sci 352: 1121-1127.
Gregory RL (2005) The Medawar Lecture 2001 knowledge for vision: vision for knowledge.
Philos Trans R Soc Lond B Biol Sci 360: 1231-1251.
Griffin DR (1958) Listening in the dark. New Haven: Yale University Press.
Grön G, Wunderlich AP, Spitzer M, Tomczak R, Riepe MW (2000) Brain activation during
human navigation: gender-different neural networks as substrate of performance. Nat
Neurosci 3: 404-408.
Günther ACLG (1864) Catalogue of the fishes in the British Museum. Volume 5. Catalogue
of the Physostomi, containing the families Siluridae, Characinidae, Haplochitonidae,
Sternoptychidae, Scopelidae, Stomiatidae in the collection of the British Museum.
London: Taylor & Francis.
Haddad V, Martins IA (2006) Frequency and gravity of human envenomations caused by
marine catfish (suborder siluroidei): a clinical and epidemiological study. Toxicon 47:
838-843.
Hafting T, Fyhn M, Molden S, Moser MB, Moser EI (2005) Microstructure of a spatial map
in the entorhinal cortex. Nature 436: 801-806.
Härmä A, Jakka J, Tikander M, Karjalainen M, Lokki T, Hiipakka J, Lorho G (2004)
Augmented reality audio for mobile and wearable appliances. J Audio Eng Soc 52:
618-639.
Harris LJ (1978) Sex differences in spatial ability: possible environmental, genetic, and
neurological factors. In: Kinsbourne M, editor. Asymmetrical function of the brain.
Cambridge: Cambridge University Press. pp. 405-522.
Hartley T, Burgess N (2003) Models of spatial cognition. In: Nadel L, editor. Encyclopedia of
cognitive science. London: MacMillan.
Hauser MD, Carey S, Hauser LB (2000) Spontaneous number representation in semi-freeranging rhesus monkeys. Proc R Soc Lond B 267: 829-833.
!
109!
Hausfeld S, Power RP, Gorta A, Harris P (1982) Echo perception of shape and texture by
sighted subjects. Percept Mot Skills 55: 623-632.
Hayman R, Verriotis M, Jovalekic A, Fenton AA, Jeffery KJ (2011) Anisotropic encoding of
three-dimensional space by place cells and grid cells. Nat Neurosci 14(9): 1182-1188.
Helal A, Moore SE, Ramachandran B (2001) Drishti: an integrated navigation system for
visually impaired and disabled. In: Proceedings of the 5th International Symposium on
Wearable Computer. Zürich: ETH.
Hill AJ, Best PJ (1981) Effects of deafness and blindness on the spatial correlates of
hippocampal unit activity in the rat. Exp Neurol 74: 204-217.
Hittner JB, Swickert R (2006) Sensation seeking and alcohol use: a meta-analytic review.
Addict Behav 31: 1383-1401.
Horvath P, Zuckerman M (1993) Sensation seeking, risk appraisal, and risky behaviour. Pers
Individ Dif 14: 41-52.
Hothorn T, Bretz F, Westfall P (2008) Simultaneous inference in general parametric models.
Biom J 50: 346-363.
Hristov NI, Conner WE (2005) Sound strategy: acoustic aposematism in the bat-tiger moth
arms race. Naturwissenschaften 92: 164-169.
Hughes RN, Blight CM (2000) Two intertidal fish species use visual association learning to
track the status of food patches in a radial maze. Anim Behav 59: 613-621.
Huntingford FA, Wright PJ (1989) How sticklebacks learn to avoid feeding patches. Behav
Process 19: 181-189.
Iachini T, Sergi I, Ruggiero G, Gnisci A (2005) Gender differences in object location memory
in a real three-dimensional environment. Brain Cogn 59: 52-59.
Ingle D, Sahagian D (1973) Solution of a spatial constancy problem by goldfish. Physiol
Psychol 1: 83-84.
Jacobs LF, Gaulin SJ, Sherry DF, Hoffman GE (1990) Evolution of spatial cognition: sexspecific patterns of spatial behavior predict hippocampal size. Proc Natl Acad Sci
USA 87: 6349-6352.
Jacobs LF, Schenk F (2003) Unpacking the cognitive map: the parallel map theory of
hippocampal function. Psychol Rev 110: 285-315.
Jammer M (1993) Concepts of space. The history of theories of space in physics (3rd, enlarged
edition). New York: Dover Publications.
Jansen RF, Pieneman AW, ter Maat A, Gahr M, Stiedl O (2010) Extracellular multi unit
recording in fear conditioning in mice using a telemetry approach in an automated
home cage (DualCage) environment. In: Spink AJ, Grieco F, Krips OE, Loijens LWS,
Noldus LPJJ, Zimmermann PH, editors. Proceedings of Measuring Behavior. p. 58.
Jeannerod M, Arbib MA, Rizzolatti G, Sakata H (1995) Grasping objects: the cortical
mechanisms of visuomotor transformation. Trends Neurosci 18: 314-320.
!
110!
Jeffery KJ, Anand RL, Anderson MI (2006) A role for terrain slope in orienting hippocampal
place fields. Exp Brain Res 169: 218-225.
Jeffery KJ, Burgess N (2006) A metric for the cognitive map: found at last? Trends Cogn Sci
10(1): 1-3.
Jog MS, Conolly CI, Kubota Y, Iyengar DR, Garrido L, Harlan R, Graybiel AM (2002)
Tetrode technology: advances in implantable hardware, neuroimaging, and data
analysis techniques. J Neurosci Methods 117: 141-52.
Jung MW, McNaughton BI (1993) Spatial selectivity of unit activity in the hippocampal
granular layer. Hippocampus 18: 1175-1185.
Kaatz IM, Lobel PS (2001) A comparison of sounds recorded from a catfish (Orinocodoras
eigenmanni, Doradidae) in an aquarium and in the field. Biol Bull 201: 278-280.
Kailola PJ, Bussing WA (1995) Ariidae. Bagres marinos. In: Fischer W, Krupp F, Schneider
W, Sommer C, Carpenter KE, Niem V, editors. Guia FAO para Identification de
Especies para lo Fines de la Pesca. Pacifico Centro-Oriental. 3 Vols. Rome: FAO. pp.
860-886
Kalmijn AJ (1989) Functional evolution of lateral line and inner ear sensory systems. In:
Coombs S, Gorner P, Munz H, editors. The mechanosensory lateral line: neurobiology
and evolution. New York: Springer-Verlag. pp. 187-216.
Kaluza T (1921). Zum Unitätsproblem in der Physik. Sitzungsber Preuss Akad Wiss Berlin
(Math Phys) 96: 69-72.
Kammoun S, Parseihian G, Gutierrez O, Brilhault A, Serpa A, Raynal M, Oriola B, Macé MJM, Auvray M, Denis M, Thorpe SJ, Truillet P, Katz BFG, Jouffrais C (2012)
Navigation and space perception assistance for the visually impaired: The NAVIG
project, IRBM 33: 182-189.
Karleskint G, Turner RL, Small JW (2009) Introduction to marine biology (3rd revised
edition). Belmont: Brooks/Cole - Cengage Learning.
Kaski D (2002) Revision: is visual perception a requisite for visual imagery? Perception 31:
717-731.
Katz BFG, Dramas F, Parseihian G, Gutierrez O, Kammoun S, Brilhault A, Brunet L, Gallay
M, Oriola B, Auvray M, Truillet P, Denis M, Thorpe S, Jouffrais C (2012) NAVIG:
Guidance system for the visually impaired using virtual augmented reality, Technol
Disabil 24: 163-178.
Katz BFG, Truillet P, Thorpe S, Jouffrais C (2010) NAVIG: Navigation assisted by artificial
vision and GNSS. Workshop on multimodal location based techniques for extreme
navigation (Pervasive 2010.
Kellogg WN (1962) Sonar system of the blind. Science 137: 399-404.
Kentros CG, Agnihotri NT, Streater S, Hawkins RD, Kandel ER (2004) Increased attention to
spatial context increases both place field stability and spatial memory. Neuron 42:
283-295.
!
111!
Kerr NH (1983) The role of vision in 'visual imagery' experiments: evidence from the
congenitally blind. J Exp Psychol Gen 112: 265-277.
Killian NJ, Jutras MJ, Buffalo EA (2012). A map of visual space in the primate entorhinal
cortex. Nature 491(7426): 761-764.
Klatzky RL, Loomis JM, Golledge RG, Cicinelli JG, Doherty S, Pellegrino JW (1990)
Acquisition of route and survey knowledge in the absence of vision. J Mot Behav 22:
19-43.
Klein O (1926). Quantentheorie und fünfdimensionale Relativitätstheorie. Z Phys A 37(12):
895-906.
Knierim JJ, McNaughton BL (2001) Hippocampal place-cell firing during movement in
three-dimensional space. J Neurophysiol 85: 105-116.
Knierim JJ, McNaughton BL, Poe GR (2000) Three-dimensional spatial selectivity of
hippocampal neurons during space flight. Nat Neurosci 3: 209-210.
Kobayashi M, Takeda M, Hattori N, Fukunaga M, Sasabe T, Inoue N, Nagai Y, Sawada T,
Sadato N, Watanabe Y (2004) Functional imaging of gustatory perception and
imagery: "top-down" processing of gustatory signals. Neuroimage 23: 1271-1282.
Koenderink JJ, van Doorn AJ (2000) Direct measurement of the curvature of visual space.
Perception 29: 69-79.
Konishi M, Knudsen EI (1979) The oilbird: hearing and echolocation. Science 204: 425-427.
Kosslyn SM (1973) Scanning visual images: Some structural implications. Percept
Psychophys 14: 90-94.
Kosslyn SM (1980) Image and Mind. Cambridge: Harvard University Press.
Kosslyn SM, Ball TM, Reiser BJ (1978) Visual images preserve metric spatial information:
evidence from studies of image scanning. J Exp Psychol Hum Percept Perform 4: 4760.
Kosslyn SM, Ganis G, Thompson WL (2001) Neural foundations of imagery. Nat Rev
Kosslyn SM, Pomerantz JR (1977) Imagery, propositions, and the form of internal
representations, Cogn Psychol 9: 52-76.
Krogh A (1929) The progress of physiology. Am J Physiol 90(2): 243-251.
Kujala T, Alho K, Huotilainen M, Ilmoniemi RJ, Lehtokoski A, Leinonen A, Rinne T,
Salonen O, Sinkkonen J, Standertskjöld-Nordenstam CG, Näätänen R (1997)
Electrophysiological evidence for cross-modal plasticity in humans with early- and
late-onset blindness. Psychophysiology 34: 213-216.
Kujala T, Alho K, Paavilainen P, Summala H, Näätänen R (1992) Neural plasticity in
processing of sound location by the early blind: an event-related potential study,
Electroen Clin Neuro 84: 469-472.
Kuznetsov MY (2009) Traits of acoustic signalization and generation of sounds by some
schooling Physostomous fish. Acoust Phys 55: 799-808.
!
112!
Kwiecinsky GG (2006) Phyllostomus discolor. Mamm Spec 801: 1-11.
Lacreuse A, Kim CB, Rosene DL, Killiany RJ, Moss MB, Moore TL, Chennareddi L,
Herndon JG (2005) Sex, age, and training modulate spatial memory in the rhesus
monkey (Macaca mulatta). Behav Neurosci 119: 118-126.
Leclerc C, Saint-Amour D, Lavoie ME, Lassonde M, Lepore F (2000) Brain functional
reorganization in early blind humans revealed by auditory event-related potentials.
Neuroreport 11: 545-550.
Legge GE, Madison C, Vaughn BN, Cheong AMY, Miller JC (2008) Retention of high tactile
acuity throughout the life span in blindness. Percept Psychophys 70: 1471-1488.
Lepore N, Shi Y, Lepore F, Fortin M, Voss P, Chou YY, Lord C, Lassonde M, Dinov I, Toga
AW, Thompson PM (2009) Patterns of hippocampal shape and volume differences in
blind subjects. Neuroimage 46(4): 949-957.
Lessard N, Paré M, Lepore F, Lassonde M (1998) Early-blind human subjects localize sound
sources better than sighted subjects. Nature 395: 278-280.
Lever C, Burton S, Jeewajee A, O’Keefe J, Burgess N (2009) Boundary vector cells in the
subiculum of the hippocampal formation. J Neurosci 29: 9771-9777.
Levy LJ, Astur RS, Frick KM (2005) Men and women differ in object memory but not
performance of a virtual radial maze. Behav Neurosci 119: 853-862.
Lewald J (2002) Vertical sound localization in blind humans. Neuropsychologia 40: 18681872.
Linn MC, Peterson AC (1986) A meta-analysis of gender differences in spatial ability:
implications for mathematics and science achievement. In: Hyde JS, Linn MC, editors.
The psychology of gender: advances through meta-analysis. Baltimore: Johns Hopkins
University Press. pp. 67-101.
Loomis JM, Golledge RG, Klatzky RL, Speigle JM, Tietz J (1994) Personal guidance system
for the visually impaired. In: Proceedings of the first international ACM/SIGCAPH
conference on assistive technologies. pp. 85-91.
Loomis JM, Klatzky RL, Golledge RG, Cicinelli JG, Pellegrino JW, Fry PA (1993)
Nonvisual navigation by blind and sighted: assessment of path integration ability. J
Exp Psychol Gen 122: 73-91.
Loomis JM, Klatzky RL, Philbeck JW, Golledge RG (1998) Assessing auditory distance
perception using perceptually directed action. Percept Psychophys 60: 966-980.
Lopez JC, Bingman VP, Rodríguez F, Gomez Y, Salas C (2000) Dissociation of place and
cue learning by telencephalic ablation in goldfish. Behav Neurosci 114: 687-699.
Lopez JC, Broglio C, Rodríguez F, Thinus-Blanc C, Salas C (1999) Multiple spatial learning
strategies in goldfish (Carassius auratus). Anim Cogn 2: 109-120.
Ludvig N, Tang HM, Gohil BC, Botero JM (2004) Detecting location-specific neuronal firing
rate increases in the hippocampus of freely-moving monkeys. Brain Res 1014: 97-109.
Lugli M, Pavan G, Torricelli P, Bobio L (1995) Spawning vocalizations in male freshwater
gobiids (Pisces, Gobiidae). Environ Biol Fish 43: 219-231.
!
113!
Luneburg, RK (1947) Mathematical analysis of binocular vision. Princeton: Princeton
University Press.
Lynch K. (1960) The image of the city. Cambridge: MIT Press.
Maccoby E, Jacklin CN. 1974. The psychology of sex differences. Stanford: Stanford
University Press.
Madsen PT, Johnson M, de Soto NA, Zimmer WMX, Tyack P (2005) Biosonar performance
of foraging beaked whales (Mesoplodon densirostris). J Exp Biol 208: 181-194.
Manner C, Blakley S, Lawrence S, O’Neill E, Raines C (2011) Understanding the predictors
of negative personal relationship experiences on Facebook. Int J Hum Soc Sci 1: 1619.
Martin A, Jin C, van Schaik A (2009) Psychoacoustic evaluation of systems for delivering
spatialized augmented-reality audio. J Audio Eng Soc 57: 1016-1027.
Mazeroll AI, Montgomery WL (1998) Daily migrations of a coral reef fish in the Red Sea
(Gulf of Aqaba, Israel): initiation and orientation. Copeia 4: 893-905.
McGee MG (1979) Human spatial abilities: psychometric studies and environmental, genetic,
hormonal, and neurological influences. Psychol Bull 86: 889-918.
McNaughton BL, O'Keefe J, Barnes CA (1983) The stereotrode: a new technique for
simultaneous isolation of several single units in the central nervous system. J Neurosci
Meth 8: 391-397.
Merritt DJ, MacLean EL, Crawford JC, Brannon EM (2011) Numerical rule-learning in ringtailed lemurs (Lemur catta). Front Psychol 2: 23.
Mershon DH, Bowers JN (1979) Absolute and relative cues for the auditory perception of
egocentric distance. Perception 8: 311-322.
Mershon DH, King LE (1975) Intensity and reverberation as factors in the auditory
perception of egocentric distance. Percept Psychophys 18: 409-415.
Middlebrooks JC, Green DM (1991) Sound localization by human listeners. Annu Rev
Psychol 42: 135-159.
Milner AD (1997) Vision without knowledge. Philos Trans R Soc Lond B 352: 1249-1256.
Minkowski H (1909) Raum und Zeit. Leipzig: B.G. Teubner Verlag.
Moffat SD, Hampson E, Hatzipantelis M (1998) Navigation in a "virtual" maze: sex
differences and correlation with psychometric measures of spatial ability in humans.
Evol Hum Behav 19: 73-87.
Mogdans J, Bleckmann H (2012) Coping with flow: behavior, neurophysiology and modeling
of the fish lateral line system. Biol Cybern 106: 627-642.
Möhres FP, Kulzer E (1956) Über die Orientierung der Flughunde (Chiroptera Pteropodidae). Z Vergl Physiol 38: 1-29.
Moller P (1995) Electric fishes: history and behavior. London: Chapman & Hall.
!
114!
Nakatani T, Irino T (2004) Robust and accurate fundamental frequency estimation based on
dominant harmonic components. J Acoust Soc Am 116(6): 3690-700.
Neuhoff JG, Planisek R, Seifritz E (2009) Adaptive sex differences in auditory motion
perception: looming sounds are special. J Exp Psychol Hum Percept Perform 35: 225234.
Noordzij ML, Zuidhoek S, Postma A (2006) The influence of visual experience on the ability
to form spatial mental models based on route and survey descriptions. Cognition 100:
321-342.
Norman JF, Todd JT (1992) The visual perception of 3-dimensional form. In: Carpenter GA,
Grossberg S, editors. Neural networks for vision and image processing. Cambridge:
MIT press. pp. 93-110.
Norris KS (1969) The echolocation of marine mammals. In: Andersen HT, editor. The
Biology of Marine Mammals. New York: Academic Press.
Norris KS (1995) Whales, dolphins, and porpoises. Washington: National Geographic
Society.
Nowak NT, Resnick SM, Elkins W, Moffat SD (2011) Sex differences in brain activation
during virtual navigation: a functional MRI study. 33rd Annual Meeting of the
Cognitive Science Society.
O'Laughlin EM, Brubaker BS (1998) Use of landmarks in cognitive mapping: gender
differences in self report versus performance, Pers Individ Dif 24: 595-601.
O’Keefe J, Burgess N (1996) Geometric determinants of the place fields of hippocampal
neurons. Nature 381: 425-428.
O’Keefe J, Dostrovsky J (1971) The hippocampus as a spatial map. Preliminary evidence
from unit activity in the freely-moving rat. Brain Res 34: 171-175.
O’Keefe J, Nadel L (1978) The hippocampus as a cognitive map. New York: Oxford
University Press.
Odling-Smee L, Braithwaite VA (2003a) The influence of habitat stability on landmark use
during spatial learning in the threespine stickleback. Anim Behav 65: 701-707.
Odling-Smee L, Braithwaite VA (2003b) The role of learning in fish orientation. Fish Fish 4:
235-246.
Parvulescu A (1967) The acoustics of small tanks. In: Tavolga WN, editor. Marine
bioacoustics, Vol. 2. Oxford: Pergamon Press. pp. 7-13.
Pietrini P, Furey ML, Ricciardi E, Gobbini MI, Wu W-HC, Cohen L, Guazzelli M, Haxby JV
(2004) Beyond sensory images: object-based representation in the human ventral
pathway. Proc Natl Acad Sci USA 101: 5658-5663.
Pinheiro JC, Bates DM (2000) Mixed-effects models in S and S-PLUS. New York: Springer
Verlag.
Pinheiro JC, Bates DM, DebRoy S, Sarkar D, R Development Core Team (2009) Linear and
nonlinear mixed effects models. Package “nlme”, version 3.1-96.
!
115!
Popper AN (1970) Auditory capacities of the Mexican blind cave fish (Astyanax jordani) and
its eyed ancestor (Astyanax mexicanus). Anim Behav 18: 552-562.
Popper AN, Fay RR, Platt C, Sand O (2003) Sound detection mechanisms and capabilities of
teleost fish. In: Collin SP, Marshall NJ, editors. Sensory processing in aquatic
environments. New York: Springer-Verlag. pp. 3-38.
Popper AN, Plachta DTT, Mann DA, Higgs D (2004) Response of clupeid fish to ultrasound:
a review. ICES J Mar Sci 61: 1057-1061.
Portfors CV, Mayko ZM, Jonson KG, Cha G, Roberts PD (2011) Spatial organization of
receptive fields in the auditory midbrain of awake mouse. Neuroscience 193: 429-439.
Post N, von der Emde G (1999) The ‘novelty response’ in an electric fish: response properties
and habituation. Physiol Behav 68: 115–128.
Pylyshyn ZW (1973) What the mind’s eye tells the mind’s brain: a critique of mental
imagery. Psychol Bull 80: 1-24.
Pylyshyn ZW (2003) Explaining mental imagery: now you see it, now you don't. Reply to
Kosslyn et al. Trends Cogn Sci 7: 111-112.
Ranck JB Jr. (1973) Studies on single neurons in dorsal hippocampual formation and septum
in unrestrained rats. Exp Neurol 41: 461-555.
Recce ML, O'Keefe J (1989) The tetrode: an improved technique for multi-unit extracellular
recording. Soc Neurosci Abstr 15: 1250.
Reese ES (1989) Orientation behaviour of butterflyfishes (family Chaetodontidae) on coral
reefs: spatial learning of route specific landmarks and cognitive maps. Environ Biol
Fish 25: 79-86.
Reis RE, Kullander SO, Ferraris CJ (2003) Methods. In: Reis RE, Kullander SO, Ferraris CJ,
editors. Check list of the freshwater fishes of South and Central America. Porto
Alegre: EDIPUCRS. pp. 4-8.
Renouf D, Davis MB (1982) Evidence that seals may use echolocation. Nature 300: 635-637.
Rice CE (1967) Human echo perception. Science 155: 656-664.
Rice CE (1969) Perceptual enhancement in the early blind. Psychol Rec 19: 1-14.
Rice CE, Feinstein SH, Shusterman RJ (1965) Echo detection of the blind: size and distance
factors. J Exp Psychol 70: 74-80.
Rieser JJ, Lockman JJ, Pick HL Jr (1980) The role of visual experience in knowledge of
spatial layout. Percept Psychophys 28: 185-190.
Roche RAP, Mangaoang MA, Commins S, O'Mara SM (2005) Hippocampal contributions to
neurocognitive mapping in humans: a new model. Hippocampus 15: 622-641.
Röder B, Teder-Sälejärvi W, Sterr A, Rösler F, Hillyard SA, Neville HJ (1999) Improved
auditory spatial tuning in blind humans. Nature 400: 162-166.
Rodríguez CA, Torres A, Mackintosh NJ, Chamizo VD (2010) Sex differences in the
strategies used by rats to solve a navigation task. J Exp Psychol Anim Behav Process
36: 395-401.
!
116!
Rodríguez F, Duran E, Vargas JP, Torres B, Salas, C (1994) Performance of goldfish trained
in allocentric and egocentric maze procedures suggest the presence of a cognitive
mapping system in fishes. Anim Learn Behav 22: 409-420.
Roentgen UR, Gelderblom GJ, Soede M, de Witte LP (2008) Inventory of electronic mobility
aids for persons with visual impairments: a literature review. J Vis Impair Blind 102:
702-724.
Roof RL, Stein DG (1999) Gender differences in Morris water maze performance depend on
task parameters. Physiol Behav 68: 81-86.
Rosenblum LD, Gordon MS, Jarquin L (2000) Echolocating distance by moving and
stationary listeners. Ecol Psychol 12: 181-206.
Roy S, Wang X (2012) Wireless multi-channel single unit recording in freely moving and
vocalizing primates. J Neurosci Meth 203: 28-40.
Salas C, Broglio C, Rodríguez F, Lopez JC, Portavella M, Torres B (1996) Telencephalic
ablation in goldfish impairs performance in a spatial constancy problem but not in a
cued one. Behav Brain Res 79: 193-200.
Sand O (1974) Directional sensitivity of microphonic potentials from the perch ear. J Exp
Biol 60: 881-899.
Sandstrom NJ, Kaufman J, Huettel SA (1998) Males and females use different distal cues in a
virtual environment navigation task. Brain Res Cogn Brain Res 6: 351-360.
Sargolini F, Fyhn M, Hafting T, McNaughton BL, Witter MP, Moser MB, Moser EI (2006)
Conjunctive representation of position, direction, and velocity in entorhinal cortex.
Science 312: 758-762.
Saucier DM, Green SM, Leason J, MacFadden A, Bell S, Elias LJ (2002) Are sex differences
in navigation caused by sexually dimorphic strategies or by differences in the ability to
use the strategies?, Behav Neurosci 116: 403-410.
Save E, Cressant A, Thinus-Blanc C, Poucet B (1998) Spatial firing of hippocampal place
cells in blind rats. J Neurosci 18(5): 1818-1826
Scarf D, Hayne H, Colombo M (2011) Pigeons on par with primates in numerical
competence. Science 334: 1664.
Schenkman BN (1986) Identification of ground materials with the aid of tapping sounds and
vibrations of long canes for the blind. Ergonomics 29: 985-998.
Schenkman BN, Nilsson ME (2010) Human echolocation: blind and sighted persons' ability
to detect sounds recorded in the presence of a reflecting object. Perception 39: 483501.
Schenkman BN, Nilsson ME (2011) Human echolocation: pitch versus loudness information.
Schmidtke D, Esser K-H (2011) Sex matters in echoacoustic orientation: gender differences
in the use of acoustic landmarks in Phyllostomus discolor (lesser spear-nosed bat). J
Comp Physiol A 197: 531-539.
!
117!
Schmitz S (1999) Gender differences in acquisition of environmental knowledge related to
wayfinding behavior, spatial anxiety and self-estimated environmental competencies.
Sex Roles 41: 71-93.
Schmitzer-Torbert N (2007) Place and response learning in human virtual navigation:
behavioral measures and gender differences. Behav Neurosci 121: 277-290.
Schnitzler H, Moss CF, Denzinger A (2003) From spatial orientation to food acquisition in
echolocating bats. Trends Ecol Evol 18: 386-394.
Schregardus DS, Pieneman AW, ter Maat A, Jansen RF, Brouwer TJF, Gahr ML (2006) A
lightweight telemetry system for recording neuronal activity in freely behaving small
animals. J Neurosci Meth 155: 62-71.
Schwitzgebel E, Gordon MS (2000) How well do we know our own conscious experience?
The case of human echolocation. Philosophical Topics 28: 235-246.
Shepard RN, Cooper L. (1982) Mental images and their transformations. Cambridge: MIT
Press.
Shepard RN, Metzler J (1971) Mental rotation of three-dimensional objects. Science 972:
701-703.
Siemers BM, Schauermann G, Turni H, Merten Svon (2009) Why do shrews twitter?
Communication or simple echo-based orientation. Biol Lett 5: 593-596.
Silverman I, Eals M (1992) Sex differences in spatial abilities: evolutionary theory and data.
In: Cosmides L, Tooby J, editors. The adapted mind: evolutionary psychology and the
generation of culture. New York: Oxford University Press. pp. 533-549.
Simon H, Gilden D, Brabyn J, Lotze A, Levitt H (2012) Acoustic cues used by blind
travelers. J Acoust Soc Am 131: 3516.
Singer AC, Karlsson MP, Nathe AR, Carr MF, Frank LM (2010) Experience-dependent
development of coordinated hippocampal spatial activity representing the similarity of
related locations. J Neurosci 30: 11586-11604.
Snyder LH, Grieve KL, Brotchie P, Andersen RA (1998) Separate body- and worldreferenced representations of visual space in parietal cortex. Nature 394(6696): 887891.
Solstad T, Boccara CN, Kropff E, Moser MB, Moser EI (2008) Representation of geometric
borders in the entorhinal cortex. Science 322: 1865-1868.
Speed MP, Ruxton GD (2005) Warning displays in spiny animals: one (more) evolutionary
route to aposematism. Evolution 59: 2499-2508.
Spelke ES, Lee SA (2012) Core systems of geometry in animal minds. Phil Trans R Soc B
367: 2784-2793.
Stanley (2006) Toward adapting spatial audio displays for use with bone conduction: the
cancellation of bone-conducted and air-conducted sound waves. Master Thesis.
Georgia Institute of Technology: USA.
Steck SD, Mallot HA (2000) The role of global and local landmarks in virtual environment
navigation. Presence (Camb) 9: 69-83.
!
118!
Stevenson RJ, Case TI (2005) Olfactory imagery: a review of its phenomenological,
experimental and theoretical bases. Psychon Bull Rev 12: 244-264.
Stiebler I, Ehret G (1985) Inferior colliculus of the house mouse. I. A quantitative study of
tonotopic organization, frequency representation, and tone‐threshold distribution. J
Comp Neurol 238: 65-76.
Stiedl O, Pieneman AW, Gutzen C, Schwarzer S, Jansen RF (2010) Fear conditioning in an
automated home cage (DualCage) environment. In: Spink AJ, Grieco F, Krips OE,
Loijens LWS, Noldus LPJJ, Zimmermann PH, editors. Proceedings of Measuring
Behavior. pp. 83-84.
Stimpert AK, Wiley DN, Au WWL, Johnson MP, Arsenault R (2007) 'Megapclicks': acoustic
click trains and buzzes produced during night-time foraging of humpback whales
(Megaptera novaeangliae). Biol Lett 3: 467-470.
Stoffregen TA, Pittenger JB (1995) Human echolocation as a basic form of perception and
action. Ecol Psychol 7: 181-216.
Strelow ER (1985) What is needed for a theory of mobility: direct perception and cognitive
maps-lessons from the blind. Psychol Rev 92: 226-248.
Strelow ER, Brabyn JA (1982) Locomotion of the blind controlled by natural sound cues.
Struiksma ME, Noordzij ML, Postma A (2009) What is the link between language and spatial
images? Behavioral and neural findings in blind and sighted individuals. Acta Psychol
(Amst) 132: 145-156.
Supa M, Cotzin M, Dallenbach K (1944) "Facial vision": the perception of obstacles by the
blind. Am J Psychol 57: 133-183.
Taube JS, Muller RU, Ranck JB Jr. (1990) Head-direction cells recorded from the
postsubiculum in freely moving rats, I: description and quantitative analysis. J
Tavolga WN (1971a) Sound production and detection. In: Hoar WS, Randall DJ, editors. Fish
Physiology, Vol. 5. New York: Academic Press, Inc. pp. 135-205.
Tavolga WN (1971b) Acoustic orientation in the sea catfish, galeichthys felis. Ann NY Acad
Sci 188: 80-97.
Tavolga WN (1976) Acoustic obstacle detection in the sea catfisch (Arius felis). In: Schuijf A,
Hawkins AD, editors. Sound reception in fish. Amsterdam: Elsevier. pp. 185-204.
Tavolga WN (1977) Mechanisms for directional hearing in the sea catfish (Arius felis). J Exp
Biol 67: 97-115.
Tavolga WN (1982) Auditory acuity in the sea catfish (Arius felis). J Exp Biol 96: 367-376.
Teng S, Whitney D (2011) The acuity of echolocation: Spatial resolution in the sighted
compared to expert performance. J Vis Impair Blind 105: 20-32.
Thinus-Blanc C, Gaunet F (1997) Representation of space in blind persons: vision as a spatial
sense? Psychol Bull 121: 20-42.
!
119!
Thompson WL, Kosslyn SM, Hoffman MS, Van der Kooij K (2008) Inspecting visual mental
images: can people “see” implicit properties as easily in imagery and perception?
Mem Cognit 36: 1024-1032.
Thuen F, Klepp KI, Wold B (1992) Risk-seeking and safety-seeking behaviours: a study of
health-related behaviours among Norwegian school children. Health Educ Res 2: 269276.
Tinti C, Adenzato M, Tamietto M, Cornoldi C (2006) Visual experience is not necessary for
efficient survey spatial cognition: evidence from blindness. Q J Exp Psychol (Hove)
59: 1306-1328.
Tittle JS, Todd JT, Perotti VJ, Norman JF (1995) The systematic distortion of perceived 3D
structure from motion and binocular stereopsis. J Exp Psychol Human 21: 663-678.
Tobler WR (1994) Bidimensional regression, Geogr Anal 29: 169-186.
Todd JT, Bressan P (1990) The perception of 3-dimensional affine structure from minimal
apparent motion sequences. Percept Psychophys 48: 419-430.
Todd JT, Norman JF (1991) The visual perception of smoothly curved surfaces from minimal
apparent motion sequences. Percept Psychophys 50: 509–523.
Tolman EC (1948) Cognitive maps in rats and men. Psychol Rev 55: 189-208.
Townsend PK (1995) The eleven-dimensional supermembrane revisited. Phys Lett B 350:
184-187.
Trullier O, Wiener SI, Berthoz A, Meyer JA (1997) Biologically based artificial navigation
systems: review and prospects. Prog Neurobiol 51: 483-544.
Ulanovsky N (2011) Neuroscience: how is three-dimensional space encoded in the brain?
Curr Biol 21: R886-888.
Ulanovsky N, Moss CF (2007) Hippocampal cellular and network activity in freely moving
echolocating bats. Nat Neurosci 10: 224-233.
Ulanovsky N, Moss CF (2011) Dynamics of hippocampal spatial representation in
echolocating bats. Hippocampus 21: 150-161.
Ungar S (2000) Cognitive mapping without visual experience. In: Kitchin R, Freundschuh S,
editors. Cognitive mapping: past present and future. London: Routledge. pp. 221-248.
Vanlierde A, Wanet-Defalque M-C (2004) Abilities and strategies of blind and sighted
subjects in visuo-spatial imagery. Acta Psychol (Amst) 116: 205-222.
Vecchi T, Tinti C, Cornoldi C (2004) Spatial memory and integration processes in congenital
blindness. Neuroreport 15: 2787-2790.
von der Emde G, Ringer T (1992) Electrolocation of capacitive objects in four species of
pulse-type weakly electric fish. I. Discrimination performance. Ethology 91: 326-338.
Vonk J, Beran MJ (2012) Bears ‘count’ too: quantity estimation and comparison in black
bears, Ursus americanus. Anim Behav 84(1): 231-238.
Warburton K. (1990) The use of local landmarks by foraging goldfish. Anim Behav 40: 500505.
!
120!
Ward SL, Newcombe N, Overton WF (1986) Turn left at the church, or three miles north.
Environ Behav 18: 192-213.
Warren DH. 1994. Blindness and children - An individual differences approach. Cambridge:
Cambridge University Press.
Waterman S, Gordon D (1984) A quantitative-comparative approach to analysis of distortion
in mental maps. Prof Geogr 36: 326-337.
Weeks R, Horwitz B, Aziz-Sultan A, Tian B, Wessinger CM, Cohen LG, Hallett M,
Rauschecker JP (2000) A positron emission tomographic study of auditory
localization in the congenitally blind. J Neurosci 20: 2664-2672.
Werner S, Krieg-Brückner B, Mallot HA, Schweizer K, Freksa C (1997) The role of
landmark, route, and survey knowledge in human and robot navigation. In: Jarke M,
Pasedach K, Pohl K, editors. Informatik ’97. Berlin: Springer-Verlag. pp. 41-50.
Westheimer G (1965) Visual acuity. Annu Rev Psychol 16: 359-380.
Wiener SI (1993) Spatial and behavioral correlates of striatal neurons in rats performing a
self-initiated navigation task. J Neurosci 13: 3802-3817.
Wiener SI (1996) Spatial, behavioural and sensory correlates of hippocampal CA1 complex
spike cell activity: implications for information processing functions. Prog Neurobiol
49: 335-361.
Williams CL, Barnett AM, Meck WH (1990) Organizational effects of early gonadal
secretions on sexual differentiation in spatial memory. Behav Neurosci 104: 84-97.
Williams CL, Meck WH (1991) The organizational effects of gonadal steroids produce
sexually dimorphic spatial ability. Psychoneuroendocrinology 16: 157-177.
Williams CL, Meck WH (1993) Organizational effects of gonadal hormones induce
qualitative differences in visuospatial navigation. In: Haug M, Whalen RE, Aron C,
Olsen KL, editors. The development of sex differences and similarities in behavior.
Dordrecht: Kluwer Academic. pp. 179-185.
Witten E (1995) String theory dynamics in various dimensions. Nucl Phys B 443: 85-126.
Wong M, Gnanakumaran V, Goldreich D (2011) Tactile spatial acuity enhancement in
blindness: evidence for experience-dependent mechanisms. J Neurosci 31: 7028-7037.
Wright JJ (2009) Diversity, phylogenetic distribution, and origins of venomous catfishes.
BMC Evol Biol 9: 282.
Yartsev MM, Witter MP, Ulanovsky N (2011) Grid cells without theta oscillations in the
entorhinal cortex of bats. Nature 479(7371): 103-107.
Yoo SS, Freeman DK, McCarthy JJ 3rd, Jolesz FA (2003) Neural substrates of tactile
imagery: a functional MRI study. Neuroreport 14: 581-585.
Yoo SS, Lee CU, Choi BG (2001) Human brain mapping of auditory imagery: event-related
functional MRI study. Neuroreport 12: 3045-3049.
!
121!
Zeddies DG, Fay RR, Gray PW, Alderks AA, Acob A, Sisneros JA (2012) Local acoustic
particle motion guides sound-source localization behavior in the plainfin midshipman
fish, Porichthys notatus. J Exp Biol 215: 152-160.
Zelick R, Mann DA, Popper AN (1999) Acoustic communication in fishes and frogs. In: Fay
RR, Popper AN, editors. Comparative hearing: fish and amphibians. New York:
Springer-Verlag. pp. 363-411.
Zipser D, Andersen RA (1988) A back-propagation programmed network that simulates
response properties of a subset of posterior parietal neurons. Nature 331(6158): 679684.
Zuckerman M (1994) Behavioral expressions and biosocial bases of sensation seeking. New
York: Cambridge University Press.
Zuckerman M, Eysenck S, Eysenck HJ (1978) Sensation seeking in England and America:
cross-cultural, age, and sex comparisons. J Consult Clin Psychol 46: 139-149.
Zuckerman M. (1979) Sensation seeking: beyond the optimal level of arousal. Hillsdale:
Lawrence Erlbaum.
Zwiers MP, Van Opstal AJ, Cruysberg JR (2001) A spatial hearing deficit in early-blind
humans. J Neurosci 21: RC142:1-RC142:5.
!
!
!
122!
!
123!
Acknowledgements
I thank PD DR. KARL-HEINZ ESSER for the great opportunity to conduct the studies for this
thesis in his working group, his support and his confidence. I further thank my co-supervisors
PROF. DR. ECKART ALTENMÜLLER and APL. PROF. DR. MANUELA GERNERT for motivating
and inspiring discussions and the evaluation of this thesis.
I would also like to thank my colleagues SÖNKE VON DEN BERG, ARNE LIEBAU, DR. SANDRA
AMMERSDÖRFER, and MATHIAS VOIGT for inspiration, intense discussions, and their warm
support, as well as students from our lab, namely SARAH GALINSKI, BIANCA BRETSCHNEIDER,
AMELIE STALLFORTH (Leibniz University, Hannover, Germany), VIVIAN HO, and REEM
SANADIKI (Northeastern University, Boston, USA) for their helping hands.
Further, I would like to acknowledge and thank all my cooperation partners:
PROF. DR. MANFRED L. GAHR, DR. ANDRIES
TER
MAAT, and LISA TROST (Max-Planck-
Institute for Ornithology, Seewiesen, Germany) for inviting me to their lab, for providing me
with the telemetric devices used for the study presented in chapter 2 of this thesis, for giving
me the opportunity to present my work in the Biological Colloquium of the Max-PlanckInstitute for Ornithology, as well as for their hospitality during my research stays in
Seewiesen.
PROF. DR. OLIVER STIEDL, RENÉ JANSEN, PH.D., and ANTON W. PIENEMAN (Center for
Neurogenomics and Cognitive Research, University of Amsterdam, Amsterdam, Netherlands)
for inviting me to their lab, for giving me the opportunity to present my work in the
Neuroscience Colloquium at the University of Amsterdam, and for their valuable advice
concerning the neuronal recordings from the hippocampus in freely behaving mice.
PROF. DR. HOLGER LUBATSCHOWSKI, DR. TAMMO RIPKEN, and MARIO SIMONS (Laser
Zentrum Hannover e.V., Hannover, Germany) for giving me access to the exiplex laser and
the scanning electron microscope used for the development of the electrode-manufacturing
protocol presented in chapter 2 of this thesis.
BERND PETERSEN, MARTIN BAASKE, and ULRIKE KRÜGER (Landesbildungszentrum für
Blinde,
Hannover,
Germany),
as
well
as
GERD
SCHWESIG
(Blinden
und
!
124!
Sehbehindertenverband Niedersachsen, Hannover, Germany) for their help in subjectacquisition for the study presented in chapter 3. I would also like to thank the 47 ANONYMOUS
SUBJECTS
for their time and effort in participating in the study.
PROF. DR. DR. H.C. JÖRG HARTUNG and DR. JOCHEN SCHULZ (Institute for Animal Hygiene,
Animal Welfare and Behaviour of Farm Animals, University of Veterinary Medicine,
Hannover, Germany) for providing me with a large-scale, temperature-controlled room to set
up the swimming pool used in the study presented in chapter 4 and for bearing the incurred
heating and water costs.
PROF. DR. URSULA SIEBERT and MICHAEL DÄHNE (Institute for Terrestrial and Aquatic
Wildlife Research, University of Veterinary Medicine, Hannover, Germany) for the
calibration of the hydrophone used in the study presented in chapter 4, as well as PROF. DR.
GÜNTHER EHRET (Institute for Neurobiology, University of Ulm, Germany) for kindly
providing the hydrophone.
I further thank the STUDIENSTIFTUNG DES DEUTSCHEN VOLKES for project funding in form of
a Ph.D. scholarship as well as a scholarship for studies abroad and PROF. DR. BERND
OPPERMANN (Institute of German and European Civil and Commercial Law, Leibniz
University, Hannover, Germany; liaison professor of the Studienstiftung des Deutschen
Volkes) for his continuous support.
In the end I would like to deeply thank my FRIENDS, my FAMILY, and, above all, my parents
IRMHILD and SIEGFRIED SCHMIDTKE, for constantly supporting me and keeping me grounded
as well as for their patience and for being there for me.
!
!
125!
Supporting Material
Supporting*Fig.*1.!Individual!of!Ariopsis'seemanni!(Tete!sea!catfish).
!

University of Veterinary Medicine THESIS

Transcription

Similar documents

PowerPoint Sunusu

FERNANDEZ STOHANZL PORTFOLIO 2014.indd

SOMEWHAT MORE THAN NOTHING”

Professor György KOCZISZKY, PhD E-mail: regkagye@uni

Entreprise Location Intelligence Brochure

City of Collinsville historic Landmarks

Towards a Danish Spatial Information Infrastructure

SES-300-XXX-ATEX eng

Creekside Connection - Creekside Church of the Brethren