- 1 - Running head: Spatial and empathic perspective

Social Cognition, Vol. 33, No. 3, 2015, pp. 187–210
ERLE AND TOPOLINSKI
Spatial and empathic perspective-taking
Spatial and Empathic Perspective-Taking
Correlate on a Dispositional Level
Thorsten M. Erle
University of Würzburg
Sascha Topolinski
University of Cologne
Empathic perspective-taking is inevitably described using spatial locatives (“understanding a point of view”) and involves a metaphorical
“merging of the self and other.” Visuo-spatial perspective-taking involves an actual embodied merging of the self and other. Yet, the two
kinds of perspective-taking are treated as independent of each other.
Four studies found that these two kinds of perspective-taking are related on a dispositional level. Individuals with high levels of empathic
perspective-taking were better at imagining different views of the same
object (Studies 1–3). This correlation could not be explained by general
intelligence (Study 2) and was nominally stronger among participants
who explicitly reported that they engaged in spatial perspective-taking
(Study 3). Study 4 showed a correlation between empathic perspective-taking and performance on another spatial perspective-taking task.
These relations can be explained by a shared mechanism of embodied
self-rotation during all kinds of perspective-taking.
Being able to tune into the thoughts and feelings of our interaction partners is a main component of empathy (i.e., perspective-taking; Davis, 1983)
and an important aspect of social life. But what does it take to be empathic?
It seems to involve spatial perspective taking, since in linguistic metaphors
we cannot speak about empathizing without using spatial locatives, such
The first author is an associated member of and would like to thank the research training group
RTG: 1253/2 (Emotions) which is supported by the German Research Foundation (DFG) by federal
and Länder funds. We thank Amelie Backfisch, Giti Bakhtiari, Niklas Barth, Jönna Hedder, Friederike
Kälke, and Dimitrios Kourtoglou for help with the data collection.
Address correspondence to Thorsten M. Erle, Department of Psychology II, University of
Würzburg, Röntgenring 10, 97070 Würzburg, Germany; E-mail: [email protected].
© 2015 Guilford Publications, Inc.
187
188ERLE AND TOPOLINSKI
as when we “put ourselves into another person’s shoes” or understand a
“point of view” (cf. Lakoff & Johnson, 1980). Thus, simulations of another
psychological perspective might be grounded in visual simulations of perspective (cf. Barsalou, 1999). From a functional perspective, this assumption makes sense as well: both empathic (Batson, Sager, et al., 1997; Davis,
Conklin, Smith, & Luce, 1996; Galinsky, Ku, & Wang, 2005) and visuo-spatial perspective-taking (Michelon & Zacks, 2006; Zacks & Michelon, 2005;
Zacks, Vettel, & Michelon, 2003) hinge on establishing a common reference
frame to simulate a mental state within another person. To empathize we
need to simulate “how something feels” and for visuo-spatial perspectivetaking we need to simulate “how something looks” (Level-2 perspectivetaking; cf. Flavell, Everett, Croft, & Flavell, 1981) for another person.
This functional similarity raises the question whether these simulations
also share a common mechanism. The underlying processes of spatial
perspective-taking have been demonstrated most clearly by Kessler and
Thomson (2010). When adopting the spatial perspective of another person,
people transform their body schema into the target’s perspective—literally
putting their body in the target’s place. Consequently, by physically rotating the perspective-taker toward or away from the target, spatial perspective-taking becomes easier or more difficult. Therefore, spatial perspective
taking is a strongly embodied process (cf. Kessler & Rutherford, 2010; Kessler & Thomson, 2010; for the general notion of embodiment see, e.g., Glenberg, 2010; Topolinski, 2010; Topolinski & Strack, 2009, 2010). Note that
this process is also employed during level-2 visual perspective-taking (see
Surtees, Apperly, & Samson, 2013a, 2013b), which is why these terms are
used interchangeably in the present context.
Although the mechanisms of spatial perspective-taking have been thoroughly investigated, only little is known about its psychological consequences. However, previous research also hints at a link between visuospatial and psychological or empathic forms of perspective-taking. For
instance, visuo-spatial perspective-taking and social skills, a measure of
general empathic functioning (exemplary item: “I enjoy social occasions”;
cf. Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001), are positively correlated (e.g., Brunyé et al., 2012; Shelton, Clements-Stephens,
Lam, Pak, & Murray, 2012; Thakkar, Brugger, & Park, 2009). Furthermore,
autism spectrum disorder—an extreme case of social and empathic dysfunction—has been associated with deficits in visuo-spatial perspectivetaking (Hamilton, Brindley, & Frith, 2009). Finally, differences in social
skills predict how spatial perspective problems are approached (Kessler &
Wang, 2012). People with low social skills engage in the abovementioned
embodied transformation during spatial perspective-taking only to a lesser
degree. Therefore, for people with low social skills, these tasks could be
embodied differently (e.g., as the rotation of only some parts of the body),
they could be solved by object-rotation rather than mental self-rotation
strategies, or completely rule-based strategies. This research demonstrates
Spatial and empathic perspective-taking189
a link between generally high functioning empathy and visuo-spatial perspective-taking skills. It does not, however, speak to the relation between
different kinds of perspective-taking, although it seems plausible to assume
that empathic perspective-taking also contributes to high social skills.
For empathic perspective-taking the research landscape looks diametrically different. The effects of empathic perspective-taking have received
ample empirical attention but relatively little is known about their underlying mechanisms. In empathic perspective-taking research, participants
usually learn about a fictional character and then are instructed to “please
imagine how [the target] feels. Try to take her perspective in the interview,
imagining how she is feeling about what is happening. While you watch
and listen to her, picture to yourself just how she feels. In your mind’s eye,
visualize clearly and vividly her reactions in this situation” (the imaginetarget/other instruction; for other instructions, see Davis et al., 2004, p. 1628;
see also Batson, Early, & Salvarani, 1997). It has been shown that such instructions lead to a plethora of important social outcomes. For instance,
perspective-taking leads to increased prosocial behavior (e.g., Batson,
Chang, Orr, & Rowland, 2002) or decreased prejudices (e.g., Galinsky &
Moskowitz, 2000), and most importantly the adjustment of own thoughts
(Epley, Keysar, Van Boven, & Gilovich, 2004), feelings (Batson, Early, & Salvarani, 1997), or even the view of oneself (Galinsky, Wang, & Ku, 2008)
toward those of the perspective-taking target. As the main mechanism of
psychological perspective-taking effects, a “merging between the self and
the other” is assumed (see, e.g., Davis et al., 1996).
One important limitation of such experiments is that they all focus on
outcomes that are measured after perspective-taking allegedly occurred.
This is not to say that these works do not focus on underlying mechanisms
of perspective-taking in some way, but to put it more accurately, they focus
on the underlying mechanisms, which are engaged once perspective-taking occurred. This leaves the important question of what actually happens
while we imagine a different psychological perspective completely unanswered. For instance, a seminal chapter by Epley and Caruso (2009) describes this initial step simply as “activating perspective-taking” (p. 298),
focusing on why or why not people engage in perspective-taking, while
the important question of how we adopt the psychological perspective of
another remains untapped. The same is true for the social skills measure
(Baron-Cohen et al., 2001) which measures empathic functioning in general, but does not speak to the processes engaged within a specific empathic
perspective-taking episode. The present work seeks to shed light on the
underlying mechanisms of psychological perspective simulations per se,
and not the psychological trickle-down consequences of these processes.
Although the conceptual metaphor would suggest so, the embodied selfother merging that happens during spatial perspective-taking has never
been related to psychological self-other merging in empathic perspective-
190ERLE AND TOPOLINSKI
taking. Rather, visual and empathic perspective-taking are strongly differentiated in social psychology. It is argued that although both kinds of
perspective-taking often make parallel predictions, they are subserved by
different mechanisms (Libby & Eibach, 2011, p. 225f). Indeed, there exists
one paper showing that instructions to engage in visual versus empathic
perspective-taking can lead to different outcomes on the same measure
(Fiske, Taylor, Etcoff, & Laufer, 1979).
Much like in empathic perspective-taking research, participants are instructed to take different perspectives in social psychological research on
visual perspective-taking, too (cf. Libby & Eibach, 2011, p. 189; see also
Kross & Ayduk, 2011). A large body of research demonstrated that imagining events from a different visual perspective can change representations
of these events (Libby, Shaeffer, & Eibach, 2009), emotions (Valenti, Libby,
& Eibach, 2011), and even behavior (Libby, Shaeffer, Eibach, & Slemmer,
2007). Although these effects are framed as specific to the visual aspects of
the simulation, they are not related to the visual perspective-taking mechanisms established in cognitive psychology (Kessler & Thomson, 2010) and
the question of what actually happens during the simulation process remains unanswered again. Even Fiske et al. (1979) state in their general discussion: “the question remains open, that is, whether role-taking in general
combines aspects of both imaging and empathy” (Fiske et al., 1979, p. 374).
We agree with this conclusion and would like to add that it is a very strong
assertion that instructions such as “take a perspective in a situation,” “picture to yourself how somebody feels,” or “in your mind’s eye, visualize
clearly and vividly somebody’s reactions in a situation” are not supplemented by any visual input whatsoever. It is an assumption, for which
direct evidence is lacking, too.
While we agree that different perspective-taking instructions can lead to
different outcomes on the same task (cf. Batson, Early, & Salvarani, 1997;
Fiske et al., 1979), we argue that the distinction between empathic and visual perspective-taking mechanisms in social psychology is premature.
These presently used instruction manipulations are unable to tease visual
and psychological simulation aspects apart and it seems quite impossible
to instruct a person to “only imagine visual, but not psychological aspects
of another person” or vice versa. We argue that simulations of both psychological and visual perspectives rely on embodied transformations of
the self into the other’s position (cf. Kessler & Rutherford, 2010; Kessler
& Thomson, 2010). This embodied self-rotation is one mechanism, which
quite literally contributes to the “merging of the self and other,” which is
presumably the main mechanism behind empathic perspective-taking effects (see, e.g., Batson, Chang, Orr, & Rowlands, 2002; Batson, Polycarpou,
Harmon-Jones, Imhoff, Mitchener, Bednar et al., 1997; Batson, Sager, Garst,
Kang, Rubchinsky, & Dawson, 1997; Davis et al., 1996). Since this mechanism is shared between the two kinds of perspective-taking, visuo-spatial
and empathic perspective-taking correlate on a dispositional level.
Spatial and empathic perspective-taking191
Data Analysis
For Study 1, sample size was determined heuristically since the expected
effect size was unknown. We ran the study as part of a data collection campaign that was scheduled to gather N = 120 participants; N = 115 showed
up. For the remainder, we determined sample sizes with a priori power
analyses using G*Power (Faul, Erdfelder, Lang, & Buchner, 2007) targeting
a power of (1-β) = .80 in a one-tailed test based on the effect of Study 1. For
Studies 2 and 3 this amounted to N = 144 participants. Study 4 was a reanalysis of an existing dataset. All exclusion criteria and measures used are
disclosed. All psychometric tests were analyzed according to their respective manuals (cf. Liepmann, 2007; Stumpf & Fay, 1983). Although our hypotheses are clearly one-tailed, reported p-values correspond to two-tailed
tests (as per convention).
Study 1
The first study was designed to test our general hypothesis. As a measure
of empathic perspective-taking, we used the German version (Paulus,
2009) of the Interpersonal Reactivity Index (IRI; Davis, 1983). This scale
is a multidimensional approach to empathy, comprising three emotional
subscales—Empathic Concern (EC; e.g., “I often have tender, concerned
feelings for people less fortunate than me”), Fantasy (FS; “I daydream and
fantasize, with some regularity, about things that might happen to me”),
and Personal Distress (PD; “In emergency situations, I feel apprehensive
and ill-at-ease”)—as well as one cognitive subscale, Perspective-Taking
(PT; “I sometimes try to understand my friends better by imagining how
things look from their perspective”). The IRI is the most frequently used
measure of empathy for which a validated German translation exists. Furthermore, its factorial structure makes it possible to establish concurrent
validity (i.e., a correlation between empathic and visuo-spatial perspective-taking), while simultaneously showing discriminant validity toward
the other subscales of empathy.
PT was the focal dependent variable in the present studies. This cognitive subscale refers to the willingness or frequency with which participants
engage in empathic perspective-taking in everyday life. The original author of the IRI considered PT as a measure of both motivation and skill
to engage in empathic perspective-taking (Davis, 1983, p. 115), although
the items arguably refer more closely to the motivational aspects of empathy. Compared to this, prior research has focused on measures related
to empathic “performance” (i.e., social skills, cf. Baron-Cohen et al., 2001).
This measure, however, describes successful empathic adjustment overall,
which can come about by means different from perspective-taking. PT as a
measure of perspective taking in particular comes much closer to the idea
of embodied self-other merging. Furthermore, prior research has shown
192ERLE AND TOPOLINSKI
that motivation for accurate empathic perspective-taking eliminates performance differences between individuals with allegedly high versus low
empathic accuracy (see, e.g., Hall & Schmid Mast, 2008; Klein & Hodges,
2001). Finally, in general, skill and motivation usually are highly correlated
and this should feed into the interpretation of the PT items. Therefore, the
subscale PT was deemed more appropriate for the present research question than the social skills subscale of the Autism Quotient (Baron-Cohen et
al., 2001), although such a measure likely contributes independently to the
relation between empathic and visuo-spatial perspective-taking.
Both empathic concern and personal distress describe feelings, which follow from empathic perspective-taking. These subscales are often of interest
in studies looking at the consequences of perspective-taking, for instance
on the distinction between imagine-self and imagine-other empathizing
instructions (see, e.g., Batson, Early, & Salvarani, 1997; Davis et al., 2004;
Lamm, Batson, & Decety, 2007). The focus of the present work is the simulation preceding such consequences, and this should correlate to the cognitive aspect of empathy, as indicated by the subscale PT and not these
emotional facets of empathy, which are temporally subsequent (the same
is true for, e.g., the social skills subscale of the Autism Quotient, see BaronCohen et al., 2001).
The subscale Fantasy arguably comes closer to PT than empathic concern
and personal distress. However, not by coincidence, it is also conceptualized as a facet of affective empathy, which is contrasted to cognitive empathy (i.e., PT) in the IRI (see Davis, 1980, 1983). The ability to appreciate the
character of the protagonist in a movie or book differs in at least three important regards from everyday perspective-taking. First, whereas perspective-taking is a cognitively demanding process, narrative transportation is
more experiential. People actively try to understand another person, but
they are passively drawn into an elegant narrative. Second, and related,
perspective-taking happens in the here and now—the real world. Fantasizing, on the other hand, refers to imagining oneself into another alternative
world and being aware of this. Third, perspective-taking depends solely
on the perspective-taker and the target of perspective taking, whereas fantasizing necessarily is also a function of the surrounding plot and many
other aspects (for a comprehensive model, see Van Laer, de Ruyter, Visconti, & Wetzels, 2014, p. 801). These differences are well documented in
literature on narrative transportation (for a recent review, see Van Laer et
al., 2014). Taken together, these differences question how well it should
relate to general visuo-spatial abilities.
As a measure of spatial perspective-taking we used the Tube Figures Test
(TFT; Stumpf & Fay, 1983), a well-established German test of general visuo-spatial intelligence. For instance, it is a mandatory part of the training
of prospective pilots and physicians in Germany, as well as of many assessment centers in larger companies. On every trial, participants are presented with a long curved tube inside of a transparent cube, the original
Spatial and empathic perspective-taking193
Table 1. Means of the Measures Used Across Three Studies
Study 1
(N = 115)
Study 2
(N = 147)
Study 3
(N = 169)
Study 4
(N = 163)
IRI (total score)
55.55 (7.96)a
54.23 (7.68)b
54.46 (7.71)
56.65 (6.91)
Empathic Concern
14.92 (2.61)a
14.49 (2.84)b
14.95 (2.65)
15.11 (2.42)
Fantasy
14.37 (3.40)
13.95 (3.23)
b
13.89 (3.53)
15.18 (2.85)
Personal Distress
11.39 (3.28)a
11.23 (3.32)b
11.27 (3.37)
11.60 (3.12)
Perspective-Taking
14.87 (2.83)
a
14.55 (2.74)
b
14.35 (3.22)
14.76 (2.96)
Tube Figures Test
10.41 (3.24)
11.27 (3.51)
10.46 (3.64)
—
—
11.40 (3.87)
—
—
Arithmetic Problems
a
Notes. SD in parentheses; IRI = Interpersonal Reactivity Index; aN = 113; bN = 120.
figure. Additionally, this original figure is simultaneously presented from
another side (left, right, top, bottom, or behind) in a second picture, the
target figure. Participants have to indicate the angle from which the target
depicts the original figure.1 The TFT is not a clear-cut measure of perspective-taking. It can alternatively be solved using object rotation strategies.
Although dissociations between these different spatial abilities have been
shown experimentally (see, e.g., Hamilton et al., 2009), they are often highly correlated in psychometric tests (r > .80; Hegarty & Waller, 2004). The
TFT has the additional benefit that its stimulus material renders the use of
non-spatial strategies unlikely—non-spatial strategies that can be applied
to solve other visuo-spatial tests such as the mental rotation test (Vandenberg & Kuse, 1978).
We predicted a positive correlation between performance in the TFT and
the subscale PT but no other subscale of the IRI. We discuss the matter of
object rotation versus perspective-taking within the TFT further in Study 3.
Method
Sample. N = 115 individuals (N = 80 female, N = 33 male, N = 2 missing;
Mage = 27, SD = 7) participated for €7. The study was the first part of a longer battery including other unrelated tasks (Topolinski, Erle, & Reber, 2015;
Topolinski, Maschmann, Pecher, & Winkielman, 2014). N = 4 participants
were excluded due to below-chance performance in the TFT; the IRI data
of N = 2 was lost due to technical problems.
Materials and Procedure. Participants completed the German version of the
IRI and the TFT in counterbalanced order.2 Both tests were administered as
stated in the respective manuals with the only difference that administration was computer-based.
1. Because the TFT is protected by copyright, we are not allowed to include pictures of the
material. Comparable stimuli can be found at: http://www.testedich.de/quiz23/quiz/1179643508/
Raeumliches-Vorstellungsvermoegen.
2. The order of measurement was completely counterbalanced in the first three studies. Since
across studies order did not affect any outcome, it is not discussed henceforth.
194ERLE AND TOPOLINSKI
Table 2. Reliability (CR-α) of the Measures Used Across Three Studies
Study 1
(N = 115)
Study 2
(N = 147)
Study 3
(N = 169)
Study 4
(N = 163)
IRI (total score)
.76a
.76b
.74
.71
Empathic Concern
.54a
.67b
.62
.59
Fantasy
.76
.72
b
.77
.67
Personal Distress
.70a
.74b
.73
.66
Perspective-Taking
7a
.6
.78
b
.83
.77
Tube Figures Test
.58
.75
.68
—
Arithmetic Problems
—
.84
—
—
a
Notes. IRI = Interpersonal Reactivity Index; aN = 113; bN = 120.
IRI. The German version of the IRI consists of 16 items (4 for each subscale) and has a 5-point response scale (cf. Paulus, 2009). We use the aggregate scores for each of the scales as dependent variables.
TFT. Participants received a maximum of 21 TFT problems. They could
work on each item self-paced but could not browse back to an earlier item.
After 12 minutes, the task was automatically stopped (cf. Stumpf & Fay,
1983). The number of correct answers is the main dependent variable of the
TFT3 but we also recorded mean response time.
Results and Discussion
Secondary Analyses. Table 1 shows the mean scores of all measures used
across the present studies, and Table 2 their reliabilities (CR-α), which were
poor to good in Study 1. Mean RT in the TFT correlated positively with
performance, r(109) = .30, p = .001, 95% CI [.13, .50]: the longer it took participants to solve the task, the higher their overall performance was.4 All
subscales of the IRI were correlated with the total score, all rs(111) > .56, p
< .001.
Main Analysis and Discussion. As predicted, there was a significant positive correlation between visuo-spatial abilities and empathic perspectivetaking, r(107) = .20, p = .033, 95% CI [.02, .40]. Participants who performed
better at the TFT indicated more empathic perspective-taking. The other
subscales of the IRI, however, did not correlate with TFT performance, all
|r|s(107) < .10, ps > .33.
3. The TFT manual suggests absolute number rather than the percentage of correct answers as the
dependent variable of the test (Stumpf & Fay, 1983). There was, however, a small number of people
who did not complete all 21 items in time (15 out of 431 participants; 3.48%). The correlation between
percentage of correct answers on the TFT and empathic PT was almost identical to the correlation
between absolute number of correct answers and empathic PT (Study 1: r(107) = .22, p = .025; Study 2:
r(118) = .20, p = .026; Study 3: r(165) = .22, p = .004; meta-analysis: r(394) = .21, p < .001).
4. Since accuracy and RT of the TFT were correlated, we also computed a partial correlation
between spatial and empathic perspective-taking, controlling for the RT on the spatial perspectivetaking task. This correlation was almost identical to the bivariate correlation: Study 1: r(106) = .18, p =
.063; Study 2: r(117) = .22, p = .014; Study 3: r(164) = .20, p = .009; meta-analysis: r(393) = .20, p < .001.
Spatial and empathic perspective-taking195
Given that the TFT is a measure of visuo-spatial intelligence, it might be
possible that the present correlation simply means that more intelligent
people are also more empathic. Study 2 ruled this out by establishing discriminative validity regarding another aspect of intelligence, namely numerical intelligence.
Study 2
Study 1 was replicated with the inclusion of a measure of numerical intelligence, the “arithmetic problems” subscale from a well-established German
IQ test (Liepmann, 2007). We again expected to find a positive correlation
between spatial abilities and empathic perspective-taking, a positive correlation between numerical intelligence and spatial abilities, but no correlation between numerical intelligence and empathic perspective-taking.
We also expected no correlation between the remaining IRI subscales and
either of the other tests.
Method
Sample. N = 147 individuals (N = 82 female, N = 38 male, N = 27 missing;
Mage = 26, SD = 8) participated for €10. The study was part of a longer battery including other unrelated tasks (Topolinski, 2014). N = 1 participant
was excluded due to below-chance performance on the TFT. Due to a programming error, IRI data were not recorded for N = 27 participants.
Materials and Procedure. In addition to the tasks from Study 1, participants completed a series of 20 arithmetic problems. The task was selfpaced but terminated after ten minutes regardless of how many problems
were solved until then. The dependent variable is the number of correctly
solved problems (cf. Liepmann, 2007). Participants were handed a pen and
a blank sheet of paper for auxiliary calculations.
Results and Discussion
Secondary Analyses. Mean scores and reliabilities of the questionnaires
can be found in Tables 1 and 2. Reliabilities were acceptable to good. While
there was a positive correlation between mean latency and performance
for spatial perspective-taking, r(144) = .32, p < .001, 95% CI [.17, .50], the
opposite was true for numerical intelligence, r(145) = -.19, p = .021, 95% CI
[-.36, -.03]. Again, all subscales of the IRI were significantly correlated with
its total score, all rs(118) > .58, ps < .001. Also, no IRI subscale correlated
with numerical intelligence, all rs(118) < .12, ps > .22.
Main Analyses. Participants who reported higher empathic perspectivetaking again performed better on the visuo-spatial perspective-taking task
(TFT), r(118) = .20, p = .026, 95% CI [.03, .39]. In line with their conceptualization as different facets of intelligence, the TFT and numerical intelligence were positively related, r(144) = .34, p < .001, 95% CI [.19, .52]. Im-
196ERLE AND TOPOLINSKI
portantly, there was no correlation between empathic perspective-taking
and numerical intelligence, r(118) = .05, p = .601, 95% CI [-.13, .23], and the
correlation between spatial and empathic perspective-taking when partialling out numerical intelligence remained on a similar level, rp(117) = .20, p
= .031, 95% CI [.02, .38]. Again, none of the other IRI subscales were related
to TFT performance, all rs(118) < .11, ps > .23. These findings rule out that
general intelligence is responsible for the observed correlation between
spatial and empathic perspective-taking.
In principle, it is possible to solve the TFT by two different strategies: (1)
using actual perspective transformation, the test-taker rotates him or herself mentally to imagine what the object would look like from a different
angle. In contrast, (2) using simple object rotation, the test-taker mentally
rotates the target figure until it matches the orientation of the original reference figure (cf. Zacks & Michelon, 2005; Zacks et al., 2003). Of course, our
conceptual argumentation predicts that only perspective transformation
but not object rotation should be correlated with empathic perspectivetaking because object rotation does not involve the self or the body schema
(cf. Kessler & Rutherford, 2010; Kessler & Thomson, 2010).
Study 3
In this study, we assessed the strategy participants used to solve the TFT,
and predicted that the observed correlation between empathic perspectivetaking and visuo-spatial ability would be greater for individuals performing perspective transformations than those using object rotation. However,
there was no reason to assume that there will be no correlation at all for
object-rotators because it has been shown that both strategies correlate substantially with each other (Hegarty & Waller, 2004). To this end, Study 1
was replicated with an additional assessment of TFT strategy (see below
and Appendix).
Method
Sample. N = 170 individuals (N = 122 female, N = 47 male, N = 1 missing;
Mage = 26, SD = 9) participated for €10. The study was part of a longer battery including other unrelated tasks (Topolinski, Boecker, Erle, Bakhtiari,
& Pecher, 2015). N = 2 participants were excluded because they performed
below-chance level on the TFT. Data of N = 1 participant was lost due to
technical problems.
Materials and Procedure. All procedural details were similar to Study 1. After the TFT, participants received definitions of perspective transformations
and object rotations, respectively, and indicated whether they predominantly used one or the other strategy, a mixture of both, or a completely
different strategy. Then participants rated their strategy on an 8-point scale
ranging from 1 (“only object rotation”) to 8 (“only perspective transformation”). Finally, participants freely described their strategy.
Spatial and empathic perspective-taking197
Results and Discussion
Secondary Analyses. Descriptive statistics and measurement properties of
the questionnaires again can be found in Tables 1 and 2. Reliabilities were
poor to good. Again, mean RT in the TFT correlated positively with mean
performance, r(165) = .29, p < .001, 95% CI [.15, .45], and all IRI subscales
correlated with its total score, all rs > .50, all ps < .001.
Main Analyses. We again found a positive correlation between spatial
ability and empathic perspective-taking, r(165) = .22, p = .004, 95% CI [.07,
.38]. Again, the irrelevant IRI subscales were generally uncorrelated with
TFT performance although PD came close, r(165) = -.15, p = .061, 95% CI
[-.30, .01] (other |r|s < .12, ps >.12).
People reporting perspective transformations as their main strategy during the TFT had a higher mean TFT performance than people reporting
predominantly employing object rotation, r(165) = .24, p = .002, 95% CI [.09,
.39]. While this finding was unexpected, it does not challenge our main
assumptions. More importantly, predominant perspective-takers also reported higher empathic perspective-taking, r(165) = .18, p = .020, 95% CI
[.03, .34].
Finally, we split the sample to compare the respective correlations for
perspective-takers and object-rotators. We hypothesized that the correlation should be larger for people who report performing perspective transformations during the TFT. This was possible because we implemented a
forced-choice 8-point strategy scale. N = 83 (60 female, 23 male) were classified as object-rotators (scores ≤ 4), whereas N = 86 (62 female, 24 male)
were classified as perspective-takers (scores ≥ 5). The correlation between
TFT performance and empathic perspective-taking was significant for the
perspective-takers, r(83) = .29, p = .008, 95% CI [.08, .51], but not for the
object-rotators, r(80) = .13, p = .25, 95% CI [-.09, .35], although these coefficients did not differ significantly, z = 1.06, p = .29.
As acknowledged before, the TFT is not a process-pure measure of spatial perspective-taking abilities because it can also be solved by means
of object rotation. These correlations, however, suggest that the relation
between the TFT and empathic perspective-taking might be stronger for
people performing spatial perspective-taking during the TFT. Of course,
the present data are far from conclusive on this matter because the correlations between the “perspective-takers” and “object-rotators” did not differ
significantly. However, it is still noteworthy that the correlation between
empathic perspective-taking and TFT performance was not statistically
significant for the object-rotators.
There are also several reasons why these differences might be understated in the present study. First, the spatial abilities of object rotation and
perspective transformation are distinct, yet highly correlated (see, e.g.,
Hegarty & Waller, 2004). Second, since these abilities are so similar, potentially participants were confused by the similar definitions of object versus
viewer rotation and this might have distorted self-reports of TFT strategy.
198ERLE AND TOPOLINSKI
Given all this, it might be difficult to show differences between the two
abilities using a correlational approach. Prior research has attempted a direct manipulation of object versus viewer rotation, although in a different
setup (Wraga, Creem-Regehr, & Proffitt, 2004). Potentially, future research
implementing such a direct manipulation of TFT strategy can make a stronger case for the presently suggested distinction.
Since limiting participants’ strategy during the TFT seemed quite difficult, the final study rather went on to generalize the correlation to a clearcut measure of visuo-spatial perspective-taking, which has been shown to
critically depend on embodied self-rotation and to involve object rotation
only to a lesser degree (cf. Kessler & Thomson, 2010).
Study 4
This experiment included both the German version of the IRI (Paulus,
2009) and a more direct measure of spatial perspective-taking from Kessler
and Thomson (2010). In this task, participants see a person sitting at a table
with one object to his left and another object to his right. On every trial,
one of these objects is the target. Participants have to indicate which hand
the person in the picture would use to grab the target object. To do this,
they have a left hand and a right hand response key. Kessler and Thomson (2010) demonstrated that the majority of participants solve this task by
mentally transposing their body schema into the target’s perspective (see
also Kessler & Wang, 2012). Therefore, task difficulty increases with angular disparity between the perspective-taker and the other person. The main
dependent variable in this study was the mean RT over all trials. This task
has the obvious advantage over perspective-taking instructions that it can
be used to induce purely visual perspective-taking. Therefore, it is possible
to tease apart visual and psychological simulation aspects that are inherently confounded when instructing participants “to visualize a person’s
[psychological] situation” (cf. Davis et al., 2004).
Based on the results of the first three studies, we expected a positive correlation between participants’ performance on the spatial perspective-taking task and their self-reported empathic perspective-taking (i.e., a negative correlation between mean RTs and empathic perspective-taking). The
better participants are at visually understanding how the world looks visually for another person, the more inclined they should be to understanding how it “looks” psychologically. This is also in line with another study,
where social skills (a construct similar to empathic perspective-taking; cf.
Baron-Cohen et al., 2001) positively predicted performance on the visual
perspective-taking task (Brunyé et al., 2012). If, however, empathic and
spatial perspective-taking were completely unrelated, there should be no
correlation between this demand-free, process-pure, and objective measure
of spatial perspective-taking and empathic perspective-taking.
Spatial and empathic perspective-taking199
Method
Sample. N = 163 psychology students at the University of Würzburg (N =
130 female, N = 33 male; Mage = 21, SD = 4) participated for partial course
credit. The study was the first part of a two-study package. Due to technical difficulties, the experiment froze for N = 8 participants before the visuospatial perspective-taking task (cf. Kessler & Thomson, 2010), and those
participants only completed demographics and the IRI. Therefore, the final
sample consisted of N = 155 participants.
Materials and Procedure. Participants completed 80 trials (16 at 0, 40, 80,
120, and 160 degrees of angular disparity, respectively) of the spatial perspective-taking paradigm of Kessler and Thomson (2010). Every trial began with the instructions to put the two index fingers on the response keys,
and to sit straight and centered in front of the monitor. After one button
press, participants were shown the target object of the trial. After another
button press, the trial started with a fixation cross that was presented in
the center of the screen for 500 milliseconds before it was replaced by the
person at the table. Participants had to respond as fast and as accurately
as possible but responded self-paced to the stimuli. When they pressed the
wrong key, an error message appeared on the screen for five seconds. Later,
they provided demographic information and completed the German version of the IRI (Paulus, 2009). All error trials (8.3%) were excluded from the
RT analyses. The main analysis focuses on the mean RT over all remaining
trials of the task, aggregated across levels of angular disparity.
Results and Discussion
Secondary Analyses. Descriptive statistics and measurement properties of
the questionnaires again can be found in Tables 1 and 2. Reliabilities were
poor to good. Again, all IRI subscales correlated with its total score (all rs >
.53, all ps < .001). On the spatial perspective-taking task, RTs increased with
increasing angular disparity as indicated by a significant main effect, F(4,
148) = 19.68, p < .001, ηp² = .35, and linear trend, F(1, 151) = 76.18, p < .001,
ηp² = .34. Thus, our stimulus materials (see Appendix) seem comparable to
those used by Kessler and Thomson (2010).
Main Analyses. We found a negative correlation between mean RT in the
spatial perspective-taking task and empathic perspective-taking, r(153) =
-.20, p = .013, 95% CI [-.04, -.36], that is, a positive correlation between spatial and empathic perspective-taking abilities. The other IRI subscales were
again uncorrelated with spatial perspective-taking (other |r|s < .12, ps ≥
.15).
This finding corroborates and extends the results from Studies 1–3 by
showing that higher visual perspective-taking performance (i.e., lower
RTs) correlates with higher dispositional levels of self-reported empathic
perspective-taking. Prior research has shown that for the majority of participants, the task of Study 4 involves embodied self-transformation (Kessler & Wang, 2012), which we assume to be the shared mechanism between
200ERLE AND TOPOLINSKI
simulations of empathic and spatial perspective-taking. Therefore, Study 4
supports our general hypothesis more directly than the first three studies.
In prior research, social skills predicted the RT difference between low
angular disparities, where no embodied self-rotation is necessary to solve
the task, and high angular disparities, where it is (Brunyé et al., 2012). Presently, there was only a relation between empathic perspective taking and
overall RTs, which was comparable for low, r(153) = -.20, p = .012, 95% CI
[-.04, -.36], and high angular disparities, r(153) = -.18, p = .028, 95% CI [-.02,
-.34]. This points to an important difference between social skills, as a measure of empathic “performance” and empathic perspective-taking, which
is a self-report of willingness to engage in perspective-taking. Arguably, it
makes sense to assume that motivation would predict RTs on all levels of
angular disparity whereas skill differentiates at the levels with the highest
difficulty. Of course, both Brunyé and colleagues’ (2012) and our study did
not include a direct manipulation of participants’ body posture (for such
manipulations in other social phenomena see, e.g., Sparenberg, Topolinski,
Springer, & Prinz, 2012; Topolinski, 2010; Topolinski & Sparenberg, 2012),
which limits the interpretability of angular disparity differences in terms
of mental self-rotation.
Meta-Analysis: Gender Effects and Effect-Sizes
Since the observed effects in our studies are rather small, we also computed a meta-analytic correlation over all three studies using the TFT with
a resulting sample size of N = 396. This correlation was identical to that
of the individual studies, r(394) = .21, p < .001, 95% CI [.11, .31]. The confidence interval informs us about the small, yet reliable, effect size of our
studies in the most accurate way.
Importantly, there was only a small number of male participants in all
individual studies, which made it difficult to inspect gender effects individually. The meta-analysis allows us to draw conclusions about potential
gender effects with higher accuracy. Gender has been shown to be related to both empathic (Davis, 1980; Davis & Kraus, 1991) and visuo-spatial
perspective-taking performance (Brunyé et al., 2012; Eisenberg & Lennon,
1983), and mental self-rotation during visuo-spatial perspective-taking
tasks specifically (Kessler & Wang, 2012). If the observed correlation were
stronger for women than for men, the present results would overstate the
true overall correlation, which would limit the implications of the observed
effects, given their as-of-now small effect size. Alternatively, if the observed
correlation were stronger for men than for women, the presently observed
correlations might understate the true overall correlation. We computed
separate meta-analytic correlations for women and men to look at this.
Although the resulting correlations were both significant, r(110) = .36, p
< .001, 95% CI [.18, .56] for male; and r(281) = .13, p = .028, 95% CI [.01, .25]
for female participants, respectively, the correlation for male participants
Spatial and empathic perspective-taking201
was significantly stronger, Z = -2.13, p = .033. This mirrors results of Brunyé
and colleagues (2012), who found that the relation between social skills, a
related construct, and visuo-spatial perspective-taking is stronger for men
than for women.
For Study 4, which used a different visuo-spatial measure, the correlation between visuo-spatial and empathic perspective-taking for men, r(30)
= -.29, p = .105, 95% CI [-.66, .06], was nominally higher than for women,
r(121) = -.17, p = .060, 95% CI [-.35, .01]. Although both individual correlations, as well as their difference (Z = 0.62, p = .535), were not statistically
significant, they were in the predicted direction, which is in line with Studies 1–3, as well as prior research (Brunyé et al., 2012). The lack of gender
effects in this study could be due to differences between social skills and
empathic perspective-taking. Alternatively, it could be a function of the
rather small number of male participants in our study, combined with the
much smaller number of visuo-spatial perspective-taking trials (80 vs. 672
trials in Brunyé et al., 2012).
These additional analyses shed some light on the magnitude of the observed correlations and that the true correlation might actually be understated in the present report, because our samples were predominantly
female. Furthermore, the not always satisfactory reliabilities could limit
the magnitude of the present correlations. The attenuation corrected correlations (cf. Amelang & Schmidt-Atzert, 2006) of our studies equal r‘ =
.32, r‘ = .27, and r‘ = .29, respectively. Interestingly, overall reliability of the
empathic perspective-taking subscale (CR-α = .76 for female, and CR-α =
.82 for male participants, respectively), and especially the TFT (CR-α = .57
for female, and CR-α = .71 for male participants, respectively), was lower
for women than for men, which also contributes to the smaller correlation
for female participants.
On the other hand, there are also some reasons to believe that the correlations should be small in general. For instance, whereas visuo-spatial
perspective-taking is determined mostly by skill, empathic perspectivetaking is always a function of skill and motivation, although these components should be strongly related. The wording of the items of the subscale
PT more closely refers to the motivational aspect of empathic perspectivetaking, that is, the willingness to engage in empathic perspective-taking
rather than to the eventual success of the process (although the original author assumed that PT measures both, see Davis, 1983, p. 115). This provides
another threshold of shared variance that can be accounted for between
visuo-spatial and empathic perspective-taking as measured with the IRI.
Finally, it is unclear how many participants performed perspective transformations in our first two studies. The highest correlation was found for
the alleged perspective-takers in Study 3 but this estimate is based on retrospective report, which can be problematic (cf. Schwarz et al., 1994). On
top of this, it seems appropriate to assume that people engaged in multiple
processes during the TFT.
202ERLE AND TOPOLINSKI
Taken together, there are both reasons to assume that the true correlation
between empathic and visuo-spatial perspective-taking should be higher
and lower in the general population. These biases notwithstanding, we
found a small, yet stable correlation across the whole sample and all subsamples in this meta-analysis, which gives us confidence in the generalizability of this correlation to other measures of empathic perspective-taking
(cf. Brunyé et al., 2012; Kessler & Wang, 2012).
General Discussion
Bridging the theorizing of two, until now separated fields of study, the
social psychology of empathy and the perceptual-cognitive psychology of
visuo-spatial perspective-taking, the present studies establish a dispositional link between empathic perspective-taking and visuo-spatial abilities
that is specific to the faculty of spatial perspective-taking and not (visuospatial) intelligence in general. Our research shows that at least some variance is shared between empathic and spatial perspective-taking and we
believe that contemplating how someone feels often automatically entails
a simulation of that person’s viewpoint, too. This assumption is contrary
to the strong dissociation between visual (imagery) perspective-taking and
empathic perspective-taking (Libby & Eibach, 2011). As mentioned in the
introduction, it seems quite impossible to instruct exclusively psychological simulations of another person’s situation without any visual input or
the other way around. However, spatial perspective-taking paradigms as
in Kessler and Thomson (2010) give us exactly this opportunity because
there is only one way of solving them. For instance, it is conceivable that
having participants adopt the perspective of a target person spatially (as in
Kessler & Thomson, 2010) would result in higher empathy for this person.
In this case, we could be certain that the simulation that caused the psychological self-other merging was purely visual because the spatial perspective-taking paradigm demands this specific simulation. In prior research,
which has demonstrated various benefits of direct instructions to empathize, for instance, stereotype reduction (e.g., Galinsky & Moskowitz, 2000)
or increased helping (e.g., Batson et al., 2002; Batson, Sager, et al., 1997), the
visual and psychological simulation aspect is inherently confounded. In
addidition, the verbal nature of the materials makes it possible for participants to respond to these paradigms without actually taking the perspective of the target at all. Replicating classic perspective-taking instruction
effects with subtle, nonverbal, and demand-free spatial perspective-taking
inductions seems like a promising future research avenue.
Another question for future research is what happens when the causality
is turned around. Is it more difficult to adopt the visual perspective of a
person that we do not want to empathize with? Prior research has shown
that empathic perspective-taking is immediately ineffective when the tar-
Spatial and empathic perspective-taking203
get is a member of a highly stigmatized group such as murderers (see Batson, Polycarpou et al., 1997, Experiment 3). However, empathic perspective-taking had an effect after one week. Since visual perspective-taking
only happens in the here and now, visual perspective taking for murderers
could be more difficult than for other stigmatized (see Batson, Polycarpou
et al., 1997, Experiments 1 and 2) or non-stigmatized targets.
If such a uni- or bidirectional causal relation between visuo-spatial and
empathic or psychological perspective-taking were to be found, future
research should return to connect this knowledge with the existing findings on the mechanisms following perspective simulations. For instance,
it seems reasonable to assume that visually caused perspective-taking resembles more closely the imagine-self perspective (Batson, Early, & Salvarani, 1997; Davis et al., 2004), because only this kind of perspective-taking
asks participants to “put themselves in the shoes of another person”—a
feat which is quite literally achieved by the visuo-spatial manipulation by
Kessler and Thomson (2010). An interesting question is, whether there is
also a relation with imagine-other perspective-taking instructions. Such a
distinction should manifest itself especially on the personal distress that is
felt after perspective-taking, that is, visuo-spatially induced perspectivetaking should have a stronger emotional signature than the imagine-other
perspective-taking instruction (cf. Batson, Early, & Salvarani, 1997; Lamm
et al., 2007).
One alternative explanation that should be tested in future research is
that the observed correlation between empathic and spatial perspective
stems from the strong metaphorical overlap (cf. Lakoff & Johnson, 1980).
People might interpret the IRI questions as referring also to visual kinds of
perspective-taking and therefore match their responses to what they know
about their visuo-spatial capabilities. Note, however, that also this alternative account implies a close relation between the two types of perspectivetaking.
This article is the first to show a dispositional link between spatial perspective-taking abilities and self-reported empathic perspective-taking. Instead of arbitrarily differentiating psychological self-other merging from
the embodied self-other merging that happens during spatial perspectivetaking, we unite the two constructs at a mechanistic level. Simulations
of psychological perspectives are grounded in simulations of the spatial
perspective of another social agent. Of course, the fact that both kinds of
perspective taking are subserved by the same mechanism begs the important question whether improving this mechanism alone is sufficient to increase the rate at which people engage in perspective-taking. If this were
the case, people could practice visuo-spatial perspective-taking to improve
their social skills. Potentially, even a dispositional lack of empathy such as
in autism (Hamilton et al., 2009) might benefit from long-term training in
visuo-spatial abilities.
204ERLE AND TOPOLINSKI
References
Amelang, M., & Schmidt-Atzert, L. (2006).
Psychologische diagnostik und intervention
(pp. 39-44). Heidelberg: Springer Verlag.
Baron-Cohen, S., Wheelwright, S., Skinner, R.,
Martin, J., & Clubley, E. (2001). The autism-spectrum quotient (AQ): Evidence
from asperger syndrome/high-functioning autism, malesand females, scientists
and mathematicians. Journal of Autism
and Developmental Disorders, 31, 5-17.
Barsalou, L. W. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22,
577-660.
Batson, C. D., Chang, J., Orr, R., & Rowland,
J. (2002). Empathy, attitudes, and action:
Can feeling for a member of a stigmatized group motivate one to help the
group? Personality and Social Psychology
Bulletin, 28, 1656-1666.
Batson, C. D., Early, S., & Salvarani, G. (1997).
Perspective taking: Imagining how another feels versus imagining how you
would feel. Personality and Social Psychology Bulletin, 23, 751-758.
Batson, C. D., Polycarpou, M. P., HarmonJones, E., Imhoff, H. J., Mitchener, E. C.,
Bednar, L. L., Klein, T. R., & Highberger,
L. (1997). Empathy and attitudes: Can
feeling for a member of a stigmatized
group improve feelings toward the
group? Journal of Personality and Social
Psychology, 72, 105-118.
Batson, C. D., Sager, K., Garst, E., Kang, M.,
Rubchinsky, K., & Dawson, K. (1997). Is
empathy-induced helping due to selfother merging? Journal of Personality and
Social Psychology, 73, 495-509.
Brunyé, T. T., Ditman, T., Giles, G. E., Mahoney, C. R., Kessler, K., & Taylor, H. A.
(2012). Gender and autistic personality
traits predict perspective-taking ability
in typical adults. Personality and Individual Differences, 52, 84-88.
Chai, X. J., & Jacobs, L. F. (2009). Sex differences in directional cue use in a virtual
landscape. Behavioral Neuroscience, 123,
276-283.
Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. Catalog of Selected Documents in
Psychology, 10, 85-99.
Davis, M. H. (1983). Measuring individual
differences in empathy: Evidence for a
multidimensional approach. Journal of
Personality and Social Psychology, 44, 113126.
Davis, M. H., Conklin, L., Smith, A., & Luce,
C. (1996). Effect of perspective taking on
the cognitive representation of persons:
A merging of self and other. Journal of
Personality and Social Psychology, 70, 713726.
Davis, M. H., & Kraus, L. A. (1991). Dispositional empathy and social relationships.
In W. H. Jones & D. Perlman (Eds.), Advances in personal relationships (Vol. 3, pp.
75-115). Greenwich, CT: JAI Press.
Davis, M. H., Soderlund, T., Cole, J., Gadol,
E., Kute, M., Myers, M., & Weihing, J.
(2004). Cognitions associated with attempts to empathize: How do we imagine the perspective of another? Personality and Social Psychology Bulletin, 30,
1625-1635.
Eisenberg, N., & Lennon, R. (1983). Sex differences in empathy and related capacities.
Psychological Bulletin, 94, 100-131.
Epley, N., & Caruso, E. M. (2009). Perspective
taking: Misstepping into others’ shoes.
In K. D. Markman, W. M. P. Klein, & J.
A. Suhr (Eds.), Handbook of imagination
and mental simulation (pp. 295-309). New
York: Psychology Press.
Epley, N., Keysar, B., Van Boven, L., & Gilovich, T. (2004). Perspective taking as egocentric anchoring and adjustment. Journal of Personality and Social Psychology,
87, 327-339.
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G* Power 3: A flexible
statistical power analysis program for
the social, behavioral, and biomedical
sciences. Behavior Research Methods, 39,
175-191.
Fiske, S. T., Taylor, S. E., Etcoff, N. L., & Laufer,
J. K. (1979). Imaging, empathy, and
causal attribution. Journal of Experimental Social Psychology, 15, 356-377.
Flavell, J. H., Everett, B. A., Croft, K., & Flavell,
E. R. (1981). Young children’s knowledge about visual perception: Further
evidence for the Level 1–Level 2 dis-
Spatial and empathic perspective-taking205
tinction. Developmental Psychology, 17,
99-103.
Galinsky, A. D., Ku, G., & Wang, C. S. (2005).
Perspective-taking and self-other overlap: Fostering social bonds and facilitating social coordination. Group Processes
and Intergroup Relations, 8, 109-124.
Galinsky, A. D., & Moskowitz, G. B. (2000).
Perspective-taking: Decreasing stereotype expression, stereotype accessibility,
and in-group favoritism. Journal of Personality and Social Psychology, 78, 708-724.
Galinsky, A. D., Wang, C. S., & Ku, G. (2008).
Perspective-takers behave more stereotypically. Journal of Personality and Social
Psychology, 95, 404-419.
Glenberg, A. M. (2010). Embodiment as a unifying perspective for psychology. Wiley
Interdisciplinary Reviews: Cognitive Science, 1, 586-596.
Hall, J. A., & Schmid Mast, M. (2008). Are
women always more interpersonally
sensitive than men? Impact of goals and
content domain. Personality and Social
Psychology Bulletin, 34, 144-155.
Hamilton, A. F. d. C., Brindley, R., & Frith, U.
(2009). Visual perspective taking impairment in children with autistic spectrum
disorder. Cognition, 113, 37-44.
Hegarty, M., & Waller, D. (2004). A dissociation
between mental rotation and perspective-taking spatial abilities. Intelligence,
32, 175-191.
Kessler, K., & Rutherford, H. (2010). The two
forms of visuo-spatial perspective taking are differently embodied and subserve different spatial prepositions.
Frontiers in Psychology, 1, 1-12.
Kessler, K., & Thomson, L. A. (2010). The embodied nature of spatial perspective taking: Embodied transformation versus
sensorimotor interference. Cognition,
114, 72-88.
Kessler, K., & Wang, H. (2012). Spatial perspective taking is an embodied process,
but not for everyone in the same way:
Differences predicted by sex and social
skills score. Spatial Cognition and Computation, 12, 133-158.
Klein, K. J., & Hodges, S. D. (2001). Gender
differences, motivation, and empathic
accuracy: When it pays to understand.
Personality and Social Psychology Bulletin,
27, 720-730.
Kross, E., & Ayduk, O. (2011). Making meaning out of negative experiences by selfdistancing. Current Directions in Psychological Science, 20, 187-191.
Lakoff, G., & Johnson, M. (1980). Metaphors we
live by. Chicago: The University of Chicago Press.
Lamm, C., Batson, C. D., & Decety, J. (2007).
The neural substrate of human empathy: Effects of perspective-taking and
cognitive appraisal. Journal of Cognitive
Neuroscience, 19, 42-58.
Lawton, C. A. (1994). Gender differences in
way-finding strategies—Relationship
to spatial ability and spatial anxiety. Sex
Roles, 30, 765-779.
Libby, L. K., & Eibach, R. P. (2011). Visual perspective in mental imagery: A representational tool that functions in judgment,
emotion, and self-insight. In M. P. Zanna
& J. M. Olson (Eds.), Advances in experimental social psychology (Vol. 44, pp. 185245). San Diego: Academic.
Libby, L. K., Shaeffer, E. M., & Eibach, R. P.
(2009). Seeing meaning in action: A bidirectional link between visual perspective and action identification level. Journal of Experimental Psychology: General,
138, 503-516.
Libby, L. K., Shaeffer, E. M., Eibach, R. P., &
Slemmer, J. A. (2007). Picture yourself at
the polls: Visual perspective in mental
imagery affects self-perception and behavior. Psychological Science, 18, 199-203.
Liepmann, D. (2007). Intelligenz-Struktur-Test
2000 R: IST 2000 R. Göttingen: Hogrefe.
Michelon, P., & Zacks, J. M. (2006). Two kinds
of visual perspective taking. Perception
and Psychophysics, 68, 327-337.
Paulus, C. (2009). Der Saarbrücker Persönlichkeitsfragebogen SPF (IRI) zur Messung von
Empathie: Psychometrische Evaluation der
deutschen Version des Interpersonal Reactivity Index. Retrieved from http://
psydok.sulb.uni-saarland.de/volltexte/2009/2363/pdf/SPF_Artikel.pdf.
Schwarz, N., Sudman, S., Brewer, W. F., Herrmann, D. J., Back, K. W., Ross, M., et al.
(1994). Autobiographical memory and the
validity of retrospective reports. Heidelberg: Springer-Verlag.
Shelton, A. L., Clements-Stephens, A. M., Lam,
W. Y., Pak, D. M., & Murray, A. J. (2012).
Should social savvy equal good spatial
skills? The interaction of social skills
206ERLE AND TOPOLINSKI
with spatial perspective taking. Journal
of Experimental Psychology: General, 141,
199-205.
Sparenberg, P., Topolinski, S., Springer, A.,
& Prinz, W. (2012). Minimal mimicry:
Mere effector matching induces preference. Brain and Cognition, 80, 291-300.
Stumpf, H., & Fay, E. (1983). Schlauchfiguren:
Ein test zur Beurteilung des räumlichen
Vorstellungsvermögens; Handanweisung.
Göttingen: Verlag für Psychologie, Hogrefe.
Surtees, A. D. R., Apperly, I., & Samson, D.
(2013a). Similarities and differences in
visual and spatial perspective-taking
processes. Cognition, 129, 426-438.
Surtees, A. D. R., Apperly, I., & Samson, D.
(2013b). The use of embodied self-rotation for visual and spatial perspectivetaking. Frontiers in Human Neuroscience,
7, 698.
Thakkar, K. N., Brugger, P., & Park, S. (2009).
Exploring empathic space: Correlates of
perspective transformation ability and
biases in spatial attention. PLoS One, 4,
e5864.
Topolinski, S. (2010). Moving the eye of the
beholder: Motor components in vision
determine aesthetic preference. Psychological Science, 21, 1220-1224.
Topolinski, S. (2012). The sensorimotor contributions to implicit memory, familiarity,
and recollection. Journal of Experimental
Psychology: General, 141, 260-281.
Topolinski, S. (2014). A processing fluencyaccount of funniness: Running gags and
spoiling punchlines. Cognition and Emotion, 28, 811-820.
Topolinski, S., Boecker, L., Erle, T. M., Bakhtiari,
G., & Pecher, D. (2015). Affective matching of oral inward-outward movements
with object valence and oral object affordance. Manuscript submitted for publication.
Topolinski, S., Erle, T. M., & Reber, R. (2015).
Necker’s smile: Immediate affective
consequences of early perceptual processes. Cognition, 140, 1-13.
Topolinski, S., Maschmann, I. T., Pecher, D., &
Winkielman, P. (2014). Oral approachavoidance: Affective consequences of
muscular articulation dynamics. Journal
of Personality and Social Psychology, 106,
885-896.
Topolinski, S., & Sparenberg, P. (2012). Turning
the hands of time—Clockwise movements increase preference for novelty.
Social Psychological and Personality Science, 3, 308-314.
Topolinski, S., & Strack, F. (2009). Motormouth:
Mere exposure depends on stimulusspecific motor simulations. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 35, 423-433.
Topolinski, S., & Strack, F. (2010). False fame
prevented: Avoiding fluency effects
without judgmental correction. Journal
of Personality and Social Psychology, 98,
721-733.
Tversky, B., & Hard, B. M. (2009). Embodied
and disembodied cognition: Spatial perspective-taking. Cognition, 110, 124-129.
Valenti, G., Libby, L. K., & Eibach, R. P. (2011).
Looking back with regret: Visual perspective in memory images differentially affects regret for actions and inactions. Journal of Experimental Social Psychology, 47, 730-737.
Vandenberg, S. G., & Kuse, A. R. (1978). Mental
rotations, a group test of three-dimensional spatial visualization. Perceptual
and Motor Skills, 47, 599-604.
Van Laer, T., de Ruyter, K., Visconti, L. M.,
& Wetzels, M. (2014). The extended
transportation-imagery model: A metaanalysis of the antecedents and consequences of consumers’ narrative transportation. Journal of Consumer Research,
40, 797-817.
Wraga, M., Creem-Regehr, S. H., & Proffitt,
D. R. (2004). Spatial updating of virtual
displays. Memory and Cognition, 32, 399415.
Zacks, J. M., & Michelon, P. (2005). Transformations of visuospatial images. Behavioral and Cognitive Neuroscience Reviews,
4, 96-118.
Zacks, J. M., Vettel, J. M., & Michelon, P. (2003).
Imagined viewer and object rotations
dissociated with event-related fMRI.
Journal of Cognitive Neuroscience, 15,
1002-1018.
Spatial and empathic perspective-taking207
Appendix
Here, we provide a complete list of all our stimuli that are not copyright protected. Because of copyright protection, materials of the Tube Figures Test (Stumpf & Fay, 1983) and
the Arithmetic Problems subscale of the IST (Liepmann, 2007) cannot be shared online. Although we translated our German materials as precisely as possible, the English translations are not always verbatim.
Interpersonal Reactivity Index (IRI; Studies 1–4)
The German version of the IRI consists of only 16 items (cf. Paulus, 2009). Each subscale
consists of four items, which are presented in a fixed order. The scale labels were 1 = nie
(never), 2 = selten (rarely), 3 = manchmal (sometimes), 4 = oft (often), and 5 = immer (always).
Table 1 shows the original German items and their corresponding item within the original
IRI (Davis, 1983) in italics.
Table 1. Items of the German Version of the IRI and the Corresponding Item from the Original
Item
1
Item wording and translation
Ich empfinde warmherzige Gefühle für Leute, denen es weniger
gut geht als mir.
Subscale
EC
I often have tender, concerned feelings for people less fortunate
than me.
2
Die Gefühle einer Person in einem Roman kann ich mir sehr gut
vorstellen.
FS
I really get involved with the feelings of the characters in a novel.
3
In Notfallsituationen fühle ich mich ängstlich und unbehaglich.
PD
In emergency situations, I feel apprehensive and ill-at-ease.
4
Ich versuche, bei einem Streit zuerst beide Seiten zu verstehen,
bevor ich eine Entscheidung treffe.
PT
I try to look at everybody’s side of a disagreement before I make a
decision.
5
Wenn ich sehe, wie jemand ausgenutzt wird, glaube ich, ihn
schützen zu müssen.
EC
When I see someone being taken advantage of, I feel kind of
protective toward them.
6
Ich fühle mich hilflos, wenn ich inmitten einer sehr
emotionsgeladenen Situation bin.
PD
I sometimes feel helpless when I am in the middle of a very
emotional situation.
7
Nachdem ich einen Film gesehen habe, fühle ich mich so, als ob
ich eine der Personen aus diesem Film sei.
FS
After seeing a play or movie, I have felt as though I were one of the
characters.
8
In einer gespannten emotionalen Situation zu sein, beängstigt
mich.
Being in a tense emotional situation scares me.
PD
208ERLE AND TOPOLINSKI
Table 1. (continued)
Item
9
Item wording and translation
Mich berühren Dinge sehr, auch wenn ich sie nur beobachte.
Subscale
EC
I am often quite touched by things that I see happen.
10
Ich glaube, jedes Problem hat zwei Seiten und versuche deshalb
beide zu berücksichtigen.
PT
I believe that there are two sides to every question and try to look
at them both.
11
Ich würde mich selbst als eine ziemlich weichherzige Person
bezeichnen.
EC
I would describe myself as a pretty soft-hearted person.
12
Wenn ich einen guten Film sehe, kann ich mich sehr leicht in die
Hauptperson hineinversetzen.
FS
When I watch a good movie, I can very easily put myself in the
place of a leading character.
13
In heiklen Situationen neige ich dazu, die Kontrolle über mich zu
verlieren.
PD
I tend to lose control during emergencies.
14
Wenn mir das Verhalten eines anderen komisch vorkommt,
versuche ich mich für eine Weile in seine Lage zu versetzen.
PT
When I’m upset at someone, I usually try to “put myself in his
shoes” for a while.
15
Wenn ich eine interessante Geschichte oder ein gutes Buch lese,
versuche ich mir vorzustellen, wie ich mich fühlen würde, wenn
mir die Ereignisse passieren würden.
FS
When I am reading an interesting story or novel, I imagine how I
would feel if the events in the story were happening to me.
16
Bevor ich jemanden kritisiere, versuche ich mir vorzustellen, wie
die Sache aus seiner Sicht aussieht.
PT
Before criticizing some body, I try to imagine how I would feel if I
were in their place.
Notes. EC = Empathic Concern. FS = Fantasy. PD = Personal Distress. PT = Perspective-Taking.
Tube Figure Test Strategy Assessment (Study 3)
Table 2 shows the original German instructions of the TFT strategy assessment and an
English translation. The analyses in the paper focused on the eight-point scale answers
because those provided higher statistical power in comparison to the nominal four-point
scale and because the free answers often were unclear about which strategy was used (e.g.,
“I imagined how the object looked after the rotation”). We refrained from using the term
perspective-taking in our questions because we were concerned that participants might
guess the purpose of the study. Also, we wanted to emphasize the contrast between the two
studies by using the same terminology (self vs. object rotation).
Spatial and empathic perspective-taking209
Table 2. Instructions, Question Wording and Response Scales of the TFT Strategy Assessment
Item
German
English
Instruction
Vielen Dank für die Bearbeitung dieser
Aufgabe! In dieser Studie interessieren
wir uns weniger für die von Ihnen
erbrachte Gesamtleistung bei diesen
visuellen Rätseln, sondern vielmehr
interessiert uns die Lösungsstrategie,
die Sie verwendet haben. Es folgen
einige Fragen dazu.
Thank you for completing this task! In
this study, we are not only interested
in your performance on these visual
puzzles but also in the strategy that
you used to solve them. A couple of
questions about this will follow.
Four-point
strategy
assessment
Unsere Probanden berichten im
Wesentlichen zwei Lösungsstrategien:
1. Man kann die zweite Figur im
Kopf drehen, bis sie so aussieht wie
die erste. Also man dreht das Objekt
und nicht sich selbst, man gleicht
die beiden Objekte aneinander an.
2. Man versucht sich vorzustellen,
wie die erste Figur aussehen würde,
wenn man selbst links, rechts, über,
unter oder hinter ihr stehen würde.
Also man stellt sich die Frage,
wie das Objekt aus einer anderen
Perspektive aussehen würde. Hier
dreht man also sich selbst um auf
die Lösung zu kommen. Bitte klicken
Sie die Strategie an, die mehr auf Sie
zutrifft. Welche Strategie haben Sie
bei der Lösung der visuellen Rätsel
verwendet?
Our participants mainly report two
strategies: 1. You can rotate the second
tubular figure until it visually matches
the first. That means you rotate the
object and not yourself and thereby
you achieve a match between the first
and the second figure. 2. You can also
imagine how the first figure would
look if you yourself would stand to
its left, its right, above it, below it or
behind it. That means you imagine
how the object would look from
another perspective. Here, you rotate
yourself to come up with the correct
solution. Please indicate which of these
strategies applies best to you.
Skala: (1) Objekt drehen, (2) Selbst
drehen, (3) Eine Mischung, (4) Eine
andere Strategie
Scale: (1) object rotation, (2) self rotation,
(3) mixture of (1) & (2), (4) other
strategy
Falls Sie eine Mischung der beiden
Strategien verwendet haben, können
Sie diese hier noch spezifizieren.
In case you used a mixture of the two
strategies, please specify the amount to
which you used the two strategies.
Skala: (1) Ich habe fast immer das
Objekt gedreht. – (8) Ich habe fast
immer mich selbst gedreht.
Scale: (1) I almost always rotated the
object. – (8) I almost always rotated
myself.
Hier haben Sie jetzt noch die
Möglichkeit, frei über die von Ihnen
verwendete Strategie zu schreiben
(zum Beispiel wenn Sie eine
andere Strategie verwendet haben).
Beschreiben Sie bitte möglichst
genau, wie Sie die visuellen Rätsel
bearbeitet haben! Beschreiben Sie
bitte auch Ihre Strategie, wenn Sie
eine der von uns beschriebenen
Strategien verwendet haben, das ist
auch interessant für uns. Wie gesagt:
Nicht Ihre Leistung, sondern die
Strategie bei der Aufgabe interessiert
uns.
Finally, you now have the opportunity
to freely describe the strategy that you
used (for instance if you indicated
using a completely different strategy).
Please specify in as much detail as
possible how you solved the visual
puzzles! Please also describe your
strategy if you used one of the two
strategies which we defined earlier,
as this is also interesting for us. As we
said: Not your performance, but the
strategy within this task is of interest
to us.
Eight-point
strategy
assessment
Free strategy
description
210ERLE AND TOPOLINSKI
Stimuli of the Perspective-Taking Task (Study 4)
The data of Study 4 was originally collected for another line of research. In addition to
the angular disparity between the participant and the target in the picture (varied within
subjects between 0°, 40°, 80°, 120°, and 160°; cf. Kessler & Thomson, 2010), we varied the
target person. Since no manipulation except for angular disparity affected reaction times (all
Fs < 1.41, all ps > .23), we collapsed the data for the present paper. Figure 1 shows only one
exemplary stimulus of the experiment but we are happy to provide our whole stimulus set
to interested researchers.
Figure 1. Perspective-taking trial with a human target at 40° angular disparity.