Sensation and perception

Neuropsychologia 79 (2015) 175–185

Contents lists available at ScienceDirect

Neuropsychologia

http://d 0028-39

n Corr UMR 72 Univers Victor H

E-m

journal homepage: www.elsevier.com/locate/neuropsychologia

Changing perspective: The role of vestibular signals

Diane Deroualle a, Liliane Borel a, Arnaud Devèze b,c, Christophe Lopez a,n

a Aix Marseille Université, CNRS, NIA UMR 7260, 13331 Marseille, France b Aix Marseille Université, IFSTTAR, LBA UMR T24, 13916 Marseille, France c Assistance Publique Hôpitaux de Marseille (APHM), Hôpital Nord, Service ORL et Chirurgie Cervico-faciale, 13915 Marseille, France

a r t i c l e i n f o

Article history: Received 20 April 2015 Received in revised form 20 July 2015 Accepted 21 August 2015 Available online 24 August 2015

Keywords: Vestibular system Multisensory integration Embodied cognition Mental imagery Perspective taking Third-person perspective Full-body motion platform Social cognition

x.doi.org/10.1016/j.neuropsychologia.2015.08.0 32/& 2015 Elsevier Ltd. All rights reserved.

espondence to: Laboratoire de Neurosciences 60, Centre National de la Recherche Scientifiq ité, Centre Saint Charles, Fédération de Rech ugo, 13331 Marseille Cedex 03, France. Fax: þ ail address: [email protected] (C.

a b s t r a c t

Social interactions depend on mechanisms such as the ability to take another person's viewpoint, i.e. visuo-spatial perspective taking. However, little is known about the sensorimotor mechanisms under- pinning perspective taking. Because vestibular signals play roles in mental rotation and spatial cognition tasks and because damage to the vestibular cortex can disturb egocentric perspective, vestibular signals stand as important candidates for the sensorimotor foundations of perspective taking. Yet, no study merged natural full-body vestibular stimulations and explicit visuo-spatial perspective taking tasks in virtual environments. In Experiment 1, we combined natural vestibular stimulation on a rotatory chair with virtual reality to test how vestibular signals are processed to simulate the viewpoint of a distant avatar. While they were rotated, participants tossed a ball to a virtual character from the viewpoint of a distant avatar. Our results showed that vestibular signals influence perspective taking in a direction- specific way: participants were faster when their physical body rotated in the same direction as the mental rotation needed to take the avatar's viewpoint. In Experiment 2, participants realized 3D object mental rotations, which did not involve perspective taking, during the same whole-body vestibular stimulation. Our results demonstrated that vestibular stimulation did not affect 3D object mental rota- tions. Altogether, these data indicate that vestibular signals have a direction-specific influence on visuo- spatial perspective taking (self-centered mental imagery), but not a general effect on mental imagery. Findings from this study suggest that vestibular signals contribute to one of the most crucial mechanisms of social cognition: understanding others’ actions.

& 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The visuo-spatial perspective an observer has on their en- vironment is commonly experienced as originating from their body, and is often referred to as first-person perspective (1PP) in the literature. Seminal psychological investigations into the first-per- son perspectival experience (Claparède, 1925; Stratton, 1899), as well as more recent virtual reality studies manipulating the par- ticipant's viewpoint (Slater et al., 2010), showed that the self is invariably bound to the origin of 1PP. Therefore, 1PP is now con- sidered by many philosophers and neuroscientists a main con- stituent of human self-consciousness (Vogeley and Fink, 2003; Vogeley et al., 2004; Blanke and Metzinger, 2009; Blanke, 2012; Serino et al., 2013). While 1PP is the common viewpoint on the

22

Intégratives et Adaptatives- ue (CNRS) and Aix-Marseille erche 3C – Case B, 3, Place 33 4 13 55 08 44. Lopez).

world, infants progressively develop the ability to take another person's viewpoint, or third-person perspective (3PP), which con- stitutes a crucial feature of social interactions and empathy (e.g. Piaget and Inhelder, 1956). In the present article, we use the term “3PP taking” to designate the mechanisms used by an observer to simulate the visuo-spatial viewpoint of another person (i.e. what the other person can see), rather than the mechanisms by which an observer reads another person's emotions, intentions and thoughts (i.e. “mentalizing”, Frith and Frith, 2006).

Research in cognitive neuroscience has traditionally in- vestigated perspective taking using own-body mental transfor- mation tasks, where participants are required to simulate another body's position and make judgments from this new position. Most tasks require participants to decide whether the colored hand of a body is a right or left hand, or whether the outstretched arm of a body is a right or left arm (e.g. Blanke et al., 2005; Mohr et al., 2010; Parsons, 1987a, 1987b; Schwabe et al., 2009; Zacks et al., 1999). Other studies used virtual reality to investigate perspective taking more directly, by measuring the participant's ability to lo- cate or count objects from a distant viewpoint (Committeri et al.,

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185176

2004; David et al., 2006; Kessler and Thomson, 2010; Kockler et al., 2010; Lambrey et al., 2012; Mattan et al., 2014; Samson et al., 2010). A common finding of these studies is that the time needed to make a judgment from a distant viewpoint increases as the angle increases between the participant's viewpoint and the si- mulated viewpoint (e.g. Kessler and Rutherford, 2010; Kessler and Thomson, 2010; Kockler et al., 2010). One explanation is that the cognitive cost of the own-body mental transformation (i.e. men- tally aligning one's own body reference with that of a distant avatar or person) increases as the angle increases between the participant's and avatar's bodies (Parsons, 1987a,1987b). Another line of research posits that perspective taking is a strongly em- bodied process and that this longer reaction time may relate to the incongruence between the participant's body posture and that of a distant avatar. In a series of behavioral studies, Kessler and col- leagues showed that participants were faster to adopt an avatar's viewpoint when their body posture matched that of the avatar (Kessler and Rutherford, 2010; Kessler and Thomson, 2010; Kessler et al., 2014). This line of research emphasized the embodied nature of visuo-spatial perspective taking, as previously proposed for mental imagery of non-corporeal objects (see Amorim et al., 2006).

To date, only few studies investigated the motor and sensory mechanisms underpinning 3PP taking. Tversky and Hard (2009), for example, asked participants to describe the spatial relation between two objects placed on a table and showed that the mere presence of another person increased the likelihood of describing the location of these objects according to this person's viewpoint. Importantly, observing another person intending to grasp an ob- ject further increased the description of object position according to the 3PP, which suggests that spontaneous 3PP taking subserves understanding another person's actions (Tversky and Hard, 2009). In addition, there is evidence that signals from one's own body are used to mentally rotate one's body in space. As mentioned above, 3PP taking is faster and more accurate when the participant's body posture is congruent with that of the avatar whose viewpoint is simulated (Kessler and Thomson, 2010). Similarly, there is a wealth of data showing that mental rotation of bodies and body parts (without perspective taking) is modulated by the observer's body posture. Thus, participants are faster to detect whether pictures of hands depict a left or right hand if the participants' hands are in a posture (e.g. palm down) congruent to the hand visually presented when compared to an incongruent posture (e.g. palm up) (Ionta and Blanke, 2009; Conson et al., 2011; Bläsing et al., 2013; Ionta et al., 2007, 2012, 2013). Clinical investigations suggest that nor- mal processing of proprioceptive and motor signals is necessary for accurate mental own-body transformations, as patients with proprioceptive deafferentation (Ter Horst et al., 2012) or premotor cortex lesion (Arzy et al., 2006) are impaired in mental rotation of body parts. In line with this finding, studies using brachial plexus anesthesia or imagined paralysis in healthy volunteers showed that somatosensory signals are required for mental rotations of body parts (Hartmann et al., 2011; Silva et al., 2011).

Another important candidate for the sensorimotor foundations of perspective taking is the vestibular system, which has recently been involved in a growing number of bodily and self experiences (reviews in Blanke, 2012; Ferrè et al., 2013; Lenggenhager and Lopez, 2015a, 2015b; Lopez, 2013; Mast et al., 2014). Yet, vestibular implication in the sensorimotor mechanisms of visuo-spatial and social perspective taking remains under-investigated. Since ves- tibular receptors encode own body rotations and translations in space, brain structures processing vestibular information are likely also involved in spatial transformations necessary to mentally si- mulate being located somewhere else and to imagine what the perspective would be from this new location (e.g. Mast et al., 2007; Lopez et al., 2011; Deroualle and Lopez, 2014). Three main lines of

arguments support this view. First, a vestibular contribution to perspective taking is suggested by observations in epileptic pa- tients reporting out-of-body experiences, characterized by the feeling of being located outside the body and of perceiving the environment from this extracorporeal position. Importantly, such disembodied 3PP is often associated with vestibular illusions, in- cluding sensations of lightness or levitation, which suggests that vestibular signal is misintegrated when 1PP is lost (Brugger, 1997; Blanke et al., 2002, 2004; Blanke and Mohr, 2005; Lopez et al., 2008, 2010a; Heydrich et al., 2011). Second, several studies used artificial vestibular stimulations to test the influence of vestibular signals on own-body mental imagery and mental rotation. Yet, these studies provided conflicting results. Three reported altered performance: Lenggenhager et al. (2008) found that galvanic vestibular stimulation increased response times for mental rota- tions of objects, and Mast et al. (2006) showed that caloric ves- tibular stimulation increased error rate for mental rotation of letters. An increase in error rate was also reported as an effect of galvanic vestibular stimulation on perspective taking tasks – without adopting another person's view point (Dilda et al., 2012a). In contrast with these three studies, Falconer and Mast (2012) showed that caloric vestibular stimulation facilitated mental ro- tation of bodies and suggested that vestibular signals could facil- itate the updating of bodily references and of the body schema. The reasons for the discrepancies in these results are not yet clear, but a common feature of these studies is that they used artificial stimulation of the vestibular receptors, providing the brain with vestibular information that has no physiological equivalent. No- tably, only one study (Van Elk and Blanke, 2014) used natural vestibular stimulation by means of a rotatory motorized chair to investigate how vestibular information influences own-body mental transformation. The participants detected faster whether the colored hand of an avatar was a left or right hand when the chair rotated in the same direction as the mental rotation needed to align their body axis with the avatar's. Finally, a third line of evidence suggesting a vestibular contribution to own-body mental imagery comes from patients with vestibular disorders. Vestibular defective patients are slower and less accurate in mental own- body transformations tasks (Candidi et al., 2013; Grabherr et al., 2011) as well as in tasks requiring to mentally navigate in space (Péruch et al., 2011). Yet, the tasks performed by these patients, as well as those by healthy participants reported above, involved own-body mental transformations (e.g. to judge whether the co- lored arm or extended arm of an avatar is a right or left arm) but, strictly speaking, they did not require changes in the participants' viewpoint. Thus, the range of tasks used so far did not require locating objects according to a simulated distant spatial location nor interacting with other individuals from this new position.

In the present study, we combined natural vestibular stimula- tion on a rotatory chair with virtual reality to test how vestibular signals are used to simulate an avatar's perspective (Experiment 1), and to achieve mental rotation of 3D objects that did not involve perspective taking (Experiment 2). In Experiment 1, we developed a virtual ball-tossing game (adapted from David et al., 2006), which the participants played from their 1PP or from the viewpoint of a distant avatar (3PP), while they were simultaneously rotated in the yaw plane. As participants were required to throw a ball to a person from the viewpoint of a distant avatar, we assume that we assessed the influence of vestibular signals on changing the viewpoint, and not only on mental own-body imagery. A second experiment was designed to test this proposition more directly with the aim of disentangling the influence of vestibular signals on 3PP taking (i.e. egocentric mental transformation) vs. 3D object mental rotation (i.e. object-based mental transformation). As previous studies found that galvanic and caloric vestibular sti- mulation influenced mental imagery of non-corporeal objects

Time (s) Onset

4 8 End

1.5 23

Stimulus Feedback

Answer

0

+ +

A cc

el er

at io

n

) V

el oc

ity

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185 177

(Dilda et al., 2012; Falconer and Mast, 2012; Lenggenhager et al., 2008; Mast et al., 2006), it was important to test whether natural vestibular stimulation influenced 3D object mental rotation in a direction-specific way. In Experiment 2, we developed a task re- quiring participants to mentally rotated 3D objects (object-based mental transformations) in the yaw plane while they were rotated in the same plane. In contrast with previous studies that used caloric and galvanic vestibular stimulations, the natural stimula- tion used here has the advantage of providing the brain with physiologically accurate vestibular signals and to assess behavior in a more ecological paradigm. Another important aspect of this study was to create visuo-spatial tasks that imposed mental ro- tation of the participant's body and viewpoint, or mental rotation of 3D objects, in a pre-defined direction (clockwise vs. counter- clockwise) in the yaw plane. In both experiments, we compared trials for which the direction of mental rotation and the direction of chair rotation were congruent and incongruent to determine whether vestibular information is used to compute visuo-spatial perspective taking and 3D object mental rotation.

of rotation of rotation

1PP taking

3PP taking

Visual detection

Fig. 1. Experimental setup and stimuli for Experiment 1. (A) Eight-second mo- tion profile used for whole-body rotations in the yaw plane. (B) Each trial consisted of an 8 s clockwise or counterclockwise rotation during which participants per- formed the perspective-taking task. After an accommodation period of 1.5 s, a vi- sual scene was presented (‘stimulus’) with avatars located around a circle. After participants answered, a feedback video on their performance was shown (‘feed- back’). A fixation cross was presented until the end of the chair rotation and during the 15 s inter-stimulus interval. (C) Example of visual scene shown for the third- person perspective-taking task (3PP taking). Participants were instructed to take the perspective of the avatar holding the ball (wearing a blue T-shirt) and to vir- tually throw the ball to the avatar wearing a red T-shirt. The correct response here was to throw the ball to the left. (D) Example of visual scene shown for the first- person perspective task (1PP taking). Participants were instructed to virtually throw the ball to the avatar wearing a red T-shirt from their own perspective. The correct response here was to throw the ball to the right. (E) Example of visual scene shown for the visual detection task. Participants were instructed to detect whether one avatar was holding a ball or not. The correct response here was ‘yes’. (For in- terpretation of the references to color in this figure, the reader is referred to the web version of this article.)

2. Experiment 1

2.1. Materials and methods

2.1.1. Participants Twenty healthy volunteers participated (13 females and

7 males, mean age7SD: 2574 years). All participants were right- handed, as confirmed by the Edinburgh Handedness inventory (74724%) (Oldfield, 1971). Participants had normal or corrected- to-normal vision and declared no history of vestibular, neurolo- gical, or psychiatric disease. Experimental procedures were ap- proved by the local Ethics Committee (Comité de Protection des Personnes Sud-Mediterranée II, 2011-A01221-40) and followed the ethical recommendations laid down in the Declaration of Helsinki.

2.1.2. Vestibular stimulation Vestibular stimulations were produced using a motorized ro-

tatory chair (Rematique, Med4, Saint-Etienne). The chair rotated in a horizontal plane around a vertical axis aligned with the parti- cipant's longitudinal body axis (i.e. yaw rotations). The motion profile was an 8-s unidirectional rotation in the clockwise or counterclockwise direction. The chair accelerated at 20 °/s² during 4 s until it reached a velocity of 80 °/s, then it decelerated during 4 s at 20 °/s² until it returned to 0 °/s (Fig. 1a). This motion profile was selected because it provides suprathreshold vestibular sti- mulation interfering with spatial cognition (Figliozzi et al., 2005; Van Elk and Blanke, 2014), induces no or weak neurovegetative side effects, and includes an acceleration phase long enough (4 s) to complete the tasks.

2.1.3. Visual stimuli and tasks Participants were involved in a virtual ball-tossing game

(adapted from David et al., 2006). Visual stimuli consisted of co- lored pictures of a scene with 5 or 6 human avatars located around a circle (Fig. 1c, d and e). Avatars differed in their morphology and the color of their clothes and hair, but they were all males and had the same posture, standing upright with their hands in front of them, as if they were ready to receive and throw a ball. From one trial to another, avatars had the same position around the circle and only the color of their T-shirt changed (white, red or blue). Visual stimuli were presented in a high-resolution head-mounted display (LDI-100BE, Sony) with a 30° horizontal field-of-view.

2.1.3.1. Third-person perspective taking task. Participants performed the virtual ball-tossing game from the perspective of a distant

avatar. Six avatars were shown in the visual scene: one was facing the participants (180°), one was in front of the participants and seen from the back (0°), two were on their right (60° and 120°), and two on their left (�60° and �120°) (Fig. 1c). Participants were instructed to adopt the visuo-spatial perspective of the avatar holding the ball (identified by a blue T-shirt) and from this

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185178

position to throw the ball to the avatar wearing a red T-shirt. The avatar with the blue T-shirt was located either at 760°, 7120° or 180° (but never at 0°, to avoid confusion with the 1PP taking task), resulting in a total of 5 angles of visuo-spatial perspective taking. For each angle of perspective taking, the ball could be thrown to the avatar wearing the red T-shirt located 60° or 120° to the right or left of this avatar (4 positions). The avatars with the red and blue T-shirts were never facing each other. A total of 20 visual stimuli were created (5 angles of visuo-spatial perspective � 4 positions of the avatar with the red T-shirt). Participants indicated as quickly and accurately as possible whether they had to throw the ball to the right or left. Participants responded on a response pad (RB-830, Cedrus Corporation, San Pedro, USA) by pressing a button with their right middle finger to throw the ball to their right, or by pressing a button with their right index finger to throw the ball to their left.

Each trial onset was synchronized with the chair rotation (Fig. 1a and b). A fixation cross was presented during the first 1.5 s of chair rotation, followed by the presentation of the visual scene during which participants performed the ball-throwing task. Thus, visual stimuli were presented during the acceleration phase, when the chair velocity was 30 °/s. After participants answered, a video was presented during 3 s, showing the ball being thrown in the direction of the avatar with the red T-shirt if the response was correct, or in the opposite direction if the response was incorrect. There was a 15 s break between two consecutive trials to allow the suppression of the rotatory post-effect before the next trial. Two consecutive trials were separated by 23 s. The experiment was a 5 Angle of visuo-spatial perspective taking (i.e. 5 locations of the avatar with a blue T-shirt: 760°, 7120° and 180°) � 4 Location of the avatar with a red T-shirt (760° and 7120° with respect to the avatar with a blue T-shirt) � 2 Direction of chair rotation (clockwise vs. counterclockwise chair rotation) design, with 2 repetitions of each of the 40 stimulus combinations. Thus, each participant completed 80 trials divided into 4 blocks of 20 trials presented in a pseudorandom order.

2.1.3.2. First-person perspective taking task. Five avatars were shown in the visual scene, one was facing the participants (180°), two were on their right (60° and 120°) and two on their left (�60° and �120°) (Fig. 1d). The hands holding the ball in the front of the scene indicated that the participants had to toss the virtual ball from their first-person perspective. As for the 3PP taking task, participants were instructed to throw the ball to the avatar wearing a red T-shirt. The avatar with the red T-shirt was located at 760° or 7120° with respect to the participant's viewpoint, resulting in 4 possible visual stimuli. Participants indicated as quickly and accurately as possible whether they had to throw the ball to their right or left, as indicated above.

Each trial was synchronized with the chair rotation and com- prised the same sequence of events as the 3PP taking task (Fig. 1a and b). After participants responded, a video was also shown during 3 s. Each of the four visual stimuli was repeated twice for counterclockwise and clockwise chair rotations so that each par- ticipant completed 16 trials.

2.1.3.3. Visual detection task. A visual detection task was used to control for potential influence of chair rotation on low-level visual processes and redirection of spatial attention (Ferrè et al., 2014a; Figliozzi et al., 2005). Participants realized the same visual detec- tion task during the rotation of the chair, and while sitting on the chair without any rotation. Visual stimuli depicted a scene with six avatars located around a circle and wearing a white T-shirt (Fig. 1e). In 10 trials, one of the avatars held a ball in his hands while in 10 other trials, there was no ball in the visual scene. Participants were instructed to detect as quickly and accurately as

possible whether there was a ball in the visual scene. Half of the participants had to press a button with their right middle finger as soon as they detected a ball, or with their right index finger when they detected there was no ball, and vise versa for the other par- ticipants. A fixation cross was presented for 1.5 s when the chair started to move, then a stimulus was presented until the partici- pant gave a response, and a fixation cross was presented until the end of the chair rotation. A fixation cross was presented on the screen for 15 s between two consecutive trials. Trials were pre- sented in a pseudorandom order, half of them during counter- clockwise chair rotations and the other half during clockwise chair rotations. The same sequence of events, with the same timing, was presented when the chair did not move. Half of the participants started with a block of 20 trials without chair rotation followed by a block of 20 trials with chair rotation. For the other participants, the blocks were presented in the reverse order. Thus all partici- pants completed 40 trials.

2.1.4. Experimental procedures Participants were comfortably installed on the chair, the head-

mounted display was put on, and they were secured with a belt. The participants' head was maintained by a support to prevent any movement during rotation and to optimize the stimulation of the lateral semicircular canals. They also wore earphones emitting white noise to eliminate any potential motion cues from auditory signals. All recordings were conducted in a darkened room.

Visual stimuli were presented in the head-mounted display using SuperLab 4.5 (Cedrus Corporation, San Pedro, USA). Re- sponses were given on a response pad (RB-830, Cedrus Corpora- tion, San Pedro, USA) placed on the participant's lap. All partici- pants first completed the visual detection task. Ten participants began with the 1PP taking task and 10 began with the 3PP taking task. Before the experiment proper, they completed a training session for each task consisting in a random selection of 20 trials for familiarization with the response pad and experimental pro- cedures. Participants took short breaks after each task.

At the end of the experiment, participants completed the Si- mulator Sickness Questionnaire, which quantifies on a four-point scale ranging from “absent” to “strong” (with 1¼no effect, 2¼weak effect, 3¼medium effect and 4¼strong effect) the oc- currence of 15 symptoms such as nausea, headache and vertigo (Kennedy et al., 1993; Van Elk and Blanke, 2014).

2.1.5. Data analysis We calculated the mean response time and percentage of cor-

rect answers. Trials yielding incorrect answers were discarded from the analysis of the response times. Data were analyzed using repeated-measures ANOVA (Statistica, Version12 SP3, StatSoft Inc.) and post-hoc analyses of significant interactions were conducted using planned comparisons with an a priori alpha level of 0.05.

2.2. Results

2.2.1. Third-person perspective taking task Response times were first analyzed using repeated-measures

ANOVA with the Direction of chair rotation (clockwise, counter- clockwise) and Angle of visuo-spatial perspective taking (�60°, �120°, 180°, 120°, 60°) as within-subject factors. Results showed a main effect of the angle (F4,76¼32.59, po0.0001, p

2η ¼0.63) (Fig. 2a), indicating that participants performed a mental rotation to take the perspective of the avatar, with longer response times for larger angles of visuo-spatial perspective taking. There was no significant effect of Direction of chair rotation and no interaction of Direction of chair rotation � Angle of visuo-spatial perspective taking.

950

1000

1050

1100

Clockwise perspective taking

Counterclockwise perspective taking

re sp

on se

ti m

e (m

s)

Clockwise chair rotation Counterclockwise chair rotation

0

5

10

15

20

25

800

900

1000

1100

1200

1300

1400

er ro

rs (

% )

re sp

on se

ti m

e (m

s)

response time % errors

Fig. 2. Results (Experiment 1). (A) Mean response times and percentage of errors as a function of the angle of visuo-spatial perspective taking. Negative and positive angles represent clockwise and counterclockwise rotations of the participant's viewpoint, respectively. (B) Mean response times are shown for the significant interaction of Direction of chair rotation � Direction of visuo-spatial perspective. Vertical bars depict standard errors of the mean (SEM).

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185 179

To further analyze the influence of vestibular stimulation on perspective taking, we calculated a second ANOVA without the 180° angle, as this angle is not associated with a clear direction of mental imagery. Thus, response times were analyzed using re- peated-measures ANOVA with the Direction of chair rotation (clockwise, counterclockwise), Direction of visuo-spatial perspec- tive taking (clockwise, counterclockwise) and Angle of visuo-spa- tial perspective taking (60°, 120°) as within-subject factors. Inter- estingly, we found a significant interaction of Direction of chair rotation � Direction of visuo-spatial perspective (F1,20¼9.46, po0.01, p

2η ¼0.33), indicating that participants responded differ- ently when the chair rotated in the same as vs. opposite direction to the mental rotation used to simulate an avatar's viewpoint (Fig. 2b). When participants took the perspective of an avatar lo- cated on the right side of the circle, the shortest path for the mental rotation would predict that they mentally rotated their whole body in the counterclockwise direction, and when the avatar was on the left side of the circle, that they realized a clockwise mental rotation. A first post-hoc analysis indicated that when participants realized counterclockwise perspective taking (i.e. avatars located on the right side of the visual scene), response times were significantly longer for clockwise when compared to

counterclockwise chair rotations (paired t-test, po0.01). By con- trast, when participants realized clockwise perspective taking (i.e. avatars located on the left side of the visual scene), response times did not differ between clockwise and counterclockwise chair ro- tations (p¼0.14). A second post-hoc analysis showed that clock- wise chair rotations facilitated 3PP taking for congruent trials (i.e. faster responses for clockwise vs. counterclockwise perspective taking; po0.05) whereas counterclockwise chair rotations did not (p¼0.39). Results showed no main effect of Direction of chair ro- tation (F1,20¼2.33, p¼0.14, p

2η ¼0.11), Direction of visuo-spatial perspective taking (F1,20¼0.92, p¼0.35, p

2η ¼0.05) and a trend for the Angle of visuo-spatial perspective taking (F1,20¼3.97, p¼0.06,

p 2η ¼0.17). Analysis of the percentage of errors revealed a significant main

effect of Angle of visuo-spatial perspective taking (F1,19¼5.13, po0.05, p

2η ¼0.21) (Fig. 2a). There was no other significant main effect or interaction.

2.2.2. First-person perspective taking task Response times did not differ for counterclockwise (mean7

SEM: 532715 ms) and clockwise (538715 ms) chair rotations (two-sided paired t-test, p¼0.59).

Since participants made less than 0.3% of errors, we did not analyze errors.

2.2.3. Visual detection task Analysis of response times was performed using a repeated-

measures ANOVA with two within-subjects factors: Direction of chair rotation (clockwise, counterclockwise) and Position of the ball in the scene (left, right side of the screen). There was no main effect of the Direction of chair rotation (counterclockwise: 728716 ms; clockwise: 747721 ms; F1,19¼0.87, p¼0.36, p

2η ¼0.04), Position of the ball in the scene (F1,19¼0.29, p¼0.59, p

2η ¼0.02) and no in- teraction between these factors (F1,19¼0.90, p¼0.35, p

2η ¼0.05). Therefore, the rotation of the chair did not redirect visuo-spatial attention to the right or left side of the screen.

An additional ANOVA showed no difference in response times when the chair rotated (738713 ms) or was stationary (737717 ms; F1,19¼0.71, p¼0.41, p

2η ¼0.04). Thus, vestibular sti- mulation did not impair visual processing of the scene.

Since participants made less than 3.8% of errors, we did not analyze errors.

2.2.4. Analysis of the spatial and stimulus-response compatibility effects

For the 3PP taking task, it was important to control for the position of the response button on the keyboard (left or right), i.e. the spatial and stimulus-response compatibility effects (Bosbach et al., 2004; Lu and Proctor, 1995). Indeed, the interaction found between the Direction of chair rotation and the Direction of visuo- spatial perspective taking could be modulated by spatial compat- ibility effects between the position of the blue avatar in the scene, the response button and the direction of chair rotation (Van Elk and Blanke, 2014). Therefore, repeated-measures ANOVAs were calculated with three within-subjects factors: the Direction of chair rotation (clockwise, counterclockwise), the Position of the blue avatar in the scene (left, right side of the screen) and the Response button position (left, right). There was no main effect of the Position of the blue avatar (F1,19¼0.55, p¼0.47, p

2η ¼0.03), Di- rection of chair rotation (F1,19¼2.40, p¼0.14, p

2η ¼0.11) and Re- sponse button position (F1,19¼1.58, p¼0.22, p

2η ¼0.08). As de- scribed above, a significant interaction was found between the Direction of chair rotation and the Position of the blue avatar

1 object 2 object

Same objects

Different objects

Answer

2 object

+ +

A cc

el er

at io

n

) V

el oc

ity

Time (s) Onset

of rotation

4 8 End

of rotation

1.5 23 0

1 object

Fig. 3. Experimental setup and stimuli for Experiment 2. (A) Eight-second mo- tion profile used for whole-body rotations in the yaw plane. (B) Each trial consisted of an 8 s clockwise or counterclockwise rotation during which participants per-

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185180

(F1,19¼8.30, po0.01, p 2η ¼0.30), because the position of the blue

avatar indicates the direction of the visuo-spatial perspective taking. By contrast, the ANOVA did not show an interaction be- tween the Direction of chair rotation and the Response button position (F1,19¼0.80, p¼0.38, p

2η ¼0.04), indicating that latency to press the button was not reduced for chair rotations in the same direction or increased for rotations in the opposite direction (i.e. stimulus-response compatibility effect). It is important to note that there was no triple interaction between the Direction of chair rotation, the Position of the blue avatar and the Response button position (F1,19¼1.34, p¼0.26, p

2η ¼0.07). Altogether, the present results indicate no spatial compatibility effect in the 3PP taking task.

2.2.5. Questionnaire data Participants reported low vegetative symptoms due to the ro-

tation of the chair (average intensity¼1.770.9). The most fre- quently reported symptoms were tiredness (2.570.9), general discomfort (2.271.0) and difficulty concentrating (2.170.9), suggesting only moderate influence of these side effects on the participants' performances.

2.3. Conclusion

The main finding of Experiment 1 was that participants changed perspective faster when they rotated on the chair in the same direction as the mental rotation needed to take the avatar's viewpoint. Thus, signals from the horizontal semicircular canals influenced 3PP taking in a direction-specific way. Yet the question remains as to whether the direction-specific influence of the vestibular stimulation revealed in Experiment 1 is specific to visuo- spatial perspective taking (a form of egocentric mental transfor- mation) or reflects a more general influence of vestibular stimu- lation on mental imagery (including object-based mental imagery) as suggested by previous results (Mast et al., 2006; Lenggenhager et al., 2008; Péruch et al., 2011). In Experiment 2, participants were also passively rotated on the chair while they realized mental rotations of 3D objects, which did not involve perspective taking. We also created trials for which the direction of chair rotation and direction of the mental rotation of 3D objects were congruent and incongruent.

formed mental rotation of a 3D object. After an accommodation period of 0.5 s, a first 3D object was shown for 2.5 s during which a superimposed hand indicated the direction (clockwise/counterclockwise) and amplitude (25°, 50° or 75°) of the mental rotation. A second 3D object was presented and participants indicated whether it was similar or different from the first object. After participants an- swered, a fixation cross was presented until the end of the chair rotation and during the 15 s inter-stimulus interval. (C) Examples of 3D objects shown as first and second objects: the first line shows two identical objects (with an angular disparity of 75°) and the second line shows different objects.

3. Experiment 2

3.1. Materials and methods

3.1.1. Participants In this second experiment, 13 right-handed healthy volunteers

participated (7 females and 6 males, mean age7SD: 2474 years; Handedness inventory: 82725%).

3.1.2. Vestibular stimulation To allow comparisons with Experiment 1, we used the same

rotatory chair and the same 8-s motion profile (Fig. 3a).

3.1.3. Visual stimuli and tasks Participants were involved in a 3D objects mental rotation task

(Hoppe et al., 2012; Shepard and Metzler, 1971). A novel aspect of this task was to cue participants to mentally rotate the re- presentation of a 3D object either in a clockwise or counter- clockwise direction, so as to create predictable conflicts between physical body rotation and mental rotation of the 3D object. Visual stimuli consisted of high-resolution photographs of complex 3D geometrical objects in gray colors presented in a head-mounted

display (Fig. 3c). Twelve 3D objects were created and photo- graphed in a chosen position (0°) and were again photographed after being rotated by 25°, 50° and 75° around their vertical axis in the clockwise and counterclockwise directions.

A 3D object was first presented at the center of the screen for 0.5 s, then a virtual hand with the index finger pointing toward the participant appeared at the center of the 3D object (in half of the trials a right or left hand was presented; Fig. 3c). The hand rotated in the horizontal plane to inform participants about the direction (clockwise or counterclockwise) and angle (25°, 50° and 75°) of the mental rotation of the 3D object. Participants were trained to mentally rotate the representation of the object in the direction indicated by the hand as long as the hand rotated. The duration of the hand rotation was 2.5 s for all angles of mental rotation. Then, a 3D object was presented at the center of the screen and parti- cipants indicated as quickly and accurately as possible whether

0

10

20

30

40

50

900

1100

1300

1500

1700

er ro

rs (

% )

re sp

on se

ti m

e (m

s)

Response time % Errors

1150

1250

1350

1450

Clockwise mental rotation

Counterclockwise mental rotation

re sp

on se

ti m

e (m

s)

Clockwise chair rotation Counterclockwise chair rotation

Clockwise mental rotation

Counterlockwise mental rotation

Fig. 4. Results (Experiment 2). (A) Mean response times and percentage of errors as a function of the angle of mental rotation of the 3D objects. Negative and po- sitive angles represent clockwise and counterclockwise object rotations, respec- tively. (B) Mean response times show no significant interaction of the Direction of chair rotation and Direction of mental rotation. Vertical bars depict standard errors of the mean (SEM).

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185 181

this object was identical to (two thirds of the trials), or different from (one third of the trials), the first 3D object after it had been rotated in the direction indicated by the hand (Fig. 3c). Participants responded on a response pad (RB-830, Cedrus Corporation) by pressing on a button defined as the response “same object” with their right middle finger, or by pressing on another button defined as the response “different object” with their right index finger. The position of the response buttons was counterbalanced across participants.

Each trial onset was synchronized with the chair rotation (Fig. 3a and b). A fixation cross was presented during the first 500 ms of chair rotation, followed by the presentation of the 3D object for 1 s. Immediately after, the hand was presented on the object and rotated in the clockwise or counterclockwise direction during 2.5 s. Thus, participants started mental rotation during the accel- eration phase, when the chair velocity was 30°/s, that is exactly as in Experiment 1. The second picture was presented until partici- pants answered. Participants performed 144 trials: the first and second objects were identical for 96 trials and different for 48 trials. The experiment was a 3 Angle of mental rotation (25°, 50°, 75°) � 2 Direction of mental rotation (clockwise, counter- clockwise) � 2 Direction of chair rotation (clockwise, counter- clockwise) design, with 12 different 3D objects. There was a 15 s break between two consecutive trials to allow the suppression of the rotatory post-effect before the next trial. 144 trials were ran- domly allocated to six blocks of 24 trials. The order of presentation of the six blocks was randomized across participants. Participants took short breaks after they completed each block.

3.1.4. Experimental procedures We used similar experimental procedures as in Experiment 1.

The onset of chair rotation and the onset of visual stimuli were synchronized using SuperLab 4.5. Before the experiment proper, participants completed a training session consisting of a random selection of 24 trials to become familiar with the mental imagery of 3D objects. Participants completed the Simulator Sickness Questionnaire after the experiment.

3.1.5. Data analysis We calculated the mean response times and mean percentage

of correct answers for each combination of Direction of chair ro- tation, Direction of mental rotation and Angle of mental rotation. Trials yielding incorrect answers were discarded from the analysis of the response times. Data were analyzed using repeated-mea- sures ANOVA and post-hoc analyses of significant interactions were conducted using planned comparisons with an a priori alpha level of 0.05.

3.2. Results

Response times were analyzed using repeated-measures AN- OVA with the Direction of chair rotation (clockwise, counter- clockwise), Direction of mental rotation (clockwise, counter- clockwise) and Angle of mental rotation (25°, 50° and 75°) as within-subject factors. As expected, the ANOVA revealed a sig- nificant main effect of the Angle of mental rotation (F2,24¼9.10, po0.005, p

2η ¼0.43), showing longer response times for larger angles of 3D object rotation (Fig. 4a). This result indicates that participants actually performed a mental rotation before they decided if the first and second objects were the same, which is compatible with experiments showing simultaneously two 3D objects (Shepard and Metzler, 1971; Péruch et al., 2011).

In contrast with the results from Experiment 1, we did not find a significant interaction of Direction of chair rotation � Direction of mental rotation (F2,24¼0.04, p¼0.84, p

2η ¼0.07) (Fig. 4b),

suggesting that vestibular stimulation has no directional influence on the mental rotation of 3D objects. In addition, there was no main effect of Direction of chair rotation (F1,12¼0.80, p¼0.93,

p 2η ¼0.01) and of Direction of mental rotation (F1,12¼0.65, p¼0.44, p 2η ¼0.05). Analysis of the percentage of errors revealed a significant main

effect of Angle of mental rotation (F2,24¼9.0, po0.005, p 2η ¼0.43),

participants being less accurate for larger angles of mental rota- tion. There were no other significant main effects or interaction (Fig. 4a).

Participants reported low vegetative symptoms due to the ro- tation of the chair (average intensity¼1.670.8). The most fre- quently reported symptoms were eyestrain (2.270.8), headache (2.171.0) and general discomfort (1.970.8).

4. Discussion

We investigated how natural whole-body rotations in the yaw plane influenced 3PP taking and object-based mental rotation in two separate experiments. 3PP taking, but not mental rotation of 3D objects, was influenced by the vestibular stimulation. We in- terpret these results in line with recent findings showing that

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185182

bodily signals – including vestibular semicircular canal signals – are required for embodied cognitive processing.

4.1. Vestibular stimulation influences 3PP taking in a direction-spe- cific way

The main finding of Experiment 1 is that response time analysis showed a significant interaction between the direction of chair rotation and the direction of 3PP taking. We found that vestibular information influenced 3PP taking in a direction-specific way: participants were faster to adopt the perspective of a distant avatar when both physical rotations and simulated viewpoint/ own-body rotations were congruent. Better performance for con- gruent physical and simulated viewpoint/own-body rotations suggest shared sensorimotor mechanisms for 3PP taking and vestibular self-motion coding. Experiment 1 provides the first de- monstration that vestibular information is involved in a task during which participants are explicitly required to take another person's viewpoint. A previous study (Van Elk and Blanke, 2014) also used natural vestibular stimulation on a rotating chair and showed a direction-specific influence of vestibular stimulation on own-body mental transformations (i.e. participants judged whe- ther the colored hand of an avatar was a left or right hand), but their task differed from ours as it did not explicitly require to lo- cate objects according to the viewpoint of a distant avatar. In ad- dition, in that study, some participants mentally rotated their body in space, while others used an opposite, object-centered, mental rotation strategy (Van Elk and Blanke, 2014). Our results and those from Van Elk and Blanke (2014) allow to conclude that vestibular signals are used to mentally simulate own-body mental transfor- mations and to change visuo-spatial perspective (viewpoint rotations).

It is difficult to directly compare results from Experiment1 with those from studies that measured how caloric and galvanic ves- tibular stimulation influenced mental imagery. To the best of our knowledge, these studies did not report a direction-specific effect of vestibular stimulation on mental imagery of letters (Mast et al., 2006), non-corporeal objects, bodies (Falconer and Mast, 2012; Lenggenhager et al., 2008), or perspective taking (Dilda et al., 2012a). These studies showed more global effects of vestibular stimulation (i.e. increase or decrease in response times as com- pared to a baseline condition). Yet, it should be noted that sen- sations of self-motion evoked by caloric and galvanic stimulations are variable across participants (e.g. Kolev, 2001) and have no physiological equivalent. In the present study, the fact that 3PP taking involved mental imagery in the same (yaw) plane as chair rotations may have favored the direction-specific influence of the vestibular stimulation on 3PP taking. This should motivate the use of natural vestibular stimulation on full-body motion platforms in future investigations of the vestibular influence on cognition. This method circumvents issues related to the variability of responses to caloric and galvanic vestibular stimulation and allows testing participants in more ecological and well-controlled situations (review in Palla and Lenggenhager, 2014).

The interaction of Direction of chair rotation � Direction of vi- suo-spatial perspective found in Experiment 1, showed a stronger congruency effect for counterclockwise 3PP taking (i.e. for avatars located on the right side of space), and clockwise chair rotations modulated more strongly 3PP taking. Such asymmetry was not reported in a previous study by Van Elk and Blanke (2014), and in the present case it may be related to complex interactions be- tween the participant's handedness, hemispheric dominance of vestibular processing and of visuo-spatial perspective taking. First, only right-handed participants were tested. Right-handed parti- cipants process right visual space in a different way than left visual space (e.g. Rubichi and Nicoleti, 2006; Casasanto and Chrysikou,

2011) and handedness also influences own-body mental trans- formation (Gardner and Potts, 2010). For example, right-handed participants answer faster to visual stimuli presented in their right visual space. Therefore, in the present study, taking the perspec- tive of avatars located on the right side of space may be different in right-handed participants. Response times were similar for avatars located on the right vs. left side of the visual scene. After Experi- ment 1 participants filled out a questionnaire about the experi- enced difficulty of 3PP taking for each avatar position. They tend to judge easier to take the perspective of avatars located on the right side of space (Wilcoxon signed-rank test: p¼0,075). Second, neuroimaging studies revealed asymmetrical cortical processing of vestibular signals in right-handed participants. There is an overall vestibular dominance in the right hemisphere in right-handers and the hemisphere ipsilateral to the stimulated ear is also more activated (Dieterich et al., 2003). Third, there is evidence that such vestibular dominance in the right hemisphere, in conjunction with a marked dominance of spatial functions in the right vs. left cer- ebral hemispheres, may account for different effects of right and left vestibular stimulation – or left and right vestibular loss – on body perception, mental imagery, spatial navigation and space perception (e.g. Lenggenhager et al., 2008; Hüfner et al., 2007; Lopez et al., 2010b; Saj et al., 2013; Toupet et al., 2014; Arshad et al., 2015). In conclusion, while it is to date unclear how hand- edness, vestibular processing and spatial processing interact, we propose that it may drive the interaction between the Direction of chair rotation and Direction of 3PP taking. Further studies will be necessary to understand the relations between handedness and the way vestibular signals influence visuo-spatial processing (for a recent account, see Arshad et al., 2013).

4.2. Vestibular contribution to perspective taking vs. object-based mental imagery

The use of identical stimulation parameters and timing in Ex- periments 1 and 2 allows us to directly compare the results and to disentangle the influence of vestibular signals on 3PP taking vs. 3D object mental rotation (i.e. object-based mental imagery). While a direction-specific influence of the vestibular stimulation was found for 3PP taking in Experiment 1, this was not the case for mental rotation of 3D objects in Experiment 2. This finding in- dicates that the vestibular influence found for 3PP taking does not rely on a mere vestibular influence on any type of mental imagery, but likely reflects a specific vestibular contribution to self-centered mental imagery. Previous studies that used caloric and galvanic vestibular stimulation also failed to report a direction-specific in- fluence of vestibular signals on object-based mental imagery (Dilda et al., 2012; Falconer and Mast, 2012; Lenggenhager et al., 2008). However, the present study is the first where participants performed 3D objects mental rotations during natural whole-body rotations and that systematically manipulated the congruency between the direction of mental rotation and vestibular stimula- tion in a same plane.

The negative results from Experiment 2 lead us to propose that vestibular signals play a predominant role in mental imagery of one's own viewpoint and self-location (self-centered imagery) rather than in object-centered imagery. This proposition is sup- ported by clinical observations in epileptic and brain-damaged patients showing that abnormal vestibular processing in the brain can evoke the experience of an altered, disembodied, visuo-spatial perspective (Blanke et al., 2002, 2004; Lopez et al., 2008). A recent investigation conducted by Ferrè et al. (2014b) provides further evidence that vestibular signals contribute to the sensorimotor mechanisms of perspective taking. These authors showed that low intensity galvanic vestibular stimulation increased the likelihood that ambiguous letters drawn on the participant's forehead were

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185 183

perceived from a 1PP, which is probably in line with the natural role of vestibular signals in anchoring the self to the body. Finally, data collected in patients with peripheral vestibular disorders also indicated that mental own-body transformations and mental na- vigation were impaired when compared to healthy participants (Candidi et al., 2013; Grabherr et al., 2011; Péruch et al., 2011). Yet, two of these studies revealed that object-centered imagery was also altered after vestibular disorders (Candidi et al., 2013; Péruch et al., 2011). This suggests that pathological vestibular processing may be more detrimental to mental imagery abilities than the natural vestibular stimulation used in Experiment 2, or that stronger vestibular stimulation may be necessary to interfere with object-centered imagery.

A different implication of vestibular signals in 3PP taking and object-based mental imagery can be related to the fact that both categories of mental imagery involve different cognitive mechan- isms (e.g. Kozhevnikov and Hegarty, 2001) and rely differently on sensorimotor processing. 3PP taking is an implicit cognitive pro- cess (Samson et al., 2010; Tversky and Hard, 2009) that relies strongly on the processing of bodily signals (for an alternative account, see Heyes (2014) and Santiesteban et al. (2014) and be- low). While there was evidence that postural signals (the partici- pant's whole-body posture) are taken into account for 3PP taking (Kessler and Thomson, 2010; Kessler et al., 2014), our findings demonstrate that vestibular signals are also involved in the mental simulation of another person's viewpoint. 3PP taking can be con- sidered a strongly ‘embodied’ cognitive process as geometrical transformations (i.e. rotations and translations) used to compute another person's viewpoint rely on a combination of visual, so- matosensory and vestibular signals. By contrast, object-based mental imagery has different characteristics (Zacks and Michelon, 2005; Zacks and Tversky, 2005): it does not require transformation of the participant's bodily reference and is probably less depen- dent upon bodily signals. For example, Kessler and Thomson (2010) showed that manipulating the participant's body posture influenced 3PP taking, but not objects rotation. Similarly, manip- ulating postural information (Ter Horst et al., 2012), as well as imagined paralysis (Hartmann et al., 2011), influenced mental imagery of body parts but not of letters.

3PP taking and object-based mental imagery have been related to partly different brain mechanisms (Zacks, 2008; Zacks et al., 2003). Neuroimaging studies of 3PP taking revealed the implica- tion of various regions, including the temporo–parietal junction, intraparietal sulcus, insula and precuneus (Committeri et al., 2004; Vogeley et al., 2004; Blanke et al., 2005; David et al., 2006; Kockler et al., 2010; Lambrey et al., 2012). Of note, these regions overlap with the main areas of the vestibular cortical network, such as the temporo–cortex, insula, parietal operculum and intraparietal sul- cus (Bottini et al., 1994; Dieterich et al., 2003; Kahane et al., 2003; Lopez et al., 2012; Mazzola et al., 2014; review in Lopez and Blanke, 2011). In a recent fMRI study, Zu Eulenburg et al. (2013) showed that remembering own-body rotations activated vestibu- lar regions, showing that recalling self-rotations involved vestib- ular regions. Object-based mental imagery revealed the implica- tion of the fronto–parietal cortex (Zacks and Michelon, 2005). We propose that the vestibular effects found in Experiment 1 may be explained within a similar framework, i.e. mental simulation of rotations and translations operated to take the viewpoint of a distant avatar activate vestibular regions such as the temporo– parietal junction and intraparietal sulcus.

We finally note that recent findings (Heyes, 2014; Santiesteban et al., 2014) may contradict the view that 3PP taking strongly rely on embodied processing as proposed here. These studies have suggested that explicit 3PP taking (as used in Experiment 1) and implicit 3PP taking (i.e. spontaneous changes of the viewpoint) reflect domain-general (attentional) processing rather than spatial

perspective taking. In these studies, participants were faster to count the number of dots visible by a distant avatar when this number was congruent with the number of balls visible from their viewpoint, an effect referred to as “egocentric intrusion” (Samson et al., 2010). As the same effect was reported when an avatar was replaced by the arrow or a rectangle, it was proposed that spatial- attentional processing related to the orientation of the avatar's body and gaze drives the egocentric intrusion effect and that perspective taking was not crucially involved (Santiesteban et al., 2014). Of note, behavioral and neuroimaging studies have recently challenged these results showing that implicit and explicit 3PP taking is a social mechanism involving Theory of Mind brain areas (e.g. Nielsen et al., 2015; Schurz et al., 2015). Altogether, these results indicate the necessity to control for attentional factors, such as the influence of gaze direction on visuo-spatial attention (see, e.g. Frischen et al., 2007), when investigating 3PP taking. In our study, the presentation of five avatars located around a circle and looking in opposite directions has likely not redirected spatial attention to one side of space (as confirmed by the Visual detec- tion task; see also Capozzi et al. (2014)) and this cannot explain the direction-specific influence of the vestibular stimulation on 3PP taking.

4.3. Spatial attention and stimulus-response compatibility effects

Additional analyses conducted for Experiment 1 ruled out any influence of spatial attention and stimulus-response compatibility effects that could have confounded the present data. First, we showed that the participant's ability to process images was not altered by chair rotations. This was demonstrated by similar re- sponse times to detect balls in the visual scene when the chair was rotating and immobile (see results from the visual detection task). Thus, residual eye movements (i.e. vestibulo-ocular reflex) not suppressed by fixation, or potential neurovegetative effects evoked by chair rotations did not alter the participants' performance. In both experiments, participants reported only moderate side effects due to chair rotations. Second, we analyzed the stimulus-response compatibility effect (i.e. Simon effect; reviewed in Lu and Proctor (1995)) as vestibular stimulations can redirect spatial attention toward the direction of chair rotation (Figliozzi et al., 2005) and mental imagery and spatial compatibility effects are closely in- tertwined (Gardner and Potts, 2011; May and Wendt, 2012, 2013; Van Elk and Blanke 2014). Our results showed that the direction- specific influence of vestibular stimulation on 3PP is neither re- lated to redirection of spatial attention nor to stimulus-response compatibility effects. Those effects, originally described during detection of simple visual and tactile stimuli (Figliozzi et al., 2005), are therefore not apparent for more complex visuo-spatial tasks. Finally, the long duration of each trial (8 s) and inter-stimulus interval (15 s) precluded testing the same participants during 3PP taking tasks while the chair was rotating and immobile. The ab- sence of data from a ‘no-rotation’, baseline, condition does not allow to draw conclusion about the general influence of natural vestibular stimulation on 3PP taking and object-based imagery. Instead, the present results reveal direction-specific influence of the vestibular signals on 3PP taking.

5. Conclusions and perspectives

In summary, the present results indicate that vestibular in- formation is part of the bodily signals processed to simulate someone else's visuo-spatial perspective. Thus, vestibular signals seem crucial for one of the most fundamental aspects of human social cognition: interacting with others. Chair rotations used in the present experiments naturally stimulate the horizontal

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185184

semicircular canals (positioned in the yaw plane). To date, semi- circular canal signals have been involved in a wide range of cog- nitive functions including memory, interpretation of self- and other-motion and location (Lopez et al., 2013; Van Elk and Blanke, 2012), decision-making (McKay et al., 2013; Preuss et al., 2014b), emotions and affective control (Preuss et al., 2014a; Winter et al., 2013). Our results add to the implication of the semicircular canal signals to self-consciousness and spatial cognition (Lenggenhager and Lopez, 2015a). Whether otolithic vestibular information also contributes to perspective taking, cognitive and affective control should be the topic of future investigations.

Acknowledgments

The research leading to these results has received funding from the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme (FP7/2007-2013) under REA grant agreement number 333607 (‘BODILYSELF, vestibular and multisensory investigations of bodily self-consciousness’). We are grateful to Guy Escoffier and Dany Paleressompoulle for their technical assistance with the rotating chair and electronic device (synchronization between chair rotations and visual stimulus presentations), and to Dr. Isabelle Virard for helpful comments on the manuscript.

References

Amorim, M.A., Isableu, B., Jarraya, M., 2006. Embodied spatial transformations: “body analogy” for the mental rotation of objects. J. Exp. Psychol. Gen. 135, 327–347.

Arshad, Q., Nigmatullina, Y., Bronstein, A.M., 2013. Handedness-related cortical modulation of the vestibular-ocular reflex. J. Neurosci. 33, 3221–3227.

Arshad, Q., Patel, M., Goga, U., Nigmatullina, Y., Bronstein, A.M., 2015. Role of handedness-related vestibular cortical dominance upon the vestibular-ocular reflex. J. Neurol. 262, 1069–1071.

Arzy, S., Overney, L.S., Landis, T., Blanke, O., 2006. Neural mechanisms of embodi- ment: asomatognosia due to premotor cortex damage. Arch. Neurol. 63, 1022–1025.

Blanke, O., 2012. Multisensory brain mechanisms of bodily self-consciousness. Nat. Rev. Neurosci. 13, 556–571.

Blanke, O., Landis, T., Spinelli, L., Seeck, M., 2004. Out-of-body experience and au- toscopy of neurological origin. Brain 127, 243–258.

Blanke, O., Metzinger, T., 2009. Full-body illusions and minimal phenomenal self- hood. Trends Cogn. Sci. 13, 7–13.

Blanke, O., Mohr, C., 2005. Out-of-body experience, heautoscopy, and autoscopic hallucination of neurological origin Implications for neurocognitive mechan- isms of corporeal awareness and self-consciousness. Brain Res. Brain Res. Rev. 50, 184–199.

Blanke, O., Mohr, C., Michel, C.M., Pascual-Leone, A., Brugger, P., Seeck, M., Landis, T., Thut, G., 2005. Linking out-of-body experience and self processing to mental own-body imagery at the temporoparietal junction. J. Neurosci. 25, 550–557.

Blanke, O., Ortigue, S., Landis, T., Seeck, M., 2002. Stimulating illusory own-body perceptions. Nature 419, 269–270.

Bläsing, B., Brugger, P., Weigelt, M., Schack, T., 2013. Does thumb posture influence the mental rotation of hands? Neurosci. Lett. 534, 139–144.

Bosbach, S., Prinz, W., Kerzel, D., 2004. A Simon effect with stationary moving stimuli. J. Exp. Psychol. Hum. Percept. Perform. 30, 39–55.

Bottini, G., Sterzi, R., Paulesu, E., Vallar, G., Cappa, S.F., Erminio, F., Passingham, R.E., Frith, C.D., Frackowiak, R.S., 1994. Identification of the central vestibular pro- jections in man: a positron emission tomography activation study. Exp. Brain Res. 99, 164–169.

Brugger, P., 1997. Illusory reduplication of one's own body: phenomenology and classification of autoscopic phenomena. Cogn. Neuropsychiatry 2, 19–38.

Candidi, M., Micarelli, A., Viziano, A., Aglioti, S.M., Minio-Paluello, I., Alessandrini, M., 2013. Impaired mental rotation in benign paroxysmal positional vertigo and acute vestibular neuritis. Front. Hum. Neurosci. 7, 783.

Capozzi, F., Cavallo, A., Furlanetto, T., Becchio, C., 2014. Altercentric intrusions from multiple perspectives: beyond dyads. PloS One 9, e114210.

Casasanto, D., Chrysikou, E.G., 2011. When left is “right”. Motor fluency shapes abstract concepts. Psychol. Sci. 22, 419–422.

Claparède, E., 1925. Note sur la localisation du moi. Arch. Psychol 19, 172–182. Committeri, G., Galati, G., Paradis, A.L., Pizzamiglio, L., Berthoz, A., LeBihan, D., 2004.

Reference frames for spatial cognition: different brain areas are involved in viewer-, object-, and landmark-centered judgments about object location. J. Cogn. Neurosci. 16, 1517–1535.

Conson, M., Mazzarella, E., Trojano, L., 2011. Self-touch affects motor imagery: a study on posture interference effect. Exp. Brain Res. 215, 115–122.

David, N., Bewernick, B.H., Cohen, M.X., Newen, A., Lux, S., Fink, G.R., Shah, N.J., Vogeley, K., 2006. Neural representations of self versus other: visual-spatial perspective taking and agency in a virtual ball-tossing game. J. Cogn. Neurosci. 18, 898–910.

Deroualle, D., Lopez, C., 2014. Toward a vestibular contribution to social cognition. Front. Integr. Neurosci. 8, 16.

Dieterich, M., Bense, S., Lutz, S., Drzezga, A., Stephan, T., Bartenstein, P., Brandt, T., 2003. Dominance for vestibular cortical function in the non-dominant hemi- sphere. Cereb. Cortex N. Y. N 1991 (13), 994–1007.

Dilda, V., MacDougall, H.G., Curthoys, I.S., Moore, S.T., 2012a. Effects of Galvanic vestibular stimulation on cognitive function. Exp. Brain Res. 216, 275–285.

Falconer, C.J., Mast, F.W., 2012. Balancing the mind: vestibular induced facilitation of egocentric mental transformations. Exp. Psychol. 59, 332–339.

Ferrè, E.R., Bottini, G., Iannetti, G.D., Haggard, P., 2013. The balance of feelings: vestibular modulation of bodily sensations. Cortex 49, 748–758.

Ferrè, E.R., Kaliuzhna, M., Herbelin, B., Haggard, P., Blanke, O., 2014a. Vestibular- somatosensory interactions: effects of passive whole-body rotation on soma- tosensory detection. PloS One 9, e86379.

Ferrè, E.R., Lopez, C., Haggard, P., 2014b. Anchoring the self to the body: vestibular contribution to the sense of self. Psychol. Sci. 25, 2106–2108.

Figliozzi, F., Guariglia, P., Silvetti, M., Siegler, I., Doricchi, F., 2005. Effects of ves- tibular rotatory accelerations on covert attentional orienting in vision and touch. J. Cogn. Neurosci. 17, 1638–1651.

Frischen, A., Bayliss, A.P., Tipper, S.P., 2007. Gaze cueing of attention: visual atten- tion, social cognition, and individual differences. Psychol. Bull. 133, 694–724.

Frith, C.D., Frith, U., 2006. The neural basis of mentalizing. Neuron 50, 531–534. Gardner, M.R., Potts, R., 2010. Hand dominance influences the processing of ob-

served bodies. Brain Cogn. 73, 35–40. Gardner, M.R., Potts, R., 2011. Domain general mechanisms account for imagined

transformations of whole body perspective. Acta Psychol. (Amst.) 137, 371–381. Grabherr, L., Cuffel, C., Guyot, J.P., Mast, F.W., 2011. Mental transformation abilities

in patients with unilateral and bilateral vestibular loss. Exp Brain Res. 209, 205–214.

Hartmann, M., Falconer, C.J., Mast, F.W., 2011. Imagined paralysis impairs embodied spatial transformations. Cogn. Neurosci. 2, 155–162.

Heydrich, L., Lopez, C., Seeck, M., Blanke, O., 2011. Partial and full own-body illu- sions of epileptic origin in a child with right temporoparietal epilepsy. Epilepsy Behav. 20, 583–586.

Heyes, C., 2014. Submentalizing: I am not really reading your mind. Perspect. Psychol. Sci. 9, 131–143.

Hoppe, C., Fliessbach, K., Stausberg, S., Stojanovic, J., Trautner, P., Elger, C.E., Weber, B., 2012. A key role for experimental task performance: effects of math talent, gender and performance on the neural correlates of mental rotation. Brain Cogn. 78, 14–27.

Hüfner, K., Hamilton, D.A., Kalla, R., Stephan, T., Glasauer, S., Ma, J., Bruning, R., Markowitsch, H.J., Labudda, K., Schichor, C., Strupp, M., Brandt, T., 2007. Spatial memory and hippocampal volume in humans with unilateral vestibular deaf- ferentation. Hippocampus 17, 471–485.

Ionta, S., Blanke, O., 2009. Differential influence of hands posture on mental rota- tion of hands and feet in left and right handers. Exp. Brain Res. 195, 207–217.

Ionta, S., Fourkas, A.D., Fiorio, M., Aglioti, S.M., 2007. The influence of hands posture on mental rotation of hands and feet. Exp. Brain Res. 183, 1–7.

Ionta, S., Perruchoud, D., Draganski, B., Blanke, O., 2012. Body context and posture affect mental imagery of hands. PloS One 7, e34382.

Ionta, S., Sforza, A., Funato, M., Blanke, O., 2013. Anatomically plausible illusory posture affects mental rotation of body parts. Cogn. Affect. Behav. Neurosci. 13, 197–209.

Kahane, P., Hoffmann, D., Minotti, L., Berthoz, A., 2003. Reappraisal of the human vestibular cortex by cortical electrical stimulation study. Ann. Neurol. 54, 615–624.

Kennedy, R.S., Lane, N.E., Berbaum, K.S., Lilienthal, M.G., 1993. Simulator sickness questionnaire: an enhanced method for quantifying simulator sickness. Int. J. Aviat. Psychol. 3, 203–220.

Kessler, K., Cao, L., O’Shea, K.J., Wang, H., 2014. A cross-culture, cross-gender comparison of perspective taking mechanisms. Proc. Biol. Sci. 281, 20140388.

Kessler, K., Rutherford, H., 2010. The two forms of visuo-spatial perspective taking are differently embodied and subserve different spatial prepositions. Front. Psychol. 1, 213.

Kessler, K., Thomson, L.A., 2010. The embodied nature of spatial perspective taking: embodied transformation versus sensorimotor interference. Cognition 114, 72–88.

Kockler, H., Scheef, L., Tepest, R., David, N., Bewernick, B.H., Newen, A., Schild, H.H., May, M., Vogeley, K., 2010a. Visuospatial perspective taking in a dynamic en- vironment: perceiving moving objects from a first-person-perspective induces a disposition to act. Conscious. Cogn. 19, 690–701.

Kolev, O.I., 2001. The directions of nystagmus and apparent self-motion evoked by caloric tests and angular acceleration. J. Vestib. Res. 11, 349–355.

Kozhevnikov, M., Hegarty, M., 2001. A dissociation between object manipulation spatial ability and spatial orientation ability. Mem. Cognit. 29, 745–756.

Lambrey, S., Doeller, C., Berthoz, A., Burgess, N., 2012. Imagining being somewhere else: neural basis of changing perspective in space. Cereb. Cortex N. Y. N. 1991 (22), 166–174.

Lenggenhager, B., Lopez, C., 2015a. Vestibular contributions to the sense of body, self, and others. In: Metzinger, T., Windt, J.M. (Eds.), Open MIND. MIND-Group,

D. Deroualle et al. / Neuropsychologia 79 (2015) 175–185 185

Frankfurt, pp. 1–38. Lenggenhager, B., Lopez, C., 2015b. Vestibular sense and perspectival experience. A

reply to Adrian Alsmith. In: Metzinger, T., Windt, J.M. (Eds.), Open MIND. MIND-Group, Frankfurt, pp. 1–9.

Lenggenhager, B., Lopez, C., Blanke, O., 2008. Influence of galvanic vestibular sti- mulation on egocentric and object-based mental transformations. Exp. Brain Res. 184, 211–221.

Lopez, C., 2013. A neuroscientific account of how vestibular disorders impair bodily self-consciousness. Front. Integr. Neurosci. 7, 91.

Lopez, C., Blanke, O., Mast, F.W., 2012. The human vestibular cortex revealed by coordinate-based activation likelihood estimation meta-analysis. Neuroscience 212, 159–179.

Lopez, C., Falconer, C.J., Mast, F.W., 2013. Being moved by the self and others: in- fluence of empathy on self-motion perception. PloS One 8, e48293.

Lopez, C., Halje, P., Blanke, O., 2008. Body ownership and embodiment: Vestibular and multisensory mechanisms. Neurophysiol. Clin. 38, 149–161.

Lopez, C., Heydrich, L., Seeck, M., Blanke, O., 2010a. Abnormal self-location and vestibular vertigo in a patient with right frontal lobe epilepsy. Epilepsy Behav. 17, 289–292.

Lopez, C., Lenggenhager, B., Blanke, O., 2010b. How vestibular stimulation interacts with illusory hand ownership. Conscious. Cogn. 19, 33–47.

Lopez, C., Vibert, D., Mast, F.W., 2011. Can imagined whole-body rotations improve vestibular compensation? Med. Hypotheses 76, 816–819.

Lopez, C., Blanke, O., 2011. The thalamocortical vestibular system in animals and humans. Brain Res. Rev. 67 (1–2), 119–146.

Lu, C.H., Proctor, R.W., 1995. The influence of irrelevant location information on performance: A review of the Simon and spatial Stroop effects. Psychon. Bull. Rev. 2, 174–207.

Mast, F.W., Bamert, L., Newby, N., 2007. Mind over matter? Imagined body move- ments and their neuronal correlates, in: Spatial Processing in Navigation. Imagery and Perception. Springer ScienceþBusiness Media, New York, pp. 353–368.

Mast, F.W., Merfeld, D.M., Kosslyn, S.M., 2006. Visual mental imagery during caloric vestibular stimulation. Neuropsychologia 44, 101–109.

Mast, F.W., Preuss, N., Hartmann, M., Grabherr, L., 2014. Spatial cognition, body representation and affective processes: the role of vestibular information be- yond ocular reflexes and control of posture. Front. Integr. Neurosci. 8, 44.

Mattan, B., Quinn, K.A., Apperly, I.A., Sui, J., Rotshtein, P., 2014. Is it always me first? Effects of self-tagging on third-person perspective-taking. J. Exp. Psychol. Learn. Mem. Cogn.

May, M., Wendt, M., 2013. Visual perspective taking and laterality decisions: Pro- blems and possible solutions. Front. Hum. Neurosci 7, 549.

May, M., Wendt, M., 2012. Separating mental transformations and spatial com- patibility effects in the own body transformation task. Cogn. Process 13 (Suppl. 1), S257–S260.

Mazzola, L., Lopez, C., Faillenot, I., Chouchou, F., Mauguière, F., Isnard, J., 2014. Vestibular responses to direct stimulation of the human insular cortex. Ann. Neurol. 76, 609–619.

McKay, R., Tamagni, C., Palla, A., Krummenacher, P., Hegemann, S.C.A., Straumann, D., Brugger, P., 2013. Vestibular stimulation attenuates unrealistic optimism. Cortex 49, 2272–2275.

Mohr, C., Rowe, A.C., Blanke, O., 2010. The influence of sex and empathy on putting oneself in the shoes of others. Br. J. Psychol. 101, 277–291.

Nielsen, M.K., Slade, L., Levy, J.P., Holmes, A., 2015. Inclined to see it your way: do altercentric intrusion effects in visual perspective taking reflect an intrinsically social process? Q. J. Exp. Psychol 2006, 1–21.

Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97–113.

Palla, A., Lenggenhager, B., 2014. Ways to investigate vestibular contributions to cognitive processes. Front. Integr. Neurosci. 8, 40.

Parsons, L.M., 1987a. Imagined spatial transformation of one's body. J. Exp. Psychol. Gen. 116, 172–191.

Parsons, L.M., 1987b. Imagined spatial transformations of one’s hands and feet. Cognit. Psychol. 19, 178–241.

Péruch, P., Lopez, C., Redon, C., Escoffier, G., Zeitoun, A., Sanjuane, M., Deveze, A., Magnan, J., Borel, L., 2011. Vestibular information is necessary for maintaining metric properties of representational space: Evidence from mental imagery.

Neuropsychologia 49, 3136–3144. Piaget, J., Inhelder, B., 1956. The Child's Conception of Space. Routledge, London. Preuss, N., Hasler, G., Mast, F.W., 2014a. Caloric vestibular stimulation modulates

affective control and mood. Brain Stimulat. 7, 133–140. Preuss, N., Mast, F.W., Hasler, G., 2014b. Purchase decision-making is modulated by

vestibular stimulation. Front. Behav. Neurosci. 8, 51. Rubichi, S., Nicoleti, R., 2006. The Simon effect and handedness: Evidence for a

dominant-hand attentional bias in spatial coding. Percept. Psychophys. 68, 1059–1069.

Saj, A., Honoré, J., Bernard-Demanze, L., Devèze, A., Magnan, J., Borel, L., 2013. Where is straight ahead to a patient with unilateral vestibular loss? Cortex. 49, 1219–1228.

Samson, D., Apperly, I.A., Braithwaite, J.J., Andrews, B.J., Bodley Scott, S.E., 2010. Seeing it their way: evidence for rapid and involuntary computation of what other people see. J. Exp. Psychol. Hum. Percept. Perform. 36, 1255–1266.

Santiesteban, I., Catmur, C., Hopkins, S.C., Bird, G., Heyes, C., 2014. Avatars and ar- rows: Implicit mentalizing or domain-general processing? J. Exp. Psychol. 40 (3), 929–937.

Schurz, M., Kronbichler, M., Weissengruber, S., Surtees, A., Samson, D., Perner, J., 2015. Clarifying the role of theory of mind areas during visual perspective taking: issues of spontaneity and domain-specificity. NeuroImage 117, 386–396.

Schwabe, L., Lenggenhager, B., Blanke, O., 2009. The timing of temporoparietal and frontal activations during mental own body transformations from different visuospatial perspectives. Hum. Brain Mapp. 30, 1801–1812.

Serino, A., Alsmith, A., Costantini, M., Mandrigin, A., Tajadura-Jimenez, A., Lopez, C., 2013. Bodily ownership and self-location: components of bodily self-con- sciousness. Conscious. Cogn. 22, 1239–1252.

Shepard, R.N., Metzler, J., 1971. Mental rotation of three-dimensional objects. Sci- ence 171, 701–703.

Silva, S., Loubinoux, I., Olivier, M., Bataille, B., Fourcade, O., Samii, K., Jeannerod, M., Démonet, J.-F., 2011. Impaired visual hand recognition in preoperative patients during brachial plexus anesthesia: importance of peripheral neural input for mental representation of the hand. Anesthesiology 114, 126–134.

Slater, M., Spanlang, B., Sanchez-Vives, M.V., Blanke, O., 2010. First person experi- ence of body transfer in virtual reality. PloS One 5, e10564.

Stratton, G.M., 1899. The spatial harmony of touch and sight. Mind VIII, 492–505. Ter Horst, A.C., Cole, J., van Lier, R., Steenbergen, B., 2012. The effect of chronic

deafferentation on mental imagery: a case study. PloS One 7, e42742. Toupet, M., Van Nechel, C., Bozorg Grayeli, A., 2014. Influence of body laterality on

recovery from subjective visual vertical tilt after vestibular neuritis. Audiol. Neurootol. 19, 248–255.

Tversky, B., Hard, B.M., 2009. Embodied and disembodied cognition: spatial per- spective-taking. Cognition 110, 124–129.

Van Elk, M., Blanke, O., 2014. Imagined own-body transformations during passive self-motion. Psychol. Res. 78, 18–27.

Van Elk, M., Blanke, O., 2012. Balancing bistable perception during self-motion. Exp. Brain Res. 222, 219–228.

Vogeley, K., Fink, G.R., 2003. Neural correlates of the first-person-perspective. Trends Cogn. Sci. 7, 38–42.

Vogeley, K., May, M., Ritzl, A., Falkai, P., Zilles, K., Fink, G.R., 2004. Neural correlates of first-person perspective as one constituent of human self-consciousness. J. Cogn. Neurosci. 16, 817–827.

Winter, L., Wollmer, M.A., Laurens, J., Straumann, D., Kruger, T.H.C., 2013. Cox's chair revisited: can spinning alter mood states? Front. Psychiatry 4, 132.

Zacks, J.M., 2008. Neuroimaging studies of mental rotation: a meta-analysis and review. J. Cogn. Neurosci. 20, 1–19.

Zacks, J.M., Michelon, P., 2005. Transformations of visuospatial images. Behav. Cogn. Neurosci. Rev. 4, 96–118.

Zacks, J.M., Tversky, B., 2005. Multiple Systems for Spatial Imagery: Transforma- tions of Objects and Bodies. Spat. Cogn. Comput. 5, 271–306.

Zacks, J.M., Vettel, J.M., Michelon, P., 2003. Imagined viewer and object rotations dissociated with event-related FMRI. J. Cogn. Neurosci. 15, 1002–1018.

Zacks, J., Rypma, B., Gabrieli, J.D., Tversky, B., Glover, G.H., 1999. Imagined trans- formations of bodies: an fMRI investigation. Neuropsychologia 37, 1029–1040.

Zu Eulenburg, P., Müller-Forell, W., Dieterich, M., 2013. On the recall of vestibular sensations. Brain Struct. Funct. 218, 255–267.