Abstract

There is growing evidence that vestibular information is not only involved in reflexive eye movements and the control of posture but it also plays an important role in higher order cognitive processes. Previous behavioral research has shown that concomitant vestibular stimuli influence performance in tasks that involve imagined self-rotations. These results suggest that imagined and perceived body rotations share common mechanisms. However, the nature and specificity of these effects remain largely unknown. Here, we investigated the neural mechanisms underlying this vestibulocognitive interaction. Participants (n = 20) solved an imagined self-rotation task during caloric vestibular stimulation. We found robust main effects of caloric vestibular stimulation in the core region of the vestibular network, including the rolandic operculum and insula bilaterally, and of the cognitive task in parietal and frontal regions. Interestingly, we found an interaction of stimulation and task in the left inferior parietal lobe, suggesting that this region represents the modulation of imagined body rotations by vestibular input. This result provides evidence that the inferior parietal lobe plays a crucial role in the neural integration of mental and physical body rotation.

INTRODUCTION

A key process underlying social cognition is spatial perspective taking (SPT), the ability to mentally adopt the spatial point of view of another person. The dance instructor uses SPT when showing her students a sequence of new dance steps, the chief surgeon uses SPT to supervise the resident's movements when fixing a fracture, and we all use SPT when giving someone directions from a different point of view. An integral component of SPT is own-body mental imagery, which is required to take a spatial perspective that differs from the actual perspective. Own-body mental imagery tasks commonly require a spatial transformation of own body coordinates (e.g., Zacks, Rypma, Gabrieli, Tversky, & Glover, 1999), and this is the reason why this cognitive ability is of interest in the context of vestibular cognition. Several studies provided firm evidence that actual body position in space, for instance, posture, can influence performance in tasks that involve imagined self-rotations (Wang, Callaghan, Gooding-Williams, McAllister, & Kessler, 2016; Surtees, Apperly, & Samson, 2013; Kessler & Thomson, 2010; de Lange, Helmich, & Toni, 2006). These findings suggest that imagined self-rotation is an embodied cognitive process, which relies on sensorimotor signals. Physical self-rotations are processed by the vestibular system, including its cortical network. Accordingly, vestibular information is involved when self-rotations are imagined, suggesting that the cortical vestibular network shares processes with spatial cognitive tasks (for an overview, see Deroualle & Lopez, 2014).

Previous research has shown that imagined self-rotations are modulated by concomitant vestibular information (Macauda et al., 2019; Deroualle, Borel, Devèze, & Lopez, 2015; Falconer & Mast, 2012; Lenggenhager, Lopez, & Blanke, 2008). Vestibular stimulation leads to impaired performance (Dilda, MacDougall, Curthoys, & Moore, 2012; Lenggenhager et al., 2008) but also to facilitation in imagined self-rotation (Falconer & Mast, 2012). Facilitation was strongest when the direction of perceived rotation matched the direction of the imagined rotation. This was confirmed by van Elk and Blanke (2014) and Deroualle et al. (2015). Interestingly, none of these behavioral studies found an influence of vestibular stimulation on object-based mental rotation. Distinctions between mentally rotating one's own body and mental rotation of objects have been shown in RTs (Keehner, Guerin, Miller, Turk, & Hegarty, 2006; Michelon & Zacks, 2006; Jola & Mast, 2005), in factor analysis (Kozhevnikov & Hegarty, 2001), and in predictions on navigational skills (Kozhevnikov, Motes, Rasch, & Blajenkova, 2006). Also, clinical studies show that vestibular dysfunction impairs performance in own-body mental imagery but to a lesser extent in object-based mental rotation (Allum et al., 2017; Grabherr, Cuffel, Guyot, & Mast, 2011). Similar impairments have been found in healthy participants during microgravity but not for object-based mental rotations (Grabherr et al., 2007; Leone, Lipshits, Gurfinkel, & Berthoz, 1995). Vestibular information influences imagined self-rotations, but not object-based mental rotations.

Taken together, behavioral and clinical evidence suggest that vestibular information contributes to own-body mental imagery in a task-specific way. In this study, we used fMRI to measure participants' BOLD signal while they were performing imagined self-rotations during concomitant vestibular stimulation. The neural mechanisms underlying vestibulocognitive interactions are still widely unclear. The cortical vestibular network is widespread (Frank & Greenlee, 2018; Lopez & Blanke, 2011), and it is not yet fully understood which of its structures and subparts also contribute to cognitive tasks like imagined self-rotations.

The Cortical Vestibular Network

The cortical vestibular network is highly distributed, and different core structures located in the posterior insula and surrounding areas have been proposed (Lopez, Blanke, & Mast, 2012). In monkeys, Guldin and Grüsser (1998) provided evidence for the parieto-insular vestibular cortex (PIVC) as the core region of the vestibular network. The exact location of the human analog of the PIVC is still a matter of debate (Frank & Greenlee, 2018; Lopez & Blanke, 2011). The multisensory area in the parieto-insular cortex (Brandt & Dieterich, 1999), the temporo-peri-Sylvian vestibular cortex (Kahane, Hoffmann, Minotti, & Berthoz, 2003), as well as the cytoarchitectonic area OP 2 in the parietal operculum (Eickhoff, Weiss, Amunts, Fink, & Zilles, 2006) have been suggested. Indeed, OP 2 has been consistently shown to be a structure responding to vestibular stimuli, showing strong functional connectivity with other structures of the vestibular network (zu Eulenburg, Caspers, Roski, & Eickhoff, 2012). Recently, it has been proposed that the core of the vestibular cortex consists of multiple areas with anatomical correlates in the retroinsula, parietal operculum, and posterior insula, subsumed as area PIVC+ (Frank & Greenlee, 2018; Frank, Wirth, & Greenlee, 2016; Frank, Baumann, Mattingley, & Greenlee, 2014).

Several additional cortical areas are associated with the processing of vestibular input. Besides activation in the temporal lobe (superior temporal gyrus, middle temporal gyrus, Heschl's gyrus), the frontal lobe (left superior frontal gyrus, left precentral/postcentral gyrus), and the anterior and midcingulate gyrus, there is substantial activation in the parietal lobe (Lopez et al., 2012; zu Eulenburg et al., 2012; Bense et al., 2003; Suzuki et al., 2001). Parietal activation related to vestibular processing has been found in the precuneus and areas including the intraparietal sulcus and angular gyrus in the inferior parietal lobe (IPL), overlapping partly with the TPJ. The TPJ plays a crucial role in self-location (Kaski et al., 2016; Blanke & Arzy, 2005) and self-motion perception (Frank & Greenlee, 2018; Smith, Greenlee, DeAngelis, & Angelaki, 2017; Blanke, Ortigue, Landis, & Seeck, 2002). Still little is known about the functions of these higher vestibular areas and the cognitive operations in which they are involved. In this study, we focus on imagined self-rotations as a core cognitive ability.

Neural Correlates of Own-body Mental Imagery

The parietal cortex has been consistently identified as the core region for mental rotation in fMRI (Tomasino & Gremese, 2016; Keehner et al., 2006; Zacks, 2008; Zacks, Vettel, & Michelon, 2003; Carpenter, Just, Keller, Eddy, & Thulborn, 1999), EEG (Ter Horst, Jongsma, Janssen, Van Lier, & Steenbergen, 2012; Heil, 2002), and rTMS experiments (Harris & Miniussi, 2003). Zacks (2008) found the superior and inferior parietal cortex and regions extending into the occipital cortex (BA 7, BA 19, BA 39, and BA 40; Brodmann, 1909) to be consistently activated when comparing higher angles of mental rotation with lower angles or no rotation. Thus, these regions are prime candidates for the transformation-specific aspect of the mental rotation task, that is, the imagined self-rotation. Imagined self-rotations have been reported to activate the posterior parietal cortex, superior and inferior parietal cortex, precuneus, fusiform gyrus, inferior and middle occipital gyrus, anterior cingulate gyrus, SMA, and the cerebellum (Wraga, Flynn, Boyle, & Evans, 2010; Creem-Regehr, Neil, & Yeh, 2007; Wraga, Shephard, Church, Inati, & Kosslyn, 2005; Creem et al., 2001). Interestingly, imagined self-rotations compared with mental object rotations led to activations in the postcentral gyrus bilaterally, left IPL and superior parietal lobe (SPL), and right angular gyrus, again emphasizing the role of the parietal cortex (Tomasino & Gremese, 2016). The same meta-analysis found that mental rotation tasks using body stimuli compared with other stimuli (without specific task instructions) led to higher activation in the left SPL and the IPL bilaterally. Blanke et al. (2005) have shown that TMS over the right TPJ interferes with own-body mental imagery. Recently, Macauda et al. (2019) provided the first evidence for a neural correlate of imagined self-rotation in PIVC+ by combining perceived self-rotations (induced by means of galvanic vestibular stimulation) with an SPT task in an fMRI paradigm. The task, which was adopted from Keehner et al. (2006), involved an imagined self-rotation to adopt a different spatial perspective and a spatial judgment about a visual scene from this new perspective. PIVC+ was activated both during task and during galvanic vestibular stimulation, suggesting a common neural correlate of imagined and perceived self-rotations.

Hypothesis

Participants performed an imagined self-rotation task while at the same time they were exposed to caloric vestibular stimulation (CVS) or sham stimulation inside the scanner. The goal was to investigate the modulatory effect of concurrent vestibular stimulation on neural correlates of own-body mental imagery. Based on previous neuroimaging studies using either own-body mental imagery or CVS, we predicted that structures in the IPL and SPL are prime candidates for integration of perceived and imagined self-rotation. We included PIVC+ as an additional ROI based on the recent findings of Macauda et al. (2019). No fMRI study has yet identified neural correlates of a modulatory effect of CVS on own-body mental imagery. Moreover, we expected a main effect of CVS in the core region PIVC+ and additional structures of the vestibular network. Regarding the behavioral results, we expected a monotonic increase in RTs with increasing angular disparity (Keehner et al., 2006; Michelon & Zacks, 2006; Jola & Mast, 2005). Furthermore, we expected CVS to lead to either decreasing or increasing RTs due to facilitation or interference effects, respectively.

METHODS

Participants

Twenty-six healthy volunteers (17 women, aged between 19 and 30 years) with normal or corrected-to-normal vision were included in the experiment. Data of six participants had to be excluded before data analysis, one owing to technical errors, three owing to left-handedness, and two owing to discomfort while lying in the scanner, leaving a total of n = 20 participants. Exclusion criteria were MRI contraindications and any history of neurological disorders or medication interfering with the experiment. Participants were recruited from the student population at the University of Regensburg and were compensated with either course credit or a modest honorarium. Informed consent was obtained from each participant. The study conformed to the Declaration of Helsinki and was approved by the ethics committee of the University of Regensburg.

Materials

CVS

We used CVS to induce activation of the vestibular network. CVS as opposed to galvanic vestibular stimulation is leading to a true sensory stimulation of the vestibular end organ (predominantly the horizontal semicircular canals), excluding the otoliths (Aw, Haslwanter, Fetter, Heimberger, & Todd, 1998). Most participants report a perception of yaw rotation, which makes CVS a suitable method to combine imagined and perceived body rotations. Bithermal CVS was applied by an MR-compatible closed-loop system developed by Frank and Greenlee (2014). The system includes MR ear protection with integrated glass pods for each ear. The water flow to each ear was controlled by closing or opening tubes for hot (45–49°C) and cold water (18–22°C) on a relay plate located in the scanner control room. If pain was reported, hot and cold temperatures were adjusted slightly (2–3°) toward body temperature to avoid any pain or discomfort during CVS. CVS and sham stimulation alternated during the experiment. For CVS left cold/right hot (CH) and left hot/right cold (HC) was provided alternately to the ear canals (Figure 1A). For sham stimulation a mixture of hot and cold water (close to body temperature, 37°C) was used. Each stimulation condition lasted 60 sec. The device needed approximately 10 sec to reach stable steady-state levels of caloric stimulation; therefore, all rotation trials within this initial 10-sec time window were excluded from further analyses.

Figure 1. 

Illustration of the experimental design. (A) Stimulation scheme. Both runs started with sham stimulation, followed by either left hot–right cold (HC) or right hot–left cold (CH) stimulation. For the rest of the run HC and CH alternated with sham in between. Stimulation duration was always 60 sec. (B) Illustration of the imagined self-rotation task performed during CVS and sham stimulation. Stimuli were rotated either along the yaw axis, the roll axis, or not rotated at all. Participants were able to respond during stimulus presentation and the subsequent 3000 msec.

Figure 1. 

Illustration of the experimental design. (A) Stimulation scheme. Both runs started with sham stimulation, followed by either left hot–right cold (HC) or right hot–left cold (CH) stimulation. For the rest of the run HC and CH alternated with sham in between. Stimulation duration was always 60 sec. (B) Illustration of the imagined self-rotation task performed during CVS and sham stimulation. Stimuli were rotated either along the yaw axis, the roll axis, or not rotated at all. Participants were able to respond during stimulus presentation and the subsequent 3000 msec.

Stimuli and Task

In the imagined self-rotation task, participants decided on each trial whether the presented 3-D male avatar stretched out his right or left arm. Participants were instructed to solve the task by mentally rotating themselves to adopt the spatial perspective of the avatar. Participants answered by pressing a button with the right index finger (right arm stretched out) or the left index finger (left arm stretched out). The pictured avatar was rotated in different orientations along the yaw or roll axis, respectively (−135°, −45°, 0°, 45°, 135°; negative angles indicate counterclockwise rotations; see Figure 1B). This resulted in 18 different stimuli, which were presented in a pseudorandomized order balanced for the different stimulation conditions. To keep interparticipant differences as small as possible, the same sequence of stimuli was used for all participants.

Participants completed two blocks consisting of 211 trials of the imagined self-rotation task each. Each stimulus was shown 10–13 times per run. Stimuli were presented for 1 sec followed by a fixation cross, which was shown for 3 sec plus jitter. Jitter times were randomized using a normal distribution (M = 4.00 sec, SD = 1.50 sec) to vary stimulus onsets relative to repetition time (TR). The duration of one run was approximately 28 min (1682 sec). In the scanner, the stimuli were back-projected using a PROPixx projector (VPixx Technologies), with a mirror reflected size of 0.40 × 0.30 m and a viewing distance from the participant to the mirror of 0.95 m. Furthermore, response buttons of the Celeritas Fiber Optic Response System (Psychology Software Tools) were used.

MRI Data Acquisition

MRI data were acquired using a 3T MAGNETOM Prisma MR scanner (Siemens) with a Siemens Head/Neck 20-channel coil. Head movements were minimized using foam pads around participant's head and neck. Moreover, a small pillow was positioned under the participant's head to alter head orientation forward by approximately 30°. This brings the horizontal semicircular canal into a roughly earth-vertical plane, thus making thermoconvection induced by CVS most effective (Barnes, 1995). At the same time, this head orientation minimizes magnetic vestibular stimulation as it occurs when exposed to large magnetic fields (Roberts et al., 2011). Functional images were acquired with a T2*-weighted gradient-recalled EPI sequence with Generalized Autocalibrating Partially Parallel Acquisitions (GRAPPA) (Griswold et al., 2002) acceleration factor of 2, sensitive to BOLD contrast (37 transverse slices, TR = 2000 msec, echo time = 30 msec, flip angle = 90°, voxel size 3 × 3 × 3 mm, interslice gap 0.3 mm, field of view = 192 × 192 mm). Each session included two runs that each lasted 1700 sec and included acquisition of 850 volumes. Structural images were collected using a T1-weighted magnetization-prepared rapid gradient echo sequence (176 sagittal slices, TR = 2300 msec, echo time = 2.98 msec, flip angle = 9°, voxel size 1 × 1 × 1 mm, no interslice gap, field of view = 248 × 256 mm) optimized for the differentiation of gray and white matter by sequence parameters from the Alzheimer's Disease Neuroimaging Initiative Project (adni.loni.ucla.edu).

Experimental Procedure

Before starting the experiment, participants filled out a questionnaire concerning demographic data and exclusion criteria and received instructions for the imagined self-rotation task. We administered 20 practice trials to ensure that the instructions were correctly understood. Then, participants were familiarized with the scanner environment and the CVS. During scanning, participants completed two runs of the imagined self-rotation task with simultaneous CVS. The start of the CVS condition (HC or CH) was balanced over all participants. High-resolution anatomical images were acquired between the two runs.

After the experiment, participants filled out an additional questionnaire about mental rotation strategies and sensations of self-motion during CVS. Participants indicated how often they used the instructed mental self-rotation strategy as opposed to a mental object rotation strategy or any other (self-reported) strategies. Additionally, they were asked to rate illusory self-motion intensity for the whole body, trunk, head, arms, and legs separately on a numerical rating scale from 0 (no motion perception) to 6 (very strong motion perception).

Preprocessing and Data Analysis

Behavioral Data

RT data were analyzed using R (R Core Team, 2013). Trials in the first 10 sec of CVS or sham stimulation and trials with RTs below 0.1 sec (18 trials) or above 4 sec (73 trials) were omitted. Only trials with correct responses were considered. Mean RTs were analyzed by means of a 2 × 3 repeated-measures ANOVA with factors Stimulation (sham, CVS) and Stimulus Orientation (0°, 45°, 135°; clockwise and counterclockwise angles were pooled). If sphericity assumption was violated (for main effect of Stimulus Orientation or interaction), Greenhouse–Geisser correction was applied, and these corrected p values as well as Greenhouse–Geisser epsilon (Greenhouse & Geisser, 1959) are reported. We report post hoc t tests (two-tailed) where necessary for the interpretation of main effects or interactions.

fMRI Data

Preprocessing and statistical analysis of imaging data were performed using SPM12 (Version 6685; Penny, Friston, Ashburner, Kiebel, & Nichols, 2011) based on MATLAB (The MathWorks).

Preprocessing.

Five dummy scans were acquired at the beginning of every run to reach a steady-state MR signal and successively excluded from the analysis. Functional images were realigned to the first image of the first run by rigid body movements to correct for head movements extracting six realignment parameters for statistical analysis. Body movements did not exceed 6-mm deviation. The images were slice-time corrected, coregistered to the high-resolution anatomical image, and normalized to the standard template in the Montreal Neurological Institute (MNI) space using linear and nonlinear transformations. Smoothing was applied using a Gaussian kernel of FWHM of 8 mm to reduce noise. A temporal high-pass filter was applied (cutoff 128 sec) to eliminate low-frequency signal drift.

First-level analysis.

A general linear model was defined for each participant's functional runs with stimulation (sham, CVS) and stimulus orientation (0°, 45°, 135°) as regressors of interest. As in the behavioral analysis, functional imaging data of the first 10 sec after the change of stimulation were dropped, and only correct trials with RTs between 0.01 and 4 sec were included. The six rigid body motion correction parameters (three translation and three rotation variables) were entered into the general linear model as confounding variables of no interest. The canonical hemodynamic response function was used to model the BOLD response. Trial-by-trial RTs were included in the model as parametric modulators. Thus, contrast images for each participant were created for all factor combinations: “0° sham,” “45° sham,” “135° sham,” “0° CVS,” “45° CVS,” “135° CVS.”

Second-level analysis.

Whole-brain group effects were calculated using a 2 × 3 “full factorial” design with within-participant factors Stimulation (sham, CVS) and Stimulus Orientation (0°, 45°, 135°). We assessed the main effects of Stimulation and Stimulus Orientation as well as the interaction between the two factors. The threshold for significance was set at p < .05 family-wise error (FWE)-corrected at the peak level. For anatomical labeling and visualization, the xjView toolbox (www.alivelearn.net/xjview) and mriCron (Rorden & Brett, 2000) were used.

ROI analysis.

ROI analysis was performed using the Anatomical Automatic Labeling 2 Toolbox (AAL2; Eickhoff et al., 2005), and parameter estimates of ROIs were extracted with Marsbar (Brett, Anton, Valabregue, & Poline, 2002). Both toolboxes were implemented in MATLAB (The MathWorks). Repeated-measures ANOVAs were calculated using R software (R Foundation for Statistical Computing). The right and left IPL and SPL were chosen as anatomical ROIs (Tzourio-Mazoyer et al., 2002; see Figure 2). Based on the recent results of Macauda et al. (2019), we included right and left PIVC+ as additional ROIs. The regions consisted of two spheres with a radius of 5 mm each around the peak voxels of PIVC and posterior insular cortex (PIC) using coordinates reported by Frank et al. (2016; see Figure 2). We chose to use PIVC+ rather than reporting PIVC and PIC separately, because Macauda et al. found overlaps of activation with both areas. For the ROIs, we extracted parameter estimates for all conditions and calculated a repeated-measures 2 × 3 ANOVA with factors Stimulation (sham, CVS) and Stimulus Orientation (0°, 45°, 135°). When the assumption of sphericity was violated, Greenhouse–Geisser correction was applied just like in the behavioral analysis. Post hoc t tests (two-tailed) were used where it was deemed appropriate. For visualization of ROIs, Connectome Workbench (Marcus et al., 2011) was used.

Figure 2. 

Anatomical ROIs IPL and SPL (Tzourio-Mazoyer et al., 2002), and PIVC+ (Frank et al., 2016) overlaid on an inflated brain template in Connectome Workbench (Marcus et al., 2011). Intraparietal sulcus (IPS), postcentral gyrus (PCG), and Brodmann's areas (BA; Brodmann, 1909) added by hand. The labeled BA 39 and BA 40 also indicate the angular gyrus and the supramarginal gyrus, respectively.

Figure 2. 

Anatomical ROIs IPL and SPL (Tzourio-Mazoyer et al., 2002), and PIVC+ (Frank et al., 2016) overlaid on an inflated brain template in Connectome Workbench (Marcus et al., 2011). Intraparietal sulcus (IPS), postcentral gyrus (PCG), and Brodmann's areas (BA; Brodmann, 1909) added by hand. The labeled BA 39 and BA 40 also indicate the angular gyrus and the supramarginal gyrus, respectively.

RESULTS

Behavioral Results

Accuracy was not analyzed because of a low number of errors (2.5%). Repeated-measures ANOVA revealed a significant main effect of Stimulus Orientation on RTs, F(2, 38) = 51.09, p < .001, ε = .578, ηG2 = .115, but not of Stimulation, F(1, 19) = 1.79, p = .197. However, there was a significant interaction between Stimulation and Stimulus Orientation, F(2, 38) = 4.55, p = .017, ηG2 = .002. Estimated RTs for all conditions are shown in Figure 3. All pairwise comparisons concerning the main effect of Stimulus Orientation were significant. RTs were significantly longer for 45° compared with 0°, Δ45°–0° = 0.05, t(19) = 3.80, p = .001, for 135° compared with 0°, Δ135°–0° = 0.18, t(19) = 7.18, p < .001, and for 135° compared with 45°, Δ135°–45° = 0.13, t(19) = 9.01, p < .001. Concerning the interaction, post hoc t tests on the difference between CVS and sham for each stimulus orientation separately show a significant difference in RTs between CVS and sham in the 0° condition, ΔCVS-sham = 0.03, t(19) = 2.97, p = .008. We did not find this difference in RTs between CVS and sham in the 45° condition, ΔCVS-sham = −0.01, t(19) = −0.88, p = .388, or in the 135° condition, ΔCVS-sham = −0.002, t(19) = −0.34, p = .740.

Figure 3. 

Average RTs for stimulus orientation and type of stimulation for n = 20 participants. The main effect of Stimulus Orientation is reflected in increased RTs for 45° and 135° compared with 0°, and 135° compared with 45°. There was no main effect of Stimulation. However, there was a significant interaction between Stimulation and Stimulus Orientation. Post hoc t tests reveal higher RTs for CVS than sham in the 0° condition, but no difference between CVS and sham in the 45° and 135° condition (significant differences are highlighted). Error bars indicate SEM.

Figure 3. 

Average RTs for stimulus orientation and type of stimulation for n = 20 participants. The main effect of Stimulus Orientation is reflected in increased RTs for 45° and 135° compared with 0°, and 135° compared with 45°. There was no main effect of Stimulation. However, there was a significant interaction between Stimulation and Stimulus Orientation. Post hoc t tests reveal higher RTs for CVS than sham in the 0° condition, but no difference between CVS and sham in the 45° and 135° condition (significant differences are highlighted). Error bars indicate SEM.

Participants reported to have used the instructed self-rotation strategy to solve the imagined self-rotation task in 65% of trials (SD = 23) on average. Some participants reported that they also used an object rotation strategy (M = 30%, SD = 24), or yet another strategy in 5% of trials (SD = 9). Upon further questioning, participants reported to have solved these few trials from memory instead of using any kind of mental rotation.

In the self-motion questionnaire, all but one participant reported illusory self-motion sensations during CVS (M = 2.23, SD = 1.32). Specifically, self-motion sensations were mostly located at the level of the head (M = 2.58, SD = 1.50). Ratings were lower at the level of the whole body (M = 1.30, SD = 1.22), the trunk (M = 0.98, SD = 1.24), the legs (M = 0.85, SD = 1.33), and the arms (M = 0.68, SD = 1.05).

fMRI Results

Full-factorial Whole-brain Analysis

CVS.

Group-level analysis revealed a significant main effect of Stimulation in the rolandic operculum bilaterally, the insula bilaterally and the right thalamus. Post hoc t tests revealed that this effect was driven by the “CVS > sham” contrast. Specifically, we found higher activation for “CVS > sham” in the rolandic operculum bilaterally, the insula bilaterally, the right thalamus, and the left postcentral gyrus (see Table 1 and Figure 4). There was no significant activation for the opposite contrast (“sham > CVS”). There was no main effect of the side where hot and cold stimulation were applied. More specifically, no difference in activation was found for the contrasts “HC > CH” and “CH > HC.”

Table 1. 
Peak Coordinates (x, y, z) in MNI Space, Cluster Size (k), z Score, and FWE-corrected p Value for the Main Effect of Stimulation
ContrastRegionxyzkz ScorepFWE
Main effect of simulation (F test) R rolandic operculum 54 −10 14 252 7.49 <.001 
R rolandic operculum 39 −16 20   7.14 <.001 
R insula 39 −4   6.65 <.001 
L rolandic operculum −39 −19 20 70 6.39 <.001 
L insula −36 −13 11   6.16 <.001 
L rolandic operculum −54 −4 11 10 5.24 .002 
R thalamus 21 −16 5 1 4.74 .020 
  
CVS > sham (t test) R rolandic operculum 54 −10 14 284 7.58 <.001 
R rolandic operculum 39 −16 20   7.24 <.001 
R insula 39 −4   6.75 <.001 
L rolandic operculum −39 −19 20 83 6.49 <.001 
L insula −36 −13 11   6.26 <.001 
L rolandic operculum −54 −4 11 16 5.36 .001 
R thalamus 21 −16 5 3 4.88 .010 
L postcentral gyrus −60 −16 20 7 4.64 .027 
ContrastRegionxyzkz ScorepFWE
Main effect of simulation (F test) R rolandic operculum 54 −10 14 252 7.49 <.001 
R rolandic operculum 39 −16 20   7.14 <.001 
R insula 39 −4   6.65 <.001 
L rolandic operculum −39 −19 20 70 6.39 <.001 
L insula −36 −13 11   6.16 <.001 
L rolandic operculum −54 −4 11 10 5.24 .002 
R thalamus 21 −16 5 1 4.74 .020 
  
CVS > sham (t test) R rolandic operculum 54 −10 14 284 7.58 <.001 
R rolandic operculum 39 −16 20   7.24 <.001 
R insula 39 −4   6.75 <.001 
L rolandic operculum −39 −19 20 83 6.49 <.001 
L insula −36 −13 11   6.26 <.001 
L rolandic operculum −54 −4 11 16 5.36 .001 
R thalamus 21 −16 5 3 4.88 .010 
L postcentral gyrus −60 −16 20 7 4.64 .027 

Regions not printed in bold belong to the above-mentioned cluster and therefore lack cluster size values. L stands for left hemisphere; R stands for right hemisphere.

Figure 4. 

The contrast “CVS > sham” shows vestibular activation in the rolandic operculum (RO) bilaterally, the insula (Ins) bilaterally, the left postcentral gyrus (PCG), and the right thalamus (Th; all ps < .05 FWE-corrected) overlaid on coronal slices of the high-resolution anatomical brain template contained in MRIcron (Rorden & Brett, 2000).

Figure 4. 

The contrast “CVS > sham” shows vestibular activation in the rolandic operculum (RO) bilaterally, the insula (Ins) bilaterally, the left postcentral gyrus (PCG), and the right thalamus (Th; all ps < .05 FWE-corrected) overlaid on coronal slices of the high-resolution anatomical brain template contained in MRIcron (Rorden & Brett, 2000).

Imagined self-rotation task.

The main effect of Stimulus Orientation revealed activation in the SPL bilaterally, as well as in the left superior frontal gyrus. Pairwise comparisons revealed that this main effect was driven by the largest difference in stimulus angle (i.e., “135° > 0°” contrast). This t test showed significant activation in the SPL bilaterally, the left superior frontal gyrus and in a single voxel in the anterior part of the right insula. We found no significant activation for any of the other pairwise comparisons. Detailed summaries of the main effect of Stimulus Orientation can be found in Table 2 and Figure 5. Neither the direction (leftward vs. rightward) nor the rotation axis (yaw vs. roll) of imagined self-rotation yielded any significant differences.

Table 2. 
Peak Coordinates (x, y, z) in MNI Space, Cluster Size (k), z Score and FWE-corrected p Value for the Main Effect of Rotation
ContrastRegionxyzkz ScorepFWE
Main effect of rotation (F test) L superior frontal gyrus −24 −10 62 12 5.23 .002 
L SPL −30 −52 59 5.17 .003 
R SPL 18 −61 59 4.62 .037 
  
135° > 0° (t test) L superior frontal gyrus −21 −10 62 35 5.61 <.001 
L SPL −30 −52 59 17 5.55 <.001 
R insula 36 23 4.60 .033 
R SPL 21 −61 62 4.59 .033 
ContrastRegionxyzkz ScorepFWE
Main effect of rotation (F test) L superior frontal gyrus −24 −10 62 12 5.23 .002 
L SPL −30 −52 59 5.17 .003 
R SPL 18 −61 59 4.62 .037 
  
135° > 0° (t test) L superior frontal gyrus −21 −10 62 35 5.61 <.001 
L SPL −30 −52 59 17 5.55 <.001 
R insula 36 23 4.60 .033 
R SPL 21 −61 62 4.59 .033 

L stands for left hemisphere; R stands for right hemisphere.

Figure 5. 

The contrast “135° > 0°” shows task-specific activation in the SPL bilaterally, the left superior frontal gyrus (SFG), and the right insula (Ins; all ps < .05 FWE corrected) overlaid on sagittal slices of the template brain contained in MRIcron (Rorden & Brett, 2000).

Figure 5. 

The contrast “135° > 0°” shows task-specific activation in the SPL bilaterally, the left superior frontal gyrus (SFG), and the right insula (Ins; all ps < .05 FWE corrected) overlaid on sagittal slices of the template brain contained in MRIcron (Rorden & Brett, 2000).

Interaction of CVS and imagined self-rotation task.

Whole-brain group-level analysis revealed no interaction of Stimulation and Stimulus Orientation.

ROI Analysis

IPL.

The parameter estimates of the ROI analysis in the right or left IPL did not show significant changes in voxel activation due to stimulation or stimulus orientation. Interestingly, a significant interaction of stimulation and stimulus orientation was found in the left, F(2, 38) = 4.32, p = .020, ηG2 = .052, but not in the right IPL, F(2, 38) = 2.99, p = .078, ε = 0.768 (see Figure 6). Post hoc comparisons revealed that only in the CVS condition parameter estimates in left IPL were increased when participants had to perform an imagined self-rotation. There was a significant difference in parameter estimates between the 0° and 45° condition, Δ45°–0° = 0.43, t(19) = 2.55, p = .020, and the 0° and 135° condition, Δ135°–0° = 0.56, t(19) = 2.35, p = .030, but not between the 45° and 135° condition, Δ135°–45° = 0.12, t(19) = 1.06, p = .304. In the sham condition, there were no differences in parameter estimates when comparing the 0° and 45° condition, Δ45°–0° = −0.30, t(19) = −1.96, p = .065; the 0° and 135° condition, Δ135°–0° = −0.07, t(19) = −0.45, p = .656; or the 45° and 135° condition, Δ135°–45° = 0.23, t(19) = 1.42, p = .171.

Figure 6. 

Results of the ROI analysis in the (A) left and (B) right IPL. Mean parameter estimates indicate a significant interaction between Stimulus Orientation and Stimulation in the left IPL, but no main effect of Stimulation or Stimulus Orientation. Post hoc t tests reveal that there are no differences between the stimulus orientation in the sham condition, but the parameter estimates are increased in the 45° and 135° compared with 0° in the CVS condition (significant differences are highlighted). No significant main effect of Stimulus Orientation or Stimulation, and no interaction was found in the right IPL. Error bars indicate SEM.

Figure 6. 

Results of the ROI analysis in the (A) left and (B) right IPL. Mean parameter estimates indicate a significant interaction between Stimulus Orientation and Stimulation in the left IPL, but no main effect of Stimulation or Stimulus Orientation. Post hoc t tests reveal that there are no differences between the stimulus orientation in the sham condition, but the parameter estimates are increased in the 45° and 135° compared with 0° in the CVS condition (significant differences are highlighted). No significant main effect of Stimulus Orientation or Stimulation, and no interaction was found in the right IPL. Error bars indicate SEM.

SPL.

The parameter estimates of the ROI analysis revealed a significant main effect of stimulus orientation in the right, F(2, 38) = 8.24, p = .004, ε = 0.715, ηG2 = .086, and in the left hemisphere, F(2, 38) = 11.92, p < .001, ε = 0.749, ηG2 = .122. Post hoc t tests revealed that there was no difference between 0° and 45° in the left, Δ45°–0° = 0.26, t(19) = 1.83, p = .083, and in the right hemisphere, Δ45°–0° = 0.13, t(19) = 0.91, p = .372. There was, however, a significant difference between 0° and 135° in the left, Δ135°–0° = 0.72, t(19) = 3.87, p = .001, and in the right hemisphere, Δ135°–0° = 0.65, t(19) = 3.01, p = .007. There was also a significant difference between 45° and 135° for both the left, Δ135°–45° = 0.46, t(19) = 4.11, p < .001, and the right hemisphere, Δ135°–45° = 0.52, t(19) = 3.82, p = .001. There was no main effect of Stimulation and no interaction of Stimulation and Stimulus Orientation in the left or right SPL.

PIVC+.

The post hoc ROI analysis reveals a main effect of Stimulation in the right, F(1, 19) = 39.94, p < .001, ηG2 = .203, and left PIVC+, F(1, 19) = 18.85, p < .001, ηG2 = .090, which is in line with the results of the whole-brain analysis of CVS. We found no main effect of Stimulus Orientation and no interaction of Stimulation and Stimulus Orientation in the right or left PIVC+.

DISCUSSION

We investigated the neural correlates of a modulatory sensory vestibular influence on own-body mental imagery. Participants were instructed to mentally rotate themselves to align with the orientation of an avatar on the screen while at the same time they received sensory vestibular input by means of caloric stimulation. We found an interaction of imagined self-rotation and the processing of vestibular information in the left IPL. Exclusively during CVS, left IPL activity is increased when participants perform imagined self-rotations (45° and 135°) as opposed to a pure left–right discrimination without self-rotation (0°). This task-specific effect was absent during sham stimulation; imagined self-rotation alone did not lead to any differences in left IPL activity. No significant interaction was found in the right IPL, SPL, and PIVC+ bilaterally. The main finding of this study suggests that vestibular stimulation modulates left IPL activity during own-body mental imagery.

IPL is a rather large structure with distinct subdivisions, each of which can contribute differently to spatial cognitive tasks like imagined self-rotation (e.g., Sack, 2009). It has been associated with self-motion perception (Lopez & Blanke, 2011; Shinder & Taube, 2010; Bense et al., 2003; Suzuki et al., 2001), self-location (van Elk, Duizer, Sligte, & van Schie, 2017; Kaski et al., 2016; Ionta et al., 2011; Blanke & Metzinger, 2009; Arzy, Thut, Mohr, Michel, & Blanke, 2006; Blanke & Arzy, 2005), vestibular imagery (zu Eulenburg, Müller-Forell, & Dieterich, 2013), imagined self-rotation (Tomasino & Gremese, 2016), and embodied perspective taking (Wang et al., 2016). Specifically the left TPJ, located in the ventral part of the IPL (Igelström & Graziano, 2017), is associated with tasks involving body representation and self-referential processing (Nakul & Lopez, 2017; Bos, Spoor, Smits, Schouten, & Vincent, 2016; Ionta et al., 2011; Blanke et al., 2005). Interestingly, the IPL is also involved in higher level vestibular computations such as integration of velocity and time for the processing of self-location (Kaski et al., 2016; Ionta et al., 2011). Therefore, converging evidence suggests that the IPL is a predestined region associated with higher vestibular processes. The findings from this study provide new evidence linking the spatial processing of self-motion information induced by CVS with the spatial processing required to imagine self-motion.

In our experiment, participants had to mentally simulate a change of body position while at the same time receiving sensory vestibular information during CVS. Behavioral studies on vestibular cognition have shown that sensory vestibular information had a specific influence on performance in tasks involving imagined self-rotation (Deroualle et al., 2015; Falconer & Mast, 2012; Lenggenhager et al., 2008). Other studies provided further evidence that imagined self-rotation is influenced by sensory information about body position and requires a transformation and mapping of egocentric body coordinates onto an avatar (Wang et al., 2016; Kessler & Thomson, 2010). Analogously, imagined hand movements are influenced by information on hand posture (de Lange et al., 2006). An increase of left IPL activity was reported when there was a larger incongruence between hand posture and imagined hand movement. In our study, left IPL activity was increased during vestibular stimulation only when an imagined self-rotation was required. The results support the hypothesis that at least parts of the vestibular cortical network are recruited in an offline mode to simulate self-motion (Mast & Ellis, 2015).

Interestingly, clinical evidence from patients suffering from heterotopagnosia following left IPL lesion underline the findings of this study. Heterotopagnosia is a selective deficit in localizing body parts of another person while localization of own body parts remains unaffected (Bassolino et al., 2019; Auclair, Noulhiane, Raibaut, & Amarenco, 2009). The projection of own body coordinates into another person's body (Auclair et al., 2009; Arzy et al., 2006) is exactly the process which is required in our task, and it is tightly linked to the left IPL. Moreover, when healthy participants imagine position and visual perspective outside their own body, the TPJ, which overlaps with the IPL, is selectively activated (Blanke et al., 2005). Several studies demonstrate a facilitation effect of vestibular stimulation on own-body mental imagery (Deroualle et al., 2015; van Elk & Blanke, 2014; Falconer & Mast, 2012). Facilitation has been explained by heightened accessibility of bodily reference frame due to vestibular stimulation (Falconer & Mast, 2012). In this regard, it is likely that increased IPL activity during imagined self-rotation under CVS reflects a recruitment of the egocentric reference frame to compute the spatial transformation necessary to solve the self-rotation task.

Although the behavioral results are in favor of facilitation, it is noteworthy that imagining self-rotation during vestibular stimulation could lead to conflicting signals, which must be resolved to achieve robust self-location. Findings in healthy participants during multisensory conflicts (Ionta et al., 2011) and in patients with out-of-body illusions (Blanke et al., 2005) suggest the TPJ as a crucial structure for integrating multisensory body signals to encode self-location and create spatial unity of self and body. Moreover, heautoscopy (seeing one's own body from a distant point of view) is a clinical phenomenon linked to pathology in particularly the left TPJ, leading to disintegration of multisensory signals (Blanke & Mohr, 2005). These patients show distorted self-location, and often two spatial positions are experienced either simultaneously or alternating (Blanke, 2012). In the context of this study, it is thus conceivable that CVS interferes with imagined self-rotation, leading to increased activation in left IPL. However, RT data do not support that vestibular stimulation interferes with imagined self-rotation, because RTs were not increased during CVS when the visual stimuli were rotated. It is more likely that concomitant vestibular stimulation facilitates the mental transformation required by the self-rotation task.

Behavioral results show the typically increased RTs as a function of stimulus orientation. During CVS, increased RTs were found in the 0° condition where no mental rotation was required, but not in the self-rotation conditions (45° and 135°). In the light of previous findings, this was unexpected. Falconer and Mast (2012) found no difference between CVS and sham in the no rotation condition, but a facilitation when performing imagined self-rotations during CVS. Increased RTs in the 0° condition during CVS could reflect a distraction caused by vestibular stimulation. However, there were no signs of distraction when the stimuli were rotated. An overall distraction effect induced by CVS is expected to manifest itself in all rotation conditions. This was not the case. It is thus possible that the distraction is compensated for by a facilitation due to CVS in the rotation conditions (45° and 135°). Based on these considerations, we suggest that there is a hidden facilitation of imagined self-rotation during CVS. The interaction in the left IPL could be a neural correlate of this facilitation, reflecting the modulatory influence of CVS on imagined self-rotation. However, this interpretation should be taken with caution because RTs do not show a direct influence of CVS on imagined self-rotation.

In contrast to our study, Macauda et al. (2019) found activation in PIVC+ not only during galvanic vestibular stimulation but also during their SPT task. We did not find any task-specific activation in PIVC+. The task used by Macauda et al. involved not only an imagined self-rotation but also a visuospatial judgment about external objects from a different perspective. PIC specifically is known to respond to vestibular and visual stimuli (Frank & Greenlee, 2018), and it is therefore possible that it is activated by the SPT task used by Macauda et al. due to its visual component. The negative correlation of BOLD response in PIC and RTs reported by Macauda et al. could also reflect the visual–vestibular integration required to solve the SPT task. Furthermore, they focused on the comparison between SPT and mental object rotation, as opposed to our study comparing imagined self-rotation to a parity task requiring no mental rotation. Our data suggest that vestibular stimulation interacts with imagined self-rotation in the left IPL rather than PIVC+.

Regarding the main effect of vestibular stimulation, we found clusters of activation in the insula and rolandic operculum, bilaterally showing that CVS activated the core of the vestibular cortical network, referred to as PIVC+. This is in line with previous findings (Frank & Greenlee, 2018; Lopez et al., 2012; Fasold et al., 2002) and demonstrates that CVS induced a clear vestibular activation. We found larger clusters of activation in the right (nondominant) hemisphere, irrespective of the CVS condition (HC or CH). This is in line with previous studies suggesting a lateralization of cortical vestibular processing to the right hemisphere (Wirth, Frank, Greenlee, & Beer, 2018; zu Eulenburg et al., 2012; Dieterich et al., 2003). However, we did not find an effect of the side of stimulation (HC vs. CH), which has been reported in previous studies (Lopez et al., 2012; Dieterich et al., 2003). Some of the cortical vestibular areas found by previous studies using CVS could not be replicated. This is likely due to differences in the design such as temperature, visual fixation, control condition, and the applied statistical threshold (for an overview, see Lopez et al., 2012). Compared with previous studies using CVS in the scanner, the most remarkable difference in our study is that participants performed a cognitive task during vestibular stimulation. This cognitive task may have distracted participants from monitoring self-motion perception. CVS inevitably leads to sensory conflicts between vestibular information signaling a yaw rotation and visual and somatosensory input signaling no change in body position (Klingner, Axer, Brodoehl, & Witte, 2016; Mast & Ellis, 2015; Kolev, 2002). Therefore, at least some of cortical brain activation patterns reported in earlier studies could be caused by sensory conflicts and metacognitive monitoring rather than the processing of vestibular information per se. This finding is of interest for future studies that aim to better investigate the functions of the cortical vestibular network.

Behavioral analysis revealed no main effect of CVS on RTs. Especially, there was no difference in RTs between CVS and sham when participants had to perform an imagined self-rotation to solve the task. Given that we continuously switched between the CVS conditions and sham stimulation, carry-over effects of stimulation conditions are possible. The effects of CVS are known to outlast the duration of stimulation (Barnes, 1995). This might account for the lack of a CVS main effect in the behavioral analysis and to the main effect of CVS being restricted to PIVC+. However, applying hot stimulation after cold and vice versa alters the temperature gradient faster toward body temperature (Barnes, 1995). Although this reduces potential carry-over effects, we cannot rule them out completely. Importantly, however, carry-over effects cannot account for our findings as their impact would reduce the difference between CVS and sham stimulation.

Own-body mental imagery activated clusters in the SPL bilaterally, as well as the left superior frontal gyrus, which is in line with previous studies on mental rotation using egocentric rotation strategies and body stimuli (Tomasino & Gremese, 2016; Zacks, 2008). However, on the whole-brain level, we did not find differential activation in other areas that have been associated with imagined self-rotation, such as occipital areas or the IPL among others (Blanke, Ionta, Fornari, Mohr, & Maeder, 2010; Wraga et al., 2010; Creem-Regehr et al., 2007; Creem et al., 2001). This difference may be due to differences in tasks and stimuli, statistical threshold, and applied correction for multiple comparisons or contrasts in the fMRI paradigm. Indeed, some studies contrasted imagined self-rotation tasks not to tasks requiring no mental rotation but rather to tasks requiring mental object rotations (Macauda et al., 2019; Wraga et al., 2005; Zacks et al., 2003). Whereas neuroimaging results on the main effect of own-body mental imagery only revealed a difference between 135° and 0°, we found differences in RTs in all pairwise comparisons. RTs were longer in the 45° condition compared with the 0° condition as well as in the 135° condition compared with the 0° and 45° condition. The increase in RTs in the 45° condition is in contrast to SPT studies that showed increasing RTs only for angles above 60° (Kessler & Thomson, 2010; Michelon & Zacks, 2006; Parsons, 1987). A possible explanation is that participants were explicitly instructed to mentally rotate themselves in all conditions and not to use any heuristic strategies (Kozhevnikov & Hegarty, 2001).

Last but not least, clinical findings add to the discussion of our results. It turned out that left but not right vestibular loss changes perception of self-orientation in space (Saj et al., 2013) and own-body mental imagery (Deroualle et al., 2019). Deroualle et al. (2019) suggest that right-sided disruption of vestibular pathways affects multisensory processing in the right PIVC+, whereas left-sided disruption affects both sides (see also Dieterich, Kirsch, & Brandt, 2017). Therefore, it is of interest that we found an interaction in the left IPL, which again demonstrates a left-hemispheric contribution to own-body mental imagery during CVS.

Conclusion

Behavioral studies on vestibular cognition have shown that vestibular information influences performance in own-body mental imagery and SPT. In our study, brain activation elicited by perceived and imagined self-rotation interacted in the left IPL, suggesting this region as a neural correlate of behavioral findings. The vestibular modulation of inferior parietal activity during own-body mental imagery provides further evidence that the vestibular cortical network is recruited when imagining self-rotations.

Acknowledgments

This work was supported by the Swiss National Science Foundation (project no. 162480) and the Center for Cognition, Learning and Memory of the University of Bern. We would like to thank Alvin Chesham for creating the stimulus material, Marvin Maechler and Alexandra Otto for their assistance in data collection, and all the participants. We are grateful for expert advice given by Gianluca Macauda, Thomas Baumgartner, and Andrea Federspiel.

Reprint requests should be sent to Fred W. Mast, Department of Psychology, University of Bern, Fabrikstrasse 8, 3012 Bern, Switzerland, or via e-mail: fred.mast@psy.unibe.ch.

REFERENCES

Allum
,
J. H. J.
,
Langewitz
,
W.
,
Sleptsova
,
M.
,
Welge-Luessen
,
A.
,
Honegger
,
F.
,
Schatz
,
T. H.
, et al
(
2017
).
Mental body transformation deficits in patients with chronic balance disorders
.
Journal of Vestibular Research
,
27
,
113
125
.
Arzy
,
S.
,
Thut
,
G.
,
Mohr
,
C.
,
Michel
,
C. M.
, &
Blanke
,
O.
(
2006
).
Neural basis of embodiment: Distinct contributions of temporoparietal junction and extrastriate body area
.
Journal of Neuroscience
,
26
,
8074
8081
.
Auclair
,
L.
,
Noulhiane
,
M.
,
Raibaut
,
P.
, &
Amarenco
,
G.
(
2009
).
Where are your body parts? A pure case of heterotopagnosia following left parietal stroke
.
Neurocase
,
15
,
459
465
.
Aw
,
S. T.
,
Haslwanter
,
T.
,
Fetter
,
M.
,
Heimberger
,
J.
, &
Todd
,
M. J.
(
1998
).
Contribution of the vertical semicircular canals to the caloric nystagmus
.
Acta Oto-Laryngologica
,
118
,
618
627
.
Barnes
,
G.
(
1995
).
Adaptation in the oculomotor response to caloric irrigation and the merits of bithermal stimulation
.
British Journal of Audiology
,
29
,
95
106
.
Bassolino
,
M.
,
Bouzerda-Wahlen
,
A.
,
Moix
,
V.
,
Bellmann
,
A.
,
Herbelin
,
B.
,
Serino
,
A.
, et al
(
2019
).
You or me? Disentangling perspectival, perceptual, and integrative mechanisms in heterotopagnosia
.
Cortex
,
120
,
212
222
.
Bense
,
S.
,
Bartenstein
,
P.
,
Lutz
,
S.
,
Stephan
,
T.
,
Schwaiger
,
M.
,
Brandt
,
T.
, et al
(
2003
).
Three determinants of vestibular hemispheric dominance during caloric stimulation: A positron emission tomography study
.
Annals of the New York Academy of Sciences
,
1004
,
440
445
.
Blanke
,
O.
(
2012
).
Multisensory brain mechanisms of bodily self-consciousness
.
Nature Reviews Neuroscience
,
13
,
556
571
.
Blanke
,
O.
, &
Arzy
,
S.
(
2005
).
The out-of-body experience: Disturbed self-processing at the temporo-parietal junction
.
Neuroscientist
,
11
,
16
24
.
Blanke
,
O.
,
Ionta
,
S.
,
Fornari
,
E.
,
Mohr
,
C.
, &
Maeder
,
P.
(
2010
).
Mental imagery for full and upper human bodies: Common right hemisphere activations and distinct extrastriate activations
.
Brain Topography
,
23
,
321
332
.
Blanke
,
O.
, &
Metzinger
,
T.
(
2009
).
Full-body illusions and minimal phenomenal selfhood
.
Trends in Cognitive Sciences
,
13
,
7
13
.
Blanke
,
O.
, &
Mohr
,
C.
(
2005
).
Out-of-body experience, heautoscopy, and autoscopic hallucination of neurological origin: Implications for neurocognitive mechanisms of corporeal awareness and self-consciousness
.
Brain Research Reviews
,
50
,
184
199
.
Blanke
,
O.
,
Mohr
,
C.
,
Michel
,
C. M.
,
Pascual-Leone
,
A.
,
Brugger
,
P.
,
Seeck
,
M.
, et al
(
2005
).
Linking out-of-body experience and self processing to mental own-body imagery at the temporoparietal junction
.
Journal of Neuroscience
,
25
,
550
557
.
Blanke
,
O.
,
Ortigue
,
S.
,
Landis
,
T.
, &
Seeck
,
M.
(
2002
).
Stimulating illusory own-body perceptions
.
Nature
,
419
,
269
270
.
Bos
,
E. M.
,
Spoor
,
J. K. H.
,
Smits
,
M.
,
Schouten
,
J. W.
, &
Vincent
,
A. J. P. E.
(
2016
).
Out-of-body experience during awake craniotomy
.
World Neurosurgery
,
92
,
586.e9
586.e13
.
Brandt
,
T.
, &
Dieterich
,
M.
(
1999
).
The vestibular cortex: Its locations, functions, and disorders
.
Annals of the New York Academy of Sciences
,
871
,
293
312
.
Brett
,
M.
,
Anton
,
J.-L.
,
Valabregue
,
R.
, &
Poline
,
J.-B.
(
2002
).
Region of interest analysis using the MarsBar toolbox for SPM 99
.
Neuroimage
,
16
,
S497
.
Brodmann
,
K.
(
1909
).
Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues
.
Leipzig, Germany
:
J. A. Barth
.
Carpenter
,
P. A.
,
Just
,
M. A.
,
Keller
,
T. A.
,
Eddy
,
W.
, &
Thulborn
,
K.
(
1999
).
Graded functional activation in the visuospatial system with the amount of task demand
.
Journal of Cognitive Neuroscience
,
11
,
9
24
.
Creem
,
S. H.
,
Downs
,
T. H.
,
Wraga
,
M.
,
Harrington
,
G. S.
,
Proffitt
,
D. R.
, &
Downs
,
J. H.
(
2001
).
An fMRI study of imagined self-rotation
.
Cognitive, Affective, & Behavioral Neuroscience
,
1
,
239
249
.
Creem-Regehr
,
S. H.
,
Neil
,
J. A.
, &
Yeh
,
H. J.
(
2007
).
Neural correlates of two imagined egocentric transformations
.
Neuroimage
,
35
,
916
927
.
de Lange
,
F. P.
,
Helmich
,
R. C.
, &
Toni
,
I.
(
2006
).
Posture influences motor imagery: An fMRI study
.
Neuroimage
,
33
,
609
617
.
Deroualle
,
D.
,
Borel
,
L.
,
Devèze
,
A.
, &
Lopez
,
C.
(
2015
).
Changing perspective: The role of vestibular signals
.
Neuropsychologia
,
79
,
175
185
.
Deroualle
,
D.
,
Borel
,
L.
,
Tanguy
,
B.
,
Bernard-Demanze
,
L.
,
Devèze
,
A.
,
Montava
,
M.
, et al
(
2019
).
Unilateral vestibular deafferentation impairs embodied spatial cognition
.
Journal of Neurology
,
266(Suppl. 1)
,
149
159
.
Deroualle
,
D.
, &
Lopez
,
C.
(
2014
).
Toward a vestibular contribution to social cognition
.
Frontiers in Integrative Neuroscience
,
8
,
16
.
Dieterich
,
M.
,
Bense
,
S.
,
Lutz
,
S.
,
Drzezga
,
A.
,
Stephan
,
T.
,
Bartenstein
,
P.
, et al
(
2003
).
Dominance for vestibular cortical function in the non-dominant hemisphere
.
Cerebral Cortex
,
13
,
994
1007
.
Dieterich
,
M.
,
Kirsch
,
V.
, &
Brandt
,
T.
(
2017
).
Right-sided dominance of the bilateral vestibular system in the upper brainstem and thalamus
.
Journal of Neurology
,
264(Suppl. 1)
,
55
62
.
Dilda
,
V.
,
MacDougall
,
H. G.
,
Curthoys
,
I. S.
, &
Moore
,
S. T.
(
2012
).
Effects of Galvanic vestibular stimulation on cognitive function
.
Experimental Brain Research
,
216
,
275
285
.
Eickhoff
,
S. B.
,
Stephan
,
K. E.
,
Mohlberg
,
H.
,
Grefkes
,
C.
,
Fink
,
G. R.
,
Amunts
,
K.
, et al
(
2005
).
A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data
.
Neuroimage
,
25
,
1325
1335
.
Eickhoff
,
S. B.
,
Weiss
,
P. H.
,
Amunts
,
K.
,
Fink
,
G. R.
, &
Zilles
,
K.
(
2006
).
Identifying human parieto-insular vestibular cortex using fMRI and cytoarchitectonic mapping
.
Human Brain Mapping
,
27
,
611
621
.
Falconer
,
C. J.
, &
Mast
,
F. W.
(
2012
).
Balancing the mind: Vestibular induced facilitation of egocentric mental transformations
.
Experimental Psychology
,
59
,
332
339
.
Fasold
,
O.
,
von Brevern
,
M.
,
Kuhberg
,
M.
,
Ploner
,
C. J.
,
Villringer
,
A.
,
Lempert
,
T.
, et al
(
2002
).
Human vestibular cortex as identified with caloric stimulation in functional magnetic resonance imaging
.
Neuroimage
,
17
,
1384
1393
.
Frank
,
S. M.
,
Baumann
,
O.
,
Mattingley
,
J. B.
, &
Greenlee
,
M. W.
(
2014
).
Vestibular and visual responses in human posterior insular cortex
.
Journal of Neurophysiology
,
112
,
2481
2491
.
Frank
,
S. M.
, &
Greenlee
,
M. W.
(
2014
).
An MRI-compatible caloric stimulation device for the investigation of human vestibular cortex
.
Journal of Neuroscience Methods
,
235
,
208
218
.
Frank
,
S. M.
, &
Greenlee
,
M. W.
(
2018
).
The parieto-insular vestibular cortex in humans: More than a single area?
Journal of Neurophysiology
,
120
,
1438
1450
.
Frank
,
S. M.
,
Wirth
,
A. M.
, &
Greenlee
,
M. W.
(
2016
).
Visual-vestibular processing in the human Sylvian fissure
.
Journal of Neurophysiology
,
116
,
263
271
.
Grabherr
,
L.
,
Cuffel
,
C.
,
Guyot
,
J.-P.
, &
Mast
,
F. W.
(
2011
).
Mental transformation abilities in patients with unilateral and bilateral vestibular loss
.
Experimental Brain Research
,
209
,
205
214
.
Grabherr
,
L.
,
Karmali
,
F.
,
Bach
,
S.
,
Indermaur
,
K.
,
Metzler
,
S.
, &
Mast
,
F. W.
(
2007
).
Mental own-body and body-part transformations in microgravity
.
Journal of Vestibular Research
,
17
,
279
287
.
Greenhouse
,
S. W.
, &
Geisser
,
S.
(
1959
).
On methods in the analysis of profile data
.
Psychometrika
,
24
,
95
112
.
Griswold
,
M. A.
,
Jakob
,
P. M.
,
Heidemann
,
R. M.
,
Nittka
,
M.
,
Jellus
,
V.
,
Wang
,
J.
, et al
(
2002
).
Generalized autocalibrating partially parallel acquisitions (GRAPPA)
.
Magnetic Resonance in Medicine
,
47
,
1202
1210
.
Guldin
,
W. O.
, &
Grüsser
,
O.-J.
(
1998
).
Is there a vestibular cortex?
Trends in Neurosciences
,
21
,
254
259
.
Harris
,
I. M.
, &
Miniussi
,
C.
(
2003
).
Parietal lobe contribution to mental rotation demonstrated with rTMS
.
Journal of Cognitive Neuroscience
,
15
,
315
323
.
Heil
,
M.
(
2002
).
The functional significance of ERP effects during mental rotation
.
Psychophysiology
,
39
,
535
545
.
Igelström
,
K. M.
, &
Graziano
,
M. S. A.
(
2017
).
The inferior parietal lobule and temporoparietal junction: A network perspective
.
Neuropsychologia
,
105
,
70
83
.
Ionta
,
S.
,
Heydrich
,
L.
,
Lenggenhager
,
B.
,
Mouthon
,
M.
,
Fornari
,
E.
,
Chapuis
,
D.
, et al
(
2011
).
Multisensory mechanisms in temporo-parietal cortex support self-location and first-person perspective
.
Neuron
,
70
,
363
374
.
Jola
,
C.
, &
Mast
,
F. W.
(
2005
).
Mental object rotation and egocentric body transformation: Two dissociable processes?
Spatial Cognition and Computation
,
5
,
217
237
.
Kahane
,
P.
,
Hoffmann
,
D.
,
Minotti
,
L.
, &
Berthoz
,
A.
(
2003
).
Reappraisal of the human vestibular cortex by cortical electrical stimulation study
.
Annals of Neurology
,
54
,
615
624
.
Kaski
,
D.
,
Quadir
,
S.
,
Nigmatullina
,
Y.
,
Malhotra
,
P. A.
,
Bronstein
,
A. M.
, &
Seemungal
,
B. M.
(
2016
).
Temporoparietal encoding of space and time during vestibular-guided orientation
.
Brain
,
139
,
392
403
.
Keehner
,
M.
,
Guerin
,
S. A.
,
Miller
,
M. B.
,
Turk
,
D. J.
, &
Hegarty
,
M.
(
2006
).
Modulation of neural activity by angle of rotation during imagined spatial transformations
.
Neuroimage
,
33
,
391
398
.
Kessler
,
K.
, &
Thomson
,
L. A.
(
2010
).
The embodied nature of spatial perspective taking: Embodied transformation versus sensorimotor interference
.
Cognition
,
114
,
72
88
.
Klingner
,
C. M.
,
Axer
,
H.
,
Brodoehl
,
S.
, &
Witte
,
O. W.
(
2016
).
Vertigo and the processing of vestibular information: A review in the context of predictive coding
.
Neuroscience & Biobehavioral Reviews
,
71
,
379
387
.
Kolev
,
O. I.
(
2002
).
The directions of nystagmus and apparent self-motion evoked by caloric tests and angular accelerations
.
Journal of Vestibular Research
,
11
,
349
355
.
Kozhevnikov
,
M.
, &
Hegarty
,
M.
(
2001
).
A dissociation between object manipulation spatial ability and spatial orientation ability
.
Memory & Cognition
,
29
,
745
756
.
Kozhevnikov
,
M.
,
Motes
,
M. A.
,
Rasch
,
B.
, &
Blajenkova
,
O.
(
2006
).
Perspective-taking vs. mental rotation transformations and how they predict spatial navigation performance
.
Applied Cognitive Psychology
,
20
,
397
417
.
Lenggenhager
,
B.
,
Lopez
,
C.
, &
Blanke
,
O.
(
2008
).
Influence of galvanic vestibular stimulation on egocentric and object-based mental transformations
.
Experimental Brain Research
,
184
,
211
221
.
Leone
,
G.
,
Lipshits
,
M.
,
Gurfinkel
,
V.
, &
Berthoz
,
A.
(
1995
).
Is there an effect of weightlessness on mental rotation of three-dimensional objects?
Cognitive Brain Research
,
2
,
255
267
.
Lopez
,
C.
, &
Blanke
,
O.
(
2011
).
The thalamocortical vestibular system in animals and humans
.
Brain Research Reviews
,
67
,
119
146
.
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
.
Macauda
,
G.
,
Moisa
,
M.
,
Mast
,
F. W.
,
Ruff
,
C. C.
,
Michels
,
L.
, &
Lenggenhager
,
B.
(
2019
).
Shared neural mechanisms between imagined and perceived egocentric motion—A combined GVS and fMRI study
.
Cortex
,
119
,
20
32
.
Marcus
,
D. S.
,
Harwell
,
J.
,
Olsen
,
T.
,
Hodge
,
M.
,
Glasser
,
M. F.
,
Prior
,
F.
, et al
(
2011
).
Informatics and data mining tools and strategies for the Human Connectome Project
.
Frontiers in Neuroinformatics
,
5
,
4
.
Mast
,
F. W.
, &
Ellis
,
A. W.
(
2015
).
Internal models, vestibular cognition, and mental imagery: Conceptual considerations
.
Multisensory Research
,
28
,
443
460
.
Michelon
,
P.
, &
Zacks
,
J. M.
(
2006
).
Two kinds of visual perspective taking
.
Perception & Psychophysics
,
68
,
327
337
.
Nakul
,
E.
, &
Lopez
,
C.
(
2017
).
Commentary: Out-of-body experience during awake craniotomy
.
Frontiers in Human Neuroscience
,
11
,
417
.
Parsons
,
L. M.
(
1987
).
Imagined spatial transformation of one's body
.
Journal of Experimental Psychology: General
,
116
,
172
191
.
Penny
,
W. D.
,
Friston
,
K. J.
,
Ashburner
,
J. T.
,
Kiebel
,
S. J.
, &
Nichols
,
T. E.
(Eds.). (
2011
).
Statistical parametric mapping: The analysis of functional brain images
.
London
:
Academic Press
.
R Core Team
. (
2013
).
R: A language and environment for statistical computing
.
Vienna
:
R Foundation for Statistical Computing
.
Roberts
,
D. C.
,
Marcelli
,
V.
,
Gillen
,
J. S.
,
Carey
,
J. P.
,
Della Santina
,
C. C.
, &
Zee
,
D. S.
(
2011
).
MRI magnetic field stimulates rotational sensors of the brain
.
Current Biology
,
21
,
1635
1640
.
Rorden
,
C.
, &
Brett
,
M.
(
2000
).
Stereotaxic display of brain lesions
.
Behavioural Neurology
,
12
,
191
200
.
Sack
,
A. T.
(
2009
).
Parietal cortex and spatial cognition
.
Behavioural Brain Research
,
202
,
153
161
.
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
.
Shinder
,
M. E.
, &
Taube
,
J. S.
(
2010
).
Differentiating ascending vestibular pathways to the cortex involved in spatial cognition
.
Journal of Vestibular Research
,
20
,
3
23
.
Smith
,
A. T.
,
Greenlee
,
M. W.
,
DeAngelis
,
G. C.
, &
Angelaki
,
D. E.
(
2017
).
Distributed visual–vestibular processing in the cerebral cortex of man and macaque
.
Multisensory Research
,
30
,
91
120
.
Surtees
,
A.
,
Apperly
,
I.
, &
Samson
,
D.
(
2013
).
Similarities and differences in visual and spatial perspective-taking processes
.
Cognition
,
129
,
426
438
.
Suzuki
,
M.
,
Kitano
,
H.
,
Ito
,
R.
,
Kitanishi
,
T.
,
Yazawa
,
Y.
,
Ogawa
,
T.
, et al
(
2001
).
Cortical and subcortical vestibular response to caloric stimulation detected by functional magnetic resonance imaging
.
Cognitive Brain Research
,
12
,
441
449
.
Ter Horst
,
A. C.
,
Jongsma
,
M. L. A.
,
Janssen
,
L. K.
,
van Lier
,
R.
, &
Steenbergen
,
B.
(
2012
).
Different mental rotation strategies reflected in the rotation related negativity
.
Psychophysiology
,
49
,
566
573
.
Tomasino
,
B.
, &
Gremese
,
M.
(
2016
).
Effects of stimulus type and strategy on mental rotation network: An activation likelihood estimation meta-analysis
.
Frontiers in Human Neuroscience
,
9
,
693
.
Tzourio-Mazoyer
,
N.
,
Landeau
,
B.
,
Papathanassiou
,
D.
,
Crivello
,
F.
,
Etard
,
O.
,
Delcroix
,
N.
, et al
(
2002
).
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain
.
Neuroimage
,
15
,
273
289
.
van Elk
,
M.
, &
Blanke
,
O.
(
2014
).
Imagined own-body transformations during passive self-motion
.
Psychological Research
,
78
,
18
27
.
van Elk
,
M.
,
Duizer
,
M.
,
Sligte
,
I.
, &
van Schie
,
H.
(
2017
).
Transcranial direct current stimulation of the right temporoparietal junction impairs third-person perspective taking
.
Cognitive, Affective, & Behavioral Neuroscience
,
17
,
9
23
.
Wang
,
H.
,
Callaghan
,
E.
,
Gooding-Williams
,
G.
,
McAllister
,
C.
, &
Kessler
,
K.
(
2016
).
Rhythm makes the world go round: An MEG-TMS study on the role of right TPJ theta oscillations in embodied perspective taking
.
Cortex
,
75
,
68
81
.
Wirth
,
A. M.
,
Frank
,
S. M.
,
Greenlee
,
M. W.
, &
Beer
,
A. L.
(
2018
).
White matter connectivity of the visual–vestibular cortex examined by diffusion-weighted imaging
.
Brain Connectivity
,
8
,
235
244
.
Wraga
,
M.
,
Flynn
,
C. M.
,
Boyle
,
H. K.
, &
Evans
,
G. C.
(
2010
).
Effects of a body-oriented response measure on the neural substrate of imagined perspective rotations
.
Journal of Cognitive Neuroscience
,
22
,
1782
1793
.
Wraga
,
M.
,
Shephard
,
J. M.
,
Church
,
J. A.
,
Inati
,
S.
, &
Kosslyn
,
S. M.
(
2005
).
Imagined rotations of self versus objects: An fMRI study
.
Neuropsychologia
,
43
,
1351
1361
.
Zacks
,
J. M.
(
2008
).
Neuroimaging studies of mental rotation: A meta-analysis and review
.
Journal of Cognitive Neuroscience
,
20
,
1
19
.
Zacks
,
J.
,
Rypma
,
B.
,
Gabrieli
,
J. D. E.
,
Tversky
,
B.
, &
Glover
,
G. H.
(
1999
).
Imagined transformations of bodies: An fMRI investigation
.
Neuropsychologia
,
37
,
1029
1040
.
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
.
zu Eulenburg
,
P.
,
Caspers
,
S.
,
Roski
,
C.
, &
Eickhoff
,
S. B.
(
2012
).
Meta-analytical definition and functional connectivity of the human vestibular cortex
.
Neuroimage
,
60
,
162
169
.
zu Eulenburg
,
P.
,
Müller-Forell
,
W.
, &
Dieterich
,
M.
(
2013
).
On the recall of vestibular sensations
.
Brain Structure and Function
,
218
,
255
267
.

Author notes

*

These authors shared first authorship.