Human Movement Science 40 (2015) 352–358

Contents lists available at ScienceDirect

Human Movement Science journal homepage: www.elsevier.com/locate/humov

Mental rotation of letters, body parts and scenes during whole-body tilt: Role of a body-centered versus a gravitational reference frame Otmar L. Bock, Marc Dalecki ⇑ Institute of Physiology and Anatomy, German Sport University Cologne, Germany

a r t i c l e

i n f o

Article history: Available online 13 February 2015 PsycINFO classification: 2300 2340 Keywords: Body tilt Reference frame Cognitive processing Mental rotation Embodiment

a b s t r a c t It is known that in mental-rotation tasks, subjects mentally transform the displayed material until it appears ‘‘upright’’ and then make a judgment. Here we evaluate, by using three typical mental rotation tasks with different degrees of embodiment, whether ‘‘upright’’ is coded to a gravitational or egocentric reference frame, or a combination of both. Observers stood erect or were wholebody tilted by 60°, with their left ear down. In either posture, they saw stimuli presented at different orientation angles in their frontal plane: in condition LETTER, they judged whether the stimuli were normal or mirror-reversed letters, in condition HAND whether they represented a left or a right hand, and in condition SCENE whether a weapon laid left or right in front of a displayed person. Data confirm that reaction times are modulated by stimulus orientation angle, and the modulation curve in LETTER and HAND differs from that in SCENE. More importantly, during 60° body tilt, the modulation curve shifted 12° away from the gravitational towards the egocentric vertical reference; this shift was comparable in all three conditions and independent of the degree of embodiment. We conclude that mental rotation in all conditions relied on a similar spatial reference, which seems to be a weighted average of the gravitational and the egocentric vertical, with a higher weight given to the former. Ó 2015 Elsevier B.V. All rights reserved.

⇑ Corresponding author at: School of Kinesiology and Health Science, Centre for Vision Research, York University, Toronto, ON, 357 Bethune Coll., 4700 Keele Str., M3J 1P3, Canada. Tel.: +1 416 736 2100x33405. E-mail addresses: [email protected] (O.L. Bock), [email protected] (M. Dalecki). http://dx.doi.org/10.1016/j.humov.2015.01.017 0167-9457/Ó 2015 Elsevier B.V. All rights reserved.

O.L. Bock, M. Dalecki / Human Movement Science 40 (2015) 352–358

353

1. Introduction When humans are asked to judge objects presented in their frontal plane at different orientations, their reaction times increase consistently with the angle between the displayed and the canonical ‘‘upright’’ orientation of those objects; this suggests that objects are mentally rotated into the upright before a judgment is made (Cooper & Shepard, 1973; Shepard & Metzler, 1971). In another paradigm, participants are asked to judge the location of objects in front of an avatar displayed at different orientations with respect to their own body. In this case, reaction times increase with the angle between the avatar and the own body, which again suggests that mental rotation took place (Ratcliff, 1979; Semmes, Weinstein, Ghent, & Teuber, 1963; Zacks, Rypma, Gabrieli, Tversky, & Glover, 1999). However, several lines of evidence suggest that the underlying processes differ between the two paradigms: the speed of mental rotation is not the same (Kessler & Thomson, 2010; Kozhevnikov, Motes, Rasch, & Blajenkova, 2006;), clinical deficits are dissociated (Fiorio et al., 2007; Ratcliff, 1979; Semmes et al., 1963) and neural activation patterns are different (Thomas, Dalecki, & Abeln, 2013; Wraga, Shephard, Church, Inati, & Kosslyn, 2005; Zacks et al., 1999). These discrepancies have been interpreted as evidence that in the former paradigm, objects are mentally rotated with respect to the own stationary body while in the latter paradigm, the own body is mentally rotated with respect to the stationary scene (Jola & Mast, 2005; Kessler & Thomson, 2010). When objects are mentally rotated, they could be judged as ‘‘upright’’ relative to the own body (egocentric reference), or relative to the gravitational vertical (gravicentric reference). To distinguish between these alternatives, some studies compared subjects’ performance in an upright posture with that in a head-tilted or whole body-tilted posture, and found that displayed objects were mentally rotated into an orientation that was in-between the egocentric and the gravitational vertical (Corballis, Nagourney, Shetzer, & Stefanatos, 1978; Friedman & Hall, 1996; McMullen & Jolicoeur, 1992). It therefore appears that the spatial reference for mental rotation of extrinsic objects is formed by combining egocentric and gravitational cues. The present study expands this work by evaluating the spatial reference for mental rotation not of extrinsic objects, but rather of one’s own body. Since the behavioral and neuronal correlates for these two mental transformations differ (see above), it is conceivable that the spatial reference differs as well. Two earlier studies asked subjects to judge the identity of visual landscapes (Gaunet & Berthoz, 2000) or hand drawings (Sekiyama, 1982), and observed a predominance of the egocentric reference; however, it is unclear whether subjects in those studies mentally rotated the displayed material or their own body, and the spatial reference for own-body rotation therefore remains unclear. Moreover, mental rotation stimuli differ in their level of embodiment, which is likely to be low for external objects, higher for human hands and highest for scenes containing an avatar (Jola & Mast, 2005; Kessler & Thomson, 2010; Wraga et al., 2005). The latter type of stimulus is therefore most likely to induce mental rotation of one’s own body, which is why we selected it for our study. To compare mental rotation of the own body with that of extrinsic objects, we included a letter rotation task. We also included a third popular variant, the hand rotation task. All three tasks were administered to the same subjects in the same experimental session, to quantify differences between tasks unconfounded by interindividual variability or by fluctuations of mood, fatigue, food intake, etc.; serial-order effects were controlled by counterbalancing the order of stimuli and of body postures. By using a letter rotation, hand rotation and own body rotation task we were also able to compare whether the reference frame for ‘upright’ judgments differs between different degrees of embodiment. Furthermore, we decided to increase the statistical power of our data by using a sample size of n = 24, while n = 10 was the typical used sample size in the previous studies on mental rotation with body tilt. The main aim of the present study was therefore to elucidate the effects of whole-body tilt on mental rotation of three common stimuli (external objects, body parts, whole body), to find out whether the reference frame for ‘‘upright’’ judgments differs between tasks.

354

O.L. Bock, M. Dalecki / Human Movement Science 40 (2015) 352–358

2. Methods 2.1. Subjects 24 right handed participants (12 female) aged 24.63 ± 5.22 years were tested. They were healthy by self-report, and were not familiar with sensorimotor research. All signed an informed consent statement for this study, which was pre-approved by the authors’ institutional ethics committee. 2.2. Setup As in our previous work, observers viewed a computer screen (Dalecki, Hoffmann, & Bock, 2012), but in contrast to the previous study, they viewed the screen through a cylindrical aperture of 21.5 cm diameter, which excluded all peripheral visual information. In condition LETTER, they saw the letter ‘‘G’’ or ‘‘R’’ and had to decide whether it was mirror-reversed or non-reversed. In condition HAND, they saw line drawings of the dorsum or palm of a hand and had to decide whether it was a left or a right hand. In condition SCENE, they saw schematic drawings of a person in top view; the person sat at a round table with a gun or knife in front of one shoulder and a flower in front of the other shoulder, and observers had to decide in front of which shoulder the weapon was located. Fig. 1 illustrates which orientation of the depicted person should yield the shortest (top row) and the longest (middle row) reaction times when observers (A) are erect, (B) are tilted and the egocentric reference prevails or (C) are tilted and the gravitational reference prevails. 2.3. Procedures In all three conditions, images were presented in a mixed sequence of twelve orientations (0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, and 330°, where 0° represents the gravity-centered upright), four objects (left dorsum, right dorsum, left palm, right palm in HAND; reversed G, nonreversed G, reversed R, non-reversed R in LETTER; gun on the left, gun on the right, knife on the left, knife on the right in SCENE) and two repetitions, for a total of 96 trials for each condition. In each condition, a short rest break was given after the first 48 trials. Overall performed each observer 588 trials, subdivided into a 98 (trials)  3 (Condition)  2 (Posture) design.

Fig. 1. Expected outcome of a mental-rotation task in which a human body is displayed in top view at different orientations, in the observer’s frontal plane. We know from earlier work that for an erect observer, reaction times are shortest when the displayed body is aligned with the observer (top square in (A)) and longest when it is opposite (bottom square in (A)). When the observer is tilted sideways, shortest and longest reaction times should shift along with the observer if mental rotation uses an egocentric reference frame (B), but should not shift if mental rotation uses a gravitational reference frame (C). Note that for clarity, Fig. 1 shows the stimuli as if they were presented above the observers’ head; actually, they were presented at eye level.

O.L. Bock, M. Dalecki / Human Movement Science 40 (2015) 352–358

355

Participants were instructed to respond quickly and accurately by depressing a key with their right or left index fingertip, where the right key coded the image of a right hand, a non-reversed letter or a weapon to the right, respectively. The next image was presented 0.5–1.0 s after the response. Incorrect responses elicited a short warning sound. The experiment began with 16 practice trials in each condition, and participants were then tested in all three conditions following a counterbalanced order. Dependent variable was the median reaction time of correct responses in each condition and orientation. The rate of false responses was about 7%, which was very similar to our previous study (Dalecki et al., 2012), and comparable for both body postures (see below, t-test: p > .05). Unlike in our previous study (Dalecki et al., 2012), participants were tested in two body postures: once while standing upright and once while tilted 60° left shoulder down. To ensure a constant angle of tilt, subjects laid left-side down on a 60° tilted, padded platform; their head was supported by a sturdy pillow such that the long axes of head and trunk were congruent, and their feet, hip, shoulder and head were strapped to the platform. The two postures were administered in counterbalanced order and affected only the subject, not the computer screen. As a consequence, the orientation of the displayed stimuli was registered – and will be reported – as angles with respect to the gravitational vertical, not with respect to the egocentric vertical. The experiment last a total of 35 min per participant, with 9 min of instruction and practice trials, followed by 13 min of testing in each body posture. 2.4. Data analysis The relationship between stimulus orientation and reaction time was assessed by fitting Gaussian regression curves separately to the data from each observer, condition and body posture. Each fit yielded the parameters mode (i.e., stimulus orientation with the longest reaction time), height (i.e., difference between longest and shortest reaction time) and width (i.e., sharpness of the regression curve). Each parameter was then submitted to a two-way analysis of variance (ANOVA) with repeated measures on Condition (HAND, LETTER, SCENE) and Posture (upright, tilted). 3. Results Fig. 2 illustrates the reaction time across observers for each stimulus orientation and condition when participants were (A) upright or (B) tilted 60° left-shoulder down. As well-known from literature, reaction time was consistently modulated by stimulus orientation and was shorter for SCENE

Fig. 2. Median reaction time for the three experimental conditions, HAND, LETTER and SCENE, in erect (A) and in tilted observers (B). The abscissa gives angles with respect to a gravitational reference, and the arrow in (B) indicates the 180° angle with respect to an egocentric reference. Symbols represent across-participant means, and brackets the pertinent standard errors.

356

O.L. Bock, M. Dalecki / Human Movement Science 40 (2015) 352–358

Fig. 3. (A) Illustrates Gaussian curve fits to the data of one observer in condition HAND, once with the observer erect and once with the observer tilted 60°. The curve for the tilted posture is somewhat lower and broader, and its mode is shifted somewhat to the left if compared to the erect posture. (B–D) Summarizes the Gaussian fit parameters across all data sets: if compared to upright observers, curves in tilted observers had a lower height (C), a larger width (C), and their mode was shifter to the left (D). Histogram bars represent across-participant means and brackets the pertinent standard errors. ⁄Denotes p < .05; ⁄⁄denotes p < .01.

than for HAND and LETTER (c.f. Dalecki et al., 20121; Harris et al., 2000; Kessler & Thomson, 2010; Parsons, 1987). More importantly, reaction time was shortest for orientations near 0° and longest for those near 180° for all three conditions and both body postures. Since the abscissa in Fig. 2 is scaled with respect to the gravitational and not the egocentric reference (see Section 2), the data indicate that reaction time was more closely associated with the gravitational reference. The arrow in Fig. 2B indicates where the longest reaction times should have occurred if the egocentric reference had prevailed. Fig. 3A illustrates Gaussian fits to the data of a given observer in condition HAND; one curve represents data collected with the observer erect, and the other those with the observer tilted. Both curves look generally similar, but closer inspection reveals that the curve for the tilted observer is somewhat lower, wider and shifted to the left. These observations were confirmed when the Gaussian fit parameters of all subjects were submitted to ANOVAs: a significance of Posture was yielded for height (F(1,23) = 6.43; p < .05), width (F(1,23) = 7.619; p < .05) as well as mode (F(1,23) = 12.43; p < .01), as illustrated in Fig. 3B–D. ANOVAs further yielded a significance of Condition for height (F(2,46) = 19.28; p < .001), which was lower in SCENE than in HAND or LETTER, and for mode

1 When comparing the median reaction times from the present upright condition (Fig. 2A) with those from our previous uprightonly study (Dalecki et al., 2012), an ANOVA with the group factor Study (previous, present) and repeated measures on Condition (HAND, LETTER, SCENE) and Orientation (0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, and 330°) yielded no significant effects of Study and its interactions (all p > .05), which confirms the similarity of data from both studies.

O.L. Bock, M. Dalecki / Human Movement Science 40 (2015) 352–358

357

(F(2,46) = 6.53; p < .01), which was shifted 16° to the left for both tilt angles in HAND if compared to LETTER and SCENE. The interaction of Posture and Condition was not significant for any parameter (all p > .05). During 60° body tilt, mode was shifted 12° to the left (from the ‘upright’ graviceptive reference frame), independent of Condition. The mean speed of mental rotation, calculated as s = 180°/height of Gaussian fit, was 370.03, 430.87 and 687.47°/s for HAND, LETTER and SCENE in the upright body posture, and was 459.52, 533.14 and 778.37°/s for HAND, LETTER and SCENE in the tilted body posture, which is in accordance to the results of height. 4. Discussion The main purpose of the present study was to find out whether metal rotation uses the same spatial reference in three typical mental rotation paradigms: external object rotation, body part rotation and whole-body rotation. If it were based on distinct references, we should expect a significant Condition (LETTER, HAND, SCENE)  Posture (0° upright body position, 60° left ear down tilted body position) term for the parameter mode. The present data document once more that reaction times in mental-rotation tasks are modulated by stimulus orientation, and that this modulation differs between condition SCENE on one side, and HAND and LETTER on the other side (cf. Dalecki et al., 2012), independent of the body posture. More importantly, we found that a 60° whole-body tilt didn’t shift the modulation by 60°, but rather by only about 12° away from the gravitational ‘upright’ to the egocentric ‘upright’ reference. The shift was comparable for all three used mental rotation tasks. From this we conclude that our participants based mental rotation, neither on a gravitational reference, since the shift was larger than 0°, nor on an egocentric reference, since the shift was distinctly smaller than 60°. Instead, they may have used a weighted average of the gravitational and the egocentric reference, assigning a higher weight to the former. This fits well with earlier studies on object rotation (Corballis et al., 1978; Friedman & Hall, 1996; McMullen & Jolicoeur, 1992) and on human hand rotation (Sekiyama, 1982), but not with a study on the rotation of complex visual scenes, in which an egocentric reference predominated (Gaunet & Berthoz, 2000). A possible explanation of this discrepancy is that in the latter study, the angle of tilt was only 33°: the egocentric reference may have a high weight at small tilt angles, but a low weight at large tilt angles. However, additional research would be needed to verify this view, for example by using a similar experiment protocol as used in the present study except including a 30°, 60° and 90° whole body tilt. Since we didn’t found a significant Condition  Posture effect we further conclude that the reference frame was independent of the degree of embodiment that differs between the three chosen mental rotation tasks (cf. Introduction). We further found that 60° whole-body tilt left-ear down not only shifted the reaction-time curves sideways, but also made them lower and broader. This is to be expected when averaging two curves, spaced 60° apart. Our finding therefore provides additional support for the weighted-average hypothesis. The same finding has yet another implication: lower and broader reaction-time curves correspond to a higher speed of mental rotation (cf. Fig. 3A), i.e., the rotation speed of our observers increased under whole-body tilt. No such increase was observed when subjects were tested supine on Earth (Francuz, 2010; Mast, Ganis, Christie, & Kosslyn, 2003), or in weightlessness (Dalecki, Dern, & Steinberg, 2013; Grabherr et al., 2007; Leone, Lipshits, Gurfinkel, & Berthoz, 1995), possibly because gravity was irrelevant in the latter studies rather than being in conflict with the egocentric reference. Conceivably, a conflict between fundamental reference frames speeds up all cognitive processes since such a situation is potentially dangerous to the organism. Summing up, we conclude that observers rotated the visual display into a canonical upright orientation located between the gravitational and the egocentric upright, but closer to the former, in all three tasks. We thus have no evidence that the spatial reference for ‘upright’ is different for mental rotation of extrinsic objects versus of one’s own body. Acknowledgments The authors wish to thank N. Elting and T. Horel for help in data acquisition, and L. Geisen, T. Kesnerus and M. Drescher for software development. Our work was supported by the German Aerospace

358

O.L. Bock, M. Dalecki / Human Movement Science 40 (2015) 352–358

Center (DLR) on the behalf of the German Ministry for Research and Technology (Grant 50WB0726). Responsibility for the contents rests with the authors. References Cooper, L. A., & Shepard, R. N. (1973). Chronometric studies of the rotation of mental images. In Visual information processing (pp. 175–176). New York: Academic Press. Corballis, M. C., Nagourney, B. A., Shetzer, L. I., & Stefanatos, G. (1978). Mental rotation under head tilt: Factors influencing the location of the subjective reference frame. Perception & Psychophysics, 24, 263–273. Dalecki, M., Dern, S., & Steinberg, S. (2013). Mental rotation of a hand, letter and complex scene in microgravity. Neuroscience Letters, 533, 55–59. Dalecki, M., Hoffmann, U., & Bock, O. (2012). Mental rotation of letters, body parts and complex scenes: Separate or common mechanisms? Human Movement Science, 31, 1151–1160. Fiorio, M., Tenazzi, M., Ionta, S., Fiaschi, A., Moretto, G., Edwards, M. J., et al (2007). Mental rotation of body parts and noncorporeal objects in patients with idiopathic cervical dystonia. Neuropsychology, 45, 2346–2354. Francuz, P. (2010). The influence of body position and spatial orientation of an object on mental rotation task’s performance. Procedia Social and Behavioral Sciences, 5, 896–900. Friedman, A., & Hall, D. L. (1996). The importance of being upright: Use of environmental and viewer-centered reference frames in shape discriminations of novel three-dimensional objects. Memory & Cognition, 24, 285–295. Gaunet, F., & Berthoz, A. (2000). Mental rotation for spatial environment recognition. Cognitive Brain Research, 9, 91–102. 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. Harris, I. M., Egan, G. F., Sonkkila, C., Tochon-Danguy, H. J., Paxinos, G., & Watson, J. D. G. (2000). Selective right parietal lobe activation during mental rotation A parametric PET study. Brain, 123, 65–73. Jola, C., & Mast, F. W. (2005). Mental object rotation and egocentric body transformation: Two dissociable processes? Spatial Cognition and Computation, 5, 217–237. Kessler, K., & Thomson, L. A. (2010). The embodied nature of spatial perspective taking: Embodied transformation versus sensorimotor interference. Cognition, 114, 72–88. Kozhevnikov, M., Motes, M. A., Rasch, B., & Blajenkova, O. (2006). Perspective-taking vs. mental rotation transformations and how they predict spatial navigation performance. Neuroimage, 20, 397–417. Leone, G., Lipshits, M., Gurfinkel, V., & Berthoz, A. (1995). Is there an effect of weightlessness on mental rotation of threedimensional objects? Cognitive Brain Research, 2, 255–267. Mast, F. W., Ganis, G., Christie, S., & Kosslyn, S. M. (2003). Four types of visual mental imagery processing in upright and tilted observers. Cognitive Brain Research, 17, 238–247. McMullen, P. A., & Jolicoeur, P. (1992). Reference frame and effects of orientation on finding the tops of rotated objects. Journal of Experimental Psychology: Human Perception and Performance, 18, 807–820. Parsons, L. M. (1987). Imagined spatial transformation of one’s hands and feet. Cognitive Psychology, 19, 178–241. Ratcliff, G. (1979). Spatial thought, mental rotation and the right cerebral hemisphere. Neuropsychologia, 17, 49–54. Sekiyama, K. (1982). Kinesthetic aspects of mental representations in the identification of left and right hands. Perception & Psychophysics, 32, 89–95. Semmes, J., Weinstein, S., Ghent, L., & Teuber, H. L. (1963). Correlates of impaired orientation in personal and extrapersonal space. Brain, 86, 747–772. Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional objects. Science, 171, 701–703. Thomas, M., Dalecki, M., & Abeln, V. (2013). EEG coherence during mental rotation of letters, hands and scenes. International Journal of Psychophysiology, 89, 128–135. Wraga, M., Shephard, J. M., Church, J. A., Inati, S., & Kosslyn, S. M. (2005). Imagined rotations of self versus objects: An fMRIstudy. Neuropsychologia, 43, 1351–1361. Zacks, J. M., Rypma, B., Gabrieli, J. D. E., Tversky, B., & Glover, G. H. (1999). Imagined transformations of bodies: An fMRI investigation. Neuropsychologia, 37, 1029–1040.

Mental rotation of letters, body parts and scenes during whole-body tilt: role of a body-centered versus a gravitational reference frame.

It is known that in mental-rotation tasks, subjects mentally transform the displayed material until it appears "upright" and then make a judgment. Her...
590KB Sizes 0 Downloads 6 Views