SHORT REPORT Comparison of Visual Sensitivity to Human and Object Motion in Autism Spectrum Disorder

Transcription

1 SHORT REPORT Comparison of Visual Sensitivity to Human and Object Motion in Autism Spectrum Disorder Martha D. Kaiser, Lara Delmolino, James W. Tanaka, and Maggie Shiffrar Successful social behavior requires the accurate detection of other people s movements. Consistent with this, typical observers demonstrate enhanced visual sensitivity to human movement relative to equally complex, nonhuman movement [e.g., Pinto & Shiffrar, 2009]. A psychophysical study investigated visual sensitivity to human motion relative to object motion in observers with autism spectrum disorder (ASD). Participants viewed point-light depictions of a moving person and, for comparison, a moving tractor and discriminated between coherent and scrambled versions of these stimuli in unmasked and masked displays. There were three groups of participants: young adults with ASD, typically developing young adults, and typically developing children. Across masking conditions, typical observers showed enhanced visual sensitivity to human movement while observers in the ASD group did not. Because the human body is an inherently social stimulus, this result is consistent with social brain theories [e.g., Pelphrey & Carter, 2008; Schultz, 2005] and suggests that the visual systems of individuals with ASD may not be tuned for the detection of socially relevant information such as the presence of another person. Reduced visual sensitivity to human movements could compromise important social behaviors including, for example, gesture comprehension. Keywords: autism; motion perception; biological motion; motion coherence Introduction When people move, their actions convey extensive social information. Social brain theories [e.g., Schultz, 2005] suggest that the visual system is typically tuned for the analysis of socially relevant information. Consistent with this, typical adult observers demonstrate greater visual sensitivity to human movement than to complex, nonhuman movement [e.g., Pinto & Shiffrar, 2009]. Because individuals with Autism Spectrum Disorder (ASD) experience compromised social abilities [APA, 2006] and atypical neural activity in social brain areas [e.g., Pelphrey & Carter, 2008], we examined whether observers with ASD exhibit or lack enhanced visual sensitivity to human movement. Studies of visual sensitivity to human motion commonly use point-light stimuli created by attaching markers to a person s body and then recording that person s movements so that only the markers are visible [Johansson, 1973]. While the resultant displays are sparse (Fig. 1), typical adults readily detect point-light actors [Blake & Shiffrar, 2007]. Typically developing infants readily orient toward point-light displays of biological motion [Simion, Regolin, & Bulf, 2008] while toddlers with ASD do not [Klin, Lin, Gorrindo, Ramsay, & Jones, 2009]. Reduced visual experience with human motion typically decreases visual sensitivity to it [Jacobs, Pinto, & Shiffrar, 2005]. Observers with ASD and control observers do not differ in their ability to verbally describe the actions performed by point-light actors [Hubert et al., 2007; Moore, Hobson, & Lee, 1997; Parron et al., 2008] but do differ in their ability to detect the presence of coherent human motion in point-light displays [Blake, Turner, Smoski, Pozdol, & Stone, 2003]. Observers with ASD also frequently demonstrate impairments in their sensitivity to coherent visual motion in general [for review see Kaiser & Shiffrar, 2009; Simmons et al., 2009]. This raises the question of whether observers with ASD are compromised in their visual sensitivity to coherent human motion, in specific, or in their visual sensitivity to coherent motion, in general. Observers with and without ASD process point-light displays of human motion differently. Activity in the posterior region of the superior temporal sulcus (STSp) is necessary for the visual perception of point-light human motion [Grossman, Battelli, & Pascual-Leone, 2005; Saygin, 2007] but not object motion [Beauchamp, Lee, Haxby, & Martin, 2003]. Observers with ASD exhibit From the Child Study Center, Yale School of Medicine, New Haven, Connecticut (M.D.K.); Department of Psychology, Rutgers University, Newark, New Jersey (M.D.K., M.S.); Douglas Developmental Disabilities Center, Rutgers University, New Brunswick, New Jersey (L.D.); Department of Psychology, University of Victoria, Victoria, British Columbia (J.W.T.) Received October 14, 2009; revised March 8, 2010; accepted for publication April 5, 2010 Address for correspondence and reprints: Martha D. Kaiser, Child Neuroscience Laboratory, Yale Child Study Center, 230 South Frontage Road, New Haven, CT martha.kaiser@yale.edu Grant sponsor: Simons Foundation; Grant number: 94915; Grant sponsors: National Sciences and Engineering Research Council of Canada, James S. McDonnell Foundation; Temporal Dynamics of Learning Center; Grant number: SBE Published online in Wiley InterScience (www. interscience.wiley.com) DOI: /aur.137 & 2010 International Society for Autism Research, Wiley Periodicals, Inc. INSAR Autism Research 3: 1 5,

2 Typical observers can use local (point-by-point) and global (spatiotemporally extended) analyses to detect coherence in unmasked point-light displays. Masking disrupts local analyses and, in typical observers, increases reliance on global analyses [Bertenthal & Pinto, 1994]. Because observers with ASD rely on local analyses [e.g., Behrmann, Thomas, & Humphreys, 2006; Dakin & Frith, 2005], masking should disrupt their visual sensitivity to human motion. Methods Participants Figure 1. Photographs of the person (A) and tractor (B) within the motion capture system. Static images cropped from the resultant point-light movies of the person (C) and tractor (D). While point-light displays are difficult to recognize when static [Johansson, 1973], observers readily detect these displays when set in motion. Slides from movies depicting scrambled human motion (E) and coherent human motion in a mask (F). Masks were constructed by scrambling the initial locations of the points from a duplicate stimulus. Then, the walker was hidden within the mask. structural and processing atypicalities in the STSp [e.g., Zilbovicius et al., 2006] that are evident during the perception of point-light walkers [Freitag et al., 2008; Herrington et al., 2007]. During development in typical children, but not in children with ASD, STSp activity becomes increasingly tuned for human motion and less responsive to object motion [Pelphrey & Carter, 2008]. To the extent that perceptual sensitivity to point-light displays of human movement depends upon STSp activity [e.g., Grossman et al., 2005; Saygin, 2007], observers with ASD may not experience enhanced sensitivity to human motion over other types of motion. To determine whether observers with ASD, like typical observers, demonstrate enhanced sensitivity to human movement relative to object movement, we conducted a motion coherence task with point-light displays of a moving person and a moving tractor. Tractor motion was selected because, like human motion, it is complex, globally nonrigid, and jointed. An ASD group (lowest reading level years) included 6 males (mean age years, SD ) with a diagnosis of High Functioning Autism (3), Asperger s Syndrome (2), or PDD-NOS (1) recruited from a school for people with ASD. They received monetary compensation for their participation. All ASD diagnoses were made by an independent clinician and confirmed by a school district child study team or another clinician. A typical group contained 32 Rutgers undergraduates (26 female) who received credit toward a course requirement (mean age years, SD 5 3.5). Mean ages across groups did not significantly differ, t(5) , P All participants were naive to the hypothesis under investigation, had normal or corrected to normal visual acuity, and provided written informed consent. The Rutgers University Institutional Review Board approved this study. Apparatus Stimuli appeared on a 14-inch Dell TM monitor (60 Hz, 1, pixel resolution) positioned 52 cm from the observer and controlled by a Dell TM Pentium computer. The experiment was programmed in E-Prime (Psychology Software Tools, Inc., Pittsburgh, PA). Movies were processed with Motion Builder 5.0 (Kaydara TM, San Rafael, CA). A ReActor motion capture system (Ascension Technology) measured sensors attached to a person or tractor moving within a 3.6 by 4.8 m area (Fig. 1). Stimuli Nine sensors were attached to the actor (head, wrists (2), elbow, shoulder, feet (2), knee, waist) and nine were attached to a John Deere Loader (Peg Perego, cm) toy tractor (wheels (4), front bucket (4), bucket pivot joint (1)). The tractor and actor each repeatedly performed three similar actions: (1) locomoting along a 3 m linear path (2) rotating the bucket (tractor) or reaching (human) downward to pick something up and (3) locomoting 1.5 m and then rotating the bucket/reaching down to pick something up. Motion capture data were converted into eight pointlight human movies and eight point-light tractor movies 2 Kaiser et al./asd and complex motion perception INSAR

3 (5 sec each). Motion direction (leftward/rightward) was counterbalanced. The point-light walker (maximum extents: degrees of visual angle (DVA)) and tractor (maximum extents: DVA) had lateral displacement distances ranging from 8.8 to 16.1 DVA and speeds ranging from 1.76 to 3.22 DVA/sec. The points defining each stimulus were white, 0.33 DVA in diameter, and appeared against a black background. One scrambled point-light movie was constructed from each coherent human and tractor movie by scrambling the starting locations of the points (Fig. 1E). Across conditions, each stimulus appeared unmasked and masked. Unmasked stimuli had nine points as described above. Masked stimuli contained an additional nine masking points. Mask construction is described in detail elsewhere [Chouchourelou, Matsuka, Harber, & Shiffrar, 2006]. Briefly, each point-light stimulus was duplicated and the starting locations of those duplicate points were scrambled. Each stimulus was then hidden within the scrambled mask that had been constructed from it (Fig. 1F). Thus, the points defining each mask and stimulus had identical velocities, sizes, and luminance. Under these conditions, coherent motion detection involves global analyses [Bertenthal & Pinto, 1994]. Design and Procedure In a blocked, within subjects design, each participant saw coherent and scrambled point-light human motion and tractor motion in masked and unmasked conditions. Trials were blocked by stimulus (human or tractor) and condition (masked or unmasked). Participants completed the unmasked condition first. Each block contained 32 movies (8 coherent and 8 scrambled movies each shown twice in random order). Block order was counterbalanced across participants within each condition. In each masking condition, participants completed two human blocks and two tractor blocks. Participants responded by pressing a button labeled yes or another labeled no. In the human blocks, participants reported whether all (unmasked condition) or some (masked condition) of the dots were stuck to a person. In the tractor blocks, participants reported whether all (unmasked condition) or some (masked condition) of the dots were stuck to a tractor. Correct responses were yes to coherent and no to scrambled motion. No feedback was provided. Reaction time data were not analyzed because participants responded after watching each movie. Results Visual sensitivities to human motion and tractor motion were assessed with d-prime measures (normalized hit rate minus normalized false alarm rate) for each participant for each stimulus type in each condition [Macmillan & Figure 2. (A) Unmasked condition: Typical adult observers demonstrate significantly greater visual sensitivity to the presence of coherent human motion than to the presence of coherent tractor motion. Observers with ASD demonstrate equivalent sensitivity. (B) Masked condition: When point-light targets are hidden within point-light masks, typical adult observers detect human motion coherence better than tractor motion coherence. Observers with ASD exhibit equivalent performance to both types of motion. (C) Unmasked condition of follow-up study with typically developing young observers aged 7 8 (35 children), 9 (26 children), and (22 children) years. (D) Masked condition with the same typically developing children. Error bars indicate the standard error. Creelman, 1991]. Both groups performed well above chance (d-prime 5 0), in all unmasked (all Pso0.001) and masked (all Pso0.002) conditions (Fig. 2A and B). Due to limited matching of participant groups, direct statistical comparison across groups was not performed. Instead, data were compared within each group. The ASD group demonstrated equivalent visual sensitivity to the presence of coherent human motion and coherent tractor motion in the unmasked, t(5) , P , and masked, t(5) , P , conditions. Conversely, the typical group demonstrated significantly greater visual sensitivity to the presence of coherent human motion than to the presence of coherent tractor motion in the unmasked, t(31) , Po0.001, and masked, t(31) , Po0.001, conditions. Performance by the six typical males, considered separately, still showed greater sensitivity to human motion (e.g., masked t(5) , Po0.05). Repeated measures analysis of variances were conducted for both groups separately, with Masking and Stimulus as within subject factors. The ASD group showed no significant main effect of Masking, F(1) , P , or Stimulus, F(1) , P , nor a significant interaction, F(1,5) , INSAR Kaiser et al./asd and complex motion perception 3

4 P For the typical group, there were significant main effects of Masking, F(1) , Po0.0001, and Stimulus, F(1) , Po0.0001, but no interaction, F(1,31) , P Discussion This study examined whether observers with ASD, like typical observers, demonstrate greater visual sensitivity to human movement than to nonhuman movement. There are three main findings. First, typical observers demonstrated greater sensitivity to human motion than to tractor motion, consistent with the hypothesis that the human visual system is typically tuned for the detection and analysis of socially relevant information [e.g., Pinto & Shiffrar, 2009; Schultz, 2005]. Second, observers with ASD demonstrated equivalent levels of visual sensitivity to coherent human and tractor motions, suggesting that their visual systems may not be tuned for the detection of socially relevant motion. Because this study assessed visual sensitivity to human motion relative to object motion, these results cannot be attributed to global deficits in coherent motion perception [Koldewyn, Whitney, & Rivera, 2010]. Consistent with past studies, the current results indicate that observers with ASD can detect coherent human movement [e.g., Atkinson, 2009; Blake et al., 2003; Moore et al., 1997]. The novel finding here is that, unlike typical observers, observers with ASD do not show enhanced visual sensitivity to human movement relative to object movement. This finding is consistent with evidence that, during development, STSp activity increasingly differentiates human motion from object motion in typicals but not in children with ASD [Pelphrey & Carter, 2008]. Finally, all participants could detect coherent human and tractor motion in masked displays [Annaz et al., 2010; Murphy, Brady, Fitzgerald, & Troje, 2009] that reduced the effectiveness of local processes [Bertenthal & Pinto, 1994]. Masking may not engage global motion processes in observers with ASD since masking reduced performance in the typical but not the ASD group. Global processes are typically important for action perception [e.g., Blake & Shiffrar, 2007]. The current data are subject to interpretive caveats. The ASD group had only six participants. Yet, each completed 256 trials, yielding 1,536 total trials. This provided far more data than previously reported in related studies. Nonetheless, we are currently running more participants to address problems associated with small sample sizes. Importantly, the current results are consistent with a recent finding that typical observers with low AQ scores (few autistic traits) show greater visual sensitivity to human than to tractor motion while typical observers with high AQ scores do not [Kaiser & Shiffrar, 2010]. Given limited matching between groups, performance patterns were only compared within each group. To determine whether performance in the ASD group reflected the inclusion of observers with lower mental ages, a shortened (one block per condition) but otherwise identical version of the current task was administered to a significantly younger group of 83 (40 males) typically developing children (Mean Age years, SD years) at the University of Victoria. Like typical adults a decade older, these young typically developing children demonstrated greater visual sensitivity to the presence of coherent human than tractor motion under masked, t(82) , Po0.0001, and unmasked, t(82) , Po0.0001, conditions (Fig. 2C and D). Thus, mental age differences are unlikely to account for the group differences reported above. Acknowledgments This study was funded by grants from the Simons Foundation (]94915), National Sciences and Engineering Research Council of Canada, the James S. McDonnell Foundation, and the Temporal Dynamics of Learning Center (NSF Grant ]SBE ). The authors thank John Franchak for assistance with stimulus construction, Zena Fermano for assistance with data collection, and the staff at the Douglass Developmental Disabilities Center for their assistance with the autistic participants. References American Psychiatric Association. (2006). Diagnostic and statistical manual of mental disorders, 4e., text rev. (DSM- IV-TR). Washington, DC: APA. Annaz, D., Remington, A., Milne, E., Coleman, M., Campbell, R., et al. (2010). Atypical development of motion processing trajectories in children with autism. Developmental Science [Published online on December 28, 2009]. Atkinson, A.P. (2009). Impaired recognition of emotions from body movements is associated with elevated motion coherence thresholds in autism spectrum disorders. Neuropsychologia, 47, Beauchamp, M.S., Lee, K.E., Haxby, J.V., & Martin, A. (2003). fmri responses to video and point-light displays of moving humans and manipulable objects. Journal of Cognitive Neuroscience, 15, Behrmann, M., Thomas, C., & Humphreys, K. (2006). Seeing it differently: Visual processing in autism. Trends in Cognitive Sciences, 10, Bertenthal, B., & Pinto, J. (1994). Global processing of biological motions. Psychological Science, 5, Blake, R., & Shiffrar, M. (2007). Perception of human motion. Annual Reviews Psychology, 58, Blake, R., Turner, L.M., Smoski, M.J., Pozdol, S.L., & Stone, W.L. (2003). Visual recognition of biological motion is impaired in children with autism. Psychological Science, 14, Kaiser et al./asd and complex motion perception INSAR

5 Chouchourelou, A., Matsuka, T., Harber, K., & Shiffrar, M. (2006). The visual analysis of emotional actions. Social Neuroscience, 1, Dakin, S., & Frith, U. (2005). Vagaries of visual perception in autism. Neuron, 48, Freitag,C.M.,Konrad,C.,Haberlen,M.,Kleser,C.,vonGontard,A., et al. (2008). Perception of biological motion in autism spectrum disorders. Neuropsychologia, 46, Grossman, E.D., Battelli, L., & Pascual-Leone, A. (2005). Repetitive TMS over STSp disrupts perception of biological motion. Vision Research, 45, Herrington, J.D., Baron-Cohen, S., Wheelwright, S.J., Singh, K.D., Bullmore, E.T., et al. (2007). The role of MT1/V5 during biological motion perception in Asperger Syndrome: An fmri study. Research in Autism Spectrum Disorders, 1, Hubert, B., Wicker, B., Moore, D.G., Monfardini, E., Duverger, H., et al. (2007). Brief Report: Recognition of emotional and nonemotional biological motion in individuals with autistic spectrum disorders. Journal of Autism and Developmental Disorders, 37, Jacobs, A., Pinto, J., & Shiffrar, M. (2004). Experience, context, and the visual perception of human movement. Journal of Experimental Psychology: Human Perception & Performance, 30, Johansson, G. (1973). Visual perception of biological motion and a model for its analysis. Perception and Psychophysics, 14, Kaiser, M.D., & Shiffrar, M. (2009). The visual perception of motion by observers with autism spectrum disorders: A review and synthesis. Psychonomic Bulletin and Review, 16, Kaiser, M.D., & Shiffrar, M. (2010). Variability in the visual perception of human motion as a function of the observer s autistic traits. In: Johnson K, Shiffrar M, editors. Visual perception of the human body in motion: Findings, theory, and practice. Oxford: Oxford University Press. Klin, A., Lin, D.J., Gorrindo, P., Ramsay, G., & Jones, W. (2009). Two-year-olds with autism orient to non-social contingencies rather than biological motion. Nature, 459, Koldewyn, K., Whitney, D., & Rivera, S.M. (2010). The psychophysics of visual motion and global form processing in autism. Brain, 133, Macmillan, N., & Creelman, C. (1991). Detection theory: A user s guide. Cambridge, UK: Cambridge University Press. Moore, D.G., Hobson, R.P., & Lee, A. (1997). Components of person perception: An investigation with autistic, nonautistic retarded and typically developing children and adolescents. British Journal of Developmental Psychology, 15, Murphy, P., Brady, N., Fitzgerald, M., & Troje, N.F. (2009). No evidence for impaired perception of biological motion in adults with autistic spectrum disorders. Neuropsychologia, 47, Parron, C., Da Fonseca, D., Santos, A., Moore, D.G., Monfardini, E., & Dereulle, C. (2008). Recognition of biological motion in children with autistic spectrum disorders. Autism, 12, Pelphrey, K.A., & Carter, E.J. (2008). Brain mechanisms for social perception: Lessons from Autism and Typical Development. Annals of the New York Academy of Sciences, 1145, Pinto, J., & Shiffrar, M. (2009). The visual perception of human and animal motion in point-light displays. Social Neuroscience, 4, Saygin, A.P. (2007). Superior temporal and premotor brain areas are necessary for biological motion perception. Brain, 130, Schultz, R.T. (2005). Developmental deficits in social perception in autism: The role of the amygdala and fusiform face area. International Journal of Developmental Neuroscience, 23, Simion, F., Regolin, L., & Bulf, H. (2008). A predisposition for biological motion in the newborn baby. Proceedings of the National Academy of Science, 105, Simmons, D.R., Robsertson, A.E., McKay, L.S., Toal, E., McAleer, P., & Pollick, F. (2009). Vision in autism spectrum disorders. Vision Research, 49, Zilbovicius, M., Meresse, I., Chabane, N., Brunelle, F., Samson, Y., & Boddaert, N. (2006). Autism, the superior temporal sulcus and social perception. Trends in Neurosciences, 29, INSAR Kaiser et al./asd and complex motion perception 5