Abstract
Humans show position-specific adaptation to simple stimuli (e.g., oriented bars), but show position-invariance to complex stimuli (i.e., faces). Such distinguishing characteristics of adaptation can potentially reveal the neural mechanism underlying visual processing. We investigated position-specific and position-invariant adaptation to biological motion. Subjects were adapted to walker or runner point-light stimuli for 6 seconds and then were given a 1-s test stimulus (a morphed action intermediate between walker and runner). Subjects reported the perceived action of the test. An Autism Spectrum Quotient (AQ score) was acquired, which measures the degree to which an adult with normal intelligence has traits associated with Autism Spectrum Disorder (ASD). We measured the percentage of trials in which a runner was reported. We computed the difference between conditions with walking and running adaptation to measure the "adaptation effect". There were significant adaptation effects when adapting and testing locations were identical (i.e., position-specific adaptation). When adapting and testing locations differed, i.e. position-invariant adaptation, we found an overall non-significant effect, which was however significantly negatively correlated with the AQ score. Splitting the subjects along the median AQ score, we found that the low-AQ group (with fewer autistic traits) showed significant position-invariant adaptation, while the high-AQ group showed no position-invariant adaptation. Conversely, adaptation effects at presumably lower visual levels (acquired by subtracting the position-specific effect from the position-invariant effect) were significantly positively correlated with AQ score, suggesting that subjects with more autistic traits had increased adaptation at low visual processing levels. Our data show both position-specific and position-invariant adaptation to biological motion. Whether adaptation effects transfer to other locations is correlated with the AQ score. This result is consistent with the hypothesis that people with ASD tend to focus more on low-level stimulus characteristics, and have impaired processing of high-level (position-invariant) aspects of biological motion.
Meeting abstract presented at VSS 2012