Do people with more autistic traits see body movement differently?
Among ordinary students, those with more autistic traits were less automatically drawn to whole-body movement and leaned more on local joint motion, which let them match others' overall performance.
Source
Impaired global, and compensatory local, biological motion processing in people with high levels of autistic traits
Study at a glance
- Design
- Human experiment — Two within-subject psychophysics experiments (distractor interference; location-specific vs location-invariant adaptation) with AQ score as a continuous individual-difference predictor.
- N
- 25 undergraduates in Experiment 1 and 30 in Experiment 2 (10 took part in both); no single combined N is reported.
- Population
- UCLA psychology undergraduates, mostly female, not screened for an autism diagnosis
- Outcome
- Central-task accuracy with intact vs scrambled peripheral point-light distractors (Exp 1); local and global walker/runner adaptation strength (Exp 2), each related to Autism-Spectrum Quotient score
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
Students completed the Autism-Spectrum Quotient (AQ) and then two vision tasks built from point-light figures. In Experiment 1 (25 students) they counted target crosses at the centre of the screen while walkers or boxers moved in the periphery, either intact or spatially scrambled so that only the whole-body form differed. In Experiment 2 (30 students) they adapted to a walker or runner and judged an ambiguous walker/runner morph shown at the same or a different screen location, separating location-specific (local) from location-invariant (global) adaptation.
What they found
In Experiment 1, students with low AQ scores were distracted by intact figures (poorer counting than with scrambled figures), whereas high-AQ students were not distracted at all; central-task accuracy with an intact walker rose with AQ (r = 0.43). In Experiment 2, global adaptation fell as AQ rose (r = -0.38) while local adaptation rose (r = 0.45), and the total of local plus global adaptation did not differ between high- and low-AQ groups, suggesting local processing compensated for weaker global processing.
The limits
What it doesn't show
Participants were typical undergraduates, not people diagnosed with autism, so the results concern autistic traits and the authors caution against extrapolating to autism itself. Samples were small (25 and 30, mostly women), group comparisons used only nine people per group in Experiment 1, and several correlations were modest or non-significant (e.g., the boxer condition). The predictive-coding explanation is an interpretation the experiments did not directly test, and correlations cannot show that autistic traits cause the processing differences.
Key terms
- Biological motion
- The movement pattern of a living body, often shown as a few moving dots at the joints (a point-light display) that people readily see as a person walking or acting.
- Autism-Spectrum Quotient (AQ)
- A 50-item self-report questionnaire measuring how many autism-like traits a person has; higher scores mean more traits.
- Global vs local processing
- Global processing uses the overall configuration of a stimulus (e.g., body shape); local processing uses individual parts (e.g., one joint's trajectory).
- Location-invariant adaptation
- An aftereffect that transfers to a different place in the visual field, taken as a sign of high-level (global) processing, since low-level neurons are tied to specific locations.
- Predictive coding
- A theory that higher brain levels predict incoming input and feed predictions back so that lower levels only signal the unexpected part.
Flashcards
0 of 12 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
In Experiment 1, what happened to low-AQ participants' counting accuracy when intact point-light figures appeared in the periphery?
Common questions
Why scramble the point-light figures instead of just removing them?
Scrambling keeps every dot's local motion identical but destroys the body shape, so any difference between intact and scrambled conditions must come from global processing rather than from the local movement signals.
If high-AQ people had weaker global processing, why weren't they worse at the walker/runner task?
Their local adaptation was stronger, and the combined local-plus-global effect was the same in both groups, so extra reliance on local cues masked the global weakness. This is the paper's point that normal performance does not prove normal global processing.
Does this mean people with autism cannot perceive body movement?
No. The study tested non-diagnosed students with varying traits and found a shift in which processing route was used, not an inability to perceive motion; the authors warn that AQ and autism do not correspond perfectly.
More on Autism