Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster B in Poster Session B: Tuesday, August 4, 2:00 – 3:45 pm, Kimmel Center, Shorin & Rosenthal Rooms

Outcome-driven decisions dominate causal inference in autism across species

Aviel Sulem¹, Benedict Praegel¹, Jaroslav Slobodskoy-Plusnin¹, Yonathan Loewenstein¹, Adi Mizrahi¹, Merav Ahissar¹; ¹Hebrew University of Jerusalem

Presenter: Aviel Sulem

Autism is typically characterized by social communication differences, often interpreted through the lens of Theory of Mind—the ability to infer others’ beliefs and intentions. However, it remains unclear whether differences observed in social contexts reflect domain-specific impairments or broader differences in how individuals infer causes from observations. Here we test whether autism is associated with a different computational strategy for guiding perceptual decisions. We contrast two possibilities: relying directly on recent observable outcomes (reward or punishment), or inferring the causes of success and failure from the structure of stimuli and tasks. Neurotypical and autistic adults performed a two-tone pitch discrimination task in which feedback from the previous trial provided no information about the correct response on the current trial. Neurotypicals showed adaptive updating of perceptual biases based on the difficulty of recent stimuli, consistent with stimulus-based inference. In contrast, autistic participants relied more strongly on previous outcomes, tending to repeat rewarded responses. Strikingly, mice carrying the autism-associated 16p11.2 deletion showed the same outcome-driven bias in an auditory Go/NoGo discrimination task. These findings suggest that enhanced reliance on observable outcomes, rather than inference about stimulus structure, may represent a conserved computational feature of autism in humans and mice.

Topic Area: Development, Individual Differences & Clinical Populations