AI-guided stimuli discovery and generation to optimize facial emotion perception studies in autism

2026-07-09

AI Summary

Traditional research on facial emotion perception has yielded inconsistent findings regarding differences between autistic and neurotypical individuals, hindering the identification of robust group distinctions. This study proposes a model-guided stimulus optimization framework: first, population-specific artificial neural networks predict individual emotion judgments from facial stimuli; then, generative adversarial networks (GANs) automatically discover or synthesize images that either maximize or minimize behavioral differences between groups. Moving beyond conventional fixed stimulus sets, this approach successfully identifies images that significantly amplify group differences in an independent sample, while synthesized images effectively reduce such disparities. The framework thus enables precise probing of perceptual mechanisms underlying neurodiversity in emotion processing.

Abstract

Understanding perceptual differences between autistic and neurotypical adults requires behavioral assays that are sensitive, reliable, and mechanistically informative. Facial emotion perception is a useful test case because group differences have been reported, but findings vary across studies. Here we show that this variability may reflect image-level sparsity: autistic-neurotypical differences in emotion judgments were concentrated in a small subset of diagnostic facial expressions rather than spread uniformly across stimuli. We trained population-specific artificial neural network models to predict image-level judgments for autistic and neurotypical participants, then used these models to select novel faces predicted to maximize group separation. In an independent cohort, model-selected images produced larger behavioral differences than matched random images. We then used the same models with a generative adversarial network to transform diagnostic images toward greater predicted group agreement. In phenotype-matched validation, synthesized images reduced behavioral separation relative to their matched originals. These results establish a model-guided framework for discovering and transforming stimuli that reveal population-specific perceptual differences. More broadly, they show how behavioral phenotyping can move beyond averaging across fixed stimulus sets toward optimized assays that identify the conditions under which neurodivergent perception diverges or converges.

Problem

• facial emotion perception
• autism
• stimuli variability
• perceptual differences
• behavioral assays

Innovation

• AI-guided stimuli selection
• population-specific neural networks
• generative adversarial networks
• facial emotion perception
• behavioral phenotyping