Luca Dstasio
Research Mentor: Greg Gage
Mentor Department: Backyard Brains, Engineering
Author(s): Luca Dstasio, Greg Gage
Session: Session 2 (10:00 AM – 10:50 AM)
Presentation Type: Poster 45
Abstract
According to Attention Schema Theory (AST), the human brain models the attention of others not as an abstract mental state, but as a physical substance, akin to beams of energy projecting from the eyes. This study investigated whether this cognitive model recruits low-level sensory machinery, specifically the motion processing system. In an online psychophysical experiment (N=120), participants viewed static faces with averted gaze (the adaptor) followed by a random dot kinematogram (RDK) moving in a congruent or incongruent direction. We observed a significant motion aftereffect: reaction times were significantly slower when the physical motion matched the direction of the static gaze (p<0.005). This negative aftereffect, a hallmark of neural adaptation, suggests that the brain processes static gaze using the same neural population as physical motion. These findings provide empirical support for the extramission model of attention proposed by AST. Furthermore, they offer a biological constraint for NeuroAI: just as the human brain recycles motion circuits to model latent social states efficiently, generative AI architectures might achieve greater social alignment by grounding abstract concepts in low-level sensory embeddings.
