Synthesizing Images on Perceptual Boundaries of ANNs for Uncovering and Manipulating Human Perceptual Variability

TL;DR

This paper introduces BAM, a computational framework that generates stimuli along neural network decision boundaries to systematically study and manipulate human perceptual variability, validated through large-scale behavioral experiments.

Contribution

The paper presents BAM, a novel method combining neural network boundary sampling with behavioral data to analyze and influence individual perceptual differences.

Findings

Generated stimuli effectively induce perceptual variability.

Personalized models can predict individual perceptual decisions.

Manipulation of perceptual decisions is achievable through adversarial stimuli.

Abstract

Human decision-making in cognitive tasks and daily life exhibits considerable variability, shaped by factors such as task difficulty, individual preferences, and personal experiences. Understanding this variability across individuals is essential for uncovering the perceptual and decision-making mechanisms that humans rely on when faced with uncertainty and ambiguity. We present a computational framework BAM (Boundary Alignment & Manipulation framework) that combines perceptual boundary sampling in ANNs and human behavioral experiments to systematically investigate this phenomenon. Our perceptual boundary sampling algorithm generates stimuli along ANN decision boundaries that intrinsically induce significant perceptual variability. The efficacy of these stimuli is empirically validated through large-scale behavioral experiments involving 246 participants across 116,715 trials, culminating in the variMNIST dataset containing 19,943 systematically annotated images. Through personalized model alignment and adversarial generation, we establish a reliable method for simultaneously predicting and manipulating the divergent perceptual decisions of pairs of participants. This work bridges the gap between computational models and human individual difference research, providing new tools for personalized perception analysis.