Wilson S. Geisler

This study measures contrast-encoding in macaque V1 using Gaussian stimuli across a wide contrast range, finding that neural responses continue to increase beyond 100% Weber contrast.

The paper trains texture discrimination and segmentation models using spatial proximity as a proxy for same-different ground truth, achieving accurate results without feedback.

This paper introduces a hierarchical Bayesian observer model for texture segmentation that mimics human visual processing and outperforms deep networks on specific tasks.

The paper develops a computational framework for target identification under uncertainty and evaluates ideal and heuristic observers against human performance.

This study investigates how similarity masking affects V1 responses in macaques, finding initial enhancement followed by suppression and proposing a model to explain these dynamics.

The paper tests if low-power optogenetic stimulation in primate V1 can mimic visual masking effects, finding it reduces detection sensitivity and shows spatial selectivity.

The paper compares human and macaque detection performance in natural backgrounds with neural responses in V1, finding quantitative agreement and support for population decoding.