Robert C. Wilson

Individual differences reveal shared computational structure in perceptual decision-making

The paper uses population-level individual differences to discover shared computational components in perceptual decision-making, revealing a six-dimensional structure across datasets.

Modeling the journey as well as the destination: a control theory account of rotational navigation

This paper develops a closed-loop model of rotational navigation that explains how heading estimates are transformed into motor commands, showing that prediction errors drive corrective movements.

Illusory path configurations reveal age-related differences in egocentric pointing variability

This study finds that older adults show different egocentric pointing variability in virtual reality tasks involving illusory path configurations compared to young adults.

The dynamics of explore-exploit decisions suggest a threshold mechanism for reduced random exploration in older adults

This paper uses a drift diffusion model to show that older adults have higher decision thresholds, which may compensate for lower signal-to-noise ratios, leading to reduced random exploration.

Human Strategy Adaptation in Reinforcement Learning Resembles Policy Gradient Ascent

This paper introduces DynamicRL, a method to estimate how humans adapt their reinforcement learning strategies over time, showing parameter updates resemble policy gradient ascent.

Semantic Influences on Object Detection: Drift Diffusion Modeling Provides Insights Regarding Mechanism

The paper uses drift diffusion modeling to analyze how semantic activation affects object detection, finding that valid labels increase signal-to-noise ratio while invalid labels require more evidence to resolve conflicts.

No Effect of Chronic or Acute Pain on Working Memory in the Sternberg Task

This study tests whether pain affects working memory in the Sternberg Task and finds no effect on response time, accuracy, or their relationship with set size for chronic or acute pain.

Deep Learning Improves Parameter Estimation in Reinforcement Learning Models

This study compares parameter estimation methods in reinforcement learning models, finding that deep learning outperforms traditional optimization in reliability and generalizability.

Frontal-midline theta and posterior alpha oscillations index early processing of spatial representations during active navigation

The paper investigates whether spatial navigation involves parallel or serial processing of self-motion and landmark cues using EEG oscillations during virtual navigation.

Combination and competition between path integration and landmark navigation in the estimation of heading direction

This study investigates how humans integrate visual and body-based cues for heading estimation, finding a hybrid strategy where cues are combined for small mismatches and compete for large ones.

A divisive model of evidence accumulation explains uneven weighting of evidence over time

The paper introduces a divisive normalization model for evidence accumulation, showing it matches human perceptual decision-making as well as the Drift Diffusion Model.

Temporal discounting correlates with directed exploration but not with random exploration

This study finds that temporal discounting correlates with directed exploration but not with random exploration in decision-making tasks.

Regulation of evidence accumulation by pupil-linked arousal processes

This study investigates suboptimalities in perceptual decision making and their relationship with pupil-linked arousal processes, finding that different suboptimalities correlate with distinct aspects of pupil response.

A causal role for right frontopolar cortex in directed, but not random, exploration

This study shows that right frontopolar cortex inhibition selectively reduces directed exploration, indicating distinct neural systems for directed and random exploration strategies.

A causal role for right frontopolar cortex in directed, but not random, exploration.

This study shows that right frontopolar cortex causally supports directed exploration but not random exploration in the explore-exploit dilemma.