Adaptive Computational
Cognition Laboratory

Human perception cannot capture every detail of the world, so the brain must decide what information to keep and what to discard. We use rate–distortion theory, the branch of information theory that deals with lossy compression, to predict how an optimal but capacity-limited perceptual system should behave. This approach explains how people identify stimuli and remember visual details. It also shows that Shepard's "universal law of generalization" follows from efficient coding, in biological and artificial systems alike.

See: (Sims, 2016) • (Sims, 2018)

People rarely face exactly the same situation twice, so learning is only useful if it carries over to new situations. We study how people build simplified internal representations that keep the features that predict reward and discard the rest. Adding this principle of efficient coding to reinforcement learning produces models that generalize as well as people do, where standard models fall short. This work links human learning to open problems in machine learning and artificial intelligence.

See: (Fang & Sims, 2025)

Skilled motor behavior, from reaching for a cup to competing at an elite level, depends on tightly coordinated perception and movement. We use mobile eye tracking and motion capture to measure how people move in fine-grained detail. We then develop models of motor control that account for the limits of information processing. We also study extreme expertise in complex domains such as competitive esports, and have developed new ways to test models of expert performance when there are only a few experts to study.

See: (Phillips & Sims, 2025) • (Lerch & Sims, 2016)

We study how limits on memory, attention, and other cognitive systems shape the decisions people make. Our models show that some well-known biases in human choice can arise from forming efficient representations of what we see. We also examine how a person's current state, such as stress or anxiety, changes the interplay between visual memory and decision making.

See: (Malloy & Sims, 2025) • (Kaper et al., 2025)