Sofia Soares

Inferring brain-wide interactions using data-constrained recurrent neural network models

This paper introduces CURBD, a method using data-constrained recurrent neural networks to infer brain-wide interactions from neural recordings across species.

Cognitive experience alters cortical involvement in navigation decisions

The paper shows that mice with different cognitive experiences use distinct cortical areas and neural activity patterns for navigation decisions, highlighting past learning as a key factor in determining cortical necessity for decision-making.

Striatal circuits support broadly opponent aspects of action suppression and production

The paper shows that striatal direct and indirect pathways exhibit functional opponency during action suppression and production in mice.

Dopamine responses reveal efficient coding of cognitive variables

The paper investigates how reward expectations are influenced by internal representations in rodents, finding that efficient, compressed representations best match both behavior and dopaminergic activity.

High-yield methods for accurate two-alternative visual psychophysics in head-fixed mice

This paper introduces high-yield methods for two-alternative visual psychophysics in head-fixed mice, showing rapid learning and accurate psychometric curves.

Striatal dynamics explain duration judgments

This study shows that striatal neuron dynamics predict and are necessary for duration judgments in rats performing a psychophysical task.

Bonsai: An event-based framework for processing and controlling data streams

Bonsai provides a visual programming framework for managing parallel data streams in neuroscience experiments, enabling flexible and rapid prototyping of integrated experimental designs.