David Soto

This study investigates how the brain represents ensemble mean and variance across different visual features using fMRI and multivariate pattern analysis, finding distinct neural pathways for mean and variance processing.

This paper introduces a co-adaptive decoding method for fMRI neurofeedback that improves training performance through real-time decoder updates, supported by simulations and real data.

The paper re-analyzes fMRI data to compare neural representations of conscious and unconscious perceptual content using convolutional neural network models, finding that these models predict brain responses across different regions and participants.

This study examines how brain representations of semantic knowledge change with healthy aging, finding increased generalization and specialization that may optimize neural resources.

This paper uses fMRI-DecNef to show unconscious modification of conceptual representations, affecting visual search behavior without conscious awareness of the target content.

PyDecNef provides an open-source framework for real-time fMRI analysis, enabling accessible and customizable decoded neurofeedback experiments.

The paper uses encoding-based representational similarity analysis to compare brain activity patterns for conscious and unconscious visual contents using fMRI data.

This study identifies neural representations of unconscious content in human brains and simulates it using deep networks, showing overlapping representations between conscious and unconscious processing.

This study uses fMRI and SVM to show that social concepts are represented in a distributed network involving semantic and social regions, with the lateral temporal lobe encoding affect and likableness best.

This study shows that deeper mental simulation during word processing leads to more generalized and connected brain representations of meaning compared to shallow processing.

The paper shows that prospective decisions about future behavior share a common representational structure, as demonstrated by machine learning predictions across different tasks.

This study shows human V1 can learn multiple concurrent complex visual sequences, including goal-irrelevant ones, using fMRI.

This study investigates how the human brain encodes memories of task performance during introspection, finding similar brain activity patterns during on-task and introspection states with distinct connectivity profiles.