Conjunctive representations that integrate stimuli, responses, and rules are critical for action selection

Area of Science:

• Cognitive Neuroscience
• Neuroscience
• Computational Neuroscience

Background:

• Humans flexibly use abstract rules to guide actions in specific situations.
• The exact neural mechanisms linking rules to sensory and motor requirements are not fully understood.
• Event files, or conjunctive representations, are theorized to integrate rules with goal-relevant features for action selection.

Purpose of the Study:

• To investigate the temporal dynamics of conjunctive representations during rule-based action selection in humans.
• To determine if conjunctive representations are functionally distinct from their constituent features.
• To establish a method for tracking these representations with high temporal resolution.

Main Methods:

• Applied time-resolved representational similarity analysis to electroencephalography (EEG) signals.
• Participants performed a cued, rule-based action selection task across two experiments.
• Analyzed spectral-temporal profiles of EEG data to decode representational content.

Main Results:

• Conjunctive representations were found to be active throughout the action selection period.
• The strength of these conjunctions predicted trial-by-trial response time variability.
• Evidence suggests conjunctive representations are functionally separable from representations of individual features.
• Results linked conjunction strength to the partial-overlap priming pattern, a behavioral marker.

Conclusions:

• Provides direct evidence for the role of conjunctive representations as critical precursors to action selection in humans.
• Demonstrates a novel method for tracking neural representations with high temporal resolution.
• Highlights the importance of integrated representations in flexible, rule-governed behavior.