Brain signatures indexing variation in internal processing during perceptual decision-making.

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Brain activity exhibits significant trial-to-trial variability, impacting cognitive task performance.
• Existing methods for studying brain activity variability often rely on experimental manipulations or spontaneous changes, potentially missing endogenous factors.
• Understanding these internal variations is crucial for linking brain dynamics to cognitive processes.

Approach:

• Utilized a data-driven modularity-maximization clustering method to identify spatial-temporal electroencephalography (EEG) patterns across individual trials.
• Applied the method to data from 25 subjects performing a motion direction discrimination task with varying motion coherence levels.
• Related identified EEG activity clusters (subtypes) to behavioral performance metrics, including response times.

Key Points:

• Two distinct spatial-temporal EEG activity subtypes were identified, differing in their patterns.
• Subtype 1, although more frequent at lower motion coherence, was associated with faster response times.
• Computational modeling indicated Subtype 1 reflects a lower decision threshold.

Conclusions:

• The study reveals endogenous, trial-to-trial variability in decision processes not typically observable.
• The developed clustering method effectively identifies distinct brain states relevant to cognition and behavior.
• Findings offer new insights into the neural mechanisms underlying performance variations in cognitive tasks.