Consistent Gradient of Performance and Decoding of Stimulus Type and Valence From Local and Network Activity

Area of Science:

• Neuroscience
• Cognitive Science
• Psychology

Background:

• Individual differences in behavior and neural activity are crucial for understanding cognitive processes.
• Emotion recognition involves complex facial and lexical processing, with significant inter-individual variability.

Purpose of the Study:

• To identify intracranial neural markers differentiating performance in a dual-valence emotion recognition task.
• To explore electrophysiological differences in resting-state and task-related brain activity.
• To assess the utility of network-based measures for capturing individual performance profiles.

Main Methods:

• Preliminary study to replicate evoked responses.
• Analysis of time series data and network connectivity.
• Multivariate pattern analysis and machine learning for neural decoding.
• Examination of resting-state versus task-related activity.

Main Results:

• Time series classification mirrored behavioral gradients for stimulus type, but not valence.
• Network-based measures successfully reflected individual hierarchical profiles for both stimulus type and valence.
• Network measures proved sensitive to distributed processes like emotional valence discrimination.

Conclusions:

• Brain network measures offer a sensitive marker for inter-individual performance variability in emotion recognition.
• Combining network analysis with classification methods can provide insights into single-subject studies.
• This approach holds potential for extrapolation to other neuroscientific techniques and research areas.