Brain network dynamics predict moments of surprise across contexts

Area of Science:

• Neuroscience
• Cognitive Psychology
• Computational Neuroscience

Background:

• Surprise is a fundamental human emotion triggered by unexpected events.
• The neural underpinnings of surprise across different contexts remain incompletely understood.
• Investigating shared neural mechanisms could reveal a unified theory of surprise.

Purpose of the Study:

• To determine if surprise shares common neural bases across distinct contexts.
• To identify and validate a predictive brain network model for surprise.
• To test the generalizability of this model in varied real-world and experimental settings.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to measure brain activity.
• A novel predictive model, the surprise edge-fluctuation-based predictive model (EFPM), was developed.
• The EFPM's predictive power for surprise was tested in three distinct groups experiencing different expectation violations.

Main Results:

• The EFPM successfully predicted surprise in an adaptive learning task.
• The same EFPM generalized to predict surprise in individuals watching suspenseful sports and videos with violated psychological expectations.
• The surprise EFPM demonstrated superior predictive accuracy compared to other models.

Conclusions:

• Shared neurocognitive processes underlie the experience of surprise across diverse contexts.
• Distinct subjective experiences of surprise can be represented within a common neural dynamic space.
• This research provides evidence for a fundamental, context-invariant neural basis of surprise.