Evaluating the Bayesian causal inference model of intentional binding through computational modeling.

Area of Science:

• Cognitive Neuroscience
• Computational Psychology
• Human-Computer Interaction

Background:

• Intentional binding, a key indicator of agency, describes the subjective time compression between an action and its sensory consequence.
• The precise neural and computational mechanisms underlying intentional binding remain poorly understood.
• Bayesian causal inference (BCI) presents a promising theoretical framework, yet requires robust empirical validation.

Purpose of the Study:

• To computationally model and quantitatively evaluate potential mechanisms of intentional binding.
• To test the efficacy of Bayesian causal inference (BCI) models in explaining observed time estimation data.
• To investigate the algorithmic basis of BCI in inferring event timing and causality.

Main Methods:

• Development and implementation of diverse computational models for intentional binding.
• Fitting computational models to individual participant data on time estimation.
• Quantitative comparison of model performance, including BCI and maximum likelihood estimation (MLE).

Main Results:

• BCI models significantly outperformed traditional models in explaining time estimation.
• Identified causal belief and temporal prediction as key parameters contributing to intentional binding.
• Estimated parameters indicated time compression arises from expecting immediate action-consequence relationships.

Conclusions:

• Computational modeling, particularly BCI, offers a powerful tool for dissecting the mechanisms of intentional binding.
• Findings support the role of action-outcome expectation in subjective time perception.
• Probability-matching may underlie the heuristic reconstruction of event timing under causal uncertainty.