A Bayesian method for inference of effective connectivity in brain networks for detecting the Mozart effect

TL;DR

This paper introduces a Bayesian topology identification method for inferring effective brain connectivity, demonstrating its superiority over Granger-causality analysis in simulations and revealing connectivity changes after listening to Mozart, especially in a subgroup of subjects.

Contribution

The paper develops a Bayesian approach for effective connectivity inference and extends it to compare groups, providing a new tool for analyzing brain network dynamics in response to stimuli.

Findings

Bayesian method outperforms Granger-causality in simulations, especially with short data.

Detected significant connectivity changes after Mozart exposure in specific brain networks.

Longer listening duration correlates with more pronounced connectivity changes in a subgroup.

Abstract

Several studies claim that listening to Mozart music affects cognition and can be used to treat neurological conditions like epilepsy. Research into this Mozart effect has not addressed how dynamic interactions between brain networks, i.e. effective connectivity, are affected. The Granger-causality analysis is often used to infer effective connectivity. First, we investigate if a new method, Bayesian topology identification, can be used as an alternative. Both methods are evaluated on simulation data, where the Bayesian method outperforms the Granger-causality analysis in the inference of connectivity graphs of dynamic networks, especially for short data lengths. In the second part, the Bayesian method is extended to enable the inference of changes in effective connectivity between groups of subjects. Next, we apply both methods to fMRI scans of 16 healthy subjects, who were scanned before and after exposure to Mozart's sonata K448 at least 2 hours a day for 7 days. Here, we investigate if the effective connectivity of the subjects significantly changed after listening to Mozart music. The Bayesian method detected changes in effective connectivity between networks related to cognitive processing and control: First, in the connection from the central executive to the superior sensori-motor network. Second, in the connection from the posterior default mode to the fronto-parietal right network. Finally, in the connection from the anterior default mode to the dorsal attention network, but only in a subgroup of subjects with a longer listening duration. Only in this last connection an effect was found by the Granger-causality analysis.