Understanding music perception through neural signal analysis

How does the brain make sense of music? This project investigates the neural mechanisms underlying auditory perception by combining acoustic signal processing with analysis of brain activity recordings. To understand how the brain processes music, we will apply a range of signal processing and feature extraction methods to describe and annotate music. These representations will then be used within encoding and/or decoding models to analyse iEEG data, revealing how different sound properties are processed across the brain.

The student joining this project should be comfortable programming independently in Python and have some familiarity with basic signal processing techniques and/or time-series analysis. The project will develop practical skills in neural data analysis, acoustic signal processing, and encoding/decoding models.

Comparing methods for estimating intrinsic neural timescales in EEG/iEEG data

Intrinsic neural timescales can be understood as the characteristic durations over which neural activity remains correlated with itself. Timescales have emerged as a key organising principle of cortical computation, yet the methods used to estimate them vary in their assumptions and sensitivity. In this project, we will implement and systematically compare several timescale estimation methods (e.g. autocorrelation function, spectral methods), assessing their reliability, validity, and sensitivity to methodological choices.

The student working on this project should be motivated to program independently in Python and to analyse EEG and/or iEEG data. The project will develop practical skills in neural data analysis, reliability metrics, and model comparison.