* | Carlsbergfondet
Til bevillingsoversigt

The neuronal basis of musical imagination

Carlsbergfondets internationaliseringsstipendier


Imagine a song that you know by heart. Without effort, you could almost hear it and sing it in your mind. You may even imagine it with twisted lyrics and, if you are musically talented, you could mentally sing it backwards or add new melodic lines to it. This is an example of how rich our musical imagination can be. Yet, we know very little about how the brain represents, holds in memory and manipulates these musical "thoughts". To investigate this issue, in collaboration with the Knight laboratory at the University of California at Berkeley, I will record neuronal activity directly from inside the brain while human participants perform a simple working memory task involving sequences of musical sounds.


The knowledge derived from this project will greatly improve our understanding of auditory perception, memory and the neural processing of music. Furthermore, the patterns of activity recorded directly from inside the brain, which have a high temporal and spatial resolution, could be used to inform research on non-invasive methods such as EEG, potentially leading to significant technological developments.


I will record activity directly from inside the brain of patients with epilepsy who require invasive monitoring of neural activity with electrodes placed inside and on the surface of the brain. This technique is known as intracranial EEG (iEEG). In the experiment, participants are asked to perform a task where they mentally hold and manipulate sequences of musical sounds. To understand the neuronal basis of musical imagination, I will use machine learning techniques to decode imagined sounds from brain activity and will analyze neuronal oscillations at different frequencies and in different brain areas.


Understanding how the brain represents, holds in memory and manipulates sequences of sounds can be the basis for future clinical applications. For example, decoding of auditory imagery from brain activity could help patients with speech impairments to partially recover the ability to communicate. Furthermore, applying the knowledge derived from iEEG research to more accessible non-invasive methods could lead to the improvement of brain-computer interfaces that can be readily used in clinical settings. Consequently, after my project finishes, I wish to further develop this line of research in Denmark, combining iEEG and non-invasive methods, and benefiting from an outstanding international network of collaborators.