Bernardino, I. | bioRxiv Science

bioRxiv2025

Neural correlates of emotional responses to self-selected music: evidence from multivariate pattern analysis

Music is a uniquely powerful stimulus for evoking complex and deeply felt emotions. While previous research has identified neural correlates of music-evoked emotional responses, less is known about how these felt emotions are represented in the brain, particularly when elicited by familiar, personally meaningful music. Here, we used a personalized fMRI paradigm in which participants (N = 20) each selected musical excerpts corresponding to the nine emotion categories defined by the Geneva Emotional Music Scale. These self-selected excerpts were presented during functional MRI scanning. We first examined the neural correlates of music-evoked emotion by comparing brain activity during music listening to that during exposure to white noise. The maps were consistent with previous research, highlighting clusters in sensory and limbic regions. We then used multivoxel pattern analysis to decode emotion categories from whole-brain activation patterns. The results revealed that music-evoked emotions could be reliably discriminated based on distributed neural activity, with consistent involvement of the superior temporal gyrus, supplementary motor area, amygdala, and cerebellum, among other auditory, motor, and interoceptive regions. These findings provide new insight into the neural encoding of musical emotions and highlight the value of personalized, music-based paradigms for research in auditory and affective neuroscience.

neuroscience↗

bioRxiv2024

Decoding Music-Evoked Valence and Arousal: Unraveling the Neural Correlates of Naturalistic Music Characteristics through fMRI

Music can convey basic emotions, such as joy and sadness, and more complex ones, such as tenderness or nostalgia. Its effects on emotion regulation and reward have attracted much attention in cognitive and affective neuroscience. Understanding the neural correlates of music-evoked emotions may guide the development of neurorehabilitation interventions based on music. Here, we used fMRI to examine the relationship between the classification of music excerpts regarding perceived valence and arousal and their neural correlates. Twenty participants were scanned while listening to 96 musical excerpts, which were classified beforehand into four categories as a function of valence (positive vs. negative) and arousal (high vs. low). Differences in valence and arousal modulated activity in cortical regions, most noticeably the music-specific subregions of the auditory cortex, but also in the thalamus and regions of the reward network such as the amygdala. Using multivoxel pattern analysis, we created a computational model able to decode the valence and arousal of the music excerpts significantly above chance. We further explored how a set of musical features relate to brain activity in valence-, arousal-, reward-, and auditory-related ROIs. The results emphasize the differential involvement of musical features in the brain, notably expressive features such as Vibrato and Tonal and Spectral dissonance in valence, arousal, and reward brain networks, while a broader set of features modulate sensory auditory networks. Using ecologically valid music stimuli, we contribute to the definition of the neural substrates of music listening and evoked emotions. Moreover, the definition of the musical features that modulate specific brain networks paves the way to developing novel music-based neurorehabilitation strategies.

neuroscience↗