Research seminar: Gláucia Laís Salomão, Stockholm University

Title: Decoding Emotional Inner Voices Using Multivariate Pattern Analysis (MVPA): An fMRI Pilot Study.

Seminar

Date:

Thursday 23 October 2025

Time:

15.00 – 17.00

Location:

C307, Södra huset

 

Abstract:

The motor imagery of speech – often referred to as covert or inner speech – offers a unique window into the neural mechanisms linking speech perception and production. Yet its emotional dimension, reflected in expressive modulations of the voice during speech, remains largely unexplored, even though emotional expression deeply permeates spoken language. 

Research on speech motor imagery using functional magnetic resonance imaging (fMRI) has traditionally relied on General Linear Models (GLMs) to identify brain regions engaged during the imagined articulation of words or sentences. In this pilot study, I applied Multivariate Pattern Analysis (MVPA) to examine patterns of neural activity underlying the motor imagery of vocalizations that either conveyed or suppressed emotional content. The aim was to determine whether neural activity associated with vocal motor imagery is modulated by emotional expression.

The results suggest that emotional expression modulates neural representations underlying vocal motor imagery, although classification performance varied considerably across participants. Moreover, classification algorithms performed better at distinguishing between imagined voices expressing positive and negative emotions – and even between two negative emotions – than between the two positive emotions. 

This study is being replicated with a larger dataset as part of the research project ”Listening to Others’ Emotions”, and is expected to contribute to models of the neural mechanisms underlying empathy in response to emotional voices. 

On Gláucia Laís Salomão