Reliability of single-subject neural activation patterns in speech production tasks

Abstract

Traditional group fMRI (functional magnetic resonance imaging) analyses are not designed to detect individual differences that may be crucial to better understanding speech disorders. Single-subject research could therefore provide a richer characterization of the neural substrates of speech production in development and disease. Before this line of research can be tackled, however, it is necessary to evaluate whether healthy individuals exhibit reproducible brain activation across multiple sessions during speech production tasks. In the present study, we evaluated the reliability and discriminability of cortical functional magnetic resonance imaging data from twenty neurotypical subjects who participated in two experiments involving reading aloud mono- or bisyllabic speech stimuli. Using traditional methods like the Dice and intraclass correlation coefficients, we found that most individuals displayed moderate to high reliability, with exceptions likely due to increased head motion in the scanner. Further, this level of reliability for speech production was not directly correlated with reliable patterns in the underlying average blood oxygenation level dependent signal across the brain. Finally, we found that a novel machine-learning subject classifier could identify these individuals by their speech activation patterns with 97% accuracy from among a dataset of seventy-five subjects. These results suggest that single-subject speech research would yield valid results and that investigations into the reliability of speech activation in people with speech disorders are warranted.