AUTHOR=Johnson Rachel , Matthews Jennifer , Diawara Norou , Carroll Rachel 

TITLE=Statistical Analysis of fNIRS Data: Consideration of Spatial Varying Coefficient Model of Prefrontal Cortex Activity Changes During Speech Motor Learning in Apraxia of Speech

JOURNAL=Frontiers in Applied Mathematics and Statistics

VOLUME=Volume 6 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/applied-mathematics-and-statistics/articles/10.3389/fams.2020.00032

DOI=10.3389/fams.2020.00032

ISSN=2297-4687

ABSTRACT=One of the key cognitive functions for new learning is cognitive flexibility. In a healthy system, the role of the prefrontal cortex (PFC) during speech is of growing interest. For individuals who have an impairment in the programing and planning of speech (apraxia of speech) secondary to a stroke or traumatic brain injury, the once hardwired activity is now a goal directed rule-based act. The literature falls short in PFC involvement for rehabilitation of speech following a brain injury. This repeated measures single case-design study involves two individuals with chronic acquired apraxia of speech. Hemodynamic changes in PFC activity during speech motor learning are measured using functional near-infrared spectroscopy. Two models - a generalized linear model (GLM) and a spatial varying coefficient model - are used to distinguish PFC activity differences corresponding the task, stage, and time of intervention. The significant differences in the pattern of activity for the regions of interest in the PFC are associated with structure of practice, feedback schedule, and the time of intervention. This pilot study provides evidence for the PFC role in speech motor learning and that neural reorganization can continue several years post-stroke. The implications include considerations for statistical methods used for fNIRS analysis with the potential to build tailored interventions to facilitate neural network reorganization for speech motor learning following an acquired brain injury.