Functional near infrared spectroscopy in the multimodal assessment of working memory impairments following traumatic brain injury

A frequent consequence of traumatic brain injury (TBI) is cognitive impairment, which results in significant disruption of an individual's everyday living. To date, most clinical rehabilitation interventions still rely on behavioral observation, with little or no quantitative information about physiological changes produced at the brain level.Functional brain imaging modalities have been extensively used in the study of cognitive impairments following TBI. However, the applications of these technologies to rehabilitation have been limited. This is due in part to the expensive or invasive nature of these modalities or because they rely on experimental tasks that are not ecologically valid in reference to real-world behaviors. Additionally, studies of cognitive impairments have most commonly depended on a single imaging modality. However, different modalities glean differ aspects of the brain activity and each may offer distinct and often complementary strengths. Therefore, combining multiple technologies could offer improved understanding of brain-behavior relationship under pathological conditions.The objective of this study is to apply, for the first time, functional near infrared spectroscopy (fNIRS), and its integration with electroencephalography (EEG), to the assessment of working memory impairments following TBI. fNIRS provides a localized measure of prefrontal hemodynamic activation, which is susceptible to TBI, and it does so in a noninvasive, affordable and wearable way, thus partially overcoming the limitations of other modalities. EEG offers a cost-effective and simple measure of brain electrical activity and it has been employed in a few studies on traumatic brain injury, showing abnormal patterns of neural activity. The combination of two modalities therefore offers information about the spatial location of the recorded activity and it takes advantage the good temporal resolution of EEG.Participants included six TBI subjects and eleven healthy controls. Standard neuropsychological tests probing attention and working memory were administered. Brain activation measurements were collected during a visual n-back task, designed to incrementally vary the working memory load and often used in neuroimaging studies.Overall, the results from the research presented in this thesis provide first evidence of the ability of fNIRS to reveal differences between TBI and healthy subjects in working memory tasks, suggesting a dysfunction in the matching between cognitive demands and cortical resources in TBI subjects. Moreover, this study has demonstrated that fNIRS measures can distinguish between the two groups and these data have been investigated relative to the results obtained by EEG alone or by the combination of fNIRS and EEG. Each of the two modalities revealed unique strengths that can contribute to the classification between the two groups. Therefore, successful combination of fNIRS and EEG for this application takes advantage of the complementary strengths of the individual modalities in order to improve the classification between normal and TBI cases.