Working Memory in Those With a History of Concussion
Loading...
Authors
Shires, Jorja
Issue Date
2024
Type
Dissertation
Language
Keywords
Concussion , EEG , Modeling , MTBI , Neurostimulation , Working Memory
Alternative Title
Abstract
Mild traumatic brain injury (mTBI), a common head injury, is expected to heal within weeks. However, deficits in visual working memory (WM) persist in undergraduates who report a history of mTBI (hmTBI). This pattern is observed years after injury. MTBI is a heterogenous group in several areas including etiology, treatment, age at injury, sex, and even WM performance. In other words, not all people with hmTBI are impaired in VWM, but a subset are. We proposed several related research questions: 1) are persistent WM deficits attributable to distributions of frontal lobe function, 2) is there a common neural signature associated with VWM impairment in hmTBI, and 3) can neuromodulation improve WM function in hmTBI? This dissertation addresses each of these related questions. Aim 1 extended our previous WM findings using a new task relying heavily on frontal lobe regions. We predicted greater deficits on this task compared to our previous WM tasks. Aim 2 investigated whether the subgroup showing WM deficits share an underlying neural signature. Electroencephalogram (EEG) data was used to train a classifier to determine whether a neural signature accurately predicts group membership (control, hmTBI low WM, hmTBI high WM). In Aim 3, we applied transcranial direct current stimulation (tDCS) with the goal of improving WM. The overall goal was to identify who has lasting traces of hmTBI, why they persist, and how to remediate VWM in those with residual deficits. Our results found no behavioral or neuronal difference between mTBI and controls. There was no group effect regardless of neuromodulation, task demands, or cognitive load. EEG data was unable to discriminate between hmTBI and controls. However, we did find a significant difference in VWM performance between the dataset collected for this dissertation and a dataset collected pre-COVID. The hmTBI population present at the University of Nevada, Reno may have changed after the pandemic. Our other major finding was in Aim 1. Performance was significantly higher on ACTS-WM than ACTS despite the increase cognitive load. This finding suggests abstract concepts, such as instructions, and concrete stimuli may be held in WM in separate stores. While our findings were mostly unexpected, they have interesting implications for WM theory and research in special populations.