Mental Imagery for the Detection of Awareness: Evaluating the Convergence of Functional Magnetic Resonance Imaging and Electroencephalographic Assessments

Abstract

<p>The accurate diagnosis of disorders of consciousness presents substantial difficulty because of the reliance on behaviour-based assessment tools. A patient may be covertly aware but unable to indicate their state due to physical impairments. Neuroimaging researchers have begun to seek alternate methods of assessment that rely on brain responses rather than behavioural ones. To this end, mental imagery has been employed as a voluntary cognitive activity that can be measured with fMRI or EEG to indicate awareness. In this dissertation I examine the advantages and limitations of these two imaging techniques and argue that EEG is more suitable for this patient population. I expand upon existing mental imagery research by exploring additional tasks that have not been applied to this problem, in order to address three previously unanswered questions that are central to the development of imagery-based diagnostic tools. First, do individuals differ on which imagery tasks produce the most reliable activation? Second, can the robustness of brain activation during imagery be predicted from familiarity with the imagined activity? Third, do fMRI and EEG provide converging evidence about individual imagery performance? In order to answer these questions, 6 mental imagery tasks were examined using simultaneous EEG and fMRI recordings, in combination with participant ratings. The findings revealed that, of the mental imagery tasks studied, mental arithmetic consistently produced the most robust activation at the single subject level. Additionally, there was no relationship between participants’ familiarity with an activity and the level of brain activation during performance. The key finding demonstrated that EEG and fMRI were in agreement on both of these questions, lending support to the increasing use of EEG over fMRI in disorders of consciousness.</p>