Cognitive Effects on the Neurophysiology and Biomechanics of Stroke Recovery

author:Solomon Gilbert Diamond
adviser:Robert D. Howe
year:2004
degree: Ph.D.
institution: Harvard University

Motor function recovery after stroke typically plateaus after six months. Case reports suggest that hypnosis has the potential to stimulate further recovery of function beyond this period. A pilot clinical investigation of hypnosis-aided recovery of motor function after stroke is presented. Baseline hand motor performance was quantified for six stroke subjects with a force-following task. Brain activity during the task was measured using functional magnetic resonance imaging (fMRI). After cognitive training with hypnosis for improved motor performance, subjects performed the task again. Reaction times and muscle contraction and relaxation rates improved significantly after hypnosis and changes persisted during follow up testing at two weeks. Hypnosis versus baseline fMRI results show increased activation extent in bilateral sensorimotor cortex with an ipsilateral shift in laterality. No significant differences were found in motor performance or fMRI results when the unaffected hand performed the task as a control condition.

The autonomic physiology of hypnosis was studied by correlating heart rate variability (HRV) model parameters with hypnotic depth and motor performance changes. Modeling the low frequency (LF) and high frequency (HF) components in HRV as a sum of two sinusoids poses a significant signal processing challenge. An Improved Prony Algorithm (IPA) is presented for estimating the frequencies of real-valued sinusoids embedded in noise. The variance in estimated frequency is exactly analyzed for the single sinusoid case and a closed-form solution is derived for the estimation of two sinusoid frequencies.

Utilizing IPA, the effects of hypnosis on HRV were studied. Ten normal subjects used a lever to dynamically rate their hypnotic depth. Principle findings were that heart rate and HF frequency were negatively correlated with self-rated hypnotic depth whereas HF amplitude was positively correlated. IPA analysis of HRV data collected from stroke subjects in the pilot clinical investigation showed persistent HRV effects immediately after hypnosis. Muscle relaxation rate was correlated with lower heart rate and higher HF amplitude indicating that the autonomic relaxation effect of hypnosis may play a role in the observed motor performance changes. The neuroimaging, biomechanical and HRV results of this pilot study justify further investigation of neurorehabilitation with hypnosis.



Harvard BioRobotics Laboratory Home