Resting state functional connectivity is associated with cognitive dysfunction in non-demented people with Parkinson's disease.

BACKGROUND: Parkinson's disease (PD) can result in cognitive impairment. Executive dysfunction often appears early, followed by more widespread deficits later in the course of the disease. Disruption of parallel basal ganglia thalamo-cortical loops that subserve motor and cognitive function has been described in PD. However, there is emerging evidence that the default mode network, a cortical network that is active at rest with reduced activation during task performance, may also play a role in disease related cognitive decline. OBJECTIVE: To determine the relative contribution of the executive control and default mode networks to parkinsonian executive dysfunction in medicated non-demented patients. METHODS: We used BOLD fMRI to measure resting state functional connectivity in the executive control and default mode (DM) networks, and examined switching, processing speed, working memory/attention and motor performance in 14 medicated non-demented PD participants and 20 controls. RESULTS: Performance on neuropsychological measures was similar across groups. Functional connectivity was not different across disease conditions in the executive control network. DMN functional connectivity was decreased in the PD group, specifically between posterior cingulate, medial prefrontal, and inferior parietal nodes. Greater DMN functional connectivity was associated with faster processing speed in the PD group. CONCLUSIONS: The continuous relationship between DMN disconnection and executive task performance indicates a possible biological contributor to parkinsonian cognitive deficits. The dynamics of executive control network change may be different than that of the DMN, suggesting less sensitivity to early cognitive deficits.

Efficacy of tailored computer-based neurorehabilitation for improvement of movement initiation in Parkinson's disease.

While Parkinson's disease (PD) is considered a motor disorder, motor signs of PD can be exacerbated by cognitive dysfunction. We evaluated the efficacy of a computer-based cognitive rehabilitation training program designed to improve motor-related executive function. Thirty people with PD and 21 controls participated in the 10-day training. Training consisted of a two-phase button press task. First, subjects produced an externally cued (EC) digit sequence, typing numbers displayed on the computer screen. Second, subjects were prompted to generate the same sequence in the absence of the number display (internally represented sequence, IR). Sequence length was automatically adjusted to maintain 87% correct performance. Participants were evaluated before and after training using a fixed version of the training task, and generalization of training was assessed using measures involving IR motor sequencing, switching and activities of daily living. PD participants were divided into two groups, those who showed impairment in IR motor sequence production prior to training (N=14) and those whose performance was similar to controls (N=16). Following training the impaired PD group showed significantly greater reduction in sequence initiation and completion time and in error rate for IR conditions compared to the unimpaired PD and control groups. All groups improved on Trails B-A, and pre-training Trails B was identified as a predictor of training-based improvement in IR sequence completion time and error rate. Our findings highlight the importance of neurorehabilitation tailored to the specific cognitive deficits of the PD patient.