Correlation transfer from basal ganglia to thalamus in Parkinson’s disease

Abstract

There is much experimental evidence that neurons located in the basal ganglia of parkinsonian primates show increased pairwise correlations, oscillatory activity, and burst rate compared to normal brain activity. Past computational work has suggested that such changes in the firing pattern of neurons in the globus pallidus internus (GPi), the main output nucleus of the basal ganglia, may compromise thalamocortical relay capabilities. To understand how changes in the patterns of basal ganglia activity affect correlation transfer, we study pairs of realistic models of thalamocortical (TC) relay neurons receiving correlated inhibitory input from the GPi, as well as uncorrelated excitatory signals from cortex. We observe that bursty firing patterns such as those seen in the parkinsonian GPi allow for stronger transfer of correlations and higher correlation susceptibility than do firing patterns found under normal conditions. We also show that removing the T-current in the TC neurons does not significantly affect the correlation transfer, despite its pronounced effects on the spiking of the neurons. Oscillatory firing patterns in GPi are shown to affect the time scale at which correlations are best transferred through the system. We obtain the same results using an integrate-and-fire-or-burst (IFB) model of TC neurons as we do with a more realistic conductance-based model of the TC neurons, suggesting that the IFB model is a good reduced model for studying correlation transfer. In a reduced point process model, we derive analytic calculations of the spike count correlation coefficient for the time-inhomogeneous case. The analysis indicates that the rhythms seen in the transfer of correlations at varying time scales are very robust to different levels of spike correlations and rate correlations between the neurons. It also points to the fact that these rhythms can be seen because of differences in instantaneous spike correlations, even when the long time scale rhythmic modulation of the neurons in identical. Overall, these results show that parkinsonian firing patterns in GPi do indeed affect the way that correlations are transferred to the thalamus.