Next we examined how the fastest rhythm in the network, the gamma

Next we examined how the fastest rhythm in the network, the gamma rhythm, was related to simultaneous theta and, in the simulations demonstrating a pattern completion phenomenon, alpha oscillations. To this end, n:m phase synchrony and phase-amplitude coupling effects were evaluated buy Epacadostat for different pairs of rhythms. The strongest phase-amplitude coupling was observed between theta and gamma oscillations (strength of modulation, PLVPAM=0.80, see Experimental

procedures) with gamma amplitude lowest at the peaks of theta (cf. Jacobs and Kahana, 2009) in accordance with the modulatory effect of theta phase on pyramidal cell firing ( Fig. 8A). The phase-amplitude modulation for theta-alpha and alpha-gamma pairs was estimated at ~0.75 and ~0.70, respectively. As can be seen in Fig. 8A, a similar hierarchy

of coupling relations was also found with 1:3 phase synchrony (PLV1:3) for theta-alpha and alpha-gamma rhythms, and 1:9 for the theta-gamma pair. In the simulations of memory replay analogous results for theta and gamma coupling ( Fig. 8B) were reported. In conclusion, gamma appeared as a basic unit in a hierarchical organization of neural oscillations consistently with biological evidence ( Basar et al., 2001, Lakatos et al., 2005, Canolty this website et al., 2006, Sirota et al., 2008, Schroeder and Lakatos, 2009, Canolty and Knight, 2010 and Palva et al., 2010). We also investigated how spiking activity was controlled within this hierarchy of LFP rhythms. The spike phase distributions indicated larger width of modulation by slower theta oscillations than faster gamma ( Fig. 8A and B). Finally, to connect our work with theories based on experimentally observed precise spatiotemporal firing patterns, we investigated the repetitive occurrence of those in the simulations

with cued memories. For 50 reactivations of the same pattern, we used a “sliding tape algorithm” (Abeles and Gerstein, 1988; see Experimental procedures) to identify all multi-neuronal sequential firing patterns consisting of at least three spikes and occurring more than twice (Fig. 9A and B). We found significantly more such patterns than expected at a chance level. In the oscillatory regime, we could observe a higher number 2-hydroxyphytanoyl-CoA lyase of spatiotemporal spike patterns that occurred at least three times within a trial despite considerably higher firing rates in the regime without gamma and alpha oscillations (25 compared to 8 s−1 on average). Finally, clear differences in the distribution of the total spike sequence durations (time span) vs. the number of their reoccurrences (Fig. 9C and D) reflected the modulatory effect of the underlying alpha rhythm on firing patterns in the oscillatory case. We used a biophysically detailed attractor network model, which with minor modifications was adapted to simulations of two memory phenomena – memory pattern completion and periodic replay of memory items.

Comments are closed.