New paper out in NeuroImage.

The structure of brain networks impose constraints on how neural signals circulate spontaneously or in response to external stimuli. We show in this study driven by Prof Esther Florin that pre-stimulus and resting-state activity indeed have similarities with post-stimulus responses. We used a range of methods, including hidden Markov modeling, to reveal associations across participants and experimental trials, with dependencies across brain areas, frequency bands and recording sessions. Esther was a post-doctoral researcher in the lab, before taking her first faculty position back in her home country (Germany).

From the paper: “Gamma activity is thought to serve several cognitive processes, including attention and memory. Even for the simplest stimulus, the occurrence of gamma activity is highly variable, both within and between individuals. The sources of this variability, however, are largely unknown.

In this paper, we address one possible cause: the cross-frequency influence of spontaneous, whole-brain network activity on visual stimulus processing. By applying Hidden Markov modelling to MEG data, we reveal that the trial-averaged gamma response to a moving grating depends on the individual network dynamics, inferred from slower brain activity (<35 ​Hz) in the absence of stimulation (resting-state and task baseline). In addition, we demonstrate that modulations of network activity in task baseline influence the gamma response on the level of trials.

In summary, our results reveal a cross-frequency and cross-session association between gamma responses induced by visual stimulation and spontaneous network activity. These findings underline the dependency of visual stimulus processing on the individual, functional network architecture.”