1 Dynamic changes in large-scale functional connectivity predict 2 performance in a multisensory task 3 4

21 Complex and variable behavior requires fast changes of functional connectivity in large-scale cortical 22 networks. Here, we report on the cortical dynamics of functional coupling across visual, auditory and 23 parietal areas during a lateralized detection task in the ferret. We hypothesized that fluctuations in 24 coupling, indicative of dynamic variations in the network state, might predict the animals’ 25 performance. While power for hit and miss trials showed significant differences only around stimulus 26 and response onset, phase coupling already differed before stimulus onset. Principal component 27 analysis of directed coupling at the single-trial level during this period revealed subnetworks that most 28 strongly related to behavior. While higher global phase coupling of visual and auditory regions to 29 parietal cortex was predictive of task performance, a second component showed that a reduction in 30 coupling between subnetworks of sensory modalities was also necessary, probably to allow a better 31 detection of the unimodal signals. Furthermore, we observed that long-range coupling became more 32 predominant during the task period compared to the pre-stimulus baseline. Taken together, these 33 results suggest that fluctuations in the network state, particular with respect to long-range connectivity, 34 are key determinants of the animals’ behavior. 35 certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not this version posted June 29, 2020. ; https://doi.org/10.1101/579938 doi: bioRxiv preprint