Scale-free dynamics concerns self-similarity in activity patterns across time, i.e. long-range temporal correlations(LRTC), which establish temporal structure. While LRTC as well as closely related temporal variability have been demonstrated in spontaneous brain activity, their exact psychological significance like for the level of consciousness remains unclear. Using fMRI, we here investigated both LRTC and temporal variability in anesthetic versus non-anesthetic state to test for the impact of the level of consciousness. Our data showed significant reduction in both, LRTC(measured by the power-law exponent(PLE)) and temporal variability(measured by the standard deviation(SD) of BOLD signal change), in the anesthetic state when compared to the awake state. Additionally, LRTC and SD no longer correlated with each other in the anesthetic state while both measures correlated significantly in the awake state. Finally, conducting exploratory logistic regression and ROC curve analysis, the combination of both PLE and SD showed a robust high(> 90%) classification accuracy in determining the individual subject’s state, i.e., anesthetic or awake state. In summary, our findings show for the first time global reduction in both scale-free brain activity and temporal variability in anesthetic state including their dissociation. Though exploratory this suggests that the resting state’s temporal structure, i.e. LRTC, and its coupling to temporal variability may be relevant in maintaining the level of consciousness. |