PT  - JOURNAL ARTICLE
AU  - Takumi Sase
AU  - Keiichi Kitajo
TI  - The metastable human brain associated with autistic-like traits
AID  - 10.1101/855502
DP  - 2019 Jan 01
TA  - bioRxiv
PG  - 855502
4099  - http://biorxiv.org/content/early/2019/11/26/855502.short
4100  - http://biorxiv.org/content/early/2019/11/26/855502.full
AB  - Recent studies suggest that the resting brain utilizes metastability such that the large-scale network can spontaneously yield transition dynamics across a repertoire of oscillatory states. By analyzing resting-state electroencephalographic signals and the autism-spectrum quotient acquired from healthy humans, we show experimental evidence of how autistic-like traits may be associated with the metastable human brain. Observed macroscopic brain signals exhibited slow and fast oscillations forming phase-amplitude coupling (PAC) with dynamically changing modulation strengths, resulting in oscillatory states characterized by different PAC strengths. In individuals with the ability to maintain a strong focus of attention to detail and less attention switching, these transient PAC dynamics tended to stay in a state for a longer time, to visit a lower number of states, and to oscillate at a higher frequency than in individuals with a lower attention span. We further show that attractors underlying the transient PAC could be multiple tori and consistent across individuals, with evidence that the dynamic changes in PAC strength can be attributed to changes in the strength of phase-phase coupling, that is, to dynamic functional connectivity in an electrophysiological sense. Our findings suggest that the metastable human brain can organize spontaneous events dynamically and selectively in a hierarchy of macroscopic oscillations with multiple timescales, and that such dynamic organization might encode a spectrum of individual traits covering typical and atypical development.Significance Statement Metastability in the brain is thought to be a mechanism involving spontaneous transitions among oscillatory states of the large-scale network. We show experimental evidence of how autistic-like traits may be associated with the metastable human brain by analyzing resting-state electroencephalographic signals and scores for the autism-spectrum quotient acquired from healthy humans. We found that slow and fast neural oscillations can form phase-amplitude coupling with dynamically changing modulation strengths, and that these transient dynamics can depend on the ability to maintain attention to detail and to switch attention. These results suggest that the metastable human brain can encode a spectrum of individual traits by realizing the dynamic organization of spontaneous events in a hierarchy of macroscopic oscillations with multiple timescales.