Fig. 1: Operationalizing network control theory in the context of human neuroimaging.
A The strength of structural connections between brain regions were determined by the quantitative anisotropy (QA) estimated from diffusion spectrum imaging data. We used beta coefficients from general linear models to specify brain activation maps during task sessions where participants were given 1 mg oral alprazolam or placebo. These maps were then fed into a network control model to analyze the energy required for transitions between different brain states. We were particularly interested to estimate the persistence energy, Pe, defined as the energy required to maintain a state. Brain regions in the cortex and subcortex were defined by the 234-node Lausanne parcellation. B Schematic of emotion identification and emotion memory tasks. In the emotion identification task, participants were required to identify the emotions expressed on the displayed faces, with variable crosshair fixation between cues. In the emotion memory task, the same sequence of faces as in the preceding emotion identification task was presented, with each target face accompanied by two foil expressions. Participants were instructed to recall the expression that matched the previously seen face.
