Neural oscillations underlie critical computational and representational functions of the brain. Oscillatory activity extends from the millisecond cycles of an interneuron-pyramidal neuron microcircuit, to flows of information over hundreds of milliseconds in columnar mesocircuits, to the coordination of long-range brain macrocircuit interactions over seconds that support higher order cognitions (Mathalon and Sohal, 2015). These three regimes correspond roughly with three orders of magnitude of frequency ranges: 100 Hz (high gamma), 10 Hz (delta to low gamma), and <1 Hz (infraslow). Oscillations arise from and interact on a neuronal scaffold, whose intrinsic property is plasticity—both developmental and experience dependent. We propose that psychiatric illnesses are pathologies of the oscillatory connectome, in which critical representational processes generated within neuronal architecture and supported by oscillatory coupling are distorted.

We define the oscillatory connectome (OC) as the patterns of oscillatory coupling of neuronal populations under given conditions, physically conjoined with a specific axodendritic and glial architecture. The OC stores information and executes computations through plasticity in topology, synaptic strength, and membrane conductance (Sejnowski and Paulsen, 2006). It reflects the interplay between an individual’s genome, exposome, developmental stage, and cognitive/behavioral repertoire.

OC pathologies frequently manifest at longer time scales across large cortical and subcortical neural populations (eg, abnormal prefrontal-subgenual network dynamics seen in depression during processing of emotionally evocative stimuli (Smart et al, 2015)). Pathology may also be observed at shorter time scales and within localized neuronal assemblies, such as impaired auditory representations in early psychosis that progress concomitant with volume reductions in Heschl’s gyrus (Salisbury et al, 2007). OC pathologies appear to be probabilistically related to clinical psychiatric features.

Defining psychiatric illnesses as oscillatory connectomopathies has three immediate research implications:

  1. 1

    Structural and physiological assessments of the brain must be integrated, ideally combining detailed information on neural architecture with measures of oscillation patterns and their coupling across different frequency bands and brain regions.

  2. 2

    As genomics begins to elucidate molecular components of abnormal synaptic and microcircuit function, we must discover how such abnormalities contribute to meaningful variations in oscillatory meso- and macrocircuits. We predict that dysplasticity mechanisms will represent key common pathways—processes that affect neural architecture or communication over time in a manner that:

    1. a

      Impedes normal developmental and experience-dependent plasticity in both micro and mesoscale oscillatory dynamics (as is likely in schizophrenia and autism), or

    2. b

      Biases macroscale plasticity toward selective enhancement of maladaptive but highly salient representations (as happens in addictions, depression, PTSD).

  3. 3

    Significant innovations in psychiatric nosology and treatment development will require an understanding of neural oscillatory connectomics in health and disease. In social anxiety, resting state connectivity metrics (indirectly measuring the infraslow OC) and tractography of right inferior longitudinal fasciculus is five times better than symptom severity at predicting improvement after CBT (Whitfield-Gabrieli et al, 2015). In schizophrenia, intensive auditory training drives changes in oscillatory dynamics across auditory and prefrontal cortices that correlate with cognitive gains (Dale et al, 2015). ‘RDoC’ and ‘successful target engagement’—the new buzzwords in psychiatric research—ultimately mean understanding and harnessing adaptive changes in the neural network oscillation patterns that give rise to human behavior.

Funding and disclosure

Sophia Vinogradov is a site investigator on an SBIR grant to PositScience, Inc., a company with a financial interest in cognitive training software; she is also a consultant to Forum Pharmaceuticals. There are no competing financial interests in relation to the work described in this article. Dr Herman has no disclosures to report.