Brain‐based ranking of cognitive domains to predict schizophrenia

Abstract

Schizophrenia is a devastating brain disorder that disturbs sensory perception, motor action, and abstract thought. Its clinical phenotype implies dysfunction of various mental domains, which has motivated a series of theories regarding the underlying pathophysiology. Aiming at a predictive benchmark of a catalog of cognitive functions, we developed a data‐driven machine‐learning strategy and provide a proof of principle in a multisite clinical dataset (n = 324). Existing neuroscientific knowledge on diverse cognitive domains was first condensed into neurotopographical maps. We then examined how the ensuing meta‐analytic cognitive priors can distinguish patients and controls using brain morphology and intrinsic functional connectivity. Some affected cognitive domains supported well‐studied directions of research on auditory evaluation and social cognition. However, rarely suspected cognitive domains also emerged as disease relevant, including self‐oriented processing of bodily sensations in gustation and pain. Such algorithmic charting of the cognitive landscape can be used to make targeted recommendations for future mental health research.

Functional specialization within the inferior parietal lobes across cognitive domains

“…To rigorously assess whether connectivity links between a specific IPL subregion and cortical parcel were reliably weaker or stronger in one task relative to the other two, a pooled permutation-based baseline was computed across all three task conditions. Task-specific functional coupling shifts were determined by statistical significance testing based on a non-parametric permutation procedure using an empirical nullhypothesis distribution 26,71 . The data-derived null model reflected the constellation of neural activity coupling strengths between a given subregion and other cortical regions that would be expected if task A induced similar patterns of brain connectivity, compared to the respective other tasks B and C. Following this fully data-driven analysis tactic, for each of the three tasks, the analysis directly provided brain maps of task-dependent functional connectivity profiles for the IPL subregions.…”

Looking within: Interoceptive sensibility in young adults with psychotic‐like experiences

“…A vast body of literature has described disrupted sleep (e.g., Batalla-Martín et al, 2020;Ong et al, 2020), aberrant sense of self (e.g., Keromnes et al, 2018;Moe & Docherty, 2014;Parnas & Handest, 2003), and impaired social cognition (Green et al, 2015;Vaskinn & Horan, 2020) as core features of psychosis. These deficits, along with anomalous bodily experiences (Nyboe et al, 2016;Stanghellini et al, 2012), and neuroanatomical and functional abnormalities within the insular cortex (Ebisch et al, 2013;Ebisch et al, 2014;Karrer et al, 2019), provide indirect evidence to suggest that interoception may be impaired in those with psychosis. Remarkably, only one study to date has directly investigated the relationship between interoception and psychosis.…”

Deep learning identifies partially overlapping subnetworks in the human social brain

“…For hyperparameter search, we tuned the maximum depth (2 or 6), the minimal split of samples (2 or 6), and the minimum samples of leaves (2 or 6). We noticed that fitting 100 decision trees showed saturation in prediction accuracy based on the out-of-bag estimates on training data from unseen UKB participants by a given decision tree 50 , 56 . Our rationale was to test for the existence of exploitable non-linear effects in our brain imaging data for predicting social traits.…”