Transdiagnostic, Connectome-Based Prediction of Memory Constructs Across Psychiatric Disorders

Abstract

Memory deficits are observed in a range of psychiatric disorders, but it is unclear whether memory deficits arise from a shared brain correlate across disorders or from various dysfunctions unique to each disorder. Connectome-based predictive modeling is a computational method that captures individual differences in functional connectomes associated with behavioral phenotypes such as memory. We used publicly available task-based functional MRI data from patients with schizophrenia (n = 33), bipolar disorder (n = 34), attention deficit hyper-activity disorder (n = 32), and healthy controls (n = 73) to model the macroscale brain networks associated with working, short- and long-term memory. First, we use 10-fold and leave-group-out analyses to demonstrate that the same macroscale brain networks subserve memory across diagnostic groups and that individual differences in memory performance are related to individual differences within networks distributed throughout the brain, including the subcortex, default mode network, limbic network, and cerebellum. Next, we show that diagnostic groups are associated with significant differences in whole-brain functional connectivity that are distinct from the predictive models of memory. Finally, we show that models trained on the transdiagnostic sample generalize to novel, healthy participants (n = 515) from the Human Connectome Project. These results suggest that despite significant differences in whole-brain patterns of functional connectivity between diagnostic groups, the core macroscale brain networks that subserve memory are shared.

Linking Individual Differences in Personalized Functional Network Topography to Psychopathology in Youth

“…Consistent with our findings, previous transdiagnostic studies have reported common disrupted patterns of functional connectivity across mental disorders (Baker et al, 2019; Barron et al, 2020; Ma et al, 2020; Qi et al, 2020; Xia et al, 2018). Moreover, recent studies have demonstrated that the overall psychopathology factor is associated with abnormalities of both within- and between-network functional connectivity (Elliott et al, 2018; Karcher et al, 2020; Kebets et al, 2019).…”

Prediction of life satisfaction from resting‐state functional connectome

“…To further improve the interpretability of our findings, we examined network anatomy in a similar manner to previous studies (Barron et al., 2020 ; Lake et al., 2019 ). Briefly, we computed the probability that iFCs are shared between the networks identified by our prediction model and within or between nine canonical RSNs.…”

Transdiagnostic connectome-based prediction of craving

“…For predictive modeling, a multidimensional approach was employed, which combines multiple connectomes and behavioral measures into a single predictive modeling framework. 25 Given that a single behavioral measurement represents a noisy approximation of any behavioral construct, and there were six craving ratings (i.e., before and after each imagery script), a principal components analysis (PCA) was used to summarize craving in a data-driven manner for predictive modeling with CPM. To maintain separate train and test groups, for each iteration, each PCA was limited to the training datasets and the PCA coefficients applied to the test dataset.…”

“…19,20 CPM is also data-driven, and is therefore used to identify neural 'signatures' in functional-connectivity data related to a specific phenotype. 18,21 CPM has been used to predict cognitive and clinical phenotypes, [22][23][24][25] and, more recently, has been used to predict treatment outcomes in SUDs. 26,27 This study aimed to identify a transdiagnostic 'craving network' by applying CPM to functional connectivity data in individuals with and without substance-use-related conditions, to predict a continuous measure of self-reported craving.…”

“…19,20 CPM is also data-driven, and is therefore used to identify neural ‘signatures’ in functional-connectivity data related to a specific phenotype. 18,21 CPM has been used to predict cognitive and clinical phenotypes, 22-25 and, more recently, has been used to predict treatment outcomes in SUDs. 26,27…”