Brain Region-wise Connectivity-based Psychometric Prediction Framework, Interpretation, Replicability and Generalizability

Wu, Jianxiao

Dissertation / PhD Thesis

FZJ-2023-02026

Brain Region-wise Connectivity-based Psychometric Prediction Framework, Interpretation, Replicability and Generalizability

Wu, J. (Corresponding author)FZJ*

2022

50p (2022) = Dissertation, HHU, 2022

Please use a persistent id in citations: http://hdl.handle.net/2128/34387

Abstract: The study of brain-behavior relationships is a fundamental aspect of neuroscience.Recently, it has become increasingly popular to investigate brain-behavior relationshipsby relating the interindividual variability in psychometric measure to the interindividualvariability in brain imaging data. In particular, prediction approaches withcross-validation can be useful for identifying generalizable brain-behavior relationshipsin a data-driven manner. Nevertheless, it remains to be ascertained what brain-behaviorrelationships can be interpreted from the prediction models, and how generalizable themodels are to fully new cohorts. In this work, we attempt to fill in the gap ofinterpretability by developing a region-wise connectivity-based psychometric prediction(CBPP) framework. This framework involves a region-wise approach where a predictionmodel is estimated and evaluated for each brain region. The prediction accuracy of eachregion-wise model is a direct indication of that brain region’s association with thepsychometric measure predicted. In study 1, we applied the framework to a range ofpsychometric variables from a large healthy cohort and demonstrated the helpfulness ofthe framework in constructing region-wise psychometric prediction profiles orpsychometric-wise prediction pattern across the brain. In study 2, we demonstrated theusefulness of the framework in assessing cross-cohort replicability and generalizability interms of brain-behavior relationships derived from the prediction models, instead of justbased on prediction accuracies. In study 3, we systematically examined existingpsychometric prediction studies, summarizing the trends in the field, calling for the useof large cohorts and external validation. Overall, our work suggested the importance ofinterpretability and generalizability for psychometric prediction, recommending the useof multiple large cohorts in evaluating the interpretability and generalizability.

Note: Dissertation, HHU, 2022

Contributing Institute(s):