High-Precision Phenotypic Estimates for Brain-Behaviour Association Analyses

This thesis aims to establish precise and reliable measurement approaches for the variables connecting brain activity to human behaviour. I find that refining brain imaging methods (through noise reduction techniques and brain connectivity estimators) has limited impact on strengthening the associations between brain and behaviour. Additionally, I show that a commonly used behavioural measure in brain-behaviour research lacks sufficient validity and reliability, leading to imprecise behavioural estimates. These findings suggest that refining behavioural measurement, rather than brain imaging techniques, may be key to uncovering robust and reliable brain-behaviour relationships.