The Use of Canonical Correlation Analysis to Assess the Relationship Between Executive Functioning and Verbal Memory in Older Adults

The roles of executive functioning, simple attention, and medial temporal lobes in early learning, late learning, and delayed recall

Fractionating performance of a verbal list-learning test can provide a nuanced interpretation of the relationship between brain networks and learning and memory abilities. Within older adult samples, including those with mild cognitive impairment and Alzheimer’s disease, cortical volumes for attention and executive functioning networks correlate more strongly with neuropsychological performance measures of early learning trials relative to late learning and delayed recall. In contrast, medial temporal lobe (MTL) structures, such as the hippocampus, are more strongly correlated to performance on late learning and delayed recall measures relative to early learning. We sought to extend these findings by evaluating the contributions of simple attention, executive function (EF), and MTL structures to learning and recall in a cognitively heterogeneous sample of older adults that included healthy controls (n = 54), adults with MCI (n = 63), and those with dementia (n = 13). We used canonical correlation analyses to test the hypotheses that the contributions of EF, simple attention, and the MTL to verbal memory would differ across phases of learning and recall. Results showed that relationships between the MTL and memory were the only ones to demonstrate a graded pattern of association, ranging from r = .46 to .57 across early learning, late learning, and delayed recall. Simple attention and EF were both significantly and moderately related to learning and recall, but those relationships did not vary across phases as hypothesized. We explore alternative interpretations for our discrepant findings, including the influence of sample characteristics and methodology, advocating for multivariate approaches.

Multi-group analysis using generalized additive kernel canonical correlation analysis

Multivariate analysis has been widely used and one of the popular multivariate analysis methods is canonical correlation analysis (CCA). CCA finds the linear combination in each group that maximizes the Pearson correlation. CCA has been extended to a kernel CCA for nonlinear relationships and generalized CCA that can consider more than two groups. We propose an extension of CCA that allows multi-group and nonlinear relationships in an additive fashion for a better interpretation, which we termed as Generalized Additive Kernel Canonical Correlation Analysis (GAKCCA). In addition to exploring multi-group relationship with nonlinear extension, GAKCCA can reveal contribution of variables in each group; which enables in-depth structural analysis. A simulation study shows that GAKCCA can distinguish a relationship between groups and whether they are correlated or not. We applied GAKCCA to real data on neurodevelopmental status, psychosocial factors, clinical problems as well as neurophysiological measures of individuals. As a result, it is shown that the neurophysiological domain has a statistically significant relationship with the neurodevelopmental domain and clinical domain, respectively, which was not revealed in the ordinary CCA.

CONFINED: distinguishing biological from technical sources of variation by leveraging multiple methylation datasets

Methylation datasets are affected by innumerable sources of variability, both biological (cell-type composition, genetics) and technical (batch effects). Here, we propose a reference-free method based on sparse canonical correlation analysis to separate the biological from technical sources of variability. We show through simulations and real data that our method, CONFINED, is not only more accurate than the state-of-the-art reference-free methods for capturing known, replicable biological variability, but it is also considerably more robust to dataset-specific technical variability than previous approaches. CONFINED is available as an R package as detailed at https://github.com/cozygene/CONFINED.