Our Lab
Dr. Stephen Aichele – Director
The overarching goals of the CHA lab are to understand how individual differences in age-related cognitive trajectories are shaped by biopsychosocial factors—and how cognitive processes in turn influence mental health (dementia and depression risk) and longevity (mortality risk) in later adulthood.
Modeling cognitive change is central to this work. Different cognitive processes change in different ways as a function of ‘normal healthy aging’, in response to training or therapeutic interventions, through dynamic and interdependent associations with other processes, and in relation to genetic, social, lifestyle, and environmental factors. Statistical approaches that we use to capture such associations include structural equation models, multilevel models, joint longitudinal and time-to-event models, and machine learning approaches. We apply these and other methods to data sets both large and small, in collaboration with colleagues based in the USA and in Europe.
Current projects primarily focus on secondary data analyses related to the Manchester Longitudinal Study of Cognition (MLSC), one of the first studies of long-term cognitive changes in middle-aged and older adults. We have recently begun to investigate the effects of long-term air pollution exposures on age-related cognitive declines and Alzheimer’s-related neuropathology. In addition to rich longitudinal cognitive data, we are fortunate to have access to genetic and human brain autopsy information and to benefit from the expertise of collaborators at the Universities of Manchester and Oxford, UK.
Other recent projects have leveraged state-of-the-art statistical methods to investigate:
- cognitive reserve(compensatory factors for age- or pathology-related brain damage)
- cognitive epidemiology(cognitive decline as predictive of mortality risk)
- cognitive factors as comparatively predictive of depression and anxiety
- long-term effects of cognitive (attentional) training on emotion regulation and personality
- cognition-depression dynamics (time-ordered, bidirectional associations)