Abstract
Stress is a strong risk factor for internalizing symptoms and can lead to changes in gray matter (GM) structure associated with psychopathology. Adolescence and emerging adulthood is a time of heightened stress, onset of stress-related disorders such as depression, and GM development, and is thus posited to be a sensitive period for the effects of stress. However, not all youth who experience stress develop symptoms. Learned helplessness theory and evidence describes how having control over stressful experiences can protect against negative affective and neural outcomes of life stress. However, despite research demonstrating this protective effect in rodent models and with acute-stress paradigms in humans, it is unknown how perceived control over real life stressors relates to internalizing symptoms (depression and anxiety). Further, little is known about the neural correlates of exposure to and perceived control over life stress humans. I present four studies, each corresponding to a chapter, that address these gaps. All studies use adolescent and emerging adult samples to capture this period of heightened risk. Chapters 1 and 2 demonstrate that perceived lack of control over life stressors is associated with depression symptoms in community and clinical samples, and that this relation is distinct from the risk conferred by amount of stress exposure. Chapters 3 and 4 demonstrate that frequency of life stress exposure and perceived control over life stressors have distinct gray matter correlates. Greater perceived control over life stress is associated with GM thickness in the frontoparietal control network (FPCN) and gender-specific associations with GM in the medial prefrontal cortex and amygdala. Conversely, relations between the frequency of stress exposure and GM thickness and surface area were all gender-specific, with opposite directions of effects seen for male and female youth in the FPCN, default mode network, and limbic regions for GM thickness, and default mode network (DMN) and dorsal attention network regions for surface area. These findings suggest that appraisals of control may shape clinical and neural stress outcomes in humans although these relations should be tested longitudinally to determine temporal precedence, and that testing for gender differences is crucial to understanding the neural impact of life stress.