Graduation Year


Date of Submission


Document Type

Campus Only Senior Thesis

Degree Name

Bachelor of Arts



Reader 1

Dr. Stacey Doan

Reader 2

Dr. Brian Duistermars

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

2022 Benjamin SW Wolters


According to the allostatic theory of stress, temporary increases in cortisol levels serve as healthy, adaptive responses to acute stressors (Sapolsky et al., 2000). Cortisol levels increase to ensure that the body maintains blood-glucose levels sufficient to sustain an adequate “fight or flight” response. Alternatively, allostatic theory suggests that cortisol hyporeactivity—the body’s inability to mount a cortisol response—is a maladaptive response to an acute stressor. Blunted cortisol responses often result from chronic stress-induced hypercortisolemia which is highly associated with and has been shown to serve as a mechanism for the pathogenesis of Major Depressive Disorder (MDD) (Amsterdam et al., 1988; Tolfoli et al., 2011; Maletic et al., 2007). Yet, empirical research that measures cortisol responses to laboratory stressors does not consistently find relationships between cortisol reactivity and MDD that corroborate allostatic theory (Morris et al., 2017). The present study terms this inconsistency the “cortisol puzzle” and produces evidence suggesting that the puzzle, rather than casting doubt on allostatic theory, reflects the failure of past research to account for whether participants actually perceived artificial stress tasks as psychologically stressful. With a design intended to match prior research, the present study measured the change in salivary cortisol levels of 374 undergraduates (Mage= 22.6, 55% cisgender female) from both the Claremont Colleges and Auburn University in response to the Trier Social Stress Test. Cortisol reactivity had no main effect on MDD—a finding in line with the inconsistent findings of previous literature. However, perceived stress significantly moderated this relationship such that more blunted cortisol responses were associated with more depressive symptoms if participants viewed the task as stressful. These findings suggest the results of prior studies were downward biased by participants who did not perceive the laboratory tasks as stressful.

This thesis is restricted to the Claremont Colleges current faculty, students, and staff.