How to Study Stress in Psychology

The stress topic has three distinct layers, and you need to understand how they connect rather than treating them as separate topics.

Layer 1 — Physiology of stress: Master two axes. The SAM (sympathomedullary) pathway: threat → amygdala/hypothalamus → sympathetic nervous system → adrenal medulla → adrenaline and noradrenaline into bloodstream. Fast (seconds), short-lived. The HPA axis: stressor → hypothalamus (CRH) → anterior pituitary (ACTH) → adrenal cortex → cortisol. Slower (15–30 min), sustained. Selye's General Adaptation Syndrome (alarm → resistance → exhaustion) describes what happens across time if the stressor persists. Know these mechanistically — not just as sequences but as systems with feedback loops (cortisol suppresses further CRH and ACTH via negative feedback).

Layer 2 — Sources of stress and measurement: Know the SRRS (Holmes & Rahe, 1967), the Hassles & Uplifts Scale (Kanner et al., 1981), and Karasek's demand–control model. These aren't just "tools" — each embeds a different theory of what stress IS (cumulative readjustment, chronic micro-stressors, structural work conditions). Know the DeLongis et al. (1982) finding that daily hassles predicted health better than life events. Understand skin conductance (GSR/EDA) as a physiological stress measure — what it detects (sympathetic activation via sweat glands), why it's sensitive but non-specific.

Layer 3 — Individual differences and management: Why does the same stressor affect people differently? Three main answers: personality (Type A/hostility, hardiness), social support (Cohen & Wills, 1985 buffering model), and coping strategies. For management: know stress inoculation training (Meichenbaum, 1977) — three phases, what happens in each. Know biofeedback — what signal is fed back, what physiological process is being trained, why operant conditioning applies.

SRRS: Holmes & Rahe (1967). 394 participants rated 43 life events in LCUs relative to marriage (=50). US Navy personnel study linked total LCU score to subsequent illness. Know the scale's methodology and its four main weaknesses (ignores appraisal, conflates positive/negative events, retrospective recall bias, modest correlations).

Daily hassles: Kanner et al. (1981) — developed Hassles Scale (117 items) and Uplifts Scale (135 items). DeLongis et al. (1982) — daily hassles predicted health status better than major life events.

Karasek demand–control: Karasek (1979). High demand + low control = job strain = greatest health risk. Extended by Siegrist (1996) with effort–reward imbalance model.

WCGS / Type A: Rosenman et al. (1975). 3,154 California businessmen, 8.5 years, Type A had ~2× CHD rate. Know the refinement: Miller et al. (1996) meta-analysis showing hostility (not global Type A) is the toxic component.

Kiecolt-Glaser immune studies: (1984) medical students had lower NK cell activity during exams. (1987) Alzheimer caregivers showed poorer immune function and slower wound healing. Know what NK cells do.

Cohen et al. cold study: Cohen, Tyrrell & Smith (1991). 394 volunteers, deliberate rhinovirus exposure, dose-dependent relationship between stress index and cold incidence. Strongest causal design in the stress-illness literature.

Kobasa hardiness: Kobasa (1979). 837 Illinois Bell executives during AT&T divestiture. Three Cs: Commitment, Control, Challenge. Know the Funk (1992) critique (overlap with neuroticism).

Social support: Cohen & Wills (1985). Main effects (social integration) vs buffering (functional support matches stressor demands). Holt-Lunstad et al. (2015) — social isolation associated with ~29% increased mortality risk.

Stress inoculation: Meichenbaum (1977). Three phases: conceptualisation, skills acquisition, application. Most effective for PTSD, medical stress, anger.

Apply the models: Use the demand–control model to analyse real work scenarios — a nurse (high demand/variable control), a surgeon (high demand/high control), a data-entry clerk (low demand/low control). Use Lazarus's appraisal model to explain why the same stressor (an exam) produces different levels of physiological stress in different people.

Compare measurement approaches: SRRS (cumulative LCUs over a year) vs Hassles Scale (frequency/intensity of daily events) vs physiological measures (cortisol, skin conductance). Each captures a different aspect. Cortisol and GSR give you objective, real-time biological data but are non-specific. Self-report scales are retrospective and subject to biases but capture the subjective meaning of stressors.

Compare SAM and HPA: SAM is neural AND hormonal, fast (seconds), short-lived, produces cardiovascular effects. HPA is purely hormonal, slow (minutes), sustained, produces metabolic and immune effects. Both activate simultaneously. Chronic SAM activation → cardiovascular damage. Chronic HPA activation → immunosuppression, hippocampal atrophy, depression.

Critique the stress-illness evidence: Most research is correlational or retrospective — reverse causation is a serious alternative (ill people experience more life events). The Cohen et al. cold study is the strongest design but used a deliberately narrow stressor (rhinovirus); generalisability to other diseases is assumed. The SRRS ignores individual appraisal — same event, different impact. The hardiness construct may overlap with neuroticism. Type A as a global predictor has not replicated reliably — hostility is the active ingredient.

Stress is one of the most consequential topics in applied psychology. Cardiovascular disease remains the leading cause of death globally, and stress-related mechanisms (hypertension, inflammation, health behaviour change) account for a meaningful proportion of that burden. Work-related stress costs the UK economy an estimated £28 billion per year in sickness absence and reduced productivity (Deloitte, 2022). The demand–control framework directly informs job redesign interventions used by organisations.

Immunosuppression research has clinical implications beyond the common cold: Kiecolt-Glaser's research group has shown that stress delays wound healing, reduces vaccine efficacy, and predicts cancer progression — the latter through effects on NK cell activity and tumour microenvironment. Social isolation has been recognised as a public health crisis, with national loneliness strategies now in place in the UK and Japan. The biological mechanisms linking social support to health (oxytocin, reduced cortisol reactivity) are well enough understood to inform intervention design. Stress inoculation training is embedded in military preparation protocols, pre-surgical programmes, and first-responder training worldwide.

Confusing SAM and HPA axes: The adrenal medulla (SAM) and adrenal cortex (HPA) are anatomically adjacent but functionally and embryologically distinct. SAM → adrenaline/noradrenaline. HPA → cortisol. Getting the gland layer wrong, or saying the hypothalamus releases cortisol, will lose marks on almost any stress question.

Describing Selye's GAS without the exhaustion stage: The third stage — exhaustion, where physiological resources are depleted and disease vulnerability soars — is the clinically important prediction and the most-tested part.

Treating daily hassles and life events as the same construct: They measure different stressor types with different measurement tools. DeLongis et al. (1982) found hassles are the stronger predictor — make sure you know why (chronic micro-activation vs infrequent major events).

Global Type A without the hostility refinement: Describing Type A as "time urgency + competitiveness + hostility = CHD risk" without noting that subsequent research (Miller et al., 1996) identified hostility as the toxic component is an incomplete answer.

Describing social support without specifying which model: "Social support reduces stress" is too vague. Distinguish Cohen & Wills's (1985) main-effects model (social integration is always beneficial) from the buffering model (functional support specifically protects under high stress). Examiners expect this distinction.

Describing biofeedback without identifying what is being fed back: Many students write "biofeedback uses machines to help you relax" without specifying the physiological signal (GSR, EMG, HRV, temperature), the modality (skin conductance biofeedback, EMG biofeedback), or the mechanism (operant conditioning of autonomic responses). The specificity is what earns marks.