Morning Bright Light Therapy and Daytime Functioning in University Students

Background and rationale: Morning light is the principal entraining signal for the human circadian system, acting through melanopsin-containing intrinsically photosensitive retinal ganglion cells that project to the suprachiasmatic nucleus. University students frequently experience irregular sleep-wake schedules and limited daytime light exposure, particularly in cooling-dominant subtropical buildings where occupants keep blinds drawn. Bright light therapy (BLT) is an established treatment for circadian and affective disorders, but its effects on sleep regularity, daytime alertness, mood, and objective cognition in healthy students are less well characterized. This trial evaluates a brief, self-administered morning BLT regimen using a within-subject crossover design.

Design: Single-center, randomized, within-subject (crossover) trial. Each participant completes two consecutive one-week conditions-morning BLT and usual-light control-in counterbalanced order with no washout period. Allocation to condition order (sequence) is randomized, and each participant serves as their own control; all crossover models include condition, period (week), and sequence terms.

Setting and population: Healthy university students residing in shared student housing in Hong Kong. Eligibility is restricted to generally healthy young adults without a diagnosed sleep disorder; sleep quality is screened (Pittsburgh Sleep Quality Index), and individuals with diagnosed insomnia or other relevant conditions are excluded.

Intervention: During the BLT week, participants self-administer a 30-minute session of 10,000 lux (photopic) bright light approximately 30 minutes after waking, using a standard light therapy lamp positioned per manufacturer guidance. The stimulus is characterized in melanopic units (melanopic equivalent daylight illuminance ≈ 8,989 lux). During the control week, participants maintain their usual indoor lighting and morning routine without the lamp.

Assessments: Sleep is monitored continuously with a wrist-worn Fitbit and a daily diary throughout both weeks. Daytime sleepiness and alertness are rated at fixed times of day (Karolinska and Stanford Sleepiness Scales). Mood and affect are assessed with the Positive and Negative Affect Schedule and the Profile of Mood States. A computer-based cognitive battery (Stroop, Flanker, Corsi block-tapping, verbal span) is administered at the end of each condition week. Post-trial semi-structured interviews capture subjective experience and acceptability.

Outcome structure: A small, a priori primary confirmatory outcome family-defined in the approved study protocol on the basis of hypothesized domains rather than observed effects-covers daytime sleepiness, sleep regularity, positive affect, and cognition; remaining measures are secondary or exploratory. Formal outcome-measure definitions are provided in the Outcome Measures module.

Statistical analysis: Outcomes are analyzed with linear mixed-effects models of the form outcome \~ condition + period + sequence + (1 \| participant), estimated by REML with Satterthwaite degrees of freedom. Multiplicity is controlled with Holm correction across the primary outcome family. Paired within-person effect sizes (Cohen's dz) accompany model-based estimates, and sensitivity analyses examine period, sequence, and carryover effects.

Conduct in waves: The trial is being conducted in waves. A first wave (33 participants) has completed data collection; a second wave is planned and will commence pending additional funding, so recruitment is currently inactive.