Caffeine Effects on Sleep Taken 0, 3, or 6 Hours Before Bedtime, PSY – Study Notes
Difficulty: Introductory | Prerequisites: Basic understanding of research methods (independent vs. dependent variables, placebo controls)
Source: Drake, C., Roehrs, T., Shambroom, J., & Roth, T. (2013). Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine, 9(11). http://dx.doi.org/10.5664/jcsm.3170
Tags: caffeine, sleep disruption, sleep hygiene, adenosine, double-blind study, placebo-controlled, circadian rhythm, sleep latency, sleep architecture, psychopharmacology, health psychology
This study sits at the intersection of psychopharmacology and sleep science, two areas that come up repeatedly in introductory and intermediate psychology courses. Caffeine is the most widely consumed psychoactive substance in the world, yet most people have surprisingly little understanding of how long its effects persist. The research matters because it challenges a common assumption: that drinking coffee in the afternoon is harmless to sleep. If you are covering topics such as states of consciousness, biological rhythms, or research methods in psychology, this study is a useful case to know well.
Twelve healthy sleepers took 400 mg of caffeine at 0, 3, or 6 hours before bedtime in a double-blind, placebo-controlled study. Even caffeine consumed six hours before bed significantly disrupted sleep. A key finding was that participants were often unaware of the disruption, which helps explain why people ignore sleep hygiene advice about afternoon caffeine.
Double-blind study
An experimental design in which neither the participants nor the researchers administering the treatment know who is receiving the active substance and who is receiving the placebo. In simple terms, this means nobody involved can let expectations bias the results.
Placebo-controlled
A study design that includes a sham treatment (the placebo) identical in appearance to the active treatment, so that any observed effects can be attributed to the substance itself rather than to participants' beliefs. Think of it as the control condition that lets you rule out the "I think I took caffeine, so I can't sleep" effect.
Independent variable (IV)
The variable the researcher deliberately manipulates. In this study, the IV was the timing of caffeine administration: 0, 3, or 6 hours before scheduled bedtime.
Dependent variable (DV)
The variable that is measured to see whether it changes in response to the IV. Here, the DV was the amount and quality of sleep participants obtained.
Sleep disturbance
Any disruption to the normal pattern or duration of sleep, including difficulty falling asleep, waking during the night, or reduced total sleep time. In simple terms, this means your sleep was worse than usual, whether or not you noticed it.
Subjective vs. objective sleep measures
Subjective measures come from self-report (e.g. a sleep diary). Objective measures come from monitoring devices that track actual sleep physiology. Think of it as the difference between how you felt you slept and what the data say happened.
Sleep hygiene
A set of behavioural and environmental recommendations designed to promote healthy sleep. Common examples include keeping a consistent bedtime, avoiding screens before bed, and limiting caffeine intake in the hours before sleep.
The study was experimental in nature: researchers manipulated caffeine timing and measured the effect on sleep
It used a within-subjects design, meaning each participant experienced all conditions (0, 3, and 6 hours before bed, plus placebo)
The design was randomised, double-blind, and placebo-controlled, which is considered a strong methodology for minimising bias
12 healthy, normal sleepers
Recruited from the Detroit tri-county area via local advertisements
Screened by a sleep specialist to exclude anyone with a history of insomnia, difficulty falling or staying asleep, or non-restorative sleep lasting a month or more
No opposing or clinical groups were included; the study focused exclusively on people with healthy baseline sleep
Participants took three pills each day
One pill contained 400 mg of caffeine (roughly equivalent to two to three cups of coffee)
The other two pills were placebos
Participants did not know which pill was the active one
Each participant completed four conditions across different nights: caffeine at 6 hours before bed, 3 hours before bed, 0 hours before bed (at bedtime), and a placebo-only night
Every morning, participants logged a sleep diary, including on placebo-only days
The study ran for approximately one week
A moderate dose of caffeine (400 mg) taken at bedtime significantly disrupted sleep
Caffeine taken 3 hours before bedtime also significantly disrupted sleep
Caffeine taken 6 hours before bedtime still produced measurable sleep disruption
Sleep disturbance was detected by both objective sleep monitors and subjective self-report diaries
A critical finding: participants frequently lacked subjective awareness of the sleep disruption caused by caffeine, even when objective measures confirmed it
The gap between what the monitors showed and what participants reported suggests that people may not realise caffeine is harming their sleep. This helps explain why many individuals continue to consume caffeine in the afternoon or evening despite sleep hygiene guidelines advising against it.
This research is directly relevant to public health messaging around caffeine and sleep. It is part of the evidence base behind the common recommendation to avoid caffeine after early afternoon. For anyone who drinks coffee, tea, or energy drinks regularly, the finding that you may not even notice the damage to your sleep is worth knowing: you cannot rely on how you feel in the morning to judge whether yesterday's afternoon coffee was harmless.
Students often assume that if you "feel fine" after afternoon caffeine, your sleep was unaffected. The study shows objective disruption can occur without subjective awareness.
Some students describe this study as purely correlational. It is an experiment: the researchers manipulated the timing of caffeine intake (the IV) and measured its effect on sleep (the DV) under controlled conditions.
Students sometimes confuse the number of conditions with the number of groups. This was a within-subjects design: the same 12 participants completed all four conditions, rather than being split into separate groups.
The dose used (400 mg) is sometimes described as "a cup of coffee." It is closer to two to three standard cups, which matters when discussing effect size.
⚠️ Know the IV (timing of caffeine) and DV (sleep amount/quality). This is a staple research-methods question.
⚠️ Be able to explain why the study is experimental rather than correlational: the researchers manipulated the IV and used random assignment to conditions.
⚠️ The discrepancy between subjective and objective sleep measures is a frequently tested point. It ties into broader themes of self-report bias and the limits of introspection.
⚠️ Understand what "double-blind" and "placebo-controlled" mean in context. You may be asked to identify these features from a study description.
⚠️ The finding that even 6 hours before bed is problematic is the headline result that exam questions tend to focus on.
True or False: Caffeine consumed 6 hours before bedtime had no measurable effect on sleep in this study. (False: it still significantly disrupted sleep.)
Fill in the blank: The independent variable in this study was the ______ of caffeine administration relative to bedtime. (timing)
True or False: Participants always noticed when their sleep was disrupted by caffeine. (False: objective measures showed disruption that participants did not subjectively report.)
Fill in the blank: The study used a ______, double-blind, ______-controlled design. (randomised, placebo)
True or False: Different groups of participants were assigned to each caffeine timing condition. (False: it was a within-subjects design; all participants completed all conditions.)
Q: What were the independent and dependent variables in the Drake et al. (2013) study?
A: The independent variable was the timing of caffeine intake (0, 3, or 6 hours before bedtime). The dependent variable was the amount and quality of sleep obtained by participants.
Q: Why is this study classified as an experiment rather than a correlational study?
A: The researchers manipulated the independent variable (caffeine timing) under controlled conditions and used random assignment to the order of conditions. In a correlational study, the researcher observes naturally occurring relationships without manipulation.
Q: What was the most surprising finding of the study, and why does it matter for sleep hygiene?
A: Participants often lacked subjective awareness of caffeine-induced sleep disruption, even when objective monitors confirmed it. This matters because it suggests people may continue consuming caffeine in the afternoon simply because they do not perceive the harm it causes to their sleep.
Q: Describe two features of the study design that strengthen confidence in the results.
A: The double-blind procedure ensured that neither participants nor researchers knew which pill was caffeine, reducing expectation bias. The placebo control provided a baseline to confirm that observed sleep disruption was caused by caffeine rather than by taking a pill or participating in the study.
Q: A student claims that caffeine only disrupts sleep if taken right at bedtime. How does this study challenge that claim?
A: The study found significant sleep disruption even when caffeine was consumed 6 hours before bed. This indicates that caffeine's effects on sleep persist well beyond the point at which most people assume they have worn off.
This study connects to states of consciousness because sleep is one of the primary altered states covered in most introductory psychology courses, and caffeine directly interferes with the transition into sleep.
It also ties into research methods: the study is a clean example of a within-subjects, double-blind, placebo-controlled experiment, making it useful revision material for any exam that asks you to identify design features from a scenario.
The subjective vs. objective measurement discrepancy links to broader discussions of self-report bias and the limits of introspection, which appear in personality, social, and clinical psychology modules.
Caffeine and sleep, sleep disruption, sleep latency, sleep architecture, adenosine receptor antagonist, psychoactive substances, double-blind experiment, placebo effect, within-subjects design, repeated measures, self-report bias, objective sleep measurement, sleep diary, actigraphy, polysomnography, circadian rhythm, sleep hygiene, Drake et al. 2013, Journal of Clinical Sleep Medicine, consciousness, biological rhythms, psychopharmacology, stimulant effects on sleep, afternoon caffeine, coffee and insomnia