Course outline
Module 1
Module 1: Foundations of Sleep Science
Foundations of Sleep Science
Knowledge check
Q1. What are the two primary states that alternate during a typical night of human sleep?
- A. Wakefulness and REM sleep
- B. REM sleep and non-REM sleep ✓
- C. Light sleep and deep sleep only
- D. Dreaming and waking rest
Why: Human sleep cycles between REM (rapid eye movement) sleep, when most vivid dreaming occurs, and non-REM sleep, which is further divided into lighter and deeper stages. These two states alternate in cycles of roughly 90 minutes.
Q2. Which sleep stage is most associated with physical restoration and growth hormone release?
- A. Stage N1 (light sleep)
- B. REM sleep
- C. Stage N3 (slow-wave or deep sleep) ✓
- D. Wakefulness before sleep onset
Why: Stage N3, also called slow-wave sleep, is the deepest non-REM stage. It is strongly linked to physical recovery, tissue repair, and the largest pulses of growth hormone secretion.
Q3. What does an actigraphy-based sleep tracker primarily measure to estimate sleep?
- A. Brain electrical activity through EEG
- B. Eye movements through electrooculography
- C. Body movement through an accelerometer ✓
- D. Blood oxygen levels through pulse oximetry
Why: Actigraphy uses accelerometers to detect movement. Sleep is inferred from periods of low movement, though this method cannot directly identify sleep stages the way EEG can.
Q4. Why is polysomnography considered the gold standard for measuring sleep?
- A. It is the cheapest method available for home use
- B. It records brain waves, eye movements, muscle tone, and breathing simultaneously ✓
- C. It only measures heart rate during sleep
- D. It requires no equipment attached to the body
Why: Polysomnography combines EEG (brain activity), EOG (eye movements), EMG (muscle tone), airflow, breathing effort, and heart rhythm. This multi-signal approach allows clinicians to score sleep stages and detect disorders precisely.
Q5. What is a key limitation of consumer sleep trackers compared with clinical sleep studies?
- A. They cannot measure sleep duration at all
- B. They are generally less accurate at distinguishing individual sleep stages ✓
- C. They always overestimate the amount of REM sleep by a fixed margin
- D. They can only be used in a sleep laboratory
Why: Consumer wearables typically estimate sleep versus wake reasonably well but are less reliable at classifying specific stages such as N1, N2, N3, and REM. Their stage estimates should be treated as approximations rather than clinical measurements.
Module 2
Module 2: Sleep Stages and Their Measurement
Sleep Stages and Their Measurement
Knowledge check
Q1. Which sleep stage is characterized by rapid eye movements and vivid dreaming?
- A. Stage N1 (light sleep)
- B. Stage N2 (light sleep with sleep spindles)
- C. Stage N3 (deep sleep)
- D. REM sleep ✓
Why: REM (rapid eye movement) sleep is the stage associated with rapid eye movements, vivid dreaming, and muscle atonia. The other stages are non-REM stages with slower brain activity.
Q2. What type of brain wave is dominant during deep (slow-wave) sleep?
- A. Beta waves
- B. Alpha waves
- C. Delta waves ✓
- D. Gamma waves
Why: Delta waves are slow, high-amplitude brain waves (0.5 to 4 Hz) that dominate N3, the deep sleep stage. Beta and gamma waves are fast and associated with alert wakefulness, while alpha waves appear during relaxed wakefulness.
Q3. Which measurement technique is most commonly used in clinical settings to determine sleep stages?
- A. Electrocardiography (ECG)
- B. Electroencephalography (EEG) ✓
- C. Pulse oximetry
- D. Actigraphy
Why: EEG measures electrical activity in the brain and is the standard for identifying sleep stages. Used together with EOG and EMG in polysomnography, it provides the most accurate staging. ECG tracks heart activity, pulse oximetry measures blood oxygen, and actigraphy measures movement.
Q4. What does EMG (electromyography) measured during sleep primarily reveal?
- A. Eye movement patterns
- B. Heart rate variability
- C. Muscle tone and activity ✓
- D. Brain wave frequency
Why: EMG records electrical activity from muscles, usually the chin, to assess muscle tone. Muscle tone is highest in wakefulness and light sleep and drops sharply during REM sleep, which is why EMG helps distinguish REM from other stages.
Q5. Consumer wearable sleep trackers most often estimate sleep stages using which combination of signals?
- A. EEG and EOG only
- B. Movement and heart rate variability ✓
- C. Blood glucose and body temperature
- D. Breathing sounds and oxygen levels only
Why: Most wearables rely on accelerometer data (movement) and heart rate or heart rate variability, sometimes with skin temperature. Unlike polysomnography, they do not measure brain waves directly, so their stage estimates are less precise than clinical EEG-based staging.
Module 3
Module 3: Wearable Sensors and Hardware
Wearable Sensors and Hardware
Knowledge check
Q1. Which sensor is most commonly used in wrist-worn sleep trackers to detect movement?
- A. Gyroscope
- B. Accelerometer ✓
- C. Barometer
- D. Magnetometer
Why: Accelerometers measure acceleration along one or more axes, allowing a wearable to detect body movement. Sleep trackers use this movement data to estimate when a person is awake or restless versus still.
Q2. What does photoplethysmography (PPG) measure in a wearable sleep device?
- A. Brain electrical activity
- B. Eye movement
- C. Changes in blood volume under the skin ✓
- D. Skin temperature only
Why: PPG uses light-emitting diodes and photodetectors to track changes in blood volume beneath the skin with each heartbeat. This allows the device to estimate heart rate and heart rate variability, which help infer sleep stages.
Q3. Why do many sleep trackers include a skin temperature sensor?
- A. To measure body fat percentage
- B. To detect hydration levels
- C. Because body temperature typically drops during sleep onset and changes across sleep stages ✓
- D. To measure blood oxygen directly
Why: Core and peripheral body temperature follow a circadian pattern, generally dropping as a person falls asleep. Tracking skin temperature trends helps wearables contextualize sleep timing and stage transitions.
Q4. What is a key limitation of using only accelerometer data to classify sleep stages?
- A. Accelerometers cannot detect any movement
- B. Accelerometers drain the battery too quickly
- C. Lying still may indicate either deep sleep or quiet wakefulness, so movement alone is ambiguous ✓
- D. Accelerometers only work when charging
Why: Stillness is ambiguous: a person can be motionless while awake or in deep sleep. This is why modern devices combine accelerometry with heart rate, HRV, and temperature data to improve stage classification accuracy.
Q5. Which wearable component is typically responsible for storing sensor data before it is synced to a phone or cloud?
- A. Microcontroller with onboard memory ✓
- B. Vibration motor
- C. Display backlight
- D. Charging coil
Why: A microcontroller processes sensor signals and manages onboard flash or other memory, buffering readings until the device connects via Bluetooth or another link to transfer the data for analysis.
Module 4
Module 4: Sleep Tracking Algorithms and Data Processing
Sleep Tracking Algorithms and Data Processing
Knowledge check
Q1. What is the primary signal that most wrist-worn sleep trackers use to estimate sleep stages?
- A. Blood glucose levels measured through the skin
- B. Heart rate and heart rate variability, combined with movement data from an accelerometer ✓
- C. Brainwave activity measured through electrodes in the device
- D. Body temperature measured only at the forehead
Why: Most consumer wearables estimate sleep stages by combining heart rate, heart rate variability, and accelerometer-based movement. Brainwave measurement (EEG) requires specialized electrodes and is not used by typical wrist devices.
Q2. What does an accelerometer primarily contribute to sleep tracking algorithms?
- A. It measures the exact oxygen level in the blood
- B. It detects body movement, which helps distinguish wake periods from sleep ✓
- C. It records the sound of snoring with high accuracy
- D. It calculates the user's total daily calorie burn
Why: An accelerometer detects motion. Periods of little or no movement are typically associated with sleep, while movement often indicates wakefulness, making it a key input for sleep/wake classification.
Q3. Why do sleep tracking algorithms often apply smoothing or filtering to raw sensor data?
- A. To increase the battery life of the device
- B. To remove noise and brief artifacts so that sleep stage transitions are more stable and realistic ✓
- C. To convert the data into a format that can be printed
- D. To hide periods of wakefulness from the user
Why: Raw sensor signals are noisy, and brief movements or sensor artifacts can cause false stage changes. Smoothing and filtering reduce this noise so the output better reflects realistic sleep patterns.
Q4. Which validation method is commonly used to assess how accurately a consumer sleep tracker identifies sleep compared to clinical standards?
- A. Comparing its output against polysomnography (PSG), the gold standard in sleep lab measurement
- B. Asking users to rate how well they slept on a scale of one to ten ✓
- C. Measuring the device's screen brightness under different lighting
- D. Counting the number of app notifications sent each night
Why: Polysomnography, which records brain activity, eye movement, and muscle tone, is the clinical gold standard. Validation studies compare wearable outputs against it. Subjective ratings alone are not an accurate benchmark.
Q5. What is a common limitation of sleep tracking algorithms that rely mainly on movement and heart rate?
- A. They cannot detect sleep at all unless the user is awake
- B. They may struggle to distinguish quiet wakefulness, such as lying still in bed, from light sleep ✓
- C. They always overestimate deep sleep by exactly two hours
- D. They require a constant internet connection to record any data
Why: A person lying still while awake looks similar to sleep in movement data, so algorithms can misclassify quiet wakefulness as sleep. Heart rate variability helps somewhat, but this confusion remains a known limitation.
Module 5
Module 5: Accuracy, Validation, and Limitations
Accuracy, Validation, and Limitations
Knowledge check
Q1. Which sleep stage is most commonly misclassified by consumer wearable devices that rely primarily on accelerometry and heart rate?
- A. Wake
- B. Light sleep (N1/N2)
- C. Deep sleep (N3) ✓
- D. REM sleep
Why: Consumer devices often struggle to distinguish deep sleep (N3) from light sleep because both involve low movement and similar heart rate patterns. EEG, which measures brain wave activity, is needed to reliably identify deep sleep, and wearables without EEG tend to show poorer agreement with polysomnography for this stage.
Q2. What is the gold standard reference method against which sleep tracking devices are typically validated?
- A. Actigraphy
- B. Sleep diary
- C. Polysomnography (PSG) ✓
- D. Pulse oximetry
Why: Polysomnography, which simultaneously records EEG, EOG, EMG, respiration, and heart rhythm in a lab setting, is considered the gold standard for sleep measurement. Validation studies compare wearable outputs against PSG to quantify accuracy for stages, total sleep time, and wake detection.
Q3. Why do wrist-worn sleep trackers often perform well at estimating total sleep time but poorly at estimating wake time after sleep onset?
- A. Wrist sensors cannot detect any movement during the night
- B. People who are awake while lying still look physiologically similar to sleeping people to a motion-based sensor ✓
- C. Battery saving modes turn off the sensor after sleep onset
- D. Wake periods are always shorter than the device sampling interval
Why: Accelerometers detect movement, and a person lying quietly awake shows little motion, so the device may classify it as sleep. This causes overestimation of sleep time and underestimation of wake after sleep onset, a known limitation of actigraphy-based methods.
Q4. Which factor is a common source of reduced accuracy in optical heart rate sensors used for sleep tracking?
- A. 低 ambient room temperature
- B. Skin pigmentation and poor sensor contact or fit ✓
- C. Using the device in airplane mode
- D. Sleeping on the opposite side of the body where the device is worn
Why: Photoplethysmography (PPG) signals are affected by melanin absorption, which can reduce signal quality on darker skin, and by loose fit or movement that disrupts sensor contact. These factors can degrade heart rate and heart rate variability estimates that feed into sleep staging algorithms.
Q5. What does it mean if a sleep tracker has high sensitivity but low specificity for detecting sleep?
- A. It rarely misses true sleep periods but often labels wake periods as sleep ✓
- B. It rarely labels wake as sleep but often misses true sleep periods
- C. It is accurate for both sleep and wake detection
- D. It only works correctly for people over 60 years old
Why: Sensitivity measures the proportion of true sleep epochs correctly identified as sleep, while specificity measures the proportion of true wake epochs correctly identified as wake. High sensitivity with low specificity means the device catches nearly all sleep but over-attributes quiet wakefulness to sleep, which is typical of motion-based trackers.
Module 6
Module 6: Privacy, Data Security, and Ethics
Privacy, Data Security, and Ethics
Knowledge check
Q1. Which type of sleep tracking data is generally considered most sensitive from a privacy standpoint?
- A. Total time spent in bed
- B. Heart rate patterns combined with timestamps of nightly sleep and wake times ✓
- C. The device's battery level
- D. The number of steps taken during the day
Why: Heart rate patterns linked with precise sleep and wake timestamps can reveal health conditions, routines, and habits, making them more sensitive than simpler metrics like battery level or step counts.
Q2. What does the principle of data minimization mean for sleep tracking apps?
- A. Collecting as much data as possible to improve accuracy
- B. Collecting only the data that is necessary for the app's stated purpose ✓
- C. Storing data indefinitely in case it becomes useful later
- D. Reducing the size of data files to save device storage
Why: Data minimization is a core privacy principle requiring organizations to collect only what is needed for a specific, disclosed purpose, reducing risk if the data is breached or misused.
Q3. Why is informed consent important when a sleep tracking app shares data with third parties?
- A. It makes the app load faster on the user's phone
- B. It ensures users understand and agree to how their personal health data will be used and shared ✓
- C. It is only required for paid subscriptions
- D. It replaces the need for encryption
Why: Informed consent means users are clearly told what data is collected, who receives it, and why, allowing them to make an meaningful choice about their personal health information.
Q4. Which security practice best protects sleep data stored on a cloud server?
- A. Storing all data in plain text for easy access
- B. Using end-to-end or strong encryption for data both in transit and at rest ✓
- C. Allowing all employees unrestricted access to user records
- D. Deleting data only when storage space runs out
Why: Encrypting data both during transmission and while stored protects it from interception and unauthorized access, which is essential for sensitive health information.
Q5. A sleep tracker company wants to use anonymized user sleep data to train an AI model. Which ethical concern is most important to address?
- A. Whether the device color matches the brand guidelines
- B. Whether anonymization can be reversed by combining data with other sources, and whether users were told about this use ✓
- C. Whether the AI model will run faster on mobile devices
- D. Whether the company can charge a higher subscription fee
Why: Data that appears anonymous can sometimes be re-identified when combined with other datasets, and using health data for new purposes without transparent disclosure raises serious ethical and legal concerns.
Module 7
Module 7: Using Sleep Data to Improve Health
Using Sleep Data to Improve Health
Knowledge check
Q1. Which sleep stage is most associated with physical restoration and growth hormone release?
- A. Stage 1 (light sleep)
- B. Stage 2 (light sleep)
- C. Slow-wave (deep) sleep ✓
- D. REM sleep
Why: Slow-wave sleep, the deepest non-REM stage, is when the body performs much of its physical repair and when growth hormone is released in larger amounts. Sleep trackers often estimate it from movement and heart rate patterns.
Q2. What does a sleep tracker typically use to estimate whether you are asleep or awake?
- A. Blood oxygen levels only
- B. Body temperature only
- C. Movement (accelerometer) combined with heart rate and sometimes heart rate variability ✓
- D. Brain wave activity measured through the skin
Why: Most consumer wearables rely on an accelerometer to detect movement, along with heart rate and sometimes heart rate variability. Brain activity requires EEG, which consumer wrist devices do not measure, so their sleep staging is an estimate.
Q3. Why is it useful to look at trends in your sleep data rather than a single night's score?
- A. Single nights are always recorded incorrectly
- B. Trends reveal patterns in habits and recovery that one night can hide, such as the effect of late caffeine ✓
- C. Trends show exactly how many hours of REM you need each night
- D. Single-night scores are legally required to be ignored
Why: One night can be affected by a stressful day or an unusual meal, so a single score is noisy. Weekly trends in total sleep, bedtime consistency, and resting heart rate are more useful for spotting habits that help or hurt your sleep.
Q4. Which change is a commonly recommended, evidence-based way to use sleep data to improve sleep?
- A. Going to bed at irregular times to match what the tracker suggests
- B. Keeping a consistent sleep and wake schedule, including on weekends ✓
- C. Sleeping in for several hours every weekend to make up for lost sleep
- D. Stopping all tracking altogether to avoid any anxiety about data
Why: A consistent schedule supports the body's circadian rhythm and tends to improve sleep quality. Irregular timing, including large weekend shifts, can disrupt that rhythm and is sometimes called social jetlag.
Q5. A sleep tracker shows you consistently getting less deep sleep after drinking alcohol in the evening. What is the most reasonable interpretation?
- A. Alcohol has no effect on sleep and the tracker is malfunctioning
- B. Alcohol may be disrupting your sleep architecture, so you could test reducing evening drinking and compare results ✓
- C. Alcohol increases deep sleep, so you should drink more before bed
- D. The tracker can only measure sleep, not what affects it, so the data is meaningless
Why: Alcohol often suppresses REM sleep and fragments the second half of the night. Using your own data to test a change, such as drinking less in the evening, and then comparing trends is a practical way to apply sleep tracking to health decisions.