How Much REM Sleep Do You Need by Age?

How much REM sleep should you get each night?

For most adults, rapid eye movement (REM) sleep commonly represents about 20%–25% of total sleep. For someone sleeping seven to nine hours, that works out to roughly 84–135 minutes.

Age changes the picture most dramatically during early childhood. Newborns spend much more of their sleep in an REM-like state, while children reach a sleep architecture closer to adult patterns within the first few years of life.

There is an important limitation, however: medical organizations do not prescribe a specific REM-minute target for every age.

Age-based REM numbers are best treated as developmental reference points rather than nightly quotas.

This guide explains how REM sleep changes across the lifespan, how percentages translate into minutes, why REM naturally varies from night to night, and why consumer sleep-stage estimates should be interpreted as trends rather than exact clinical measurements.

Quick Answer: How Much REM Sleep Do You Need by Age?

The largest age-related change happens early in life.

Newborns may spend around 50% of sleep in active or REM-like sleep.
By around age 1, REM may account for roughly 30% of sleep.
By about age 2–3, REM approaches roughly 20%–25%.
Children, teenagers, and adults generally remain near this broad proportion, with considerable individual variation.
Older adults may show somewhat less REM in some studies, although the age-related change is modest and not consistently linear.

For most adults, a practical reference is approximately:

20%–25% REM sleep

If you sleep:

Total Sleep	20% REM	25% REM
6 hours	72 minutes	90 minutes
7 hours	84 minutes	105 minutes
8 hours	96 minutes	120 minutes
9 hours	108 minutes	135 minutes
10 hours	120 minutes	150 minutes

These calculations illustrate typical sleep architecture. They are not clinical minimums.

REM Sleep by Age: Reference Percentage and Minutes

The following table combines developmental REM-sleep research with commonly used age-based total sleep-duration ranges.

The REM percentages are reference values, and the minutes are estimates calculated from total sleep time. They should not be interpreted as prescribed REM requirements.

Age	Typical REM Reference	Common Total Sleep Reference	Approximate REM Minutes
Newborn, 0–3 months	About 50% active / REM-like sleep	About 14–17 hours per 24 hours	About 420–510 min/day
Around 1 year	About 30%	About 12–16 hours per 24 hours	About 216–288 min/day
1–2 years	Declining toward 20%–25%	About 11–14 hours per 24 hours	Roughly 132–210 min/day
3–5 years	About 20%–25%	About 10–13 hours per 24 hours	About 120–195 min/day
6–12 years	About 20%–25%	About 9–12 hours per 24 hours	About 108–180 min/day
13–18 years	About 20%–25%	About 8–10 hours per 24 hours	About 96–150 min/day
Adults	About 20%–25%	Commonly 7–9 hours	About 84–135 min/night
Older adults	Often still near 20%, with wider individual variation	Example: 7–8 hour night	Roughly 84–120 min/night at 20%–25%

The early-life rows require additional caution because infant sleep architecture differs from mature adult sleep. In newborns, researchers commonly use the term active sleep, which develops into mature REM sleep as the nervous system and EEG sleep patterns develop.

There Is No Official REM Sleep Requirement by Age

Total sleep recommendations are well established for different age groups.

Stage-specific requirements are much less precise.

For example, health organizations can recommend that a teenager sleep approximately eight to ten hours, but they do not prescribe:

“A 15-year-old must get exactly 118 minutes of REM sleep.”

The reason is that sleep architecture varies with:

Individual biology
Total sleep duration
Sleep timing
Previous sleep loss
Stress
Exercise
Alcohol
Medication
Sleep disorders
Measurement method

Use age-based numbers to understand broad physiology rather than to grade every night.

Reference Is Different From Requirement

This distinction is central to interpreting REM sleep.

A reference value

Describes what researchers commonly observe in healthy populations.

A requirement

Would define how much REM an individual must obtain for optimal health.

Research provides much stronger evidence for the first than the second.

That is why a statement such as:

“Healthy adults commonly spend around 20%–25% of sleep in REM”

is more defensible than:

“Every adult needs exactly two hours of REM.”

What Is REM Sleep?

REM stands for rapid eye movement.

It is one of the two major categories of sleep, alongside non-rapid eye movement (NREM) sleep.

During REM:

Brain activity becomes relatively high.
Rapid eye movements occur.
Most skeletal muscles become strongly inhibited.
Breathing becomes more variable.
Heart rate can become more variable.
Dreaming commonly occurs.

REM sleep has been associated with processes involving memory, learning, emotional processing, and brain development.

It is one part of a complete sleep architecture that also includes N1, N2, and deep N3 sleep.

For a broader explanation, see what light, deep, and REM sleep mean.

Why Babies Have So Much REM-Like Sleep

Early brain development is extraordinarily active.

Full-term newborns may spend approximately half of their total sleep in active sleep, the developmental precursor of mature REM sleep.

This is dramatically different from adult sleep.

Newborn sleep also differs structurally:

Sleep cycles are shorter.
Sleep is spread across multiple periods in 24 hours.
Infants may enter active/REM-like sleep directly.
Mature NREM-stage features are still developing.

As the nervous system matures, NREM sleep becomes more organized and the proportion of REM-like sleep declines.

How REM Changes During the First Year

The first year contains one of the fastest changes in human sleep architecture.

A simplified developmental sequence looks like:

Developmental Stage	Approximate REM / Active Sleep Pattern
Full-term newborn	Around 50% of total sleep
First several months	REM-like sleep gradually decreases as NREM develops
Around 1 year	Approximately 30% in some normative datasets
Around 2–3 years	Approaches approximately 20%–25%

This developmental transition is much larger than most REM changes that occur later in adulthood.

Why Infant REM Minutes Are So High

Two factors occur simultaneously:

Infants sleep many more hours per day.
A much larger proportion of that sleep is active/REM-like sleep.

If a newborn sleeps 16 hours over 24 hours and approximately half is active sleep, the arithmetic works out to roughly:

8 hours, or 480 minutes, of active/REM-like sleep

That number should be understood as developmental physiology rather than a target parents should try to optimize manually.

REM Sleep in Toddlers

By the toddler years, sleep architecture becomes much closer to the pattern seen later in life.

At around age 2, REM commonly represents approximately 20%–25% of total sleep.

Total sleep remains much longer than in adults, however.

A toddler sleeping 12 hours may therefore spend approximately:

144–180 minutes

in REM if 20%–25% of that sleep is classified as REM.

The longer REM duration compared with many adults is driven largely by longer total sleep.

REM Sleep in Preschool Children

Children ages 3–5 commonly need substantially more total sleep than adults.

If total sleep is approximately 10–13 hours and REM is around 20%–25%, the mathematical range is approximately:

120–195 minutes of REM

Again, this is a calculated reference range rather than a medical requirement.

Napping can also contribute to total 24-hour sleep at younger ages, so nighttime REM minutes alone do not capture the complete sleep period.

REM Sleep in School-Age Children

By school age, the percentage of REM sleep is broadly similar to later life.

What remains different is total sleep duration.

Children ages 6–12 are generally recommended to obtain more total sleep than teenagers and adults.

If total sleep is 9–12 hours and REM occupies approximately 20%–25%, that produces roughly:

108–180 minutes

across the full sleep period.

REM Sleep in Teenagers

Teenagers generally need around eight to ten hours of total sleep.

A 20%–25% REM reference would correspond to approximately:

96–150 minutes

The more important issue for many teenagers is obtaining enough total sleep opportunity.

A short school-night sleep window can reduce the later portion of sleep, where REM episodes tend to become longer.

Why Waking Too Early Can Reduce REM Disproportionately

REM is distributed unevenly across the night.

Early sleep cycles contain more NREM and deep sleep.

As the night progresses:

Deep N3 sleep generally becomes less prominent.
REM episodes become progressively longer.

This means the final one or two hours of a full night can contain substantial REM sleep.

Cutting an eight-hour sleep opportunity down to six hours removes more than two hours of generic sleep time. It can remove a particularly REM-rich portion of the night.

Seven Hours vs Eight Hours: Why the Extra Hour Matters

Consider two hypothetical nights.

Night A: 8 Hours

The sleeper completes several full cycles and remains asleep through longer later-night REM periods.

Night B: 7 Hours

The final hour is removed.

The total sleep reduction is 12.5%, but the reduction in REM minutes could be proportionally greater because REM becomes more concentrated late in the sleep period.

This is one reason total sleep opportunity matters before trying to optimize a stage percentage.

How Much REM Sleep Do Adults Need?

Healthy adults commonly spend approximately 20%–25% of total sleep in REM.

For common nightly sleep durations:

Total Sleep	Approximate REM at 20%	Approximate REM at 25%
7 hours	84 min	105 min
7.5 hours	90 min	113 min
8 hours	96 min	120 min
8.5 hours	102 min	128 min
9 hours	108 min	135 min

This is why consumer sleep articles often describe adult REM as approximately 90–120 minutes.

The wider 20%–25% calculation shows that healthy-looking nights can reasonably fall outside that narrower range depending on total sleep time.

Is 90 Minutes of REM Enough?

It can be completely compatible with a normal adult sleep pattern.

For example:

7.5 hours total sleep × 20% REM = 90 minutes

A person sleeping nine hours may naturally accumulate more REM minutes.

The number should therefore be interpreted relative to:

Total sleep time
REM percentage
Recent baseline
Night-to-night variation
How sleep was measured

Is Two Hours of REM Normal?

Yes, two hours can fit a normal adult night.

For example:

8 hours × 25% REM = 120 minutes

A longer sleeper could accumulate even more.

There is no reason to force REM downward simply because it exceeds 90 minutes.

Is One Hour of REM Too Little?

Context determines how useful the number is.

Sixty minutes represents:

About 14% of a 7-hour night
About 12.5% of an 8-hour night

That is below the common adult 20%–25% reference.

One night can still occur because of:

Natural night-to-night variation
Short sleep
Early awakening
Alcohol
Stress
Sleep fragmentation
Medication
Stage-classification error

A repeated pattern matters more than one isolated wearable estimate.

How Much REM Sleep Do Older Adults Need?

There is no separate official REM-minute requirement for older adults.

Research on adult aging is more nuanced than the common statement that REM simply decreases year after year.

Some large lifespan analyses have found a modest decrease in REM percentage from young to later adulthood.

Other systematic analyses of healthy adults have found little or no statistically significant age-related change in the percentage of REM sleep.

A useful practical reference remains around one-fifth of total sleep, while recognizing wider individual variation.

Adult REM Percentage Changes Less Than Many People Expect

One analysis of adult sleep studies estimated approximately:

Age	Estimated REM Percentage
19	About 21.7%
40	About 21.2%
75	About 18.8%
85	About 20.4%

The pattern is not a simple straight decline.

This illustrates why age-adjusted REM interpretation should use broad references rather than strict decade-by-decade targets.

REM Percentage vs REM Minutes

These two measurements answer different questions.

REM Percentage

Shows how much of your total sleep was classified as REM.

REM Minutes

Shows the absolute amount of estimated REM sleep.

Both matter because total sleep duration can change the interpretation.

Example: Same REM Percentage, Different REM Time

	Night A	Night B
Total sleep	8 hours	6 hours
REM percentage	22%	22%
REM time	106 min	79 min

The stage proportion stayed exactly the same.

Absolute REM duration fell by about 27 minutes because total sleep was shorter.

This is why increasing total sleep opportunity can be more important than chasing the REM percentage itself.

Example: Same REM Minutes, Different Percentage

	Night A	Night B
Total sleep	7 hours	9 hours
REM time	100 min	100 min
REM percentage	23.8%	18.5%

The REM minutes are identical.

The percentage changes because total sleep changes.

Neither metric should be interpreted without the other.

The Percentage → Minutes → Total Sleep Framework

When reviewing REM, use this order:

1. Percentage

How much of the sleep period was estimated as REM?

2. Minutes

How much absolute REM time did you accumulate?

3. Total Sleep

Was the night long enough to allow the later REM-rich cycles to occur?

This prevents a percentage from hiding a short total night.

Why REM Naturally Changes From Night to Night

Sleep architecture is dynamic.

Even under healthy conditions, REM can change from one night to the next because of:

Total sleep duration
Bedtime and wake time
Recent sleep debt
Stress
Alcohol
Exercise
Illness
Medication
Night awakenings

A wearable graph should therefore show some variation.

A perfectly identical REM percentage every night is not the goal.

One Low REM Night Usually Needs Context

Suppose your recent REM pattern is:

Night	REM
Monday	104 min
Tuesday	112 min
Wednesday	96 min
Thursday	71 min
Friday	109 min

If Thursday was also a short or fragmented night and REM returned to its usual range afterward, that single value carries much less weight.

Persistent REM Changes Matter More

Review the broader pattern if:

Estimated REM stays substantially below your usual range for many nights.
Total sleep is consistently adequate but the stage pattern remains different.
Sleep is increasingly fragmented.
You repeatedly wake unrefreshed.
Daytime sleepiness is increasing.
Other sleep or respiratory signals also change.

Wearable data can identify a trend. It cannot determine the medical cause.

Why Late Wake Time Can Increase REM Minutes

Later sleep cycles contain longer REM periods.

If you normally wake at 6:00 a.m. but sleep until 8:00 a.m. after sleep restriction, the extra time can include substantial REM.

This contributes to the phenomenon sometimes described as REM rebound after previous sleep loss or REM suppression.

A higher REM percentage after recovery sleep does not automatically mean something is wrong.

Sleep Debt Can Change REM Architecture

Sleep pressure responds to previous sleep history.

After insufficient sleep, the body does not simply distribute every recovered minute evenly across sleep stages.

The architecture of recovery sleep can change according to:

How much sleep was lost
Which part of the night was lost
Recent REM suppression
Circadian timing

This is another reason one night's percentage should not become a permanent target.

Alcohol Can Reduce REM

Alcohol can alter sleep architecture.

Acute evening alcohol exposure commonly suppresses REM during the earlier part of the night and can lead to changing sleep patterns as alcohol is metabolized.

You may see:

Lower estimated REM
Higher sleeping heart rate
Lower HRV
More fragmented later-night sleep

The combination provides more useful context than the REM number alone.

Caffeine Can Affect REM Indirectly Through Sleep Timing

Late caffeine can delay sleep onset or shorten total sleep opportunity.

Because REM becomes more concentrated later in the night, a delayed bedtime combined with a fixed wake time can reduce REM minutes simply by shortening the sleep period.

This is an example of why total sleep timing should be reviewed before trying to change a specific stage.

Stress Can Change Sleep Architecture

Stress can affect:

Sleep latency
Night waking
Heart rate
HRV
Sleep-stage distribution

A stressful night may therefore produce a different REM estimate even when bedtime is unchanged.

Exercise and REM Sleep

Exercise can support overall sleep health over time.

The immediate effect on REM varies with:

Exercise intensity
Timing
Training load
Previous sleep
Individual fitness

A single hard workout does not guarantee either more or less REM that night.

Look for repeated patterns across similar training days.

Why Consumer Sleep Trackers Cannot Measure REM the Same Way as a Sleep Lab

Clinical sleep staging uses polysomnography, or PSG.

A clinical sleep study can record signals such as:

EEG brain activity
Eye movements
Muscle activity
Heart rhythm
Breathing
Blood oxygen

REM is clinically scored using these direct physiological signals.

How Wearables Estimate REM

Consumer wearables usually work from a smaller sensor set.

Depending on the device, sleep algorithms may use combinations of:

Movement
Optical pulse signals
Heart rate
HRV
Breathing-related information

The algorithm then estimates whether a period is most compatible with:

Wake
Light sleep
Deep sleep
REM sleep

This makes consumer sleep staging useful for longitudinal tracking, with expected classification error.

Sleep/Wake Detection Is Easier Than Exact Stage Detection

A person who is moving around has a strong signal of wakefulness.

Distinguishing quiet light sleep from REM is more difficult because both can involve relatively little movement and overlapping cardiovascular patterns.

Validation studies comparing consumer wearables with PSG continue to find misclassification between sleep stages.

This means a displayed REM duration such as:

87 minutes

should be interpreted as an estimate.

Why You Should Avoid Chasing Exact REM Minutes

Suppose your wearable reports:

89 minutes on Monday

and:

101 minutes on Tuesday

The 12-minute difference could contain:

Real biological variation
A longer sleep opportunity
Different sleep-stage distribution
Different sensor quality
Algorithmic classification variation

The data does not justify treating Tuesday as automatically healthier.

The Stage Estimate → Trend → Context Framework

A practical way to read wearable REM is:

Stage Estimate

Start with the displayed REM percentage and minutes.

Trend

Review several nights rather than one result.

Context

Add:

Total sleep
Bedtime
Wake time
Night waking
Alcohol
Caffeine
Stress
Training

This gives the estimated stage a useful physiological context.

Use a 7-Night REM Trend

A weekly pattern reduces the influence of one unusual night.

Track:

Metric	What to Review
REM minutes	Typical weekly amount
REM percentage	Typical share of total sleep
Total sleep	Was sleep opportunity sufficient?
Wake time	Was the REM-rich final portion shortened?
Night waking	Was sleep fragmented?
Next-day state	Did you feel rested and alert?

Seven Nights Are Better Than One, but Longer Trends Are Better Still

A single week may contain unusual travel, training, work stress, or illness.

Four weeks provide a stronger personal reference.

Compare:

This week vs. last week
Workdays vs. weekends
Alcohol-free vs. drinking nights
Normal sleep vs. shortened sleep
High-stress vs. lower-stress periods

Repeated patterns become more meaningful as the dataset grows.

Do Not Compare REM Percentages Across Different Devices Too Aggressively

Different sleep systems can use different:

Sensors
Sampling intervals
Artifact filters
Stage algorithms
Sleep-window definitions

One device may classify a 20-minute period as REM while another labels part of it light sleep.

If you switch devices, establish a new baseline before interpreting small differences.

Why Total Sleep Is Usually the First Metric to Check

If REM is low because you slept only five hours, the most direct issue is insufficient sleep opportunity.

Trying to “increase REM percentage” while continuing to sleep five hours addresses the wrong part of the problem.

Start with:

Total sleep opportunity
Sleep timing
Sleep continuity
Then stage distribution

Example: Low REM Caused by a Short Night

	Normal Night	Short Night
Total sleep	8 h	5.5 h
REM percentage	22%	19%
REM minutes	106 min	63 min
Wake time	Normal	2.5 hours early

The largest difference is the shortened sleep opportunity.

The early wake time also removed a REM-rich portion of the night.

Example: Low REM Percentage With Enough Total Sleep

Now consider:

	Recent Baseline	Current Night
Total sleep	8 h	8 h 10 min
REM percentage	21%–24%	13%
REM minutes	101–115 min	64 min
Night waking	Low	High

Here, the change is less easily explained by total sleep duration.

Review fragmentation, alcohol, stress, medication, illness, and whether the pattern repeats.

When a Low REM Trend Deserves More Attention

Consider a broader sleep review when several factors occur together:

REM estimates remain substantially below your established range.
Total sleep is adequate.
Sleep remains fragmented.
You consistently wake unrefreshed.
Daytime sleepiness or concentration problems are increasing.
Breathing or SpO2 trends also look unusual.

A healthcare professional can evaluate symptoms and determine whether formal sleep testing is appropriate.

Can You Have Too Much REM Sleep?

A higher-than-usual REM percentage on one night is not automatically concerning.

Possible explanations include:

Longer sleep duration
Recovery after previous REM restriction
Sleeping later into the morning
Normal biological variation
Wearable stage-estimation error

Persistent unusual sleep patterns require context rather than one universal upper REM threshold.

Is More REM Always Better?

Sleep health depends on the complete architecture.

A healthy night includes:

N1
N2
N3 deep sleep
REM

Maximizing one stage is not the goal.

Better questions include:

Did I sleep long enough?
Was the timing consistent?
Was sleep reasonably continuous?
Did I move through multiple cycles?
Do my stage trends remain near my normal pattern?
How do I feel during the day?

REM Sleep and Deep Sleep Are Both Important

REM and deep N3 sleep have different physiological characteristics.

	REM Sleep	Deep N3 Sleep
Typical timing	More prominent later in the night	More prominent earlier in the night
Brain activity	Relatively active	Dominated by slow-wave activity
Muscle tone	Strongly inhibited	Reduced but present
Heart rate	More variable	Generally slower and steadier
Breathing	More irregular	More regular

Neither stage needs to “beat” the other in a sleep score.

How RingConn Helps Track REM Trends

RingConn estimates sleep stages as part of its overnight wellness tracking.

Useful sleep information can include:

Total sleep time
Sleep timing
Estimated light sleep
Estimated deep sleep
Estimated REM sleep
Night waking
Heart rate
HRV
SpO2
Respiratory rate

The strongest use is repeated within-device comparison.

Instead of treating one REM number as a pass-or-fail score, ask:

Is my REM trend close to my recent range?
Did total sleep become shorter?
Did I wake earlier than usual?
Did alcohol or stress change?
Was the entire night more fragmented?

The RingConn App guide explains how to review sleep-stage estimates alongside heart rate, HRV, and other wellness trends.

Consumer Sleep Stages Are Best Used as Trends

Exact clinical sleep staging requires polysomnography.

RingConn provides consumer wellness estimates designed to help users understand repeated sleep patterns.

That makes the most useful comparison:

your recent RingConn nights vs. your own previous RingConn nights

rather than:

one wearable stage value vs. an exact laboratory reference.

Our guide to why wearable sleep scores and sleep stages can differ explains how sensor data and algorithms affect stage estimates.

How to Review Your REM Data in Five Steps

1. Check Total Sleep First

Was there enough sleep opportunity for several complete cycles?

2. Check REM Minutes

How much estimated REM did the night contain?

3. Check REM Percentage

What share of total sleep was classified as REM?

4. Compare With Your Recent Baseline

Review at least several nights rather than one isolated result.

5. Add Context

Consider:

Bedtime
Wake time
Alcohol
Caffeine
Stress
Training
Travel
Illness

What Is a Good REM Sleep Percentage on a Wearable?

For adults, approximately 20%–25% is a useful population reference.

Your wearable percentage should still be interpreted as an estimate.

A value of:

18%, 21%, 24%, or 27%

on an individual night can occur without automatically indicating poor or abnormal sleep.

Look at your established range and multi-night trend.

Should You Try to Increase REM Sleep?

Start by improving the conditions that support a complete night of sleep.

Useful priorities include:

Allow enough total sleep time.
Keep a reasonably consistent schedule.
Avoid repeatedly cutting the final portion of sleep short.
Limit alcohol close to bedtime.
Manage late caffeine according to your sensitivity.
Address repeated night waking.
Maintain regular physical activity.

The goal is healthier overall sleep architecture rather than forcing one stage upward.

Protect the Final Part of the Night

Because REM periods become longer later in sleep, repeated early waking can be especially relevant.

For example:

Going to bed at midnight and waking at 5:30 a.m. can remove REM-rich cycles.
Going to bed earlier while maintaining the same wake time provides more opportunity for complete sleep cycles.

Total sleep duration remains the first priority.

What If Your REM Percentage Looks Good but You Are Still Tired?

A normal-looking REM percentage does not guarantee a fully restorative night.

Review:

Total sleep
Sleep efficiency
Night waking
Sleeping heart rate
HRV
Respiratory trends
How long daytime fatigue has persisted

A person can obtain 22% estimated REM while still sleeping too little overall or experiencing fragmented sleep.

What If Your REM Is Low but You Feel Great?

One low wearable estimate can occur while you feel completely normal the next day.

Check whether:

The device had good overnight contact.
Total sleep was normal.
Other sleep metrics remained close to baseline.
The REM estimate returns toward normal on subsequent nights.

A multi-night pattern provides a stronger signal than one chart.

How Long Should You Track REM Before Judging the Trend?

A useful starting point is at least 7 nights.

Two to four weeks provides stronger context because it captures more variation in:

Weekdays and weekends
Training
Work stress
Sleep schedules
Social routines

This trend-based approach is consistent with the broader principle of building a personal wearable baseline.

Use RingConn Gen 3 for Long-Term Adult Sleep Trends

RingConn Gen 3 supports continuous day-and-night wellness monitoring and provides sleep-stage estimates alongside heart rate, HRV, SpO2, respiratory rate, activity, and other supported signals.

For adult users, this creates a broader context around REM trends.

Users interested in long-term sleep and wellness monitoring can explore RingConn Gen 3.

When Wearable REM Changes Deserve Medical Context

A wearable stage estimate alone cannot diagnose a sleep disorder.

Consider discussing sleep with a healthcare professional when you repeatedly experience:

Severe daytime sleepiness
Persistent inability to feel refreshed despite adequate sleep opportunity
Frequent prolonged awakenings
Repeated loud snoring or observed breathing interruptions
Unusual movements or behaviors during sleep
Major unexplained changes in sleep quality

Clinical evaluation may include history, examination, and appropriate sleep testing.

REM Sleep by Age Checklist

Newborns spend far more sleep in active/REM-like sleep than adults.
REM falls rapidly during infancy and approaches roughly 20%–25% by early childhood.
Children obtain more REM minutes mainly because they sleep longer overall.
Teenagers and adults commonly spend around 20%–25% of sleep in REM.
Adult aging produces much smaller REM-percentage changes than early development.
There is no official age-by-age minimum REM requirement.
Always read REM minutes alongside total sleep time.
Remember that later-night sleep contains longer REM periods.
Avoid judging one low or high REM night.
Consumer sleep-stage values are estimates rather than clinical PSG measurements.
Compare trends using the same wearable over multiple nights.

Final Takeaway

REM sleep changes dramatically during early life and then becomes much more stable as a proportion of total sleep.

A useful developmental summary is:

Newborn: about 50% active/REM-like sleep
Around 1 year: about 30%
Around 2–3 years: approaching 20%–25%
Older children and adults: commonly around 20%–25%

Adult aging can modestly change sleep architecture, but REM percentage does not simply decline in a straight line with every decade.

The more useful framework is:

Age Reference → Total Sleep → REM Percentage → REM Minutes → Multi-Night Trend

Percentage and minutes should always be interpreted together.

An adult sleeping eight hours with 22% REM has about 106 minutes of REM. The same 22% during a six-hour night provides only about 79 minutes.

Total sleep opportunity therefore matters greatly, especially because REM becomes more concentrated in the later part of the night.

Consumer wearables add another layer of uncertainty. Sleep-stage algorithms estimate REM from indirect signals such as movement and cardiovascular patterns, while clinical polysomnography directly measures brain, eye, and muscle activity.

Use wearable REM as a longitudinal estimate:

one night → context
one week → pattern
several weeks → stronger personal baseline

RingConn can support this trend-based approach by placing estimated REM alongside total sleep, heart rate, HRV, respiratory rate, SpO2, and other supported wellness information.

RingConn products are intended for personal health and wellness awareness and are not medical devices. Sleep-stage estimates, heart rate, HRV, SpO2, respiratory rate, and other RingConn wellness information should not replace polysomnography, professional medical advice, diagnosis, emergency assessment, or treatment.

FAQ: How Much REM Sleep Do You Need by Age?

How much REM sleep should an adult get?

Healthy adults commonly spend around 20%–25% of total sleep in REM. For a seven-to-nine-hour night, that works out to approximately 84–135 minutes. There is no official minimum REM-minute requirement that applies to every adult.

How much REM sleep should children get?

REM proportion falls rapidly during early childhood and approaches approximately 20%–25% by around age 2–3. Children still tend to accumulate more REM minutes than adults because their total sleep requirement is longer.

Do babies need more REM sleep?

Newborns spend about half of their sleep in active or REM-like sleep, reflecting very different early-life sleep architecture. The proportion decreases rapidly as the brain and NREM sleep patterns mature.

Do older adults need less REM sleep?

There is no established recommendation that older adults should aim for a lower REM amount. Some research shows a modest age-related reduction in adult REM percentage, while other analyses find little significant change. Total sleep duration and fragmentation often change more noticeably with age.

Is 90 minutes of REM sleep enough?

Ninety minutes can fit a normal adult sleep pattern. For example, 90 minutes is 20% of a 7.5-hour night. Interpret the value with total sleep duration, your recent REM baseline, and how consistently the pattern appears.

Why do I get more REM toward morning?

REM episodes become progressively longer across successive sleep cycles. Deep NREM sleep is concentrated more heavily earlier in the night, while REM becomes more prominent later. Waking earlier than usual can therefore disproportionately reduce REM time.

Can a wearable accurately measure REM sleep?

Consumer wearables can estimate REM using movement, optical pulse, heart rate, HRV, and related signals. Validation studies show useful trend information but also meaningful stage-classification errors compared with polysomnography. Use repeated trends rather than treating exact nightly minutes as clinical measurements.

When should low REM sleep concern me?

One low wearable estimate usually needs context. Persistent low REM estimates alongside adequate total sleep, repeated sleep fragmentation, severe daytime sleepiness, breathing concerns, or consistently unrefreshing sleep may justify discussion with a healthcare professional.