Can People Tell I'm High? Signs of Cannabis Use Others Actually Notice

Lifestyle & Identity

Red Eyes + Smell

Red eyes from vasodilation and smell are the two most detectable signs of cannabis use regardless of tolerance, while non-users are consistently the most accurate at noticing behavioral changes.

Berchansky et al., Clinical Toxicology, 2026

Berchansky et al., Clinical Toxicology, 2026

View as image

This is one of the most common anxieties among cannabis users, from first-time consumers to daily veterans. The answer depends on several factors: how much you consumed, your tolerance level, who is observing you, and which specific signs they know to look for. Research on cannabis intoxication cues, drug recognition training, and social perception provides a surprisingly detailed picture of what is actually detectable and what is mostly in your head.

The Physiological Signs You Cannot Fully Control

Detection

How Detectable Are You? Sign-by-Sign

Likelihood that a non-user observer would notice each sign

Physical (no tolerance fade)

Behavioral (fades with tolerance)

Smell on clothes/breathAlways visible

95%

Red / bloodshot eyesAlways visible

88%

Delayed responsesTolerance helps

70%

Dry mouth effectsAlways visible

60%

Inappropriate laughterTolerance helps

55%

Wandering conversationTolerance helps

50%

Appetite increaseTolerance helps

35%

Anxiety / paranoiaTolerance helps

30%

Your nose adapts within minutes — you cannot reliably assess your own smell. Ask a non-user.

Non-users and former users are the most likely to notice

Cannabis High Detectability Signs

Certain physical effects of THC are directly caused by its pharmacological action and occur regardless of your experience level, setting, or mindset. These are the hardest to conceal because they are not behavioral. They are biological.

Conjunctival injection (red eyes). THC acts as a vasodilator, causing blood vessels to relax and expand. The blood vessels in the conjunctiva, the thin membrane covering the white of the eye, become engorged and visible, producing the characteristic red or bloodshot appearance. This effect is dose-dependent. A low dose may produce barely noticeable reddening, while a high dose can make the eyes dramatically bloodshot.

Eye redness typically begins within fifteen to thirty minutes of inhalation and can persist for two to four hours. It occurs with all routes of administration including edibles, though the onset with edibles is delayed to match the overall absorption timeline. Importantly, this effect does not diminish significantly with tolerance. Daily users still experience vasodilation; they may simply be more accustomed to ignoring it.

Over-the-counter eye drops containing tetrahydrozoline or naphazoline, marketed as redness relievers, constrict the conjunctival blood vessels and are reasonably effective at reducing visible redness. They work within one to five minutes and last two to four hours. However, they do not fully eliminate redness from heavy use, and regular use of these drops can cause rebound redness when discontinued.

Dry mouth (xerostomia). THC binds to CB1 receptors in the submandibular salivary glands, reducing saliva production. This produces the classic cottonmouth that makes speaking, swallowing, and eating feel noticeably different. While others cannot see dry mouth directly, its effects on your speech, including slightly altered articulation, frequent lip-licking, and constant drinking, are observable to attentive people.

Altered pupil response. Research findings on cannabis and pupil size are mixed, but some studies suggest THC can cause mild pupil dilation in low-light conditions and slightly impaired pupillary response to light changes. This is subtle and rarely noticed in social settings, but it is detectable with careful observation.

Changes in heart rate. THC increases heart rate by fifteen to fifty beats per minute in the first hour after use. You cannot see someone's heart rate, but the downstream effects, including visible pulse at the neck in thin individuals and subjective feelings of anxiety that affect behavior, can be indirectly observable.

The Behavioral Signs That Vary By Person

Unlike the physiological signs above, behavioral indicators of cannabis intoxication vary enormously between individuals and are heavily modulated by tolerance, setting, and dose.

Delayed reaction time. THC impairs processing speed and reaction time in a dose-dependent manner. In conversation, this manifests as slightly longer pauses before responding, occasional mid-sentence derailment where you lose the thread of what you were saying, and a general sense of being half a beat behind the conversational flow. This is more noticeable to people who know your normal cadence well.

Tangential thinking. Cannabis affects working memory and executive function, making it easier to drift from one thought to another without maintaining a clear through-line. You might start telling a story, branch into a related thought, and never return to the original point. To others, this can look like scattered thinking or difficulty focusing.

Inappropriate laughter. THC modulates activity in brain regions involved in humor processing and emotional regulation, lowering the threshold for finding things funny. Laughing at things that are not particularly amusing, or being unable to stop laughing once you start, is a well-recognized sign. This is especially noticeable in formal or serious settings where your reaction is incongruent with the social context.

Psychomotor changes. At moderate to high doses, cannabis can affect fine motor coordination, gait, and balance. Fumbling with objects, moving slightly more carefully or deliberately than usual, or having a subtly altered walk are observable to careful observers.

Speech patterns. Beyond the effects of dry mouth, THC can slow speech rate, increase the frequency of filler words like "um" and "uh," and affect prosody, the musical quality of speech. Some users speak more quietly while high; others become unusually verbose.

Who Notices and Who Does Not

The detectability of cannabis intoxication depends as much on the observer as on the user.

Non-users are generally the most perceptive observers of cannabis cues. They are not habituated to the smell, behavioral patterns, or physical signs. A non-user is likely to notice red eyes, cannabis odor, or behavioral changes that would not register with a regular user. Research on drug recognition indicates that people who have never used a substance are often more attuned to its effects in others than people who use it regularly.

Former users are also highly perceptive because they have personal experience with the signs and are no longer habituated to them. A former daily user can often identify cannabis intoxication with remarkable accuracy because they recognize the patterns from their own experience.

Regular users are often the least reliable detectors, particularly in social settings where cannabis use is common. They are habituated to the smell, they are familiar with the behavioral patterns, and their own potential intoxication reduces their observational acuity. The fact that your stoner friends cannot tell does not mean your boss or your parents would have the same difficulty.

People who know you well detect behavioral changes more readily than strangers because they have a strong baseline for your normal behavior. A subtle change in your conversational pace, a slight difference in your emotional expressiveness, or an unusual pattern of eye contact stands out against the background of how you normally present. A stranger interacting with you for the first time has no baseline and may simply attribute any unusual behavior to personality.

What Research Says About Detection Accuracy

Studies examining the accuracy of cannabis intoxication detection reveal some interesting patterns.

Drug recognition experts, law enforcement officers trained specifically in identifying drug impairment, correctly identify cannabis intoxication at rates significantly above chance but far below certainty. In controlled studies, accuracy rates typically range from 60 to 80 percent, varying with the dose and the time since consumption. This means that even trained observers miss cannabis intoxication 20 to 40 percent of the time.

Untrained observers perform worse. In social psychology studies where participants are asked to identify which members of a group have consumed cannabis, accuracy hovers around 50 to 65 percent for moderate doses, only modestly above chance. However, accuracy increases substantially at higher doses and when the observer knows the person well.

The most commonly cited cue in detection studies is eye appearance, followed by smell, then behavioral indicators. When smell is removed as a factor, either through experimental design or because the person consumed edibles, detection accuracy drops significantly.

The Tolerance Factor

Regular cannabis users develop tolerance to many of the behavioral effects of THC. The same dose that would produce obvious impairment in an occasional user may produce minimal behavioral change in a daily user. This is a real pharmacological phenomenon: chronic CB1 receptor activation leads to receptor downregulation and desensitization, requiring more THC to produce the same effect.

However, tolerance development is not uniform across all effects. Behavioral tolerance, the ability to function normally despite intoxication, develops faster than physiological tolerance. A daily user may have a nearly normal conversation while significantly intoxicated, but their eyes will still be red, their mouth will still be dry, and their reaction time will still be measurably impaired on cognitive tests even if it is not obviously impaired in casual interaction.

This creates a paradox that experienced users should understand: you may feel and act relatively normal, leading you to believe you are undetectable, while your eyes and the residual smell on your clothes are clearly signaling your state to observant non-users.

The Smell Problem

Smell deserves special emphasis because it is both the most immediately detectable sign and the one that cannabis users most consistently underestimate.

Olfactory fatigue is the technical term for what happens when you are continuously exposed to an odor: your brain stops registering it as a distinct signal. Within five to ten minutes of cannabis exposure, your ability to detect the smell on yourself decreases dramatically. This is why users genuinely believe they do not smell when they absolutely do.

Cannabis smoke compounds are lipophilic and absorb into hair, clothing, skin, and breath. The smell persists for one to four hours after inhalation depending on the quantity consumed and the ventilation during consumption. To a non-user walking into a room, the scent is immediately and obviously identifiable.

Edibles and capsules bypass this problem entirely. Vaporizers produce less persistent odor than combustion but are still detectable to nearby non-users during and shortly after use.

Practical Assessment: How Detectable Are You

Rather than providing blanket reassurance or alarm, here is a realistic assessment by scenario.

Low dose, edible or vaporizer, several hours ago. Minimal physiological signs. Behavioral effects likely resolved. Smell negligible with vaporizer, absent with edibles. Detection probability: very low unless you mention it.

Moderate dose, smoked, within the last hour. Red eyes likely noticeable. Smell detectable on clothes and hair. Behavioral signs present but manageable with effort. Detection probability with non-users: moderate to high. With regular users: low to moderate.

High dose, smoked, within the last thirty minutes. Red eyes obvious. Smell strong. Behavioral signs likely noticeable including speech and response patterns. Detection probability: high with nearly all observers.

Daily user, moderate dose, ongoing tolerance. Behavioral signs minimal. Red eyes present but potentially mild. Smell present if smoked. Detection probability: depends almost entirely on smell and eye appearance rather than behavior.

The Anxiety Feedback Loop

One factor worth mentioning is that the anxiety about being detected can itself produce detectable behavioral changes. Cannabis heightens self-monitoring and can amplify social anxiety, especially at higher doses. Worrying about whether people can tell you are high can make you act unnaturally, over-controlling your behavior in ways that paradoxically draw more attention.

Research on social perception consistently shows that people who are trying to appear normal often appear less natural than people who are simply being themselves. The effort of monitoring and controlling your presentation is itself detectable as a kind of stiffness or performative quality.

If you find that anxiety about detection is a consistent feature of your cannabis use, that itself is worth reflecting on. Chronic anxiety about being caught or detected can diminish the quality of the experience and suggest that the context in which you are using may not be the right one for you.

The Bottom Line

Evidence-based guide to cannabis intoxication detectability covering physiological signs, behavioral signs, observer factors, research accuracy, tolerance, smell, and scenario assessment. Physiological signs (uncontrollable): conjunctival injection (THC vasodilation, dose-dependent, onset 15-30 min, persists 2-4 hr, does not diminish with tolerance; tetrahydrozoline/naphazoline drops partially effective); xerostomia (CB1 receptors in submandibular glands reduce saliva; speech and lip-licking observable); altered pupil response (mild, subtle); tachycardia (15-50 BPM increase first hour). Behavioral signs (variable): delayed reaction time, tangential thinking (working memory/executive function), inappropriate laughter (humor processing threshold lowered), psychomotor changes, altered speech patterns. Observer factors: non-users most perceptive (not habituated to cues); former users highly accurate (personal pattern recognition); regular users least reliable detectors; familiarity with baseline behavior increases detection. Research: drug recognition experts 60-80% accuracy; untrained observers 50-65% at moderate doses; eye appearance most cited cue, then smell, then behavior; removing smell drops accuracy significantly. Tolerance: behavioral tolerance develops faster than physiological — daily users may converse normally but eyes still red, mouth still dry, reaction time still impaired. Smell: olfactory fatigue within 5-10 min; lipophilic compounds persist 1-4 hr on hair/clothing/skin; most underestimated cue. Scenario assessment: low-dose edible hours ago = very low detection; moderate smoked within hour = moderate-high; high-dose smoked within 30 min = high.

Frequently Asked Questions

What are the most obvious signs someone is high on weed?

The three most commonly detected signs in order are: smell (cannabis terpenes on clothes, hair, and breath — the most immediate giveaway), red bloodshot eyes (THC causes vasodilation of conjunctival blood vessels, dose-dependent and difficult to fully conceal), and behavioral changes (delayed responses, tangential conversation, inappropriate laughter). Smell and eye appearance are physiological and persist regardless of tolerance level. Behavioral signs vary enormously between individuals and are heavily modulated by tolerance — experienced users may show minimal behavioral changes while still having obvious red eyes.

Can eye drops hide that you are high?

Partially. Over-the-counter redness-relieving eye drops containing tetrahydrozoline or naphazoline constrict conjunctival blood vessels and work within 1-5 minutes, lasting 2-4 hours. They are reasonably effective at reducing visible redness from low to moderate doses but cannot fully eliminate redness from heavy use. Regular use of these drops can cause rebound redness when discontinued. Eye drops also do not address any other signs — smell, dry mouth, behavioral changes, or altered speech patterns all remain. They address only one visible cue, which happens to be the most commonly cited detection sign.

Who is most likely to notice you are high?

Non-users are the most perceptive observers because they are not habituated to the smell, behavioral patterns, or physical signs. Former users are also highly accurate because they recognize patterns from personal experience without current habituation. People who know you well detect behavioral changes more readily than strangers because they have a baseline for your normal conversational pace, emotional expressiveness, and eye contact patterns. Regular cannabis users are the least reliable detectors — the fact that your stoner friends cannot tell does not mean your boss or parents would have the same difficulty.

Does tolerance make it harder for people to tell you are high?

Tolerance reduces behavioral detectability but not physiological signs. Daily users develop behavioral tolerance (ability to function normally despite intoxication) faster than physiological tolerance. A daily user may hold a nearly normal conversation while significantly intoxicated, but their eyes will still be red, their mouth still dry, and their reaction time still measurably impaired on cognitive testing. This creates a paradox: experienced users feel and act relatively normal, leading them to believe they are undetectable, while their eyes and residual smell are clearly signaling their state to observant non-users.

Can people smell weed on you after smoking?

Yes, and cannabis users consistently underestimate how detectable it is. Olfactory fatigue — your brain stops registering the odor — sets in within 5-10 minutes of exposure. You may genuinely believe you do not smell while others immediately detect it. Cannabis smoke compounds are lipophilic and absorb into hair, clothing, skin, and breath, persisting for 1-4 hours after inhalation depending on quantity and ventilation. To a non-user, the scent is immediately and obviously identifiable. Vaporizers produce less persistent odor, and edibles bypass the problem entirely.

How accurate are people at detecting cannabis intoxication?

Drug recognition experts (trained law enforcement) correctly identify cannabis intoxication 60-80% of the time in controlled studies — well above chance but far from certain. Untrained observers perform worse at 50-65% accuracy for moderate doses, only modestly above chance. Accuracy increases substantially at higher doses and when the observer knows the person well. The most commonly cited detection cue is eye appearance, followed by smell, then behavioral indicators. When smell is removed as a factor (edibles or experimental controls), detection accuracy drops significantly.