Yunus Tuğ For Unsplash+
Study: If Someone Passes This Drunk Memory Test, a Blackout Almost Certainly Isn’t Happening
In A Nutshell
- About half of all drinkers have experienced a blackout, and most had no idea it was happening at the time.
- A University of Missouri study found a simple smartphone photo-recall test can help determine whether a blackout is occurring during a drinking episode.
- Passing every recall test was a strong indicator no blackout was underway, with a 92% accuracy rate (and 99% for the most severe type of blackout).
- The test is better at ruling blackouts out than confirming them, but researchers say it could advance prevention, research, and even legal proceedings.
Alcohol blackouts are not something that happens only to people with drinking problems. About half of all people who drink have experienced one, and most had no idea it was happening at the time. They were awake, talking, making decisions, and carrying on what looked like a normal night out. Their brain had quietly stopped recording all of it.
That invisibility is exactly the problem. A new study published in the journal Addiction tested whether a simple smartphone memory quiz, given to people while they were actively drinking, could flag when memory may be failing, and more reliably, show when a blackout was unlikely. For more than 50 years, the only way to assess whether someone had blacked out was to ask them afterward.
Blackouts have been linked to injuries, arrests, sexual victimization, and costly emergency room visits, yet for all that harm, they have remained essentially undetectable as they occur.
Blackouts Strike Nearly Half of All Drinkers, Regardless of Alcohol Use Disorder
Here is what makes blackouts so hard to catch. During one, the brain loses its ability to store new long-term memories, but short-term memory stays intact. A person can hold a conversation, drive a car, and recall something from two minutes ago. What they cannot do is hold onto any of it past the moment. Nobody present, including the person blacking out, necessarily knows it is happening.
Researchers at the University of Missouri enrolled 63 young adults between 18 and 30, all of whom reported multiple blackouts per month. About 78% were female and 51% identified as White. Over 30 days, participants logged their drinking on a smartphone app. When participants filed a drink report, including hourly follow-up reports during a drinking episode, the app showed them a photo for five seconds. Fifteen minutes later, it asked them to recall what they had seen, first by typing a description, then by identifying the correct image from three options. That gap was the key: short-term memory works during a blackout, but storing a new image for 15 minutes requires the brain to actually consolidate it. If a person remembered the photo, their memory was likely functioning. If they didn’t, it may not have been.
Participants completed about 85% of the memory prompts they were sent, even on nights of heavy drinking. Sixty percent reported at least one blackout over the course of the study.
Passing the Alcohol Blackout Memory Test Was a Strong Sign the Brain Was Still Recording
When participants failed at least one recall test on a given day, they self-reported a blackout 39% of the time. That is a real signal, but not a reliable one on its own. Being drunk, distracted, or simply inattentive can cause someone to forget a photo without a blackout occurring. So a failed test alone is not enough to draw conclusions.
More telling was what happened when participants passed. On days they correctly recalled every memory test, they reported no blackout 92% of the time. For the most severe type of blackout, where memories are permanently gone rather than just hard to retrieve, that figure climbed to 99%. In other words, passing the test was a strong sign that the brain was still recording.
Each extra drink above a person’s usual nightly intake also raised blackout odds by about 74% on average, underscoring that how much someone drinks relative to their norm matters as much as total consumption.
Real-Time Alcohol Blackout Checks Could One Day Help Friends, Bartenders, and Courts
A real-time signal like this opens up possibilities that simply did not exist before. A friend checking on someone who has had too much, a bartender gauging a patron’s state, or an app-based alert system could all potentially use a quick image recall prompt to assess whether memory is still working. And unlike most tools in alcohol research, this one requires nothing more than a phone. Researchers also suggest the test could move blackout science away from next-day surveys and toward in-the-moment study of how and when memory begins to fail.
There is also a legal dimension worth noting. Blackouts are sometimes cited in criminal and civil cases. Researchers suggest the test would be more appropriate for arguing that a blackout was unlikely than for proving one occurred, given the range of other reasons someone might fail a recall test while intoxicated.
Further work is needed to sharpen the test’s ability to positively identify blackouts rather than just rule them out. Researchers point to heavier memory tasks and alcohol biosensor data as two possible ways to improve accuracy. But for a problem that has quietly evaded in-the-moment scientific study for well over half a century, even a real-time clue is a long time coming.
Disclaimer: The findings of this study are based on a small sample of 63 young adults who reported frequent alcohol-induced blackouts, the majority of whom were female. Results may not apply to all drinkers. This article is intended for informational purposes only and does not constitute medical or legal advice.
Paper Notes
Limitations
Participants were predominantly young adult social drinkers who reported frequent blackouts, and women were significantly overrepresented in the sample. No objective biological measure of blackout was available as a comparison standard, so the study relied on participants’ own self-reports of memory loss, which introduces the risk of errors in both directions. Recognition memory showed little variability across participants, limiting its usefulness as a diagnostic indicator. The test performed better at detecting full, permanent blackouts than partial ones where some memories can be retrieved with prompting.
Funding and Disclosures
This work was supported by the University of Missouri’s Center for Addiction Research and Engagement. Investigator effort was supported by the National Institute on Alcohol Abuse and Alcoholism (grant numbers K23AA026895 and T32AA013526). The National Institute of Health had no role in study design, data collection, analysis, interpretation, or the decision to publish. The authors report no conflicts of interest.
Publication Details
Authors: Mary Beth Miller, Sydney D. Shoemaker, Lindsey K. Freeman, Ashley F. Curtis, Jennifer E. Merrill, and Edgar C. Merkle. Miller and Shoemaker are affiliated with the Department of Psychiatry and the Department of Psychological Sciences at the University of Missouri; Curtis is with the College of Nursing at the University of South Florida; Merrill is with the Center for Alcohol and Addiction Studies at Brown University; Merkle is with the Department of Psychological Sciences at the University of Missouri. Title: “Predictive value of real-time memory tests in identifying alcohol-induced blackouts in situ.” Journal: Addiction. DOI: 10.1111/add.70446. Published 2026.







