Photographic Memory: Is It Real, and Do You Have One?

Here is the honest test. A photographic memory would let you look at a page for half a minute, look away, and then read off the fourth word of line 14, or read the whole page backwards as easily as forwards. Haber's criteria for a real eidetic image work along the same lines. People describe the image as if it were projected on a surface in front of them, they talk about it in the present tense, and they can move their eyes across it while it stays where it is. Someone remembering in the ordinary way talks in the past tense and rebuilds the page from what they know.

Almost everyone who tries this finds they can't. What they usually have is something else: a good memory for where things were (top left, under the diagram), fast reading of a subject they already know well, or a knack for grouping material into meaningful chunks. Those are real skills, and they can be measured.

That's what our memory test does. Each of its 50 questions first shows you something to remember on a screen of its own, with a countdown: a list of words, a string of digits, names paired with details, a grid with some squares filled, or a short notice with times and prices. Then it disappears and you answer one multiple-choice question about it. There are five areas of ten questions, each area gets harder as it goes, and you can't go back. It takes about 15 minutes. Your total out of 50 is free when you finish; the full report breaks it down by area and explains it. A perfect score in an area means a strong short-term memory for that kind of material. It doesn't mean a photographic one.

In everyday speech the two mean the same thing. In research they don't. A photographic memory, as people imagine it, would be permanent and exact: a stored copy you could consult years later. Eidetic imagery is neither. It's a picture that lingers for a while after you look away, and then fades.

Haber's tests were built to separate an eidetic image from two other things. An afterimage, like the spot you see after looking at a lamp, moves when your eyes move and lasts only a moment; an eidetic image stays in place while you look around it, and lasts longer. Ordinary memory feels like knowing, not seeing, and people describe it in the past tense ("there was a cat on the left"), where a child with an eidetic image says "the cat is on the left".

Among adults, as far as research can tell, essentially no one. Among children aged 6 to 12, a small percentage show eidetic imagery, and in Haber's long-term study it stayed remarkably stable through the elementary school years. He also found no higher rate among preschool children, or among people who couldn't read in studies across cultures, which argues against the idea that it's an immature kind of memory that children grow out of as they learn to read.

The best-known claim of an adult with it is a 1970 Nature paper by Charles Stromeyer and Joseph Psotka (doi:10.1038/225346a0). They described a woman who looked at a pattern of random dots with one eye, then later at a second pattern with the other eye, and saw the hidden figure that appears only when the two are combined. The case was never repeated with anyone else in the published research we could find. When John Merritt printed similar tests in newspapers and magazines and invited readers to try, his 1979 report was titled "None in a million".

In 2006 researchers at the University of California, Irvine described a woman, called AJ in the paper, who could accurately recall vast amounts about her own life, across most of her lifetime, without using memory techniques (Parker, Cahill and McGaugh, Neurocase, doi:10.1080/13554790500473680). By 2012 the same group had studied eleven people with the ability and named it highly superior autobiographical memory, or HSAM (LePort and colleagues, Neurobiology of Learning and Memory, doi:10.1016/j.nlm.2012.05.002).

Two findings make it unlike the photographic memory of films. First, on most standard laboratory memory tests the HSAM group scored like age- and sex-matched controls. Their gift is for days and events, not for lists or pages. Second, their memories are not recordings. In a 2013 study they were at least as prone to false memories as anyone: like controls, they "recognized" words that had never been on a list, and they absorbed misleading details about a slideshow of photos more than controls did (Patihis and colleagues, PNAS, doi:10.1073/pnas.1314373110). The authors suggest that even these remarkable memories are put together by the same fallible, reconstructive process as everyone else's.

We find that reassuring. If the best autobiographical memories ever measured still rebuild rather than replay, nobody should feel bad about a memory that does the same.

In 2003 Eleanor Maguire and colleagues scanned the brains of people who competed in events like the World Memory Championships and gave them a battery of tests (Nature Neuroscience, doi:10.1038/nn988). Their superior memory was not explained by exceptional intelligence or by differences in brain structure. They used a spatial learning strategy, and while memorizing they engaged brain regions such as the hippocampus that matter for spatial memory.

That strategy is usually the method of loci, also called the memory palace. A 2017 study shows how far it goes. Seventeen memory athletes recalled on average 70.8 of 72 words twenty minutes after seeing them; matched controls recalled 39.9 (Dresler and colleagues, Neuron, doi:10.1016/j.neuron.2017.02.003). Then 51 men with no experience of memory techniques were divided into groups, and one group trained in the method for 30 minutes a day over 40 days. In training their score on a five-minute word task rose from about 17 words to about 42. Their brain networks shifted toward the pattern seen in the champions, and the more they shifted, the larger the memory gains, up to four months after training.

To try it, pick a route you know well, such as the walk from your front door to your kitchen, and fix ten spots along it in order. To remember a list, place each item at one spot as a vivid, slightly absurd picture: milk pouring out of the letterbox, eggs balanced on the coat hook. To recall, walk the route in your head. It's built for lists and sequences; for understanding a textbook chapter, the habits in the last section of this guide matter more.

In 1956 George Miller wrote his famous paper on "the magical number seven, plus or minus two" (Psychological Review, doi:10.1037/h0043158). Later work found the real limit is smaller. Nelson Cowan, reviewing a wide range of studies, put it at about four chunks and noted that Miller's seven "was meant more as a rough estimate and a rhetorical device than as a real capacity limit" (Behavioral and Brain Sciences, 2001, doi:10.1017/S0140525X01003922).

The key word is chunk. A chunk is whatever you can hold as one unit. For a child learning to read, a letter is a chunk; for you, a whole word is. A ten-digit phone number is easier as three groups than as ten separate digits, and SF's 79 digits came from a system for grouping them that he built up with practice. Most of what looks like a bigger memory is better chunking, which is good news, because chunking can be practiced.

That's also why our memory test gets harder in each area by adding more to hold: the point where items stop sticking tells you more about your memory than the total does.

The two habits with the strongest evidence are simple and slightly uncomfortable. The first is testing yourself instead of rereading. In a 2006 experiment, students who studied a passage and then practiced recalling it remembered less five minutes later than students who simply reread it, but remembered substantially more two days and one week later; rereading made them more confident, not better (Roediger and Karpicke, Psychological Science, doi:10.1111/j.1467-9280.2006.01693.x).

The second is spacing your practice out. A review of 317 experiments found that study sessions spread out beat the same time crammed together, and that the longer you need to remember something, the longer the best gap between sessions (Cepeda and colleagues, Psychological Bulletin, 2006, doi:10.1037/0033-2909.132.3.354). A large review of ten study techniques rated these two, practice testing and distributed practice, as the only ones of high utility. Highlighting and rereading, two techniques students report relying on heavily, were rated low (Dunlosky and colleagues, Psychological Science in the Public Interest, 2013, doi:10.1177/1529100612453266).

Brain-training games are the third thing people ask about. A review of the studies their makers cite found good evidence that they improve the games themselves, less that they help with similar tasks, and little that they improve everyday memory or thinking (Simons and colleagues, Psychological Science in the Public Interest, 2016, doi:10.1177/1529100616661983). If you have a quarter of an hour, quizzing yourself on something you actually need to remember is the better bet. And since spacing only works if you start early, our procrastination test may be more useful to your memory than it sounds.

A memory test like ours measures how much you can hold in mind for a few seconds and pick out from close wrong answers, on one day, under one set of conditions. It's a fair snapshot of short-term memory for words, digits, names, positions and details. A tired evening, a buzzing phone and whether you used a strategy can all move the score, so one sitting is an estimate, not a fixed trait.

It can't measure long-term memory, which is what most people mean when they worry about forgetting, and it can't tell you anything about dementia, memory loss or any medical condition. If you're worried about changes in your memory, that's a conversation for a doctor, not a quiz.

It also can't tell you whether you have a photographic memory, because no short multiple-choice test could. What it can show is where your memory is strongest and where it starts to slip, and that's the part you can do something about. If you want the wider picture of how you reason, our IQ test covers that, and our guide to IQ scores explains what the numbers mean.