Sherlock Holmes and the infamous brain attic

Illustration: tisserande

Maria Konnikova is the author of Mastermind: How to think Like Sherlock Holmes, out January 2013 from Viking, and a blogger at Scientific American. Visit her website and follow her on Twitter.

One of the most widely held notions about Sherlock Holmes has to do with his supposed ignorance of Copernican theory. “What the deuce is [the solar system] to me?” he exclaims to Watson in A Study in Scarlet. “You say that we go round the sun. If we went round the moon it would not make a pennyworth of difference to me or to my work.” And now that he knows that fact? “I shall do my best to forget it,” he promises.

It’s fun to hone in on that incongruity between the superhuman-seeming detective and a failure to grasp a fact so rudimentary that even a child would know it. And ignorance of the solar system is quite an omission for someone who we might hold up as the model of the scientific method, is it not? Even the BBC series Sherlock can’t help but use it as a focal point of one of its episodes.

But two things about that perception bear further mention. First, it isn’t, strictly speaking, true. Witness Holmes’s repeated references to astronomy in future stories—in “The Musgrave Ritual,” he talks about “allowances for personal equation, as the astronomers would have it”; in “The Greek Interpreter,” about the “obliquity of the ecliptic”; in “The Bruce-Partington Plans,” about “a planet leaving its orbit.” Indeed, eventually Holmes does use almost all of the knowledge that he denies having at the earliest stages of his friendship with Dr. Watson. (And in true-to-canon form, Sherlock the BBC series does end on a note of scientific triumph: Holmes does know astronomy after all, and that knowledge saves the day—and the life of a little boy.)

In fact, I would argue that he exaggerates his ignorance precisely to draw our attention to a second—and, I think, much more important—point. His supposed refusal to commit the solar system to memory serves to illustrate an analogy for the human mind that will prove to be central to Holmes’s thinking and to our ability to emulate his methodology. As Holmes tells Watson, moments after the Copernican incident, “I consider that a man’s brain originally is like a little empty attic, and you have to stock it with such furniture as you choose.”

When I first heard the term brain attic, all I could picture in my seven-year-old head was the cover of the black-and-white Shel Silverstein book that sat prominently on my bookshelf, with its half-smiling, lopsided face whose forehead was distended to a wrinkled triangle, complete with roof, chimney, and window with open shutters. Behind the shutters, a tiny face peeking out at the world. Is this what Holmes meant? A small room with sloped sides and a foreign creature with a funny face waiting to pull the cord and turn the light off or on?

As it turns out, I wasn’t far from wrong. For Sherlock Holmes, a person’s brain attic really is an incredibly concrete, physical space. Maybe it has a chimney. Maybe it doesn’t. But whatever it looks like, it is a space in your head, specially fashioned for storing the most disparate of objects. And yes, there is certainly a cord that you can pull to turn the light on or off at will. As Holmes explains to Watson, “A fool takes in all the lumber of every sort that he comes across, so that the knowledge which might be useful to him gets crowded out, or at best is jumbled up with a lot of other things, so that he has a difficulty in laying his hands upon it. Now the skillful workman is very careful indeed as to what he takes into his brain-attic.”

That comparison, as it turns out, is remarkably accurate. Subsequent research on memory formation, retention, and retrieval has proven itself to be highly amenable to the attic analogy.

The attic can be broken down, roughly speaking, into two components: structure and contents. The attic’s structure is how our mind works: how it takes in information. How it processes that information. How it sorts it and stores it for the future. How it may choose to integrate it or not with contents that are already in the attic space. Unlike a physical attic, the structure of the brain attic isn’t altogether fixed. It can expand, albeit not indefinitely, or it can contract, depending on how we use it (in other words, our memory and processing can become more or less effective). It can change its mode of retrieval (How do I recover information I’ve stored?). It can change its storage system (How do I deposit information I’ve taken in: where will it go? how will it be marked? how will it be inte- grated?). At the end, it will have to remain within certain confines—each attic, once again, is different and subject to its unique constraints—but within those confines, it can take on any number of configurations, depending on how we learn to approach it.

The attic’s contents, on the other hand, are those things that we’ve taken in from the world and that we’ve experienced in our lives. Our memories. Our past. The base of our knowledge, the information we start with every time we face a challenge. And just like a physical attic’s contents can change over time, so too does our mind attic continue to take in and discard items until the very end. As our thought process begins, the furniture of memory combines with the structure of internal habits and external circumstances to determine which item will be retrieved from storage at any given point. Guessing at the contents of a person’s attic from his outward appearance becomes one of Sherlock’s surest ways of determining who that person is and what he is capable of.

Much of our attic’s original intake is outside of our control: just like we must picture a pink elephant to realize one doesn’t exist, we can’t help but become acquainted—if only for the briefest of moments—with the workings of the solar system should Watson choose to mention them to us. We can, however, learn to master many aspects of our attic’s structure, throwing out junk that got in by mistake (as Holmes promises to forget Copernicus at the earliest opportunity), prioritizing those things we want to and pushing back those that we don’t, learning how to take the contours of our unique attic into account so that they don’t unduly influence us as they otherwise might.

While we may never become quite as adept as the master at divining a man’s innermost thoughts from his exterior, in learning to understand the layout and functionality of our own brain attics we take the first step to becoming better at exploiting its features to their maximum potential—in other words, to learning how to optimize our own thought process, so that we start any given decision or action as our best, most aware selves.


  1. However interesting Sherlock Holmes may seem, we have to remember he’s not and was never a real person. He came from a man’s imagination, and is entirely fictional. Different aspects of his personality have probably been inspired by real people Conan Doyle knew, but there is no evidence a single person existed with his abilities and demeanor.

    Demeanor which was downright nasty – being a fake human being, he would make a terrible real one. However much I can sometimes admire some of the imagined traits of the character, I certainly have no intention to emulate him in others.

    The article’s focus on compartmentalizing thoughts and memories especially seems like a bad idea. I’m a knowledgivore myself, and the strength of my abilities is not so much what I remember (despite having a very good memory, I am not so good at rote memorization) but what I understand. The links I can discern between various datum that are seemingly unrelated, but which I can relate to one another and then to other information that may at first have seemed unrelated is a skill that I have honed over the years

    In the end, I think that I can achieve results much better than the imaginary methods of an imaginary character. We may think those methods would work because of how they are portrayed as working, but the fact is that however entertaining the character and his methods are, they are still imaginary.

    Maybe I’m wrong. It’s also highly plausible that everyone has a differently organized mind and optimizing its use is different for everyone – what works for me might not work for others.

    1.  Actually Sherlock was directly inspired by Dr. Joseph Bell, a teacher/employer of Arthur Conan Doyle. That isn’t to say it changes all of your argument, but the character is actually based off of a person and their habits, just taken to an extreme.

    2. I am quite sure that if you use a device such as “an attic” to compartmentalize knowledge it will not actually stay that way when your attention is elsewhere. Instead other non-attic-locality associations will form based upon the most recent experience, the most pronounced experience, discovered properties of the subject, representations of it (including where knowledge of it is physically kept, eg books), and other prior experiences with it. I am sure you are aware of this, though.

      Like you I am not confident that my brain works in exactly the same way as everyone else. Having stated my disclaimer, to me my principle concern is attention. My understanding of intelligence is not an analog of brain “horsepower” but of having good filters to keep irrelevant information out of the process, the signal and noise model. For instance I find that brown noise from electric heaters and ac units allow my sense of hearing to relax and I gain increased rote concentration. Why is that? I think this is because I know with certainty that it is noise without any useful attributes and I can thus filter it out. I could fit a metaphor of an attic into this kind of self-awareness. My attic would not just be a storage location but also a place to be apart from the noise of the household while being with the stored information.

      I don’t actually find the “attic” to be useful to me, though. It is likely that Arthur Conan Doyle resonated with the metaphor when he thought about his own creative process, perhaps keeping a writing room with objects and information that contributed to and did not detract from his creative work. Many writers do this. However I do imagine I am hearing brown noise when I need to concentrate such as during an exam or interview so I do artificially carry my metaphor with me and this seems to work for me.

  2. AFAIK, Arthur Conan Doyle was no neurologist. Everything Shelock Holmes did or said was from this guy’s imagination. So in using him as a model for brain science, you can pretty much say whatever you want and there;s no way to falsify any of it. It’s a lot like basing political arguments on the events of Battlestar Galactica. Fun, maybe, but please don’t call it science.

    1. The point here isn’t that Arthur Conan Doyle was a neurologist. The point is to use a popular fictional character as a framing device to draw people in and help them conceptualize and learn about real science. You’re getting it backwards if you think that the point is to show us all how much Doyle knew or how smart Sherlock Holmes hypothetically was. 

      1. I think the point was to demonstrate how memory works. A real life example- like the famous incident where Albert Einstein forgot his own name- would have been more realistic. The attic metaphor casts memories as static nouns that lie about and wait for us to pull them out. But the Einstein story suggests each memory is a verb building on other verbs. The mental cathedrals of medieval times suggest an interlocking webwork of memory that’s interdependent for its context. The difference between an attic and a church is that people don’t really use an attic for anything in particular.
         Sherlock Holmes is a poor model to draw from- especially when there are so many real life examples from which to draw.

        1. Your Einstein example is as apocryphal as Sherlock Holmes is fictional (I read of it being his phone number he couldnt recall).  Either example can be useful in attending to the point.  I can’t argue with what you think the core message of the article was, (I agree with Maggie), but I do think you are reaching when you way “people don’t use an attic for anything”.  While I can’t speak for people in general, I can speak for myself and a small circle of with whom I am familiar.  And we use our attic space to store things that are important but which we don’t expect to need immediately or continuously.  My wife and I have a continual tug of war about whether a Halloween costume or an outdated textbook deserves to take up some of that limited space.  The trade-off is that we can put more “stuff” in the attic at the cost of something else which has to get thrown away.

          It is reasonable that the use of Shelrlock Holmes (probably as familiar to many people as Einstein) does not speak to you in this context, but I think you are reaching to conclude it is a poor model on a sample size of one.

          1. This newest reboot of the Sherlock mythology does some pretty cool things with the character. Mostly when we see him on TV, he’s portrayed as someone whose you’d actually enjoy. Not so much this fellow! He’s as aspie as they come, and  more than a little spooky, as seen by others.

            (those parts of his character are what I relate to the most. Not so much the brilliant crime fighting bit)

            I think we are meant to infer that his brilliance comes at a price. If his emotional life were in synch with those around him, he wouldn’t be able to pick the world apart the way he does.

             When I think about Doyle’s attic, the lumber is dry and straight and uniform, chosen at leisure, no pushy salesmen or termites to worry about.

            In my own experience, it’s hard to tell what’s lumber sitting around waiting for use, and what’s actually a part of the ceiling. The wood is never the same species or cut, and it’s usually warped by water damage.

             In other words, the emotional state I was in when I acquired the memory, has everything to do with where that memory goes and how it’s handled. I think Sherlock Holmes has a very simplistic view of his own mind, and I know better than to follow his example.

        2. Google “memory palace”.  It does not matter one whit whether one uses an attic, a cathedral, or a public bathroom. 

          Though I don’t think that actually speaks to the point of the OP, which struck me more like Michael Pollan’s passage in Botany of Desire about the importance of forgetting.  Some of Oliver Sachs’ writings about folks with photographic memory are similar.  It’s important not to remember everything, as strange as that sounds.

          1. Another passage in fiction that comes to mind, is when Lazarus Long complains that too long a life makes for muddled memory- he might have to rearrange his schema somehow if it begins to be a problem.

             I think the Doyle passage is actually a little offensive to me, because it’s been suggested before that I am a poor craftsman, picking up lumber willy nilly. If memories took up exclusive space that couldn’t then be used for other memories, that would make sense. But in my experience memories tend to fit inside each other like nesting russian dolls.

            If one could carefully choose what memories to retain, and what to ignore, there would be no such thing as PTSD. It would also mean that you could predict, for the rest of your life, the kinds of experinces that you’d have. _The Eternal Sunshine of the Spotless Mind_ had fun with that one!

  3. Apropos of nothing: You know what I love about boingboing? How they manage to leave a couple hundred pixels of white space on the left side of the page, and still manage to get the StumbleUpon button to overlap the text. Fuckin’ awesome!

    (Yeah, I know they’re not the only one. But I just happened to pick this moment to be particularly annoyed by it.)

  4. What the deuce is it to me how a fictional character, a drug addict, created in small spurts over several decades by a guy (who also believed in ghosts and fairies) who was cranking them out even when he didn’t want to, thinks?

    1.  No one can answer that question but you.  And no one else is in a position to close your browser window and find you something else to do instead.

  5. “Knowledge is of two kinds. We know a subject ourselves, or we know where we can find information upon it. When we enquire into any subject, the first thing we have to do is to know what books have treated of it. This leads us to look at catalogues, and at the backs of books in libraries.”
    — Samuel Johnson 

  6. Well, then.  I guess I’d better stop reading all the random crap on BoingBoing right now.  All this useless information is crowding out stuff I really need to know.

    1. Considering the time I’ve spent on TV tropes, it’s probably just flat out too late for me. Unless I become a TV writer.

    1.  Yes, I was going to mention Giordano Bruno. Not a fictional character, although John Crowley uses him in his fiction. As Crowley describes the ‘system’, each piece of information is set in one room or another and this helps to find it again later. But the real value is that when you come looking for it, it has been affected by its surroundings.

    2.  Yeah – I was expecting the article to segue into memory palaces but instead it.. I’m not sure what it did. Rambled I guess.

  7. And how is one to decide which stuff is worth keeping? He tends to use some of the most obscure, out-of-the-way facts to tie his cases together.

    1. Seasons, Tides are linked to Planetary model. If Sherlock finds the color of mud stains important to his work then so could weather, longitude and latitude. Our brain seems to have a huge capacity to not only absorb massive amounts of details but also organize it into usable and re-linkable information (much like the web). Given the right context 2 “widely” separated facts could prove crucial to solve a case. It seems that Holmes’ primary skill was to build his case on obscure but pertinent facts together into a web of corroborating evidence. So in the words of another fictional character “important” is as “important” does.

      On a slightly parallel note it seems like a very judgemental approach to sifting information rather than a storage decision would be required. Personality typing might classify Sherlock as an INTJ (Introverted, iNtuitive, Thinking and Judgemental personality):

  8. Doyle does spend a lot of time speculating on the value of information and knowledge. His brother Mycroft was a pretty close approximation of the internet, a person who was the sum total of governmental knowledge and could retrieve it across almost any index on demand.

    Sherlock on the other hand was all about the practicality of knowledge, how looking at a system from all it’s relative viewpoints can provide unexpected (and entertaining) results. Just spouting off that the earth revolves around the sun to sound like a smarty-pants probably pissed him off to no end.

    1. Letting it steep, I also wanted to add that Sherlock as the perfect reasoner, uses his full intelligence to solve the problem by only using the perfect bits of knowledge that are necessary. Anything extra would be at best a distraction, and at worst a clue to the wrong solution.

      Of course only in fiction can someone know for certain exactly what needs to be known to solve a problem. Every fact and law we have are only good until something comes along to disprove it, but in those heady days of discovery, one could almost believe in our mastery of knowledge.

  9. There was also a point in one of the Sherlock Holmes books, however, where he made nearly the opposite statement, something to the effect of, “it’s best to learn all one can because one can’t foresee what knowledge will be required in the future”. I can’t recall where this comes from (ironically), and I hope I’m not imagining it, but the memory is vivid because I remember how diametrically opposed this seemed to Holmes’ (i.e. Doyle’s) earlier comparison of the brain and its memory’s contents as something akin to a master craftsman’s toolbox.

  10. As a cognitive psychologist, I have to point out that this conceptualization of memory is at least 40 years out of date.  [Quick clarification–the idea of a “Memory Palace” is certainly an excellent mnemonic strategy, but it’s not a theory of memory, which the author is arguing here; totally different concepts.]  There’s a certain usefulness to the analogy, but, as Maria partially concedes, there are a lot of caveats that need to be included: your attic needs to be able to change sizes (maybe it’s holographic?); of course when you find something in your attic, it’s possible to misplace it, but it’s also possible to turn it into something else; etc. 

    But there are a lot of ways in which this analogy directly contradicts current understandings of memory.  Conceiving of pieces of information as items in an attic brings to mind the old analogy of memory as video recordings–the idea that a memory is sort of a physical object that you can revisit whenever you need to.  But as I pointed out, we know that the mere act of accessing a memory can change that memory; making it more or less likely to be recalled in the future, changing how it’s retrieved, or adding to or taking away from aspects of the memory (including confabulated, i.e., made up, aspects).  Similarly, the idea of your memory as being like a room suggests that it can be “filled up,” and that it would be better easier to both store and to find things when the room is emptier than when it is more crowded.  But this is just wrong–we know, instead, that a person’s accumulated knowledge actually makes it *easier* to create new memories, as well as to retrieve related memories (within limits–organization counts for a lot, as well).  Memories are not isolated curios on shelves–they have definite unconscious linkages, which themselves can be modified over time (see Proust & madelines). 

    Now, what I’m describing comes from research on typical populations.  It’s true that there is variation in memory structure and abilities–some of these limitations and structures might be different for, say, a Kim Peek-like savant.  But I don’t think that Sherlock Holmes is a good model for understanding one’s own mind.  I had my own youthful infatuation for the detective, but I’ve come to realize that his famed “powers of deduction” are about as realistic as Superman.

  11. Unlike a physical attic, the structure of the brain attic isn’t altogether fixed. It can expand, albeit not indefinitely, or it can contract, depending on how we use it

    So it’s really a Brain Tardis…

  12. It seems like a lot of people are scoffing at this, but they should know the guys who compete in the world memory championships memorize whole decks of cards at a time, and they use a method just like this to do it. There was a writer who interviewed one of the champions about his method, and then applied it himself and became champion the next year, he wrote a book about it.

    To summarize the idea, it involves creating an imaginary place to store memories, wherein you place objects that you associate with what you want to retain. And then to recall this information later you walk through this imaginary structure in your mind and look for the object associated with what you’re remembering.

    It’s supposedly easiest to break it down into floors with five rooms to a floor, and five objects to a room. That way you know there’s 25 memories to a floor and if you’re trying to remember a specific thing you associate it with it’s coordinates, say 1-3-5 for example, the first floor, third room, fifth object.

    I don’t think you could live your life memorizing everything like this, that seems impractical. But for memorizing tedious things like vocabulary words, decks of playing cards, historical dates, etc, it is a viable mnemonic device.

Comments are closed.