Medicine Icon Medicine

Understanding Memories: Lovely Metaphors Belong in Songs, Not Science

Michael Egnor


Biologist P. Z. Myers doesn’t agree with my observation that the notion that the brain stores memories makes no sense. Myers writes:

Egnor claims that it is impossible for the brain to store memories. Yes, he knows that neural damage can cause loss of memory, that certain delicate areas of the brain, if harmed, can destroy the ability to make new memories, and he waves those awkward facts away to announce that there is simply no way memory or information of any kind can be stored in a meat-organ like a brain. He doesn’t say where memories are kept, then, nor does he account for any of the physiological correlates of memory, nor does he seem to give a damn about any of the neuroscience experiments that have teased apart the underlying molecular mechanism. By pure reason alone, if we can call his argument a product of reason at all, he deduces that the brain could not possibly have any way of storing memories.

Myers explains his view:

… [I]f memory is patterned activity in the brain, then of course that pattern can have a physical cause: the spatial arrangement of axons and dendrites, the localization of proteins at synapses, subtle changes in synaptic boutons that modify their electrical properties. Aplysia [a worm] can learn to associate a touch with an unpleasant stimulus, and will remember that association when touched in the future, which is a psychological thing. We know how that psychological thing is stored in the brain of Aplysia, as changes in the strength of synapses.

Myers continues:

[Egnor] lapses immediately into dualist assumptions. There is a separate you from your brain, which has to go searching through the brain like a garbage picker rummaging through the rubbish bins to find that portrait of Nana. But that’s clearly not how it works. There is no external entity that has to trace through a series of memory locations — memory is a set of invoked associations. It’s you. There isn’t a homonculus somewhere rifling through the stacks of memories, but instead, those memories are part of the youness of you, and triggering that pattern of activity is part of the consciousness being expressed by your brain…that is, your mind.

This imaginary engram search story only makes sense if you assume dualism and that memory is co-dependent on finding a memory in a disorganized heap, and as Egnor points out, it doesn’t work. By his own reasoning, his model fails.

MInd-and-Technology3.jpgIn the mid 20th century, a group of Oxford philosophers and some allies advanced a philosophical school called ordinary language philosophy. The ordinary language school included many very prominent philosophers — Gilbert Ryle and Ludwig Wittgenstein — and the movement continues today, with John Searle and Stanley Cavell among others. Their goal is laudable. They seek to clear up the morass of conceptual confusion in science and philosophy caused by use of language that did so much violence to the ordinary meanings of words that it had become nonsensical. In their view — a view to which I hold as well — the idiosyncratic use of words, divorced from logic and common sense, reduces to an attempt at private language, which Wittgenstein proposed was incoherent.

The contemporary criticism of such phrases as "memory is stored in the brain" and "the brain evaluates propositions" and "the occipital cortex perceives images" — criticism made by neuroscientists and philosophers like Maxwell Bennett and Peter Hacker among others — is in keeping with the salient critiques by ordinary language philosophers who insist that we need to be honest and careful with the meanings of words in our scientific discourse. Ordinary language philosophy in neuroscience is an appeal to conceptual hygiene.

"Memories are stored in the brain" is simply unintelligible. Memories aren’t storable. It is akin to the assertion "the square root of 10 is red." It is not logically or empirically wrong. It doesn’t rise to the level of testability. It is simply incoherent.

I’m reminded of a phrase from one of my favorite songwriters, Paul Simon, in "The Boxer": "I’ve squandered my resistance for a pocket full of mumbles."

A lovely metaphor (I lived in New York City at the time and took the lyrics to heart). But a neuroscience proposal for a laboratory investigation of pocket-able mumbles would be unlikely to get NSF funding, and would perhaps warrant a psychiatric evaluation of the lead investigator. "Memories stored in the brain" is no less unintelligible than "a pocket full of mumbles."

Intelligibility is prior to logical or empirical analysis. Unintelligible assertions simply can’t be discussed or debated or proven or disproven or be the object of experiments. Unintelligible assertions merely degrade understanding.

Often, unintelligible assertions in neuroscience are simply the careless use of metaphors. We say metaphorically that we store our memories in our photo albums or on our hard drives or in our brains. But memories are psychological things. They have neither mass nor volume nor location, and the assertion that they can actually be stored in anything is unintelligible — no less unintelligible than the assertion that the square root of a number can be a color or that mumbles can be stuffed in pockets.

Representations of memories can certainly be stored. Synapses can be altered in ways that correlate more or less with learning, and we can write "mumbles" on scraps of paper and stuff the scraps in our pockets. And no doubt that is the kind of thing Myers means by invoking storage of memories, sort of. No one doubts that the brain can store representations or at least things that seem like representations — proteins and dendritic arrays and electrochemical gradients and the like. Much of modern neuroscience is the study of these things.

But Daniel Dennett, no less, cautions us: "The recognition that there are two levels of explanation [in neuroscience] gives us the burden of relating them…" (Content and Consciousness, pp. 95-96).

If Myers believes that there are two levels of explanation for memory — the psychological and the physical — he needs to shoulder the burden of relating them. I believe that there are two levels of explanation for memory, and I relate them via Thomistic metaphysics and psychology. It’s a venerable approach to philosophy of the mind and neuroscience.

If Myers believes that there is only one level of explanation for memory, then he believes either: 1) psychological memory is the same thing as the spatial arrangement of axons and dendrites, etc. or 2) psychological memory is folk psychology that doesn’t really exist, and all that exists is spatial arrangement of axons and dendrites, etc.

If Myers believes the first, he adheres to identity theory, either token identity theory or type identity theory. If he believes the second, he adheres to eliminative materialism.

Myers doesn’t make it clear which theory he accepts. One doubts that he understands either theory well enough to accept it or reject it. His neurophilosophy is little more than a pocket full of mumbles.

Image: By Eddie Mallin [CC BY-SA 2.0 (], via Wikimedia Commons.

Michael Egnor

Senior Fellow, Center for Natural & Artificial Intelligence
Michael R. Egnor, MD, is a Professor of Neurosurgery and Pediatrics at State University of New York, Stony Brook, has served as the Director of Pediatric Neurosurgery, and award-winning brain surgeon. He was named one of New York’s best doctors by the New York Magazine in 2005. He received his medical education at Columbia University College of Physicians and Surgeons and completed his residency at Jackson Memorial Hospital. His research on hydrocephalus has been published in journals including Journal of Neurosurgery, Pediatrics, and Cerebrospinal Fluid Research. He is on the Scientific Advisory Board of the Hydrocephalus Association in the United States and has lectured extensively throughout the United States and Europe.



Continuing SeriesMind and TechnologyPhilosophyScience