An item in today’s Physorg newsletter describes some remarkable neurological research: scientists at CalTech, by showing pictures to test subjects while monitoring brain activity, have managed to associate individual neurons in the medial temporal lobe with specific perceptions.
We read:
Dr. Moran Cerf of the California Institute of Technology in Pasadena and colleagues conducted their experiment by showing the subjects images of people, places or objects that were familiar to them, including pictures of celebrities such as Michael Jackson, Marilyn Monroe, and Bill Clinton. They then looked for the neurons that fired when the subject was shown each image.
In each of the subjects they found individual neurons fired when the person looked at a specific image. So there was a “Michael Jackson neuron”, a “Marilyn Monroe neuron”, and others that fired when the person was shown an image of the Eiffel tower, a spider, or other familiar objects or places.
When the neurons corresponding to particular images had been identified, the researchers hooked the electrodes up to a computer that displayed the image corresponding to the neuron that fired. The subject was then asked to think about one of the images. So, for example, a subject was asked to think about Marilyn Monroe. The Marilyn Monroe neuron in the subject’s brain fired, and the information was relayed to the computer, which then displayed Monroe’s image.
Another experiment designed to test how well the subjects could control the single neurons was a fade experiment in which the subject was shown a combined image of two faces: Josh Brolin (star of Goonies) and Marilyn Monroe, and told to think of Josh Brolin. The electrodes sent data on the Josh Brolin and Marilyn Monroe neurons to the computer, which brightened the image of the one causing most neuron firing. As the subject thought of Brolin, the image of Monroe faded out.
This is an impressive result, with enormously important implications.
Next questions: if I have a neuron in my brain that reliably activates when I see or think of Marilyn Monroe, does it really fire only when I see or think of Marilyn Monroe? I’ll be surprised if neural resources are allocated in such a “dedicated” way — a one-to-one mapping of the kind suggested here, with each neuron representing exactly one intentional referent, and doing nothing else; it seems so limiting. I’ve always imagined that the hardware implementation of our memory and intentionality would take the form of configurations of groups of neurons; it seems you’d get more for less that way. And is this “Marilyn” cell activated if I see a picture of Marilyn Monroe without being conscious of it? How about if I dream about her?
And then the much harder question: why that neuron?
Read the story here.
3 Comments
Oh. That neuron.
I’m curious as to what we’ll discover, as a consequence of this research, about how physically similar our brains are to each other. Does the similarity really extend down to the neuronal level?
Hi Kevin,
Well, of course it is awfully improbable that we could map particular neurons in my brain onto corresponding individual neurons in yours. The point of the “why that neuron?” question is that to answer it we would have to understand the brain’s “operating system”, in the same way we need to understand a computer’s operating system in order to know why a particular memory location had been allocated for specific data.
It would also lead us toward an implementation-level understanding of just how a single neuron can “represent” a particular intentional state.
Lacking that OS-level understanding it is going to be impossible to establish neural correlates of intentional states without collaborative reporting by the subject, so being able to put a helmet on an uncooperative witness, for example, and “reading off” the contents of his or her mind is not in the cards anytime soon, I think, though I do think it is possible in principle. Indeed, it would be an enormous task, perhaps impossibly so, simply to look at the contents of a computer’s hard drive and understand what all the ones and zeroes represent without a thorough understanding of the OS and ancillary software that created and stored the data in the first place.