Two teams of researchers, Langston et al, and Wills et al, found that at just 16 days after birth, rats possess adult-like direction cells and place cells in the hippocampus and entorhinal cortex, areas known to be critical for spatial cognition. A couple of days later, grid cells appear.Bear in mind that rats are born much earlier in their development than people are. Human babies are born with fully-functioning senses and are able to see. The eyes of baby rats however are sealed shut until about day 14. So this is very convincing evidence that the hippocampal system is "hard-wired" to store representations of space in the way that it does. Sensory input provides the actual data about particular places, but the hard work of designing a way of coding space has already been done by evolution.
That's great news for baby rats, because otherwise they would have to learn not just about their environment, but about the very concept of space. That would be asking a lot because baby rats are not the smartest of creatures. It seems very likely that the same is true of humans. It's tricky to stick electrodes into the brains of babies (the parents tend to object), but we know that in adults, damage to the hippocampus and related areas causes spatial processing and memory deficits just like those seen in rats. It's probably not just space, either - Noam Chomsky has built his career on the theory that we also possess a specialized language-learning mechanism.
These data would have come as no surprise to Kant, whose philosophy was based on the notion that our knowledge of the world is dependent upon the existence of innate mental "categories", such as space and time, which we do not learn by experience, but which rather allow us to make sense of experience. Incidentally, Kant looked a bit like a rat.
Yet a mystery remains, and it has nothing to do with the brain. What's the story behind this pair of all-but-identical articles? They're literally the two most similar scientific papers I have ever read - they found the same results, using the same methods, with only a few minor technical differences.The fact that two independent groups have shown the same thing is great evidence for the reliability of these findings, but it's very rare for journals to publish simultaneously in this way, although arguably it should happen more often. So what's the story? Did these two groups not know they were working on the same thing, or was there a "Space Race" to publish first? Are they friendly collaborators or bitter rivals? We can only imagine...
Langston, R., Ainge, J., Couey, J., Canto, C., Bjerknes, T., Witter, M., Moser, E., & Moser, M. (2010). Development of the Spatial Representation System in the Rat Science, 328 (5985), 1576-1580 DOI: 10.1126/science.1188210Wills, T., Cacucci, F., Burgess, N., & O'Keefe, J. (2010). Development of the Hippocampal Cognitive Map in Preweanling Rats Science, 328 (5985), 1573-1576 DOI: 10.1126/science.1188224
9 comments:
A very interesting post indeed.
This may be a overly pedantic observation, but according to Kant space and time were forms of (passive) intuition. In his opinion the intuition filters the sense-data (gognition kind of begins in sensory perception) and after that the understanding which faculties the (constructive) categories are, commands the filtered sense-data into statements.
So space and time were not constructive categories for him but passive receptors. It may be that Kant's idea was to say that when we form propositions we are in need of constructive categories: All our truthfull statements have to be under a particular quantity, quality, relation and modality. Rats could not form propositions (I guess?) so their "mental" faculty would not be comparable to categories of Kant's faculties.
Sorry to reponse so heavily metaphysically but this all just came to my mind because interpreting Kant in a naturalistic sense is quite problematic.
Surprised you think that co-publication rarely happens. These days, researchers find out about each other quite often at meetings and subsequently communicate with editors to get their papers co-submitted. I've done it myself 5-6 times in the last couple of years.
Noah: Oh, ok. It's not something I've ever encountered myself, although now you mention it, I can think of a few other examples, although generally it was two related experiments rather than two replications of the same one.
Perhaps this is evidence that the hippocampal system of researchers is "hard-wired" to perform this sort of research on rats in the way that it does.
The two papers actually differed significantly in their treatment of grid cells. Langston et al. suggest that grid cells mature slowly, whereas Wills et al. say the opposite.
The answer plays a key role in understanding which cells drive learning in the other cells.
(described in the 'Perspectives' section of the issue)
Good on ya for the Kant reference, but is it really apt? That a representation of space is hard wired doesn't mean a category or concept "space" mediates the ding-an-sich and the mind.
The way I understand Kant, it does, although this may mean that I don't properly understand Kant. I suppose the point is that space is not something we learn about, but a system that our minds/brains inherently use to structure our perceptions.
But that "system" may correspond to the way things "really are" in themselves.
@ Ian Smith:
The Kant reference is very good. Burgess has actually written about the relation between his own findings and Kant's synthetic a priori and then tried to put the rats in a simulation of non-Euclidean space but did not manage to make out much of the data. What happened was that it was quite 'noisy', but my guess is that one wouldn't know what kind of arrangement there should be in the grids for non-Euclidean (because if you look at a point in curved space it looks like your normal 3d).
Also, since these are scientists, I don't expect them to worry so much about the body-mind problem.
@makislav- I think Kant couldn't have known that space-time is processed in the hippocampus, and that rats have such a nice one compared to humans. But I don't think Kant's view on rats matters at all; I think what is more important is the idea of categories and how it evolves.
Post a Comment