First of all, I'm not knowledgeable of philosophy by any means (engineering student). However, I was wondering if from a philosophical point of view we could reason about what might happen to a brain that has never received sensory data. Obviously, it would have no conception of language. Could it still have thoughts? Would the thoughts be simply primal urges, or would even those cease to exist without some sort of input? Would a functionalist theory say that since there are no sensory inputs, there could be no outputs if we later hooked up the brain to a machine that could somehow express outputs without altering the brain's state of not receiving inputs? Please let me know what you guys think from a philosophical standpoint. This question has been on my mind for a while, so to speak. Thanks!

  • First and last. Don't you think we all are already in this vat? Brain will take himself as input. If he is not point-like. If he is point-like then its another story. – Asphir Dom Feb 6 '14 at 14:41
  • Why would it obviously have no conception of language? – val Feb 6 '14 at 15:34
  • Because language is a social phenomenon, so it seems likely that because this particular brain would receive no sensory input (including inputs such as the speech of other people etc) that it would not be able to develop a language as we conceive of. Although it may have some primal thoughts, it would not have a conception of anything perceivable in our world. Although let me know if there are schools of thought that allow for language to arise separately from socialization. It seems like it wouldn't hold in the functionalist theory of the mind. – nschager Feb 6 '14 at 18:13
  • Pure conjecture, but: it might be possible, that, devoid of stimulation, the brain-in-vat might not develop a sense of identity, of self (the sense of self might be a bi-product of having a body and interacting others-with-bodies). Without the sense of identity, or perhaps for some other reason, the brain might form multiple identities which would interact with each other. Or the brain would imagine (create) an "other", a world with which to interact. – obelia Feb 6 '14 at 18:53
  • Would passage of time be considered a sensory input? – adeena Feb 14 '14 at 12:53

Probably not.

You may like Lakoff and Johnson's Metaphors We Live By, which argues that language is deeply dependent on metaphors, like "argument is war". The most basic metaphors are founded in experience. No experience, and your building blocks for how to speak and think are much less determined. I would also suggest Philosophy in the Flesh by the same authors. In particular, they describe embodied cognition (or 'the embodied mind'), which claims that how we think hugely depends on what we can do with our bodies and senses.

Another way to think of this issue is to consider the difference between logically possible neuron configurations, and meaningful neuron configurations. I suspect that the ratio of meaningful : meaningless is extremely small, perhaps smaller than the ratio of solutions : possible answers for NP-complete problems. Ockham's razor is one way to think of this: we tend to only be able to learn a tiny bit at a time about some complex system. We work by successive approximation, with constant checking of our ideas with reality. But what if there is no [accessible] reality?

We seem to need sense-experience and motor-control to constrain our conceptions of reality to meaningful ones. We seem to need these constraints to sufficiently narrow down possible thoughts to thoughts sufficiently likely to be meaningful ones. This very problem will be explored in depth as we try to construct strong AI.


To be conscious means to be conscious of something. If the brain received no inputs of any kind what would it be conscious of? The brain would not have any content of consciousness.

From a functional standpoint we could say that the brain would have no function. A function, such as f(x), must have an x to operate on. The brain you describe would have no inputs to operate on, so a researcher would have no method of even sensing of the brain was responding to any inputs.

The brain would be blocked off from any interaction with the real world. There is no reason to believe that such a lump of matter would ever develop any complex functions. And if it had, there wouldn't be any way to detect them. If the brain, for instance, started outputting a signal through some particular neurons, and that signal turned out to be the number Pi, we would immediately begin asking how the brain had received the information.

Primal urges are urges to satisfy sensory input conditions. So a brain without inputs could have no such desires because it would have never had such an experience.

You would never be able to ask it such questions, and it would never know you asked such questions, and would not be able to answer.

In computer programming there is a phrase which applies to this question. The phrase is "garbage in, garbage out" or GIGO. But in this case it would be NINO, "nothing in, nothing out".

The information content of the brain would be indistinguishable from background noise.

  • 1
    I disagree with the comparison to computers because computers, as they exist today, are designed to "sit and wait" for input, and brains might not be. Brains, if starved of input, might create their own problems to solve, even if detached from all external reality. There might be a primal urge to do something, and if a meaningful task isn't available, a meaningless task might be created. – obelia Feb 6 '14 at 19:13
  • 1
    GIGO does not just refer computers "as they exist today". It refers to all computers, human or machine. To process information is to process information about that which exists. If the brain has no inputs of any kind from the external world it would not have any information content to process, let alone having any method of validating the results of it's processing. – inTEGraTOR Feb 6 '14 at 19:25
  • Information exists in computers but they are (generally) designed to not do anything with it, after boot up, until input is received. This has been the popular architecture of man made computers since the beginning. I don't know of anything to suggest that brains have that constraint. – obelia Feb 6 '14 at 19:43
  • The evidence that our minds operate on input is the fact that we always try to validate our knowledge by asking if it is "real". The human mind would not be able to develop it's own "problems", as you say, because it would have no criteria of reality to judge it's whirlwind of signals against. It would have no way of validating it's whims and so every wild phantasm would be as "true" as any other. The information processing function of the brain would never get off the ground without first having information to process. To be conscious is to be conscious of something. – inTEGraTOR Feb 12 '14 at 7:10

One approach would say yes.

Consider a brain that has some noise factor to it. If there is any structure to the brain (which is true for most brains), that structure will color the noise. If one divides the brain into two sections, an inner part and an outer part, we can re-frame the brain-in-a-vat. If we treat the inner part as the "brain" and the outer part as a "vat," we now have sensory input and the ability for output, and the brain may seek to make sense of this outer vat. We have now returns to the traditional brain-in-a-vat puzzle, with sensory input.

Of course, this came with the assumption that the brain had structure in a form which facilitates drawing such a boundary.


The notion of a 'brain-in-vat' is a vivid way of describing using imagery from science fiction a traditional problem of philosophy: that is, is all that that there is are minds. In traditional philosophical terms this position is a form of idealism.

From the empirical view, we see brains do need 'sensory input'.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.