Slippery Humanity

wise menThere were a number of reports recently that a robot had passed ‘one of the tests for self-awareness’. They seem to stem mainly from this New Scientist piece (free registration may be required to see the whole thing, but honestly I’m not sure it’s worth it). That in turn reported an experiment conducted by Selmer Bringsjord of Rensselaer, due to be presented at the Ro-Man conference in a month’s time. The programme for the conference looks very interesting and the experiment is due to feature in a session on ‘Real Robots That Pass Human Tests of Self Awareness’.

The claim is that Bringsjord’s bot passed a form of the Wise Man test. The story behind the Wise Man test has three WMs tested by the king; he makes them wear hats which are either blue or white: they cannot see their own hat but can see both of the others. They’re told that there is at least one blue hat, and that the test is fair; to be won by the first WM who correctly announces the colour of his own hat. There is a chain of logical reasoning which produces the right conclusion: we can cut to the chase by noticing that the test can’t be fair unless all the hats are the same colour, because all other arrangements give one WM an advantage. Since at least one hat is blue, they all are.

You’ll notice that this is essentially a test of logic, not self awareness. If solving the problem required being aware that you were one of the WMs then we who merely read about it wouldn’t be able to come up with the answer – because we’re not one of the WMs and couldn’t possibly have that awareness. But there’s sorta,  kinda something about working with other people’s point of view in there.

Bringsjord’s bots actually did something rather different. They were apparently told that two of the three had been given a ‘dumbing’ pill that stopped them from being able to speak (actually a switch had been turned off; were the robots really clever enough to understand that distinction and the difference between a pill and a switch?); then they were asked ‘did you get the dumbing pill?’  Only one, of course, could answer, and duly answered ‘I don’t know’: then, having heard its own voice, it was able to go on to say ‘Oh, wait, now I know…!”

This test is obviously different from the original in many ways; it doesn’t involve the same logic. Fairness, an essential factor in the original version, doesn’t matter here, and in fact the test is egregiously unfair; only one bot can possibly win. The bot version seems to rest mainly on the robot being able to distinguish its own voice from those of the others (of course the others couldn’t answer anyway; if they’d been really smart they would all have answered ‘I wasn’t dumbed’, knowing that if they had been dumbed the incorrect conclusion would never be uttered). It does perhaps have a broadly similar sorta, kinda relation to awareness of points of view.

I don’t propose to try to unpick the reasoning here any further: I doubt whether the experiment tells us much, but as presented in the New Scientist piece the logic is such a dog’s breakfast and the details are so scanty it’s impossible to get a proper idea of what is going on. I should say that I have no doubt Ringsjord’s actual presentation will be impeccably clear and well-justified in both its claims and its reasoning; foggy reports of clear research are more common than vice versa.

There’s a general problem here about the slipperiness of defining human qualities. Ever since Plato attempted to define a man as ‘a featherless biped’ and was gleefully refuted by Diogenes with a plucked chicken, every definition of the special quality that defines the human mind seems to be torpedoed by counter-examples. Part of the problem is a curious bind whereby the task of definition requires you to give a specific test task; but it is the very non-specific open-ended generality of human thought you’re trying to capture. This, I expect, is why so many specific tasks that once seemed definitively reserved for humans have eventually been performed by computers, which perhaps can do anything which is specified narrowly enough.

We don’t know exactly what Bringsjord’s bots did, and it matters. They could have been programmed explicitly just to do exactly what they did do, which is boring: they could have been given some general purpose module that does not terminate with the first answer and shows up well in these circumstances, which might well be of interest; or they could have been endowed with massive understanding of the real world significance of such matters as pills, switches, dumbness, wise men, and so on, which would be a miracle and raise the question of why Bringsjord was pissing about with such trivial experiments when he had such godlike machines to offer.

As I say, though, it’s a general problem. In my view, the absence of any details about how the Room works is one of the fatal flaws in John Searle’s Chinese Room thought experiment; arguably the same issue arises for the Turing Test. Would we award full personhood to a robot that could keep up a good conversation? I’m not sure I would unless I had a clear idea of how it worked.

I think there are two reasonable conclusions we can draw, both depressing. One is that we can’t devise a good test for human qualities because we simply don’t know what those qualities are, and we’ll have to solve that imponderable riddle before we can get anywhere. The other possibility is that the specialness of human thought is permanently indefinable. Something about that specialness involves genuine originality, breaking the system, transcending the existing rules; so just as the robots will eventually conquer any specific test we set up, the human mind will always leap out of whatever parameters we set up for it.

But who knows, maybe the Ro-Man conference will surprise us with new grounds for optimism.

Selfless Transfers

transferThe new film Self/less is based around the transfer of consciousness. An old man buys a new body to transfer into, and then finds that contrary to what he was told, it wasn’t grown specially: there was an original tenant who moreover, isn’t really gone. I understand that this is not a film that probes the metaphysics of the issue very deeply; it’s more about fight scenes; but the interesting thing is how readily we accept the idea of transferred consciousness.
In fact, it’s not at all a new idea; if memory serves, H.G.Wells wrote a short story on a similar theme; a fit young man with no family is approached by a rich old man in poor health who apparently wants to leave him all his fortune; then he finds himself transferred unwillingly to the collapsing ancient body and the old man making off in his fresh young one. In Wells’ version the twist is that the old man gets killed in a chance traffic accident, thereby dying before his old body does anyway.
The thing is, how could a transfer possibly work? In Wells’ story it’s apparently done with drugs, which is mysterious; more normally there’s some kind of brain-transfer helmet thing. It’s pretty much as though all you needed to do was run an EEG and then reverse the polarity. That makes no sense. I mean, scanning the brain in sufficient detail is mind-boggling to begin with, but the idea that you could then use much the same equipment to change the content of the mind is in another league of bogglement. Weather satellites record the meteorology of the world, but you cannot use them to reset it. This is why uploading your mind to a computer, while highly problematic, is far easier to entertain than transferring it to another biological body.
The big problem is that part of the content of the brain is, in effect, structural. It depends on which neurons are attached to which (and for that matter, which neurons there are), and on the strength and nature of that linkage. It’s true that neural activity is important too, and we can stimulate that electrically; even with induction gear that resembles the SF cliché: but we can’t actually restructure the brain that way.
The intuition that transfers should be possible perhaps rests on an idea that the brain is, as it were, basically hardware, and consciousness is essentially software; but isn’t really like that. You can’t run one person’s mind on another’s brain.
There is in fact no reason to suppose that there’s much of a read-across between brains: they may all be intractably unique. We know that there tends to be a similar regionalisation of functions in the brain, but there’s no guarantee that your neural encoding of ‘grandmother’ resembles mine or is similarly placed. Worse, it’s entirely possible that the ‘meaning’ of neural assemblages differs according to context and which other structures are connected, so that even if I could identify my ‘grandmother’ neurons, and graft them in in place of yours, they would have a different significance, or none.
Perhaps we need a more sophisticated and bespoke approach. First we thoroughly decipher both brains, and learn their own idiosyncratic encoding works. Then we work out a translator. This is a task of unimaginable complexity and particularity, but it’s not obviously impossible in principle. I think it’s likely that for each pair of brains you would need a unique translator: a universal one seems such an heroic aspiration that I really doubt its viability: a universal mind map would be an achievement of such interest and power that merely transferring minds would seem like time-wasting games by comparison.
I imagine that even once a translator had been achieved, it would normally only achieve partial success. There would be a limit to how far you could go with nano-bot microsurgery; and there might be certain inherent limitations. Certain ideas, certain memories, might just be impossible to accommodate in the new brain because of their incompatibility with structural or conceptual issues that were too deeply embedded to change; there would be certain limitations. The task you were undertaking would be like the job of turning Pride and Prejudice into Don Quixote simply by moving the words around and perhaps in a few key cases allowing yourself one or two anagrams: the result might be recognisable, but it wouldn’t be perfect. The transfer recipient would believe themselves to be Transferee, but they would have strange memory gaps and certain cognitive deficits, perhaps not unlike Alzheimer’s, as well as artefacts: little beliefs or tendencies that existed neither in Transferee or Recipient, but were generated incidentally through the process of reconstruction.
It’s a much more shadowy and unappealing picture, and it makes rather clearer the real killer: that though Recipient might come to resemble Transferee, they wouldn’t really be them.
In the end, we’re not data, or a program; we’re real and particular biological entities, and as such our ontology is radically different. I said above that the plausibility of transfers comes from thinking of consciousness as data, which I think is partly true: but perhaps there’s something else going on here; a very old mental habit of thinking of the soul as detachable and transferable. This might be another case where optimists about the capacity of IT are unconsciously much less materialist than they think.

Unimportant Consciousness

whistleAn interesting paper in Behavioural and Brain Sciences from Morsella, Godwin, Jantz, Krieger, and Gazzaley, reported here: an accessible pdf draft version is here.

It’s quite a complex, thoughtful paper, but the gist is clearly that consciousness doesn’t really do that much. The authors take the view that many functions generally regarded as conscious are in fact automatic and pre- or un-conscious: what consciousness hosts is not the process but the results. It looks to consciousness as though it’s doing the work, but really it isn’t.

In itself this is not a new view, of course. We’ve heard of other theories that base their interpretation on the idea that consciousness only deals with small nuggets of information fed to it by unconscious processes. Indeed, as the authors acknowledge, some take the view that consciousness does nothing at all: that it is an epiphenomenon, a causal dead end, adding no more to human behaviour than the whistle adds to the locomotive.

Morsella et al don’t go that far. In their view we’re lacking a clear idea of the prime function of consciousness; their Passive Frame Theory holds that the function is to constrain and direct skeletal muscle output, thereby yielding adaptive behaviour. I’d have thought quite a lot of unconscious processes, even simple reflexes, could be said to do that too; philosophically I think we’d look for a bit more clarity about the characteristic ways in which consciousness as opposed to instinct or other unconscious urges influence behaviour; but perhaps I’m nit-picking.

The authors certainly offer explanation as to what consciousness does. In their view, well-developed packages are delivered to consciousness from various unconscious functions. In consciousness these form a kind of combinatorial jigsaw, very regularly refreshed in a conscious/unconscious cycle; the key thing is that these packages are encapsulated and cannot influence each other. This is what distinguishes the theory from the widely popular idea of a Global Workspace, originated by Bernard Baars; no work is done on the conscious contents while they’re there. they just sit until refreshed or replaced.

The idea of encapsulation is made plausible by various examples. When we recognise an illusion, we don’t stop experiencing it; when we choose not to eat, we don’t stop feeling hungry, and so on. It’s clearly the case that sometimes this happens, but can we say that there are really no cases where one input alters our conscious perception of another? I suspect that any examples we might come up with would be deemed by Morsella et al to occur in pre-conscious processing and only seem to happen in consciousness: the danger with that is that the authors might end up simply disqualifying all counter-examples and thereby rendering their thesis unfalsifiable. It would help if we could have a sharper criterion of what is, and isn’t, within consciousness.

As I say, the authors do hold that consciousness influences behaviour, though not by its own functioning; instead it does so by, in effect, feeding other unconscious functions. An analogy with the internet is offered: the net facilitates all kind of functions; auctions, research, social interaction, filling the world with cat pictures, and so on; but it would be quite right to say that in itself it does any of these things.

That’s OK, but it seems to delegate an awful lot of things that we might have regarded as conscious cognitive activity to these later unconscious functions, and it would help to have more of an account of how they do their thing and how consciousness contrives to facilitate. It seems it merely brings things together, but how does that help? If they’re side by side but otherwise unprocessed, I’m not sure what the value of merely juxtaposing them (in some sense which is itself a little unclear) amounts to.

So I think there’s more to do if Passive Frame Theory is going to be a success; but it’s an interesting start.