Wednesday, October 27, 2010

TSRM's Shark Example 1


Here is TSRM's discussion of a case that is quite similar to the FAPs case, so I'll discuss it here:
Sharks electrically detect things to eat and things that impede locomotion (Kalmijn, 1974). An edible living thing such as a flatfish differs in ionic com• position from the surrounding water, producing a bioelectric field partially modulated in the rhythm of the living thing's respiratory movements. A flatfish that has buried itself in the sand will be detectable by a shark swimming just above it. Reproducing the bioelectric field of the flatfish artificially, bypassing a current between two electrodes buried in the sand, invites the same predatory behavior. The shark digs tenaciously at the source of the field departing from the site when the act fails to reveal an edible thing (Kalmijn, 1971). Now there is no intelligible sense in which it can be claimed that the source ought to have appeared inedible if the shark's perception were free of error and if the shark's perception of affordances were direct. In the niche of the shark 'an edible thing' and 'electric field of, say, type F' are nomically related. To predicate of the shark (a) 'detects electric field of type F' and (b) 'takes to be an edible thing' is not to refer to two different states of affairs, one (viz. (b)) that is reached from the other (viz. (a)) by an inference. Rather, it is to make reference in two ways to a single state of affairs of the shark-niche system. The linking of (a) and (b) is not something that goes on in the "mind" of the shark, as the Establishment would have it. The linking of (a) and (b) is in the physics of an ecological world, namely, that system given by the complementation of the shark and its niche.
First of all, this case is much like the stickleback case.  "Detects electric field of type F" is like the "detects red body of type R"  and "takes to be an edible thing" is like "takes to be a male stickleback".  (This is why I place the post here.)

Second, "Now there is no intelligible sense in which it can be claimed that the source ought to have appeared inedible if the shark's perception were free of error and if the shark's perception of affordances were direct"  The claim of "no intelligible sense" seems to me to be a bit strong, but it is probably right that just because a shark can (directly) perceive "edible" does not mean that it can (directly) perceive "inedible".  But, this is largely a misdirection on TSRM's part.  It will still turn out, I shall argue, that in the "non-standard" environments (i.e., in the experimental manipulation), the shark has the false perceptual belief that there is something edible below.  But, how can the shark pick up the information that there is something edible below if there is nothing edible below?  That's what the critic of EP is driving at.

22 comments:

  1. But, how can the shark pick up the information that there is something edible below if there is nothing edible below?
    Simple. It can do this when an evil psychologist has artificially simulated the bioelectric field of an edible fish (i.e. when the information has been made present).

    I like the following paragraph, which refines the 'intelligible sense' claim and addresses the 'false perceptual belief' claim:
    "But what of the shark’s actions? Should we not classify them as being in error, as being wrong? After all, the source of the electric field proved not to be an edible thing. Given the nomic relation between ‘electric field of type F’ and ‘edible’ there is no intelligible sense in which it can be said that the shark’s act of investigating the source of the field was wrong. The wrong action for the shark, given its niche and its appetite, would be not investigating the source of the field."

    The shark detects the field, and (by virtue of the nomic relation) detects 'edible fish'. The experiment violates the nomic relation but the shark still has the same perceptual experience (at least until exploration begins to reveal the problem).

    ReplyDelete
  2. I address that next paragraph in post #4 of the series, so I'll wait on that one.

    "The shark detects the field, and (by virtue of the nomic relation) detects 'edible fish'. The experiment violates the nomic relation but the shark still has the same perceptual experience"

    Ok, but that's no substantive reply to the objection. You're just saying that the shark has a false perceptual experience, rather than a false perceptual belief. There is still no "edible fish".

    ReplyDelete
  3. That's true. But until further exploration reveals that, it's weird to say the shark is in error at all, because he successfully detected the information which, in his niche, specifies fish. He can certainly come to learn that there's actually a problem; but that requires more information. At the point where he detects the field specifying 'fish', he isn't in error; perhaps more specifically, he can't know he's in error, and the only person able to describe the situation this way is the evil psychologist who has some privileged access to the ruse.

    This used to weird me out a bit too; but my PhD advisor used to talk about the fact that a perceiver doesn't ever get to 'peer behind the curtain' - information is all you ever get and from the perspective of the perceiver there is no possibility of the privileged access required to talk about 'false beliefs'. You only perceive the world by virtue of information's relation to the world. If that information has been completely recreated by an evil psychologist, there's no basis to know this (until, perhaps, further exploration reveals the inconsistency, a la the Ames Room).

    I guess one way to think about this is as two levels of analysis; from the perspective of the shark, there's no obvious way to talk about him having a false belief at time t. From the perspective of the evil psychologist, that is a way to talk about the shark. But actually this smacks of James' 'psychologist's fallacy'; taking your description and making it the explanation.

    ReplyDelete
  4. "That's true. But until further exploration reveals that, it's weird to say the shark is in error at all"

    Actually, I think it's pretty standard to say that when the shark first detects the electric field, then the shark believes (incorrrectly) that there is something edible in the sand. So, given this (false) belief, it begins to dig around in the sand. But, after digging it abandons the (false) believe that there is something edible in the sand, so stops digging.

    If you say that "But until further exploration reveals that, it's weird to say the shark is in error at all", that means that the whole time the shark believes that there is nothing edible in the sand, right? So, why is the shark digging in the sand in the first place, if it doesn't believe that there is something in the sand? And, why does it give up digging in the sand?

    "my PhD advisor used to talk about the fact that a perceiver doesn't ever get to 'peer behind the curtain' -" Yes, philosophers often use the metaphor of a "veil of ideas". But, these days the "veil" is often described as representations. (So, your advisor runs the risk of being a heterodox EPist.) But, really, there is an issue of how one can know if one's beliefs are true.

    ReplyDelete
  5. Geoff's running no such risk because he wasn't talking about your 'veil of ideas'. He's talking about the fact that perceiving organisms detect information, not the world. We have knowledge about the world because the information we detect stands in an informative, specification relationship to the world. So you only ever get to know if your beliefs are true if you have access to information about that; you never get to 'peek behind the curtain' of information to see whether you're being tricked by a psychologist in the world.

    It can only be standard to say there's a false belief from an objective, third person perspective which isn't how perception works; thus describing it that way, it's you who runs the risk, of making the psychologist's fallacy.

    If you say that "But until further exploration reveals that, it's weird to say the shark is in error at all", that means that the whole time the shark believes that there is nothing edible in the sand, right?
    No. At time t, the shark believes there's a fish there because that's what the information he detected specifies. He's correct in that belief. You're cursed with too much knowledge in this case, hence you know he's in error because his niche (within who's scope that specification relation holds) has been breached. He doesn't know that, nor will he till the exploration fails to reveal the fish.

    For eco psych, the job of perception is not to bring your mental representation of the world into correspondence with the world.

    ReplyDelete
  6. "At time t, the shark believes there's a fish there because that's what the information he detected specifies. He's correct in that belief. You're cursed with too much knowledge in this case, hence you know he's in error because his niche (within who's scope that specification relation holds) has been breached. "

    Ok. Let's be a bit more careful still. At time t, the shark believes that there is a fish in the sand. That belief is false. You can say that the belief is "correct" in the sense that it was formed by way of what is usually an ecologically reliable method. Nonetheless, the correctly formed belief is false.

    We do, indeed, have to be careful to distinguish our epistemic position from that of the shark. We know at time t that the shark has a false belief, even though the shark does not. We know at time t+1 that the shark has a true belief (there is no fish to be eaten). We know all along that there is no fish, but the shark at first believes there is a fish, but then later believes there is no fish.

    So, yes, it is an objective, third person claim that the shark is in error.

    ReplyDelete
  7. So, the "veil of information" and the "veil of ideas" are not the same, but they have the same apparent consequences. Once you have the veil, then there is the risk that what is behind the veil does not correspond to the world. This opens the door to scepticism, which is apparently what motivated Gibson to try to get rid of representation and to replace it with something for which there has to be a correspondence. Once information fails to uniquely specify the world, the problem of scepticism looms once again.

    ReplyDelete
  8. But, go back to the phenomena. For a while the shark searches, then the shark stops. Why?

    Standard account: At first the shark (falsely) believes that there is a fish in the sand, then after digging comes to believe (truly) that there is no fish in the sand. That's why it digs for a while then stops.

    EP account: _________________________.

    ReplyDelete
  9. So, yes, it is an objective, third person claim that the shark is in error.
    And this is important, because the science of perception should be interested in what it is possible for an individual to know about, and the mechanism supporting that. Hence Turvey et al and my insistence that the claim the shark is in error isn't the right way to talk about perception.

    Once you have the veil, then there is the risk that what is behind the veil does not correspond to the world.
    Well, for eco-psych, what's behind the veil is the world; it's you that's in front of it (if that makes sense). And correspondence isn't the game.

    But besides that; this is why this paper is so important. If eco-psych just moves the veil, so to speak, then we haven't really achieved anything. This paper, though, is laying out a veil that is transparent in one very specific sense; that is, because the veil is created by lawful processes, you can perceive the world by virtue of detecting the veil. The potential for ambiguity or error in the signal is then removed when the veil is actually created by those lawful processes.

    Now, nothing is perfect, and this set up has compromises like every evolved system:

    1. You can be in error by not detecting the information for some reason; plenty of ways that bit can break (lack of skill, a broken detector, etc)

    2. Laws have scope, and the laws governing perception are ecological in scope. They are thus more breakable than the laws of physics, although they are still sufficiently robust to support this analysis. You can get "errors", problems, etc like the shark example when the law has been broken (ie when the evil psychologist does his thing).

    But, go back to the phenomena. For a while the shark searches, then the shark stops. Why?
    The shark detects information which (within the scope of it's niche) specifies the presence of a fish, and this begins to dig. After digging for a while, it detects that there was no fish (maybe the power gets turned off, or it tries to eat the fake) and thus stops digging.

    This is much more a theory of perception than your account, because I'm not enriching my analysis with anything the shark cannot possibly know. It has mechanism too, which your account doesn't after you strip out the 'false' and 'true' bits you aren't allowed (and they aren't especially good mechanisms anyway; what do they mean?)

    ReplyDelete
  10. "If eco-psych just moves the veil, so to speak, then we haven't really achieved anything."

    Precisely. That's what I'm driving at. If one concedes, as you do, that nothing is perfect as in case 2. of laws having scope, then the move to Gibsonian information doesn't solve the sceptical problem.

    Now, the appeal to scope might. This is among the moves that reliabilist epistemologists, such as Dretske, make within a form of representationalism.

    ReplyDelete
  11. "because I'm not enriching my analysis with anything the shark cannot possibly know. It has mechanism too, which your account doesn't after you strip out the 'false' and 'true' bits you aren't allowed (and they aren't especially good mechanisms anyway; what do they mean?) "

    True and false are not part of the account, hence the reason I bracketed them. So, the explanation is at time t, the shark detected the electric field, so came to believe that there was something edible in the sand. After digging in the sand, it came to believe that there was not something edible in the sand. So, it stopped digging.

    On the TSRM account, the shark detects the electric field the whole time, so takes there to be a fish there the whole time. But, what you want to say is that by digging around the shark gets more Gibsonian info, so takes there not to be a fish. But, the shark then only comes to have inconsistent beliefs, namely, there is a fish and there is not a fish.

    ReplyDelete
  12. You still lack mechanism, so science-wise I think I'm ahead. So far you just have the sentence 'the shark came to the belief'. How? On what basis? TSRM get to say, on the basis of information obtained via it's activities. And they aren't inconsistent; the shark changed it's belief with new evidence. So your last sentence should read, 'namely, that there was a fish and then that there is not a fish'.

    And Gibsonian information does solve the sceptical problem, iff we can also establish that the scope is adequate. I admit that without the second bit we would have just moved the veil; but part of the eco psych research programme is defining the scope!

    An example: tau, the proposed variable for perceiving time-to-contact, came to be recognised as inadequate because the scope over which it works (objects moving at constant velocity right towards you) just isn't big enough. But it was a perfectly sensible idea and the studies that were run were productive and eventually established the problem with the scope was a real issue.

    Scope is always an part of the analysis (this is the task analysis stuff we were talking about ages ago). My work on coordination dynamics is busy identifying the information and the scope over which it is used.

    ReplyDelete
  13. "You still lack mechanism, so science-wise I think I'm ahead. So far you just have the sentence 'the shark came to the belief'. How? On what basis? "

    The mechanism will have to do with shark anatomy and physiology, which I personally do not happen to know. I am a philosopher, after all.

    "TSRM get to say, on the basis of information obtained via it's activities. And they aren't inconsistent; the shark changed it's belief with new evidence. So your last sentence should read, 'namely, that there was a fish and then that there is not a fish'."

    But, I'm not saying that TSRM are inconsistent. I'm saying that the shark will have inconsistent beliefs. Even more explicit, it looks like the TSRM account is that the shark initially thought there was a fish and then there is a fish and there is not a fish. What keeps the shark thinking that there is a fish the whole time is that it is detecting an electric field the whole time and that (according to TSRM) is the same thing as taking there to be a fish. So, we still don't know why the shark gives us looking. Why does the contradiction lead to the stopping of digging, rather than the continuation of digging?

    ReplyDelete
  14. Sorry, I meant the shark wasn't inconsistent, not TSRM. That bit should read 'and the beliefs aren't inconsistent'.

    This isn't a reflex; the digging isn't mandatory while the signal is being detected. If the digging isn't being successful, then the shark can just give up once it's perceived there's nothing edible to be had.

    ReplyDelete
  15. Ok. Digging is not a reflex and not mandatory (unlike the FAP, I suppose), but, when the shark eventually gets inconsistent beliefs, there is a fish and there is no fish, why does the shark quit digging? Sure, it would give up digging if it just perceives there is nothing edible. But, it doesn't just perceive that there is nothing edible. It perceives both that there is something edible and that there is not something edible. So, why does it give up?

    ReplyDelete
  16. TSRM seem to me to get into a jam here because of their view that "detecting an electric field" is the same thing as "taking there to be something edible". I say they are different, so the shark can continue to detect the electric field, but then on the basis of further evidence gathered by digging some to abandon the belief that there is something edible in the sane in favor of the belief that there is nothing edible in the sand. That's why the shark digs for a bit, then gives up. You tell what sounds prima facie like the right story, because you are omitted the embarrassing part about "detecting electric field F" being the same thing as "taking there to be something edible".

    ReplyDelete
  17. It's hardly embarrassing; it's the core of the laws argument. Perception of 'edible' needs a mechanism; lawful information is a lot less embarrassing than beliefs, and it actually gets you out of a jam, by being a mechanism.

    And what was the further evidence gathered by digging? Why, it was more information - perhaps olfactory, or visual - about the absence of a fish even in the presence of the field F. Turns out we have multiple perceptual systems which we use to inform and calibrate themselves and each other over extended periods of time spent exploring. The shark gives up because it turns out there isn't a fish there. It probably also gives up because it only has so much energy to waste on unsuccessful foraging that is generating conflicting information.

    Smartness gets back in here; this is an equivalent configuration the shark isn't equipped to disambiguate because it doesn't arise in it's niche without being meticulously reverse engineered by an evil psychologist (well, biologist here).

    ReplyDelete
  18. But, you keep talking around the problem. The shark continues to take there to be something edible. So, why give up? The shark later takes there not to be something edible, but still takes there to be something edible. So, why quit digging?

    ReplyDelete
  19. Moreover, the issue does not depend on belief. The problem can be run entirely on "takes to be an edible thing".

    ReplyDelete
  20. And you still need an account of how it takes there to be an edible thing.

    The shark gives up because it turns out there isn't a fish there. How is this hard?

    ReplyDelete
  21. It's true that there isn't a fish there and that the shark takes there to be no fish there. The problem for TSRM is that the shark also takes there to be a fish there. It's easy if you ignore the second sentence. But, that's the problem.

    ReplyDelete
  22. This looks like the "shoe"-shoe problem again. You know the behavior, so you can freely invoke whatever Gibsonian information you want to "explain it". You know the shark swims away, so the law must invoke there not being a shark.

    ReplyDelete