A blog about problems in the field of psychology and attempts to fix them.

Saturday, January 19, 2013

Why it is important that we see the things we see

A little bit ago there was a nice post over on Gary William's blog about "Panpsychism vs. Inogranicism" - usually taken as two extremes at which either everything is said to have mental properties to some degree (panpsychism) or the existence of mental things is denied across the board (inorganicism). This grew out of a book review from a few months ago, in which Gary claimed the author endorsed panpsychism, for the slightly less extreme position of claiming at least some inorganic things should be considered to have mental properties. At any rate, the comments on the post got interesting, and I was posed a question that would take too long to answer in there... so I'll try to answer it here... The answer requires a bit of discussion about radical empiricism and realism, in particular regarding color perception. These are crucial issues for creating a psychological science with any chance of making sense. 


The rest of the back story is pretty quick: Someone named "ledge" started saying things that sounded in line with William James, to which I replied...
have you read much William James? (Late career William James?) I think you would like it. He makes the types of points you are making, and places them as the basic facts philosophers have to work with. (e.g., When I experience something as red, the one thing we shouldn’t deny, for God’s sake, is that I am experiencing something as red!)…
Charles Wolverton (who also asked a question that also lead to the first post on this blog) wanted me to elaborate on the parenthetical. As he put it:
I’d like to understand precisely what you mean by this. In one interpretation, it seems to me a fundamental error.

... once you’re past light emitted by an object and stimulating the retina, and focus on the brain, there’s nothing to observe but neural activity. Ie, there’s nothing (and in particular, no thing) to “experience as red” in any sense that involves color other than some words uttered (or written, etc) in response to that activity (eg, “red”, “rot”, “czerwony”, etc)....
Since it’s considerably more likely that I would make a fundamental error than that you would, probably I’m misinterpreting that quote. But if not, I’d like to know where you think I’m on the wrong tack.
"Odd," I must reply, with my William James hat on, "it seems to me that the basic fact of experience we have been asked to explain is that we see a red object, and I think all types of trouble enter psychology when we lose track of that basic fact."

It took me a while to wrap my head around this way of thinking, and when I finally did, I had a big "Aha!" moment regarding William James. Not necessarily "Aha, he is right!" but definitely an "Aha, THAT is what he's trying to do!"

Imagine if I was a car repairman, and you came into my garage and asked me to fix his engine because it is making odd knocking sounds. In response to your request, I replied, "Well, you don't really have an engine, just a collection of metal bits. And, there isn't a knocking sound, just some oddly compressed air waves." You insist that you hear the engine knocking, and I keep trying to explain that there is no engine, and, even if there was, you certainly couldn't "hear" it knocking because there is no such things as "knocking noises", only airwaves. You might, at some point give in and ask me to fix your metal bits and airwaves. As soon as you do, I ask on what basis you judge the metal bits flawed and certain airwaves more desirable than others - especially given that you can never know anything about cars based on them. We would certainly be at an impasse, wouldn't we?

This same problem happens all the time to psychologists. You ask "Can you explain what is happening when I see the red apple?" and the first reply you get is "On what basis do you determine the category 'apple.' Oh, and by the way, there is no such things as 'red' so you certainly couldn't have seen it." At that point, any attempt at doing psychology is at an impasse.

After a few back and forths, you might give in and ask "Can you explain my belief that I am seeing an apple and my belief that it is red?" When you do that, the first reply you are likely to get is "Well, you don't have a justified belief that it is 'an apple', that's for sure. And what do I care how people form un-justified beliefs? Oh, and you believe you see 'red' because you have neurons that are configured in some way I don't care to specify and firing in some way I don't care to specify."  Again, we are at an impasse... and it is no longer clear we even have psychological questions! We are going down some odd philosophical rabbit hole and some odd neuroscience rabbit hole, and whatever we find at the bottom, it will not do much to answer your original question, which we have both given up on.

To answer your original question, at some point we must allow that you saw a red apple. That is the basic, factual, thing-to-be-explained. We might well explain your seeing the red apple by reference to photons of a given wave length, photo-receptors in the retina, neuronal firings, etc. But we can't let such explanatory attempts "explain away" the descriptive fact that you saw a red apple. This is in the same sense that basic premises of physics can't be allowed to paralyze the automobile repairman, or our cars will never get fixed.

Imagine if we really took the "you only see photon's" talk seriously. A preschool teacher asks a perceptual psychologist the best way to teach colors to her students. I.e., she wants them to say "red" when she holds up the red card, and "green" when she holds up the green card... But we tell her that's impossible because there really is no red card or green card. That would be, to say the least, Not Helpful.

No, something else must be possible. What if, when you ask "Can you explain what is happening when I see the red apple?" I say, "Sure." And then we go about trying to explain how you see the red apple. That seems like the only way to create a science of psychology - we determine which factors affect your perception, how their relate at the moment and in the course of development, etc. So long as we allow that you see a red apple, we can do these things... but as soon as we let someone assert a priori that you do not see a red apple... even when a red apple is right in front of you and you are looking at it... then we have a big problem.

Sure, at some point we will have to reconcile our findings with physics and neuroscience, but that is a completely separate issue. Nobody makes the biologist stop doing biology until they can translate their work into some other science's language, and we shouldn't refrain from doing psychology just because a bunch of neuroscientists and physicists have their panties in a bunch.

Does that make any sense?

22 comments:

  1. Thanks for the long reply, Eric. Ironically, after touting the importance of choosing the appropriate vocabulary for a purpose, I chose entirely the wrong one to make my point about vision. You are, of course, correct that the vocabulary of ontology is inappropriate for an auto repair shop and that the vocabularies of philosophical idealism and photons are inappropriate in the elementary school classroom. I'm certainly inclined to agree that talk of photons and neurons would not be helpful in psychological counseling, and it clearly wasn't helpful in making my point since it misled you. Mea culpa.

    But my comment/question wasn't intended to suggest that the vocabulary of neurology is appropriate for psychological counseling but rather was an attempt understand how you are interpreting expressions like "experiencing something as red" and "allow[ing] that you see a red apple". Motivated partly by Gary's post and its comments, I've been doing yet another reread of my "bible" for such matters, the DeVries and Triplett exposition of Sellars' "Empiricism and Phil of Mind". Hopefully I can now frame my concerns in terms that perhaps are more relevant to psychology.

    Our every day use of the word "seeing" ignores significant ambiguity. For his epistemological purposes, Sellars interprets "I am seeing that a red apple is over there" to mean that the subject is asserting the proposition "there is indeed an apple over there and it indeed is red". However, in some situations, the apple actually isn't "red", in others it isn't even "over there". I assume that such illusions can be concerns in psychology. But while it obviously would be pointless to challenge a naive subject's misuse of "seeing that", wouldn't it be appropriate for a psychologist talking to a peer to switch to a vocabulary (in Sellars' terms, the vocabulary of "looking as if") that resolves any ambiguity by clearly distinguishing those three situations?

    The word "seeing" is also ambiguous as to who is seeing what where and by what process. In casual conversation, the common assumption is that "seeing an apple over there" is essentially a partial description of the content of the speaker's current FOV. A slightly more sophisticated speaker may instead claim to be describing the "mental image" representing the content of the FOV. A speaker familiar with blindsight may be reluctant to think of "seeing" as having much - if anything - to do with visual images, instead preferring to speak in terms of detecting patterns of neural activity (that would be me). Aren't there situations in which it would be appropriate for a psychologist to shift from the imprecise every day use of "seeing" to the more precise vocabulary of one of the other sciences?

    Perhaps even more problematic is "experiencing". Sellars distinguishes between that word's use in the non-epistemic sense of "undergoing" (eg, pain) and its epistemic use as a precursor to knowledge acquisition. The expression "experiencing something as ..." suggests the latter use, but without some context it strikes me as reasonable to wonder exactly how "When I experience something as red, the one thing we shouldn’t deny ... is that I am experiencing something as red!" is to be interpreted. I can imagine several possible interpretations, including assuming "experience" to be synonymous with "experiencing" - in which case the assertion is a tautology and hence trivially undeniable.

    Some comments in response to Gary's post seemed to verge on "givenism", the target of Sellars' EPM, so I was concerned that someone (quite possibly yours truly) might be missing something.



    ReplyDelete
  2. Charles,
    Thank you for the thoughtful comment. Lots to think about. I'll reply in parts to make it easier. There will be some repetition; sorry about that.

    that the vocabulary of neurology is appropriate for psychological counseling but rather was an attempt understand how you are interpreting expressions like "experiencing something as red"

    I am suspicious of both.

    Sellars interprets "I am seeing that a red apple is over there" to mean that the subject is asserting the proposition "there is indeed an apple over there and it indeed is red".

    So, "I see a red apple." is considered equivalent to "See, a red apple!" I'm not sure about that, but I'm willing to consider it. Nick Thompson, used to claim that all these statements should have to specify the organism and the conditions, so that a full statement would be something like "To me, from here, a red apple there." That always struck me as good convention, but I was unsure if it was necessary.

    However, in some situations, the apple actually isn't "red", in others it isn't even "over there". I assume that such illusions can be concerns in psychology.

    Indeed, such events are a prime concern of psychology, but to pre-label them as "illusions" is problematic. That is, to call those events illusions is to already be offering a complex explanation for the events in question - presumptuously prioritizing one person's judgment over the other's. One key to understanding William James is that he was very suspicious of circumstances in which we let one person's experience trump another persons. Who gets to decide the apple is not red, and on what basis?

    Note the double bind if the privileged "scientist" tries to simultaneously declare 1) that "true color" is determined by wavelength, 2) that apple is "really green" and 3) that "color" is only something that occurs in the brain.

    ReplyDelete
    Replies
    1. Indeed, such events are a prime concern of psychology, but to pre-label them as "illusions" is problematic.

      Here is a paragraph from the chapter Mathematical Models and Social Representation by Valsiner and Rudolph in Qualitative Mathematics for the Social Sciences; Jaan is entirely innocent of it.

      [T]he traditional use of the word “illusion” is tendentious; it can be disputed along the following lines (see also Carini, 2007). Start with the axiom that a ‘whole’ that is perceived is ipso facto ‘correctly’ perceived. Then, for the person who perceives a whole, what is—or may be—‘illusory’ is not the whole: it is the felt need or imposed demand to identify the perceptually present and correctly perceived whole as something else, namely, a certain unperceived whole that is perceptually and physically absent from the present situation of the perceiver (and might even be physically absent from the entire universe, past, present, and future—if, say, it is a “round triangle”). Contrariwise, for a(nother, or the same) person (perhaps a psychologist) who is observing the situation, what is illusory is the conviction that the ‘whole’ known to the perceiver is in some manner or degree less (or more) ‘real’ than the ‘unwhole’ known to the observer, which the perceiver somehow should and would be perceiving—were not the universe (or the observer) somehow setting successful snares. On this view, the ascription of ‘illusion’ is a category error, a failure of the ascriber’s (formal or informal) ontology and epistemology to adequately fit the phenomena of construction by the human mind (starting with the human perceptual system).


      Delete
    2. Interesting take. I'm not sure all illusions would fit under this umbrella, but certainly the majority of so-called "perceptual illusions" would.

      Delete
    3. I find it difficult to decipher the quoted passage.

      First, "a ‘whole’ that is perceived is ipso facto ‘correctly’ perceived" seems not an "axiom" but arguably false. I infer from "'correctly' perceived" that the perceiving is epistemic, ie, can cause a statement like "I am perceiving that an X is over there", which - like "seeing that" - may be intended to endorse assertion of the proposition "over there is an X". But we know that such assertions can be incorrect. There may be no X over there or anywhere else. In which the supposedly perceived X is an illusion. (See note below.)

      Putting "correctly" in quotes may be intended by the authors to suggest non-epistemic perceiving, presumably a special case of the kind of event that Sellars calls a non-epistemic experiencing, Then I take the point of the passage to be that one must be careful not to confuse these:

      a. non-epistemic experiencing of sensory stimulation

      b. verbal responses consequent to that experiencing

      I agree but don't see what that has to do with "illusion". A possible verbal response to such an experiencing could be "I am perceiving a red apple". We've addressed such a response when considered to be assertion of a proposition but are now addressing non-epistemic perception. So, it has to be interpreted differently. One option is to interpret it as a description, but not of something external but of something internal. I question this interpretation (I think "description" is misleading), but many seem to accept it. So, for present purposes I'll acquiesce. But I don't know what it would mean to say that such a "description" is correct or incorrect. To say that it is one or the other is to suggest that it agrees or disagrees with some other "description" of one's experiencing, one known to be accurate. But what could that possibly mean? Who else can describe your experiencing at all, never mind more accurately than you?

      In short, I think the quoted passage unconvincing in its attempt to disparage use of the word "illusion". Like any word, it can be misused, and the hypothesized psychologist may be doing so. But my use was explicitly in an epistemic context, so I don't see the passage's relevance.

      Note: dictionary.com defines "illusion" in terms of correspondence to reality. I would define it in terms of "truth", interpreting that word in the sense of assertions accepted based on justifications offered in the space of reasons - what I take to be the serious sense behind Rorty's quip "truth is what your peers let you get away with saying".

      Delete
  3. wouldn't it be appropriate for a psychologist talking to a peer to switch to a vocabulary (in Sellars' terms, the vocabulary of "looking as if")

    Sure... if we really had a better handle on that vocabulary, and it better served our technical purposes. "I See" seems more concrete to me than "It looks to me as if". Perhaps Sellars is trying to under that apparent concreteness, but I find it appealing. To switch entirely to a vocabulary of "looking as if" seems to invite dualism with open arms, if not full idealism.

    The word "seeing" is also ambiguous as to who is seeing what where and by what process.

    Yes, yes! And there the task of the psychologist begins... but no amount of analysis should be allowed to lose our site of the whole.

    Aren't there situations in which it would be appropriate for a psychologist to shift from the imprecise every day use of "seeing" to the more precise vocabulary of one of the other sciences?

    Yes again... but "more precise" cannot be understood to mean "provided by another science, which we assume to be more precise". Part of psychologists' physics and neuroscience envy is the blind assumption that their vocabularies are more precise. I am not convinced that vague hand waving at neurons, or parts of the optic cortex, or photons, provides anything more precise. To say that someone has certain neurons we can't specify firing right now, is not more precise than saying that they see a red apple. And (returning to the prior problem) if we could specify the neurons, then we would be more precise, but in a way that adds only a small insight into the larger process being inquired of.

    ReplyDelete

  4. it strikes me as reasonable to wonder exactly how "When I experience something as red, the one thing we shouldn’t deny ... is that I am experiencing something as red!" is to be interpreted. I can imagine several possible interpretations, including assuming "experience" to be synonymous with "experiencing" - in which case the assertion is a tautology and hence trivially undeniable.

    Well... I wouldn't be altogether sad with that interpretation. People who get their foot stuck to this philosophical Tar Baby often seem to deny things that I think are tautologically true. In some approaches to dualism and idealism, for example, it is taken as axiomatic that if I believe I have experienced X, then whatever I might have experienced, it sure as hell wasn't an actual X. To accept my claim even as a tautology would undermine these approaches.

    Some comments in response to Gary's post seemed to verge on "givenism", the target of Sellars' EPM, so I was concerned that someone (quite possibly yours truly) might be missing something.

    My interest is piqued... tell me more! What would it mean to endorse "givenism"?

    ReplyDelete
  5. Eric -

    First, let me say how much I appreciate the opportunity to have an exchange like this. Having to defend a position against informed critique highlights the deficiencies in one's understanding of the position. In this case, yet another trip back to a well-worn section of my deVries/Triplett exposition of EPM revealed a couple of misconceptions I've had for the 2-plus years I've been battling the essay. (Altho pretty slow, I'm not quite as slow as that suggests. As philosophy profs, they spent 10 years, some of those with other colleagues, trying understand EPM, finally producing a 180 page exposition of the 100 page essay!)

    I'll start with "the given" since that will introduce ideas that should make responses to your other issues easier to phrase. The general idea of the given is that for each person there is a foundational set of primitive epistemic entities that do not depend on any other "knowledge" (defined in a way that will emerge later) and are epistemically efficacious in that acquiring other knowledge does depend on them - directly or indirectly - as part of learning, inferring, et al. The entities in that set can be known "directly", "immediately" - ie, in some way that doesn't require intersubjectivity. The existence of such foundational sets is what Sellars calls "the myth of the given". Detailing the many variants on the myth occupies several chapters of EPM, but I think this summary captures what we need here.

    The issue about "seeing that" versus "looks" has to do with "endorsement", an aspect that Sellars considers key and has to do with justification. His classic statement defining "knowledge" is roughly that to be in a state of knowing is to be in the "space of reasons", ie, of being able to justify a propositional assertion to one's peers by giving reasons for it. Asserting a proposition amounts to endorsing it, so Sellars wants to convert the quotidian "I see a red object" into explicit assertion of a proposition. He does that by stating that "I am seeing that p" is to be understood as asserting p. The problem is that in general, one can't endorse p (ie, be prepared to justify the assertion to peers) without additional knowledge about the environment. Eg, if p = "over there is a red object", to assert p one needs additional knowledge that indicates the presence of the supposedly "seen" object, knowledge of the lighting conditions, et al. In other words, seeing may be believing, but by itself it can't be knowing because it can't meet the independence requirement of "the given". (And if something as basic as a quale (color) can't be a given, what can?) In situations in which inadequate additional knowledge is available, one can only assert "I see that over there is an apple and it looks red", which is endorsement of the apple's presence but not that it "is red". Or only assert that "it looks as if there is a red apple over there" which endorses neither.

    Someone who actually understands EPM would no doubt tear that paragraph to shreds, but I hope it adequately captures the points relevant to this exchange and explains why I am concerned by statements that could be interpreted as something like "I just know red when I see it". Not so! One reacts to a range of non-epistemic sensory experiences by uttering "red" because of having learned to do so. If challenged to provide reasons, one can say in response that "red" is the word one has been taught to utter when having such sensory experiences and that doing so seems to allow one to cope in a society. But the intersubjectivities involved in that response violate a requirement of the supposed foundational set that constitutes the given.

    ReplyDelete
    Replies
    1. Hmmmm...

      It sounds in the beginning like givens are supposed to be bad (e.g., the myth of the given), but later on it seems like "the independence requirement of the given" is something we want.

      Based on the first part, I thought I was in agreement, as I too don't like "innate" givens of any type. But I'm also suspicious around "justification" talk. Surely we allow for "knowledge" in some situations where a person cannot justify their claim... don't we?

      What do you think Sellers would make of affordance talk? I.e., that "I see a P" is to assert that certain actions I may take will have certain consequences?

      Delete
    2. It's not a matter of either "the given" or the independence requirement for it being good or bad. The question are whether there can be a knowledge base that is given rather than acquired, and assuming there can be, what characteristics must its members have. The "myth" according to Sellars is the assumption that there can be such a base.

      Surely we allow for "knowledge" in some situations where a person cannot justify their claim... don't we?

      Apparently some heavyweights have argued that justification is not required for certain things things that we "just know", and deVries and Triplett disagree on how effective those arguments are. My simplistic take is that there are degrees of requiring justification. Eg, some propositions are justified by essentially being definitions; eg, "the capital of Texas is Austin". Others could be justified if necessary, but never need be because no one cares to dispute them; eg, "I was born in Texas".

      Assertion of some propositions may be justified partially by the reliability of the person asserting and the quality of any referenced sources. I've come to consider determining when that should be accepted as sufficient justification for one's knowledge to be an acquired skill, one that is much rarer than many think (recall the saying that "there are those who know, those who don't know, and those who don't know they don't know"). I think the source of many current problems in the US is that there are way too many in that third group. So, I'd be inclined to accept that if you can't justify your assertions, they simply don't qualify as knowledge, but to have a low threshold for what counts as justification for a lot of uncontroversial assertions.

      The only Sellars I've read is EPM and therefore have no opinion on what he'd think about affordances.

      Delete
    3. Googling Sellars and affordance led me to this blog by Teed Rockwell and in particular, this very interesting paper which compares ideas of Dewey and Sellars relevant to our discussion, explores the relationships between those ideas and connectionism, and notes some relationships with Gibson's ecological approach. I have skimmed the paper and plan to revisit it soon. I assume it will be of interest to you as well.

      Although I had heard of connectionism before reading the paper, I didn't know anything specific about it. Rockwell explains connectionism by viewing the individual neurons in a (biological) network as component values in a very large dimension vector space and specifying that the response to an input is determined by the region of input vector space into which the input vector falls. This is similar to the way I think of the process except that I think in terms of banks of special electronic filters called "matched filters" that effectively also divide input space into "regions". Here too the response to an input viewed as a "signal" is determined by the region of "input signal space" into which the input signal falls.

      Delete
    4. Thanks for the link. I will check it out!

      Delete
  6. On to specific responses.

    EC: Sellars interprets "I am seeing that a red apple is over there" to mean that the subject is asserting the proposition "there is indeed an apple over there and it indeed is red". (from previous CW comment)

    So, "I see a red apple." is considered equivalent to "See, a red apple!"

    CW: I would avoid both. As noted, for Sellars the precise phrase "seeing that p" is to be understood as asserting proposition p. Because of the subtlety of this material, I try to be very careful in wording things. Of course, I often fail, but here the only change I might make would be to separate the existential and qualitative propositions to emphasize that they can be asserted and endorsed independently.

    EC: Nick Thompson, used to claim that all these statements should have to specify the organism and the conditions

    CW: Since we're addressing the epistemic role of seeing and requiring intersubjectivity, the organism presumably is human. And per my previous comment, I clearly agree about the importance of being explicit about conditions/environment when "statements" are understood as being assertions of propositions.

    EC: However, in some situations, the apple actually "is not red", in others it actually "is not over there". I assume that such illusions can be concerns in psychology. (from previous CW comment)

    ... to pre-label them as "illusions" is ... to already be offering a complex explanation for the events in question - presumptuously prioritizing one person's judgment over the other's.

    CW: The issue you raise is part methodology, part epistemology. Re the former, I assume that a relevant experiment meets requirements for credibility such as being multiply witnessed, repeatable, documented, etc - what I take to be requirements of the scientific method. Re the latter, I assume that the experimenter(s) could, if challenged, justify (ie, give reasons for) their positions re the existential and qualitative states of the apple supposedly seen by the subject. Assuming acceptance of those reasons by peers, they can be said to "know" the status of the apple and that the subject's statement "I am seeing that ..." (explicitly postulated in my comment) indicates that the subject is experiencing illusions. (Under these assumptions, W. James' suspicions re an isolated "knower" seem inapplicable.)

    EC: Note the double bind if the privileged "scientist" tries to simultaneously declare 1) that "true color" is determined by wavelength, 2) that apple is "really green" and 3) that "color" is only something that occurs in the brain.

    CW: The way I like to express the idea presumably behind "true color is determined by wavelength" is that light at a certain wavelength will typically evoke from "normal" subjects the same word, eg, "red" or if a subject is especially good at distinguishing and naming color sensory experiences, perhaps a variant like "crimson". The phrase "true color" is probably better avoided since as I recall, it has a technical meaning that is irrelevant here.

    that apple is "really green" presumably is equivalent to p = "the apple over there is green". If the scientist means "I am seeing that p" and is credible, then s/he should be able to justify the assertion as just described.

    IMO, we shouldn't say obscure things like "color" is only something that occurs in the brain, since whatever is intended by such statements surely can be precisely and understandably phrased using some appropriate vocabulary.

    ReplyDelete
    Replies
    1. CW: Since we're addressing the epistemic role of seeing and requiring intersubjectivity, the organism presumably is human.

      Well... I'm not sure if I agree with it... but Nick thought we need to be even more specific. Not, "To a human, from a general angle, red apple.", but "To THAT person, from EXACTLY here, red apple."

      ------

      Re: Illusions, with the William James hat on\:
      Surely both the experimenter and the participant can provide SOME sort of justification for their opinions on whether a pair of lines are angled or parallel. Surely, the experimenter can get other experimenters to agree that the lines are straight; just as surely the participant could get other participants bent. A science of psychology (remember... James Hat On!) needn't try to declare one right and the other wrong. Instead, it might simply work to describe the prior conditions that lead a person to experience the world in a particular way. Both the experimenter's experience and the participant's experience are the result of a set of past experiences and a current condition, and those are determinable.

      --------

      Re: Color

      I agree with your presumption: "light at a certain wavelength will typically evoke from 'normal' subjects the same word, eg, 'red'."

      And think we can proceed from that simply by acknowledging that by "typical" you mean that sometimes it happens differently, and that we can explore those situations. To label the typical "true" and the exceptions "illusions" muddies the whole enterprise. That is, we still get caught up in weird (and unnecessary) complexities when a participant says "Green!" and we say "No, it is red, you are the victim of an illusion."

      Note, when you make claims about "typical" responses, you are simply making claims about the current state of people in the current world.... not claims about the object!

      Delete
    2. re. "illusions"

      I think I see the problem. You (and WJ and presumably Valsiner and Rudolph) are focusing on perception. I'm focusing on cognition. So, suppose that typical subject S is viewing a white surface illuminated with red light. S will (epistemically) perceive the surface as being red and may say "the surface looks red". That's what I'd call an "optical (or in general, perceptual) illusion". If S instead asserts "I am seeing that the surface is red", that's an epistemic assertion which in this case is incorrect. I've been calling such errors "illusions", which appears to be a poor choice since that word apparently is generally taken to indicate perceptual error. I'll happily drop it and just call the assertion an epistemic error.

      we still get caught up in weird (and unnecessary) complexities when a participant says "Green!" and we say "No, it is red, you are the victim of an illusion."

      If you and I are in looking at a surface from the same vantage point and I say it's red but you say it's green, assuming that we are in essentially equivalent epistemic states re the surface I agree that it would be appropriate to explore possible explanations for the difference rather than each declaring the other wrong or the victim of an illusion.

      But I'm assuming a laboratory environment in which a subject and an experimenter are in distinctly different epistemic states. Suppose a subject is viewing a surface which is reflecting light that the experimenter knows typical subjects would respond to by saying "that looks red". Then a subject's response "that looks green" is clearly "atypical". But I again agree calling it an "illusion" isn't a good idea.

      I agree with your "note". I use tediously long expressions like "a surface which is reflecting light that the experimenter knows typical subjects would respond to by saying "that looks red" instead of simply "a red surface" to make precisely that distinction.

      Delete
    3. Well... classically psychologists have considered "perceptual illusions" to be the prototypical variety. However Holt, myself, and, I suspect, Lee, would be happy to extend the same principles to anything else you might call illusory. In the tradition of American Philosophy, of which James and Gibson are a part, there is a strong suspicion about sharp divides between things like "perception" and "cognition".

      I'm not sure you will fare any better (in a conversation with us) by referring to "epistemic errors". To call an error an "epistemic error" is to presume considerable insight both into the nature of the error and how the world would work without the error. How can such insight be justified? That is neither a completely honest nor a completely rhetorical question... my point is that it is a difficult position to maintain against scrutiny. For James, both in person and in philosophy, such arrogant presumption was unbearable.

      ----

      This brings us to the experimenter and participant. The experimenter has presumes that people "typically" in "typical" circumstances respond to that wave legnth by saying "red". He then creates "atypical" circumstances, and waits for someone to say "green".

      At this point the traditional experimenter, based on his odd presumption, declares that this participant is responding to the wave length in the wrong way.

      However, there is a completely different way to go, which James would prefer:

      What if the original so-called typical people in so-called typical circumstances were responding to other than than just the wave length? What if they were responding to some higher-order property displays that the experimenter has failed to identify? What if the person saying "green" is responding to the manipulated higher-order property, which the experimenter has unknowingly set in the process of building his display? If that is the case, then there is no "illusion"---perceptual, epistemic, or otherwise--- worth speaking of. There is just an misguided experimenter on the path towards discovering the variables to which people actually respond when they use color words.

      This same logical framework can be used---James, Holt, myself, and probably Lee, would argue---in any case of so-called error: The experimenter is the one in error, they don't know what the participant was responding to, and their assertion that the participant is in error is just a projection of their own failure. As Skinner would say in later, and mostly unrelated, circumstances: The organism is never wrong.

      ---------

      It is no accident that, coming from this tradition, Gibson argued that perception was a sensitivity to higher-order invariants.

      Delete
    4. OK, I now have a better idea of how to fit all this together. This will be redundant with what I've said before, but hopefully it will be more coherent.

      Since we've kind of beaten red apples into cider, I'll instead use Müller-Lyer as an basis of comparison. Assume a subject facing a flat white surface filling the FOV on which the M-L figure has been drawn. I see three related but distinguishable types of reaction to this:

      1. non-epistemic perceptions, in which the subject is merely aware of some irregularity in an otherwise uniform expanse.

      This type might be called "raw experiencing", or per Sellars, "undergoing". The subject may react but not in a way that would be called cognitive. Although required for the next two types, for our purposes this type doesn't seem of independent interest.

      2. epistemic perceptions, in which the subject, responding to a request to describe the scene, says something like "I see what looks to be a uniform white expanse with two parallel horizontal lines of different lengths". (Mention of "arrow heads" is omitted for simplicity.)

      The reply indicates to the experimenter that the optical illusion has worked, but it is not intended to assert anything about what is "really there". I assume "the organism is never wrong" applies to this type when the subject is assumed to be neither intentionally lying nor using language in an atypical way.

      3. an epistemic seeing, in which a subject asserts "There is a uniform white expanse with two parallel horizontal lines of different lengths".

      A subject having this type of reaction asserts that what "looks to be" the case actually "is" the case. And this type of reaction can be "wrong" in the sense that there is consensus within a relevant peer group that the assertion is "false". I'm not implying anything metaphysical here (idealism, realism, etc), just stating an assumption about how that peer group would label that assertion.

      In terms of that trichotomy, it appears to me that your primary interest is in type 2, and I mostly agree with your statements as applied to that type. In particular, I agree that a type 2 perception can't be "wrong" because I don't know what that would mean - "wrong" relative to what standard"? However, I don't see why you object to describing a simple behavior that few if any other subjects exhibit in the same situation as "atypical". But since type 2 perception isn't my primary interest, I'm not inclined to pursue that.

      My interest is in the assertion aspect of type 3, and it is generally assumed that an assertion can be "false", again not necessarily as compared to any esoteric concept like "reality" but in the straightforward sense that almost everyone in a relevant peer group (here, the experimenters, readers of papers about the experiment, et al) agree that it is.

      The complexities of visual perception may be obscuring the very simple epistemic point I'm trying to make. Suppose I say to you "I think the capital of Texas is Houston". Absent reasons to believe that I'm intentionally lying, you presumably accept that statement as true. However, if I assert "The capital of Texas is Houston", you surely consider that statement false. I take the relationship between "looks as if" and "seeing that" to be analogous.

      Delete
  7. Continuing ...

    EC: "I See" seems more concrete to me than "It looks to me as if".

    CW: Again, Sellars uses "seeing that", but you're on the right track. Just interpret "more concrete" as "endorses" the implicit "less concrete" as "doesn't endorse".

    EC: "more precise" cannot be understood to mean "provided by another science, which we assume to be more precise"

    Again, I think my injecting neurology clouded the issue, and I take your point that "precise" isn't the best word since it suggests a qualitative laddering. To repeat (ad nauseum, no doubt), I think the issue is choosing a vocabulary appropriate for the immediate purposes. One vocabulary isn't necessarily better or worse than another, but it may be appropriate in a situation in which the other isn't.

    You might find Sellars' full quote about the space of reasons interesting. (After this last rereading, I think I understand it much better):

    In characterizing an episode or a state as that of "knowing", we are not giving an empirical description of [it but] placing it in the logical space of reasons, of ... being able to justify what one says.

    Moving to a neurological vocabulary is an attempt to hypothesize an explanation rather than to "give an empirical description", but I think the sentiment applies. And as you suggest, such an explanation isn't appropriate in many contexts. Anyway, that's why I'm inclined to use the neurological vocabulary, but don't dispute that it may be inappropriate for many activities in psychology.

    And finally (finally!) to the quote that motivated all this:

    When I experience something as red, the one thing we shouldn’t deny ... is that I am experiencing something as red!"

    The first problem I have is with "experience/experiencing something as red". If the something is an object, I'd argue that the quote is ill formed. We "experience" events, not objects. Objects can cause experiences, but then the phrases should be "experience/experiencing the result of sensory stimulation caused by something". But then I don't know how to interpret the "as red" part. So, I might try:

    When I experience sensory stimulation caused by an object and in response I utter "red", the one thing we shouldn’t deny ... is that I am experiencing sensory stimulation caused by a red object!"

    But this essentially says that my judgment that the color of an object is infallible, which we know isn't the case.

    Or the quote might be edited to say:

    When I am experiencing a sensory event caused by an object and in response I utter "red", the one thing that shouldn’t be denied ... is that I am experiencing that sensory event!"

    This seems clearly a tautology (the utterance is irrelevant and could be deleted). If some do deny it, I'd be interested in their arguments for doing so.

    I could go on, but instead perhaps you could elaborate on the quote with these concerns in mind.

    ReplyDelete
    Replies
    1. But we don't experience sensory stimulation!!! Perhaps, maybe, some very, very, very, specially trained people can experience "raw" sensory stimulation. There is a problematic shifting of levels here. As you said above, I experience objects and events (as also argued by James... who leads to the phenomenologists and ecological psychologists).

      Part of the explanation of my experiencing objects and events surely involves sensory stimulation... and if X explains Y, then X and Y must be distinct!

      --------

      Many varieties of idealism have asserted that someone thinking that they experience X is ipso facto evidence that they are not experiencing X, because, a priori, all we experience is the product of our own mind (which might or might now be a merely a small dark corner of the over-mind).

      Most forms of dualism don't go so far, but only assert (again a priori) that my experiences provide little to no reliable evidence about the state of the world. Thus, my experiencing "a red object over there" cannot, under any circumstances, provide evidence for there being a red object over there.

      To the dualist or the idealism it makes no difference if we are more specific and say "To me, from here, a red object over there", because for them ALL THREE parts of the claim are suspect.

      Thus, if people were to take dualism or idealism seriously, they could not take comfort in what one scientists can convince his or her peers... because the consistent dualist and idealist will admit no evidence as to the reality of any of them!

      It is a deep, deep rabbit hole.

      Delete
    2. we don't experience sensory stimulation!!!
      I experience objects and events

      Yet another problematic word.

      As I have emphasized, Sellars distinguishes non-epistemic and epistemic experiencing, explaining the former as "undergoing", which I take to mean something like "happening to". With that understanding, strictly speaking we do "experience sensory stimulation". But I would agree that we aren't aware of the stimulation itself, may be aware of its consequences, and should distinguish the stimulation and its consequences. That's why in my comment of 1/24 3:21PM (to which the quotes above are a reply) I initially used the phrase "experiencing the result of sensory stimulation caused by something". Unfortunately, I subsequently dropped the "result of", thereby losing the distinction and creating the confusion I sought to avoid. Once again, mea culpa.

      The pervasive problem I see with this area is the absence of a consistently applied all-inclusive vocabulary. "Experience" and "perceive" are multiply ambiguous: objects, events, other? epistemic or not? If events, what kinds of events? External or internal? If internal, neural activity at the stimulation point or later in the processing path? "aware of" or "conscious of" - and what, if anything, is the difference? Etc, etc. How can we communicate if there's no effective vocabulary. One reason my comments are so long is that I try to define my terms so as to make such distinctions and use them consistently, though clearly often failing to do so. And I sometimes "shift level" down to neural activity because I don't know how else to make some distinctions. There may be words/phrases commonly used by people schooled in this area in order to effect such distinctions, but I don't know them.

      I don't see the relevance of idealism or dualism to the discussion. For present purposes, I assume the existence of the external world, and I'm about as non-dualist as they come.

      Delete
    3. There may be words/phrases commonly used by people schooled in this area in order to effect such distinctions, but I don't know them.

      Fear not! You are at the heart of the deep, deep problems in psychology. There are no words/phrases commonly used that effect such distinctions...

      Don't misunderstand me here: There are quite a few psychologists who pretend that such a vocabulary exists, and who resolutely ignore any shortcomings the use of such terms might entail. However, I doubt you would find indoctrination into that culture satisfactory.

      Delete
  8. Mental note.... new long-term book idea:

    Down the Rabbit Hole: Dualism, Hardcore!

    ReplyDelete