From lan@U.Arizona.EDU Mon Sep 6 11:16:25 1999 Return-Path: Received: from Arizona.EDU (Penny.Telcom.arizona.edu [128.196.128.217]) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) with ESMTP id LAA01867 for ; Mon, 6 Sep 1999 11:16:24 -0700 Received: from DIRECTORY-DAEMON by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) id <01JFNR46HPQOB8SOD8@Telcom.Arizona.EDU> for chalmers@paradox.soc-sci.arizona.edu; Mon, 6 Sep 1999 18:17:21 MST Received: from paradox.soc-sci.arizona.edu by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) with ESMTP id <01JFNR44WPGWB8S95N@Telcom.Arizona.EDU> for chalmers@Arizona.EDU; Mon, 06 Sep 1999 18:17:19 -0700 (MST) Received: from orion.U.Arizona.EDU (orion.U.Arizona.EDU [128.196.137.206]) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) with ESMTP id LAA01863 for ; Mon, 06 Sep 1999 11:16:16 -0700 Received: from localhost (lan@localhost) by orion.U.Arizona.EDU (8.8.6 (PHNE_17190)/8.8.6) with ESMTP id SAA11696; Mon, 06 Sep 1999 18:17:15 -0700 (MST) Date: Mon, 06 Sep 1999 18:17:14 -0700 (MST) From: Lonnie A Nelson Subject: Re:a couple of questions... In-reply-to: To: Logan T Trujillo Cc: scicon@paradox.soc-sci.arizona.edu Message-id: MIME-version: 1.0 Content-type: TEXT/PLAIN; charset=US-ASCII Status: RO I have a couple of questions that have stuck in my mind, any and all input is welcome. 1) With regard to Denett's proposition that we "only think there is phenomenal consciousness" via many complex functions; my question, "is there "something it is like" to think that we have phenomenal consciousness? :) 2) With regard to phenomenally conscious states: It was given in the first lecture that a state "Y" is phenomenally conscious if there is something it is like to be in "Y". What followed was a list of "primary" or "foundationally" phenomenal conscious states. I may have mistaken this, but it seemed to imply to me that if a state was not "foundational" it did not belong on the list. THis could have been because the list was of "primary" states, though that is not in my notes (i may have missed it). If this was meant to be a list of the kind of possible phenomenal states and not only of foundational or primary phenomenal states, it seems monumentally incomplete given that conscious experience is often multimodal. It would also seem to be the case that one could mix these primary states and come up with phenomenal experience that as a whole cannot be predicted accurately by the sum of its parts (a sort of systemic interaction of experience) as was proposed to be the case with "drunkenness" or "hypnotic states". Would it then be the case that one could think of the list that was put on the board as a "periodic table of phenomenal states" out of which all other complex states are constructed? If so, what would one propose as the formula for "what it is like" to be in any given complex state (pick one) And / or how might we go about delineating empirically what the "formula" is for any given complex state? Just thoughts, -Lonnie ___________________ It is a common fate of all knowledge to begin as heresy and end as orthodoxy. -Thomas Huxley Lonnie A Nelson Department of Psychology Human Energy Systems Laboratory University of Arizona lan@u.arizona.edu From chalmers@paradox.soc-sci.arizona.edu Mon Sep 6 16:45:51 1999 Return-Path: Received: from Arizona.EDU (Penny.Telcom.arizona.edu [128.196.128.217]) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) with ESMTP id QAA02562 for ; Mon, 6 Sep 1999 16:45:51 -0700 Received: from DIRECTORY-DAEMON by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) id <01JFO2MMIB5SB8S9HW@Telcom.Arizona.EDU> for chalmers@paradox.soc-sci.arizona.edu; Mon, 6 Sep 1999 23:46:48 MST Received: from paradox.soc-sci.arizona.edu by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) with ESMTP id <01JFO2MKXIIOB8T6OJ@Telcom.Arizona.EDU> for chalmers@Arizona.EDU; Mon, 06 Sep 1999 23:46:46 -0700 (MST) Received: (from chalmers@localhost) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) id QAA02557 for scicon@paradox.soc-sci.arizona.edu; Mon, 06 Sep 1999 16:45:44 -0700 Date: Mon, 06 Sep 1999 16:45:44 -0700 From: David Chalmers Subject: Re:a couple of questions... To: scicon@paradox.soc-sci.arizona.edu Message-id: <199909062345.QAA02557@paradox.soc-sci.arizona.edu> MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Status: RO Lonnie writes: >2) With regard to phenomenally conscious states: It was given in the first >lecture that a state "Y" is phenomenally conscious if there is something >it is like to be in "Y". What followed was a list of "primary" or >"foundationally" phenomenal conscious states. I may have mistaken this, >but it seemed to imply to me that if a state was not "foundational" it did >not belong on the list. THis could have been because the list was of >"primary" states, though that is not in my notes (i may have missed it). >If this was meant to be a list of the kind of possible phenomenal states >and not only of foundational or primary phenomenal states, it seems >monumentally incomplete given that conscious experience is often >multimodal. It would also seem to be the case that one could mix these >primary states and come up with phenomenal experience that as a whole >cannot be predicted accurately by the sum of its parts (a sort of systemic >interaction of experience) as was proposed to be the case with >"drunkenness" or "hypnotic states". Would it then be the case that one >could think of the list that was put on the board as a "periodic table of >phenomenal states" out of which all other complex states are constructed? Well, the list wasn't intended to be taken too seriously, but there's a sense in which it was trying to get at the "basic" sorts of conscious states. As you say, there are any number of multimodal states, but those get awfully complex. And one could argue that these are decomposable to compositions of their various "modal" parts. If they can't be so decomposed, then there may be something basic in these states too (as we said e.g. for the category of "background states"), and the only "periodic table" we would get would be an indefinitely extendible one. If it turns out that many or most such cases can be decomposed, though, then just perhaps it could turn out that we could have just a small number of classes of basic conscious states. In a way this might be a little bit like a periodic table. One big disanalogy, though, would be that within each class, there would be a huge number of "elements" (e.g. different sorts of sensory experiences) which may not be easily catalogable, and may not even be finite. >If so, what would one propose as the formula for "what it is like" to >be in any given complex state (pick one) And / or how might we go about >delineating empirically what the "formula" is for any given complex state? Good question. I suppose the simple answer to try would be to say that being in a composition of X and Y is like simultaneously being in X and being in Y. Of course, if decomposability fails, then your multimodal states won't really be compositions after all and this formula won't work. But I suspect that if decomposability fails, then there won't be any good formula, precisely because in that case the "whole" won't be a function of the "parts". The question of how we go about determining whether a given complex state (e.g. hypnosis, drunkenness) is wholy decomposable into more specific elements is awfully tricky. I haven't thought about it a great deal and don't have any immediate ideas about it. It seems that phenomenology has to provide the primary database here. In some cases it seems clear phenomenologically that it would be possible to have states A and B without state C, so state C can't be reducible to A and B. (E.g., take C = the experience of humor and A and B = sensory experiences.) But in other cases it may be less clear. I think patient and reflective attention to phenomenology can go a long way here, but maybe there are "fringe" properties that are hard to resolve phenomenologically. It's not immediately obvious how empirical investigation can settle the matter, but maybe there are ways to bring it to bear. Any suggestions here are very welcome! --Dave. From lnielsen@azstarnet.com Wed Sep 8 13:07:15 1999 Return-Path: Received: from Arizona.EDU (Penny.Telcom.arizona.edu [128.196.128.217]) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) with ESMTP id NAA06698 for ; Wed, 8 Sep 1999 13:07:14 -0700 Received: from DIRECTORY-DAEMON by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) id <01JFQ8VCH9V4B8T9LJ@Telcom.Arizona.EDU> for chalmers@paradox.soc-sci.arizona.edu; Wed, 8 Sep 1999 13:07:14 MST Received: from paradox.soc-sci.arizona.edu by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) with ESMTP id <01JFQ8V98RR4B8TB5Y@Telcom.Arizona.EDU> for chalmers@Arizona.EDU; Wed, 08 Sep 1999 13:07:10 -0700 (MST) Received: from cepheus.azstarnet.com (cepheus.azstarnet.com [169.197.56.195]) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) with ESMTP id NAA06694 for ; Wed, 08 Sep 1999 13:07:00 -0700 Received: from zippo (dialup19ip062.tus.azstarnet.com [169.197.39.62]) by cepheus.azstarnet.com (8.9.3/8.9.3) with SMTP id NAA09885 for ; Wed, 08 Sep 1999 13:06:09 -0700 (MST) Date: Wed, 08 Sep 1999 13:11:33 -0700 From: Lis Nielsen Subject: RE: a couple of questions... In-reply-to: <199909062345.QAA02557@paradox.soc-sci.arizona.edu> To: scicon@paradox.soc-sci.arizona.edu Reply-to: lnielsen@u.arizona.edu Message-id: MIME-version: 1.0 X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) Content-type: MULTIPART/MIXED; BOUNDARY="Boundary_(ID_Zj04cnl5dasI7crTdwNM5Q)" Importance: Normal X-Sent-via: StarNet http://www.azstarnet.com/ X-MSMail-Priority: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 X-Priority: 3 (Normal) Status: RO This is a multi-part message in MIME format. --Boundary_(ID_Zj04cnl5dasI7crTdwNM5Q) Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: 7bit I have a couple of comments in response to Lonnie and Dave's chat about decomposability of complex conscious states, which went something like this: > >If so, what would one propose as the formula for "what it is like" to > >be in any given complex state (pick one) And / or how might we go about > >delineating empirically what the "formula" is for any given > complex state? (Lonnie) > Good question. I suppose the simple answer to try would be to say > that being in a composition of X and Y is like simultaneously being in > X and being in Y. Of course, if decomposability fails, then your > multimodal states won't really be compositions after all and this > formula won't work. But I suspect that if decomposability fails, then > there won't be any good formula, precisely because in that case the > "whole" won't be a function of the "parts". > ... It seems that > phenomenology has to provide the primary database here. In some cases > it seems clear phenomenologically that it would be possible to have > states A and B without state C, so state C can't be reducible to A and > B. (E.g., take C = the experience of humor and A and B = sensory > experiences.) But in other cases it may be less clear. I think > patient and reflective attention to phenomenology can go a long way > here, but maybe there are "fringe" properties that are hard to resolve > phenomenologically. It's not immediately obvious how empirical > investigation can settle the matter, but maybe there are ways to bring > it to bear. Any suggestions here are very welcome! (Dave) My comments: First, in our lab, we ask subjects to report on their experience of emotion (elicited by visual images) using an instrument called the Self-Assessment Manikin (Likert-like scales developed by Lang and colleagues at the University of Florida) that taps the dimensions of valence and arousal in emotional experience. (See attached document for a copy of the scales and instructions for their use). This procedure is based on the assumption that valence and arousal are two basic dimensions on which all emotional experiences vary. The support for this assumption comes from factor analysis work on emotional language, where valence and arousal turn out to be the two factors that best account for subjects' classifications of emotion words. There are a number of advantages to this type of decomposition. It allows us to find reliable correlations between distinct features of phenomenal states and distinct components of emotional physiology. It allows us to compare groups of subjects on these distinct measures and determine whether the two dimensions dissociate. For example, normal subjects will usually report higher arousal to images they have rated as highly negative or highly positive, while reporting lower arousal to neutral pictures. Some patients with frontal brain injuries give valence (positive or negative) ratings similar to normal subjects, but report diminished and undifferentiated arousal across image types. This suggests that the two components are separable. The problem with the scales (and the dimensions in general) is that it is not at all clear (though I tend to vacillate on this issue) that they represent more than linguistic classifications subjects are able to make. Moreover, the scales are presented as linear scales with specific endpoints. Negative and positive are presented on the same dimension with neutral as a midpoint, ruling out the possibility of experiences being a little negative and a little positive. The advantages of simplicity for data collection can lead to serious limitations in our understanding of the fundamental dimensions (or elements) of conscious experience. That said, it does seem possible for us to abstract dimensions from our emotional experience, just as we can extract visual dimensions from our experience of objects (color, shape, etc.). Most subjects have no problem making the ratings, and report that they were able to rate their experiences on both dimensions. For a science of phenomenal experience to develop, one needs to begin by assuming some sorts of dimensions, in order to make progress. The problem comes in making assumptions about what is basic. Reservations that we have had about what counts as basic in emotional experience have led us to consider the use of open-ended emotion narratives as data. This demands that we develop a coding scheme that can render these narratives scientifically interpretable. In the development of this scheme, we begin by making some general assumptions about the elements that constitute emotional experiences (based on a lot of background theory) without committing ourselves to any particular map of the phenomenal space. We remain open to the question of what is basic and what is complex, hoping that the data will reveal this. In implementing this type of measurement in our lab, we will also be using the SAM scales, to determine whether there are any reliable mappings between the classifications our data reveal and the SAM dimensions. My guess is that we may be able to identify certain emotion prototypes, and that these may turn out to vary along certain basic dimensions. Perhaps these will look something like the SAM dimensions, but I expect they will be a bit more complex. I do not think we will be able to identify primitive or basic qualia into which emotional experiences can be reduced. I am always surprised at the desire to implement some form of atomic reductionism at the phenomenal level, and wonder whether it doesn't result from our entrenchment in a predominantly reductive-materialist Western scientific mindset. Lis Lis Nielsen Department of Psychology University of Arizona Tucson, AZ 85721-0068 USA lnielsen@u.arizona.edu From chalmers@paradox.soc-sci.arizona.edu Wed Sep 8 19:06:38 1999 Return-Path: Received: from Arizona.EDU (Penny.Telcom.arizona.edu [128.196.128.217]) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) with ESMTP id TAA07541 for ; Wed, 8 Sep 1999 19:06:38 -0700 Received: from DIRECTORY-DAEMON by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) id <01JFQLF17JO0B8TABQ@Telcom.Arizona.EDU> for chalmers@paradox.soc-sci.arizona.edu; Wed, 8 Sep 1999 19:06:41 MST Received: from paradox.soc-sci.arizona.edu by Telcom.Arizona.EDU (PMDF V5.1-12 #24137) with ESMTP id <01JFQLF07EZKB8T2JA@Telcom.Arizona.EDU> for chalmers@Arizona.EDU; Wed, 08 Sep 1999 19:06:40 -0700 (MST) Received: (from chalmers@localhost) by paradox.soc-sci.arizona.edu (8.9.3/8.9.3) id TAA07518 for scicon; Wed, 08 Sep 1999 19:06:34 -0700 Date: Wed, 08 Sep 1999 19:06:34 -0700 From: David Chalmers Subject: physics, emotion, and baseball To: scicon@paradox.soc-sci.arizona.edu Message-id: <199909090206.TAA07518@paradox.soc-sci.arizona.edu> MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Status: R Some miscellanous comments on things that have come up: (1) Logan raises the interesting possibility that as well as the role that intrinsic properties play in grounding extrinsic properties (on a panprotopsychist view), there may also be the possibility of further "direct interactions" between intrinsic properties which would then be empirically testable. I guess my question here is what the principled difference is between this and standard "extrinsic" interactions. After all, when one particle affects another in standard ways, a panprotopsychist can construe this as the intrinsic properties of one directly having effects on the intrinsic properties of the other. Of course physics doesn't "see" the intrinsic part and characterizes only the extrinsic structure. But the same would go for Logan's new "direct" interactions, it seems to me. Even in parapsychology, one could characterize the causal structure extrinsically. All these would have an extrinsic structure that an objective science could characterize, and all would basic intrinsic properties that are really doing the causing and being affected. On either case, the extrinsic structure is just a way of getting at the basic causal relations among the intrinsic properties. So I'm not sure that I see the difference in kind here. (2) Lis raises lots of interesting issues re her study of dimensions of emotional experience. The distinction between identifying dimensions of experience and reducing experience to basic components is important. It may well be that we can do the former but not the latter. For certain experiences, we can do both: e.g. for color, it's arguable that the experience is reducible to the components of hue, brightness, and saturation. But maybe it will turn out that for emotional experience, we will find at best a limited number of parameters that don't come close to exhausting the phenomenal character of the experience. I guess it's partly an empirical question whether the sort of reductive analysis that works for color will work elsewhere, but there's no obvious reason to believe that it should. Philosophically, one might put the issue here in terms of supervenience. Just say one has identified a number of parameters of experience, and are wondering whether an experience is reducible to those parameters. I guess this comes down to whether the character of the experience supervenes on those parameters. I.e., would it be possible in principle (even logically possible) to have two different experiences with the same values of those parameters but different overall character? With color, one might well argue that it is logically impossible to have two experiences with the same hue, saturation, and brightness, but a different color quality. (Though maybe there are issues about further parameters such as shimmer, etc.) But with emotion, for any set of parameters we have identified at least so far, the parameters don't seem to determine the quality of the experience in the same way. Maybe that just means we have to come up with better parameters, but again, it's not obvious that any full set of such parameters should exist. I note that when I put forward the (very tentative and speculative) suggestion about five or six basic sorts of phenomenal states, I wasn't really suggesting anything quite this reductive. Here, the basic categories were as broad as perceptual experiences, emotional experiences, cognitive experiences, etc, and the claim was in effect that the overall character of a subject's experiences is fully determined by the character of these components. This is consistent with the view that within each category, there may be a huge variety of states that are not reducible to any more basic building blocks (as e.g. Lis suggests re emotion). In effect we have two questions: (i) whether a subject's overall state of experience is determined by the character of their experiences in the five or six categories, and (ii) whether the character of an experiences within those five or six categories is determined by some set of more basic parameters. Both of these seem to be interesting open questions, and the answer isn't obvious (at least to me) one way or another. (3) Re Lonnie's and Joel's discussion of the baseball player: Actually there is recent empirical work that bears on this. The research of Milner and Goodale, which we'll be discussing, suggests that there are two pathways in the visual system, one for making cognitive and conscious identifications, and one for controlling online motor action. In certain cases (e.g. due to brain damage), representations in these two pathways can come apart, and one finds subjects who can perform fine-grained motor actions on objects even when they apparently don't have conscious knowledge of those objects. (E.g., a subject who can "post" a letter through a horizontal or vertical slot even though she doesn't consciously know whether it is horizontal or vertical.) It even turns out that one can find such subtle versions of such dissociations in ordinary subjects under certain conditions, and in these cases one finds that how a subject acts with respect to (say) the position of an object may be out of line with their conscious judgments. It's not out of the question that Joel's observations re the phenomenology of baseball could be partly grounded in this framework. It could turn out that representations in the "cognitive" system are less fine-grained than representations in the "motor" system. Milner and Goodale believe that the contents of visual consciousness are determined by the cognitive system, not the motor system. If that's the case, then one would expect that one would be able to performed fine-grained motor tasks (e.g. catching a baseball) that rely on visual information that is not present in consciousness at all). --Dave.