Meetings

---

Ruth Garrett Millikan (UConn)

Thursday, March 22nd, 2012, 4 to 6pm

Unicepts and the Epistemology of Phenomenology

TBA

Susan Carey (Harvard)

Friday, April 20th, 2012, 4-6 p.m.

TBA

TBA

Ken Nakayama (Harvard)

Friday, December 9th, 4 to 6pm, Room B19 of STH

Subjective Contours

The concept of the receptive field in visual science has been transformative. It fueled great discoveries of the second half of the 20^th C,   providing the dominant understanding of how the visual system works at its early stages.   Its reign has been extended to the field of object recognition where in the form of a linear classifier, it provides a framework to understand visual object recognition (DiCarlo and Cox, 2007).

Untamed, however, are areas of visual perception, now more or less ignored, dubbed variously as the 2.5 D sketch, mid-level vision, surface representations. Here, neurons with their receptive fields seem unable to bridge the gap, to supply us with even a plausible speculative framework to understand amodal completion, subjective contours and other surface phenomena.  Correspondingly, these areas have become backwater, ignored, leapt over.

Subjective contours, however, remain as vivid as ever, even more so.

Everyday, our visual system makes countless visual inferences as to the layout of the world surfaces and objects.  What’s remarkable is that subjective contours  visibly reveal these inferences.

Peter Godfrey-Smith (CUNY)

Friday, February 17th, 2012, 4 to 6pm, Room B19 of 745 Commonwealth Avenue

Representation and the Brain

Representation remains one of the most controversial concepts in cognitive science. I'll argue that new work is converging to give us a much better view of when the concept is useful and when it is not – when it yields a mere illusion of explanation. The talk will look back at how the concept of mental representation came to seem viable, in the 1970s, and argue that a specific wrong turn was taken at that stage. After sketching an alternative view, I will apply it to some simple models of adaptive behavior, revisiting the "cognitive map" concept as a case study.

Nancy Kanwisher (MIT)

Friday, October 28th, 4 to 6pm

Functional Specialization in the Human Brain: Evidence from fMRI

To what extent is the human brain composed of functionally specialized components, each optimized to solve a single mental task?  My lab has used fMRI to identify three regions that are very specifically engaged in the recognition of faces, places and bodies. The very existence of these regions--in more or less the same place in every normal subject--raises a host of new questions that I will address here. Are specialized regions found only for evolutionarily old functions? Are they found only for perceptual and motor systems, or also for abstract higher-level thought? Can we develop new methods that can discover such specializations even for functions that we do not hypothesize in advance? How do specialized regions arise in development? Why do some mental functions get their own private patch of real estate in the brain and others (apparently) do not? In addition to highly specialized regions, does the brain also have general-purpose regions that are engaged across a wide range of tasks? We do not yet have answers to all of these questions, but tantalizing clues are turning up all the time.

Reading: http://www.neuphi.com/images/readings/Kanwisher.pdf

Louise Antony (UMass Amherst)

Friday, September 30th, 4 to 6pm

"In Praise of Loose Talk: Three Ways of Following a Rule"

Abstract: This is (meant to be) a conciliatory paper, aimed at reconciling philosophers who want cognitive scientists to clarify their foundations, with cognitive scientists who say that all will be revealed as empirical work progresses. On the one hand, I’ll argue that empirical inquiry need not wait for the clarification of its foundations, and that empirical scientists should not be pressed to define terms that are serving their needs perfectly well. On the other hand, I’ll offer some philosophical clarification – a taxonomy of “rule-following” cognitive systems: rational-causal, intelligible-causal, and brute-causal.

Interested participants can read Michael Devitt's Ignorance of Language as background.

Reading: http://www.neuphi.com/images/readings/Antony.doc

Peter Tse

Friday May 13th, 2011 from 4 to 6 pm

A criterial neuronal code underlies downward mental causation and free will

TBA

David Rosenthal

Monday, March 21, 2011 from 4 to 6pm

Two Concepts of Mental Quality

TBA

Uriah Kriegel

Friday, May 6, 2011 from 2 to 4 pm

Deflating the Problem of Consciousness without Deflating Consciousness

TBA

Patricia Churchland

Friday, April 1, 2011 from 4 to 6pm

Brain-based values

TBA

Christopher Hill

Friday, March 4, 2011 from 4 to 6pm

Locating Qualia: Do They Reside in the Brain or in the Body and the World?"

TBA

Paula Droege

Friday, February 18, 2011 from 4 to 6pm

Now or never: How consciousness represents time

TBA

Robert Briscoe

Friday, January 28, 2011 from 4 to 6pm

Space, Consciousness, and Control

TBA

Dan Dennett

Friday, December 10, 2010 from 1 to 3pm

Using Humor to Reverse Engineer the Mind

TBA

Imogen Dickie

Friday, December 3, 2010 from 1 to 3pm

Perception as a Source of Knowledge-Which

TBA

Joshua Knobe

Friday, October 29, 2010 from 4 to 6pm

Intuitions about Consciousness: Experimental Studies

TBA

Jesse Prinz

Friday, September 24, 2010 from 1 to 3pm

Consciousness = Attention

Daniel Pollen

Friday, April 16th, 2010. 4-6pm.

"On the Emergence of Primary Visual Perception"

Coming Soon!

Selmer Bringsjord

March 26, 2010

When it Comes to Understanding Human-Level Intelligence and Consciousness, the Blue Brain Project is a Prodigious Waste of Time

You've probably seen at least snippets in the media about BBP, a joint research venture by Switzerland's Ecole Polytechnique Federale de Lausanne (EPFL) and IBM Research.  From a business perspective, BBP is for IBM a win-win venture.  [Despite the fact that the grand goals of the project are -- as I shall explain -- provably unreachable, technological spinoffs will be myriad, and from a marketing standpoint Big Blue's research arm continues to be, well, just simply cool -- especially when you consider that another project (which I believe *will* succeed in revealing aspects of human-level intelligence) is underway:  "Watson" (http://www.research.ibm.com/deepqa) .] But from the perspective of human-level intelligence and consciousness, BBP is irrelevant -- which is to say:  it will shed no light on, let alone support the robust digital simulation or replication of, these phenomena.  Why?  There are many reasons; three are discussed in this presentation:  (1) Human-level intelligence consists in no small part in (sometimes infinitary) abstract thought the nature of which is not reducible to the behavior of small physical objects (like molecules and cells).  (2) Human-level consciousness is not reducible to *any* physical process, period.  (3) A cluster of highly relevant theorems spells bad news for BBP.  The import of this cluster is that studying the computations of a computing machine is insufficient to determine what functions this machine is computing.

Peter Kugel

February 24, 2010

Minds may be computers but…

 "Minds may be computers but, because they evolved, they cannot be programmed the way the computers that many of us have on our desks can be.  Therefore, they must acquire the algorithms they apply in other ways.  Looking at how they do that could help us understand how minds work and why people need intelligence to play chess, why computers can play chess without it, and why animals cannot play chess at all."

Ed Boyden

Tuesday, April 27, 2009. 4-6pm.

Engineering Consciousness: An Interactive Workshop

How would YOU "make" a conscious machine? Ed Boyden will be leading us in an interactive workshop with the goal of designing consciousness from the ground up. For more details on Ed Boyden's work visit http://edboyden.org/

David Gamez

Thursday, October 29, 2009. 4-6pm.

The Science of Human and Machine Consciousness

From http://www.davidgamez.eu/pages/philosophy/index.php

In my recent research on machine consciousness I used the relationship between the phenomenal and the physical to develop a systematic approach to the science of consciousness . I have also carried out innovative work on synthetic phenomenology and I am currently working on a book about human and machine consciousness.

Patrick Cavanagh

Tuesday, December 1, 2009. 4-6pm.

The Position Sense

How do we know where things are? And how do we know which one we are looking at? Our studies suggest that position is constructed by a small set of pointers that attempt to follow and, where possible, predict target positions. When these constructed locations lag or overshoot their targets, we can catch the visual system in the act. These errors in construction underlie several perceptual mislocalizations and reveal possible mechanisms used to create perceived location. We show, in addition, that there is one privileged pointer that corresponds to our sense of “where we are looking”. This direction of gaze can be detached from the center of vision under special circumstances so that we report looking directly at targets that are far from the center of vision. This set of pointers is maintained in one map in the cortex and it serves both spatial attention and eye movement programming as well as the perception of location.

Jeremy Wolfe

Friday, October 23rd, 2009. 4-6pm.

"Bags, breasts, and lots of little cells: New adventures in visual search"

Modern civilization has created many demanding, socially important visual search tasks. These include airport baggage screening and many medical diagnostic procedures (mammography, cervical cancer screening, etc.). We perform these tasks with brains built for other tasks like foraging for food and recognizing scenes. The collision of these tasks and our brains produces a variety of interesting problems. What happens when you search for something that is very rare? Do expert searchers develop the ability to find complex targets in a single glance? My talk will focus on a series of problems with one foot in the lab and one foot in the real world.

Michael Tye

Friday, September 18th, 2009. 4-6PM

"Attention!"

Among the questions to be addressed are the following: What is it to attend to a thing?  In the visual case, what is the connection between attending to a thing and seeing it? Is attending an activity with an intentional content?  Does attention have a sui generis phenomenology?  Do the phenomenological changes that go along with shifts in attention pose a problem for representationalism about phenomenal character?

Alva Noe

Thursday, April 23rd, 2009. 4-6PM

"Out of Our Heads"

Coming Soon!

David Chalmers

Wednesday, April 22nd, 2009. 4-6PM

"The Extended Mind, Revisited"

Coming Soon!

Alex Byrne and Pawan Sinha

Wednesday, March 18th, 2009. 4-6PM

On Molyneaux's Problem

Coming Soon!

Steve Grossberg

Wednesday, February 18th, 2009. 4-6PM

"How to Link Mind to Brain: Towards Solving the Mind/Body Problem"

Coming Soon!

Reading: http://www.neuphi.com/images/readings/Gro1999ConCog.pdf

Christof Koch

Tuesday, February 10th, 2009. 4-6PM

Consciousness and Attention: Related but Distinct Processes

Coming Soon!

Reading: http://www.neuphi.com/images/readings/koch-tsuchiya-07.pdf

Ted Gibson

Monday, February 23rd, 2009. 4-6PM.

Sentence and text comprehension: Evidence from human language processing

Coming Soon!

Reading: http://www.neuphi.com/images/readings/Gibson_2006.pdf

Susanna Siegel (Harvard University)

Thursday, December 11th, 2008. 4-6PM

"What Do We See?"

.

Reading: http://www.neuphi.com/images/readings/Siegel3.pdf

Georg Northoff (University of Magdeburg, Germany)

Thursday, November 20th, 2008. 6-8PM

"The Self and its Brain - A Neurophilosophical Account"

.

Reading: http://www.neuphi.com/images/readings/Northoff06.pdf

Download mp3: http://www.neuphi.com/images/mp3s/Georg Northoff talk.mp3

Hakwan Lau (Columbia University)

Monday, October 6th, 2008. 4-6PM

"Visual Awareness and the Dynamic Threshold Hypothesis"

.

Reading: http://www.neuphi.com/images/readings/Lau2007_HOB.pdf

Peter Cariani (Harvard University)

Wednesday, September 17th, 2008. 4-6PM

"Organizational closure through neuronal signal regeneration as a possible basis for conscious awareness"

Reading: http://www.neuphi.com/images/readings/CarianiALife2008Corrected.pdf

Takeo Watanabe

Tuesday, April 19, 2008. 4-6PM

"Processing of invisible signals informs the processing of consciousness"

Coming Soon!

John Symons

Friday, April 18, 2008. 4-6PM

"Pain and Embodiment"

Coming Soon!

Justin Broackes

Tuesday, March 18, 2008. 4-6PM

"What do the Colorblind See?"

March 18, 2008, 4-6PM
Note-taker: E. I.
Last Updated: April 18, 2008

 

PRESENTATION & DISCUSSION  [J. Broackes presentation]

[Q = Question, A = Answer, C = Comment, R = Response]

This particular talk has almost nothing to do with neuroscience, but is about equally a work in cognitive science, psycho-physics, and with slight stretching philosophy of mind/psychology. 
Basic question I am interested in: What do the colorblind see? 
There is a theoretical answer that was made into a systematic theory in the 1950s by Helmholtz and Maxwell and a little bit of help from Grassmann. 
Three color channels imputing from the retina into a three dimentional color space.  Red-green axis, yellow-blue axis, and white-black axis form basis of representations for hue, saturation and brightness. 

I will focus on study of dichromats (if you only have two cone receptor types in the retina instead of three). 
Helmholtz and Maxwell conducted tests with spinning color disks.  Asked “is this distinguishable with.  Concluded that dichromats lose red-green axis completely and only have blue-yellow axis.  i.e. Green and red experiences collapses onto red-blue and dark-bright axis. 
This is a very elegant theory that has been dominant for 150 years, but probably cannot be true.  It is interesting that in c. 1854, 1905, 1940s, 1970s, 1980s, there has been evidence by distinguished people that this “standard theory” could not be right, but no one has been able to incorporate it in a new theory.  Thus, the standard theory (though it does not cohere with the evidence) has been

Q. This theory will say that there are only two hues—yellow and blue.  In other words, there is no hue experience except for two? 
A. Yes. 

14.00
Dalton (great chemist of late 18th century, gave name for “colorblindness” in many European languages) was colorblind (confused reds, greens) and wrote about the characters of his confusions.  1798: “Crimson by day has a blue tinge, but by candlelight it becomes yellowish red.  Pink seems to me three parts yellow and a reddish yellow…etc”
Here, we have a source of skepticism of the standard theory of colorblind vision—he talks as if he has more sensations than just yellow and blue. 

17:30
Experiments have been done that seem to show that the colorblind have more sensations that what the standard theory will predict. 

Nagel developed the gold standard test (Nagel anomaloscope) for determining the type of colorblindness that a person has. 
20:00-explanation of test
However, Nagel noted that dichromats will never confuse reds and greens if the stimuli is very large (e.g. 70cm radius as opposed to 1cm in radius). 

Suppose we are talking about different quantities of R, G, B light. 
The standard theory predicts that if you are a protonope (missing red receptor), then addition or subtraction of red will not make a difference to the viewer.
30:00
Judd reviewed approx 100 cases unilateral dichromacy, and concluded that the standard theory was correct.  However, when we examine the articles that Judd reviewed, it appears that Judd’s generalization is unwarranted.
e.g. Example of vonHippel & Hongren’s disagreement about  phenomenology of 18 year old patient (35:00-)
Contrary to what the standard theory predicts, colorblind patients DO seem to experience a variety of sensations under a variety of conditions.

40:45
Nagel (1910) “Among thirty dichromats of both types…I have found none that experience dichromatic vision also in large field vision.”  With area .5-.75m and distance .5-.75 meters, dichromats can distinguish reds, greens and

43:51
James & Hurvich Experiment
“Dichotomous 15 test”—Subjects must assemble caps (of 15 different colors) in the order of similarity.  Protonopes may mix up reddish/greenish ones.  Deuteronopes

Stage 1. Ask them do order the caps. 
Stage 2. Then ask them to name the caps.
They get far more right than they should—they mix up the ordering, but
Stage 2’. Order the caps again.  (They are just as inaccurate as stage 1)
Stage 3. Give color names in order of apparent similarity.  (Linguistically, they know perfectly well what the color sequence is)
 
Conclusion by H&J→ The subjects use a rule-of-them where they call dark appearing things “red” and call light appearing things “green”.  They do not actually experience the colors. 
This conclusion by H&J does not seem likely.
49:30

Prof. Broakes’ First Rival Hypothesis: Different colors have a unique way of changing (in appearance) as a function of the type (intensity, hue, saturation) of incident light.  This function may be what the dichromats detect as different “kinds” (colors) of light.  The information is there in the stimulus, but is to be recovered through this method. 

Prof. Broakes’ Second Hypothesis: The macula may be used.  (Pigmented yellow filter that we see through when we look at things directly but not when we see via peripheral vision.)  The effect of this yellow filter is similar in character to seeing objects in direct sunlight. 

Prof. Broakes modeling: What happens to colors when the incident light changes from direct yellowish sunlight to blueish skylight?: Reds are proportionally darker and Greens are lighter.  With a particular mathematical modeling (that makes a projection taking into account how the greens and reds will shift in darkness according to shifts in the incident light), greens will be projected to the green area, reds will be projected to the red area, while blues and yellows would not be affected too much. 

The macula acts in an incredibly similar way.  Transmittance of macula is 100% for RG light, but 40% of B light.  Effective sensitivity of the receptors is different with and without the macula.  The macula adds an extra dimension to vision. 

1:03
It is perfectly possible that the R-G colorblind patients get only Y and B in their experience, but are also in some way able to distinguish R and G from each other by saying that if the color changes in this way, it is R and if it changes in that way it is G. 

Do we have any evidence that the color blind actually “see” these colors? 
I think there is evidence.  It comes from the fact what we might call “aspect shift in color appearance”.  There are cases where red things can look red and instead come to a green to a color blind person.  This is an almost unrecognized phenomenon about color vision (top researchers, major literature/theorists in the field have not recognized it), so let me tell you about it.

I am a color-blind person.  I am not a person who has only two receptor types.  I am probably a person who has two medium wave length receptors and no long wave receptor.  (e.g I have no R receptors but  have two versions of G receptors)
The title of this book is written in red and green alphabet symbols (alternating in color).  When I first showed this book [shows image of book cover.  Title is printed in red and green letters, colors alternating at each letter], my friend delighted in the fact that the book on colorblindness had the title printed in red and green.  I felt a little uncomfortable because until he pointed that out, I was unaware that the title was printed in two different colors.  All the letters just look dark to me.  But later, as I held it under a bright light, eventually, the colors began to emerge.  The red letters went forward a little, and the green went backwards.  I wondered whether there was a pattern to the colors, and I initially thought that perhaps the first three letters were red, but as I looked at it more closely, I realized that the second letter was green, not red, after all.  But I could think of the green letter as red if I wanted to.  If I told myself that it was red, it came to look a little different.  But it looked different. I think this is a phenomenon not unlike the well-known “aspect shift” situations with the rabbit-duck, the necker cube, etc.  What you have is a 2-D input, which is in some sense thought into a 3-D space.  When you take 2-D input and you think it into a 3-D space, then obviously there are ambiguities—you can think it this way into a 3-D space, or you can think it that way. Is your internal space merely two dimensional?  No.  Your internal space is three dimensional, but you are projecting two dimensional information into it, in a way that will occasionally make you wrong about it and sometimes you will have to shift since you thought it was such-and-such layout but actually it’s a different layout.   You were wrong on the third dimension.  You quickly have to jump and change your mind on it.  Now, if I make those mistakes being an anomalous dichromat, it seems very likely that a real dichromat makes those mistakes too.  Other people have reported to me that they have experienced the same thing.  So I’m not the only person that this happens to, it’s quite common, it seem (although it has to be backed up by a lot of concrete evidence) to me that we do have real evidence that it is not the case that the vision of the colorblind just collapses to yellow and blue.  The colorblind have shown that they are capable of making discriminations of reds and greens—we’ve seen that from Hurvich and Jameson example, we’ve seen that from the Nagel reports on the people with large samples—they do not prove insensitive to the difference between red and green. And as for the difference, I mean, do they see reds as red and greens as green, I can’t immediately provide evidence that clearly they have the experiences of reds and greens, but it seems that it is very possible that they do have experiences that are not quite as saturated, but it seems to me very likely that even if their input corresponds to something without determinacy on the red-vs-green, and given a certain hue you may, in a sense, push it out and see it as red, or instead if you enter here/counterweight it [refers to diagram], then you’ll end up seeing it as green.  Down here, it can come to look green-ish or it can come to look red-ish according to the stimuli and the character of the surrounding hues.  There are philosophical implications and future experiments that I would like to see done, but [as we are running out of time] I will leave it at that and will be very happy to take questions. 

1:09:26
Q: Could you briefly talk about the philosophical conceptions? 
A: There’s a variety of philosophical views of what colors are.  If we go back to a very crude view, then you could imagine somebody saying, “colors are only experiences of the mind and sensations in the mind.”  A more advanced view would be, “That’s not what we’re talking about, when we talk of this thing as red, we know that it’s not always not going to look red, at any one moment, it would look a whole variety of different kinds of red, as the light that falls upon it is slightly different.”  So it would be a step forward to say “Redness is not just a sensation, it’s a power of a particular thing, or a disposition, to produce a certain kind of sensation/experiences of red under a normal circumstance.”  But it would be better yet to say “it would be a disposition to produce a whole range of experiences under a range of different conditions, so with this kind of light it looks this way, with that kind of light it looks very different and so on.”  That seems to me, so far, all helpful/taking steps forward.  There’s yet another conception that one might think of, that the dispositional power that we should be interested is not the power of the thing to produce a certain experience in us, but rather a power in the thing to transform the light that falls upon it in a certain way, which then of course will in turn have an effect upon the eye.  Think about what we mean by “shiny surface.”  It is a surface which has a certain kind of specular reflectance pattern.  A matte surface obviously does reflect light, but it doesn’t reflect light in such a way that you could see the shapes of objects when they are reflected upon it, but a shiny surface does enable you to see the shape of the object reflected on it.  What is the difference between a shiny surface and matte surface?  It’s a matter of the pattern of reflectance—what it does with the light that falls upon it.  Now what’s a white surface?  A white surface is one that give roughly 100% back of the light that falls upon it.  But of course the same is true for a mirror.  A mirror gives back 90-100% but in a different way.  It reflects the light in such a way that you can see the forms, whereas a white thing has an entirely diffuse reflectance so you can’t see the forms.  It’s the way in which the thing changes the incident light—that, I think is the characteristic of what a surface kind is.  A blue thing would be something that substantially reflected back the blue part of the spectrum, a red object would be one that very much reflects the red part of the spectrum, and so on.  And if we have that conception of what the difference between a red thing and a green thing is, then it may not matter so much precisely what the sensation of a person is in order to perceive the color of the thing.  Picture a person wearing sunglasses, if you’re wearing sunglasses, in some ways your sensations change, but would you have any difficulty identifying white things, black things, brown things—no, actually you would be pretty good at least in black and white and the basic color categories and so on.  You see the colors because what you’re doing is identifying these kinds of reflective surface in the world perhaps through the deployment of slightly different sensations from the ones you get when you take your glasses off, but in actual fact of course our eyes are subject to varying kinds of illumination all the time, (e.g. as the sun changes, the character of the sky changes, and artificial light in some sense compensates for all this), so this is not so puzzling if what you think of as a thing you are tracking as being these reflectance patterns, these reflectance kinds.  And in that case, you can imagine a person if you were stuck on the idea of what is really different of a red thing versus a green thing was a particular sensation, then you might well say “well, a person who always has sunglasses on would have different sensations so they’ll never know what redness really is.”  I prefer to say that the really characteristic thing about the red/shiny/white thing is not which particular sort of sensations it produces but what pattern of variation it produces in one’s experience.  And that pattern of variation can be just as a cue for the person who always wears sunglasses as the person who doesn’t wear sunglasses, and similarly for color blind person—even if the sensations are slightly different, you might well say that they’ve kept the ability to see the external character.  But that is of course, jumping over a whole lot of fine distinctions that have yet to be made and details to be thrown in there, but that is the general area that I wanted to work in, and that is the general direction I wanted to take the argument. 

Q. Can we summarize your view in a few points: One point is that the dispositional power plus the reflection power plus the illumination will uniquely determine what is your perception of light.  The second point is that even if there is a lot of variation, with glasses or whatever, there is a pattern.  In the scope of this pattern, no matter what, we will call this red.  That’s the categorization of our experience.  Would this summary be correct? 

A. I think that’s just about right.  If you think of a person learning to identify shiny things, they do it obviously by using their eyes.  But it’s not very important when the light is bright yellow, or bright green, or whatever, you can see that constant feature which is the way of changing the incident light, whether its reddish/incidental bluish/yellowish light, there is a pattern of change.  In actual fact, however, I would like/am tempted to go one stage beyond in my own interpretation of this, it also seem to be quite possible that in a certain sense people “correct” their sensations—I mean, there is a sense of normalization in the sensation.  It seems to be also that we may end up wanting to say that the sensations of the color blind are not as far off as the sensation of the normal person as one might first think.  What my argument is is that we would need to get people to talk a bit more about the structure of our experiential space.  Now, one thing you could think of is ordinary/normal observers—how do you know that the difference between red and green is like the difference between blue and yellow.  It’s a difference in hue, and it’s not like the difference between blue and pale blue (i.e. same hue different saturation), but when you move from blue to red, it’s not that kind of change (i.e. change in saturation), it’s a different kind of change.  How do we know this?  Well, the answer is you can get people to talk about what’s similar to what, and in what respect is it similar.  We did this—if you ask people to arrange solid colors in terms of their similarity, they would arrange this in a three dimensional space.  People know how to follow instructions such as “could we have the wall two shades lighter than the ceiling?”  And for ordinary people there are three dimensions of variation: saturation, hue, brightness.  Now, could we work with the same network of patterns of similarity of variation and so on, with the color blind too?  I think we need to ask question of dichromats—when they’re shown a large red thing and asked if they’re seeing it, then they describe it as being genuinely red.  Then if you say “can you give me something else in this room that is like that but a little more yellow-er?”  Can they identify an orange thing as being fairly similar to this red but a little bit closer to the yellow?  You can easily find out that the structure of experiences was the same for them, or it wasn’t the same for them as it is for other people.  Obviously if this dimension has completely collapsed, then the transition here from blue to red (call it T1), would effectively be the same as transition 2 from blue to gray.  And you could show with a series of large samples, take them through T1 and T2 and say “How do you describe this…?” And if they say “this is looking a little paler, this is a bluish red…etc “ then they no doubt know that this doesn’t look the same as that to them, and for that kind of thing we can say “Clearly, they’ve got a sense in which there is a different hue involved here from what’s involved there”—this is one of the experiments I would like to see done.  We need to look at the structure of color experience.    But with all of that, I think we might get us to a position where we have to say, “It wouldn’t be very surprising if the color blind had the same sensations as the normally sighted, after all, they have the same brain as the normal sighted.  What they haven’t had is the very same input from the retina but they’ve had other cues from the retina which might have stimulated the very same experiences through the activity of the brain. 

Q. I think your idea is good but I also detect some determinism in it.  The subject’s sensation is completely determined with some collapse of some structure so there are various different kinds of projections.  If these structures/inputs are the same, then your sensations should be exactly the same—is this your assumption? 

A. Yes. 

C. Actually, there is some neurophysiological evidence about color sensitive cells to different hues.  And what you suggested actually, that some how colorblind signals get transformed into hue coordinates otherwise colorblind people wouldn’t be able to experience a whole range of hues, as you suggested.  But on the other hand, there were experiments with similarities and dissimilarities of our sample with color blind people, and those experiments were using only one dimension—they do not distinguish between colors. 

A: Who is reporting that? 

C. Sokoloff and Parami(?) tried to reconstruct the ??? space from the matrix of similarities and dissimilarities between color samples.  And for color blind people that come up with disused shut up dimension. 

A: Okay, Shepard has come up with this.  (C. Yes, Shepard was one of them.)  I have to say I have some difficulty reconciling—with the other evidence that we have here.  It seems to me that it is very likely that there are wide individual variations here.  It seems to me that there are dichromats who care very much about what they are missing and work very hard to make up for it and they manage to compensate very well.  There are dichromats who don’t seem to care very much, and they might not develop all the usage of all the cues that are around.  So it seems to me that it’s possible that the experiences of these two groups after 5, 10, 50 years, may be very different.  It is not a very trendy topic now to talk about how to train the color blind to improve their vision.  But there have been times where it was quite trendy to claim that there were people who managed to get the number of errors on such-and-such a test down from 95% to 32%.  This suggests (although we should be skeptical) that training does make a real difference. 

C: Yes, colorblindness will not make anyone disabled (people are able to navigate in the world even if they are color blind), so maybe that is why there is not a demand for training. 

A: I also think that it’s interesting that many colorblind people do have interest in what they’re missing out on, and try to do better.  I think that those who learn and make the effort to recover missing dimensions so-to-speak, may end up with something close to the normal experience.  Not of course for tiny samples only seen very briefly, only with relatively large samples.  Actually, I would love to know whether the neurological studies that you’re talking about—did they involve large samples? I mean, when Nagel says “I cannot find a dichromat who does not identify with complete accuracy samples that are 70cm by 70cm—if he says that, it would be very weird if the neurologists can come along and say “they get the exact same sensation with gray.”  How in the world are they doing it then?  This is a puzzle.  The fact is, it is very likely that the neurological studies show that they get the same sensations as they get from gray—using relatively small samples…I don’t know what they’re doing.  The tradition has been not to take seriously the idea that the colorblind might have a wider range of sensations than the Maxwell-Helmholz theory says.  The popular wisdom is, experience collapses to two dimensions.  So people in the adjacent areas say “Wouldn’t it be nice if I link my research with that” and you get some things that are just false in that area.  And I have no idea whether these reports are false with respect to what they are talking about—maybe they are entirely accurate with what they are talking about but different if you use large samples.  But let me give you of things that are genuinely false.  I’ve seen people who have done long studies on Benham’s top: Shown the top, they claim that dichromats have no color experience outside of yellow and blue.  I’m not a dichromat, but very close to one.  I was asked by a researcher to look at a Benham’s top for the very first time—and I saw pink. greenish, yellow and blue…and I saw exactly the same colors in exactly the same places as he did.  I was rather pleased!  I don’t know whether my sensations were exactly the same has his, and he said “Look, I tested a number of pure dichromats and they’re reporting exactly the same thing.”  Now it’s just a puzzle that there are other people who have published exactly the opposite on this.  And I just have to say there is an element of fashion involved here, an element of experimental technique that is no 100% unchallengeable among all of us. 

Q: Is anything you’re saying about compensation for deficits or aspect shifts call to question the traditional distinction between judgment and sensation altogether in general? 

A: In a way no, but in a way I’m open to doing that.  If you take the duck-rabbit, you can see it as a duck or a rabbit.  There’s one description of this phenomena, which is: whichever total set of mind you’re in, you got the same sensation but a different judgment.   In one case you think here is a rabbit image, and in the other you think here is a duck image.  Two different judgments but only one sensation. 

C: Are you sure? 

A: This is somebody else’s description, not mine.  I’m not sure that’s the right way to describe it, because so-to-speak patterns of similarity…if people are shown 20 photographs and are told to pick out the one that is most similar to that, you’ll chose a different set of things depending on whether you’ve seen it as a duck or as a rabbit.  I don’t think judgment is in play there.  There seems to be some not-very-well understood level of pictorial representation that is different in the case of seeing it as a rabbit or seeing it as a duck.  That is, pictorially they are different even if the sensations of where you see white marks and where you see black marks is exactly the same.  This is still a puzzle to me—I’ve spent weeks reading what people have to say about this and I still do not know what I should say about this.  The particular case that I am interested in (color shift), though, will be a challenge to the standard description, or it would be a phenomenon that went outside the range of cases of the standard mode of descriptions as usually applied to.  The standard mode of description is that you got the same sensation but a different judgment.  Whereas in my case it’s clear that you got a different sensation.  Sometimes you get a red sensation, sometimes you get a green sensation from this thing.  The case I’m describing has got to be a case of variation in sensation. 

C: Yes, and the introspective case that you described about yourself, seems to be saying that “as I change my judgement, my sensation of the color changes”. 

A: That top-down view was never denied by the sensation-distinct-from-judgment theorists, I’m not a fan of that view, but I want to show respect to how it was developed.  They are fine with the idea that the input can change as a result of the judgment you make.  I heard long time ago when Christopher Peacock was defending his sensational representational distinction—that was a version of the modern day sense datum theory and he was certainly explicitly ready to say “and sensations can be influenced by judgments.”

[15:00]
Q: I wanted to know how far you will take your view that color perception is related to the effort that you put in.  Specifically, I am thinking of the example like Mary the neuroscientist.  We can perhaps take a stretched version of your view and say that the Mary case is impossible because the type of brain that we have regardless of whether your input is black and while, or RGB, you will experience color. 

A: I’m not unsympathetic to that, but it’s also not something I’d like to be a consequence of my views.  It could be that humans are built so that if they close their eyes and push their eyeballs that they’re going to get color sensations, but anyway, what I wanted to say was that you’re going to get color experiences from colored things even if you have somewhat of a reduced input.  I didn’t want to say you’re going to get color experiences from black and white things even if there’s a reduced input.  If Mary is brought up in a room where there are only black and white things and her body is painted in black and white, then I have no particular view. 

C: If she uses her yellow macula filter—it seems as if she could see color in a black and white room. 

A: I’d like to be encouraging to that, but it in some ways it points to the opposite direction from where I’m generally going.  Where I am generally going is that color is a phenomenon in the world and we have a brain that is training itself to the identification of—the key thing about color is that there is a difference between these/those things.  So in a certain sense the sensations don’t matter too much if the things we are interested in are the kinds, the features out there.  Now, what you’re focusing on are the oddities in the processing mechanism—could those give us a conception of an external kind, even if you don’t come across the external kind. 

C: I was trying to see whether your view would collapse into the view that colors just don’t exist in the world.  You said that you don’t view that, but I can imagine someone ungenerous to your view saying if you’re going to say that we know that these people experience color, when they seem to not have the normal input of color they have the same experience of color, isn’t this just taking it further away from thinking that colors are real things rather than things that are created by our brains. 

A: I think it’s exactly the opposite way around, actually.  I would say that by taking the color as exactly the sun reflectance pattern, kind of spectral reflectance, that we made it most plausible that the colorblind can get the same perception of the thing and I don’t want to say “and their sensations are radically different from normal people.”  I actually want to say “and their sensations are largely the same as normal people.”  So I’m not against sensation (some people seem to have bad conceptions of sensations), I’m rather in favor of sensations. 


Q: But in your view, where is the location that the sensation is created?  In the retina?  Cortex?  At what stage? 

A: Certainly not the retina.  It seems to me it’s better to say that the sensations are in the mind, rather than saying that they’re in V1, V2…  The activity in the V1, V2, V4…is going to result in patterns of experience which are in the mind.  I’m not suggesting an immaterial Cartesian mind, I’m suggesting that there’s a certain way of talking about the whole activity of a human being, thanks to what’s going on in the brain, the sensations don’t have a physical location…
C: You mean [inaudible]
A: When I feel a pain in my toe, there is some sense in which the pain’s in my toe, there’s some sense that the pain is due to something going on in the brain, but I don’t want to say that the pain is really in this part of the pain, not in my toe. 
C: It is projected in the brain?
A: This is a very complicated issue that I do not have a pat answer. 
C: But at least you agree that the sensation is not in the retina, right?  This gives you a ??? view to accommodate very many difficult things because the it would require the process with a lot of filters, so what sensation is is not just the ??? of the hues/brighness/saturation, but it’s more than that. It also depends on the processing mechanism there, and a lot of filterning out, a lot of characterizations.  Are you sympathetic to that?
A: I have no special objection to that. 

Q: My first question is a clarifying question—I thought that you had said that the sensation of red is in the pattern variation that the object…that it’s in the object, it is the way that the object changes according
A: I don’t actually want to say that the sensation is in the object
C: So color…what would you say is in the object? 
A: The first way we talk about colors is, “I would like to have the wall to be painted in a darker shade of blue that it is in right now.” What do we want?  We want a paint with a particular property. 
C: Right, so you want more pigments of this type. 
A: It’s obviously true that if ask “What is this thing we are talking about?” we have to fix upon the institution, the conceptual scheme within which we first fix upon it, so we say things like…if you said the colors are sensations, there would immediately be a problem of this from the fact—suppose someone asks for a red tie, and if I say “well, no, because all my ties are currently covered because it’s dark” you notice at some level it’s just a silly thing to say.  Because what does the person mean when they say they want something red, they want something that looks red under normal circumstances—they don’t mean that they want something that is currently producing a sensation of red.  They obviously know that things remain red whether or not they’re being sensed.  So of course I do not want to say that a color is a sensation.  Color is a characteristics of surfaces, lights etc that is when I happen to be looking at it in the light is good enough, able to produce experience in me. 
C: So there is two senses of color in some sense.  Sometimes I’ll be talking about the thing in the object and sometimes I’ll be talking about how it appears. 
A: But I would say that the same is true of shapes and sizes, too.  You can be asked “pick out from these photographs the things that are really elliptical” or you can be asked “pick out the things that look elliptical”, so yes, there are two things (being vs looking elliptical), but there is also being a horse and looking like a horse.  For every characteristic that we can perceive at all, there is also looking that way.  I would say it comes with the fact that the mind is a representational system.  That doesn’t mean that there are two senses of the terms, no, the only sense of a horse is a four-legged, member of such-and-such a species.  That’s what horses are, and there’s looking a horse.  A donkey might be dressed up to look like a horse, it’s not like there is one property of being a horse and one property of having the sensation of a horse.  You got horseness and the appearance of horseness.  In the first place you got the property in the world, and we got the representation of it. 

Q: I don’t think there is anything black and white in biology—everything comes in gradiations.  So you can have twins but they will be slightly different, and you can have cone receptors and I think that even though they may be produced in the same way during embryonic development, I don’t think that every cone receptor is going to be exactly the same.  So in some sense even if we have one type of cone only, you would have a gradiation (I think) and so you would be able to triangulate somehow, and you would have more than one color.  Do you think that could possibly explain why the colorblind can see different hues? 

A: That seems to be extremely interesting and at some level quite possible.  What we do know is that there are many varieties of long-wave receptors among the normal population if you’re normal trichromat, it doesn’t mean there are only one kind of L-receptor found in every person.  But whether you get variety within one person, I haven’t seen much reported but it seems to be very conceivable, I’d love to know. 

Reading: http://www.neuphi.com/images/readings/DJMH.pdf

Sean Kelly

Wednesday, February 27, 2008. 4-6PM

"Content and Constancy"

February 24, 2008, 4-6PM
Note-taker: E. I.
Last Updated: April 18, 2008

DISCUSSION  [after S. Kelly presentation]

[Q = Question, A = Answer, C = Comment, R = Response]

1:26:27
Q: What is it that we’re talking about?  What are these things that shift?  I want the fuller story. 
A: I find this very hard.  Here’s what I can say so far.  The short version: I think that when we see the size of an object (e.g. size of blackboard) and we see it to remain constant throughout the variations of the distances at which it’s presented, but nevertheless, something in our experience of the size of the blackboard changes, the thing that changes is our sense of how far away the thing is.  That’s a very preliminary account.  The immediate question is, and for the phenomenologist what that means is something like I want to think of this as the horizon, or the visual background, some sort of background context that we are sensitive to, that’s a part of our experience but that’s not a part of our experience that we’re focused on.  Now of course you could try to focus on the distance to the object, and I might look at how far away the thing is.   And then maybe I can even make a judgment about how far that is (e.g. 12ft).  But however the distance to the object is present in my experience of the size of the object it’s not that way.  And again, I can switch to the attitude in which distance is presented that way.  But I don’t think that’s the way it’s present in experience when I’m focused on the size.  It’s present as a background phenomenon and then the question is, “what kind of presence is that?”  And here’s where Merleau-Ponty is my guide. 
C: Yes, not the virtual presence of Noe.
A: Yes, certainly not Alva’s virtual presence.  Here’s where Merleau-Ponty is my true guide.  This I think is his deepest insight and the one that I would like to try to develop.  He says in various places things that suggest that the way I experience the distance to the object, or the angle of presentation of the object, or the lighting context when I’m focused on the color of the thing, is “as a tension around a norm”.  He says “I don’t experience it as a determinate distance or a determinate angle of presentation, I experience it as a tension around a norm.”  And he says in a very different place something that I think explains that.  He gives an example where he says “there you are in a museum, you are walking close to the wall, you can hear and you can see the picture, and it pushes you back.”  I think he really means that.  I think he thinks it’s part of the experience that either you need to move back or you’re pushed to go back, or it would be better if you were back, and it pushes you back so that you can see it better.  And so what I think he thinks, and what I think is the right thing to think is that: the way I experience the distance to the object while I’m focused on its size is in terms of how well that allows me to see the size of the object as it really is in my experience of it.  So I experience the distance in terms of “needing to get a little closer” or “being pushed behind in order to see it a little better”.  And this is really part of my visual experience of the thing and if that’s right that’s bizarre because what is means is that having a visual experience of even something as simple as the size/shape of an object, already involves a normative engagement with the world.  It’s not a mere descriptive representation of the way the world is.  In particular it’s not the case that what I’m representing is some apparent property that the object has.   No, in order to see the size of the thing, I’m already involved in having a sense for how well the context it’s represented allows me to get a visual grip on it.  And I think you can spell that out in more detail, but that’s roughly speaking what I think. 
C: Yes, but part of the detail too, I think would be to cash out this notion of horizons.  Both Husserl and Merleau-Ponty.  So part of what could explain the fact that the identical sensation I have when I see this part of the table here is just what I would have if I were seeing a yellow sensation under different light, then how is it that the same sensations present different properties.  And Husserl of course had this story, “well there’s hyletic data..”  but he also has tools for a better theory which is it’s the presence of these horizons that describe what future experiences would be like were I to situate myself differently, were I to do this.  That might be close to the virtual presence account standing on how you cash it out…
A: Yes, I think you have to be careful. 
C: And I don’t know what to think of the virtual account—I haven’t paid a whole lot of attention to it.  
A: Oh, I can make up your mind in three minutes.  Let me try.  I think this is deeply flawed.  I think that a lot of the things Alva does are great, but the virtual thing I think is a real mistake and here’s how I think you can see that very quickly.  Alva’s view is that in the case of seeing an object to be three dimensional, which is a case that he is very good at describing, it really is true that how I see this cup—it looks to me as it’s three dimensional and one way of saying that (Husserl said this already) it looks to me as though it goes beyond what I can see from here.  It looks to me as though it’s got a back side.  You might think, how could it possibly do that?  But never mind how it does it, I think that’s a good preliminary characterization of the experience of something as three dimensional.  The object looks as though (in Husserl’s terminology) it “transcends my experience of it”; goes beyond what I can see of it from here.    I think that’s a good preliminary characterization of the experience of voluminousness (which is what Alva calls it), but Alva wants to give an explanation of that.  And here’s how his explanation goes in terms of what he calls the virtual presence of the thing.  He says, “what it is for me to experience the cup as voluminous, is for me in having the experience of it from here also to have a practical knowledge of the sensory motor contingencies”.  What does that mean?  It means something like: part of the experience of seeing the object from here is this practical knowledge that if I were over here, it would look like that, if I were here I would look like that…and so on.  Now notice, this involves some weird commitment to the appearance properties too, because what you get is a sequence of appearances of the thing, all of which I have some sort of practical sense knowledge of when I see it from here.  I think that’s like Husserl’s view (middle stage), and I think it’s kind of a phenomenalist view that you might find in someone like C. I. Lewis.  Here’s what I think is wrong with that view.  If that were the right way to characterize my experience of an object as three dimensional, then the world could be a very peculiar way and satisfy that experience.  Namely, the world could be as follows: there could be before me, a two dimensional object that presents this image to me when I’m standing here, and goes out of existence but reappears in this two dimensional aspect when I’m here, and at this two dimensional aspect when I’m here…etc.  In other words, the object that I experience to be voluminous could actually be a series of two dimensional appearances that come into or go out of existence depending upon where I am with respect to the thing.  And that’s not the way my experience represents the world to me.
C: If you had that view that’s right, if you had that experience that would be an illusional experience
A: That’s exactly right.  So there’s an important distinction.  If the world were that way, it would trick me into having an experience as of there being a voluminous entity there.  But what’s important is that we’re trying to characterize the conditions of satisfactions of experience per say, we’re try to characterize how the world would have to be in order for this experience to be satisfied.   And it’s an indication that you got the wrong conditions if it turns out that the experience can be satisfied in a very unusual way like this. 
C: I wonder if someone can say something like this: When you’re talking just about visual phenomena you’re talking about visual schemata.  We don’t have the full constitution of the thing until you integrate the ?? which you already did, and also the tangle and all sorts of other things…  To really flesh out the experience, it’ll also have to be true that I can put my hand on it.  That’s also prescribed in horizon of my visual experience is that I’ll have a certain handle and that changes the condition of satisfaction, and then I can ….way
1:37:56
A: But you could be tricked in a similar way.  I mean, it could be that the part of the object that comes to existence is the part that is in contact with my hand. Something of a relevant weight, or something like that.  It looks like what it fails to do is to make the object look like it has the features independent of my experience of having those features.  And that’s what you need to take into account of.  It’s got a Berkeley sort of thing built into it, and that’s what it means to say it’s virtual.  That’s how Alva defines virtual—remember he says it’s virtual just in the sense that the stuff in my computer desk top is available whenever I go to get it.  It looks like it’s really there, but it’s not.  It only comes to existence when I need it.  He has this account because of the way he thinks certain kinds of change blindness experiments work and other kinds of things, but I just think it’s misapplied in this kind of context because it gives rise to this pretty immediate kind of counter example. 

1:39:00
Q: It’s pretty well known in memory that we make associations and we generalize many many things.  You might make a mistake in memory, but Moshe Bar a former speaker says that’s when memory is working properly, when it makes these mistakes and generalizes.  So how would you know that the priming effect in case of a penny is not just a fault of memory being due to the priming effect.  Because in subliminal priming it’s not available but if you’re doing conscious priming it’s possible that the priming effect is coming from memory.  Thus the reason we would be primed for a circle would because our memory, not our experience, generalizes the shape to the standard view of the penny. 
A: Let me try this.  Some people complain that the weakness in our experiment is that we used a picture of the actual penny, and since people know that pennies are round, they might have a certain association “roundness” with the image, and it’s that association that’s giving rise to the differential priming effect.  There are two things.  One thing you could do is to redo the experiment not with a picture of a penny but with a 3-D disk at an angle.  But somehow it seems you’re pushing on something different. 
C: Yes.  Because memory in general extracts information when it can.  It can’t do that for a complicated shape, it would have to remember the exact shape, maybe.  And it might be that in the initial eight shapes that you gave, if there was one that was very different it may have primed for that one too, I don’t know.  But my point is, whereas for abstract shapes maybe memory doesn’t abstract information away as you get priming that doesn’t have anything to do with memory, maybe in your case because these are things that fairly common.  Even a disk would be something that we would be able to…
A: This sounds like something that is a kind of criticism not in my experiment in particular, but on the shape priming paradigm in general. 
C: Well, I’ve always understood it as the subliminal, so I’m not really aware of the conscious priming.  Subliminal would never have those problems because it’s never going through working memory. 
A: Subliminal things could go through working memory. 
C: If it doesn’t go through the prefrontal, it doesn’t go through the working memory.  So anything that’s definitely unconscious, lets say, wouldn’t be going through hippocampus.  I don’t even know if the hippocampus is really necessary for what I’m talking about. 
C2: I guess the question is, as soon as you have conscious priming, you cannot exclude the possibility that the effect you are getting is due to something more than just the bare experience.  There could be some kind of a process.  A filter, a memory of some kind, that is giving you the effect.  That doesn’t tell you anything about how you experience the thing. 
C:  Exactly.  And because we know in particular that memory does this, that would be grounds for…
A: So let me show you another version of this experiment and see if you think it’s a problem for this version too.  Because this is a pretty well known version in the literature.  Yes, it was in the paper—here is the picture.  I wanted to use this in the context of temporal primacy.  The prime is this complicated image.  You have four different pairings (circle & pac-man, circle & circle, pac-man & circle, pac-man & pac-man).  You’re asked to attend to the lower image in the prime, and the question is, what gets primed?  In this case what gets primed is similarity judgments in circles, in this case what gets primed is similarity judgments in pac-man, and the question is what happens in this case.  And the answer is it depends on how long the prime is presented for.  Short presentation (75msec or so), then the pacman gets primed, with longer presentation (200msec) the circles get primed.  So this is a sort of standard thing in the literature.  These of course are all conscious presentations not subliminal. 
C: That seems to be more of gist perception—the 75 msec could just be gist perception. 
C2: Wouldn’t that support an opposite conclusion?  Because in that case, the apparent property is the one where you have the pacman.  Whereas the real property so-to-speak, ..
If it is true that for the short presentation the priming that occurs is the pac-man, it would seem that the apparent property is what’s caused temporally. 
A: Good.  You are reproducing the part of the talk that I skipped.  Prima facie, that’s what it seems. But I don’t think we’re justified in drawing that conclusion.  During the short presentation, (we haven’t done it yet, but if we do) I think what we’ll find is that the subject reports seeing an ellipse.  And you’ll get differential priming for ellipses.  So doesn’t that just show that during the first 75msec, you experience the apparent property, and only later in time do you experience the real property?  And the answer is, no, I don’t think it shows that because: What we wanted to know was the during the longer presentation, you first experience the apparent property, and then the real property.  The fact that when the thing is presented for only 75msec you experience the apparent property is no indication that during the first 75msec of the longer presentation you experience the apparent property.  Because there’s a crucial difference.  Namely, in the shorter presentation, at 75msec the image is turned off.  And it could be, and I think it is, that the turning off of the image is what gives rise to the apparent property, and in the longer presentation you don’t get that. 
C2: But the alternative explanation could be that in the 200msec presentation, the initial image is overridden. 
A: Yes that is correct.  The date does not suggest which of the two interpretations is correct. 
C: I’m not even sure if you did interpret it to be in favor of Noe’s hypothesis that it would work even then because if it were just phenomenon.  Because what he means by apparent properties, I’m assuming, are not just phenomenon.  If you look at Ode Oliva’s work at MIT, she basically shows two pictures of a hallway and a building overlapping each other, and has found out that the duration of presentation changes which image you see.  So again, this could be just gist phenomena, not a perceptual phenomena.  Like something that has to do with apparent properties in general.  To be able to test that you would have to have a more refined experiment.  Gist phenomenon is known to pick up strong black and white properties in very short amounts of time.  So that all that this would have shown is that the gist perception in the 75msec, not that an apparent property has been picked up.  Because apparent properties, I’m hoping, are not just gist properties. 
A: How do you define gist? 
C: Gist is not well defined, but Ode Oliva and other people who work on it claim that it’s the short wave properties of what we experience.  Long wave being subtle grays.  So the shape of something, say a ball is coming at your head, you would experience that in gist perception, but those are not the apparent subtle properties.  Because the subtle properties are what the exactly how the world looks.  Gist properties are the course grain properties. 
A: So would it make a difference if instead of being black and white, these were red and green? 
C: I’m not sure because I think that gist could also pick up on strong color distinctions.  I’m not even sure how you would even make sure that it was apparent and not gist. 
A: What my experiment shows is that somehow during the short presentation the pacman is what does the priming.  You’re idea is that there are two different ways to describe the pacman.  It could be the short wave stuff in the image or it could be the stuff of a certain shape.  And of course the short wave stuff is of a stuff of a certain shape.  I guess the way to distinguish it would be to ask whether…
C: So if you did this Ode Oliva test where you overlaid the hallway with the building, the apparent property would be this entire picture.  But the gist property would only be the black and white building and not the gray hallway.  And that would be a problem, because then you would find a split in these experiments.  Although this experiment is picking up on gist perception, it’s not picking up on what Alva Noe would call apparent. 
A: I think it’s unlikely that there is a confound here, but I think the way to tell, if I understand what a gist is, to do the experiment with shades of different color patches.  And I bet that if you do that, you’ll priming both for green shades and for red shades. 
C: Or even see if it primes for that exact shape.  (the two shapes together). 
A: Well, what you’re supposed to attend to is the lower shape.  I’ll have to think more about this. 

Reading: http://www.neuphi.com/images/readings/Content.rtf

Jeremy Wolfe

Tuesday, January 29, 2008. 3:30-5:30PM

"What can Visual Attention Tell us About Consciousness and Volition?"

January 29, 2008, 4-6PM
Note-taker: E. I.
Last Updated: April 18, 2008

 

PRESENTATION & DISCUSSION  [J. Wolfe presentation]

[Q = Question, A = Answer, C = Comment, R = Response]

3:15
My primary interest—visual search task, looking for something you’re looking for in fields full of things you are not looking for.  This ties into the question of what you experience when you’re actually looking at the world. 
If I tell you to find a tiger in this picture…looking at this, you have to search for the tiger and you will not know whether there is a tiger until you find it. 

Problem can be divided into three parts:
1) preattentive: before finding the lion, what is going on experientially?  It is not the case that there was a black hole…there was some visual experience beforehand.  We’ll call that preattentive. 
2) attentive: once attention is on lion and lion is recognized
3) postattentive status:  what happens to the representation once attention is moved away from the lion. 

6:42
To some extent I will address the question of “how do I measure the ‘speed of volition’”?  What I will tell you is that the ability to deploy attention around the field is actually very fast, but your ability to do that under volitional control is much slower. I will show how you can measure that.  What difference does it make once you have attended to something—does it change your perceptual properties in any way that is interesting? 

Visual search in lab: Usually you put an array of things on a computer screen and vary the set size and measure how long people take to report is there an “x” present?  What we find is that the amount of time it takes for people to find something is independent of how many items are on the screen.  It turns out that there is only a limited set of properties that allow us do search: color orientation size, holes, line terminations, curvatures, 3-D properties…  We have some sort of access to across the entire field in “one-step”. 
If I tell you to find the green X, you can use info about color and orientation.  Guided search (I have been working on for 20 years). 
If we do not have these “one-step” features, we will have to look at each item, i.e. search for it.  (e.g. finding a T among Ls).  In this case, the greater the number of items, the longer it takes to complete the search (approx 20msec added per item).  Eyes can only go to 3-4 items per second as opposed to

Q. Do you think about this in terms of gestaltist, pop-out?  The Ls were in the same orientation. 
A. Yes, we have to purposely change the orientations of each item so that people cannot select using the basic features (basic shapes etc). 
Q. What is the difference between preattentive perception and gist perception? 
A. I will get to that in more detail later, but the quick answer is: potentially not very much.  Preattentive = what happens before something is selected by attention. 
Gist = What you can do if you never select at all.  Non-selective vision. 

14:30
The object identity turns out to be something that does not “jump out” at you.  E.g. sketch of chicken.  The reason it is difficult to identify a particular chicken is because the mutant distracter chickens share the same local properties with the chicken you are searching for, and because we do not know how these properties hang together until we attend to them (1 version of the binding problem.   Binding problem (in neuroscience): Input from eyes-> Visual cortex with 8 million cortex specialized in color, shape, etc…how are all of these coordinated together to form a coherent image?) Before attention arrives, preattentively chickens with the same local features are identical in their preattentive representation.  Until attention, don’t know how basic features (e.g. color and orientation) hang together. (e,g, Finding red & vertical among figures that each have red/green/vertical/horizontal features much more difficult than when each of the entities have either red or green and either horizontal or vertical feature)

18:15
Asking for the preattentive representation is difficult because as soon as you try to describe it, you would employ your attention. 
Historical way of investigating it: Have subject attend to a particular place and ask what is going on in another location.  However, not ideal as it results in endless debates whether it was truly the case that all of the subject’s attention was focused in the place she was asked to attend. 

22:10
Now, how fast can we bind features with attention? 
Impression/gist—can recognize that the field is full of line-line entities and that they’re green and red, but CANNOT recognize an overall pattern of combined features within the filed, for example that this half of the field contains only red-verticals where as the other half are green-verticals. 
How can we figure out how fast we can move around the field? 
Many experiments, but one example.  [Figure presented with letters arranged in a circle, as well as one located in the center]  Focus on center and move attention clockwise on entities (letters) located around the field starting at the top.  Locate first letter that is mirror-reversed and indicate whether this letter is a “P” or a “S”. 
We will measure the “speed of volition” by changing the location of the mirror reversed letter is. 

It should be intuitively clear to you that it took longer for you to find the mirror reversed image that was located farther away from the starting point, compared to if it was located closer to the starting point.  We measure the rate by (reaction time/number of letters subjects had to consider before reaching the target letter)

We will compare with “anarchy case” where there is one mirror-reversed letter and we do not specify how you can identify this (i.e. do not have to focus attention anywhere, do not have to consider entities in any particular order).  Vary letters on the number of letters on the screen to compare rate of identification (time it took to ID letter/number of letters presented). 

Volitional situation rate: 200msec/entity
Anarchich situation rate: at most 30msec/entity
i.e. Search is much slower when you are telling yourself what to look for (data suggests 3-10 times slower).  Analogy: Analyzing an areal picture—going grid by grid is slower than freely perusing over entire field. 

33:00
What is the status of the chicken after you have attended to it?   Change-blindness type experiment where screen will show several ‘chickens’, will flash to another screen where one of the chickens may/may not fall apart, and will flash back to the original screen.  Subjects must attend to the chickens, recognize and report back: Was there a chicken that fell apart?  Results show that until I direct my attention to it, I do not know how the chicken binds together. 
Analogy with face recognition: I cannot recognize Ricardo until I attend to him (Until I attend to him, he is just this collections of “Ricardo bits”).  Once I deflect my attention away from Ricardo, the chicken experiment suggests that he goes back to being “Ricardo bits” again.
The reaction time of finding a “fallen apart chicken” from a completely new set of chickens is the same as finding it from chickens that have previously been attended to. 
Thus, we can infer that there is nothing about the act of binding (via attention) that survives perceptually in any way that tells you that changed it.  Basically there is no evidence that looking at the stable display helps on spotting the-fallen-apart-chicken any faster. 

Q. If you changed the basic features (e.g. had an array of red dots and changed the color of the dots), would the results stay the same? 
A. Yes.  Somewhat surprisingly, even with basic features the results are the same.

If we measure how long it takes to locate a particular letter in a screen (say “T”) and have the subject repeat that search task while presenting the “T” in the same location every time…500 trials later, the subject’s reaction time would not have gotten any faster. 
If we change the location of the “T” every time…→ Same result!  No change in speed of search after 500 trials. 

You may think that you have visual short-term memory to hold items. 
Why aren’t they using their memory to do this task?  The reason people search the display again rather than use their memory is because searching their memory is much slower. 
Therefore, there is nothing about living with that stimulus for an extended period of time, or having prior experience, that gives you anything special (special advantage for the next search). 

46:47
In the real world, you do actually tend to use your memory (instead of doing an random search every time) in two senses. 
1) If things are sufficiently complicated, you will go to your memory (e.g. where was that cranberry juice last time I saw it?).  This access may give you a comparative advantage. 
2) If there are 30 items on a screen, but you only ever ask about six of them, then repeated search will be a repeated search over a relevant six entities, and you learn not to search among the other. 

The results presented here shows only that there is nothing about the visual stimulus that isn’t changed by the fact that you have attended to it. You do not gain a special property that you didn’t have before.  As visual searches go, the first search and the nth search look the same.  That is the critical piece. 

[Change blindness demo with the Sistine Chapel image.] 
Finding changes between two images is a difficult task (no matter how long you look at the first image).  This demo also shows that changes in basic features can also be difficult to detect. 
[Another change blindness demo with array of red and green dots.  Was the cued dot a different color before?]  Still difficult to detect (even though we are looking at basic features).  With time, subjects tend to get better because they memorize a particular pattern or items (1-4 items) in the array and answer those accurately and guess at everything. 

1:01:50
It is not the case that if you are attending to a particular chicken, the rest of the world disappears. (Except perhaps very extreme cases where people get tunnel-visioned upon extreme stress, e.g. mugging situations where people may report that they remember the weapon but nothing else about the experience). 
My current favorite way of describing this is that there are two pathways:
Selective pathway that does object recognition and a non-selective pathway that fills in the rest of the experience (gist-y experience). 

Aude Oliva has done very interesting work on the gist of scenes.  Her results show that if I take a particular scene and I take raw image statistics within a fraction of a second, that is enough to give me a semantic label (gist, categorical status) of the scene. 

Non-selective pathway fills the world up with visual stuff/experience for you.  In a preattentive sense, we’ll know something about broad image (e.g. bunches of “+” exist on the screen), and once we apply attention to each one of the “+”, we will be able to recognize whether or not it is a particular “+” that you were looking for. 

Same with “coffee bean search”: You can tell immediately as you see the picture that they are a bunch of coffee beans.  Once you get your attention here, you will recognize that there is a guy’s head hidden in the picture.  The non-selective pathway says “round, shiny, coffeebean.”  Attentional deployment to each part of the picture confirms “yes, coffee bean, yes coffee bean, they’re still all coffee beans”…and they guy is not a guy until you run it through this bottleneck. 

1:12:00

Q. There seem to be situations where people seem to have the where representation unconsciously, whereas you seem to characterize that as a conscious representation.  What makes you hold a theory of rich perception where all of our visual field has visual stuff in it?  The two-pathway distinction does not seem to require that.
A. The what (temporal)/where (paraetal) two pathway model is popular in neuroscience.  I have been using the where pathway, or the more theoretically-neutral pathway, as an example of conscious experience because it helps explain why we see all of the peripheral “stuff” in the visual field.  Now there are alternative hypothesis that the visual field is not full of stuff all the time—that we are not

I had a chance to study a Balint’s patience (bilateral paretal damage).  They behave as if they can only see one object at a time.  These patients are not like agnosics who cannot recognize (e.g. faces) even when they see them.  Balint’s patients are able to identify objects they attend to.  It is very difficult, however, to understand what their pre-attentive visual experience is like.  But to go back to your previous question…

Q (restated). There are situations where people could have the where-pathway unconsciously, but you seem to require that it is conscious. 
A. That [the where pathway is conscious] is certainly not a requirement of what and where stuff.  It seems to be that the notion of this where-pathway (I prefer using the term “non-selective” pathway), I’ve been using that as an account of a conscious pathway because I want to use it to describe why it is that the visual field is full of stuff all the time.  Now, there is an alternative position, which is that the visual field is not full of stuff all the time—that you are only aware of the current object of attention and everything else is a grand illusion.  I can understand that this may be theoretically possible, but do not think this to be the actual is the case.  If I am in a dark room and someone suddenly switches on the lights, I seem to experience immediately that the room is full of “stuff”.  It is terribly difficult however, for methodological reasons, to try to pin this down in the lab.  This debate has been going on for at least fifteen years without any solid closure. 

Q. Do you have the same intuitions for other (non-visual) senses—that it is always full of things?  There was an inconclusive study done in California
A. My intuition is that visual experience is different at least from the chemical (smell, taste) senses.  It may be, perhaps because the chemical senses adapt extremely quickly and in the absence of change, one cannot detect any smell/taste.  For auditory experience my intuition is that it may be there continuously. 
C. Auditory perception has been shown to be non-adaptive. 

Q. Could there exist evolutionarily ancestral organisms with the where-pathway, without object recognition? 
A. You would have to go pretty far down the evolutionary tree to find such organisms. 
Q. What about blind sight effects that show that frogs (for example) avoid objects even without visual experience (i.e. when the front of the brain is removed)? 
A. Blind sight effects can be found in humans too.  There are clearly visual processes that occur outside of consciousness, but what is difficult to pin-down is trying to explain what it is that may be the contents of visual consciousness. 

Reading: http://www.neuphi.com/images/readings/IsWasVisCog06.pdf

Moshe Bar

Tuesday, November 27, 2007. 4-6PM

"The Proactive Brain: Using Analogies and Associations to Generate Predictions"

November 27, 2007, 4-6PM
Note-taker: E. I.
Last Updated: January 17, 2008

 

DISCUSSION [Following M. Bar presentation]

Q&A  

[Q = Question, A = Answer, C = Comment, R = Response]

Q. [I agree with most everything you said.  I was wondering however, whether it is better to substitute your notion of “prediction” with a broader term, “anticipation”?]  So you might say that anticipation is almost a prediction but not quite.  Could that be because anticipation is not always conscious? 
A. I’m not sure I agree with you that prediction is always conscious. 
Q. If prediction is unconscious, how can you say it is a prediction? 
A. Well, even a reflex is some sort of prediction.  And this is done at a very limbic, superficial level without one being aware of it.  Thus, I do not have any evidence to argue that all prediction is conscious.
Q. I try to draw the distinction between prediction that is more or less conscious and anticipation that is not necessarily conscious.  Even in the resting state there is a background activity [of the brain] that always makes our brain in a state to anticipate something.  This anticipation is not necessarily conscious, and it is related to a kind of preparedness, where one is alert, prepared for something but not necessarily clear of what it is going to be.  But once the input comes, which is something that fits into the spectrum of what was anticipated, then it will select [[relevance]] with the anticipation.  My major point is trying to draw a distinction with your notion of prediction with an even broader notion of anticipation in terms of preparedness and in terms of the broader background activity [of the brain].  What do you think [about substituting your “prediction” with my “anticipation”]? 
A. For us, many processes, as we do them more often, they become automatic/non-conscious.  e.g. driving from home to work, your body starts to perform without… 
C. You mean habituation.
A. Habituation, but your body starts to perform it without bothering you.  There are psychological measures to determine whether or not something is automatic.  This seems to indicate that not all predictions are necessarily conscious because there are predictions that we do often.  For example, if you are about to bite into an apple after having done it 100000 times in your life so it is automatic.  You don’t think consciously what the apple is going to taste like, your body anticipates it and prepares its state unbeknownst to you. 
Anticipation can be more general, more vague, less specific than prediction.  So, it may be possible to come up with a finer definition where anticipation is less specific and preparedness is more [specific].  For example, “Something big is going to happen tomorrow” would be anticipation, which is less specific than “I’m going to a party/movies tomorrow” (prediction).  And in this regard, anticipation is less useful. 
C. However, “anticipation” may be more appropriate to describe brain state which has no explicit cause, but is just background activity.  We cannot stop making associations. 
A. The brain doesn’t know what to do with a general non-specific anticipation, or arousal state (e.g. amygdala).  Akin to general anxiety—where people are just feeling anxious in general, not from anything specific.  I don’t see anything much a cerebral cortex can do with non-specific anticipation, other than just arousal, making the system more alert, making sure the system will respond, but that seems quite wasteful (since the whole system will be fired up).  I agree with you that there seems to be a difference between anticipation and prediction, and that we would have to investigate more carefully the specifications of tasks performed by the brain.  

 

Q. Different contexts lead to different associations.  How do these context dependent differences in associations show up in the fMRI?
A. The associations will likely look more like a switchboard than the map that was shown in the slide. What are consistently activated are the three regions that I’ve shown.  Regarding context, the more contextual cues the experimenter provides, the more [strongly] the associations will be made. 

Q. I will have particular associations when I am shown a cat, and will have other associations when I am shown a hairdryer.  What is the difference in the fMRI? 
A. So your question is, how is the activation different when I show you a chair vs when I show you a slide projector.  One way to investigate it would be to show you hundreds of slides of a chair and find out what is consistent.  In this case, the visual cortex will recall the specific features of these items.  Regarding the association areas—for certain types of association, it will activate in a certain way.  What we will see is an average of all types of activations of our cortex for all of our associations. 
 

Q. What influence does genetics have on these associations vs regularity [environment]?  Do you think the underlying brain formation has a direct impact on how we experience things?  Would an infant have a very different experience than adults due to the lack of associations it (and the brain) has made in its life? 
A. What we have found out so far clearly suggests that the more experience you have, the more associations you will potentially be able to make.  If you see something for the first time, you may have no idea what it is.  For example, my child did not at first understand that my gesture of pointing indicated that they were supposed to look at the direction I was pointing in, not look at my finger.  With experience, they learned to associate pointing as a directive gesture. 
Q. What about perception itself?  I’ve heard a story that when Columbus came over to America, the Native Americans couldn’t see the ship because it was novel to them.  Are children with less experience, likewise, unable to detect stimulus that they have no previous associations with? 
A. I’m not sure about the accuracy of the ship story, as it would have to be the case that the ship was nothing whatsoever like what the Native Americans have seen (including figures, shadows, colors, contours).  For the situation to work, the novel situation must be completely novel (like showing something in infrared when we’ve only been exposed to visible light).  Some is due to development, some is due to genetic wiring.  For humans, we also know that some of the cortex is pre-wired.  We are born with some basic tools, including some associations. 
C. I think the argument here is that if every experience depends on associations, no truly novel experience is possible. 
A. I think this again comes down to the problem of definition.  In the absolute sense, no novel experience is possible because if you show me anything that I had never seen before (e.g. a 3-D movie), I will still be able to relate it (or elements of it) to something I have experienced previously.  

 
Q. What is completely original about your results? 
A. Just like a novel electronic devise is dependent on the production of multiple different parts, my results exist because of prior works/discoveries of multiple other scientists.  However, this philosophical vibe makes me think of the famous question about the blind man who was suddenly able to see.  
C. Have you read “Mary the Neuroscientist” by Jackson?  What will Mary the neuroscientist see if she is only exposed to black and white, and is one day exposed to red? 
A. This is the only think I can think of that is close to a truly novel experience.  If we acquire a new sense (e.g. see electromagnetic waves [beyond visual]).  The question about vision is really a philosophical question, because how can they suddenly “see”…It almost means that novel experience is experiencing something without the sensors.   
C. Regarding that point, I have heard that the literature is controversial, but I have also read an article about a person who was blind most of his life, who had surgery to be able to see.  In his case, after the surgery, he was able to see splotches of light, but was not able to associate them with anything.  He thought an inanimate object was a dog, he couldn’t identify faces.  I’m not sure it got any better, but my guess would be that his brain may have been too rigid to learn too many new associations. 
A. Most people like this who make this kind of decision get majorly depressed.  It may be the case that you may not have plasticity to learn these things to associate, so in some ways this person is like an infant who can’t learn fast enough to benefit from the new kinds of stimuli.  

 

Q. Making associations on one hand seems like making use of existing pathways of associations and activating these certain pathways, in other ways, it seems like constructing.  Do you see these as two independent functions? 
A. Activation of pathways vs making new associations:  These are two different but obviously related functions.  Making new associations involves the following kinds of scenarios: I realize that appearance of A is always followed by B.  Or I realize that the appearance of A in this context is always followed by B.  It is a formation of a new/stronger associative (Hebbian) connection from a connection that didn’t exist.  The use of this connection refers to the situation when one thing is activated and consequently the most related things are also activated.   One mediates the other, so they are related, but they are not the same thing.  When we make associations, we’re primarily activating existing associations in memory.  When you think about thinking process, it’s much like a conversation, where you can think of  a certain thing for a second, and after some time, we jump to something else, and again to something else…I think that every one of us can think of an example where we are first thinking about our pants and end up thinking about the moon.  It happens all the time.
 

Q. What are the natures of the things that are associated with one another?  Are associations between what pictures, concepts, symbols?
A. That is a wonderful and complex question that I don’t think will be solved any time soon.  We have some working definitions and working assumptions about what we’re activating when we’re making associations.  If associations being made were between representations of this object and this object, it’s visual, so it relates to Carolyn’s question before about why it isn’t in the parietal cortex because if it were directing between this visual image and this visual image, the whole thing would take place in the visual cortex.  And if it were between this sound and this sound, it would be in the auditory cortex.  If it were between touch, it would be in the somatosensory cortex.  This is a very interesting question that we’re thinking about almost every day, but I can’t say that experiments can be done to directly reveal the content of this. 
Q. Are they black boxes for you in this research? 
A. I wouldn’t say black boxes.  More of “put it aside for now until we solve something else”. 
 

Q. Coming back to summarizing your thesis: We have associative structures, which is past experience.  Once the input comes, we have the present.  The interaction between the present and the past will predict the future, right?  There are many association structures that exists, and once there is an input, we immediately jump to the future.  Thus, we’re actually always living in the future because of the interactions of the present and the past.  Does this accurately summarize your thesis?  One cannot avoid living in the future because one has such a rich associative structure.  And once an input comes, we immediately have some anticipation for the future. 
A. The only reservation I have in saying that we’re always living in the future is that (even though I love this idea), sometimes we can fantasize about seeing ourselves about the past.
R. The past is the mirror image of the future. 
A. You have quite accurately summarized my thesis, but I would qualify it that it’s not necessary that these predictions are triggered by external inputs.  They can be generated by internal processes—desires, thoughts, memories—can activate and generate predictions. 
R. These are inside the body, but still external to the mental processes. 
A. If I am thinking, while I drive, about writing an e-mail when I get home, this process is completely internal.  Nothing in my immediate environment affected my prediction of the episode of writing the e-mail. 

 

Q. Do you know what happens to the brain during sleep compared to what happens in your default state? 
A. That is a difficult question.  People are studying it now, and I don’t think there is an answer yet.  We know that people do something that is related to memory during sleep, but there are no final answers regarding the extent it relates to predictions. 

Q. I read that depressed patients and Alzheimer’s patients are less able to make predictions. 
A. It has been suggested that depressed people are less able to make predictions, less able to activate associations. This brings in the disruptive, subversive part of thinking, e.g. rumination/dwelling on negative thoughts without being able to snap out of it.  Also, for certain we know that anxiety brings about depression, and I think anxiety is in direct relation to the inability to create predictions.  Many researchers are working directly on the relation between inability to create predictions, increasing uncertainty, and increasing anxiety and depression. 
Q. You also mentioned a study that claimed “use it [brain function] and lose it”.  I was very intrigued by that. 
A. Yes, I was referring to another study that discovered that the Alzheimer network is the first cortical areas being atrophied by Alzheimer’s.  So in a way, we expect the Alzheimer’s patients who have less associative thinking patterns and less predictions (from the little evidence I am familiar with) show less priming. 

 

Q. How could associations/generations of predictions be related to consciousness? 
A. You’re asking to speculate much further than my comfort zone.  Consciousness and prediction/association is certainly connected in my mind.  When I am fixating completely on a particular object, allocating all of my mental resources on this object, I may be no longer conscious, or less conscious, about other things in the periphery.  And perhaps, if I am focused on a single object, and I lose this single point of attention, the result is that I fall asleep.  Or in another way, the moment we stop allocating our attention, we lose attention.  My intuition is that association/generation of prediction is tightly related to consciousness, but I am not sure exactly how as this is a big question.

 

Q. Nonspecific anticipation: I thought you said that it happens in the amygdala that is separate from specific anticipation.  Is this correct? 
A. Lymbic system is part of the reptilian brian that causes general alertness/arousal of the organism upon activation, and so has nothing to do with prediction.  It does [have to do with prediction] in the sense that if you hear the lion roar again, you have both the higher arousal of wanting to run away, but you may also have explicit conscious predictions of what is happening in the environment.  So these two systems, I would say, are non-overlapping, yet never-the-less highly interacting.  Actually, it’s not even true that they are non-overlapping, because the prefrontal cortex, it seems, in both networks activates its prediction regions (though it’s too strong to say that they are prediction regions). 
Q. We had a speaker in the past who mentioned that a hungry cat is aroused such that it attends to more highly to food stimuli.  Would that be specific or nonspecific anticipation? 
A. This reminds me of how I noticed once that people entering the cafeteria look at our food, while people exiting the cafeteria tend to look at our faces.  When we are hungry, we don’t think about mating.  In these ways, our current states/goals/needs shape our perception, in the same way a hungry shark seeing a person on a surfboard sees a seal (instead of a human).  Our predictions shape our perceptions to such a large extent, that we will not notice any changes unless they violate our predictions to such a large extent.  Our perceptions rely on our predictions to a huge extent, and the perception is counted delayed, counted a few milliseconds afterwards. 

 

Q. How do you view consciousness (generally)?  And how do you think it is related to your work?  Specifically, I am interested in knowing what you think about a theory claiming that our perception produces biases that we are unable to change on the spot (e.g. our consciousness cannot control us to “not be racist”).  Do you think this is true?  Or do you think that consciousness can dictate how/what associations can be made? 
A. We do know that people very quickly make associations, and these associations can last for a very long time.  It is true that if someone appears very threatening to you in the first impression, it is very difficult to eliminate that association (even if that person turns out to be extremely nice).  It’s likely that consciousness can interfere in a very limited capacity, only with things that do not threaten one’s survival.  If it concerns survival, however, the subconscious takes over (which is not necessarily a bad thing). 
Q. Is consciousness only an emergent feature that doesn’t cause anything?  Or can consciousness tell our brain what associations to make?
A.  I think both.  I think that the subconscious is more stubborn about things to do with survival, but top-down causality by consciousness is also possible when decisions do not concern survival. 

Reading: http://www.neuphi.com/images/readings/Bar-TICS_2007.pdf

Dan Schacter

Tuesday, October 16, 2007. 4-6PM

"The cognitive neuroscience of constructive memory: Remembering the past and imagining the future"

DISCUSSION [Following D. Schacter presentation]

Q&A  

[Q = Question, A = Answer, C = Comment]

Q: Common problem in experimental retrieval task is to how to distinguish self-generation vs retrieval of memory. 
A: Hard to say whether hippocampal activations are a reflection of recombining details or in fact a reflection of creating novel episodes.  Our design doesn’t allow us to make that distinction.  We need other kinds of evidence to speak to that. 
 

Q: Do you plan to do anything with more with temporal acuity? 
A: We have a potential study, similar to this study, that tries to look at temporal course of activity, related to the past and future, in addition to neural imaging study.  In terms of describing these activations, instead of asking future>past, we can ask whether “imagining > remembering”.  Our results don’t require that simulating, remembering events.  We’re hoping to do a follow-up study trying to distinguish imaging future and remembering (imagine retrospectively) the past.  So far, it seems that the two imaginative activities hang together, so the functional side seems to relevant to our hypothesis. 
 

Q: What if people imagine impossible or trivial events (e.g. events in outer space)? 
A: Good question.  We’ve thought about it, but still haven’t taken this into the scanner.  Spooner & McDermann did this study behaviorally.  They found that impossible/imagined events are much less detailed than familiar ones.  I suspect that neural correlate studies will show something similar.  Although we would need to find a way to show that the amount activation is not a reflection of the amount of detail that is projected. 
 

Q: Autobiographical nature—would it make a difference if the memory is about what you are doing as opposed to about someone else? 
A: Interesting question that that requires more data.  Imaging study by Spooner & McDermann: Subjects showed a number of areas showing greater activation for imagining what they did in the future as opposed to Bill Clinton doing something in the future.  They didn’t get the full network as we did, so it is possible that some of the regions will be the same (regardless of whether you are imagining your own future or someone else’s). More studies are needed.  
C: Regarding your result that younger people are able to imagine their immediate past/future more vividly and in more detail than younger people: My grandmother may have a difficult time imagining/describing her own future compared to youngsters, but may be able to imagine someone else’s future (e.g. imagining the future of “a girl with an unfortunate background”) in a richer more detailed way compared to people who have lived less and experienced than her. 
Q: Is it possible within science we conduct today to give any kind of evidence to claim that “memory is for the sake of the future”?  This is what Aristotle considers to be a teleological cause rather than an efficient or material cause. 
A: This is part of what we’re struggling.  It’s certainly going to be impossible to do one experiment to show something that broad.  The kind of data that we are getting at least seems to suggest that some aspect of that is plausible. 
C: Instead of thinking of memory as being useful for thinking about the future, what about the brain structures that mediate the kinds of memory you are talking about are also responsible for mediation/thinking about the future in an adaptive sense.  It is adaptive for us to be able to think about the future.  So it’s not necessarily memory that’s important for thinking about the future. 
C: Right, but the key point is, I think, how you would be able to show that it is “useful for the future”.  That part would be very difficult to “prove” with correlation-based data. 
A: I would agree with that—it would be very difficult to support that kind of idea from purely correlational imaging data.  Certainly it would be impossible to get that from one fMRI experiment. 

Q: Remembering is about the past, whereas memory distortion is a function of flexible recombinations/variance that happens in/about the future.  But people recognize memory distortions as happening in the past.  So doesn’t this indicate that memory will shift to the future/imagining the future?  Couldn’t this suggest that memory is for the future? 
A: That sounds similar to our line of reasoning.  We’re trying to come up with an experimental design to test the hypothesis that: “individuals who are more susceptible to certain kinds of memory distortions are better at projecting the future.”  This would be one way to link the two (i.e. seemingly undesirable memory distortions in recollection are actually desirable for predicting the future), but we don’t have evidence for that yet.  The age experiment is as closest we have gotten. 

Q: How do these ideas connect with philosophy of perception?  Do you think that our limits to imagine will also limit our ability to perceive novel events?  Also, can you share with us what you think about visual chunking (mechanism & how it relates to memory)? 
A: I agree that they are related but I haven’t thought about it in much depth.  It is a possible direction one can take in understanding memory. 
Q: What about novelty?  If our past memories limit the ways in which we can envision the future, does it also limit our ways to recognize novelty?  Specifically, would amnesics be better at perceiving something that is very new to them than someone with a very rich memory? 
A: No, I don’t think they would.  The distinction between the novel and familiar are more blurred, if anything, in these experiments.  Amnesics they false alarm less on gist-related lure words but they do tend to false alarm more to totally unrelated items, i.e. they can’t distinguish between the novel and familiar. 

Q: Is there a difference between the near and distant future. 
A: There is.  We’ve done a followup analysis.  We’ve found a difference in hippocampal response between recent and remote, past and future.  Hippocampus modulates stronger to recent than to remote past (reflection of detail).  It looks and behaves differently than recent and remote future, where it (hippocampus) seems to be more engaged with more distant future events (requires more combination of events projecting into the future).
C: There was a study in which people were given the same choices.  Depending on whether the choice was placed in the near or distant future, people made different rational decisions. 
A: Yes, these are the kinds of paradigms that may be necessary in order to distinguish more clearly imaginative activity that is applied to the past and the future when there may be differential consequences of events at different time points in the future (as opposed to our experiments—you can imagine any experience at any time). 

Q: If you are able to get past information about a family member, and tell them a story that actually happened to them, but tell them “imagine you were in this place ten years from now…” and tell the story until they realize at some point that it was something that actually happened to them.  Do you think the hippocampal differences will be the same as your results?  Is this even possible to do? 
A: If they imagined in the future, something that actually happened to them?  I would expect certainly that they would get the same kind of overlap that we saw.  The interesting question to me would be the “future activity greater than past activity” in the three regions—that effect might be mitigated because we are doing less recombining activities and already projecting what already happened in its current form.  We may be able to get at the core of the extent to which this recombination activity is really critical. 
There was an interesting paper on Nature Neuroscience about temporally encoded memory where you get something like a replay of the temporal steps, but you will also get a “pre-play” past memories when the animal is about to go down a remembered path. 

Q: Could it be the case that depressed patients have decreased cognitive capacity in general, or is this something specifically related to memory? 
A: There is a lot of research on trying to understand whether the reduced specificity effects past and future are due to a general working memory decline as opposed to something restricted to memory.  I think that evidence suggests that it is more general, that it is not “just memory”.  I don’t know of any imaging studies off-hand, that have looked for activation of autobiographical memory in depressed patients.  There will be an interesting study coming out in Nature Neuroscience: Positive and negative events, and were how the areas are used differently in people who are more and less optimistic about the future.  Their experiment asked about positive and negative events, and project ahead.  They found activations in addition to the same regions that we did, in the amygdala (emotional), and part of the rostral interior singulate.  What they regions showed an interesting dampening down when you projected negative events (vs positive events, or past -/+ events), moreover, they showed that the dampening down correlated with trait-optimism of the individual measured by a separate scale.  What’s interesting is that the rostral interior singulate (i.e. the region that correlates with optimistic behavior) shows an abnormality in depression. 

Q: How do you think about the underlying neural representations and operations of the recombination operations (aside from populations being activated/suppressed)? 
A: I don’t have any novel ideas. 
Q: Do you think computational models will aid in understanding?
A: I think they will provide a stylistic enhancement. 
Q: What about computational representations/approaches towards memory? 
A: Certainly many classical computational models do a good job explaining certain kinds of memory distortion, though none have been proposed that allows prediction of the future events.  I think that it would be an interesting exercise to pursue. 
Q: It is often assumed that memory is just a feature that goes along with other kinds of processes (e.g. sensory, motor processes), which makes it difficult to conceive of what exactly the hippocampus does—as it apparently does “just memory”.  This seems to be an answer to the question, in that the processing that the hippocampus does is relational, recombining processing and what we call episodic memory is the outcome of this processing. 
A: Yes.  One outcome of this processing. 
C: Need not be in the sense that we’re turning everything on its head.  Experiments of future processing says that it’s not memory, it’s a control for memory.  The fact that you get more activation in the future than the past suggests that it’s not memory, it’s the application/processing of the memory than just the relational recombination itself. 
A: Yes, it does kind of reverse the relationship between memory and imagination/future planning.  That’s ahead of the train and memory is a helper. 

Q: Could photographic memory be considered as the extreme case of highly inefficient (unable to recombine) memory (and the other extreme would be highly imaginative memory)?  And perhaps it is the saliency of the experience that influences where the memory is placed on the continuum (e.g. we seem to remember more details about emotionally charged experiences).
A: I prefer to avoid photographic memory.  No convincing documentation.  However, we can think of a continuum you proposed.  It would be interesting to investigate the contributing factors to where the memories fall on this continuum.  It may be age related (e.g. older people at the cost of less specific memory and less specific future imagining, may have more “wisdom”).  It would be interesting to map out individual differences in

Q: If memory recombination can happen almost in the way that you “want it to happen”, it seems that memory is highly subjective, and unreliable as a source for determining “objective” phenomena. 
A: I wouldn’t go quite as far.  There is no question from eye-witness literature, that one must be very careful (high confidence =/= high memory).  We must strike appropriate balance between trusting memory too much (because it does have vulnerabilities in various contexts) and trashing memory as being useless because under many conditions it is reasonably accurate.

Reading: http://www.neuphi.com/images/readings/nrn_2007.pdf

Carolyn Suchy-Dicey

Tuesday, September 18, 2007. 4-6PM

"Consciousness in the Cortices--A Reply to Block"

Reading: http://www.neuphi.com/images/readings/Neuphi_Presentation_Paper.doc

Howard Eichenbaum

Tuesday, May 15, 2007. 4-6PM

"The Neural Basis of Recollection"

Neuphi
May 15, 2007, 4-6PM
Note-taker: E. I.
Last Updated: June 15, 2007

What we know about phenomenology of conscious recollection
Many parts of the brain contribute to information processing of memory (e.g. Frontal, paraetal, temporal), but we’ll focus on the hippocampus (medial temporal lobe) because it is known that selective loss of memory occurs with hippocampal damage/removal (e.g. H.M.). 

What do we mean by recollection?  We will refer to humans and animals. 
(So, valid animal model is required, which is a challenge. There is much debate on whether animals even have a conscious recollection.)

Which features of conscious recollection are supported by hippocampus? 

Familiarity vs Recollection.  Aristotle may have been the first to distinguish this.

• Familiarity: Simpler, extrapolation of perception.  persists and can be resurrected again. Like wax inprint, or photograph.  

• Recollection: Episodic memory.  Has subjective features as well as objective features.  Has feature of mental time travel.  Sense of self, or autonoetic consciousness (i.e. sense of I’ve-been-there-before), is crucial part of memory. [i.e. Autonouetic—I’ve experienced this before and I “really” remember.  Familiarity—I’ve seen this before, but I can’t really say that I remember.]

Example describing Familiarity vs Recollection, & Objective features of recollection
After poster session of scientific meeting.  Young women who looked vaguely familiar approached me.  I knew that I know this person, but didn’t really know who she was.  But, the moment she spoke, I got a rush of information, immediately knew who she was, and knew the context of how I knew her (remembered her standing in front of the poster).  Then, I was able to do mental replay of her telling me about her experiment.  It seemed that different episodes are connected to one another to form network of memories. 

Thus, three objective features of conscious recollection seem to be:
1. Information is encoded as an item in its context. There is a threshold level of information that is recollected, which is fundamentally different from familiarity. 
2. Multiple types of information involved in memory. 
3. Mental replay and networks.  Spacio-temporal features, as well as associativity of our experiences are preserved. 

Experiment about feature #1 (Items are contextual in recollection):
Question: Do rats with hippocampal damage differ in their ability to recall events? 
Procedure: Recognition memory test (typical in psych) with rats & odors.  Let them dig.  Cups and sand.  Dig if it’s new, but not if not.  Variables were height of cup, # of fruit loops, type of odor (cups with new odors have food, old odors have no food). 
Observation: Mice w/ hippocampal damange behaves differently from controls.  Rats w/ hippocampal damage have graphs that do not have the curve representative of “recollection” tasks. 
Conclusion: Hippocampus supports recollection, but not familiarity. 

Experiments about feature #2 (Multiple types of information are recollected):
Question: What kind(s) of memories do hippocampal cells encode?  Are hippocampal cells responsible for encoding location, type of oder/information, or some/all of the above? 

A) Rats—Record individual nerve cell activity in the hippocampus during odor recognition tasks.
Do hippocampal nerve cells encode type of odor and location of odor? 
Answer: Yes.  Some cells encode the type of odor, some encode location, some encode a combination of information.

B) Humans
UCLA study.  Virtual reality game taxi cab.  Find different shops, find 1 of three shops. 
Concl: Same as rats.  Some cells respond to where you are, some respond to what shop you’re going to, some fire only with a certain combination.

C) fMRI study
Horizontal sections of brain shows level of activation in hippocampus.  Human word recollection task.  Only when “truly remember”—hippocampus activation. 

Conclusion: Hippocampal cells encode a variety of information, including: location, type of stimulus/object, combinations

Experiment about feature #3 (Recollection involves sequential encoding of information as well as networking of different information):
A) Regarding event sequence:
Aristotle’s claim that: Episodes are represented as sequences of events. 
Look at spacial memory folding over time.  When I stroll from MIT to restaurant in Fenway Park, hippocampus will encode each of these as sequence of events.  Does it only encode what’s going on at that moment (ongoing event), or does it care about what happened before and after?   

Instantiate in rats T-maze alteration experiment.  (Rat starts at base of the “T” and gets reward for alternating turns.) 
Hypothesis: If hippo encodes events as sequence of events, then, left turns and right turns should be encoded differently. While it’s running at the base of the maze. 
Observation: there are cells that fire only when rats are about to do left turns. 

Conclusion: Sequences of events are encoded in the hippocampus. 

Same thing has been shown in human brain imaging.  Subjects are presented with images of route view vs bird’s-eye view.  Hippocampal activity has been observed for route type view but not for bird’s-eye view. 


B) Regarding Network of memories:
Can be observed via associative transitive inference task.  (If people are shown pairs of stimuli. A=B,  B=C  then, does the hippocampus say A = B = C?)

Procedure: Train rats to associate A&B, X&Y…  Observe, where does the rat dig?  (% time spent digging in correct location) 

Observation: Rats without hippocampus cannot do association/inference task. 
Conclusion: Hippocampus is involved in building networks between memory. 

Same result observed with humans.  (e.g. With association task about college dorm living situations.)


So, how does recollection emerge from circuitry? 
Cortical association areas→ medial temporal region outside of hippo→ Hippocampus

Process of remembering events as an item in a context
Studies so far suggest: Visual information in V1 goes to two different cortical areas.  Information about “what” particular objects are go to the ventral striatum, then to the “blue regions” (in the slide) in the parahippocampal area.  Information about “where” the object is goes to the Dorsal stream, then to the “green regions” of the parahippocampal area.  The two streams of information converge in the hippocampus.

Consequent hypothesis: The hippocampus is where different information (i.e. the object identity information from the “what” stream and the contextual information from the “where” stream) converges to become an event/episode of something occuring. 

Evidence[1.19]: 
Rats and monkey hippocampal areas are known to respond more to familiar stimuli.

fMRI: For object identity, perirhinal (blue) cortex responds more to novel than familiar stimuli.  Reacts less to stimulus a second time.  For spacial context, the green area in parahippocampus neurons responds to scenes, but not to objects.  In perirhinal cortex, no extra stimulation if context is remembered in addition to the item.  In hippocampus, item and context gets activation, but not for only one.  i.e. The hippocampus responded only when people remembered words as well as where they saw this object. 

Perihippocampal region get strong activation for objects in context, and for objects strongly associated with context (e.g. fridge), than objects with less contextual association. 

Thus, Possible Mechanism of Recollection of an Event [1:23:11]:
Retrival cue (e.g. see a woman’s face) goes into the blue part of the parahippocampal region.  If I recover only the extra features of the face, I might just be “familiar” with her.  (Goes to frontal lobe. )  But, if it goes to Hippocampus (can recover context information)  , then the information might get sent out to Parahippocampus where context is resurrected.  Then, it could go to the lower cortex where the details of the episode can be retrieved which leads to the phenomena of “I recall!”

So hippocampal region acts more like a pointer to information rather than storage of information. 

Finally, on subjective features of recollection:  Data so far shows that medial cortical networks, and not hippocampus, seems to be the self-referential parts.  (e.g. “I remember thinking about this…”)  So, perhaps objective properties of recollection are shared with animals, but not the self, subject aspects (supported by cerebral cortex).  People have the same “when/where” system as animals, but have an additional cortical input that encodes subjective/self “why” information.  Subjective features are part of recovered system.

DISCUSSION

Q&A  

Q. Episodic vs Semantic memory?
A. Episodic—Remembering learning about the capital of France in grade school.  Semantic—Capital of France = Paris  Hippocampus linked more with episodic memory.  Cortex often correlated to semantic memory.  Hippocampus learns more rapidly compared to the cortex.  Computer network model simulation can show same feature.  (Hippocampus is a fast learner, but has catastrophic interference when working by itself.   Cortex doesn’t learn as quickly, but can work in conjunction with hippocampus to receive/provide information to help hippocampus recognize things it has observed in the past.  Prefrontal cortex is working memory.)

Q. Does plasticity create a generalized sense of memory, separate from the hippocampal?  A. Plasticity is feature of all parts of the brain and is responsible for familiarity—which can be unconscious--which I will talk less of today.  For example, motor skill learning is largely unconscious, but still memory.  The motor memory seems dependent on that general plasticity. 

Q. Are you saying that even if rats had phenomenal consciousness, it would not have self-consciousness b/c it doesn’t have the neural structure?? 
A. Yes.  An interesting recent speculation about cortical evolution is that what we usually associate as evolutionarily more simple and “lower” animals have only the “higher” cortical areas, undifferentiated association areas (e.g. snake recognition), whereas higher animals have additional “lower” more specialized functions in addition to the higher ones. 
C(comment). Still that doesn’t seem to answer whether there is something to be like Mr. Rat (phenomenal consciousness)? 

Q. Other than the hippocampus, what other brain structures mix information (e.g. sensations, emotions)?
A. It is know that stages such as the visual cortex collapse information.   Even higher areas (e.g. Wernickes) collapse together higher order information.  Even higher areas such as the Hippocampus, put together figure with ground even higher order information which seems to be linked to long-term storage of information.  The prefrontal cortex is the other major area (runs show.  working, thinking, processing).  Hippocampus is probably evolutionarily older.  It’s the final association pathway.  All mammals have it.  For all mammals, the major cortical-hippocampal connection is the same. 
Q. Birds? 
A. Birds seem to have an analogue of hippocampus.  But unlike mammals, they don’t seem to have a cortex. 

Q. Is it true that the evaluation itself (e.g. of “this is X”) affects the content of the memory?
A. Yes. 
Q. So, then memory cannot be completely objective.  Because the subjective evaluation would affect the memory. 
A. True.  There is no such thing as a pure retrieval event because every retrieval affects/contributes to the memory.   Experiments can tend to simplify this complicated process. 

Reading: http://www.neuphi.com/images/readings/CompCog.pdf_.pdf

Walter Freeman

Tuesday, May 1, 2007. 4-6PM

"Explaining animal minds as intentional, self-organizing, creative systems in prelude to understanding consciousness"

Neuphi
May 1, 2007, 4-6PM
Note-taker: J. T.
Last Updated: May 5, 2007

Talk part 1, part 2, and part 3

DISCUSSION

Q&A (Multiple questioners numbered where applicable)

Q: So I can articulate an experimental philosophy: there are many
events in the brain one can measure; and one can analyze those signals
in a variety of ways, but at the end of the day, one must be able to
account with these analyses for something the subject can do.  Can you
use this theory to account for what the animal is doing on a moment by
moment basis?
A: I think that the thrust of our recording now is on multivariate
emissions, and thus have to be analyzed offline.  The kind of response
we were testing were simple, and all our experiments were performed
with full knowledge of the animal's response.  We can discriminate
between three different odors given the EEG images, but more than that
is too complex.  The big difficulty right now is working with large
numbers of neurons.

Q: You pointed out [Claude] Shannon's recommendation not to use info
theory for brain modeling.  I think you fall into the same trap when
you try to use a thermodynamic approach.  You have an alternative, say
what Wolf Singer (sp?) is saying.
A: I think that computational models are a crutch that can be used to
interpret the data, but that don't actually explain what the brain is
doing.  The key problem I see with the application of info theory is
that the response of the animal depends on its expectation, which in
turn depends on its knowledge base.  You cannot measure knowledge with
information.  You might be able to measure the information in terms of
the molecules you give to the rabbit to smell, but you cannot measure
the amount of information that finally arrives in the rabbit's neural
system.  I will say that the averaging that takes place in
thermodynamics is matched by the spatial averaging that dendrites do,
so in that way the theory is a better fit than you suggest.  However,
it is true that a statistical thermodynamic approach that doesn't
account for the individual properties of neurons is bound to fail.

Q1: It seems to me that your discussion of the cinematographic property
of perception should be able to explain our conceptual capacity to fuse
images.
Q2: It seems this whole perceptual framing thing has been lost, which
is too bad.
Q1: Well maybe now… Matt Wilson's group has some interesting ideas
about the usefulness of theta [frames].  There it's a sort of way to
perform temporal compression.

Q1: I noticed you mentioned that there are both excitatory and
inhibitory neurons in the same system, and so it looks like there would
need to be something else controlling them.
Q2: If we accept Freeman's position, the activity of the system doesn't
get its meaning from an external source, it is internally propagated in
the system.  But this discussion of individual neurons isn't meaningful
philosophically.  Their only meaning is collectively, with respect to
background activity in the entirety of the system.
A: That's right, there is an amplification of the meaning of an
individual neuron's activity to the rest of the system.

Q: As an engineer, something that is absent to me [from your
explanation] is the Turing test.  The model given in the theory seems
entirely distinct from what I would normally take a person to be.
A: We're starting work on a field called intentional robotics, in fact.
  Hoping to develop some robotic applications out of this.

Q: You've criticized models in cognitive science.  I define models as a
parsimonious set of hypothesis that allows prediction of the unknown.
I think that applies to models of the brain as well as physics…you toss
off these remarks about models as not explaining what brains are doing.
A: I think what I was trying to do was make a distinction between a
model and a theory.  What I was discussing in my history of models of
the brain was a breakdown in theory, and I believe that info theory has
had its day — we should return to an energy, or power-based theory.

 

 

Reading: http://www.neuphi.com/images/readings/AD-.Category_.errors_.pdf

Ned Block

Tuesday, March 20, 2007. 4-6PM

"Consciousness, Accessibility and the Mesh between Psychology and Neuroscience"

Neuphi
March 20, 2007, 4-6PM
Note-taker: J. T. , E. I

Last Updated: March 26, 2007

Preliminary Statement of the Puzzle 

Cognitive accessibility vs. reportability
    Are representations inside a Fodorian module phenomenal or not?
    One suggestion: If a state is unreportable, it is not phenomenally conscious.
This doesn’t work because the ultimate data are themselves reports.  Furthermore, we know that it is possible for unreportable states to be phenomenally conscious (Locked-in Syndrome).
The real question, then, is whether cognitively inaccessible states can be phenomenally conscious.  In other words: is cognitive accessibility a part of phenomenal consciousness?

A real case
    Binocular Rivalry – red/green alternating image.  This seems explicable.
Fusiform face tests.  Subject experiencing a face without being able to cognitively access that experience?
Can’t use induction, because we are interested in whether cognitive access is actually a part of what phenomenal consciousness is.
Medical importance
Terry Schiavo case, e.g.  Whether or not the patient is having experiences would seem to be important for the decision of whether or not to pull the plug.
Importance for science of consciousness
Cognitive Accessibility
    Using a Global Workspace Model of Cognitive Global Accessibility.
    -Perceptual mechanisms are suppliers of representations.
    -Mechanisms of reasoning, reporting, evaluating, deciding, etc, are “consumers”.
“Accessibility” is somewhat ambiguous here: could mean “in the global workspace” or “potentially in the global workspace”; the latter is too broad to be useful.
Correlationist methodology
Generally restricts discussions of consciousness to conscious access, since   consciousness seems to be too difficult to cover in a broader sense.
Better methodology
    Inference to the best explanation.
Empirical Argument
    We have a sense of experiencing more than we can grasp cognitively.
    Ex: Change blindness
    But is this inattentional blindness or inattentional inaccessibility?
    Thesis: it is blindness to change, but not blindness to the features that change.
Argument: Inattentional inaccessibility implies that there are different bases for phenomenal consciousness and attentive consciousness.

Refrigerator Light Illusion: subjects mistake the easy potential accessibility of all Sperling's letters with actual consciousness of them.
-    Replies: but how can the way that it seems that it seems not actually be the way that it seems?  
-    Why wouldn’t this illusion vanish when we find out about it?
-    No positive reason given for believing that the illusion exists—only the philosophical doctrine that access and consciousness must coincide.
What about the “illusion of seeing” noted in the change blindness discussion?
-    Reply: The conclusions of these change blindness experiments are debatable.

Mesh Argument:
-    Psychological argument: phenomenology overflows accessibility.
-    If strong recurrent activations in the back of the head that are not globally broadcast are phenomenal (cf. discussions of V5 region), we have a neural mechanism for overflow.
-    This is a reason for accepting that phenomenology doesn’t require global broadcasting.
In other words, abduction/inference to the best explanation (as in the Mesh Argument) dissolves the puzzle abstractly.

DISCUSSION

Q: Is it possible for phenomenology to be distinct from attention?  Could cognitive access be distinct from attention?
A: Phenomenology can be distinct from attention.  One recent article has a good argument that describes difference between phenomenology and attention.  One of the dramatic cases involves attentional manipulation of blindsight patients.  Subjects got faster with practice at the task of utilizing predicting/anti-predicting cues in their blind field.  This suggests they can pay attention in the blind field, even without the phenomenology.  So it looks like attention and phenomenology can be pretty separate.  The relation between attention and cognitive accessibility is less clear.  There is a recent study that shows that things (like faces) shown in the periphery can still be reported even if all the attention was soaked up in the center of the visual field.  This seems to show that you can have access without attention, but it could also be that all the attention was not soaked up in the center of the field as the experimenters assumed.  


Q: Can’t this V1-V5 recurrent processing loop simply be construed as a causal condition, along the lines of the necessity of the retina, or blood flow in the brain?  
A: Some view that what the recurrent processing does is to amplify and refine the process, and that the activation itself (and not the recurrent process) is the real neural basis of phenomenology.  I think this is a possibility, and that the experiments that have been done so far do not distinguish between the two possibilities.  But I think that this is an issue that can be investigated; one might e.g. cut the connection between the fusiform face area and the lower area and stimulating the fusiform face area, to see whether you still get phenomenology (even with the absence of recurrent feedback).  

Q: All of this discussion seemed to leave out any mention of memory, but memory might be an issue in any discussion of attention and experiments measuring attentiveness.  When we talk about a subject experiencing phenomenology, shouldn’t there be a distinction regarding when a subject is recalling/remembering something as opposed to experiencing something right at that moment?  
A: I did actually discuss memory, the global workspace I referred to is often referred to as working memory.  Although long-term (permanent) memory probably affects phenomenology (e.g. tasting wine may utilize categories built into the long term memory system through previous experience), it is mostly likely not required for phenomenology.  One of the conclusions of this paper is that phenomenology and working memory are importantly distinct, and that there is much more in phenomenology than in working memory.  

Q: When I think about the difference between a computer/camera and a human brain, one difference seems to be that the former merely takes in energy, whereas a mind requires more than merely physical input—a comparative judgment of some sort, or a detection of difference, seems to be required.  It doesn’t seem that there is consciousness in the former case.  The V5 loop you mentioned might be getting closer to consciousness because it is informational…but it still seems we would need something like the frontal cortex or limbic system in order to have conscious experience.
A: Judgment is bringing some system of categories that you have in long term memory to apply to the current experience. I actually deny the premise that judgment is necessary for consciousness—I believe it is possible for an animal without long-term memory/concepts to still have conscious experience.

Q: Certain recognition processes are built into the brain, while others are learned.  So when you speak of cognitive access, which sort of process do you have in mind?
A: You seem to be raising the issue, to what extent does the discussion of phenomenology of perception require the application of concepts?  My view is that you can have experiences even when you’re not applying concepts (i.e. no activation of long term memory)—but you have a different experience when you do apply concepts from long term memory.  Wine tasting is a great example of this.  Concepts come in, but they’re not necessary.

Q: What is the difference between brain-state and neural account?  
A: I don’t mean any distinction between those, they are used interchangeably.

Q: You had three accounts of the relationship between phenomenal consciousness and awareness: same order, higher order and automatic, and you think higher order is not necessary.  I’m wondering, couldn’t either automatic or same order more congenial to your account?  The same-order view is a version of a traditionally held Kantian categorical view which espouses that there is always not only awareness not just of color, but also awareness of “I’m seeing this color” (thus there must be Kantian categories).  It seems that for you to espouse this view would make it sound like you’re building too much into it and hence would like to claim that the automatic view is more congenial.
A: I see your point.  If I had to choose between automatic view, and the same-order-plus-Kant view, the same-order-plus-Kant structure fits more with my account.  I would certainly agree with that.  
Q: Some philosophers espouse that when I experience red, I also have to be aware that “I” am experiencing red.  Your account doesn’t seem to be committed to that.
A: I may have some sympathy to that, although not talked about in this paper.  There’s a paper referenced in my paper by Pollen that espouses the view that phenomenal consciousness must involve some parts of the brain related to the conception of the self.  Kandisher’s (2001) paper writes that phenomenology requires binding in space and time (great echos to Kant).  This is a place where neuroscience seems to be rediscovering some points that philosophers have been discussing for a long time.

Q: Mesh between psychology and neuroscience.  You seem to try to establish some correspondence btw neural basis and the inner phenomenology of accessibility.  The conclusion seems that if you have a certain neural basis then you have phenomenology.  With a richer neural basis you can have phenomenology plus accessibility.  So we can have a very mechanical scheme.  Accessibility presumes phenomenology, i.e. that we have a richer neural basis.  If some neural basis is taken away then you only have the phenomenology without accessibility.  This seems to be the reduction of psychology to neural sciences.  How do you account for the content of phenomenology within that reduction?  You can locate some neural structure, neural phenomena, neural content of phenomenology.  For example, how can we structurally account for the fact that although I see and recognize a lot of faces, but in with one face I have a special emotional attachment.  I’m not sure about the idea that there is a 1-to-1 relation between such content and neural structure.  Even the mesh idea seems too strong a structure.  
A: You’re right that I am presupposing a reductionist point of view.  However, it’s not exactly that I believe in such a view, but rather I am interested in how far you can push such a view, and what you can get out of this, and are there unanswerable questions (even in a reductionist point of view).  My opponents are also reductionists, but think that there are unanswerable questions even on such a view.  One thing that fits with your point: I’m not certain about is whether or not phenomenology is multiply realizable (on a machine for example).  For now, I exclude these views.  Re: 1-1 content-structure relatonship:  The reductionist view that I’m presupposing is that consciousness itself (there being a conscious field) is identified with some kind of upper brain stem relation to the cortex (perhaps something like thalamal cortical oscillation).  The specific contents, I’m identifying with specific activations within specific cortical systems (e.g. visual motion--recurrent loops between V5 and lower face areas; face experience--fusiform face area and lower area interaction).  So I have corresponded phenomenal field with one and contents with another.  I don’t know if this is precisely right, but it’s worth pushing it to see how far it can go.  

Q: Consciousness here doesn’t seem like a natural kind, but there are reasons to think it ought to be.  
A: You’re right in a sense, recurrent activation of V5 doesn’t seem rigid enough to be a natural kind, or perhaps too specific to human beings.  One way to go about it might be to bring in multiple-realizability, but this would involve so many complexities.  Thus for now, I choose to limit my investigation to human processes.  I think that it would be premature at this stage of the inquiry to relax our assumptions about the reduction.

Q:  What do you mean by cognitive accessibility does not "constitute" phenomenology?
A: I want the paper to be about the fact that cognitive accessibility is not included in phenomenology.  What do I mean by that?  It is not a part of it, in the standard sense.  I claim that the machinery of phenomenology does not necessarily include the machinery going on in the front of the head (i.e., used in accessibility).  My argument for this is the mesh argument.  

Q: The quote from the authors in the paper who later retracted their view mentioned that their conclusion was logically flawed.  Why did they say this?
A: Their point was that they just assumed the effect was the result of sparse representation, without considering other possibilities.  

Q: One of your replies to the “refrigerator light illusion” was that it did not vanish when we found out about it…why is this a response to their argument?
A: Because it is a different type of illusion than standard perceptual illusions, it is a cognitive one, an illusion about how things seem.  In general, with cognitive illusions (say, affirming the consequent or the gambler’s fallacy), once we learn about them, the illusion vanishes.  In this case, the illusion doesn’t disappear when we learn about it. 

Q: Let me ask about the slide discussing the relationship between perceptual awareness and brain-states (neural correlates).  Since it’s the case that we’re not even reliable reporters of our own visual stimuli, how is it that we can ever make an equation between percepts and brain-states—it seems to be getting close to qualia.
A: If I understand you rightly, I agree with most of what you said.  However, I think that it is possible for you to have experiences or perceptions that don’t make it all the way to the area of the brain involved in attentive consciousness.  I’m arguing that there’s no bar for us understanding phenomenal consciousness physically. 

Q: Classical phenomenology is very concerned with, e.g., figure-ground experimentation.  What sorts of things are discussed in the phenomenology you’re after?
A: Many of the same issues come up; lots of work in this field is being done on figure-ground relationships.


Let me ask you a question now.  I think the overlap between neuroscience and philosophy is very interesting, and I’m curious whether your department has any interest in starting a section of neuroscientific study, or giving credit for neuroscience classes, etc.?
A: Yes, there have been more and more students pursuing this area of study in the past ten years…where the students go, philosophy follows.

Reading: http://www.neuphi.com/images/readings/Block_BBS.pdf

Rebecca Saxe

Thursday, March 1, 2007. 5-7PM

"Overview of Recent Conceptual Advances (and Puzzles) in Cognitive Neuroscience"

Neuphi
March 1, 2007, 5-7pm
Note-taker: J.T. , E.I.
Last Updated: March 7, 2007

Welcome from Prof. Hintikka

When it comes to two different disciplines working together, the question of how these disciplines will interact becomes important.  The way I see philosophy and neuroscience coming together here is in cooperation rather than the old model on which scientist produce facts, and philosophers come onto the science after the fact.   One model—mind can assist analysis of such models at the computational level as well as the conceptual level.  For example, David Marr’s study of the visual system distinguished different levels/aspects of the visual system.  Computational analogy, like studying mathematical algorithms, coming close to software and the counterparts, study of missing organs.  This is a good model because philosophers and philosophical logicians can make interesting suggestions at the computational level, and perhaps contribute to science.

Presentation by Dr. Rebecca Saxe

“Conceptual Advances (and Puzzles) in Neuroscience”

These puzzles may not be philosophical, but are significant.  Neuroscience studies the mind through the function of healthy, human brains.  Very successful and popular, but in its popularity certain conceptual puzzles arise with the use of fMRI technology to understand the mind. 

1st puzzle: What constitutes the same brain region from individual to individual. 
1) Why do we care about isolating brain regions? 
WhyWeCare 1): Allows us to “triangulate” mind brain associations.  To make it possible to speak about one study in relation to other studies.  Single studies can never pin down function of brain.  Particularly important when we try to draw an inference from two independent studies. 

Example from my field of research
Two separate studies seemed to suggest that: The “Right Posterior STS” responds robustly to biological motion (relative to other kinds of motion/nonhuman objects) AND to belief attribution.  The inference was: There must be something common between the belief attribution and biological motion. 
However, upon closer investigation, they were actually different brain regions. 

WhyWeCare 2) Allows us to make reverse inferences from neuroimaging results
Example context: When unexpected recruitment in the brain from imaging, we want to come up with hypothesis to explain the other
Example context: Get a brain region we know well enough so that we can use to ID how people think about that task. 
e.g. Jessica’s story (Jessica broke her neighbor’s plates), manipulate subjects’ own experience with Jessica (she’s either nice or unfair).  Subjects will make different inferences AND this will manifest differently in brain function. 
Cannot make reverse attributions unless region can be isolated. 

2) Why is it difficult to ID brain regions? 
--How to define same-region-ness?  There are three prominent different methods of detecting sameness-in-regions of the brain. 
(A) Normalize to 3D coordinates by transforming the subject’s brain to the standard one.  [most common way]
(B) Consider the curvature of the actual subjects’ brain and locate based on prominent sulci/gyri. 
(C) Characterize brain regions via cytoarchitecture (cell type, brain layer) etc.  (But they have to be dead). 
The problem is that the data from these three different methods do not agree. 

3) Possible approach to address the brain region issue? 
The method I advocate is the “Functional Regions of Interest” method, involving anatomy and function.  Method comes from visual neuroscience. 
The line between V1 and V2 is increasingly and decreasingly peripheral vision. 
But of course, this is not necessarily applicable to all brain regions. 
Fortunately, belief attributions work like this as well. 

Experimental Procedure: “Sally-Anne test” adapted for grownups:
Task Type 1) False belief task: She parked a sports car in the driveway, but at night he substituted a minivan.  Where does she think her sports car is in the morning? 
Task Type 2) False photo task: Tell story of volcano erupting and island being destroyed.  Show photo of intact island.  What did the island look like in the photograph?  Must base response on a representation and not on reality. 

Result: fMRI shows that Right Posterior STS has preference for beliefs over photographs
i.e. There’s a region that is recruited in false belief stories while not in photo task. 

Assumptions inherent in this inference:
1) Stable relationship btw anatomical structure and function across domains.  One can meaningfully generalize from one experimental context to another. 
2) Structure to function links must be general in individual, and across experimental conditions. 
3) Approach is good for telling average response for a whole region to stimulus, but bad at finding finer grain spacial patterns.  Brain must have a much finer resolution than what we see in fMRI. 

2nd puzzle [35:00]: Which cognitive function gets ascribes “special” regions?  Once they got the “Special” brain regions, how do we know what they “do”? 
Foder’s Hypothesis: Special domain specific mechanisms might exist for input systems only, i.e. only for perceptually defined domains. Central mechanisms (“thought”) will only be domain general.  In fact, early fMRI (8-10yrs) results gave supportive evidence, that brain domains delimited by stimuli.  Only faces, places, shapes, human bodies, motion had special brain regions in vision. 

However, counter-evidence seems to be posed by belief-attribution study.  Data indicates that contra-Foder, perhaps there IS a brain region that is specifically recruited when people think about other people’s thinking. 

False Belief but not False Photographs is consistent with three hypothesis. 
(1) Region reacts to anything about a person
(2) Involved in attributing any internal, subjective and visible state
(3) Involved in attributing representational mental states (e.g. thoughts/desires)

Test hypotheses by three new story types:
A. About a person but not beliefs (“He was ganglion-looking, wore baggy jeans”)
B. Subjective, internal, visible states (hungry, achy, thirsty), but not representational mental state (“Skipped breakfast.  Her stomach was rumbling.”)
C. About a belief but not about false beliefs (“He knew flight was his sister’s.  It was the only one that was delayed.”) 

Data & Inference [41:00]: 

RTPJ (Right temporo-parietal junction) responds only to story type C → Seems to support hypothesis #3, that is, RTPJ is involved specifically in attributing representational mental states, rather than reacting to anything about a person or about any kind of state of a person.

Additionally…

-Modality of presentation (audio or video), and type of story does not make a difference.   

-Does not make a difference whether mental beliefs are explicitly stated (e.g Story C—esplicitly stated “He knew….”) or implied (e.g. Jessica story—implied but never stated what Jessica was thinking).  As long as subjects can be provoked to infer or attribute beliefs to the characters, the RTPJ responds.   

-RTPJ also responds to identical, non-verbal stimuli. 
    Evidence is from “Girl & Chocolate experiment”. 
    Prompt 1 [instructs to follow rules]: “If girl is facing box, respond where it went last.  If girl is looking at box,         respond where chocolate went first.” 
    Prompt 2 [instructs to attribute belief]: “Where does the girl think the chocolate is?”      

    Only prompt 2 elicits RTPJ response.

3rd Puzzle: Is mind specialization innate?  How do specializations develop? 

Foder: Modular mechanisms are innate. 
OR
Neuroanatomical development as a process of Canalization, increasing specialization with development. 
Any evidence where non-innate specialization? Ex. Words.  (It appears that specialization regions develop to handle words, clearly not innate.)  Left fusiform gyrus recruited in looking for words in own language/language you read than in any other language. 

But what about RTPJ to belief attribution?  Innate or derived/developed?   

Experimental Task: Children listened to short stories with three subsections.  (A. Pond is… [physical scene information] B. Mr. McFeeglebee is old.  Georgie is Mr. McFeeglebee’s nephew [social but non-mental-state information].  C. Georgie likes…, knows…, decides… [mental state information]) 
Response of RTPJ in children (6-11yrs old) to different subsections of the story was monitored and compare to adults.   

Result: As kids get older, the less the RTPJ responds to B.  For 6 year olds, the RTPJ responds non-selectively to stories B and C.  Specificity increases as the subjects become older.  12 year olds’ and adults’ RTPJ have no response to B. 
[54:00]

DISCUSSION

Q: You are interested in the “region”.  Naively, we understand the region as spacialy defined, but in addition it may (you say) be architecturally or functionally defined—this seems to be a problem.  How do we know that there is a 1-on-1 correspondence between region and function?  Maybe the same region will appear in different functionalities. 
A: We believe that the brain functions are correlated with specific behavior of neurons, and it happens empirically that these cells group together systemically into a region.  I don’t think a region is an ontological unit, certainly not a natural kind.  In fact, we can’t even make a claim as to the causal efficiency of regions. 
Q: Was the order of the chocolate box test fixed?
A: No, random in order, but the subjects saw all combos.
Q: How different can brains be?  How large a transformation is necessary to normalize them? 
A: Very different.  That makes such procedures highly inaccurate. 
Q: Are there any data about how the specialization of belief attribution brain region develops in autistic people?
A: Not yet, tests are starting.
Q: Are regions involved in justification of beliefs? 
A: We don’t yet know whether the regions involved cares about, e.g. the justification of a belief. 
Q: Clinical studies of patients with damage in that region?
A: The best studies of such patients agree with these results, though they tend also to be impaired in other ways (esp representational cognition, maps and signs)
Q: The “regions of the brain” story is fairly plausible, but it is surprising.  Have you given any thought to why these regions develop? 
A: I have no good answer, but I think it has a similar reason as the fact that we have a special region of the brain for face recognition—instead of simply relying upon general shape recognition function.  There would seem to be a mechanical or computational advantage. 
Q: Do you make use of the pragmatics of speech vs logical inference? 
A: I am very interested in this.  Because of the high cost of belief attribution, it would seem impossible for us to attribute belief every time we use the word “the”.  Wer are currently exploring the differences that show up with propositions like “It is raining” and “She believes it is raining”. 
Q: Why is being hungry not a representational state? 
A: This analysis comes out of a theory meant to explain the development of children’s conceptualization of minds.  The trouble is to explain why children are fluent with certain concepts of mind but ignorant of other aspects (e.g. the difference btw ignorance and being wrong).  It turns out that (attribution of) representation is one of these aspects.  This indicates that hunger is not conceptualized as representational. 

Download mp3: http://www.neuphi.com/images/mp3s/070301 Neuphi Saxe 1.mp3