Recently returning to Anil Seth’s book at reference 1, I took another look at the review at reference 2 which may well have been what got me going in the first place, where I found and then downloaded a copy of reference 3 to my Kindle. Which at £3.99 is proving very good value.
The first part of the book is about a boy, Liam McCoy, born near blind, going through a programme of treatment which culminated with supplementing his own lenses with plastic ones when he was in his teens – then a new procedure – and about his long road to improved vision. A book with plenty of human interest, but written, nevertheless, by a professor of biological science and neuroscience at Mount Holyoke College in Massachusetts, a women’s college that is gender diverse. At one point, we get a little story about teaching horsetails, at another one about teaching leaves – so she really is teaching both botany and neurons.
She is also qualified, if I can call it that, to write the present book by her own problems with vision, being cross-eyed from infancy, a defect which, with therapy, she largely was able to fix in middle life, when a new world of stereoscopic vision opened up for her. She, for example, now appears to herself to be in a mirror rather than on a mirror.
Liam, now a young adult, has a job and a good life. His vision, while some way from normal, continues to improve. And although I don’t think it is put in so many words, I think he is glad to have had the operations, to have made the transition, to have put in the work needed to make it work. Some people in a comparable position have not been so lucky.
Along the way, Barry tells us of all manner of interesting features of vision, particularly those parts of vision responsible for assembling the rather geometrical bits and pieces from the first part of the vision pathway into the more coherent images of known objects, nicely arranged in space, further along. Some of this I share below.
The snap at the head of his post started out as a coloured drawing created by Liam when he was a young adult, seven years after his corrective surgery. It nicely illustrates part of his continuing problem: he can see all the basic shapes and their colours, but he does not tend to group them into things, in the way of most of the rest of us. Other people seeing the drawing find all kinds of interesting things in it, particularly animals of one kind or another, which he does not see at all. He just sees the shapes and their colours. He has to consciously work to assemble them into objects, although he is getting better at it.
She tells us that Rubin’s face/vase image (above) is all about boundaries. If the boundary line between black and white is attached to the faces, we see faces. If it is attached to the vase we see a vase. An arrangement echoing, as it happens, one version of LWS (of reference 8), in which boundaries between two objects are attached to one or the other, rather than being shared. With the object with the boundary being the one that is in front. In this case, the brain may be unsure and the subjective experience flickers in a more or less random way between the two interpretations. An alternation rather than a blend.
Noting that the slightly different Gestalt version (for which see below) is that the brain can’t decide what is figure and what is ground. With one link being that Rubin, active at the beginning of the twentieth century, was one of the first people to describe the subjective visual experience of being made up of a small number of foreground objects plus background, very much in the way of a proscenium arch theatre with its flats and backdrop. An approach I find appealing. If nothing else, there must be a reason why proscenium arch theatres are so popular and successful.
Barry goes on to point up the importance of boundaries generally, with their being a presumption that collections of lines which, more or less, form simple closed loops are boundaries of objects. With the converse being that if you break the loop, the object is apt to vanish. Considerations which are very much to the fore in the matter of camouflage, in both its natural and military forms.
She reminds us of Oliver Sacks’s patient Dr. P, who could provide an accurate description of a glove in terms of what one might call its topology, but who did not know that it was a glove – even though he clearly knew about gloves in general, as he could recognise the same glove by handling it.
While Liam, we are told, laboriously and consciously assembles the bits and pieces which he sees into objects, a process which improves with time and effort. His brain is able, up to a point anyway, able to complete consciously what it cannot manage unconsciously. I associate to the way that a manual skill, like working a carpenter’s chisel, moves from consciousness to unconsciousness as one builds the skill. One has to attend to what one is doing, but one does not have to think about it anymore.
She tells of getting, or getting back, stereoscopic vision, and how mirrors suddenly sprang into life. Instead of appearing on the surface of the mirror she is now inside it. Complicated objects like trees moved from being flat to having depth. She tells us that Oliver Sacks (bravely) relates the same experience in reverse in the book at reference 4.
She goes on to provide various stories which are very suggestive of top-down processing, suggestive that the subjective experience of the visual scene results from the interaction of bottom-up processing coming up from the eyes and top-down processing coming down from higher regions of the brain. Very much what Seth might call prediction processing or controlled hallucination, what a psychiatrist might simply call a hallucination – and what a Freudian might call a projection. I associate to the ease with one can project all kinds of things, particularly animals and faces onto cumulus clouds.
I also tried a small experiment of my own, trying to visualise various things that I knew well. Which I could do, but it was all a lot of effort and the results seemed rather meagre, although there were hints that this was something that I might get better at if I worked at it. While in the ordinary way, I take quite a lot of interest in the visual world and I do respond to paintings, the interiors of familiar houses and to buildings. I used to draw. No idea, just presently anyway, how one might compare sighted person with another in this regard.
While in an entry dated December 2007, Sacks describes the way that his own brain sometimes filled in the big hole in his right-hand visual field, the result of treatment for a small tumour behind his retina. Quite good, for example, at regular patterns in carpets and at brick walls seen from a decent distance. It could even manage his feet after a fashion – but not other peoples’ feet.
There appear to be a lot of memoirs written by people with problems with their eyes, mostly coming but some going, and both Barry and Sacks refer to several of them. With some of the latter pointing up how different people can respond in very different ways to very similar problems. For example, some people who go blind, rather than being born blind, develop remarkable powers of inner visualisation. There is no one way to go blind.
Odds and ends
Liam could talk, read and write, although perhaps not with normal fluency. I imagine that this is less likely to be the case for those with problems hearing, rather than with problems seeing. Absence of language being a major barrier to development. Or as one might put it, being blind cuts one off from the world of things, while being deaf cuts one off from the world of people. The world outside, rather than the world inside.
I note the challenge that Barry’s binocular vision examples present to the two and a half dimensional world of LWS-R, hypothesised to make use of what is in effect a monocular, more or less two-dimensional view of the world. The current response is that LWS-R includes the capability to annotate its objects, annotation which could include information about depth. Which might work, but which certainly needs to be elaborated.
It is not just blind people who can have very different visual lives. Sacks reports hit mother doing something quite remarkable in the snap above. While I recall an eminent crystallographer lecturing at the Royal Institution telling us about a colleague who could rotate a complicated crystal in her mind. A rotation which presumably had some practical value.
Gestalt
I have been prompted by something in all this to take a quick look at the Gestalt view of perception, which appears to have been well ahead of its time when it was first formulated, around 100 years ago. And, in some ways, to look ahead both to the Tononi’s Integrated Information Theory and to Friston’s free energy principle. There is an accessible introduction to be found at reference 6.
A gloss might be that brains like to organise things into nice tidy wholes, to which end it applies various organising principles to the rather chaotic signals coming in from the outside, in particular from the eyes. A view which fits nicely with that about subjective vision being the result of interaction between bottom-up and top-down processes.
A view which is said to be in opposition to that of the structuralists who hold that the whole can be fully explained as the sum of its parts, with Gestaltists arguing that the whole is greater than the sum of its parts.
There is a tendency to express Gestalt perception in terms of axioms and principles, terms which to me rather overstate their case. Nevertheless, there does appear to be a lot going for them and I dare there is a fair amount of support from today’s work on vision, machine or otherwise. I offer a short glossary – about the details of which I dare say one might argue for a long time. Elementary texts such a reference 6 usually offer neat illustrative diagrams – which I do not include here.
Closure. When we see a complex pattern, we want a simple explanation – and will, if necessary, supply the bits that are missing which reduce the complex to the simple. For example, arcs missing from a circle.
Common fate. Things that move together, are together. So we see a shoal of fish or a murmuration of starlings not a whole lot of fish or a whole lot of starlings. An example of the latter is snapped above. Videos are more impressive, for an example of which see reference 9.
Continuity. The brain prefers continuity and smoothness. So one letter on top of another will usually be resolved, correctly, into two letters. A bicycle behind a picket fence will usually be resolved, correctly, into a bicycle rather than a collection of more or less unrelated slices. The alternative explanation being a good deal less probable, at least most of the time.
Emergence. The way that the perception of, for example, a tiger in the undergrowth or a snake on a bed of dead leaves, can suddenly pop-out or emerge from the background.
Figure and ground. There is a tendency to organise a visual scene into figures against a background, with the details of this last being rather blurred out. For which see the related remarks about Rubin and proscenium arches above.
Frames of reference. The visual field usually has a frame of reference; there is an up and down, left and right. Sometimes we need to think in terms of more than one such frame: we have at least one, determined by posture, position of head and eyes. Then the frame of reference for our visual field – perhaps the inside of a railway carriage – might be moving with respect to some larger frame of reference. Think of the confusion sometimes caused by sitting in a railway carriage next to another train which then moves off. I also associate to the way that hunting animals simplify all this for their brains by pointing at potential prey, where by pointing we mean a posture where the whole body is lined up and pointing at the prey. Humans do this sort of thing too. See reference 10 for a snap of a pointing dog.
Invariance. Most objects exhibit a high degree of invariance, even under what might be called plastic transformation. A table is still a table even if you pull it about a bit. A chair is still a chair when it is upside down. But, contrariwise, we find it much harder to compute faces when they are upside down.
Memory. The brain uses memories of objects past in building perceptions of objects present from the raw materials coming in from the eyes,
Multi-stability. Sometimes there is more than one viable perception available, in which case the subjective experience may flip-flop between them.
Pragnanz, or simplicity and conciseness. We go for a simple explanation if one is available, discarding or not attending to a lot of the detail.
Proximity. Other things being equal, things that are in close proximity to each other are grouped together.
Reification. The brain likes to organise bits and pieces, otherwise parts, into wholes.
Similarity. Items that are similar tend to be grouped together regardless of whether or not the similarity or relationship actually exists. For example, we tend to join up the dots.
Symmetry. The brain is alive to the sort of symmetry expressed by opening and closing brackets, to link the brackets and what lies between into an object.
I believe that some people use all this to argue that we humans do not start with a blank sheet as regards the visual world. We come ready equipped with neural machinery which has tendencies and preferences regarding the organisation of the visual world.
Gestaltists have also moved into psychology and management consulting – organisation development – generally. This I find less attractive.
Conclusions
Blindness apart, all good stuff – with lots of support for our vision being a mixture of bottom-up and top-down processes – if not for Seth’s predictive processing. It was well worth looking up Hutson’s reference.
Barry is a fluent and accessible writer with nicely chosen, uncomplicated illustrations of her points, a sample of which is included above.
I should add that while I have (rather quickly) read the whole of Sacks (reference 4), I have only read Part I of Barry (reference 3) and just started Part II – finding a suggestion there that the Gestalt may copy across from seeing to hearing. There may be more to come!
PS: search for the snap of the drawing at the top of the post, including Google’s image search, fails to turn up a colour reproduction. So perhaps, while his name is in the public domain, Liam still has some privacy.
References
Reference 1: Being You: A New Science of Consciousness – Anil Seth – 2021.
Reference 2: Our brains exist in a state of “controlled hallucination”: Three new books lay bare the weirdness of how our brains process the world around us – Matthew Hutson, MIT Technology Review – 2121
Reference 3: Coming to Our Senses: A Boy Who Learned to See, a Girl Who Learned to Hear, and How We All Discover the World – Susan Barry – 2021.
Reference 4: The Mind’s Eye – Oliver Sacks – 2010.
Reference 5: http://psmv4.blogspot.com/2019/11/more-on-making-regions-into-objects-and.html.
Reference 6: The Legacy of Gestalt Psychology: Since its inception early in this century, Gestalt theory has made significant contributions to the study of perception, learning and social psychology. These contributions remain influential today - Irvin Rock, Stephen Palmer, Scientific American – 1990.
Reference 7: https://en.wikipedia.org/wiki/Strabismus. Being cross-eyed is one part of this.
Reference 8: https://psmv4.blogspot.com/2020/09/an-updated-introduction-to-lws-r.html.
Reference 9: https://youtu.be/eakKfY5aHmY. Pity about the music.
Reference 10: https://psmv4.blogspot.com/2020/01/on-as-aspect-of-attention-and.html.
Group search key: sre.
No comments:
Post a Comment