Friday, February 9, 2018

The philosophical meaning of the object assembly test and its connection to "collapse into fantasy / reality"



Virtually every practicing psychologist had encountered chidren who could not grasp what they were assembling in the object assembly task (when they were not told in advance what the object was supposed to be).  These children usually progress by trial and error, erratically, grope in the dark, and only if and when they finish the task, and the object is complete, they say in surprise: "Oh, that's a dog!" (There is no dog there but I will not reveal items from the test here ...).

There is no doubt that these children have a disadvantage compared with children who see and say instantly, before beginning to assemble the pieces: "This is a dog."  Children who know instantly that it's a dog know where they are going.  They can monitor their work and check on their progress towards the goal, because they know what the goal is.  They hold the object in its entirety in their mind's eye.

Is it possible that the style of work and perception that a child exhibits in the object assembly task will appear only in this test? Perhaps this style characterizes the way he approaches the world, the way he interprets the world? And what are the implications of such a style?

Andy Clark is professor of logic and metaphysics at Edinburgh University in Scotland. Clark wrote an article in 2013 that was received with a wave of reactions from researchers and philosophers,  and was quoted about 1700 times in the professional literature. This is a very difficult article to read (for me anyway). Clark presents a theoretical concept that I understand was first introduced by Karl Friston in 2005. The theory is called HIERARCHICAL PREDICTIVE PROCESSING or HPP. The basic idea is familiar from Kant's philosophy, but it is presented here in new and interesting costume, which may help to understand what happens to a child who does not see Gestalt in object assembly.

But not only object assembly.  This theory offers an explanation of perception as well as other phenomena such as attention, mental imagery, dreaming, hallucinations and delusions, emotions, autism, and even reading difficulties. In this post I will focus on some of these phenomena.  I'll also try to walk on less secure ground for me and offer some links between cognition and psychoanalytic concepts (in the spirit of the work model presented in the previous post).

When I look out the window and see countless little green bits swaying in the breeze, my brain must understand that I see a tree swaying in the wind. How does that happen? According to HPP theory, the brain does not wait passively until sensory signals arrive, and then interprets them. The brain constantly produces predictions about the sources of signals it will receive from the senses. These predictions are the best guess, the best hypothesis the brain has about the sources (the objects in the world, in this case: the tree) from which sensory data are obtained. The brain creates these predictions based on its expectations and beliefs about how the world is structured, that is, based on its pool of knowledge.

As these predictions "descend" to sensory organs they affect the way the sensory signals are perceived. When there is a good match between the predictions and what is actually absorbed in the senses, all is well: the prediction worked. The world is really what we predicted it would be. 

When there is a mismatch between the predictions and what is actually absorbed in the senses, prediction errors, ie, the differences between what the brain predicted and what sensory organs have absorbed, "ascend" from the sensory organs to the brain.  Prediction errors cause slight changes in subsequent predictions to fit them more with what exists in reality. Because predictions are based on models we have of reality, that is, on our crystallized knowledge, prediction errors lead to changes in the models, changes in the crystallized knowledge - that is, learning. The goal of the brain is to minimize prediction errors as much as possible. The brain aspires to be in a state where it perfectly predicts the sources of signals it receives from the senses. There is a perpetual cycle of predictions and error correction.

This means that what we perceive is determined by our predictions as much as by the signals that come to our senses from the objects we perceive. The predictions that "flow" down are rich in content, while the sensory information that "flows up" is composed only of prediction errors. The rich prediction that "flows down" enables us to know what is in the world, to "grasp" the world before it is actually caught in our senses. If this theory is true, then the role of sensory contact with the world is only to verify the best guess of the mind about what is out there, and when necessary - correct it. According to this approach, our predictions are the main source of all the contents of our perceptions, even though these contents are continually tested and constantly corrected by the prediction errors generated by sensory input.

The sensory signals always come to us with a certain "noise". For example, when we listen to a person talking, we do not hear the speech sounds in a completely clear way. There are background noises, the person "swallows" some of the sounds and does not separate the words (the speech sounds flow in a continuous sequence). We are able to understand what is said because of the predictions that the brain projects toward the sensory signals. The brain expects to hear certain things (according to the context of the conversation, according to the knowledge we have about syntax and grammar) and the expectations help it bridge imperfections in input quality, complete auditory Gestalt and understand what is said.





Perceptual illusions can also demonstrate HPP theory. Square A in the left chess board appears to be darker than Square B. But as indicated by the gray lines drawn on the right chessboard, the two squares are identical in color. We do not see the color of the squares "objectively".  What we see is our best guess about the color of the squares. We know that a chess board is made up of dark and light squares, and that between four dark squares there must be a light square. Our expectations about the structure of a chessboard affect the way we perceive it.

We see the green object lying on the chessboard as a cylinder even though it is really a two-dimensional object. This is because we interpret the "shadow" painted on the left side of the green object and the circle painted at the top of it as indicating 3D, and the object being a cylinder. We do not perceive what is actually drawn but our interpretation of what is drawn. And so it should be: the role of the visual system is not to be a physical measure of light, but to help us perceive reality. And the predictions that the brain produces do indeed help us perceive reality.

Our ability to generate predictions about objects in reality is at the basis of our ability to generate mental images and to dream. Predictions are perception without sensory stimulation. In a normal process of perception, predictions are always compared with sensory data. We use the same system that generates predictions about what is in the world to create mental images or dreams that are not contrasted with sensory data.

Could it be that the infant's evolving cognitive ability to create a mental image of his mother or caretaker (a mental image of the mother as a whole or partial object, a mental image of the mother's look, voice, touch, etc.) helps him to survive the moments when the mother is not with him? We know that there are individual differences in the ability to create mental images. How do babies with poor mental imagery endure in these difficult moments?

Our predictions not only affect what we perceive but also our actions in the world. When we want to pick up a cup of coffee, for example, we imagine or predict the state of the future goal (our hand holding the cup of coffee) and generate motor commands that will help us realize the prediction. Of course, we also predict the touch of the cup, the smell of the coffee and its taste. As predictions flow down, prediction errors flow up and set the next predictions flowing downwards.

Hallucinations and delusions

We have seen so far that what we perceive is always a subtle combination of "top-down" knowledge-based predictions and "bottom-up" prediction errors.  Normal perception is always a balance between sensory signals coming from the world and our expectations for these signals.

Could it be that precisely here -  at the point of connection between expectations / predictions and sensory signals, between the internal and external realities, between the subjective and the objective - is the potential space that Winnicott speaks of?

Our perception of the world is a kind of "controlled hallucination," a fantasy that fits reality. The predictions of the brain are restrained by sensory information from the world. We have partners to this controlled hallucination: different people who look at the same object (for example, the image of the cylinder on the chessboard) usually agree about what they see. When we agree about our hallucinations we call them reality.

But sometimes there is an imbalance between the prediction errors that flow up and the predictions that flow down. When the scale is tipped too much toward the predictions (the brain over- weights predictions and under- weights prediction errors), we may be under the impression that we perceive certain objects that are not actually there, and then we experience a hallucination.  Alternatively, hallucinations and delusions may also arise when something goes wrong at the level of sensory data, which causes us to generate false prediction errors.  False prediction errors (ie. incorrect corrections of our predictions) cause unjustified changes in our internal model of the world (delusions).  From this erroneous model, we generate false predictions that flow down and affect the way we interpret sensory data. Thus more erroneous prediction errors are created that reinforce the wrong internal model and thus a cycle of misperception and misinterpretation of the world is created.

I suggested earlier that one can think of the point of balance, the delicate connection between the inner world (the predictions that flow down) and the external world (the prediction errors that flow up) as the potential space. If this is the case, we may think of tilting the scales toward the predictions (attributing excessive weight to predictions and not enough weight to prediction errors) as a collapse of the potential space to the fantasy pole. Ogden argues that in trauma or great anxiety, we may experience fantasy as reality. A frightening thought is experienced as a frightening reality, a crocodile doll a child plays with as a real crocodile about to devour the child. The child loses the ability to play.

Perhaps in moments of collapse into the pole of fantasy, perhaps when the brain attributes too much weight to predictions and too little weight to prediction errors, a child will tell TAT stories that are not well linked to the stimuli in the picture. Perhaps such a child will give a very unusual interpretation to the stimuli in the picture. Perhaps such a child will see the objects in the picture as objects in the real world and respond to them as such.

Object assembly

I suspect that difficulties of children who do not see the whole object when working on an item in object assembly may result from tilting the scales to the sensory signals (the brain over- weights prediction errors and under- weights predictions). When predictions are underweighted, they can't guide perception and can't form expectations for the whole object.   Perception is guided primarily by sensory signals. Thus, the ability to integrate what is perceived and to interpret it is greatly weakened. The ability to create categorization is weakened, and each piece of stimulus is perceived as a separate object.

In the spirit of Winnicott and Ogden's theories, one may think of tilting the scales toward the sensory signals (attributing excessive weight to the senses and under-weight to the predictions) as a collapse of the potential space to the reality pole.  This situation can also occur, according to Ogden, following trauma and when people feel anxious.  A person may cling to reality defensively, as a defense against fantasy.  In this state a person can not play or imagine.

Perhaps in moments of collapse into reality, when the brain attributes too much weight to the senses and too little weight to predictions, a child will not be able to tell a story in the TAT test, and instead describe in detail the objects he sees in the drawings. In this way he clings to reality and leaves no room for fantasy.

I am reminded of a short story by Jorge Luis Borges, "Funes, the memorious" , that my friend Garciella Neumann introduced me to. In this fictional story, Borges tells the story of Ireneo Funes, a 19-year-old boy who was hit in the head after falling from a horse. Borges writes:

"On falling from the horse, he lost consciousness; when he recovered it, the present was almost intolerable; it was so rich and bright.  A little later he realized that he was crippled. This fact scarcely interested him. He reasoned (or felt) that immobility was a minimum price to pay. And now, his perception and his memory were infallible.  In effect, Funes not only remembered every leaf on every tree of every wood, but even every one of the times he had perceived or imagined it.

…He was, let us not forget, almost incapable of general, platonic ideas. It was not only difficult for him to understand that the generic term dog embraced so many unlike specimens of differing sizes and different forms; he was disturbed by the fact that a dog at three-fourteen (seen in profile) should have the same name as the dog at three fifteen (seen from the front).

…He was the solitary and lucid spectator of a multiform world which was instantaneously and almost intolerably exact.

…Without effort, he had learned English, French, Portuguese, Latin. I suspect, nevertheless, that he was not very capable of thought. To think is to forget a difference, to generalize, to abstract. In the overly replete world of Funes there were nothing but details, almost contiguous details".

Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and brain sciences36(3), 181-204.  http://www.fil.ion.ucl.ac.uk/~karl/Whatever%20next.pdf

Your brain hallucinates your conscious reality | Anil Seth  https://www.youtube.com/watch?v=lyu7v7nWzfo&t=876s

Borges, J. L. (1962). Funes, the memorious (pp. 59-66). na.  http://marom.net.technion.ac.il/files/2016/07/Funes-the-Memorious.pdf
Ogden, T. H. (1993). The matrix of the mind: Object relations and the psychoanalytic dialogue. Jason Aronson

 

 

No comments:

Post a Comment