Words blocking thought

I have commented in the past on a particular weakness in the way that we make difficult and important decisions. See for example, references 2 and 3.

To recap, suppose we have to make a difficult decision about something, perhaps whether X or Y is the villain in some country house murder, or in some story by Simenon. We mull the thing over, mull over points for and against. Then all of a sudden we come to a decision, not a very consciously made decision, rather a light bulb moment, powered by something going on in the unconscious (UCS). But the UCS having made the decision for us, we then take it into consciousness by expressing it in words, publish it and hold to it against all comers for ever more. We grab all the evidence for the decision and fail to consider or even see all the evidence against, except perhaps evidence against which we are confident we have already rebutted. We only look in places where we are going to find evidence for.

This phenomenon is pervasive. With mathematicians one talks of so and so having good intuition. Some people have gut feelings. Some people like to sleep on problems and wake up to the solution. Government purchasing decisions, at least in my own modest experience, can also be the subject of light bulb moments. But all these feelings and moments are notoriously unreliable, and bigger organisations, particularly police forces and governments, have worked hard to keep them under control with process, governments being reinforced in this by their equally pervasive belief that it is process which enables them to extract silk purses from the ears of sows (see reference 7). Have a bit of process to support them and you really can get monkeys to type out a Shakespeare. So important decisions are subordinated to structured and transparent processes and process police are pushed into the foreground, often to the annoyance of the people who are hot on light bulbs.

Rather different is talking oneself, writing oneself or being talked into a decision. So one talks away and after a while one realises that one has expressed a decision without having consciously made it. But while one might go along with such a decision, one might stick with it, there is often a nagging sense that the internal logic of the words, the flow of the words has come to the wrong answer, without one being able to put one’s finger on quite what it is that is wrong with the logic. In this case at least, one is actually quite grateful to have some process to fall back on.

That aside, my point is that once we articulate a decision which has been made by the UCS in words, we lock on. Which might be good for getting the business of the moment carried through, but which does not always get one to the right place.

Against this background I was interested a few days ago to come across reference 1, a paper which was the work of a husband and wife team getting on for thirty years ago and which appears to have started life as the husband’s doctoral dissertation, a husband who is now a professor of psychological and brain sciences at the University of California, Santa Barbara.

This paper described some work testing aspects of the then prevalent theory that verbalising things helped to lock them down into memory. Quite a lot of space was given to discussing the work that had been done by others on visual and verbal aspects of visual memories, including the idea that this processing is both time and resource limited and that the brain has to decide about how to allocate resources between the various threads of processing, just in the way that a computer would. And maybe the brain thinks that using words about some image in memory is a lot quicker and easier than going back to the image itself, or whatever condensed version of the image has been retained.

I dare say that a lot of work has been done since on finding out at what level the brain tries to match features when it does face recognition. Does it do crude least squares matching at the level of pixels? Does it do stuff at a rather mathematical level, say estimating the geometry of facial features, say the distance between the point of the nose and the point of the chin or the distance between the two ear lobes? Does it do stuff at the level of the sort of features that can easily be described in words, things like the presence of blue eyes, ear rings, long hair or a bent nose? With the work reported at reference 5 and noticed at reference 6 suggesting that the answer, in the case of at least a couple of monkeys, neither of whom do words, is indeed rather mathematical.

However, the core of this much earlier paper is an account of six experiments, experiments which involved using possibly as many as hundreds of subjects, students from the University of Washington (the one at Seattle), students who were described as ‘who participated for course credit’. I suppose the not unreasonable idea is that one learns by participating.

The basic structure of the experiments was as follows:

• Show the subjects a short film, involving a face
• Pause
• Get one group of subjects to write a description of the face, get the other to do some unrelated activity
• Pause
• Test the ability of both groups of subjects to pick the original face out of a small collection of similar faces.

The basic result, tested and elaborated over the course of five experiments, was that verbalising the face made it harder to pick the first face out of a line-up up later.

Part of this was that where there where a face had easily verbalised cues – say a beard or a tattoo – which distinguished it from other faces in the line-up, then verbalising helped. But not otherwise. Somehow, the business of verbalising a visual memory seemed to be interfering with visual access to that memory later.

There was a sixth experiment (actually third in sequence) which did something similar for colours, another area where peoples’ recollections are very unreliable.

So there is something going at this more basic level which nicely reflects the problem with which I started. With verbalising the face taking the role of verbalising the light bulb moment.

PS: as it happens, at about the same time I happened to read a few pages about Cicero from the beginning of Trollope’s life (reference 4). Where I found a discussion of the relative merits of making one’s mind up and crashing on with it and those of swinging in the wind, drifting this way and that, honestly and reasonably unable to decide. With, on this story, the former being the way of both Julius and Augustus Caesar, the latter that of Cicero. Perhaps the former were more prone to light bulb moments than the latter, who managed to keep an open mind for ever and ever. Of the three, only Augustus died in his bed, so there is no simple moral to be drawn.

References

Reference 1: Verbal overshadowing of visual memories: some things are better left unsaid – Schooler, J.W. and Engstler-Schooler, T.Y. – 1990.

Reference 2: http://psmv3.blogspot.co.uk/2017/06/flou.html.

Reference 3: http://psmv3.blogspot.co.uk/2016/10/a-choice.html.

Reference 4: The Life of Cicero, Volume One – Anthony Trollope – 1880.

Reference 5: The Code for Facial Identity in the Primate Brain – Le Chang and Doris Y. Tsao – 2017. Open access.

Reference 6: http://psmv3.blogspot.co.uk/2017/06/reading-minds-of-monkeys.html.

Reference 7: https://link.springer.com/chapter/10.1007/978-1-4020-8427-0_21.