Opportunities for PCT

I’ll give it a shot

Bruce – I’ve been delving into analytic philosophy recently (initially to track down “ghost in the machine”) and latched onto Ryle’s work. For me, it’s tough stuff, way outside my training. Subsequently, through a few links, came across Julia Tanney’s entries in Stanford’s encyclopedia where she has supplements on perception (https://plato.stanford.edu/entries/ryle/perception.html) and imagination ( https://plato.stanford.edu/entries/ryle/imagination.html). I would welcome a PCT interpretation of these, perhaps starting with perception. It seems to me these Ryle/Tanney interpretations are open loop.
Comments welcomed.


After reading (only once) both linked articles, my first comment is that Ryle/Tanney deal only with conscious perception. The philosophical issue of “experiencing” is separate from PCT. That’s my conclusion. Read the following only for the skeleton of my argument to reach this conclusion.

What did Powers suggest about conscious perceptions on the few occasions when he did talk about them on CSGnet? Without searching the newly provided archives (congratulations and thanks to all concerned with that effort), my main takeaway is that he argued that any conscious perception is likely to be one that could be produced by a perceptual function that exists in the control hierarchy (non-conscious, of course, like everything in the control hierarchy).

Both of the linked papers ignore the possibility of non-conscious, therefore not experienced, perceptions being sometimes available to consciousness. Perhaps they would call reference to the control hierarchy (probably totally unknown to either Ryle or Tanney) “mechanism”, which is explicitly of no interest to them. However, to Powers, consciousness had the job of influencing, and perhaps helping to reorganize, the control hierarchy, changing the actions that would be observed to correct experienced error. Does not the success of MoL support the “reorganizing” part of that idea?

Imagining is a part of all that, is it not? Tanney uses as an example the ability to hear and experience a live performance of a Bach partita without previously knowing “how it goes”, an ability not available to “hearing it in the mind’s ear”. That difference seems to be a real problem to Ryle/Tanney. For PCT, to experience the partita in one’s head requires one to know “how it goes”, because that perception is derived from the (?)sequence level perceptions in the unconscious hierarchy, invoked when “controlling in imagination”.

But to invoke the perceptual control hierarchy is to explain mechanism. and Ryle/Tanney are explicitly not interested in mechanism. I think their many questions can be reduced to two related questions: What is consciousness? and “Why do we have consciousness?” Those questions may be linked to PCT, evolution, and reorganization some day, but for me right now, they are entirely independent of PCT.

Ryle is an ‘ordinary language’ philosopher. The first (Ryle on perception) seems thoroughly kerfluffled with semantic categorizations of words. The second (Ryle on imagination) goes a little farther, declaring that language that is appropriate for talk about perceptions derived from the environment is not licensed for imagined perceptions. He talks of neural signals in the brain, and shows the folly of supposing that replicas or images or representations of the world reside in the brain (and an infinite regress of homunculi or ‘ghosts’ to perceive them).

The error is to place the difficulty between perception and language, when it is between perception (signal) and perception (experience), and language as always stands off to the side.

Drawing your metaphoric sword Bruce and, just like Alexander, sliced right through the Gordian knot of Julia Tanney’s exegesis of Ryle’s views. Nice.

Aw, gosh, thanks. Oh what tangled webs we weave when first we talk of how we perceive.

I think the problem is that people understand “control” as control of action. It can be that some person or organization controls how I or some other person may act (or even think) or then I control myself. I am not a native English speaker but for me the name “perceptual control theory” first sounded like a theory about how to control someone by affecting the perceptions of that someone.

What I think that should be communicated as meaning of the name of PCT is that it is a theory of action (or behavior if you necessarily want) as control of the perceptions of the actor.

One suggestion for a replacement (which is probably not a realistic idea at all) could be APCT = “Action as Perceptual Control Theory”. However, it could be better to use old PCT as an abbreviation but open it with an explanation that it means a theory
which tries to explain the action (or behavior) of a human being or any organism as control of the perceptions of that same person or organism.

I think that your description is technically correct, Eetu. However, I think of PCT as how we control something out there as evidenced by our perceptions. As control systems, we do indeed control our perceptions but doing so is our means of controlling things out there.

I agree Fred. But when you are introducing PCT to outsiders, they certainly already understand action (or behavior) as controlling or affecting something out there. What PCT adds there is the idea that when the subject acts by affecting something out there she is actually controlling her own perceptions.

I think this is in a way quite near to the classic or humanistic idea of free action: we do not act because something forces or controls us but we act because we have a goal and we want something. And thus it is the actor oneself who is controlling - not controlled.

Perhaps it is this kind of idea which we should stress to make a more positive impression of PCT.

1 Like

Most people I know are well aware that all they know of the world out there is known to them by way of their perceptions. To tell them that they really control their perception of something instead of the something is likely to elicit a quizzical look followed by “So what?” So, I’ll ask you, Eetu, So what?

my post was mainly a reaction to this Martin’s post:

So how to introduce PCT so that it would not suggest those negative connotations? You say that the formulation I suggested would cause a “So what?”. In my experience it causes rather “Aah, yes, it is possible”. Any way from that you can continue to explain more about PCT.


I don’t try to interest people in PCT to interest them in the theory; I try to interest them in ways that theory can improve their practice of improving human performance, especially in training. Here’s one example:

Years ago, I ran the Navy’s programmed instruction writers training course. I redesigned it based on the notion that if they could properly evaluate programmed instruction materials they could also create them. The redesign proved my notion was correct. They adjusted their “writing behaviors” so as to produce a satisfactory product.

Similarly, many years later, registrants for a certification test were doing a lousy job of filling out the registration form and many had to be returned to the registrants. Clearly, they did not know how to tell if they had done a good job filling out the form. Two factors were in play. One, they didn’t have adequate reference signals for a correctly filled out form and there was a “disturbance” in the form of a code list that was improperly organized. Did PCT help me resolve that problem? Of course it did but the people I work with aren’t interested in PCT for PCT’s sake; instead, they want to know what it can do for them.


That’s the difference!

I perfectly understand and appreciate your position. That is effective and easier for you and for the people you help. I am helplessly too theoretical for that but fortunately theory happens to be my work. I think that if I manage to make
people to understand their action in a new way then they could help themselves and I don’t need to do it. That is just what helps you to help your customers: that you understand their action from the PCT perspective. Of course also your broad experience, creativity
and other personal properties are needed – much of which I am missing. So I just try to get people think differently than they used to do.

Also I believe that if scientists of many fields of research would understand PCT theory and way of thinking our sciences would proceed better that now.

If PCT meant “Purpose Control Theory” it wouldn’t meet these particular confusions. Maybe in introductions that could be given as an alternate spelling-out of the acronym. There’s even a degree of technical correctness to it: for anything above Intensities, a comparator is controlling the purposes (the reference signals) of comparators below it. It is an easily accepted abbreviation that we use all the time: “I’m controlling being in Chicago” vs. “I’m controlling a perception of being in Chicago” as I decide whether to fly or take the train.

Yes, by means of varying its error output, as ramified by neural branching and diversely amplified by multiplying synaptic connections, that higher-level comparator is actually controlling its perceptual input, which is assembled from the perceptual signals that those lower systems are controlling. Each perceptual signal is experienced as an aspect of the environment. When the word “perception” refers to experience rather than signals, it feels obvious that we control our purposes. That’s where attention and intention goes. When “perception” refers to signals, it is necessary to say that we control our perceptual input. Unpacking PCT as “Purpose Control Theory” speaks to the world of experience, from which we can draw people into the world of perceptual signals and other partly demonstrated theoretical entities in the PCT model.

I like it.


RM: I don’t like it at all. “Purpose Control Theory” makes no sense, mainly because control IS purposeful behavior. If you’re trying to eliminate “control” from the name of the theory, I think the best way to do it is to rename the theory “Perceptual Purpose Theory” (PPT). This would not only be a correct description of the theory (the objects of our purposes are states of perceptual variables) but it would have the added advantage of no longer confusing the theory with the Pacific Coast Trail;-)

FWIW, I prefer Performance Control Theory.

On reflection, I withdraw my approval for Purpose Control Theory on
the grounds that the purpose is the reference, which is not subject
to control. Nor do I like Fred’s suggestion of Performance Control
Theory, this on the grounds that it sounds as though it is based on
observable behaviour in the environment, and hence is talking about
control of output rather than of input. If the name and acronym must
be changed, in my experience, the word that causes problems is
“control”. Perhaps something suggesting that what matters is
perceptual stabilization or maintenance in a changing world might be
substituted for “control”. I don’t have a suggestion at the moment
other than “conditioning”, which I don’t much like.


“Perceptual Control Theory” still sounds good to me. Rather than changing the name it seems that publishing a great deal more research based on PCT is the way to go in order to spread PCT–good additional research, if possible, that will capture the minds of scholars such as those previous correlations of predicted and observed behavior in the order of 0.96 to 0.98 or a bit less.

How about flipping it around: the Control of Perception Theory?