Self concept and collective control

[From Bruce Nevin (20200531.07:45)]

I understand that conventional psychotherapists are much concerned with identifying what they call defense mechanisms, conceptualized as defending the ego–in PCT, to control a systems concept perception of self one defends it from disturbances. MoL doesn’t seek out particular controlled perceptions, but if control of such perceptions generates conflict the MoL process is likely to disclose them.

Even if ‘defenses’ have no prominent place in PCT psychotherapy (MoL), it may be useful to figure out what psychoanalytic clients may be controlling when their analysts perceive defense mechanisms at work.

Control of self-concept has been studied by verbally contradicting what the experimenters imagined (with reason) that the subjects believed to be true (or untrue) about themselves.

Words such as honesty, friendliness, mastery, and courage may be used for such self-attributes, and words such as lying, hostility, incompetence, freeloading, cheating, and cowardice. It is clear to me that these are collectively controlled perceptions. (If that’s controversial, I encourage an interesting and worthwhile discussion.) Consider friendliness, for example. To control a perception that one is friendly requires control of a number of perceptions, including a perception that others perceive you as being friendly. A direct contradiction of this attribute is a disturbance, as demonstrated in the above research.

These are socially important attributes, that is, other’s perceptions of these attributes in your behavior are important inputs for their control of cooperation and conflict involving you, and therefore your control of these self-perceptions is critically important for your social, business, and career success, which in turn is essential for a self-concept as a successful person. The ramifications are endless.

So how does this become a Freudian defense mechanism?

A couple of decades ago, the social psychologist Roy Baumeister and others reframed the conventional psychoanalytic concepts of defense mechanisms in a way that seems to transfer well to PCT.

Here is the relevant passage in the Wikipedia article “Psychological projection”:

Research supports the existence of a false-consensus effect whereby humans have a broad tendency to believe that others are similar to themselves, and thus “project” their personal traits onto others. This applies to good traits as well as bad traits and is not a defense mechanism for denying the existence of the trait within the self.[36]>>Instead, Newman, Duff, and Baumeister (1997) proposed a new model of defensive projection. In this view, repressors try to suppress thoughts of their undesirable traits, and these efforts make those trait categories highly accessible—so that they are then used all the more often when forming impressions of others. The projection is then only a by-product of the real defensive mechanism.[37]>>36. Baumeister, Roy F.; Dale, Karen; Sommer, Kristin L. (1998). “Freudian Defense Mechanisms and Empirical Findings in Modern Social Psychology: Reaction Formation, Projection, Displacement, Undoing, Isolation, Sublimation, and Denial”. Journal of Personality. 66 (6): 1090–92. doi:10.1111/1467-6494.00043>>37. Newman, L. S., Duff, K. J., & Baumeister, R. F. (1997). A new look at defensive projection: Thought suppression, accessibility, and biased person perception. Journal of Personality and Social Psychology, 72(5), 980–1001. https://doi.org/10.1037/0022-3514.72.5.980
I have not read the references cited at 36 and 37, but here is the abstract for the latter:

It has long been assumed that people perceive in others qualities that they wish to deny in themselves, but empirical evidence for defensive projection is limited and controversial. A new model of projection is presented in this article. People might try to actively suppress thoughts about the possibility that they have undesirable personality traits, but it was hypothesized that this response to threat ultimately causes thoughts about the unwanted traits to become chronically accessible. As a result, those trait concepts will be used to interpret others’ behavior. Studies 1–4 showed that those people who both avoid thinking about having threatening personality traits and deny possessing them (repressors) also readily infer those traits from others’ behavior. Studies 5–6 provided experimental support for the model. Unfavorable traits were attributed to participants, who, when they were asked or predisposed to not think about the traits, subsequently projected them onto someone else.

Real-time perceptions of your behavioral outputs as others perceive them (based on their statements or inferred from their participation with you in collective control) can be a disturbance to control of self-concept perceptions.

Suppose a person controls a self-perception of mastery of a complex subject matter–conventional Cognitive™ psychology, for example. Because of its complexity, because it changes in an open-ended way, because its complexity contains contradictions, due to things like emergence of new discoveries and proposals, integration with and intrusions from other systems concepts, exigencies of teaching to newcomers and populations, and inherently due to its incompleteness, such a subject matter can never be completely mastered. Perceptions of how others are or have been esteemed for their mastery or critiqued for their lapses establish in a student’s mind perceptual input functions and reference values for participating in collective control of these perceptions. Perceptions of how others perceive one’s own mastery (perhaps by overt statements, as in the above-mentioned experiments, but there are many ways to perceive the judgements and attitudes of others) may disturb control of one’s self-perception of mastery, and one way to resist such disturbances is to deny them (as above). But just as these are complex perceptions, there are many ways to resist disturbance to collectively controlled perceptions that are important inputs for control of self-concept. In addition to explicit denial and refutation, one may claim that another or others are the source of disturbance to the collectively controlled perceptions.

Most disputes within a field arise from its complexity and open-endedness, through emergence of new discoveries and proposals, integration with and intrusions from other systems concepts, exigencies of teaching to newcomers and populations, and the like. Resolution seems to follow the path observed in MoL, that is, by each of the contestants shifting attention from the specific contradictions between their concepts as expressed to a level above that, a level of perceptions which apparently can only be controlled by means of that specific lower level of control; and then shifting again to a yet higher level to control perceptions of the field as a whole, especially its development and success as a collectively controlled systems concept; and from that level perceiving also the position taken by the opponent, and the perceptions that he or she is controlling by those means.

If these ideas stand up to our collective control of PCT, we may be able to elaborate from them some practical guidance for dealing with scientific conflict in a responsible way.

Discourse proposed that a couple of threads from 1995 are related to this. They would no doubt be differently and more fully expressed today. Such is the growth and development of the field. FWIW, here they are:


Instead of self-concept consider an intentional stance that tells story

Every control loop has an ʽintentional stanceʽ that is established, maintained, and expressed by its structure–its perceptual input, its reference input setting a value for that input, and its means of influencing that input. The reference signal is its intention, and the rest of the loop provides the stance in the environment.

To tell a story is to use language as a means of influencing collectively controlled variables. Language itself is a collectively made system of collectively controlled variables.

Telling a story, and narrative generally, uses the collectively controlled resources of language as means of controlling the perceptions that the story is about. The perceptions that the story is about are collectively controlled, or you would not be telling a story about them–every story has an audience, else it would not be a story. The audience participates in collectively controlling the perceptions that the story is about, else the story would be incomprehensible to them. The audience also participates in collectively controlling the narrative and the language in which the narrative is constructed, else they would not perceive the story at all.

Consider a person who is hearing a story about herself. Upon hearing some particular attribution, she denies that it is true. The attribution might be “you are very brave”, or “you are a chronic liar”. Whatever it is, her denial is evidence that it disturbs a perception of herself. Her denial resists that disturbance. That perception of herself, which we have called a self-concept, is controlled at the (posited) systems concept level, according to the theory. Experimental work has been done by Dick Roberts and David Goldstein, among others.

Language is not a necessary intermediary for disturbing a self-concept, nor is language always the means for resisting such disturbances. Most people that I know are not able completely to represent their self-concept in the form of a story about themselves. It’s much easier to say that some attribute is true or untrue of them. Dick and David’s Q-sort methodology asked subjects to rank each of a list of attributes on a scale, 0-5 or 1-10, something like that.

You said “instead of self-concept, consider …”. This is evidence that the term “self-concept” does not fit well into your prior understanding of psychology. Is that true, and can you identify perceptions that you could not control as you wish if you were to accept this notion of self-concept?

1 Like

I do not mean the intentionality but the intentional stance as described by Daniel Dennett in which an attribute of an object is credited with the cause of a perceptual change. There is no reason for this to require language.

Yes it is true that the term “self-concept” does not fit well into my prior understanding of psychology, as it is the arbitrary “self-control” that Powers describes in chapter “Conflict and Control”. Rather, I believe a more coherent self-concept is the credit assignment problem (for control system reorganisation) that credits perceptual changes to an object’s attributes (think of becoming very averse to a new food if you get sick, or animals learning to separate even arbitrary water and food signs when they are thirsty or hungry).

Ah yes, Dennett. It is the stance of an observer wishing to predict the behavior of a rational agent (curiously referred to as an object, rather than as a subject). Many assumptions are made as to what the subject must be perceiving and what the subject’s goals must be.

Here’s Dennett on p. 17 of The Intentional Stance:

Here is how it works: first you decide to treat the object whose behavior is to be predicted as a rational agent; then you figure out what beliefs that agent ought to have, given its place in the world and its purpose. Then you figure out what desires it ought to have, on the same considerations, and finally you predict that this rational agent will act to further its goals in the light of its beliefs. A little practical reasoning from the chosen set of beliefs and desires will in most instances yield a decision about what the agent ought to do; that is what you predict the agent will do.

But predicting the subject’s behavior is not the main concern in PCT. The observer’s stance in PCT is purposely subordinated to the stance of the subject so as to determine what the subject is perceiving and which of those perceptions the subject is controlling. Nor do we have any commitment that the subject is always rational. But to the extent that Dennett’s apparent appeal to logic is no more than an expectation that observed behavior subserves antecedent goals, this corresponds to the PCT finding that ‘behaviors’ (=actions) are the observable, variable means by which behavior (=what a control loop does) controls perception.

The PCT model bears some parallels to Dennett’s physical stance and design stance, and talk about the theory (PCT) aligns notionally with the intentional stance. Physical stance, design stance, and intentional stance are increasingly abstract. With the intentional stance Dennett is concerned with explanation of mental phenomena at a level of abstraction that requires no knowledge of either physical structure or design. PCT theorizing, even discussion of its epistemological consequences, etc., is firmly grounded in the design of control systems and subject to verification in the substance of physical instantiations.

PCT rather uncomfortably straddles the relationship between perception=neural signal and perception-as-experienced. It is clear experimentally and experientially that what is controlled is a perception in both senses concurrently so that as to their unity there is no doubt, only a lack of explanation. Dennett is all about explanation, but of what? And (to your objection) is explanation possible without language?

Dennett is entirely on the side of perception-as-experienced. Neural signals are way down there at the lowest level of abstraction, the physical stance, and I am not aware of any indications that principles of negative-feedback control enter into his considerations at the next level up, the design stance. And again, none of these are stances of the observed subject, which is the concern of PCT.

You refer here to Chapter 18 of B:CP. He writes there of the conflict that is generated when one tries to control perceptions of behavioral actions, either another’s or one’s own. That form of “self control” is a different matter. The self-concept or self-image is a system concept, at the highest level proposed in HPCT. System concepts are discussed briefly in Chapter 13, pp. 172-174. For some empirical investigation, see

Robertson, R. J., Goldstein, D. M., Marmel, M., & Musgrave, M. (1999). Testing the self as a
control system: Theoretical and methodological issues. International Journal of Human-Computer Studies, 50, 571-580.

1 Like

I have thought about what you said for the last few hours need a bit to put what I’m thinking into words, but by self-concept do you also include things like bodily ownership (i.e. how do I know the hand I’m looking at is mine and not someone else’s or a rubber hand illusion) and the disturbances in dissociative PTSD. For example, why is it easy to jam a fork into your lunch than to jam a fork into your hand. See you soon.

You can’t account for this at Dennett’s intentional stance level of abstraction. Cognition and behavior are inextricable and are inextricably embodied.

I know this hand is ‘mine’ because it is an organ of perception as well as an organ of control output. I know that it is mine precisely to the extent that I know that my intentions are mine, as I use my hand as means of controlling perceptions according to my intentions. I do not need to have a self-concept in order to know that this is my hand and not a rubber simulacrum; rather, this continual inextricability of intention and coincident perception is part of a perception of body which contributes to a self-concept or self-image perception.

You can confuse this proprioception to examine what is going on. Place your hand palm to palm against a partner’s hand. With the fingers of the other hand against the back of your partner’s hand, and the thumb against the back of your hand, rub up and down. Does it feel like one side of your hand is numb?

The first order of defense of corporeal integrity arises in the ‘primitive’ limbic systems at the base of the brain which seem not to be directly available to awareness. To give yourself an injection your higher-level control systems (I need insulin, or I need vitamin B-12) must overcome a ‘natural resistance’ to sticking something sharp through your skin.

Perceptual signals that those systems control are copied to slower cortical functions that are available to awareness. Perceptions of somatic conditions that are changed in the course of limbic control of perceptual input (what we call ‘feelings’) contribute to the input functions that construct the perceptions that we call emotions.

In a healthy process there is time for higher functions to construct their perceptions of the situation and the fight-flight-freeze-fawn alarm state either subsides or the somatic resources marshalled to readiness by the limbic systems are employed to control perceptions constituting a larger strategy.

If attention dwells on the ringing alarm bells, as it were–on those aroused somatic resources–the sensations of body states evoke from memory additional perceptual input to both limbic and cortical systems, apparently forming a positive feedback. Everybody has had the experience of being tongue-tied. Higher cortical functions may be constructing appropriate perceptions but we’re not able to pay attention to those perceptions until afterward, which is why ‘what I should have said’ (or done) becomes obvious in hindsight. It was there at the time, but we couldn’t pay attention because our limbic systems committed our attention to its level of comprehension. Attention is a function of the limited capacity of our resources for perceiving and for influencing what is perceived. The Captain can’t look through the periscope because the gunner’s mate won’t let go of it.

The application to traumatic memory is obvious but it is not the only application. Control without awareness is a large subject.

Here are CSGnet threads from 1997-1998 discussing the Roberts et al. work on ‘testing self as a control system’.





1 Like

I will read that shortly but regarding your previous reply I agree I can’t believe I forgot about the proprioception I’m literally using all the time. Which also reminds me of these 2:

And also makes me think of the hybrid of touch typing and hunt and peck where keys are the goals and the nearest finger is recruited in real time. Or for example that my fingers feel like they are extending from my hand even though the actual muscle is around the forearm.

Attention is a function of the limited capacity of our resources for perceiving and for influencing what is perceived. The Captain can’t look through the periscope because the gunner’s mate won’t let go of it.

Regarding this actually you might find this interesting

He also recently wrote about “phylogenetic refinement” and a hypothesis regarding foraging selection and winner-takes-all in Cambrian era. I will just take the relevant figure and put it here.

Actually after I read the articles you posted I need to think about that hypothesis I put on hold about Histamine-Substance P circuits from Skin to Forebrain and Wakefulness.

I agree with Hommel et al. where they say

We conclude that what would serve scientific progress best would be to drop the term “attention” as a label for a specific functional or neural system and instead focus on behaviorally relevant selection processes and the many systems that implement them.

Attention in my view is an effect due to the directionally and focally limited capacity of perceptual organs and perceptual input functions, and of behavioral means of influencing the perceptions that they construct. If my means of perceiving and controlling are occupied here they cannot simultaneously be occupied there. Then the question becomes why do we control this perception in preference to that?

I will read the paper with interest, but looking at your excerpt from a related paper (the figure is partly identical to a part of Fig. 2 in the linked Hommel et al. paper) it appears that they conceive matters in terms of stimulus-response linear causation. If they were to approach this as control of proximity, controlling to be close to e.g. food and controlling to be distant from e.g. predators, the “winner takes all” notion would be swallowed by the above question, as though by a predator.