Discourse dysfunction

Who said that the power law tells us something important about how behavior works? How did they say it?

A generative model whose output data conform to the power law would tell us something important about ‘how behavior works’. Did someone point out that the data of investigations into the power law are important for this? That would not be the same as a claim that the power law alone tells us something important about how behavior works.

Hi Bruce,

Thanks for explaining how I could have reduced the dysfunction in the power law argument. I asked you to explain it for me because I thought it might clarify what you saw as a non-dysfunctional discussion on Discourse. And indeed it did.

Clearly what you mean by a non-dysfunctional discussion is one where everyone ends up in agreement. And, in particular, they should all agree on what you think is the correct argument. In the power law discussion, you thought those arguing against me were right and I was wrong. So you saw my answers to their arguments the way my opponents saw them, as being non-responsive, evasive, or just plain wrong – that is, dysfunctional. But I see things very differently, as I’ll explain below.

No, I did not assume that it was a side effect; I demonstrated that it was an irrelevant side effect with models that accounted for data from many different movement studies. This work was described in many of the net discussions on CSGNet and Discourse and in Marken & Shaffer (2017) and Marken & Shaffer (2018) yet it was treated by my opponents in the argument as though it didn’t exist. I think this shows that, while I was seeking agreement about the power law based on the presentation of data and the models that accounted for it, there wasn’t much reciprocal agreement-seeking on the part of my opponents.

Which is exactly what I demonstrated in those two papers (Marken & Shaffer, 2017; 2018) that Dennis and I published in Experimental Brain Research. The CVs are the one’s controlled by the models, different CVs for the different models. The behavior of these models was fit to actual movement behavior and the fit of models to data was excellent. And, of course, the model behavior, like the real movement behavior, exhibited the power law as a side effect.

This side effect was irrelevant to the control process that produced the movement in the sense that there was nothing in the models that specified that a power law should be a side effect of the movement. So those opposing my arguments either didn’t understand what Shaffer & I did or they were lying by saying that we hadn’t done it. Either way, not a very good example of them seeking agreement, I think.

No, my modeling and testing against real movement data showed that the observed relationship between speed and curvature is a side effect of control. The mathematical argument demonstrated why that relationship is typically close to a 1/3 power relationship. Again, either through ignorance or mendacity those opposing my arguments did not seem to be seeking agreement.

Yes, they did. Not very agreeable, were they?

Yes, and I demonstrated how I (and others, e.g. Moab et al) had derived that mathematical result. So I guess I wasn’t being very agreeable either. But, in my defense, at least I didn’t say that their arguments were “bullshit” or that nothing but “shit comes out of their mouth” or that they do “sloppy work”. After enough prodding I did finally succumb to the lure of the ad hominem retort by suggesting that the ability to see that the power law is an irrelevant side effect of control is a good test of one’s understanding of PCT, a test that my opponents have failed. I apologized for saying it. So far, no apologies from my opponents for their ad hominem comments, which is fine with me. I don’t need them.

I know. They asserted that I did those things. Indeed, they asserted a lot of things. Not very agreeable, were they? I tried to address their specific assertions about the math by showing how I and others arrived at the rather elementary algebraic solution to the relationship between the equations that power law researchers use to measure velocity and curvature. I don’t know what more I could have done. I think I was being as agreeable as anyone could possibly be.

Actually, what I was asserting was that the power law is an irrelevant side effect of producing curved movement. It’s irrelevant to the control process that produces the movement and, therefore, is irrelevant to the goal of research on the power law, which is to learn how organisms produce movement. So it was actually my opponents in the power law discussion who were being less than agreeable by attributing an argument to me that I wasn’t making.

So my arguments would not have been dysfunctional if the other parties judged them to “directly address their specific statements about the math”? Does this work for all parties or just for the one you like? Shouldn’t my opponents arguments also count as dysfunctional since they never directly addressed my specific question about what was wrong with the math?

I think this “test” makes it pretty clear that what you see as dysfunction in the power law discussion (and probably all other discussions on Discourse involving me) is simply my continuing to disagree with everyone else. You are saying that my arguments were dysfunctional because opponents disagreed with them. If they had agreed with me (which you deem, correctly, to have been unlikely) then I wouldn’t have been making dysfunctional arguments. But, again, doesn’t that count for them too. I disagreed with their arguments so it must be that their arguments were also dysfunctional.

And more evidence that the dysfunction is my failure to agree with my opponents’ arguments. All I had to do to not be dysfunctional was to agree with my opponents.

Of course I would agree with them about that. But in saying that there was a “need for experimental work to identify, and model control resulting in the appearance of the power law” they were implying that I hadn’t already done such work. In fact, I had done a considerable amount of this work. I started way back in 2016 when this whole power law thing started on CSGNet and all of my arguments about the power law have been based on this work, which continues to the present.

My main take-away from your analysis of the power law discussion is that it would not have been as dysfunctional as it was if I had just agreed with my opponents; admitted to all the things that they claim I did wrong and agreed that they were right about everything they claim. And, I presume, that I should also understand that their many unpleasant (to me, anyway) ad hominem remarks were actually my fault, caused by my persistent and annoying unwillingness to agree with them.

So pardon me if I don’t agree with your analysis of my behavior in the power law discussion or accept your recommendations for how to make the discussions less “dysfunctional”. I’m afraid I’ll continue to be “dysfunctional” from your (and all my opponents) point of view.

But there is a good way to stop my dysfunctional behavior, a way which is perfectly consistent with your analysis of my behavior in the power law debate; just stop disagreeing with me. Actually, I wouldn’t want that to happen. I benefit from the disagreement and often learn from it. I just would like it to be done politely.

So my recommended approach to reducing the dysfunctionality of discussions on Discourse is to just hold people accountable when they make impolite, ad hominem comments in their posts.

Best, Rick

For me a discussion is functional if it is (the participants are) searching for truth in spite of necessary disagreements. For someone else it seems to be functional if everybody either agrees with that person’s opinion or leaves the forum. That is very natural if that person deeply believes to own the truth, to have inherited it.

I’m sure the researchers who study the power law are doing it because they think it tells them something important about how movement behavior is produced. They might not say it explicitly in their papers but they have certainly said it implicitly by having produced over 40 years of research and publications on the power law. The publications are in journals having to do with motor control and the neural basis of behavior, which suggests that the researchers thought that their work on the power law had something to do with understanding motor behaviors, such as curved movement.

Yes, it certainly would and it certainly has. Generative control models of movement behavior have shown that the power law is an irrelevant side effect of control.

Here is a pertinent segment copied from Bill Powers’ 1978 Psych Review paper that speaks to this question.

image

The data from investigations of the power law were collected to solve an “old puzzle” – the puzzle of the power law relationship between speed and curvature that is observed when organisms produce curved movements – that the new PCT paradigm shows to no longer need solving. So, from a PCT perspective, the data from investigations of the power law are no more important to the study of movement behavior than the data from investigations of the properties of phlogiston are to the study of combustion.

My aim in the apparently never ending power law debate has been to show that this is the case – that the power law is a problem that no longer needs solving – and to encourage power law researchers to drop their investigations of the power law and redirect their considerable skills to the study of movement behavior as a control phenomenon. Unfortunately (or, possibly, fortunately) this discussion has shown Powers to have been right about what he said in the highlighted portion of the Psych Review quote. Consistent with Bill’s prediction, my explanation of the PCT view of the power law has resulted in a battle between the new paradigm and those who are unwilling to bypass altogether the old power law puzzle that they insist needs solving.

Paradigm shifts, like breaking up, are (as Neil Sedaka told us) hard to do.

Best, Rick

First, a PSA about quote tags. Your embedded quotes, after the first, include no author information.

Thus you begin with

The initial quote tag looks like this:

[quote=“bnhpct, post:63, topic:15960, full:true”]

The attribute ‘full:true’ indicates that you selected the entirety of the post and pressed the Quote widget, then inserted your rejoinders into the entire quotation. The result is that after your rejoinder to the first part, the quotation continues without author information, like this

There are no attributes after the [quote] tag.

This makes it difficult for readers to reconstruct who said what. To prevent this, use the Quote widget to quote each part that you want to reply to just before your reply.

I have gone through and pasted =“bnhpct, post:63, topic:15960” where it was missing, to keep it clear who said what even after someone perhaps quotes parts of this topic in another topic.




It is true that a fundamental aim of any field of science is to reach agreements about the subject-matter of the science. It’s also true that historically these agreements (‘scientific consensus’) are challenged by new findings and then superseded by new agreements. Consequently, it is perfectly possible for a substantive disagreement to persist in a science, sometimes for years, without the parties resorting to fallacious argument forms such as

  • contradiction with little or no supporting evidence
  • criticism of the tone without addressing the substance of the disagreement
  • attack upon the characteristics or authority of the other without addressing the substance of the disagreement
  • name-calling

All of these are ‘dysfunctional’ in scientific discourse. A ‘non-dysfunctional discussion’ is avoiding dysfunctional (logically invalid) argument forms, including but not limited to this list from Graham’s hierarchy.

For all but the last there are limited circumstances in which they may actually be functional. For example:

  • Supporting evidence may be included by reference to prior discussion, and sometimes that reference is not explicit. It may happen especially among people familiar with the terms of argument, and the missing ‘taken for granted’ references can always be made explicit. This can easily degrade into a gray area.
  • Criticism of the tone of discourse is functional for those who feel unsafe asking a question or making a proposal, and therefore (for just one example) functional for all who want to invite newcomers.
  • Questioning whether someone has got familiar with the basic terminology, methodology, and literature of the field is perfectly appropriate. However, to be functional this should not be an attack, it should direct them in a friendly and supportive way to what they need to learn. This can be tricky with folks who are already invested in being experts in what they perceive as the same or intersecting field.

It’s not a personal matter of what I think. There is universal agreement that some forms of argument are logically invalid, and that they should be excluded from scientific discourse. Fallacious argument forms assuredly do occur in scientific discourse, scientists are human, but nobody claims they’re proper.

See ambiguity of ‘irrelevant’, below.

I don’t think there was any disagreement about it being a side effect of control and not a cause of behavior. For example, Martin said “I finished my critique with the statement that perhaps the power law is indeed a behavioural illusion, though M&S sheds no light on that issue”.

Yes, they said there was a fundamental error in your mathematical argument and you said there was not. (More on that below.)

Attributing this to flaws in their education or qualifications (ignorance) or attributing it to a character defect (mendacity) is an ad hominem argument.

Mathematics is not subject to a lot of ambiguity, and conflicts in math are not resolved by exerting greater force until one or both reaches maximum output capacity.

I freely submit that I am not qualified to weigh in on the mathematical argument. The rustic but proud school system that I attended in central Florida offered no pre-calculus and but limited algebra, and it is not at all to my credit that I failed to remedy those deficiencies when I got into college. I have learned some since, but I claim no strength there.

Equivocation is one of the logical fallacies that Graham does not mention.

Equivocation depends upon or exploits ambiguity. “Disagreeable” (“not agreeable”) has two meanings. Disagreeing about the substance of the argument need not be ‘disagreeable’ in the sense of dysfunctional discourse. As I said above (and I believe that you and I have previously both remarked, maybe even agreed), scientists disagree fairly often about how to interpret findings, etc. Whether they do so ‘disagreeably’ or collegially is a different matter.

Here’s one place where it seems to me that you could have done more:

Martin:

since the formula for D was velocity (V)
times a constant in spatial variables, the equation is not an
equation from which one can determine V. The M&S claim that it
is an equation from which one can determine V is the core of my
critique.

Rick:

We said that what you said about it not being an equation that can be used to >predict V using linear regression is wrong. Which it is.

This sounds like “You’re wrong because we said so.” Note that I’m not taking sides as to who is correct. I’m pointing out that you contradicted without evidence. The very least “more” that you could have done could have been a reference to where you “said so”, i.e. the location of the specific refutation with evidence.

For other readers, a protracted portion of the exchange can be found in the CSGnet archive for 2018 by searching for “Bogus mathematics, (was Re: L’état de PCT, c’e st moi (was …))”, and selecting More… at the bottom of the initial list.

Here again I was trying to disclose the form of the argument. I identified this as a non-sequitur, changing the subject from the mathematical argument to the status of the power law as a side effect. I think the response was kinda like “Who cares, that’s probably true, but it’s not what we’re talking about. The math is wrong.”

But you emphasize that it’s not only a side effect, it’s irrelevant.

I think we’re running into equivocation over the word ‘irrelevant’. It is ambiguous as to its reference and scope.

  1. For the control system, the power law is irrelevant. I think everyone agrees that control systems do not carry out power-law calculations, and that the observed speed-curvature relation (which is not all that precise) is a side effect of control.
  2. For researchers building and testing generative models of behavior when organisms create or follow (trace) curved paths, behavioral data from research on the power law is deeply relevant.

Saying it’s irrelevant in sense (2) dismisses and trivializes any lab work on these lines that they are doing.

You asked for examples of your usage, that’s why yours are more in focus. You’ve called attention to their name-calling, so they’re not left out. They haven’t invited me to identify less severe fallacies in their argument. You’ve said that their mathematics is “bogus”, which looks like a claim that it is incorrect, but then you retracted that to say you just meant that it didn’t apply in the context of experiment. You’re welcome to identify other logical fallacies in their argument.

I said “I could be wrong”. What did I say I could be wrong about? I said I could be wrong about my entire attempt to summarize the form of the argument. I could be mistaken in various ways, including the ways that I listed. If any of those things is true, then the corresponding part of my attempt to summarize would be wrong.

For you to paraphrase this as you have done is a non sequitur. The non sequitur was created by taking a quotation out of context and changing its meaning. The full quotation means that if any of the items in the list is true, then my attempted summary is wrong in that respect. You took an excerpt from it to mean that these were tests whether or not your arguments were dysfunctional (i.e. fallacies).

No, it would just mean that I was mistaken about that.

This is not a test of who is at fault. It is a ‘test’ of the accuracy of my summary. If true, then I was mistaken in my perception that they still disagree with you.

Again, this only means that I was mistaken to think that there had been disagreement about this.

That’s a good start, but it’s even more necessary to identify the less extreme fallacies and hold people accountable for them before annoyance at them escalates to the point of name-calling (the extreme form of ad hominem argument).

Who inherited truth? How does one inherit truth?

Best, Rick

Got it. Will do.

Of course.

I was referring to your agreement with the substance, not the form, of my opponent’s arguments, though I assume you agree with both.

Actually, the rest of your post is a very detailed attempt to say that it’s the form and not the substance of my arguments that you are criticizing. I think it actually shows that you strongly agree with the substance of my opponents’ arguments and that what you consider to be dysfunctional about my arguments was that they didn’t answer the substance of their arguments to your satisfaction.

So I will stop my dysfunctional communication about the power law by arguing about it no more on Discourse. My last word (I hope;-) on the power law topic here on Discourse will by my previous post. If power law researchers want to keep chasing this Boojam there is clearly nothing I can do to convince them to stop.

Eetu, this looks to me like an ad hominem argument. You’re saying that for some unnamed person(s) others must either agree with them or leave, and that they believe they own the truth by inheritance. That would be a character flaw, wouldn’t it?

If in place of “someone else” and “that person” we understand that you mean “Rick Marken” it’s very close to outright name-calling. He could perceive it as a personal insult.

It sounds like an allusion to an opinion that I have heard from several people over the years, Rick, the opinion that you regard yourself as the heir apparent to Bill Powers’ achievements. Eetu, is that what you’re alluding to with the word ‘inherit’?

The image that you use as your self-portrait or avatar with your posts, Rick, with Bill seated closer to the camera, can be perceived as consistent with that. This may never have occurred to you. I’m sure that for you this picture evokes poignant memories of Bill and the great importance of your relationship with him and Mary professionally and personally. Using it to represent the writer insinuates that the writer and Bill speak with one voice, or even that the writer is the posthumous voice of Bill Powers. This appearance, presumably unintentional, is liable to have additional unintended side effects.

I have no opinion for or against either. I lack competence or standing.

That is correct.

I can see why it appears that way to you. I tried to talk only about invalid forms of argument, but it’s not possible to quote illustrative examples without dragging along references to the substance. I talked about your usage of invalid argument forms and not theirs because you asked me to and they did not. As a consequence, in the examples that I quoted you were not making a good case for your point of view because at those moments you were using one of the invalid forms of argument (as e.g. contradiction while assuming they had accepted as evidence things said elsewhere, rather than presenting, summarizing, or at least referencing evidence). This lends itself to the appearance that I am agreeing with their point of view and disagreeing with yours.

For me to do that would be off topic in this subcategory about discourse dysfunction, and that’s pretty close to non sequitur.

If you feel a need to even things up, you could locate places where Adam or Erling or Bruce Abbott or Martin or others slipped off the rails into one of the less onerous fallacies (aside from name-calling, which is an obvious terminal symptom). With the caveat that they have not invited that as you invited me.

Yes, it sounded like that to me too. I was going to deny that I think of myself that way but, truth be told, I kindda do. And I think Bill kindda did too. He did write this nice little inscription in my copy of Making Sense of Behavior:

But I don’t like the implication that I inherited my heirship the same way I inherited my good looks. I think Bill considered me to be his heir because I have produced the largest body of what he considered good work on PCT, other than that produced by himself, of course.

But while I think Bill might have considered me to be his heir, I don’t think that that gives me any particular claim to being right in arguments about things PCT. Indeed, I’d rather not be thought of as being Bill’s heir because I think it has caused me a lot of problems in the discussions about PCT that have occurred after Bill passed away. It seems to me that, in an effort to delegitimatize my presumed heirship, people have been particularly vicious in their arguments against me. So it would be nice if everyone stopped thinking that I’m thinking that I’m Bill’s heir and just think of me as someone who happens to know a lot more about PCT than they do;-)

Best, Rick

Well, this is how Bill inscribed both LCS and LCS II for me.

I guess that pretty well inoculates me against any imputation of inherited authority. Another thing to be grateful for.

You might consider a different profile photo for your account.

BN: Eetu, looks to me like an ad hominem argument.

It was not so much any argument than an expression of frustration and I am sorry that I let it out as such.

BN: You’re saying that for some unnamed person(s) others must either agree with them or leave, and that they believe they own the truth by inheritance. That would be a character flaw, wouldn’t it?

No, believing something is not a character. All the time we are discussing our beliefs.

[Sorry, again discourse cut the end of my email message away. Here is the missing end – hopefully.]

BN: If in place of “someone else” and “that person” we understand that you mean “Rick Marken” it’s very close to outright name-calling. He could perceive it as a personal insult.

That is up to you to understand it how you want. At the moment I have no time to go to archives to search evidence, so this time I did not want to point any named person. But again in this kind of forum it should not be a personal insult to say about some discussers that they seem to act like the owners and defenders of the pre-existing truth.

I really think that that the most dangerous dysfunction – a mother of all dysfunctions – in scientific discussion forums is that if some (or all) act so as if they already know the truth and their duty is to keep all the suspects, heresies and alternatives out of the forum. From this can then follow flawed arguments and other dysfunctions.

Eetu

Off topic but anyway.

RM: Side effects are behavioral illusions when, as in the case of the power law, they are thought to tell you something important about how behavior works.

“Important” here is a subjective statement. Behavior works in an environment and always causes side effects. It is interesting and often important to study why certain side effects follow in certain cases – especially if there are somehow surprising regularities (or irregularities) in these side effects. An interesting research topic is why curved movements often cause regular power law phenomena – that requires explanation. You have been saying that it does not require explanation because a) the phenomenon is not really existing; or b) there is a mathematical explanation; or perhaps for some other reason. But measurements show very convincingly that the phenomenon is real, surprisingly regular, but not at all universal. Because it is not universal, no mathematical explanation can do. Your mathematical explanation is tautological saying that if there is 1/3 power law then there is 1/3 power law. But why there is sometimes other power law and sometimes none?

If something is a side effect it can still be important, interesting and relevant.

I agree with you, Eetu, that this post is off topic here.

The power law discussion is scattered in many places over many years. To consolidate these in one place (at least by reference links) I have created a new topic in the Research/Motor Control category, called Discussion of the speed-curvature Power Law.

Please move the text of this post to the new power law topic. I think the easiest way for you to do that is

  1. Select and copy the off-topic text that you posted.
  2. Open the new topic Discussion of the speed-curvature Power Law and click the Reply button
  3. In your new post, paste the text which you copied from your post.

If you wish, you can precede it with an intro in brackets, something like

Well, the suggested text got munged, but it’s really unnecessary. You might want to delete the post here, however, after setting it there.

I hope this means that you will now stop acting as if you already know the truth about the power law and considering it your duty to keep heresies, such as the idea that it is an irrelevant side effect of control, out of this forum.:wink:

You asked for examples in your practice, Rick. You’re the only one of us who has extended permission for this. Let me add myself. If I write something that is disruptive to our good collegial relations I would like to know about it.

Here’s an example, Rick, from our recent discussion of the “About” post for the Phenomena category.

Yes, taken out of context as you have done it is possible to read the first sentence that way. But read together with the immediately following sentence that interpretation is not possible:

To avoid this myopic piecemeal interpretation, I reversed the first two sentences and added “so defined” to make the link between them unignorable:

PCT explains purposeful behavior as control: acting to keep variable aspects of the environment in states which the subject organism prefers. This category is for discussion of the observable phenomena of behavior, so defined.

However, instead of this, in the current state of that discussion I have invited you to substitute your own preferred “About” text, so these changes do not appear there.

Perhaps you can now see that this was not left out.

This ‘shoot from the hip’ style recurs frequently in your CSGnet responses, and persists to some extent even now. My guess is that you are controlling correctness of talk about PCT with gain so high that any disturbance requires an immediate reply to correct it, without any time allowed to perceive the larger context or to consider how matters might be perceived from the other person’s point of view.

The effect of this myopic hair-trigger style on others is not pleasant and does not enhance how you are perceived (reputation, credibility, esteem, etc.).

Actually, my interpretation is even more possible in this case because you don’t say that behavior is control, you say PCT explains behavior as control. Big difference.

No, I still can’t see it. Maybe its my PCT glasses;-)

I want people to know that there is a phenomenon to be explained; the phenomenon is control. And that phenomenon should be described. Your description of control is a mix of explanation and description: “acting to keep variable aspects of the environment in states which the subject organism prefers”.

As description, it should be added these states are aspects of the environment are protected from disturbances by the organism’s actions. It’s explanation when you say that these states are what the organism prefers. To the extent that control has been noticed by conventional psychologists, it has been explained as a caused phenomenon, where external events cause variables (like reinforcements) to be in the state that appears to be the one you prefer (consumed).

I think my suggested change to the description of the Phenomena category does a much better job of making it clear that control is an observable phenommenon and of describing what it is that one can observe:

“Control is an objective phenomenon that is seen when variable aspects of the world – controlled variables or CVs – are maintained in fixed or variable reference states , protected from the effects of disturbances.”

I’ll go with Bill’s evaluation of my ‘shoot from the hip’ style. The quote below is from one of Bill’s CSGNet posts that Dag was nice enough to copy to this thread in his efforts to show that I’m a disruptive character:

[From Bill Powers (2005.08.19.0822 MDT)]

I wish that people would examine what preceded the blowups, which was almost always someone trying to push some idea that was really not consistent with PCT – and which Rick , lacking my tact (or the tact that I once thought I had), immediately objected to, quite correctly if not always gently. In reply, Rick was attacked for being a thought policeman, for being closed-minded, for being arrogant in insisting on “PCT purity.” And of course that infuriated him, and Rick infuriated is not the Rick of Teaching Dogma in Psychology. When he is angry Rick pours gasoline on the flames. Or he used to. But I have yet to see a case where he lit the match.[emphasis mine - RM]

I am not in this to be liked; I’m in it to teach PCT. But don’t worry, I’m liked just fine by the people out there – most of whom are apparently not on Discourse – who understand PCT as well (or, at least, in the same way) as I do.

Then that would have been an appropriate critique to have made of the two sentences together. The point of this post is the piecemeal putdown of one sentence at a time, and too often only responding to the bits that you can put down. This has the appearance, as I said, of controlling “making a mistake about PCT” with high gain. High gain results in quick response to slight error, leaving not enough time for higher cognitive functions to consider context, etc.