Sunday, December 31, 2006

At the Bar with B.F. Skinner

(Note: This is an first post in a series that I plan to do over the next year and a half. This series will introduce and talk about well known psychologists/psychiatrists/neurologists, who have now passed on. The goal of this is to show the basic premises and ideas that these persons advocated. The format will be an informal conversation at a bar). While creative modifications were inevitable, I think this captures an accurate picture of the ideas and defenses used by Dr. Skinner.

Interverbal: Dr. Skinner, thanks for coming to get a drink with me

Skinner: Oh, you are welcome. What did you want to talk about?

Interverbal: So, much mythos seems to surround you. Some people hold you in the profoundest sort of respect as a great humanitarian and some accuse you of being a stepping stone to full dictatorship, the door-keeper to the loss of all the makes us human. So……maybe you can explain yourself and your position more fully….

Skinner: Well…for myself I want to see human culture develop so that we can accomplish two goals, 1, to survive, 2, to have all of us be happy. You see, what we do is a reflection of what we learn...

Interverbal: [Interrupting] But not everything we do is learned…..What about reflexes, what about innate behavior, what about behavior related to problems in the neurology, like seizures?

Skinner: There is behavior that is not related to classical or operant conditioning, but much of it is and it is that portion that I choose to look at. I will say this though; no good student would ever say that behavior is purely conditioning or that we are a tabula rasa (scraped tablet/blank slate).

Interverbal: So, then John Locke is out of the picture.

Skinner: In that regard, yes, Locke is out of the picture.

Interverbal: So, would you say that the fields of neurology and genetics play an important role in teaching us about our behavior.

Skinner: To an extent, yes.

Interverbal: And cognitive neuroscience?

Skinner [hesitates] Not necessarily, that depends on their use of mentalism and circular explanations for phenomena.

Interverbal: So, if they invoke thought process in their answer they are wrong?

Skinner: No, a mentalism is not the invocation of a process we can’t see. What I mean by “mentalism” is the circular explanation of phenomena. Why is the boy crying? Because, he is hungry. How do we know he is hungry, because he is crying.

Interverbal: So, you believe in parts of our mind?

Skinner: I think this type of labeling is tip-toeing into a mentalistic fallacy. However, yes, I do think there is thought and feelings and that these things are important in human behavior. I will go so far as to say that if we are to tell the whole story we must include those things in our analyses.

Interverbal; What about cognitive psychology?

Skinner: I view cognitive psychology as mentalistic and archaic. I see it as for psychology, what creationism is for evolution.

Interverbal: That seems very strong, many people must disagree, even behavior analysts.

Skinner: Correct….

Skinner: Before we go much further, I would like to switch roles and ask you a few questions.

Interverbal: Sure…

Skinner: Why are you here tonight?

Interverbal: Because, I wanted to talk to you.

Skinner: And how do you know you wanted to talk to me?

Interverbal: [laughs] you are trying to get me…

Skinner: [smiles] To be honest I have no problem with you describing your own feelings as “wants”. It is simple and efficient, just like we say that “sun rises”, or the “Stars come out”.

However this sort of simplistic language has no place in a genuine discussion of a science of human behavior. This is the precise problem we face in psychology; it is analogous to some astronomer telling us to really believe the sun orbits around the earth.

Although to be fair this really was true at one time. But here is the difference; biology and astronomy and medicine have all moved on beyond the earliest theoreticians; not so in psychology. Aristotle could stroll into a professional presentation of many types of psychology here and now and probably keep up with the conversation.

Interverbal: [smirks] Probably, he might even do better than some.

Skinner: You laugh, but that is true. Psychology is rutted down in simplistic descriptions that have little to do with science and a great deal to do with language.

Skinner: Sir, you are sounding all Wittgenstein-ish now.

Skinner: And for good reason, the logical positivism makes some good points, it actually was necessary for behaviorism to arise.

Interverbal: A lot gets made of the fact that you are a determinist. Our genes in combination with our environment determines how we will behave.

Skinner: That is a concise summery.

Interverbal: That seems so simple to disprove. I mean…I can get up on the bar right now and sing a song about how I have free will. That IS free will…

Skinner: [smiles] Unless of course your behavior was controlled by the reinforcing feeling of supposing you proved me wrong….

Interverbal: I could do it much later too, maybe just by spontaneously remembering.

Skinner: What would precede the behavior?

Interverbal: Well, my thought of proving you wrong…

Skinner: Exactly…

Interverbal: Okay, but people do off the wall things all the time. I mean let’s just start with kids. How many times does a kid do some ludicrous thing and when we ask them why they did it, what do they always say….”I don’t know”.

Skinner: I think you are confusing the fact that you don’t always know why a child does something with the idea that they must have done so because of free will. But I would argue that you probably haven’t even really looked for a reason why.

The more I look, the more it seems that people behavior is logical, even if they themselves can not determine this system. Research backs this up by the way.

Interverbal: So, why don’t people behave the same…. or at least the people within the same culture?

Skinner: Careful, you are bouncing into illogic. It would be a mistake to assume that culture alone tells you what you need to know to determine behavior. In reality culture almost never tells us enough to determine why a person does a given behavior. That is much more specific.

Interverbal: So you discount the work of social scientists.

Skinner: No, but I often wonder if their work searches too broadly and misses critical details.

Interverbal: Well what about a class then. In a classroom, the same rules apply for everyone and produce the same outcomes.

Skinner: Actually that is simply not always true. Also, maybe it shouldn’t be true, but that is another story. Anyway the classroom is not a child’s only environment, maybe not even the most important one.

Also, I think you are gunning for “all or nothing” comparisons. You see the most extreme options, but miss the subtle middle ones. You simply can’t ignore these.

Remember a child does not walk into a class as a bundle of unwritten and untapped potential. They arrive with several years of previous learning and possibly a different set of established reinforcers and preferences.

Skinner: Anyway I have another question fro you. What kind of beer are you drinking?

Interverbal: Duffs..

Skinner: And why do you drink Duff’s?

Interverbal: Well let see…. That would be because it’s a rich hops mix with excellent texture and smooth taste.

Skinner: You misunderstand me a bit. What I mean was why do you drink Duff’s as opposed to McCallagans?

Interverbal: Because Duff’s tastes better.

Skinner: Yes, and that is a perfectly acceptable answer all by itself. Answer me this though; are you more likely to order Duff’s or McCallagan’s.

Interverbal: Duff’s, clearly.

Skinner: We are talking about what you as one individual value. You would prefer Duff’s when given a choice. Humans are like that, we value different things, possibly because our genes determined that we would, possibly because it was paired with another reinforcer at some earlier time.

This fact permeates us and it deeply relates to our behavior. There is individuality in this deterministic system. We can never be totally sure even after lots of successful reinforcement that a person will do a given behavior at the “right time”. We can not speak of probability in out science only of likelihood or what we expect to happen.

Interverbal: This ties in well with certain mathematical theories I think.

Interverbal: Dr. Skinner thank you very much for having a drink with me. I deeply appreciate it.

Skinner: You’re welcome

Interverbal: And now, [smirks] let us choose of our own free will to have another round…

Skinner: I will drink to that.

Thursday, December 21, 2006

Tribute to Carl Sagan

The 50th Skeptic's Circle is up and involves a loving and fitting tribute to one of my
childhood heroes.

Wednesday, December 20, 2006

A Review of Informed Consent #1: Philosophical Overview

Ethics is a major branch of philosophy. It concerns rights, behavior, treatment of others, and dichotomies of good and evil.

Ethics are established in the field of human services by stating a rationale for should be done and reconciling the rationale to ethical guideline and positions. The implication of this is that more than one ethical position is possible and that the depth of the ethics of an action is related to the excellence of reconciliation to ethical guidelines.

In research we rarely deal with ethical vs. unethical. Usually it is more subtle. We must consider and attempt to match ideals or guidelines to the behavior of ourselves and others. Using this method to understand ethics, the best way to be unethical, is to fail to provide ethical justification for a treatment.

It is my observation that creating ethical guidelines calls into use another field of philosophy, being “natural law”. Natural laws, are the inherent rules of our existence, inalienable and independent of political law. The following could be said to be natural laws:

1. Children are cared for until ready to live on there own.

2. People have a duty for the welfare of one another

It should be noted that natural laws often are almost always legislated, examples of legal documents heavily based on natural law would be the Magna Carta and the US Constitution.

Interpretations of natural law, however, may lead to differences in views of ethics, different ethics may lead to different actions. I would strongly argue in favor of clearly understanding this connection when attempting to debunk the ethics of an action.

Thursday, December 07, 2006

Pub Skep

Get thee hence

Wednesday, December 06, 2006

A Review of the Parent Designed Report Form

The Parent Designed Report Form is an assessment method designed to record the potential change in autistic child undergoing therapy and specifically methyl-B12 injections. This tool, while embracing some of the qualities of objective assessments does not embrace them all. The threats to the validity and reliability of this method are numerous and have the potential to mislead the decision makers in a child’s life. Methods of alleviating these problems are discussed.


The Parent Designed Report Form is an assessment tool created by parents, for parents to measure the progress of their autistic children under specific treatments. It is advertised as the most specific and sensitive measure of change via methyl-B12, by its advocates. It’s creation and use seems closely, but not exclusively tied to the treatment protocol of methyl-B12 injections for autistic children.

The purpose of this paper is to 1) give a loose overview of the Parent Designed Report Form (PDRF). 2) Point out the problems present in within the PDRF and the justification of this tool. 3) Suggest specific changes that might help validate this tool.

Overview of PDRF

The PDRF is an assessment tool. It is periodically filled out by the parents who are often using the vitamin B12 injection treatment for their children’s autism. The tool is a retrospective Likert based scale assessment. This means that the child is not necessarily being directly observed by the parents at the time the scale is being filled out. The scales are to be filled out based on parent impression of child behavior within the recent period (6 weeks). The question include a second portion where parents are asked for specific examples of the change.

The Likert scale is based on 1-5, 1 being “Mild”, 5 being “Significant”. The questions typically refer to a “behavior” and a “change”.

For example:

# 125 “Tasks: New things never done before or never accepted before: Give the before and after examples necessary to explain you observation”.

#76 “Laughter, more normal amounts and ore normal types of laughter: Why are you saying this? Support your observation as to what is was before and now.”

An additional part of this assessment tool involves the Parents Specifics Documentation Letter. In this portion the parents are invited to write down in detail the changes they have seen in their child in what can accurately be called an emotional format.

There seems to be no evidence in support of the claim that the PDRF is the most subtle measure of a child’s progress. This claim is offered by the creators or advocates of this measure. It is seemingly based on the types and number of questions asked. Moreover there are no data that shows any sort of inter-observer-agreement on the test. We have no idea if the test is reliable between, two caregivers who were with the child over the same period. Further, we have no idea what the validity of this test might be. It has not been compared meaningfully to other established scales or to quantitative behavior assessments.

I report that some advocates of this measure claim that an affect is noticed following treatment via this measure, but not on other measures and that this constitutes a proof that this measure can detect smaller changes than the larger and better established tools. However, that is not necessarily the case, these could be false positives; I note again, that there has never been an attempt to validate this measure, the validity is assumed.

One of the great advantages claimed by proponents of this tool is that objective diagnosticians, using a reliable/valid observation based tool, only watch for a limited amount of time. They might miss the rare but increasing utterances of child first beginning to use words. The parents, they argue, who are always with the child, can observe this, remember it, and regurgitate it on the form. This seems acceptable, accept that parents like all people, are prone to the fallacy of confirmation bias, or only identifying those data, which confirm their current position possibly overlooking those data which could refute it, such as the child already increasing her utterances before the treatment began. I am reminded in this situation, of the claims of ESP advocates; it seems the really great evidence is never present when the objective skeptic is looking (Sagan, 1996).

The wording within the questions also seems to invite subjective opinion rather than objective observations on numerous occasions. For example:

#84 “Mood swings, fewer: Describe.”

#47 “Flexible, less rigid, can be redirected: Describe why you feel this to be a true statement. Be specific and give as many examples as you can think of”

#141 “Improvement in the quality of life of others e.g. therapists, etc. (indicate a title or position of the person)”

#82: Loving, cuddly, more so now than prior to the shots even though your child may always have been loving.

I identify particular problems with the Parents Specifics Documentation Letter.
It would not be an error to say that the description of this letter is a remarkable invitation to anti-science.

“Please note that a piece of paper with your “facts”, no matter how well written, will not convey some of the most important information for us to determine how well methyl b-12 is working for your child- that information being the “emotion of the moment”, that you felt when something your child did for the first time or at a special moment.”

“The specifics Documentation Letter is so much more powerful, and so much more believable, when you let us into your hearts and feelings by describing those items that made an impact or created a special memory.”

“We only want you to include comments on the responses that “blew you away”, had you utterly amazed!”

“it is this that MAKES MUSIC AND GIVES BEAUTY to being alive and working together with your child and you”.

The above being an excellent description of a post hoc, ergo propter hoc fallacy and a non-sequitur, the feeling of the parents, has absolutely no ability to establish the accuracy of a change caused by methyl B12.

Next, I wish to take issue with the claim via advocates of this measure, that researchers must find a way to validate parental experiences that show improvement via certain treatments in autism. This assertions seems to be assuming what it should first be proving, that parental observations in such situations are in fact “valid”. We know from research that parent’s description of child behavior changes based on what they are told, irregardless of the reality of the situation. This was demonstrated by Hoover & Millch (1994) who told parents to rate their child’s behavior as much worse following the children being given a sweet drink that the parents were told contained sugar (It contained an artificial sweetener).

My final problem with this assessment is that, it the comparison is not to a normed distribution, but to the child’s previous assessment. These assessments are typically done every six weeks for up to two years. The problems presented here are the concept of regression to the mean and maturation for eh child. Pre-school aged can and do mature rapidly (if unevenly at times). Moreover several of the questions in the form seem to focus on behaviors very closely tied to a developmental schedule as opposed to learning.
For example:

#92 (Concerning the child having opinions)

#81 (Listening better)

#78 (Use of language)

#110 (self confidence is greater)

#111 (Better structuring of sentences)

#112 (Less silliness)

#114 (New level of singing)

#100 (Reading at higher levels)

#115 (Improved sitting behavior)

#124 (Can multi-task)

#25 (Less clumsy)

#26 (More intelligent)

#27 (Improved compliance)

#45 (Able to better express feelings)

#1 (Abstract thinking)

#23 (Understands cause and effect)

There is not a single one of the above that is unlikely to improve, perhaps even significantly on a pre-school aged child, and especially a younger pre-school aged child over the course of 12 weeks. As matters stand the PDRF seems to be focusing on developmental tasks that are very likely to increase on just about any child.


I propose that in attempt to gain an understanding of the reliability of this measure that the following be investigated by those who took the time and effort to create this measure.

1) Typically developing children be assessed and their average scores compared to autistic children both in and out of the methyl-B12 program.

2) That objective data counts by observers (observing for a prolonged time) be compared to parent’s outcomes for this test.

3) That an attempt be made to assess this tool against other tools whose reliability validity is already established

4) That inter-observer reliability of this measure be assessed by comparing scores between caregivers of the same child.


Deprey LJ, et al "Double-Blind, Placebo-Controlled, Cross-Over Trial of Subcutaneous Methylcobalamin in Autism: Preliminary Results" AACAP 2006; 33:F47.

Hoover, D. W., Milich, R. (1994). Effects of sugar ingestion expectancies on mother-child interactions. Journal of Abnormal Psychology. 22(4), 501-15.