Sometime in the past few months I came across this article:

It bears on this article from the previous year:

Both of which have relevance to this article, perhaps in an oblique way (but there is also an ulterior motive here):

Behaviorists passing in the night

Long ago (circa late 1970s, early 1980s) and not so far away (here in Morgantown, WV), I knew an interbehaviorist whose name shall remain unspoken (yep, I’m typing, not speaking), but let’s refer to that person as Linda. We had many loud and sometimes inebriated arguments related to interbehaviorism.

Morris alludes to the unpopularity of interbehaviorism with the radical behavioral crowd as the result of what radical behaviorists considered a too philosophical and conceptual approach. That was me. My canned response to much of what Linda had to say was, “How would you design a pigeon experiment to test that philosophy?”

Years later I got a note from Linda saying that she was beginning to do research with pigeons, and that she had a better appreciation of my question.

Now, many years after that, I say to Linda: after reading Morris’ article and reviewing issues I have had in thinking about behavior analysis, I have a better appreciation of your position. Furthermore, I wish I had understood interbehaviorism better back then.

Media of stimulation

See the section titled “Interbehavioral Fields and Three-Term Contingencies” in Morris (1984, pp 199-201).

Of particular interest to me is the notion of media of stimulation (p 200):

  • “The media of stimulation refer to the sensory means by which contact is made between the organism and the stimulus environment. The media are not properties of the organism or of the stimulus, but are physical conditions (e.g., light) that permit contact between the two.”

Radical behaviorism and interbehaviorism seem to agree on “the stimulus”. From my perspective, there is overlap between “the stimulus” and “the media”. Does it matter? For all intents and purposes, can we treat them as separate?

What matters to me is that stimulation is transmissive following the laws of physics (as known to this lay person).

Case #1: What we call “direct observation” is not direct. What is usually meant is visual observation, in which the medium is radio waves within the limited frequency band which humans refer to as “light”. What we observe with our organic light-to-neural-signal transducers is the result of reflection of those radio frequencies (“colors”), minus whatever frequencies (“colors”) were absorbed by the thing being “observed”.

Case #2: But there can be any number of transductions and re-transmissions. For example, live TV “direct observation” involves the above radio wave transmissions from the thing observed, to the transducer/encoder equipment we call a “camera”, then transmission of that signal via different media to a decoder/transducer we call a “video display”, and finally transmission of a somewhat representative sampling of the “light” radio wave transmissions from the video display to the organic transducers we call “eyes”.

Now assume that the distance between two observers and the observee are identical, and the time between some stimulus (motion) of the observee and the arrival of the radio wave “light” signal to the “eyes” of the two observers is identical. The only difference is that in Case #1, there is nothing but air between the observer and the observee, while in Case #2 there are electronic transducers, transmitters, and receivers.

Are the two interactions functionally identical?

Now add in two way sound. Assume identical responses on the part of the two observers. If you would consider the interaction between the observer and the observee in Case #1 to be social behavior and verbal behavior, do you consider it to be so in Case #2?

Now add in speech to text on one end, and text to speech on the other. All participants are responding as one might expect in a “conversation”. Textual behavior is both social behavior and verbal behavior, right?

Is there some point at which one should no longer consider the interaction to be a social interaction? Say, the text is put into a text message and there is a 5 second delay in delivery. Or a 60 second delay. Or in a mail message and there is a 24 hour delay?

Social behavior, media, transductions, delays

This brings us to Hake, et al., 1983.

Starting in late 1977 when I was an undergraduate in Don Hake’s social behavior class, we had a baffling (to me) disagreement over the above issue. It was baffling to me because I was pretty sure that in one of Don’s examples of social behavior he was suggesting the above continuum of interaction. The example involved using a porch light to signal to a friend that they would be welcome to come in and have a glass of wine. If the porch light was on, welcome! Reinforcers are available. If the porch light is off, well, I’ve gone to bed and knocking on the door would at best be greeted with extinction.

Seems pretty straight forward: media of stimulation. And of course, it was immediately obvious what the extrapolations would be: various media including audio (phone, tape recordings), video (live or recorded), immediate vs delayed, and so on. But no, that is not what he intended. Though he finally came around in the summer of 1982, I never understood the reasoning behind his disagreement. I still do not. Very rude of him to die.

In any case, the “General Discussion” section (pp 20-22) is about various stimulus media, various sources of the physical phenomena that are treated as stimuli, and how difficult it is to tease out various sources of control. It is especially a problem in researching social behavior. It is very difficult to get the kind of control one might want over the behaviors that are to serve as stimuli. Thus it becomes difficult to say which resulting behaviors are social (i.e., under the stimulus control of the behavior of another organism), and which are not. Plus it really is not a Boolean decision (is/is-not). It is more of a matter of degree. It is a continuum from “not” (e.g., 0.0) to “is” (e.g., 1.0).

This seems like it could especially be a problem when stimulus media go through transductions and delays. At some point, do these processes alter the nature of the “stimuli” to the point that they no longer function in the same manner as they would without the transductions?

As a type of social behavior, all of this applies to verbal behavior. Which is part of why dealing with verbal behavior can be very messy.

Emergents, redux

‘In my post, Emergents: Social & Verbal Behavior, I mention Andy Lattal’s question about social behavior as emergent.

Hake, et al. (1983, p 22) says:

  • “Neither the social-emergent position that behavioral stimuli are unique such that they require separate laws, nor the reductionistic position that they are the same as nonsocial stimuli and therefore do not require any separate treatment, seems warranted …”

Thus, I agree with my original statement that social behavior is not an emergent, in the traditional sense stated above. I also agree with the Donahoe, Burgos, and Palmer (which quote I cannot find) treatment of emergents in the sense of “as if”. Social behavior (and therefore verbal behavior) look to be special because of the complex interactions among factors. So complex that it is difficult or impossible to disentangle sources of control, thus giving the appearance of some phenomenon that is “greater than the sum of its parts”. But then, “sum” is vastly too simple a mathematical operation to be useful in most cases, anyway. And strictly mathematical treatments probably cannot easily represent the complexities.

Critically, there is this clarification regarding behavioral and nonbehavioral stimuli a few sentences later:

  • “The present results … suggest that behavioral stimuli do differ from nonbehavioral stimuli to the extent that (1) behavioral stimuli are more variable than non-social stimuli, since they are not as easily controlled with respect to onset, offset, and consistency from one presentation to the next …, and (2) behavioral stimuli are usually operative in complex discrimination procedures involving nonbehavioral as well as behavioral stimuli and the two must be disentangled before the role of behavioral stimuli can be assessed.”

Thus, behavioral and nonbehavioral stimuli are more difficult to manage. But there is no theoretical difference. It is operational.

Non-mathematical interpretation of complex behavior

Which brings us to this article, where we might begin to think about a possible route for creating Computational Verbal Behavior software. Maybe.

Okay, I’m burning out. Time to bring this to a close for the day.

The interpretation method Donahoe discusses is based on his work with José Burgos and Dave Palmer. This is the best primer for that work:

Along with this earlier article: