The Experimental Analysis of Halloween

Wanna know how to tell if you have a successful science of behavior?

Here’s a test. Pick any random context featuring behavior, and see if your science has something to say about it. And I’m not talking about free-form interpretation (i.e., what behavior analysts like to call a “conceptual analysis”). No, I mean good old peer-reviewed primary empirical studies, i.e., data you can sink your teeth into.

Or maybe I should say “sink your fangs into,” because to make my point I’ve chosen, as my random context featuring behavior, Halloween (aka All Hallows Eve), that odd collection of rituals and superstitions that resurfaces each year on or around October 31, depending on your location. Oh, by the way, if you’re not IN one of the locations that observes Halloween** I apologize that the premise of this post may land a little flat, though I think the take-home message is pretty general, so bear with me.

** Places that celebrate Halloween include but are not limited to: the U.S., Canada, Mexico, United Kingdom, Australia, Belgium, Czech Republic, Ireland, Dominican Republic, France, Greece, India, Italy, Hong Kong, Poland, Japan, Malaysia, New Zealand, Romania, Serbia, Singapore, Poland, Sweden, South Korea, Switzerland, Taiwan, United Arab Emirates. A number of other nations blend Halloween with other holidays.

For the uninitiated, a further apology, but I won’t explain Halloween (get the run-down here) except to say that it has a spooky vibe and (in the U.S. at least) is represented by a number of fear-inducing symbols and motifs like those shown below.

For my acid test of our science of behavior, I ask: What does it have to say about these 12 topics?

You might think that’s a facetious question, a shameless click-grab for the good old ABAI blog site, but actually I’m dead (intentional Halloween pun there! I’m hilarious!) serious. At the end of this post, you’ll find selective bibliographies for each of the 12 topics. Sure, in some cases I’ve stretched a little to find relevant studies, and yeah, I might have included one or two tongue-in-cheek references. Also, in my enthusiasm to compile the lists I let some review/discussion type articles slip in, so it’s not all straight-up empirical reports. But otherwise these lists are 100% on the level. And, by the way, most of them are far from comprehensive, having wrapped up because I got tired of looking rather than because I stopped finding sources. I could have included a ton more stuff.

And so what? Well, As I see it, there are two lessons to be derived from the lists.

The first is: Yay for our team! Given the apparently silliness of my test, it’s shocking — dare I say scary? — how well the science holds up [though see Postscript 1]. And, to be clear, I didn’t even grab the lowest-hanging fruit. Our science has contributed to countless studies on how to treat fear and avoidance of things like those shown above, but to most people that’s so obvious that I didn’t bother to create a list. Doing so would have expanded this post by at least an order of magnitude.

As I mentioned in a previous post, some years ago Cognitive Psychologist and President of the American Psychological Society Henry Roediger wrote that “behaviorism won.” To some observers, he said, it may seem as if behavioral psychology faded into obscurity, but this is mainly because certain aspects of a behavioral perspective have succeeded so well that they are now part of the zeitgeist of contemporary infrastructure of psychological science. Because behavioral methods influence those of every credible scientific effort to study behavior, the influence tends to get taken for granted.

To get a feel for just how true this is, give the bibliographies at the end of this post an extended skim. I think you’ll be impressed.

Now, if you’re attentive while skimming, maybe you’ll pick up on the second important lesson. Let’s call it Don’t pat yourself on the back just yet.

Did you notice that the vast majority of the sources I listed were in journals you’ve never thought to crack open (or didn’t know existed)? Did you see a lot of authors you’ve never heard of, and research questions you hadn’t thought to ask? Did it strike you that many of the investigations are inherently interdisciplinary in nature?

Those definitely were my reactions, and with them came a humbling — no, call it frightening — realization. If I’m unaware of vast literatures with clear relevance to the mainstream of behavior analysis… if I’m not familiar with journals in which there’s clear behavior-science (or at least behavior-science-adjacent) action… and if I don’t have skills that clearly qualify me to participate in free-ranging interdisciplinary conversations about behavior… can I really call myself a behavior analyst?

At the most basic level, even preliminary to what Skinner said about the experimental analysis of behavior and what Baer-Wolf-Risley said about applied behavior analysis, are a couple of features that ought to describe us all:

We’re behavior analysts first, and specialists in some targeted domain second. What we know about behavior is highly general.
We are curious about anything and everything that helps to make behavior make sense.

And herein lies the curse of “behaviorism won.” No matter how big the science of behavior looks to you, it’s waaaay bigger than that. As Roediger noted, a behavioral approach now influences research of most types in some way, and in many research areas a behavioral viewpoint has taken on a life of its own. I promise you that the authors of many of the papers in those bibliographies have never been to a behavior analysis conference or peeked into a behavior analysis journal. And they certainly don’t know that you exist.

Which begs this question: If the science of behavior has become so diffuse and scattered that its parts no longer interact, is it really one science?

I say yes, it can be, IF there is glue to hold it together, and I believe the adhesive should come from us. Who better to evaluate and clarify how all of the varied expressions of behavior science fit together, how the parts inform a whole, than we who are Skinner’s most direct descendants and have, presumably, the firmest grasp on his conceptual system?

Easy to propose, but not so easy to do, given the vast amount of stuff out there that requires integrating. Nobody can read everything [also see Postscript 2].

To illustrate I will digress and share a Sisyphean tale of terror, a real-life Twilight Zone episode, from my grad school days. I swear I am not making this up. When I arrived to begin my program, I met a student, let’s call her Jane Bierce, who was in her 11th year of graduate study. When I asked why she said, “My dissertation proposal is almost ready to go! But every time I think I’m finished, new journal issues come out and I have to update the literature review.” When I graduated four years later, Jane Bierce still had not proposed (insert Rod Serling commentary here). Yes, it’s every grad student’s nightmare, even scarier than black cats and evil clowns: the perpetual dissertation (cue sinister music, fade to black).

But there’s an alternative to everyone-reading-everything: crowdsourcing. We can all make a contribution by seeking out, and casually keeping up with, one or two areas of study that maybe don’t get a lot of attention in our main journals and our training programs. And whenever the occasion arises in our interactions with other behavior analysts, we can share what we’ve learned. I’ve written elsewhere about the value to individuals and to a discipline of intellectual breadth (see my “Connections” series of posts, in which I assert that the more things you know, the more likely you are to know something pertinent to some new problem). If we’re all just a tiny bit more broadly educated, the discipline becomes richer.

AND, if we do this, opportunities arise to make connections with all of those folks who are using our tools to explore questions we never thought to ask. We have things to teach them, and they us. But they probably won’t seek us out, so the onus falls to us. We already do a bit of reaching across boundaries — the B.F. Skinner Lectures at the ABAI Convention are one of my favorite examples — but the 12 bibliographies show us that so much more is possible.

The result can be a unified science of behavior science that is widely distributed, rather than segregated into numerous little fiefdoms. And that, my friends, would be quite a treat.

Postscripts

The one Halloween icon on which I couldn’t find any behavior science was witches/witchcraft, except for one curious book that equated behavioral programs in correctional institutions, somehow, with witch burnings (and I swear that I am not making that up), and some near-miss sources involving magic, like these:

Flora, S. R. (2000). Praise’s magic reinforcement ratio: Five to one gets the job done. The Behavior Analyst Today, 1(4), 64-69.

Miguel, C. F. (2017). The generalization of mands. The Analysis of Verbal Behavior, 33(2), 191-204.

Anyway, students, if you’re looking to make your mark with a novel thesis or dissertation topic, here’s your shot. You’re welcome!

2. Another thing that’s critical to integrating the varied expressions of behavior science is a knack for separating the wheat from the chaff. Because many domains in which behavior science has had an impact are interdisciplinary, you’ll find terms and concepts from many disciplines and a lot of different types of research design. Behavior Analysis Purists tend to dismiss anything that’s this “impure,” and that’s a shame. Because there are many interesting and replicable findings out there just waiting to be incorporated into a thoroughgoing account of behavior. By ignoring them we miss important opportunities to test the generality of our framework and, to be quite honest, to people out there who know they are onto something, we look foolish for being ignorant of their findings while proclaiming that we have THE science of behavior.

When you look closely at work not conducted by inner-circle behavior analysts, you have to be able to distinguish between the following three things, all of which exist in quantity outside of our journals:

Work that validates the theoretical framework we’ve already put into place (even if the researchers didn’t intend their work as a test of our ideas, or even if they don’t see the correct connection)
Work that points toward something not currently emphasized in our framework
Work that’s claimed to be relevant to our framework, but probably isn’t

Like all discriminations, mastering this one requires practice with lots of exemplars, so there’s no time like the present to start working on your breadth.

A Selective Bibliography of
“Halloween Behavior Analysis“

Within each section, items are listed in no particular order.

Bats

Black Cats

OK, NOT BLACK PER SE, BUT DEFINITELY CATS.
LOTS AND LOTS OF CATS. THORNDIKE WOULD BE PLEASED.

Clowns

Ghosts

“Monsters”

Leave a Reply Cancel reply