How intelligent is behaviourist learning?

If you read my (annual? are they that frequent?) posts on this blog, you know that I puzzle about the extent to which Moshe is a behaviourist, and the extent to which I think this is or isn’t an adequate approach to a) human learning and b) the kind of learning we do in Feldenkrais.

It so happens that in the other 9/10ths of my life teaching bioethics, I also have an open bucket in the back of my head that is the question set: “what presuppositions and blindspots might there be in stories about learning as routinization, which are common in the critical thinking literature in medicine; and what are the prospects and pitfalls for extending these accounts of critical thinking to critical thinking about ethical and social issues in medicine?”

From the Stanford Encyclopedia of Philosophy, here is an overview of philosophical perspectives on behaviourism:

The New York Times Magazine has an article this weekend: How Companies Learn Your Secrets. It’s an interesting read for many reasons, but specifically related to behaviourism and learning, I was struck by its summary of the history of thinking about habit formation.

It first tells a story of rat learning. When I was reading the story about rat learning, I was imagining very active and curious rats, with creative strategies for approaching situations that contain unknowns, and with definite goals and interests. A more limited set, certainly, than humans (food, presumably some companionship at least with other rats…). They ‘generate hypotheses,’ try out going this, that and the other way in search of the chocolate. Gradually, and intelligently, they discard things that don’t work, and focus in on, repeat, perhaps even refine, and extend things that do work. Sounds like active, intelligent, goal-direction learning to me. And on such an account of learning, imagine how much richer and more complex the process would be if carried out by creatures with a broader range of goals, an ability to share strategies for innovation and gather feedback socially, and so on (like humans).

The author returns a few paragraphs below to parse the outcome of that history. The story now is: What we know is that people repeat behaviours more if they are rewarded for doing so. This is how habits are formed. In this summary, all of the sudden, it matters only minimally what the person wants or is trying to do. It matters minimally that the person has to be active and creative in generating options for how to achieve what they want. It matters minimally that the person has to have any kind of refined awareness of the world and of themselves in order to gather the right lessons in the feedback loop.

These struck me as radically different views of the process of learning. From another perspective, are they just two descriptions, at two different levels, of the “same” phenomena? Certainly, the latter focuses on the “steady state” after learning has taken place, when behaviour has been routinized. But the fact that we gravitate to the image (part metaphor, part deployment of what we understand goes on in myelination) that pathways are laid down and then strengthened, rather than the understanding that options are generated, tested, and then selectively discarded (which is what the neuroimaging studies may be metaphorically deployed to suggest), construes it as a rote process that could well be externally imposed, rather than an internal, creative, and intelligent process of negotiating self and world.

Leave a Reply

Your email address will not be published. Required fields are marked *