Operant Conditioning and Brain Training

Professional bias in Conventional Wisdom

Much of the early work in brain training was done by psychologists. Psychologists were the early publishers of the research, and guided much of the later publishing. They understood and described what they were doing in the terms of psychology—especially behaviorism.

Quite naturally they developed a sense of ownership of the field. It worked, after all, with human behavior, thoughts and feelings—mental health fields. Brain training began to be defined by them as a treatment tool. They set up certifying and professional bodies run by psychologists and began to try to limit who was qualified to do it.

When we train a brain, per the psychologists, we are conditioning brain cells to respond in a different way. It is a model that psychologists understand, and it seems to put them in charge of the technology, but is it a fact—or an assumption?

The Conditioning Model

You’ll find it stated right up front in most any book or article you read: brain training is “operant conditioning”.

Skinner taught pigeons to peck a button when a light turned red (but not green) by giving them food when they did it. Barry Sterman trained cats to relax completely (thus making a specific brain rhythm in a specific area) by giving them drops of milk.

It’s understandable that psychologists would present brain training in their own terms—they understand that way of thinking, and it keeps them in charge. But really, is that the only—or even the best—description of what’s happening when we train human brains?


Immediately receiving food after doing something is pretty clear and positive. But when a computer plays music or runs a video, is that the same thing? Pecking a button or relaxing (which most cats can do quite well) are easy tasks. But teaching a brain to release a longstanding anxiety pattern by playing music or running a video or giving points in a game seems a lot different.

Sterman himself has claimed that the way to train increases in frequency activity is to give feedback when something is done, then wait for a refractory period and start another trial. Few trainers bother with this—and those who do not get equivalent results to the “proper” conditioning model.

What if the client doesn’t like the music? What if the video or game is boring? That’s not the same thing as getting food. Can you train the pigeon by giving him a penny each time he pecks? Doubtful.

Every trainer asks these questions, but I’m unaware that anyone has ever tested them. We just accept that we’re conditioning the brain, because that’s what we’ve read everywhere.


Cats already know how to relax; pigeons to peck. The people many trainers face don’t KNOW how to pay attention. They’ve rarely done it for more than a few seconds. Depressed people don’t know how to see the world in a positive way. We’re not training them to perform a known action. We’re asking their brains to produce, in many cases, a state for which they have no reference. It’s like trying to train a pigeon or cat to change a tire.

We don’t expect the pigeon to peck bars when lights flash outside the lab. Pigeons peck for food and cats relax naturally. But brain training clients are expected to take the practice of 20-40 sessions at a computer and literally change the way they relate to the world. We want the ability to focus to appear in the classroom, in a chat with friends, when doing homework or a boring task at work. We aren’t teaching a reaction, we’re trying to change a capacity—a whole new way of dealing with the world.


What happens to the pigeon’s behavior when you stop giving the food? Eventually the pigeon learns that pecking the lever at certain times no longer provides the feedback, and he stops doing it. But brain training should result in lasting changes that generalize throughout a client’s experience and last well beyond the end of the “training” period. If the changes are extinguished when the feedback is no longer given, then what’s the benefit of training?

My point here is not necessarily to say that behaviorists are wrong, that operant conditioning doesn’t work in humans (though there’s a lot less evidence that it does) or that it’s not a viable way of understanding brain training. I’m simply saying that there are other ways of understanding what training does that make as much or more sense.

Share this post: