Talk:Reinforcement

This talk page

Hello! I have been doing some revisions to this page. I also went ahead and archived the old talk page which was getting to be a disorganized mass. Please discuss topics relevant to this page here, with each topic under a new heading, going down the page with each successive topic.

In other words, PUT NEW TOPICS AT THE BOTTOM OF THIS PAGE. Thank you!

Quick tip. Put ":" in front of replies to topics so people can see where the reply begins. :) Jcbutler (talk) 16:20, 31 January 2024 (UTC)

Hello, I would love to contribute but I am still learning how to.

I think the beginning of the page needs clarification. The word reinforcement was generally used in Skinner’s sense, in the context of operant conditioning. However the distinction between operant conditioning and classical conditioning is tricky.

On its face the example describes the conditions used in autoshaping. If a light turns on reliably before food is presented the classical contingency dominates and the pigeon pecks the key compulsively, regardless of the consequences. The key light is a conditioned stimulus. The food is the unconditioned stimulus. The whole point of operant conditioning is that the increase in behavior is determined by the consequences, the reinforcement.

So rewriting the reinforcement page to be readable yet accurate is going to require lots of effort from lots of people.

Rewriting the introductory definition to be more like the APA definition is needed.

If the intention is to distinguish reinforcers from unconditional stimuli, the example should ideally be something where the animal “spontaneously “ generates a behavior and starts doing it more and more because it gets food right after performing the behavior. The fact that there are inevitably antecedent stimuli, even if the experimenter doesn’t control them is a complication that Skinnerian behaviorists traditionally glossed over. Could it be glossed over here? Bsplendens (talk) 17:05, 24 February 2024 (UTC)

Hello and welcome, Bsplendens. I think a detailed discussion of the difference between operant conditioning and classical conditioning would be a little off topic, and better suited to one or both of those pages. It is, as you say, tricky, and even a point of contention over the years. But perhaps we should use Skinner's term "discriminative stimulus" to make this more clear. I agree that a different introductory example could be useful. But I like the existing lab example because it is very clear and matches the adjacent picture. Jcbutler (talk) 15:15, 26 February 2024 (UTC)

Next steps

We still need to clean up the text and references on reinforcement schedules and applications. Perhaps we should add a bit about intrinsic vs. extrinsic reinforcement. What else needs to be done? Jcbutler (talk) 17:27, 31 January 2024 (UTC)

Reinforcement vs Reinforcer

OK... I will admit that reinforcer is technically more accurate than reinforcement, though they tend to be synonymous in common usage, as is "reward" much of the time. I see the point you are making, Bsplendens. But I still wonder if it isn't better to use the term that is used in the title. I would also observe that the paragraph is now confusing because the definition still says that reinforcement (not reinforcer) is the consequence of the behavior. As you say, the whole thing could be revised, if you feel up to it. Perhaps defining "reinforcement" and then distinguishing it from "reinforcer" in the terminology section. Jcbutler (talk) 19:50, 11 March 2024 (UTC)

I rewrote the introductory definition slightly, so I think the paragraph reads a bit better now. Jcbutler (talk) 19:13, 13 March 2024 (UTC)