(Sorry this is taking so long. I want to get it out right, so I choke up and get it out not at all. An agile lesson, I spoze.)
Apparently stimuli are selected as cues dependent upon the nature of the subsequent reinforcer.
It sounds harmless, but the last sentence of the abstract for John Garcia and Robert Koelling’s 1966 paper Relation Of Cue To Consequence In Avoidance Learning was the beginning of the end for the then-prevalent psychological theory of behaviorism.
A Quickie Summary
Teach rats not to drink when some stimulus is present by punishing them if they do.
For example, teach a rat not to drink by punishing them with nausea if they do. Rats aren’t stupid, you know. Any rat can learn not to drink when there’s a taste in the water that is paired in their minds with nausea. (Didn’t you see Willard? Some rats can organize the horrible death of all the awful people around you.) Tests were run including means of nauseating the rats without tainting the water with a taste. No problem.
But, let’s teach a rat not to drink when drinking makes a bright light and a clicking noise. They do okay in the daily instant trials while they’re still learning the lesson. But during the post-test, where we test to see how well they learned the response, they, they, well, they got *nothing*.
“Nothing” is a very important result.
What’s It Prove About Rats?
It proves that the rat’s mind is not a generalized problem-solving machine. It can learn that action A when cue B is on causes C, and to act accordingly, provided only that the nature of the cue (B) and the nature of the consequence (C) somehow fit together.
These tests are not tests of whether the rat can *detect* A, B, and C. In other words, this is not a perceptual problem. All of the various B’s and C’s used were cross-tested against each other. The rats clearly perceived the B and the C in every case: because they responded in every single case during the learning period.
If you’ve no great love for science, let me just make it clear how earth-shattering this result was at the time. One reviewer wrote “these results are about as likely as opening a cuckoo clock and finding birdshit inside”.
By now, of course, it’s been demonstrated over and over again, tens of thousands of times.
What’s It Prove About People?
Well, it doesn’t prove anything about people. Rats aren’t people. Though some people are sometimes rats.
Yeah, Yeah, What’s It Suggest About People?
Thank you. It’s important that we get the difference between proof, evidence, and hint, and this one is a hint, that’s all.
As a hint, it makes me wonder whether human brains are generalized problem solvers.
Are there problems we see day after day, simple perceivable events triggering simple perceivable consequences, that — with all our tremendous mental acuity, we still just can not get?
Is there a problem-space where our minds work and a problem space where they don’t?
It’s possible. Some would say it’s even likely. I personally believe it’s the everyday case. It launched me, by the way, on my initial investigations of thrownness. More, later, on that.
What Do Garcia’s Rats Mean To A Coach?