For almost an hour after it happened, I sat in this room and just wept. I realize that it is not the style in our culture for a grown man to weep, but there are times when cultural taboos must be forgotten. Again, had I thought much about the sort of experiments He must have had in mind, I most probably could have predicted the next step. Even so, I most likely would have repressed the knowledge.
One of the standard problems which any learning psychologist is interested in is this one—will an animal learn something if you fail to reward him for his performance? There are many theorists, such as Hull and Spence, who believe that reward (or "reinforcement," as they call it) is absolutely necessary for learning to occur. This is mere stuff and nonsense, as anyone with a grain of sense knows, but nonetheless the "reinforcement" theory has been dominant in the field for years now. We fought a hard battle with Spence and Hull, and actually had them with their backs to the wall at one point, when suddenly they came up with the concept of "secondary reinforcement." That is, anything associated with a reward takes on the ability to act as a reward itself. For example, the mere sight of food would become a reward in and of itself—almost as much a reward, in fact, as is the eating of the food. The sight of food, indeed! But nonetheless, it saved their theories for the moment.
For the past five years now, I have been trying to design an experiment that would show beyond a shadow of a doubt that the sight of a reward was not sufficient for learning to take place. And now look at what has happened to me!
I'm sure that He must lean towards Hull and Spence in His theorizing, for earlier today, when I found myself in the jumping stand room, instead of being rewarded with my usual protein balls when I made the correct jump, I—I'm sorry, but it is difficult to write about even now. For when I made the correct jump and the door opened and I started towards the food trough, I found it had been replaced with a photograph. A calendar photograph. You know the one. Her name, I think, is Monroe.
I sat on the floor and cried. For five whole years I have been attacking the validity of the secondary reinforcement theory, and now I find myself giving Him evidence that the theory is correct! For I cannot help "learning" which of the doors is the correct one to jump through. I refuse to stand on the apparatus and have the life shocked out of me, and I refuse to pick the wrong door all the time and get an icy bath time after time. It isn't fair! For He will doubtless put it all down to the fact that the mere sight of the photograph is functioning as a reward, and that I am learning the problems merely to be able to see Miss What's-her-name in her bare skin!
I can just see Him now, sitting somewhere else in this spaceship, gathering in all the data I am giving Him, plotting all kinds of learning curves, chortling to Himself because I am confirming all of His pet theories. I just wish....
Almost an hour has gone by since I wrote the above section. It seems longer than that, but surely it's been only an hour. And I have spent the time deep in thought. For I have discovered a way out of this place, I think. The question is, dare I do it?
I was in the midst of writing that paragraph about His sitting and chortling and confirming His theories, when it suddenly struck me that theories are born of the equipment that one uses. This has probably been true throughout the history of all science, but perhaps most true of all in psychology. If Skinner had never invented his blasted box, if the maze and the jumping stand had not been developed, we probably would have entirely different theories of learning today than we now have. For if nothing else, the type of equipment that one uses drastically reduces the type of behavior that one's subjects can show, and one's theories have to account only for the type of behavior that appears in the laboratories.
It follows from this also that any two cultures that devise the same sort of experimental procedures will come up with almost identical theories.