PHILOSOPHER-SCIENTIST: It certainly can. Although a painful stimulus is involved, it is the removal of the shock that the rat would consider a most rewarding experience.
TINNY: How might what is occurring in the holistic analysis be described?
PHILOSOPHER-SCIENTIST: As soon as the shock began the holistic analysis would be seeking information which could assist in finding a way to escape the pain. Initially the lever would have no particular significance as information which might be useful in escaping from the pain. In its panic, since no means of escape would be obvious, the holistic analysis would advise carrying out actions which had no known connection with bringing an end to the pain. This is done because memory provides information that such a tactic, emitting varied responses if there is no obvious solution, sometimes hits upon a solution. There are not too many different possibilities in an experimental learning chamber, so pressing the lever would soon be one of the actions tried. As soon as the lever was pressed, and the shock immediately stopped, lever pressing would suddenly figure significantly in the holistic analysis. The holistic analysis would, even though by no means certain of the relationship where lever pressing terminated shock, advise that the lever be pressed again in hopes that it would once more allow an escape from the pain. After a few instances of the response terminating the electric shock, holistic analysis would accept the knowledge that lever pressing is an appropriate response in the circumstances and advise a high rate of lever pressing.
TINNY: Now an example of negative reinforcement in human learning, please.
PHILOSOPHER-SCIENTIST: Sometimes even behaviour which seems obvious has actually been conditioned. Given that a person does not want to get soaked, when it begins to rain they would seek shelter. In this situation the rain is the ongoing aversive stimulus and seeking shelter is the response. The response terminates the ongoing aversive stimulus. Seeking shelter stops the person from becoming soaked. Because this obvious relationship has been part of human history for so long everyone knows the appropriate response if you don't want to be soaked by the rain is to seek shelter. Even so, seeking shelter when it rains is still a response conditioned by negative reinforcement. If the response of seeking shelter did not stop people getting soaked when it rained, then no one would seek shelter when it rained.
TINNY: There would virtually be no behaviour if there was no operant conditioning.
PHILOSOPHER-SCIENTIST: What about the actions determined by physics or chemistry?
TINNY: Operant conditioning is usually believed to exist only at the animal and human level; but, all things have envolved and their development can be traced back to the beginning of the physical universe. The physical and chemical reactions are early stages in the developmental progression of operant conditioning.
PHILOSOPHER-SCIENTIST: Although it is hard to see the relationship of the laws of physics and chemistry to the laws of learning, I see the logic behind your claim.
TINNY: To me it seems obvious.
PHILOSOPHER-SCIENTIST: Soon the true nature of the laws of learning will be obvious to all.
TINNY: As I came to understand the laws of learning I was most amazed.
PHILOSOPHER-SCIENTIST: You said you would tell me about when you first realised the potential of these laws of learning.
TINNY: I'll never forget that experience. I had learned about operant conditioning as a scientific concept, but somehow had not developed any great enthusiasm for its practice or potential. I did want to try it for myself, though, so I built an experimental learning chamber that could be used for teaching birds. I chose to try these conditioning procedures with birds because we don't often think of birds as good learners, and I wanted to find out how effective these learning principles were. I wanted to present a real challenge to operant conditioning. Many birds and animals live around here so I chose a chicken that I liked a lot. I used to feed her sometimes and she would follow me around.
PHILOSOPHER-SCIENTIST: Do you think that was because she liked you too or was following you around a conditioned response.
TINNY: I think she followed me for both those reasons. Human emotions are reserved for humans, but animals feel and express emotions on their level also. I don't believe she considered me as nothing more than a food dispenser.
PHILOSOPHER-SCIENTIST: Does it particularly matter whether or not you like the animals that take part in learning experiments?
TINNY: It may not have much effect on the results of the experiment, but I think it is important for the general well being of the animals.
PHILOSOPHER-SCIENTIST: Did this chicken have a name?
TINNY: I called her Aphrodite because she was so beautiful.
PHILOSOPHER-SCIENTIST: Tell me in detail everything you did during this conditioning experiment.
TINNY: I waited until I could tell by the way Aphrodite was acting that she was hungry and then picked her up, held her for a while to make sure she was feeling comfortable, and then placed her in the experimental learning chamber. She seemed to feel alright in there and she began exploring around the chamber. I had the food hopper hooked up so I could push a button that would raise the food tray so Aphrodite could get at the grain that was to serve as a reinforcing stimulus. I kept the hopper up for a short while so that she could find out where to get food in the chamber. Then I got her used to the food hopper sometimes being up with grain available, and sometimes being down, out of sight. Each time the hopper was raised there was a fairly loud clicking sound. This served as an auditory cue that food was available, so when Aphrodite happened to be exploring another part of the chamber while the hopper was down and she heard the click she would know the food hopper was up. She would then run over and begin eating. She learned all that in a couple of minutes which I thought was very fast, but nothing compared to what was to come.
I chose a very complicated response to condition by positive reinforcement. I wanted to see if Aphrodite could learn to turn a full circle to the right while standing on one leg. This was a response that she would probably never have done without being taught. If I could condition this difficult response I would be sure of the effectiveness of the laws of learning. Without knowledge of the principles of operant conditioning it would be virtually impossible to teach such a complex behaviour to a chicken. I realised that a complex response could not be taught as a whole but must be taught in parts. I decided to first condition Aphrodite to stand on one leg. As most chickens do, she would fairly often stand on one leg for short periods of time while scratching with the raised leg. First I pushed the button bringing up the food hopper just after she had raised her left leg. Each time I waited until she had been standing on that one leg for a longer time before reinforcing that response by raising the food hopper. In only several minutes Aphrodite was standing on one leg almost constantly. Luckily chickens have very good balance and seem to stand on one leg easily.
Now that she was standing on one leg I wanted to condition the other part of the complex response, turning a full circle to the right. Chickens are always turning their heads so they can look around. I waited until she, still standing on one leg, first turned her head to the right and I then pressed the button raising the food hopper. She would stand on one leg right in front of the hopper so she could bend over and eat while still standing on one leg. Each time before raising the hopper I waited until she turned her head further to the right. At first it was just her head that turned, then her head and neck, then she began twisting her body until she could move no further to the right without jumping around on one leg. After each of these further movements to the right I would raise the food hopper. Eventually she was turning more than half way around, and it was easier for her to complete the full turn to get to the food than turning back to the left. After only a few more minutes Aphrodite was consistently turning full circles to the right while hopping on one leg. Each time she emitted that full, complex response I would reward her by making the grain available.
This whole learning process took about ten minutes. It was hard to believe how successfully the laws of learning worked. I could see Aphrodite's behaviour changing as the contingencies of reinforcement changed. The directness of the relationship between the response and the stimulus astounded me. I realised that if the learning experience could be seen this clearly and worked so effectively that there are virtually no limits to the development of human potential. I'm sure most people are in the same position as I was in regarding their understanding of learning before I performed this experiment. I took learning for granted. It was just something that happened because of the nature of people and animals. I had never even considered that each tiny step in every learning experience is a reaction to the relationships of the responses and stimuli involved. The contingencies that affected each step could be known. If they could be known they could be planned for and organised so as to allow human potential a degree of expression never before possible. People need no longer have their lives and abilities arbitrarily determined by contingencies in the environment that they were neither aware of nor had any control over.
A person who truly understood these principles of learning could exercise control over their own destiny. I set a goal for Aphrodite, determined which relationships between her natural responses and the following stimuli would increase the probability that the goal behaviour would be reached, and set up the environmental conditions in which the goal was achieved through positive reinforcement. What I did for Aphrodite I could do for myself. We could each do for ourselves. We could set our goals, determine which experiences we must undergo to achieve those goals, and organise the environmental conditions so that we will undergo those experiences which would influence our developmental progression toward our goals. What a beautiful future it would be. We could be as we chose to be.
PHILOSOPHER-SCIENTIST: Apparently the knowledge of how real and how powerful the laws of learning are has affected you greatly. You saw operant conditioning with a new clarity.
TINNY: Sometimes I think it was that moment of insight into the true nature of the learning process which allowed my consciousness to expand, thereby allowing me to see all knowledge as a whole. I saw the unified nature of all that exists. After that experience I could never again view the world from the traditional perspective. The new worldview became my perception of reality.
PHILOSOPHER-SCIENTIST: It must have been a good feeling.
TINNY: It was possibly the single most important moment in my life. I had always loved learning, knowledge, and truth; but, this new insight into the ultimate reality was almost like a rebirth. At first it was sometimes difficult to handle this new knowledge. As I suddenly saw the laws of learning so intimately and inextricably bound to every instant of my life the veil of ignorance and self-deception fell away. The honesty was almost too much. It took me some time to adjust to a world without illusion.
PHILOSOPHER-SCIENTIST: Did you ever regret having undergone the experience?
TINNY: Never once, even at the most difficult moment, did I ever regret gaining this new understanding or becoming a new person.
PHILOSOPHER-SCIENTIST: Was the overall effect of the experience really so great you became a new person?
TINNY: There is no better way I could describe the change I underwent. It was a moment of transformation.
PHILOSOPHER-SCIENTIST: What was at the heart of that transformation?
TINNY: It was the great increase in freewill. I suddenly became aware of and responsible for my every thought, word, and deed in a way that had never before been possible.
PHILOSOPHER-SCIENTIST: It had to happen.
TINNY: To me?
PHILOSOPHER-SCIENTIST: To the human race. But let's not get away from our discussion on the laws of learning. Your description of the conditioning process with Aphrodite introduced several aspects of operant conditioning that you hadn't yet described. I'd like to know more about them.
TINNY: I'll be glad to explain them. As we discuss these learning principles I try to let the knowledge unfold in a natural order. Each new bit of information will contain seeds of further truths.
PHILOSOPHER-SCIENTIST: Would you tell me more about the significance of the clicking sound the food hopper made when it was raised to present the grain as a reinforcing stimulus?
TINNY: During the learning process it is important that any response being conditioned be followed immediately by some change that will be noticed. If the lapse of time between the response and the following stimulus was even one second long the chances of the response being effectively conditioned may be significantly decreased. The longer the period between the response and the following stimulus the less obvious the relationship between the two becomes. In many learning situations it isn't possible to follow the response immediately with the reinforcing stimulus. In the case of Aphrodite's training it would have taken a second or more before she could actually begin eating the grain, which was being used as a reinforcing stimulus. Some way had to be found to make sure she realised she was being reinforced almost the instant she responded, not a second or more later. In this learning experiment that gap was bridged by a clicking noise which sounded the instant the response being reinforced was performed.
PHILOSOPHER-SCIENTIST: Why would a clicking noise be reinforcing?
TINNY: In itself the sound would be a neutral stimulus. Holistic analysis would, in the beginning, ignore the click as having no useful significance. After the clicking sound had occurred a number of times, very shortly before the grain was made available, it came to have significance in that it signaled the food was forthcoming. The significance of the clicking sound was conditioned by being paired with the positive stimulus, the grain. The grain in this case would be called a primary reinforcer in that it directly satisfies a physiological need.
PHILOSOPHER-SCIENTIST: Do all primary reinforcers directly satisfy physiological needs?
TINNY: They all do.
PHILOSOPHER-SCIENTIST: Could you give me some examples of other primary stimuli?
TINNY: Food, water, air, physical pleasure, or pain are all primary stimuli. A primary stimulus can be either positive or negative.
PHILOSOPHER-SCIENTIST: Is there a general name for a stimulus, such as the clicking sound, which serves the purpose of signaling that the response is to be followed by a primary stimulus?
TINNY: Stimuli of that type are called secondary. A secondary stimulus gains its reinforcing quality by being paired with a primary stimulus.
PHILOSOPHER-SCIENTIST: Can a secondary stimulus also gain punishing properties?
TINNY: A secondary stimulus gains its influence by being paired with a primary stimulus. If the primary stimulus is being used as a reinforcer then the secondary stimulus gains reinforcing properties; and, if the primary stimulus is being used as a punisher then the secondary stimulus gains punishing properties.
PHILOSOPHER-SCIENTIST: You say it is critical that the reinforcing stimulus follows immediately after the response being conditioned. Is it also critical that the punishing stimulus follow immediately after the response?
TINNY: In operant conditioning of all kinds there must be immediate feedback after the response. It is the result of the response which must be known for conditioning to be effective. The feedback received as to what changes come about after the response is referred to as the stimulus. During the process of reinforcement the stimulus may be positive, and be brought about by the response; or, the stimulus might be negative, and be removed by the response. In both of those learning situations, positive and negative reinforcement, it is important that feedback be given immediately after the response. Additionally, during the process of punishment the stimulus may be negative, and be brought about by the response; or, the stimulus may be positive, and be removed by the response. In both of those learning situations, presentation and withdrawal punishment, it is also important that feedback be given immediately after the response.
PHILOSOPHER-SCIENTIST: And in all those types of conditioning can a secondary stimulus be used to give immediate feedback where there might be a time lapse until the primary stimulus is perceived?
TINNY: Secondary stimuli can, and sometimes must, be used in all different forms of conditioning.
PHILOSOPHER-SCIENTIST: Would the secondary reinforcer increase the rate of the response it followed if it was no longer paired with the primary reinforcer?
TINNY: Only for a short while, until it became clear during the holistic analysis that the secondary stimulus no longer served as a cue that the primary stimulus was forthcoming.
PHILOSOPHER-SCIENTIST: The relationship between the secondary stimulus and the primary stimulus in operant conditioning is very similar to the relationship between the conditioned stimulus and the unconditioned stimulus in classical conditioning.
TINNY: They are virtually identical. In classical conditioning the conditioned stimulus is a previously neutral stimulus which gains its ability to affect the response by being paired with an unconditioned stimulus which naturally elicited that response. Both the conditioned stimulus of classical conditioning and the secondary stimulus of operant conditioning begin to lose their ability to affect the response when they are no longer paired with the unconditioned stimulus or primary stimulus respectively. The unconditioned stimulus is a primary stimulus. Both the conditioned stimulus and the secondary stimulus gain their ability to affect the response by being seen in the holistic analysis as cues which allow accurate predictions about future happenings to be made more quickly.
PHILOSOPHER-SCIENTIST: It seems there are many different kinds of stimuli.
TINNY: It seems that way but all of the apparent differences refer only to the relationship of the particular stimulus with a certain response. There is no essential difference in the stimuli; all differences are in the way the stimuli are perceived.
PHILOSOPHER-SCIENTIST: We have discussed unconditioned stimuli, conditioned stimuli, positive stimuli, negative stimuli, primary stimuli, secondary stimuli, aversive stimuli, neutral stimuli, reinforcing stimuli, and punishing stimuli. Is the only difference between all those various types in their relationship with certain responses?
TINNY: That is the only difference, although it could also be expressed in terms of how those various stimuli are perceived. In the right conditions any one stimulus, such as food, could fulfill the function of any of those seemingly different stimuli.
PHILOSOPHER-SCIENTIST: Are there more types of stimuli than those we have already discussed?
TINNY: There are more. I don't know how many, there are probably some not yet known.
PHILOSOPHER-SCIENTIST: Another question I had about the training procedure with Aphrodite has to do with the way that you rewarded her for small parts of the goal response. Why couldn't you just reward her for the full response?
TINNY: I'll use turning a full circle as an example. I don't think it would have been possible to condition Aphrodite to turn a full circle by providing a reinforcing stimulus only after the full response. Even if it could be done it would take a very long time. There are several major difficulties in trying to reinforce that response as a whole. One of the problems is quite obvious. If a response doesn't occur it can't be followed by a reinforcing stimulus. Before Aphrodite learned to turn circles in the experimental chamber, turning circles would have been a very rare occurrence. Even if she did turn a full circle and received some grain as a reward I doubt she would have realised the connection between turning the circle and receiving the food.
PHILOSOPHER-SCIENTIST: How could she fail to make that connection? During the holistic analysis wouldn't turning in a circle be considered relevant information as to how to obtain food?
TINNY: Positive reinforcement influences the future occurrences of the response that is performed immediately before the reinforcing stimulus is presented. When I say immediately I mean the very instant when the reward is given. If I waited until Aphrodite had completed a full circle she would be standing where she began when she had previously received the reinforcing stimulus. I might be conditioning her to stand in that position.
PHILOSOPHER-SCIENTIST: Why wouldn't she realise turning a full circle was the response being reinforced?
TINNY: It might take her several seconds or more to make a full turn. There is no reinforcement at any of the positions that make up the full circle, only at the end position. From the point of view of the holistic analysis no position but the final position gives any information as to when food will be forthcoming. Only the response of standing in the final position would be subject to the influence of the conditioning process. Only standing in the final position would be likely to occur more often.
PHILOSOPHER-SCIENTIST: It's still not quite clear why the response of turning in a circle wouldn't be seen by Aphrodite as the whole response.
TINNY: The truth is that turning in a circle is not one response. It is a long chain of small responses. In fact all behaviours are chained responses. It would be very hard to isolate any single response in the chain and consider it to be 'turning a circle'. It is just convenient for us to conceptualise certain chains of responses as a single response. It would make talking about a behaviour, such as turning a circle, almost impossible unless we referred to it as a single response. Just because we call chains of responses a single response for our convenience doesn't mean that holistic analysis, particularly at the animal level, treats that chain like a single response. Holistic analysis is objective and logical. Holistic analysis treats each of the links in the chain separately, although it can learn to treat a chain of responses as a single response. I not only taught Aphrodite to turn a full circle, but I also taught her to consider the chain of responses which take place during the turning of the circle as a single response.
PHILOSOPHER-SCIENTIST: The first link in the chain that makes up a full circle, which you reinforced, was turning the head to the right.
TINNY: That was the first link in the chain which I reinforced, but it was not actually the first link in the chain.
PHILOSOPHER-SCIENTIST: What was actually the first link in the chain?
TINNY: It is very difficult to say exactly which response is first in the chain. Some responses are internal, some are even mental processes. The first link is the process of turning in a circle was perhaps a mental command from the brain to the muscles which control the head and neck directing them to begin turning the head to the right. Since I couldn't see any of those internal responses I had to wait until the first motion of the head, turning to the right, until I could push the button making the grain available which would reinforce that head motion. I reinforced an early link in the chain of responses which makes up a complete turn.
PHILOSOPHER-SCIENTIST: How many links would you say would complete the chain making up a full turn?
TINNY: Because of the complexities of human and animal bodies there would be many thousands of links in that chain. Most of those responses are beyond the conscious perception of either the experimental subject or the experimenter.
PHILOSOPHER-SCIENTIST: Since there are so many linked responses which make up the chain, how can you know which responses to reinforce?
TINNY: The answer to that question involves another new concept, called shaping. All conditioning has certain basic characteristics. There is the behaviour as it occurs before the conditioning, and there is a goal for behaviour after conditioning. In the example where I trained Aphrodite, the behaviour before conditioning consisted of standing in the experimental learning chamber, occasionally scratching and looking around. The goal behaviour was standing on one leg and turning a full circle to the right. The steps in the conditioning process resulting in the change from standing and looking around to turning full circles on one leg are called the shaping process. Shaping could be defined as successive approximations toward the goal behaviour which are conditioned through a series of small steps. The behaviour before conditioning began could be called the base or baseline condition. The baseline is where all conditioning procedures must begin. During my experiment, before I started reinforcing any response, Aphrodite was just standing and looking around. This would be called her baseline behaviour. It was the starting point for the conditioning process.
PHILOSOPHER-SCIENTIST: Must conditioning always begin with the baseline point?
TINNY: It isn't feasible to begin anywhere else. You can't condition responses that are not occurring. The baseline behaviours are the only responses available to conditioning, at least before shaping has taken place. As behaviour changes when shaped through conditioning, the new responses being emitted become available to the conditioning procedures.
PHILOSOPHER-SCIENTIST: Would you explain that shaping process in relation to your example of conditioning turning behaviour? How did you know which response to reinforce first?
TINNY: Since the goal behaviour was turning a full circle to the right I looked at Aphrodite's baseline behaviour to see which was closest to the goal behaviour. The only thing she was doing that was part of the goal behaviour was occasionally turning her head to the right.
PHILOSOPHER-SCIENTIST: It wasn't very close.
TINNY: It isn't important how close to the goal behaviour the first response you reinforce is. What is important is that out of all the baseline responses going on, that the one being reinforced is a response in the direction of the goal behaviour.
PHILOSOPHER-SCIENTIST: What do you mean it must be in the direction of the goal behaviour?
TINNY: Well, in this case, if I wanted Aphrodite to turn circles to the right I wouldn't begin by reinforcing a head movement to the left. Movement to the left would be in the opposite direction to the desired goal behaviour. The way chickens turn is by turning their head and neck first. Any head movement to the right is in the direction of the goal behaviour and so that head movement is a good place to begin reinforcement. When I reinforced that first head movement to the right I was careful to take note of how far to the right that movement was.
PHILOSOPHER-SCIENTIST: Why was it necessary that you know exactly how far to the right the first reinforced head movement was?
TINNY: Because during the shaping process each reinforced response should be a successive approximation of the desired goal behaviour. Each reinforced response should be closer to the goal. After I reinforced the first head movement to the right it was important that the next head movement I reinforced be further to the right. If the second response I reinforced wasn't quite as far to the right as the first, then I would be conditioning a response in the opposite direction from the desired goal behaviour. For learning to be as fast and efficient as possible it should always be clear when progress is being made toward the goal. Each time I reinforced a head turning response I made sure it was farther to the right than the previous reinforced response. Each reinforced response was a closer approximation of the goal behaviour. The sequence of reinforced responses was - standing on one leg; the first head turn to the right; head turns increasingly further to the right; head and neck turns to the right; head, neck, and body turns to the right; head, neck, body, and hopping on one leg to the right; and finally turning full circles to the right. The final approximation reinforced is the goal behaviour, the last link in the chain.
PHILOSOPHER-SCIENTIST: Understanding the shaping process must be very important if you want to learn anything or teach anything.
TINNY: The more clearly shaping is understood the faster and more efficient the learning and teaching which takes place.
PHILOSOPHER-SCIENTIST: As you explained all those steps you went through as you conditioned Aphrodite to turn circles while standing on one leg, the process seems quite simple. With this knowledge anyone should be able to teach a chicken to do the same things. As you said, though, without the knowledge of operant conditioning it would be almost impossible to teach a chicken such a task.
TINNY: That is the power of operant conditioning; it changes the impossible to the simple. That is why it is such a powerful tool that it could be used to transform human society into a veritable utopia.
PHILOSOPHER-SCIENTIST: It will be used for that purpose.
TINNY: As we talk I am seeing more and more clearly the solution to the threatened extinction of the human species. I have no trouble now understanding how the new worldview, when spread through humanity, will bring about the solution to the many social problems we now face. It is easy to believe the human race will achieve its grand destiny, will progress to perfection in every aspect. The present state of human society can be considered the baseline condition. Perfection in every aspect can be considered the goal condition. Using the principles of learning as a tool, human society can undergo a shaping process which could positively reinforce those responses which move from the baseline condition in the direction of the goal condition. Reinforcement of successive approximations toward that goal of perfection can continue until the goal is reached. This can really happen. We can know where we are now, where we want to go, and how to get there. It can really happen like that, can't it?
PHILOSOPHER-SCIENTIST: The future of humanity as you describe it is in accord with the natural order. It can happen that way, it should happen that way.
TINNY: But will it happen that way?
PHILOSOPHER-SCIENTIST: It will if we don't destroy ourselves first.
TINNY: It will take a much better understanding of the laws of learning than the simple examples of conditioning that we have discussed so far to provide an effective tool for world social change.
PHILOSOPHER-SCIENTIST: Please tell me more then. You have convinced me of the potential of these learning principles. I want to know all there is to know about them.
TINNY: I'm sure I don't know all there is to know about the laws of learning; but I'll be glad to tell you what I do know.
PHILOSOPHER-SCIENTIST: You're pretty good at explaining things simply. Some of the things you have told me about have been extremely difficult subjects, some known only to a few scientists and philosophers. I have been able to understand almost everything you have told me so far.
TINNY: Knowledge isn't nearly as useful when it cannot be explained simply. Knowledge belongs to everyone and shouldn't be made inaccessible by being incomprehensible.
PHILOSOPHER-SCIENTIST: Sometimes you're funny in the way you use big words to express simple ideas.
TINNY: We could all benefit from having a larger vocabulary. The more words you know the more ideas you can communicate and the more things you can understand.
PHILOSOPHER-SCIENTIST: You're a very serious little girl.
TINNY: There's nothing wrong with being serious is there?
PHILOSOPHER-SCIENTIST: Nothing at all is wrong with being serious.
TINNY: Serious people have fun too; it's just that they find enjoyment in things that others don't appreciate. Serious doesn't mean unhappy.
PHILOSOPHER-SCIENTIST: What word do you think best describes the opposite of serious?
TINNY: Probably frivolous.
PHILOSOPHER-SCIENTIST: Who do you think would be living most in accord with the natural order, a serious person or a frivolous person?
TINNY: Definitely the serious person; but you don't mean everyone who is not serious is frivolous do you?
PHILOSOPHER-SCIENTIST: Words have many meanings.
TINNY: I'd like to tell you more about the laws of learning.
PHILOSOPHER-SCIENTIST: I think you do more than just realise the potential of these learning principles; I think you really like them.
TINNY: The laws of learning have become an important part of my life. I think about them all the time, and I really enjoy talking about them.
PHILOSOPHER-SCIENTIST: Do you use them in your everyday living?
TINNY: I use all the laws of learning to evaluate my every thought, word, and deed.
PHILOSOPHER-SCIENTIST: Why do you do that?
TINNY: My goal is to become perfect in every way. I want my every thought, word, and deed to be directed toward that goal. Every thought, word, and deed is a response which is subject to the conditioning influences from the environment. I want to make sure that the effect of conditioning on my every response is in the direction of perfection. If I was not aware of the laws of learning I might not even be aware when the effects of conditioning were shaping my thoughts, words, and deeds in the opposite direction from my desired goal. By understanding the principles of learning I am able to control my own destiny. I am then truly able to exercise my freewill.
PHILOSOPHER-SCIENTIST: What do you want to tell me about operant conditioning?
TINNY: I should make it clear that operant conditioning does not just occur by planning. Our every interaction with the environment is subject to the influences of the conditioning process. Operant conditioning is a natural part of our lives. Usually the stimulus following a response is brought about by that response, but there are occasionally situations when the stimulus following a response is totally unrelated to that response.
PHILOSOPHER-SCIENTIST: What happens then? Does holistic analysis ignore a stimulus which comes after a response but was not brought about by that response?
TINNY: Every response is influenced by the stimulus which occurs immediately after it. The holistic analysis looks for the relationship in time, contiguity, between the response and the following stimulus; it doesn't look for the cause. The effect is the same whether the stimulus is brought about by the response or if the occurrence of the stimulus after the response was accidental.
PHILOSOPHER-SCIENTIST: Would that particular relationship between a response and the following stimulus be called accidental conditioning?
TINNY: That would be an acceptable name; but, it is usually called superstitious learning.
PHILOSOPHER-SCIENTIST: Why is it called that?
TINNY: Because it is that accidental relationship between response and the following stimuli which reinforce superstitious beliefs. For instance, suppose someone had heard that a black cat crossing their path was bad luck, and right after a black cat crossed their path they returned home and found their house was on fire. There is no real connection between the black cat and the fire; but, because the stimulus, the house on fire, occurred right after the response, the cat crossing their path, the holistic analysis would assign some weight to that relationship between the response and the following stimulus. It would increase that person's belief that a black cat crossing their path meant bad luck.
PHILOSOPHER-SCIENTIST: So a belief can be conditioned just like a response.
TINNY: A belief is a response. It is a mental response. There are basically three kinds of responses - mental, verbal, and physical. As far as the laws of learning are concerned there is no difference between the three.
PHILOSOPHER-SCIENTIST: Does superstitious learning mainly affect mental responses, like beliefs?
TINNY: All types of responses are equally affected by superstitious conditioning. Many athletes show examples of superstitious learning. After an athlete has a particularly good performance, that performance can act as a reinforcing stimulus for responses which came just before the performance. Athletes will sometimes go through ritual behaviour before participating in their sport. They usually can't explain why they do so except that they believe it brings them luck. Sometimes this superstitious behaviour results in certain items of clothing being worn during every athletic performance. The only reason for doing this is that the particular item of clothing was worn during some notably successful athletic feat. The article of clothing in reality has nothing to do with the exceptional performance, but it becomes linked by accident in the holistic analysis. Each time an exceptional performance occurs while wearing the particular item of clothing it serves as an additional reinforcement for wearing that piece of clothing. The relationship between the article of clothing and the athletic success, while never any more real than in the beginning, becomes firmly established as significant in the holistic analysis.
PHILOSOPHER-SCIENTIST: Are animals also susceptible to superstitious conditioning?
TINNY: It isn't necessary to have human level consciousness to be affected by accidental conditioning. Animals develop ritual behaviour just as do humans. It is possible to set up the experimental chamber to condition superstitious behaviour.
PHILOSOPHER-SCIENTIST: How would you do that?
TINNY: The food hopper can be programmed to pop up at random time intervals. This means that reinforcement will be presented every once in a while regardless of what the animal is doing at the time. Whatever behaviour the animal is performing the instant before the reinforcing stimulus is presented will occur more often in the future. Not only will some response become more likely to occur, but that response will also undergo shaping.
PHILOSOPHER-SCIENTIST: How can shaping occur during accidental conditioning? Shaping is the conditioning of successive approximations toward some goal behaviour. There could be no goal response in accidental conditioning, since responses are reinforced at random. Wouldn't any response then have an equal chance of being reinforced?
TINNY: Even when reinforcers are presented randomly some responses will have a higher probability of being reinforced than others. The response most likely to be reinforced is the one that already occurs most often. Any response that is reinforced, accidentally or not, will occur more often in the future. The more often a particular response occurs the more likely it is to be accidentally reinforced. As the rate of any one response increases the rate of other responses will decrease. By this process after a lengthy period of random reinforcement some response will have been accidentally conditioned to occur at a fairly high rate. As far as the holistic analysis is concerned this response which is occurring at a high rate plays a significant role in obtaining the reinforcing stimulus. From the perspective of the holistic analysis there is a high degree of certainty in the relationship between the response being emitted at a high rate and the positive stimulus which follows that response, even though there is no true connection.
PHILOSOPHER-SCIENTIST: I can see how the rate of some particular response might be increased by accidental reinforcement, but the characteristics of the response aren't being changed by shaping are they?
TINNY: Once some response is established at a high rate during accidental conditioning then the shaping process really gets underway. I think this can best be explained by an example. If a rat was in an experimental chamber and the food hopper was programmed to rise, on the average, every twenty seconds, this might give the rat time enough after each reinforcement to get tired of standing by the food hopper waiting for more food. It might turn its back to the source of food and begin walking away. If walking away from the food hopper was the response occurring when next the food hopper was raised, the response of walking away from the food hopper would be reinforced. If this occurred several times the rat's holistic analysis might conclude that walking away from the food hopper was indeed related to receiving food. After walking away from the food hopper was established as a response of high probability there would be the opportunity for this response to be rewarded, still accidentally, at different distances from the wall with the food hopper. If several times in a row the rat was reinforced after having walked further from the food source then walking a greater distance would be the response that increased.
It wouldn't take many accidental reinforcements of this response until the rat had reached the far wall of the experimental chamber. This would provide a physical limit to the response of walking away from the food hopper. If when the rat, upon reaching the far wall, came to a stop and stood still then this behaviour would likely be occurring when next the reinforcement took place. The behaviour of standing still by the far wall might very well then become firmly established by further accidental reinforcement. If you looked in the experimental chamber you would see the rat standing very still by the wall opposite the food hopper. Each time the food hopper was raised the rat would run over, eat until the food hopper dropped out of reach, then run back to the far wall to stand still until the next time the reinforcing stimulus was presented. The rat's holistic analysis would interpret these environmental circumstances as sufficient proof that standing still by the far wall was the way to obtain food. It would be considered desirable to spend as much time as possible standing still by the far wall. With no actual connection between the food presentation and any response the random reinforcement conditioned one behaviour, walking away from the food source, to occur at a high rate; then, using that response as a baseline condition, shaped walking away from the food hopper until it became a very different response, standing still by the far wall.
PHILOSOPHER-SCIENTIST: Did you make that example up?
TINNY: No, it was the actual result of an experiment with accidental conditioning. That same random reinforcement procedure was tried a number of times. Some form of superstitious behaviour invariable developed each time.
PHILOSOPHER-SCIENTIST: What other kind of behaviour did the experimental animals end up doing?
TINNY: Common types of superstitious behaviour conditioned were swaying back and forth, hopping up and down, and lying down in certain positions. One rat ended up lying on its back with its legs in the air like it was playing dead. As soon as the food hopper was raised it would jump up, eat until the food hopper was dropped, then lie back down and put its legs up in the air.
PHILOSOPHER-SCIENTIST: For any of those behaviours to be conditioned by food being randomly presented the rats would have to be hungry wouldn't they?
TINNY: If the rats weren't hungry the food wouldn't have been a positive stimulus, so therefore would not have been able to act as a reinforcing stimulus. Hunger can be considered as a deprived condition. The opposite of deprivation is called satiation. Being hungry is a state of deprivation and being full is a state of satiation. To a deprived animal, food is a positive stimulus; but to a satiated animal food becomes either a neutral or an aversive stimulus.
PHILOSOPHER-SCIENTIST: Since the subject during the conditioning process must be in a state of deprivation, doesn't that place quite a limitation on the effective use of these procedures?
TINNY: There are ways of getting around that limitation. With animals of higher consciousness, like monkeys and with human beings, conditioning can be successfully achieved without the use of any primary reinforcers. A special type of secondary reinforcer can be used in place of the primary reinforcer.
PHILOSOPHER-SCIENTIST: Didn't you say that secondary reinforcers soon lose their reinforcing ability when not paired with the primary stimulus?
TINNY: They soon do lose their reinforcing power except for this one special class of secondary reinforcers. They are called generalised secondary reinforcers. This means they can represent a large number of primary reinforcers. There is a simpler name for these generalised secondary reinforcers. They are also called tokens. A token is some item which may have in itself no value as a primary stimulus, but which can be traded in for primary reinforcers. The token can be given as a reward for the response being conditioned and this will result in the increased probability of that response occurring in the future. The effect of the token is the same as the effect of a primary reinforcer. The reinforcing value of a token remains even if there is a long time delay until the tokens can be traded for primary reinforcers.
PHILOSOPHER-SCIENTIST: Would you give an example of tokens being used in a conditioning procedure?
TINNY: In training monkeys, wooden discs have been given after various responses to serve as reinforcers. First, of course, the monkey has to learn that the token can be traded for some primary reinforcer such as food or a drink. Monkeys will then work to get the tokens just as they would to get food or drink. The monkeys perform responses which get them tokens, save up these tokens, and later turn the tokens in for primary reinforcers. In one experiment I read about, each token could be traded in for a slice of banana.
PHILOSOPHER-SCIENTIST: That description of the use of wooden discs as tokens sounds somehow familiar.
TINNY: Tokens are widely used in human society. We call our tokens money. Money has no real value of its own. The value of money comes from what it can be traded for. We learn as children that money can be traded for primary reinforcers. Money is very influential in conditioning human behaviour. If money could no longer be traded for primary reinforcers it would lose its ability to affect the future occurrence of the responses it follows. Money would then have no value.
PHILOSOPHER-SCIENTIST: There can be no doubt as to the effectiveness of tokens as reinforcing stimuli. Some people will do anything for money.
TINNY: Money has become such a powerful generalised secondary reinforcer that it has practically become a primary reinforcer in its own right.
PHILOSOPHER-SCIENTIST: Is that a good thing?
TINNY: I think it is quite harmful. The stimuli which are primary reinforcers have been determined by the natural order. Money's reinforcing power is artificial. This seems to bring about problems in humanity's relationship with money. This is a relationship without natural harmony.
PHILOSOPHER-SCIENTIST: Perhaps that which is natural is always better than that which is artificial.
TINNY: Perhaps.
PHILOSOPHER-SCIENTIST: As you were explaining accidental conditioning you said that as the rate of one response increases, the rate of other responses decrease. Why is that?
TINNY: In any given period of time only so many responses can take place. If one response is to increase, other responses must decrease to allow for that increase. In some cases when one response increases certain other responses must decrease because they are incompatible with the reinforced response.
PHILOSOPHER-SCIENTIST: Would you give me an example of incompatible behaviour?
TINNY: When I was training Aphrodite to turn to the right is a good example. As she increased turning right, left turns decreased. Turning left is obviously incompatible with turning right. If a dog is being trained to sit, then responses such as standing, running, and jumping must decrease since they are all incompatible with sitting.
PHILOSOPHER-SCIENTIST: Is all of this knowledge necessary to bring about positive change in human society?
TINNY: Every bit of it is important. Every child should know all this and more at a very early age. These things we are discussing now form the basis of a much more sophisticated understanding of learning at the human level.
PHILOSOPHER-SCIENTIST: What should I hear next about the laws of learning?
TINNY: Our discussion has so far involved the most simple learning situations. Learning a new body of knowledge occurs by means of a shaping process. To be most successful this learning should take place through a series of rewarded small steps. Each of the steps can be thought of as a link in the chain of complete knowledge. Each step should be successfully learned before moving on to the next step. This is so the chain doesn't have weak links. Also no step should be missed or the result will be several separate chains, less functional than one complete chain.
PHILOSOPHER-SCIENTIST: I don't want you to rush through your explanation of learning principles. I would rather get too much detail than too little.
TINNY: So far we have been discussing the conditioning of one simple response. Much conditioning in real life entails different responses depending on the circumstances.
PHILOSOPHER-SCIENTIST: Is there a name for the conditioning of different responses depending on the circumstances?
TINNY: All of these different aspects of the laws of learning have names. Not all of the names are particularly useful. Since these names have been used mainly in the scientific world, the language to describe them often tends to be rather technical. The basic concepts are much more important than the names. When these concepts become widely known a more simple language will develop naturally to describe them.
PHILOSOPHER-SCIENTIST: Until that time we can use the technical terms, but remain ready to accept the new language when it develops.
TINNY: In that case there are several names that can be used to describe the conditioning of different responses depending on the circumstances. That process could be called differential reinforcement, or it could be called discrimination training.
PHILOSOPHER-SCIENTIST: Give me a simple example of discrimination training.
TINNY: Let's go back to an earlier example when a rat was conditioned to press a lever at a high rate by following each lever press with a positive reinforcer, a food pellet. That example can be changed very easily to one of discrimination training. If you'll remember, when I described an experimental learning chamber I said there was a light above the lever. We could have the light on or the light off. When the light is on pressing the lever will result in the food hopper being raised. When the light is off pressing the lever has no effect on the food hopper, it stays down out of reach. Reinforcement can occur when the light is on. No reinforcement can occur when the light is off. If those are the contingencies in effect the rat would soon learn to press the lever at a high rate when the light was on, and stop pressing the lever when the light was off. The rat discriminates between the two different learning situations.
PHILOSOPHER-SCIENTIST: What might be going on in the holistic analysis while this discrimination training is taking place?
TINNY: Before the discrimination training took place the rat had already been conditioned to press the lever at a high rate. At this point the holistic analysis had become quite certain of the relationship between the response of lever pressing and the positive stimulus of receiving food. As long as the rat was hungry holistic analysis would advise pressing of the lever. When the new contingencies came into effect, holistic analysis must reassess the relationship between lever pressing and food. Now sometimes pressing the lever isn't followed by food. A function of holistic analysis is to make the most efficient use of all energy expended. It is a legitimate expenditure of energy to press the lever when it brings food, but it is a waste of energy when no food is forthcoming. The holistic analysis would like to know when to advise pressing the lever, and when to advise not pressing the lever. The question is, how to know which is appropriate. From past experience holistic analysis has learned that often there are cues in the environment which signal appropriate behaviour. While pressing the lever the rat perceives a light in its environment. This light is sometimes on, and sometimes off. Holistic analysis becomes aware of this information. This information is of no use at first because no connection involving the light has initially been made with the presence or absence of the reinforcing stimulus after lever pressing; but, soon it becomes clear, the light is always on when lever pressing is followed by food, and the light is always off when lever pressing is not followed by food. When this connection has been made by the holistic analysis the realisation takes place that it is possible to predict accurately whether or not lever pressing is worthwhile. When the light is on lever pressing is an efficient use of energy, so holistic analysis advises responding. When the light is off lever pressing would be an inefficient use of energy, so responding is not advisable.
PHILOSOPHER-SCIENTIST: The holistic analysis doesn't miss much, does it?
TINNY: It misses nothing, but it can be fooled.
PHILOSOPHER-SCIENTIST: How can holistic analysis be fooled?
TINNY: Accidental conditioning is an example of holistic analysis being misled. This happens because holistic analysis while being very logical is also very trusting. It accepts information at face value.
PHILOSOPHER-SCIENTIST: Isn't that naive?
TINNY: Holistic analysis which takes place other than at the human level has good reason to be trusting. All material existence from the simplest form to the most highly developed animal lives in a harmonious relationship with the natural order. The natural order very seldom misleads. The natural order is essentially logical and truthful. Holistic analysis envolved to deal with logical and truthful information. The human species does not live in a harmonious relationship with the natural order. Human beings are not always logical and truthful. Holistic analysis at the human level, having developed to deal with logic and truth, finds it sometimes difficult to cope with information that is not logical or truthful.
PHILOSOPHER-SCIENTIST: The name discrimination training must refer to the fact that this type of conditioning allows the subject to discriminate between different environmental conditions. For example, when lever pressing would be reinforced and when lever pressing would not be reinforced. The light allowed the rat to discriminate between those two different learning situations.
TINNY: The light, which provides information as to what behaviour is appropriate in the particular circumstances, is another type of stimulus. A stimulus of this type is called a discriminative stimulus. A discriminative stimulus provides information as to whether or not a response is likely to be reinforced. It could also provide information as to whether or not a response is likely to be punished. Discriminative stimuli give cues which allow the contingencies operating in the conditioning process to be known.
PHILOSOPHER-SCIENTIST: Are discriminative stimuli common in everyday life?
TINNY: Much of our behaviour, both mental and physical, occurs below the level of conscious awareness. Actually every response is preceded by some stimulus, and every response is followed by some stimulus. Any preceding stimulus may serve as a discriminative stimulus, and any following stimulus may serve as a reinforcing or punishing stimulus. Whether a stimulus is discriminative, reinforcing, or punishing is determined by the function it fulfils in those particular circumstances. It is often the case that a single stimulus may function as a discriminative stimulus and a reinforcing stimulus at the same time.
PHILOSOPHER-SCIENTIST: How could that happen? You said a discriminative stimulus comes before the response and the reinforcing stimulus comes after the response.
TINNY: All behaviours are actually chains of responses. In these chains each response is followed by some stimulus, perhaps a reinforcer. After that reinforcing stimulus another response occurs. In relation to the first response the stimulus following it functioned as a reinforcing stimulus. In relation to the second response the stimulus that preceded it functioned as a discriminative stimulus. The stimulus which followed the first response and preceded the second response was the same stimulus.
