Animal Intelligence Essay, Research Paper
Philip Weiss
11/22/00
Animal Intelligence
“Animal Intelligence”
The learning theory of Thorndike represents the original S-R framework of behavioral psychology: Learning is the result of associations forming between stimuli and responses. Such associations or “habits” become strengthened or weakened by the nature and frequency of the S-R pairings. The classic example of Thorndike’s S-R theory was a cat learning to escape from a “puzzle box” by pressing a lever inside the box. After much trial and error behavior, the cat learns to associate pressing the lever (S) with opening the door (R). This S-R connection is established because it results in a satisfying state of affairs (escape from the box).
In the article “Animal Intelligence” Edward Lee Thorndike states that the research that has been done on animal behavior was not sufficient. They only relied on observations and collected anecdotes. In George Romanes “Animal Intelligence” he gathered observations and anecdotes from observers around the world to put together his theory on animal intelligence. According to his theory, the animal possesses a fairly high level of intelligence. Thorndike did not agree with this and the method of research that Romanes used. Thorndike did not accept the fact that if there is milk inside the house, the cat comes every day to get it. He wanted to know what the animal is thinking when he comes. What is going on in his mind?
Thorndike’s main purpose in studying animal intelligence is to trace back the origins of humans in general.
Thorndike felt that in order to avoid all these problems it would be more useful to do experiments on the animal rather than observations and collection of anecdotes. Through experiments we are able to have control of the stimuli in the environment and we are able to repeat the procedure more than once to get a better understanding of the occurrences during each trial. We are therefore also able to see whether the behavior of the animal is a coincidence, if he learns anything new over each trial or he stays in the same state as when he began.
To prove this he did an experiment with a wooden box that had a rope that would open a door that led to food. A cat was starved and then placed inside the box to see how he would react to the stimuli presented to him and if he would figure out how to get to the food. During the first trial the cat reacted with very frantic movements. He clawed and bit every hole available to him, he banged his head against the roof of the box. After about eight to ten minutes of this behavior the cat fell by accident on the lever , the door opened and the cat ran to the food. According to Romane’s theory, since the cat has a fairly high intelligence level he should have learned for the second trial to pull on the cord immediately when placed in the box. Thorndike found that this was not so. The second time the cat was placed in the box, it went through the same processes that it went through the first time. It scratched and bit all the possible holes in the box. Then after a while it fell on the lever that opened the door to the food. The thing that Thorndike realized is that the amount of time that it took for the cat to go through the rituals of biting, clawing, banging his head on the top of the cage and then falling on the lever decreased with each trial until it reached a point where it would do it almost immediately after it was placed in the box. Thorndike’s theory is that during each trial there is an association building up in the animals mind. Each trial that association becomes stronger and stronger until it gets stamped in the animals head exactly what he has to go through to get to the stage of pressing the lever but the animal does not sit there and contemplate how he is going to pull of a great escape from the box. The pleasure that the animal receives when he presses the lever stamps in an association between the impulse of the animal to respond to the stimuli and his sense impressions that accompany it. These are the mental events that occur in the animals mind during the experiments. He adds that the more the response is rewarding the more it gets stamped in until he responds automatically whenever his sensations are aroused.
In all of Thorndike’s experiments he never observed immediate improvement in the animals response. The animal never knew right away where the food was and how to get it. Every time it took a long time and many trials until it was stamped in the cat’s head. Latencies, however did sometimes drop significantly after a recent success but after a ser
Another problem Thorndike has with others’ observations of animals is that they are not based on the psychology of animals rather on an abnormal event or a great feat that one animal out of his whole species is able to achieve. For instance, millions of cats get stuck in trees and can’t get down and nobody blinks an eye but one cat saves its master by waking it up during a blazing fire and all of the sudden cats become the highest form of intelligence and the cat gets his picture on the front of every local newspaper. There is also a bias involved. The ones who observe animals anticipate that there is a great deal of intelligence in the animal especially if they own it or know the owners personally and as a result will report faulty and inaccurate information about his observations.
J.B Watson who is a behaviorist argues with Thorndike on how to analyze an animal during each trial. He believes that there is no way to measure an animal’s behavior based on its mental processes because we don’t have accurate information on the subject. He states that the way we tend to determine what the animal is felling during the experiments are based on how we think we would feel if put in that same situation. We must then look at actual behaviors of the animal in conjunction with hereditary influence to judge a certain behavior rather than base it on cognitive processes that we are not able to see or test.
Thorndike stated that the animal learns to associate certain behaviors with the satisfaction of succeeding such that these behaviors become more and more likely to occur. He called this “Instrumental Learning”. The animal learns to produce an instrumental response that will lead it to “satisfaction”.
There are many controversial issues in regard to some of the terms he used in his writings that were not scientifically accepted because were not words that could be measured. One of the controversial issues is that Thorndike used the term satisfaction, which refers to a mental state in the animal and there is no way for us to test mental states and therefore is not scientific. A second controversy with Thorndike’s puzzle box is that the relation between the lever and the food was not visible to the cat most of the time. An Austrian psychologists by the name of Kohler believed that Thorndike’s experiment was too complicated for the cat to learn the connection between the CS(lever) and the US(food). He did a more simplistic experiment to show learning processes. He put a monkey in a cage and outside the cage he put a bunch of bananas and a stick. All the monkey had to do was to grab the stick and bring the bananas to him. This does not disprove Thorndike’s method it just introduces a second way of looking at learning.
I believe that Thorndike’s experiment of observing a cat’s learning abilities and responses, in an unfamiliar environment, were he would have to find the rewarding stimuli, added a great deal to the advancement of behavioral psychology. Through the experiment with the cat in the box, he was able to show to a certain extent that animals use a lower level of thinking than humans do. When he placed the cat in the box in each trial the cat did not sit there and map out how he is going to escape from the box rather he acted wildly and went through the same “rituals” of biting, clawing and hitting his head on the top of the cage as he did in the first trial until he fell on the necessary stimuli (the lever) that would give him the stimuli that is rewarding to him (the food), which shows that the animal did not use his cognitive processes rather he resorted to the lower level of thinking which is association. I think that there are pros and cons to using experiments as opposed to observations. The advantage putting a subject in a laboratory environment is that he will not act and react the same to stimuli presented to him as in his natural environment. On the other hand an experiment is most effective when you are able to control and isolate as many stimuli in the environment as possible and by doing an experiment in a controlled environment, even though you are never able to control all of the stimuli around, you have more control than in a subjects natural environment. I agree with Watson’s criticism on Thorndike’s method because there is no real way to scientifically prove something that you are not able to assess and analyze. I have learned more in depth on Thorndike’s “Law of Effect” which states “Responses which have satisfying consequences are stamped in and those followed by discomfort or annoyance are stamped out” but this is still inconclusive.