page 1 | page 2
Rats, of course, cannot tell us how they think. But by watching how they attempt to control effects by manipulating potential causes, we can draw conclusions about their perception of causality. And, as it turns out, rats, as in the situation above, behave in ways that are completely analogous to humans: both can arrive at causality without trial-and-error training. Because coming up with different explanations for such analogous behavior just doesn’t fly from a scientific perspective, Miller said he is forced to conclude either that rational thought has little to do with rat or human behavior, or that rational thought processes are also widespread among non-human species.
Seen through the lens of his theory, results such as these begin to explain how something that might appear to involve a rational thought process is actually the result of purely mechanistic processes. Miller’s opinion is that the rats aren’t “reasoning out” the connection between the lever and the food, but are instead quite mechanistically calling upon temporal information from two independent learning episodes to create a meaningful association between events that were not directly paired.
Unlike most theories of associative learning, temporal coding theory also maintains a clear distinction between learning and the behavioral expression of that learning. Miller believes, in fact, that learning itself is ultimately an automatic process. He likens the mind to a garbage pail.
“Everything goes into it, and the real questions are about the conditions under which we are going to be able to retrieve and make use of things,” he said.
“The problem is not so much about what you learn, but rather, what you can retrieve and how you use the information.” Again, our mathematically challenged student might have “learned” the times tables just fine, thank you, but for a variety of reasons might not be able to access that information.
Miller’s research also distinguishes between information processing that occurs during training and processing that occurs at test. His findings here contradict and reverse what he describes as the general view in academia that initial learning is a complex chore for an organism and that once something is learned, translating that information into behavior later on is relatively trivial. He thinks instead that learning is relatively simple, and that complexity arises in retrieving and using that which has been “encoded.”
“The conditions of testing are absolutely crucial to what is going to be retrieved,” Miller says, “with much better recall produced by matching conditions of retrieval to conditions of training.
“This is fine,” he adds, “if you can control conditions of retrieval as you can in a laboratory. But in the real world, we don’t control them to any great extent. Hence, we should try to match the conditions of training to the conditions of retrieval, as imposed by real-world situations.”
In successful 19th-century classrooms, Miller said, the focus was
on teaching students by introducing them to
problems drawn from the real-world experiences they would encounter
when they left school. A math test, for instance, might ask them
to determine, based on the measurements of a wagon, the optimal
size of hay bales to be transported in it.
Today, teaching often takes a more abstract approach, but Miller’s research suggests that educators who focus on trying to teach students to “think” might be misguided. There might be a small, highly elite part of the population that can benefit from training with abstractions, he said, but the vast majority of people learn better when training is specific to a real application.
“We tend not to generalize from how to solve one task to being able to solve parallel tasks that are superficially quite different,” he said.
Teaching to the test, which is another common approach these days, could, according to Miller’s research, be a better strategy — but only if the tests are based on real-world problems. Medical and business schools that are increasing the use of case studies to train students by exposing them to real-life problems and conditions are, according to his research, on the right track.
page 1 | page 2