Some cognitive scientists argue that human intelligence is similar to a computer. The intelligence is the programs (software) running on the hardware of the brain. We'll call this approach the "symbolic approach." Later you'll understand what "symbol" has to do with it.
In the symbolic approach an important distinction is made between the mind (cognition, thought, intelligence, etc.) and the brain. These people believe that this distinction is like the distinction between the software and hardware of a computer. They believe that it is possible to describe cognition in a form for which the hardware doesn't matter. Just as you can write a program in a high-level programming language like C, Pascal, or Lisp, and run it on any computer, they believe that we can write a program which simulates the mind and not worry about whether it runs on a computer or a brain. (Of course these people don't actually run their software on somebody's brain. They are just claiming that their computer programs are the same sort of thing that we have running in our brains when we think.)
There is another, competing approach in cognitive science, one that argues that the hardware matters. The brain is very different from a computer. In particular, it is parallel; that is, unlike a computer, it can do many things at the same time. And, unlike a computer, it has no central control, no place that gives instructions to all of the other places. On this view of cognition, you can't ignore the fact that the mind exists in a brain. You can't simply write an intelligent program which is designed to run efficiently on a computer and assume that human intelligence works the same way. We will refer to this alternative way of thinking about cognition as the "distributed approach." Again we'll motivate the term later on.
Note, however, that this does not mean that the people working in the distributed framework cannot write computer programs to test their models. Any process which be described precisely as a set of steps can be simulated on a computer. It's just that these programs, because they are parallel and lack a central control, are somewhat awkward for computers, and we need to use tricks to make computers appear to be running them. We'll learn more about this later.
In sum, for the "symbolic" people, the difference between the hardware (for a computer, the chips, the hard disk, etc., for a person, the brain and, to a lesser extent, the rest of the body) and the software (for a computer, the programs that run on the hardware, for a person, the thoughts which happen in the brain). For these people, we don't have to bother about the hardware in order to understand the software. For the "distributed" people, on the other hand, the hardware matters. Brains are different from computers, so the kind of programs that are most efficient for running on computers are not likely to be the best for brains, and vice versa.
Some cognitive scientists, especially people in the "symbolic" camp, tend to believe that if we look carefully at the behavior of adults, we can figure out what sort of knowledge they have that allows them to be intelligent. These people focus on the product of learning, the end-state, and they write computer programs which they believe characterize the end-state. They are not so interested in the process that got the person to that state.
Others, especially people in the "distributed" camp, believe that it is possible to simply describe in any concise way the knowledge that we have as adults. We gained that knowledge through years of experience, and we need to include that experience in our explanations. That is, the models need to be learning models. These people are interested in the process that gets people to the state where they can do intelligent things.
In the early days of cognitive science, researchers focused on tasks such as playing chess and translating from one language to another. While some progress was made, some people felt that we would not really begin to understand complex abilities if we did not understand the simpler abilities that they are based on. This idea has led not only to an interest in explaining the intelligence of children but also in explaining the intelligence of other animals, even insects. For example, the way in which an insect walks turns out to be extremely difficult to understand, and there are researchers who focus entirely on the "intelligence" that allows a cockroach to scurry across a kitchen floor without being stepped on.
Nowadays problems that were once considered easy may be taken very seriously. Some aspects of rhythm belong to this category.