Stephen van Vlack

Sookmyung Women`s University

Graduate School of TESOL

Human Learning and Cognition

Spring 2006

Week 2 - Answers - Terry, Chapter 1 & Lamb, Chapter 1 

1. What is learning? (T)

There is no simple definition for something as complex as learning. Obviously any definition or conceptualization of learning is going to be tightly bound by the idea of knowledge, for the intake of new knowledge would indeed seem to the central core of learning, but this kind of process is impossible to observe and very hard to measure with any kind of accuracy as well. What also makes it even harder to measure is that there are also different types of knowledge. Knowledge can be represented by or take the form of specific facts or skills. These two are actually quite different particularly in how they are going to affect behaviour.

Knowledge can only be indirectly observed and this is often done through overt behaviour. So if we are thinking about knowledge of a skill like bicycle riding for example even though it is obvious that somebody has learned certain things in order to ride a bike it is not necessarily very clear exactly what they have learned by watching them ride the bike. Put in another way, the specific bits of knowledge required to ride a bike might not be directly manifested in the actual riding all the bike. Researchers there for often need to guess what might be underlying causes of particular types of behaviour.

It might at this point be important to understand that behaviour can also be separated into involuntary and voluntary or controlled behaviour. For psychologists, therefore, the clearest indication of learning is a change in behaviour. To extend this simple idea, it is also readily understandable too that for learning to have occurred the change should be permanent or at least be held for a measurable period of time. Following this, it would seem fair to say that learning must really occur in what we will come to know as long-term memory.

As teachers of language the important thing for us to remember is that learning is supposed to result in a change in behaviour. This basically means that the learned material is supposed to allow the learner to behave in a new way. They should be able to do things as a result of the learning that they were not able to do before. This is key. Language in involves doing not just reciting what might be done or evaluating the correctness of something that has been done. These are also certain types of decision-making behaviour but really what I comes down to is the ability to actually do new and novel things.

2. How are learning and performance related and why do we need to be aware of this? (T)

As was stated above, the idea of learning is centred around the acquisition of knowledge, but this is problematic because it is definitely possible to have knowledge which does in no way affect behaviour or does not do so in any observable way, at least not the time frame which is practical for observation and testing. For us as language teachers this is a particularly important point to keep in mind. What we are trying to do is not inject bits of knowledge into our students heads simply for the point of having them know them. Ultimately, we are trying to help them be able to establish a whole new repertoire of behaviours. We want them to be able to do things as a result of the learning which goes on in and out of our class.

At the same time we also need to be aware that there is such a thing as latent knowledge. This is the knowledge that sits in someone`s head and stews there seemingly dormant until it is somehow needed and made use of. At this point it kicks into life and starts affecting behaviour on a large scale. From this we can derive the idea that learning and the behavioural idea of learning is closely tied with experience. Without relevant experience or opportunities learning is unnecessary and even if it occurs it may never manifest itself. All learning, all that is being learned must, therefore, be tied to experience in the real world; that is, interaction with and in the real world.

3. What is the relationship between learning and memory? (T)

The relationship between learning and memory is an obvious one. In order to say you have learned something it must have been committed in some way, shape, or form to some kind of memory system. This means that memory is central for learning and this simple statement opens up the floodgates of memory research. Things committed to memory are supposed to be somehow measurable and that is where the entire field of testing came from. Tests are there to measure how well a group or people or animals have committed to memory information presented or encountered. Such research has out of necessity turned to the actual construct of memory and how different types of memories work and interact in allowing us to be able to learn information. Unfortunately, there are many mitigating forces that affect memory and learning. Following this, the study of learning has become a rather complicated field indeed and we will be spending really the rest of the semester answering this seemingly simple question.

4. What are some of the main approaches to studying learning and which seems to be the most promising? (T)

In the book Terry mentions three main approaches to the question of how to research the phenomenon of learning. Here they are presented in their order of conception: the behavioral approach, the cognitive approach, and the neuropsychological approach (following what we are also reading and lamb we might be better put to referring to this same approach as a neurocognitive approach). Behaviourism looks at behaviour without really trying to figure out the internal mechanisms which really control that behaviour. It looks for the external causes of behaviour, such as getting a smack or the wind blowing in your face but it really doesn`t look at all about what is going on inside the brain which tempers such behaviour and allows it to recur. From our description of learning above we see that, yes, behaviour is important but it is also extremely important to know not just why certain behaviour will occur on the surface but really we need to see what’s going on inside the head or the brains of people as they behave. The cognitive approach, and here we might be better off renaming it the analytical approach (again following terminology in Lamb, 1999)), tries to create mental representations which will explain on the inside behaviour. It is analytic because it does not examine the behaviour itself as being very important but seeks merely to use the behaviour to come up with an internalized system which may explain where the behaviour is derived from. Of course the universal grammar model of Chomsky is a perfect example of an analytical approach to learning and behaviour. The problem with this analytical approach is twofold. First of all analytical approaches argue for a strict delineation between behaviour itself and the underlying system which allows behaviour to be created. It is way too absolute and that there is no real feedback loop between behaviour and the system which creates it. This is a problem because it does not allow for the internal representational system to change. The second major problem of the analytical approach is that often the model which is created to describe certain behaviour has absolutely no psychological or cognitive reality. What this basically means is that these models are not capable of being supported by the brain based on what we know currently about how the brain functions. In effect while such representational models might seemingly work to describe how behaviour is generated they do not accurately describe how this would occur in the brain. Since the brain is where this actually does occur than this is a major problem.

While both the behaviorist and the analytical models are promising, they also need to be tempered with the last and most recently developed approach. The truth of the mater is that neuropsychological/neurocognitive data is necessary to prove any of the concepts presented in either of the first two approaches. Any knowledge we have all of behavior or systems needs to be tempered by and explain with current understanding of how the brain works. This seems so obvious but sadly it is generally not done. And that is why we have to take this class.

5. What is parallel distributed processing and why does it seem to be better than previous models of cognitive processing? (L)

The parallel processing model is a kind of connectionist model which allows multiple operations to occur simultaneously. We can contrast this type of processing model with what are called serial processing models which claim that operations must occur in a particular order. Let us now look at a very simple example.

(1) The doors opened and Mary was all set to step into the e________`s interior.

In the above example we are focusing our attention on the incomplete word e_______. The question is how do we process language to determine what units are being uttered or, as in the case of writing, encountered. In a serial model we would need to go through and hear every individual phoneme before being able to determine what exactly the word is and it is not until we have gotten the entire representation of the word that we are then able to access the possible meaning of this word. In short we need to go through a fairly long and tedious process which focuses on one specific type of information at a time. In a parallel processing type of model we use every different possible system simultaneously in order to as quickly and as accurately as possible access the necessary information about the item in question. Most of us will be able to guess that the missing word above is in fact elevator and probably not elephant. We can do this because we have used other knowledge we know of that other systems within the language to determine what is real and logical in this type of linguistic situation. We don`t need to find all the different phonemes to be able to process it. In parallel processing we can fill in the gaps quickly and relatively easily using all available systems and knowledge. So in this case we readily are able to understand that the target item is elevator partially because we know that elevators have doors and that they are open spaces with an interior while elephants may have doors leading to their access they are not empty on the inside and even if they were you probably would not want to walk inside one. We need to use this knowledge to figure these things out.

Following is we can see how effective a parallel distributed processing model will be. The advantage of this model is that it allows different types of linguistic systems to be operating simultaneously. Since we know that language is indeed composed of many different systems but that at the same time all the systems overlap and work together to form what we know his language then we certainly need a model which allows us to do this. The other advantage of this parallel distributed processing model is that it is psychologically real. Parallel distributed processing models were originally based on what are called in your own network computers and were originally built by and investigated by computer scientists. The underlying construct of the parallel distributed processing model is a simplified version of the brain wherein certain nodes or nections are connected together in a special type of computer in much the same way that neurons are connected together in the brain. The neural network learns information not by storing it in holding it but through connections to the other nodes or nections. It is psychologically real and fits well into what we know about the complex nature and multifaceted nature of the different systems of language.

6. Why is a focus on people so important in any investigation of language? (L)

In the behaviorist model researchers focused not on people but on their behavior the assumption being that there were certain universal constraints on behavior. All living creatures, from sea slugs right up to human beings all behave in similar ways based on certain basic principles. Two simple.

In the analytic model researchers focus only on the language which people produce. They don`t think it all about the actual people that are producing the language because again the belief is that by looking carefully at this particular language the universals of this underlying representational system can be derived. In this model individual people are not at all interesting.

In the neurocognitive model focussing on people as a way of investigating language is important because it impels us to focus on brains and not just bits of linguistic evidence. Since language is created in the brains of speakers and interpreted in their brains as well then it makes sense to focus on people and their brains. We want to know what goes in there to allow them to DO such things. The other important thing to remember and a whole fleet you learn this from human to reading is that everybody`s brain is different. It is very important that we embrace this idea as we try to not only learn about learning but teach our students or help our students to learn about our subject, English. It is only by looking at people and their brains that we will be able to finally pieced together the answers to very difficult questions about language and language learning.

7. What are the four kinds of evidence we need for a linguistic investigation and why do we need all of them? (L)

Lamb mentions for different kinds of evidence that researchers need for a linguistic investigation. They are:

First Base - speech mechanisms

Second Base - the linguistic text produced by people

Third Base - How speech is created, understood, and learned

Fourth Base - How the brain encodes and produced all the above mentioned points

It is important to remember that each of these comprises a different system and while it is obvious that the systems somehow work the same way, under the same set of principles imposed by the brain, they are also going to deal with different types of information. They necessarily interact and we need to see how this can happen. In this class, since this is especially not an introduction to linguistics class, we going to be focusing for the most part on the fourth base.

8. What is transparency and why does Lamb say it is an illusion? (L)

The basic idea behind transparency and the brain is that we cannot figure out what is inside the brain by carefully studying what comes out or what comes in. Unfortunately this is exactly what people have been doing for a very long time. Simple assumption is that because the language that comes out is composed of words and specific phonemes or sounds then these must also be inside the brain. This is a basic idea that the brain is transparent and it is dead wrong. This idea of the brain as transparent has lead to some seriously mistaken ideas about what the building blocks of language might be and how language may work, but it is still the model that almost all linguists follow.

Lamb`s challenge, and sadly it really is a challenge because it runs so much against current linguistic thought, is that we get rid of the idea that the brain is transparent and can be judged simply by what comes out of it. This means that we need to actually come up with models of language which function according to what we know about the brain internally. Thinking that this we can already see that things like word meaning are really an illusion of language use and are certainly not stored in the brain ib the way we might think they are. This has already been widely acknowledged by certain linguists. More difficult still, however, will be abandoning the illusion of such steadfast constructs of linguistics like the mental lexicon and words and phonemes. As I briefly mentioned may be one of the best ways of dealing with this according to the brain would be to really push the idea of convergence zones as a central part of linguistic processing.

If this all seems extremely vague and fuzzy at this point hopefully it will become clearer as we delve deeper into such questions both in this class and later in the Introduction to Linguistics class.