Cognitivism Research Paper ⋆ Research Paper Examples ⋆ EssayEmpire

This sample Cognitivism Research Paper is published for educational and informational purposes only. Free research papers are not written by our writers, they are contributed by users, so we are not responsible for the content of this free sample paper. If you want to buy a high quality research paper on any topic at affordable price please use custom research paper writing services.

Abstract

In the broad sense of the term, cognitivism encompasses all those theoretical approaches that consider the study of the human mind and its constituent processes as the main objective of psychological science. In this sense, psychology, as an experimental enterprise, was cognitive from its very beginning. For example, W. Wundt considered the task of psychology to be the analysis of conscious processes into its constituent elements, and many chapters in William James’ book The Principles of Psychology were dedicated to analyze processes such as attention, memory, and consciousness, the core cognitive processes. In a more strict sense, cognitivism refers to a particular theoretical approach to studying cognitive processes that originated as a reaction against behaviorism in the midst of the 20th century and is known as the information-processing approach. This research paper primarily concentrates on the origin and subsequent development of this approach, though attention is also paid to other cognitive views influencing psychological research.

Outline

Cognitivism During the Behaviorist Era
The Information-Processing Approach
Cognitive Science
Tenets of Cognitivism
Cognitivist Influence

1. Cognitivism During The Behaviorist Era

From a cognitive point of view, the contribution of behaviorism to our knowledge of psychological processes was almost restricted to the realms of learning, motivation, and emotion. Little if any concern was shown about cognitive processes such as perception, attention, memory, and consciousness in the writings of Hull or Skinner. Even when they made a foray into the territories of thought (C. Hull) or language (B. Skinner), their nonrepresentational approach turned out to be too restricted and had little influence on subsequent research. Among behaviorists, E. Tolman was one of the few who did not show disdain for using terms and concepts with a cognitive flavor. Even in his case, however, those terms did not refer to processes to be studied for their own interest; rather, they were conceptualized as intervening variables useful to predict behavior.

It should be mentioned, however, that behaviorism did provide a lasting contribution to cognitive and psychological research in general. Emphasis on the measurement of variables, the careful design of experiments, and the wide-ranging use of objective scientific methods is part and parcel of the invaluable legacy of behaviorism. Thanks to this methodological heritage, modern cognitivism could establish itself on a solid scientific foothold.

Despite the general sway of behaviorism during the second quarter of the 20th century in the United States, interest in cognitive research remained active within restricted domains of psychological inquiry. This was particularly so in Europe, where the influence of behaviorism took longer to be felt and was never as strong as it was in the United States. The following streams of investigation deserve to be pointed out:

The contribution of Gestalt psychologists to the development of perceptual theory.
Research on concept formation, problem solving, and thinking by researchers such as Edouard Clapare` de, Wolfgang Ko¨ hler, and Frederic Bartlett.
The sociohistorical approach to cognitive development carried out in Russia by Lev Semenovich Vygotsky.
Jean Piaget’s research program on cognitive development and genetic epistemology in Switzerland. Piaget was one of the first thinkers to realize the interdisciplinary nature of cognitive research. In 1955 he founded the International Center for Genetic Epistemology in Geneva, where scientists from different disciplines tried to set up the foundations of a true cognitive science. Piaget influence on cognitive development and educational psychology is still noticeable today; research along his line of thinking is an integral part of contemporary cognitive research.
It is important to recognize that, in some areas of psychology, behaviorism never played the dominant role it did within experimental psychology. References to mind and mentalistic concepts never disappeared from social and clinical psychology. Authors such as Kurt Lewin, Leon Festinger, Erich Fromm, Carl Rogers, and Abraham Maslow are only a few among the many who did not adhere to the dominant behavioristic approach.

All these streams of investigation paved the way for what has come to be known as the cognitive revolution.

2. The Information-Processing Approach

Cognitivism, as it was articulated within the information-processing approach, was from its beginning the result of multiple lines of thought coming from different scientific disciplines. Within psychology, there was a profound dissatisfaction with behaviorism as a method of tackling complex human behavior. World War II compelled psychologists to face practical problems, such as the manipulation of aircraft, radar, and other highly technical devices involving multiple stimuli and responses and posing heavy demands on human operators. In these situations, concepts from the animal learning labs had little application, so psychologists turned to communication engineering and information theory for help. The human operator was conceptualized as an information transmitter and decision maker, and research on selective attention followed as a matter of course.

In 1959, the linguist Noam Chomsky published a devastating review of Skinner’s book Verbal Behavior; his review was also an attack on other behavioristic accounts of language. As an alternative to studying linguistic performance, Chomsky emphasized linguistic competence, knowledge about the rules of grammar that the native speaker of a language owns. Under Chomsky’s influence, psychologists became interested in the cognitive processes responsible for the structural aspects of human language. Subsequently, the significance of linguistic structures for our understanding of memory and thought was increasingly acknowledged.

The most determining influence, however, came from the realm of computer science. Computers are general purpose machines that can be programmed to perform any well-defined task. As information-processing systems, they are able to receive, store, transform, and retrieve information. A computer performance may be analyzed at the physical level of circuits used to implement the system (hardware), but there is a different, more abstract program level (software) that lends itself as an adequate analogy for the human mind. Programs are sets of instructions ready to act upon data; they may take diverse sequences of actions depending upon different conditions. As both instructions and data are stored in the same symbolic form, programs can also act upon themselves in a recursive manner, changing their own instructions if the task demands it. When computers first appeared, they were largely considered to be mathematical devices, number manipulating machines that would make calculations easier. But it soon became apparent that they were able to deal not only with numbers but also, more generally, with symbols. Thus, computers came to be conceptualized as symbol-manipulating systems. Computers can take symbolic input, recode it, store it, retrieve it, make decisions based on the recoded information, and produce symbolic output. These operations closely resemble mental processes and offer a precise and mechanistic way of thinking about them. Hence the computer became the best metaphor to lead the scientific research of the human mind.

There is general agreement that the critical time period giving a boost to cognitivism took place between 1954 and 1960. In 1956, a summer seminar took place at Dartmouth, where Allen Newell, Herbert Simon, and other major workers on artificial intelligence gathered to establish a research program for the new discipline interested in building programs able to generate intelligent behavior. In 1958, another summer seminar was organized by Newell and Simon at the RAND Corporation in Santa Monica in order to acquaint psychologists with computer simulation techniques and their application to the study of cognitive processes. In 1960, the Center for Cognitive Studies was created at Harvard, where Jerome Bruner and George Miller served as a point of reference for the new cognitive psychology.

3. Cognitive Science

The time from 1960 to approximately 1976 was a consolidation period for cognitive psychology. Processes such as attention, memory, language, and thought moved to the forefront of psychological research. In 1967, Ulric Neisser published a textbook, titled Cognitive Psychology, that served as a reference for the psychological research that was being done within the new cognitive framework. The heuristic value of the computer analogy was widely accepted, though it was not understood the same way by all researchers. For some, the computer was just a metaphor that helped to conceptualize human mind as an information-processing system. For others, however, the computer was much more than a metaphor; they assumed that both the computer and the mind were examples of a type of system with specific properties. In 1976, Newell and Simon proposed the term physical symbol system as a name for this type of system and as a fundamental concept for a general and unifying approach to cognition that came to be known as cognitive science. That same year, the Sloan Foundation became interested in this approach and created an interdisciplinary research program to explore its possibilities. In 1977, the journal Cognitive Science was founded, and the Cognitive Science Society followed 2 years later. Nowadays cognitive science has been established as an academic discipline in several universities in different countries.

Cognitive science was an interdisciplinary enterprise in which different branches of scientific knowledge dedicated to the study of cognition merged. Originally, there were six disciplines involved, namely, psychology, computer science, linguistics, philosophy, anthropology, and neuroscience. Each had developed a particular way of looking at cognition, and some already had productive bilateral relations that had given rise to new fields of cognitive research, as was the case for psycholinguistics or computational linguistics. Nevertheless, there was the shared belief that progress within each discipline largely depended upon progress in the rest of them. Today, links between the six disciplines exist, though some are stronger than others, and new fields of research, as is the case for cognitive neuroscience, have emerged.

4. Tenets Of Cognitivism

4.1. Levels of Explanation

Cognitivism is not reductionist; it maintains that complex information-processing systems, such as the computer or the human mind, cannot be understood as a simple extrapolation of the properties of its elementary components. To understand a complex system, we have to take into consideration the fact that its behavior can be described at different levels of abstraction, each related to particular questions that may be asked about the system. If an explanation is conceptualized as an answer to a specific question, then we must be open to expect different explanations depending upon the description level associated to the question. The classical view of cognitive science distinguishes at least three different levels of analysis, named according to David Marr terminology.

4.1.1. The Computational Level

The computational level is the level of abstract problem analysis and refers to what the system does and why. At this level, explaining why people or computers do something requires pointing to their goals and objectives and to the strategies and means to carry them out. Take, as a simple example, the case of a calculator able to perform number addition. An explanation of what the machine does and why should be found in arithmetic theory, because the task to be performed by the system has to comply with the principles of this theory.

4.1.2. The Algorithmic Level

The algorithmic level is the program level in which a representation for coding the system’s input and output should be specified, together with an algorithm providing the appropriate output for a given input. This level refers to how the system brings about its task. It is also called the symbol level, because the goals established at the computational level are encoded by symbolic expressions. In the case of the calculator from the example above, explaining how the system resolves an addition has to do with the representation used to encode numbers (Arabic, Roman, or binary, for instance) and the specific algorithm employed, for example, adding the units first and carrying to the tens if the sum exceeds 9.

4.1.3. The Level of Physical Implementation

The level of physical implementation is concerned with the particular technology used to physically realize the system. In our example, an explanation at this level would point to the working of electronic circuits, but electronic circuits are not the only way to implement an addition machine. Old cash registers, for example, were not electronic but mechanical; however, they could use the same algorithm as an electronic calculator to perform addition.

These three levels are frequently related and influence each other; for example, the choice of electronic circuits to build an addition machine may favor a binary over a decimal representation of numbers. However, it is convenient to note that, because the three levels are not necessarily bound, some phenomena may find an explanation at only one or two of them. Above all, it should be noted that it is pointless to search for only one valid explanation of the behavior of complex information-processing systems. An explanation only makes sense in relation to a question referred to a particular level of analysis. Therefore, we need explanations at different levels in order to understand complex information-processing systems.

4.2. Systems Approach

Cognitivism views human cognition as an activity emerging from the interaction of a system of components. Though the search for components is considered a valuable aspect of the scientific activity, cognitivism emphasizes the significance of structure and functional architecture for our understanding of the mind.

From the beginning, cognitive psychology made wide use of flowchart models as a means of specifying the component processes involved in a particular situation. D. Broadbent’s model was one of the first to specify the attention and memory components needed to explain human behavior in situations in which a person has to select one out of several different streams of information, as exemplified by a cocktail party, where a person has to follow one among many surrounding conversations. At first, models tended to be bound to specific tasks or situations, but soon interest for building general models became dominant. The model of human memory advanced by R. Atkinson and R. Shiffrin in 1968 is a good example of this trend. These authors were able to develop a memory model consisting of three different components, namely, sensory registers, short-term memory, and long-term memory. Extensive previous research had been carried out on each of these components, but Atkinson and Shiffrin managed to put them together within a general framework that was influential for years. Memory research also offers a good illustration of the analytic procedure that cognitive research employed to determine the components of a cognitive system. Five years after the model of Atkinson and Shiffrin was published, E. Tulving argued in favor of the partition of long-term memory into an episodic component and a semantic component. Later, in the 1980s, it was proposed that both episodic and semantic memory were part of a declarative or explicit component of memory having to do with conscious knowledge, but research also focused on an implicit, largely unconscious component of memory dealing with knowledge related to cognitive, motor, and perceptual skills. Though there is no general agreement about the psychological reality of all these components and some authors consider the increasing number of them as a plain violation of the scientific principle of parsimony, a large group of psychologists views this tendency as a result of many years of analytic research and as a valuable way to make contact with brain research.

With the advent of cognitive science, the term cognitive architecture came to be used as a way to indicate that the structure of the cognitive system had some sort of primitive and permanent character. Search for this character divided the field of cognitive science into two conflicting views. On the one hand, the classical view considered the concept of the physical symbol as the building block of cognitive architecture and defined a cognitive system as a physical symbol system. On the other hand, a new connectionist view considered the concept of the neural network to be the basic unit of analysis for understanding the human cognitive system. A physical symbol is a physical entity able to designate or refer to another entity that is the meaning of that symbol. For centuries this capacity to refer to something else was thought to belong only to the realm of ideas, and an idea was considered to be nonmaterial. As mentioned previously, the computer became a model for the human mind when scientists became aware that a computer was a symbol-manipulating machine rather than a machine dealing only with numbers. The physical symbol concept, by attributing to physical entities the capacity of having meaning, became a key concept in bridging the gap between brain and mind. In turn, a physical symbol system was defined as a system constituted by physical symbols. The adaptive control of thought (ACT) and the Soar systems, developed by John Anderson and Allen Newell, respectively, are two of the most influential current symbolic architectures.

The connectionist view represented an attempt to associate cognitive science with brain theory. Though the origin of the concept neural network goes back to Donald Hebb, and some interesting work on neural networks had already been done during the first years of cognitivism, the impulse that brought connectionism to the fore took place in the mid-1980s. Connectionists tried to build information-processing systems that were neurally inspired and called computation on such a system brain-style computation because they thought that the primitive concepts of a cognitive model should resemble the primitive units of the brain, neurons.

A neural network is composed of a set of elementary, neuron-like processing units connected to each other in a specific way. The different strengths of the connections among units defines a pattern of connectivity that may change through experience according to a particular learning rule. An activation rule determines the way different inputs to a unit combine to determine the unit’s state of activation, and an output function is in charge of mapping the unit state of activation into output. The only other component of a neural network is an environment within which the system can operate. Systems of this kind have been successfully used to explain performance on cognitive tasks as well as to model cognitive processes. For some time there were major disputes between defenders of symbolic and connectionist architectures. Nowadays, the arguments have abated, and many cognitive scientists make free use of both symbolic and neural network architectures depending upon the problem at hand. Neural networks seem best suited to model knowledge dependent on long, repetitive practice, as is usually the case in implicit knowledge. On the other hand, symbolic architectures appear preferable for modeling high-level cognitive processes such as reasoning or thought.

4.3. Representation

One of the main characteristics defining a particular system as cognitive is its capacity to reinstate a prior experience in the absence of a current external stimulus. We say that such a system owns the capacity to represent its environment.

Interest in mental representation as such was not a relevant feature of the first cognitive models. As discussed previously, they mainly dealt with mental structure and its components. Reference to mental representation was implicit in the use of terms such as encoding and recoding, but interest in the mental code itself had to wait until the early 1970s.

To understand what representation is we should differentiate the representing world from the represented world. The representing world has to somehow imitate the represented world, though not every possible aspect of the latter must be reflected in the former. For any representation, the following three features must be specified:

What aspects of the represented world are chosen to be encoded;
What elements of the representing world are doing the encoding;
What the correspondence is between the two worlds.

Research on mental images and memory brought to the forefront the very problem of mental representation. Was the representation responsible for mental images different from abstract semantic or syntactic representations? Were there analogical representations? These questions divided the field of cognitive psychology in two halves, one for and the other against analogical representations, a debate that still continues, although recent research in cognitive neuroscience seems to favor their existence. In the field of semantic memory, a plethora of representational formats for abstract knowledge were advanced within the framework of symbolic architectures. Prototypes, semantic features, semantic networks, schemata, and production systems were some of the main formats employed, but none attained general acceptance. Some investigators tried to build semantic representations based on a handful of primitive structures, but most models using symbolic architectures turned to local representations. In a local representation, there is a one-to-one correspondence between entities in the represented world and elements in the representing world.

Connectionist systems use distributed representations. In a distributed representation, each entity is represented by a pattern of activity distributed over many elements in the representing world, and each element represents many different entities. Distributed representations have proved to be very effective in dealing with aspects of cognition that were difficult to implement on symbolic architectures. Content addressable and reconstructive memories, ability to generalize automatically to novel situations, and capacity to be modified according to a changing environment are genuine achievements of distributed architectures.

Cognitive science, symbolic cognitive science in particular, has been criticized for the way it has elaborated the notion of representation. Criticisms range from denying a particular characteristic of the representation, like that of being symbolic or abstract, to calling the entire notion into question. A common trait for these views is their concern about treating mind, body, and environment as coupled systems, dynamically interacting with each other; hence the term situated cognition is often used for this approach. It emphasizes that representations undergo continuous changes in tune with changes in the environment and encourages research on the changing patterns of behavior over time, using nonlinear differential equations to do the modeling. From this point of view, the best way to construct cognitive models is to build real agents. So far, the main contributions of this viewpoint are to be found in the fields of perception and motor processes. To what extent this approach may be useful in understanding higher cognitive processes remains to be seen.

4.4. Process

Processes are operations acting on a representation either to generate it or to transform it in some way. They may be involved in translating external input into a particular representation, modifying the representation itself, or generating output. A process may be considered a function mapping an input to an output.

Interest in elementary mental operations goes back to the Dutch physiologist F.C. Donders in 1868. He assumed that the time spent by an agent performing a task could be divided into a series of stages, each of which corresponded to an elementary component process of the task. By using simple tasks, supposedly consisting of few elementary components, and subtracting the time spent in performing one task from the time corresponding to the next more complex one (with an extra elementary component), Donders thought to have isolated the time for simple cognitive processes such as detection or discrimination. Two assumptions relevant for cognitive psychology underlay Donders’ experimental program: the existence of elementary mental operations and their serial organization as component processes of a complex task.

The information-processing approach has tended to rely on the assumption that a small number of elementary mental operations are sufficient to specify the complexity of human cognitive performance. However, there is little agreement about what the elementary operations are and about the ultimate level of description appropriate for an elementary operation. Connectionist models offer an alternative because they use only activation and inhibition as elementary processes acting on neuron-like elementary units, but many researchers prefer descriptions in more intuitive terms referring to information-processing stages.

A central issue for information-processing theories is the way in which mental processes are assembled in order to perform a task. Early cognitive models assumed a serial organization of processes; they were carried out one after another, each process taking as input the output of the previous one, and starting operation when the preceding process had finished. It soon became apparent that many cognitive processes in perception, memory, and other cognitive areas could not be understood on the basis of serial organization only; rather, they seemed to operate simultaneously, that is, in parallel. Parallel functioning of processes is typically associated with division of labor and modularity. A module, much like a subroutine in a computer program, is made up of a set of operations carrying out a particular assignment in the service of a larger task. It has been much debated whether the entire human mind or only some cognitive processes are modularly arranged. Nevertheless, there seems to be general agreement that modularity is a characteristic of low rather than high-level cognitive processes. Connectionism has made parallel processing a distinctive feature of its approach. The term parallel distributed processing, often used as an alternative name for connectionism, highlights the two defining characteristics of this view, namely, distributed representation and parallel processing. Even so, many connectionist models group their simple, neuron-like units into layers, thus allowing for some serial organization of information processing stages.

An additional dimension introducing differences within cognitive theory has to do with the direction of the processing flow. Some theories and models emphasize bottom-up processing, in which the entire job is carried out by algorithms acting on input data to build increasingly complex representations. David Marr’s influential theory of vision is a good example of a cognitive theory built from a bottom-up point of view. On the other hand, top-down processing emphasizes the feedback influence that complex cognitive component processes, such as goals, beliefs, and knowledge, exert on the functioning of simpler ones. A process is said to be data driven when its working is independent of feedback influences from goals, intentions, or other higher level processing stages. A process depending upon these influences is said to be conceptually driven. The most powerful cognitive models are usually interactive, meaning that they include both bottom-up and top-down processing.

Another frequently used classification of cognitive processes is the division between automatic and controlled processes. An automatic process usually refers to a sequence of component subprocesses that is triggered by a particular input. The sequence may include both bottom-up and top-down processing. An automatic process may be built-in as part of the physical implementation of a system, or its components may be assembled as a result of practice. Thus, to the extent that automaticity depends on practice, it should be considered a gradual rather than an all-or-none property for processes. On the other hand, when a new sequence of component processes is required, top-down attentional influences are needed to do the assembling. In this case, the sequence is called a controlled process, meaning that it is under attentional control. Automaticity and dependence upon attentional control are complementary characteristics of mental processes; the more automatic a process is, the less attentional control it needs to act. Distinguishing between automatic and controlled processes has become very useful in psychological research. This distinction partially overlaps with that between explicit and implicit or between conscious and unconscious processes. Further research on this area will likely reveal fundamental characteristics of the human cognitive system.

4.5. Methods

Cognitivism brought to psychology new research methods and new ways of approaching more traditional ones. Reaction time came to be one of the main dependent variables used to understand cognitive processes. As mentioned before, Donders’ subtractive method and more sophisticated later developments, like S. Sternberg’s additive factors method, served as a guide in that reaction time was not studied by itself, but as an index for making inferences about nonobservable processes. Accuracy measures were also developed with the same purpose. In 1954, signal-detection theory was applied to perception by W. Tanner and J. Swets, providing indexes of perceptual functioning uncontaminated by the observer’s decision biases. Subsequently, signal detection was also applied to memory and became a valuable tool for studying diagnostic systems.

Modeling cognitive processes was a research method characteristic of the cognitive approach. Mathematical as well as flow-chart models were widely used in the early days, but computer models, either symbolic or connectionist, were soon considered the high road to understanding cognitive processes. Computer modeling helped psychologists become aware of cognitive processing intricacy and forced them to be specific about the mechanisms involved in a particular task.

Recently, methods from the realm of neuroscience have been incorporated into cognitive research. These days, electrophysiological and brain-imaging techniques allow online recording of brain activity while a person is performing specific cognitive tasks. These technological advances provide a unique opportunity to relate brain activity to cognitive performance and to establish connections between cognitive constructs and brain structures and processes. An entirely new and promising field of research, known as cognitive neuroscience, has appeared, bringing new horizons for cognitive research.

5. Cognitivist Influence

The influence of cognitivism pervades every field of psychological research and spreads to the entire realm of social sciences. The term cognitive, considered vague and nonscientific during the behaviorist era, has now become respectable and scientific. Different approaches within cognitive science provide new ways of thinking about mental processes and intelligence. The study of human intelligence, once restricted to the statistical analysis of psychological tests, has now expanded to the working of mental machinery. The intelligence factors isolated by means of factor analysis are being studied in relation to cognitive constructs, thus bridging a long-lasting gap between differential and experimental psychology. As almost every research paper in this section shows, cognitive factors are frequently mentioned as relevant causes in the explanation of different practical situations. We learn about cognitive ergonomics, cognitive behavior modification, cognitive neuropsychology, and other cognitive enterprises as truly influential approaches within applied psychology. Each one deserves to be treated separately.

References:

Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press.
Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes, In K. W. Spence, & J. T. Spence (Eds.), The psychology of learning and motivation (pp. 89–195). New York: Academic Press.
Broadbent, D. E. (1958). Perception and communication. London: Pergamon Press.
Clark, A. (1997). Being there: Putting brain, body and world together again. Cambridge, MA: MIT Press.
Gardner, H. (1985). The mind’s new science: A history of the cognitive revolution. New York: Basic Books.
Gazzaniga, M. S., Ivry, R. B., & Mangun, G. R. (1998). Cognitive neuroscience: The biology of the mind. New York: W. W. Norton & Company.
Gruber, H. E., & Vone` che, J. J. (1977). The essential Piaget. New York: Basic Books.
Lachman, R., Lachman, J. L., & Butterfield, E. C. (1979). Cognitive psychology and information processing: An introduction. Hillsdale, NJ: Lawrence Erlbaum.
McClelland, J., Rumelhart, D., & Hinton, G. (Eds.) (1986). Parallel distributed processing: Explorations in the microstructure of cognition, Volume 1: Foundations. Cambridge, MA: MIT Press.
Neisser, U. (1967). Cognitive psychology. New York: Appleton-Century-Crofts.
Newell, A. (1990). Unified theories of cognition. Cambridge, MA: Harvard University Press.
Posner, M. I. (1978). Chronometric explorations of mind. Hillsdale, NJ: Lawrence Erlbaum.
Posner, M. I. (Ed.) (1989). Foundations of cognitive science. Cambridge, MA: MIT Press.
Posner, M. I., & Raichle, M. E. (1994). Images of mind. New York: Scientific American Library.
Wertsch, J. V. (1985). Vygotsky and the social formation of mind. Cambridge, MA: Harvard University Press.

See also:

Free research papers are not written to satisfy your specific instructions. You can use our professional writing services to order a custom research paper on any topic and get your high quality paper at affordable price.