TERM

MEANING

REFERENCE

Adaptation

In Piaget's Theory of Development, there are two cognitive processes that are crucial for progressing from stage to stage: assimilation, accommodation. These two concepts are described below.

Assimilation

Taken together, assimilation and accommodation make up adaptation, which refers to the child's ability to adapt to his or her environment.

1. Siegler, R. (1991). Children's thinking. Englewood Cliffs, NJ: Prentice-Hall.
2. Vasta, R., Haith, M., & Miller, S. A. (1995). Child psychology: The modern science. New York, NY: Wiley.

Alzheimer's Disease

Alzheimer's Disease (AD), a term coined by Alois Alzheimer in 1907, is a relentlessly progressive disease characterized by cognitive decline, behavior al disturbances, and changes in personality. Current estimates of prevalence of AD in Canada suggest that 5.1% of all Canadians 65 and over meet the criteria for the clinical diagnosis of AD, which translates into approximately 161,000 cases. AD prevalence is slightly higher in women than in men. It may be that this difference is due to the longer life expectancy of women although other factors have not been ruled out. The prevalence of dementia is strongly associated with age, affecting 1% of the Canadian population aged 65 to 74, 6.9% of individuals 75-84 and 26% of individuals 85 years and older (Canadian Study of Health and Aging, 1994).

The diagnostic criteria for dementia of the Alzheimer's Type (DAT) are as follows:

(A) The development of multiple cognitive deficits manifested by both:

1. Memory impairment (impaired ability to learn new information or to recall previously learned information)

2. One or more of the following cognitive disturbances:

   • aphasia (language disturbance)

   • apraxia (impaired ability to carry out motor activities despite intact motor function)

   • agnosia (failure to recognize or identify objects despite intact sensory function)

   • disturbances in executive functioning (i.e., planning, organizing, sequencing, abstracting)

The diagnosis of AD is based on exclusionary criteria (i.e., the absence of an identifiable cause) with diagnosis confirmed at autopsy. Treatment strategies to date have been largely ineffective, with experimental treatments mainly directed toward overcoming the cholinergic deficit.

1. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington,DC: Author.
2. Canadian study of health and aging: Study methods and prevalence of dementia. (1994). Canadian Medical Association Journal, 150(6).
3. Whitehouse, P.J. (1993) Dementia. Philadelphia: F.A. Davis.

Analogy

In cognitive psychology, analogy is considered an important method of problem solving. The problem solver attempts to use his or her knolwedge of one problem to solve another problem about which she or he has very little or no information. Barsalou (1992) provides the following example of problem solving by analogy:

"...someone who has worked at the complex for a while could simply explain to you that the layout is analogous to a starfish. On hearing this analogy you might transfer knowledge about starfish to the office complex. Thus the knowledge that a starfish has a circular body, with five legs extending from it radially and symetrically would lead to the belief that the office complex contains a center circular body, with five tapered buildings extending from it in a radially symmetric pattern." (p.110)

Obviously people do not use all of their knowledge about one problem to solve another problem. In the context of his starfish example Barsalou points out that we would not begin to think that the office complex is alive, or that it lives underwater. One problem facing cognitive psychologists is to determine how people decide upon the extent to which an analogy applies. Determining how this may be done is more difficult than it may seem. Consider that, given enough time people can find analogies between any two phenomena. We might want to say that, like the starfish, the office complex is alive--its heating ducts are like blood vessels, its doors are like mouths eating the people who enter the office complex every day. As a cognitive process analogy seems limitless. In a science that strives for regularity and lawfulness the limitlessness of analogical thinking poses a serious problem.

1. Barsalou, L. (1992). Cognitive psychology: An overview for cognitive psychologists. Hillsdale, NJ: Lawrence Erlbaum Associates.

Apparent Motion

This is a perceptual phenomenon that occurs when we perceive motion in two or more static images that are presented in succession with appropriate spatial and temporal displacements. The ability to perceive this phenomenon is mediated by the visuo-spatial pathway of the visual association regions of the brain.

We see examples of this phenomenon almost everyday when we view television or movies. This is an example of a cognitively impenetrable perception. That is, even though we know that the images are not moving, we still perceive motion.

1. Marr, D. (1982). Vision. Freeman: San Francisco, pp.159-182.
2. Zeki, S. (1992). The visual image in mind & brain. Scientific American, 241(3), 150-162.

Articulatory Loop

The articulatory loop (AL) is one of two passive slave systems within Baddeley's (1986) tripartite model of working memory. The AL, responsible for storing speech based information, is comprised of two components. The first component is a phonological memory store which can hold traces of acoustic or speech based material. Material in this short term store lasts about two seconds unless it is maintained through the use of the second subcomponent, articulatory subvocal rehearsal.

Prevention of articulatory rehearsal results in very rapid forgetting. Try this experiment with a friend. Present your friend with three consonants (e.g., C-X-Q) and ask them to recall the consonants after a 10 second delay. During the 10 second interval, prevent your friend from rehearsing the consonants by having them count 'backwards by threes' starting at 100. You will find that your friend's recall is significantly impaired! See Murdoch (1961) and Baddeley (1986) for a complete review.

1. Baddeley, A. (1986). Working memory. Oxford: Clarendon Press.
2. Murdock, B.B. Jr. (1961). The retention of individual items. Journal of Experimental Psychology, 62, 618-625.

Artificial Intelligence

Artificial intelligence is concerned with the attempt to develop complex computer programs that will be capable of performing difficult cognitive tasks. Some of those who work in artificial intelligence are relatively unconcerned as to whether the programs they devise mimic human cognitive functioning, while others have the explicit goal of simulating human cognition on the computer.

The artificial intelligence approach has been applied to several different areas within cognitive psychology, including perception, memory, imagery, thinking, and problem solving.

There are a number of advantages of the artificial intelligence approach to cognition. Computer programming requires that every process be specified in detail, unlike cognitive psychology which often relies on vague descriptions. AI also tends to be highly theoretical, which leads to general theoretical orientations having wide applicability. The main disadvantage of AI is that there is a lot of controversy about the ultimate similarity between human cognitive functioning and computer functioning.

Some of the major differences between brains and computers were spelled out in the following terms by Churchland (1989, p.100):

"The brain seems to be a computer with a radically different style. For example, the brain changes as it learns, it appears to store and process information in the same places...Most obviously, the brain is a parallel machine, in which many interactions occur at the same time in many different channels."

1. Churchland, P.S. (1989). From Descartes to neural networks. Scientific American , July, 100.
2. Eysenck, M.W. (Ed.). (1990). The Blackwell Dictionary of Cognitive Psychology. Cambridge, MA: Basil Blackwell.

Associative Memory

At its simplest, an associative memory is a system which stores mappings of specific input representations to specific output representations. That is to say, a system that "associates" two patterns such that when one is encountered subsequently, the other can be reliably recalled. Kohonen draws an analogy between associative memory and an adaptive filter function [2]. The filter can be viewed as taking an ordered set of input signals, and transforming them into another set of signals---the output of the filter. It is the notion of adaptation, allowing its internal structure to be altered by the transmitted signals, which introduces the concept of memory to the system.

A further refinement in terminology is possible with regard to the associative memory concept, and is ubiquitous in connectionist (neural network) literature in particular. A memory that reproduces its input pattern as output is referred to as auto-associative (i.e. associating patterns with themselves). One that produces output patterns dissimilar to its inputs is termed hetero-associative (i.e. associating patterns with other patterns).

Most associative memory implementations are realized as connectionist networks. Hopfield's collective computation network [1] serves as an excellent example of an auto-associative memory, whereas Rosenblatt's perceptron [3] is often utilized as a hetero-associator. There are many practical problems implementing effective associative memories however, most notably their inefficiency; the tendency is for them to fill up and become unreliable rather quickly. This is a long running open problem for both connectionism and adaptive filter theory---one that Kohonen refers to as the "problem of infinite state memory" [2].

1. J.J. Hopfield. Neural networks and physical systems with emergent collective computation abilities. Proceedings of the National Academy of Science. 79:2554-2558, 1982.
2. T. Kohonen. Self-Organization and Associative Memory. Springer Series In Information Sciences, Vol.8. Springer-Verlag, Berlin, Heidelberg, New York, Tokyo, 1984.
3. F. Rosenblatt. Principles of Neurodynamics. Spartan, New York, 1962.

Attention

"Attention" is a term commonly used in education, psychiatry and psychology. The definition is often vague. Attention can be defined as an internal cognitive process by which one actively selects environmental information (ie. sensation) or actively processes information from internal sources (ie. visceral cues or other thought processes). In more general terms, attention can be defined as an ability to focus and maintain interest in a given task or idea, including managing distractions.

1. DeGangi, G., & Porges, S. (1990). Neuroscience foundations of human performance. Rockville, MD: American Occupational Therapy Association.
2. James, W. (1890). Principles of psychology. New York: Holt.

Attention Getting

Attention getting is more than just the orienting reflex, it is the "initial orientation or alerting to a stimulus." Though this may be considered an automatic act, in fact it requires complex active thought processing. Attention getting is reliant on the qualitative nature of the stimulus. The stimulus must be strong enough to elicit a response.

DeGangi and Porges (1990) explain the types of stimuli that are attention getting vary according to past experiences of the individual, what they already know, individual reactivity to sensory stimuli, and what an individual has determined to be important to them. A hungry person may be more apt to pay attention to the smell of food than the sounds surrounding them in a traffic jam!

Attention getting is important to psychologists, particularly developmental psychologists because of its role in learning. A child's chosen attention getting stimuli can guide his/her learning abilities. "A child who learns better through the auditory channel will orient more readily to a song about body parts than a picture of a body."

1. DeGangi, G., & Porges, S. (1990). Neuroscience foundations of human performance. Rockville, MD: American Occupational Therapy Association.

Attention Holding

Attention holding is the "maintenance of attention when a stimulus is intricate or novel." Stimuli that hold our attention must be both novel and complex in order to encourage information processing. Attention holding is measured by how long one engages in a cognitive activity involving that stimulus.

Attention holding is important because of its role in learning. If an activity or stimulus is moderately complex, the person will expend energy in information processing. In other words, the person will expend energy in learning. Unfortunately, this can be complicated by poor motivation. Low motivation may present a challenge as the psychologist (or other professional) must determine if the decreased motivation is due to sensory processing problems, cognitive impairment, or other learning-related problems (of which poor attention holding may be identified).

1. DeGangi, G., & Porges, S. (1990). Neuroscience foundations of human performance. Rockville, MD: American Occupational Therapy Association.

Attention Releasing

Attention releasing is the final stage in DeGangi and Porges' (1990) process of sustained attention. Attention releasing can simply be defined as the "releasing or turning off of attention from a stimulus." Attention releasing can occur for a variety of reasons. A person can fatigue physically or mentally requiring release of attention. Arousal level can decrease, therefore a different type/strength of stimuli becomes required to maintain an alert and active state.

Attention releasing provides a person with a method to reach closure on a given activity, task, or event thereby allowing that person to switch attention to something new. As with attention getting and holding, attention releasing (the ability to shift focus) plays an important role in the learning process.

1. DeGangi, G., & Porges, S. (1990). Neuroscience foundations of human performance. Rockville, MD: American Occupational Therapy Association.

Behavioral Indeterminacy

The claim that in principle psychology is restricted to establishing weak equivalence. Weak equivalence is equivalence with respect to input/output behavior . Therefore, measuring behavior al data is unable to establish equivalence at the level of functional architecture. behavior al studies are indeterminate with respect to strong equivalence.

This issue is of importance to cognitive psychology because, if true, it implies that cognitive psychology cannot generate insight into cognition without importing knowledge based on non-behavior al observations from other disciplines.

1. Pylyshyn, Z. W. (1989). Computing in cognitive science. In M. I. Posner (Ed.), Foundations of cognitive science, Cambridge MA: MIT Press.

Biological Naturalism

Promoted by John Searle, Biological Naturalism states that consciousness is a higher level function of the brain's physical capabilities. The neurophysiological processes in the brain cause mental phenomena, which are also a feature of the brain. However, such features as consciousness are not reducible to neurophysiological systems. Not all brains produce this higher level functioning, and there are many questions still open in Biological Naturalism, which Searle himself points out, for example: how does neurophysiology account for the range of mental phenomena? how does consciousness come about? how advanced does a neurophysiological system have to be to produce consciousness?

1. Searle, John. The Rediscovery of the Mind. MIT Press, Massachusetts. 1994

Bottom-Up Processing

The cognitive system is organized hierarchically. The most basic perceptual systems are located at the bottom of the hierarchy, and the most complex cognitive (e.g. memory, problem solving) systems are located at the top of the hierarchy.

Information can flow both from the bottom of the system to the top of the system and from the top of the system to the bottom of the system. When information flows from the bottom of the sytstem to the top of the system this is called "bottom-up" processing. Lower level systems categorize and describe incoming perceptual information and pass this descriptive information onto hiher levels for more complex processing.

Broca's Area

Named for Paul Broca who first described it in 1861, Broca's area is the section of the brain which is involved in speech production, specifically assessing syntax of words while listening, and comprehending structural complexity. People suffering from neurophysiological damage to this area (called Broca's aphasia or nonfluent aphasia) are unable to understand and make grammatically complex sentences. Speech will consist almost entirely of content words.

Auditory and speech information is transported from the auditory area to Wernicke's area for evaluation of significance of content words, then to Broca's area for analysis of syntax. In speech production, content words are selected by neural systems in Wernicke's area, grammatical refinements are added by neural systems in Broca's area, and then the information is sent to the motor cortex, which sets up the muscle movements for speaking.

1. Gray, Peter. (1994). Psychology. New York, NY: Worth Publishing.

Cascade Processing

Under the assumption that a complex task can be broken down into distinct stages of information processing, and that these stages can be sequentially ordered, the complex task can be performed by completing each distinct stage.

Unlike discrete processing, with cascade models the latter stages of information processing can begin operating before the completion of earlier information processing stages. Connectionist models of information processing operate in a cascade manner and are important for the way in which these models can learn relationships between stimule and responses.

Depending on the complexity of the information being processed, it may be transmitted between some processing stages in a cascade manner, but in other stages it may be processed in a discrete manner.

1. Eysenck, M.W. (Ed.). (1990). The Blackwell Dictionary of Cognitive Psychology. Cambridge, MA: Basil Blackwell.

Central Executive

The central executive, the most important yet least well understood component of Baddeley's (1986) working memory model, is postulated to be responsible for the selection, initiation, and termination of processing routines (e.g., encoding, storing, retrieving). Baddeley (1986, 1990) equates the central executive with the supervisory attentional system (SAS) described by Norman and Shallice (1980) and by Shallice (1982).

According to Shallice (1982), the supervisory attentional system is a limited capacity system and is used for a variety of purposes, including:

• tasks involving planning or decision making
• trouble shooting in situations in which the automatic processes appear to be running into difficulty
• novel situations
• dangerous or technically difficult situations
• situations where strong habitual responses or temptations are involved

In other words, patients suffering from frontal lobe syndrome lack flexibility and the ability to control their processing resources, functions attributed to the central executive.

1. Baddeley, A.D. (1990). Human memory: Theory and practice,. Oxford: Oxford University Press.
2. Baddeley, A.D. (1986). Working memory. Oxford: Clarendon Press.
3. Norman, D.A., & Shallice, T. (1980). Attention to action. Willed and automatic control of behavior. University of California San Diego CHIP Report 99.
4. Shallice, T. (1982). Specific impairments of planning. Philosophical Transactions of the Royal Society London B 298, 199-209.
5. Rylander, G. (1939). Personality changes after operations on the frontal lobes. Acta Psychiatrica Neurologica, Supplement No. 30.

Cognitive Development

Generally it is referred to the changes which occur to a person's cognitive structures, abilities, and processes. The most widely known theory of childhood cognitive development was proposed by Jean Piaget in 1969. He proposed the idea that cognitive development consisted of the development of logical competence, and that the development of this competence consists of four major stages:

1. sensor-motor
2. preoperational
3. concrete operational
4. formal operational

Cognitive Mapping

Cognitive mapping is a general term that applies to a series of methods for measuring mental representations. These techniques attempt to describe mental images that subjects use to encode knowledge and information. Most researchers treat cognitive maps as a tool that can usefully summarize and communicate information rather than as a literal description of mental images

1. Huff, A.S. (1990). Mapping Strategic Thought Chichester, John Wiley & Sons

Cognitive Penetrability

An approach to testing strong equivalence. The cognitive penetrability approach seeks to establish whether phenomena are equivalent at the level of functional architecture by investigating whether phenomena are independent of beliefs and goals, that is if they are primitive. If manipulation of beliefs and goals systematically alters the empirical phenomenon then the phenomenon is not describing functional architecture and is cognitively penetrable.

The cognitive penetrability approach was used in the imagery debate in cognitive science in the 1980's.

1. Pylyshyn, Z. W. (1989). Computing in cognitive science. In M. I. Posner (Ed.), Foundations of cognitive science. Cambridge, MA: MIT Press.

Cognitive Psychology

Cognitive psychology is concerned with information processing, and includes a variety of processes such as attention, perception, learning, and memory. It is also concerned with the structures and representations involved in cognition. The greatest difference between the approach adopted by cognitive psychologists and by the Behaviorists is that cognitive psychologists are interested in identifying in detail what happens between stimulus and response.

Some of the ingredients of the information processing approach to cognition were spelled out by Lachman, Lachman, and Butterfield (1979). In essence, it is assumed that the mind can be regarded as a general purpose, symbol processing system, and that these symbols are transformed into other symbols as a result of being acted on by different processes. The mind has structural and resource limitations, and so should be thought of as a limited capacity processor.

A key issue in the field is the extent to which human and computer information processing systems resemble one another. The consensual view is probably that there are indeed striking similarities between computer minds, but there are also probably substantial differences. In recent years, explicitly cognitive approaches have been adopted in social and developmental psychology, as well as in occupational and clinical psychology.

1. Eysenck, M.W. (Ed.). (1990). Blackwell Dictionary of Cognitive Psychology. Cambridge, MA: Basil Blackwell.
2. Lachman, R., Lachman, J.L., & Butterfield, E.C., (1979) Cognitive psychology and information processing. Hillsdale, NJ: Lawrence Erlbaum Associates.

Cognitive Science

"the study of intelligence and intelligent systems, with particular reference to intelligent behavior  as computation" (Simon & Kaplan, 1989)

Simon, H. A. & C. A. Kaplan, "Foundations of cognitive science", in Posner, M.I. (ed.) 1989, Foundations of Cognitive Science, MIT Press, Cambridge MA.

1. Simon, H. A. & C. A. Kaplan, "Foundations of cognitive science", in Posner, M.I. (ed.) 1989, Foundations of Cognitive Science, MIT Press, Cambridge MA.
2. Eysenck, M.W. ed. (1990). The Blackwell Dictionary of Cognitive Psychology. Cambridge, Massachusetts: Basil Blackwell Ltd.

Connectionism

Connectionism is an alternate computational paradigm to that provided by the von Neumann architecture. Originally taking its inspiration from the biological neuron and neurological organization, it emphasizes collections of simple processing elements in place of the monolithic processors seen more commonly within computing. These simple processing elements are typically only capable of rudimentary calculations (such as summation), however possess a high degree of weighted inter-connectivity with one another and generally operate in parallel [2].

A particular organization of inter-connected processing elements (a network), is paired with a mathematical basis by which the connection weights are adjusted (or simply calculated directly). This allows a network to either learn a task by iterating on training examples (induction learning), or to provide a system in which solutions to particular problems can be computed. Arguably the most widely used example of the former is the multi-layer perceptron trained via error back-propagation (see [5], for example); whereas the latter is typified by networks such as the Hopfield and Tank model for combinatorial optimization [3].

To the casual reader, "connectionism", "parallel distributed processing" (PDP) and "neural networks" may be entirely synonymous. The term "neural network" is somewhat misleading to begin with as, aside from the original inspiration coming from biology, there is nothing particularly "neural" about them and any perceived biological relevance is often debatable. There is also merit in making a philosophical distinction between PDP and connectionism. For example, over time, PDP has been disposed to seek biological relevance for their models, tended to emphasize learning oriented tasks and follow a largely empirical approach. The field of neural networks has become richer than is encompassed by the traditional view of PDP.

Connectionism distinguishes itself by also viewing the network model as a computational architecture. This encompasses a wider range of network structures for which biological relevance is not an issue or for which a learning process per se is not utilized. Falling into areas such as these include a wealth of recent work which has sought to establish the formal relationship between computational power of connectionist networks and abstract machines (for example [1],[4]), and even harkens back to the aforementioned Hopfield and Tank model which computes solutions to problems by minimizing energy within a pre-wired system of weights [3].

In this respect, connectionism subsumes PDP. That is to say that PDP researchers are connectionists, however not all connectionists consider themselves to be PDP researchers. Although debatable, this point is one that this author, among others, feels is an important one.

1. C.L. Giles, B.G. Horne, T. Lin. Learning a class of large finite state machines with a recurrent neural network. Neural Networks. 8(9):1359-1365, 1995.
2. J. Hertz, A. Krogh and R.G. Palmer. Introduction to the theory of neural computation. Addison-Wesley, Redwood City, 1991.
3. J.J. Hopfield and D.W. Tank. `Neural' computation of decisions in optimization problems. Biological Cybernetics. 52:141-152.
4. S.C. Kremer. On the computational power of Elman-style recurrent networks. IEEE Transactions on Neural Networks. 6(4):1000-1004, 1995.
5. D.E. Rumelhart, G.E. Hinton, and R.J. Williams. Learning internal representations by error propagation. In D.E. Rumelhart and J.L. McClelland, editors, Parallel Distributed Processing, volume 1. MIT Press, Cambridge, 1986.

Consciousness

Consciousness refers to awareness of our own mental processes (or of the products of such processes). This awareness can be made manifest by introspective reports, in which an individual provides information about his or her mental experience.

There has been a considerable amount of controversy over the centuries concerning the value of psychology of assessing the contents of consciousness by means of introspective evidence. Aristotle claimed that the only way to study thinking was by introspection. Others, such as Galton (1883), argued that the position of consciousness "appears to be a helpless spectator of but a minute fraction of automatic brain work. Behaviorists tend to agree with Galton that psychologists should not concern themselves with consciousness and introspection.

There are certain cognitivists who would disagree with these definitions. Marvin Minsky (1985), maintains that human consciousness can never represent what is occurring at the present moment, but only a little of the recent past. This is due both because agencies have limited capacity to represent what happened recently and partly because it take time for agencies to communicate with one another. Consciousness is difficult to describe because each time we attempt to examine temporary memories, we distort the very record we are trying to interpret.

1. Eysenck, M.W. (Ed.). (1990). Blackwell Dictionary of Cognitive Psychology . Cambridge, MA: Basil Blackwell.
2. Galton, F. (1883). Inquiries into human faculty and its development. London: Macmillan.
3. Minsky, M. (1985). The society of mind. New York, NY: Simon & Schuster.

Content Addressable Memory

In a symbolic system information is stored in an external mechanism. In the example of the computer it is stored in files on the disks. As the information has been encoded in some form of file system in order to retrieve that information one must know the index system of the files. In other words, data can only be accessed by certain attributes. In a connectionist system the data is stored in the activation pattern of the units. Hence, if a processing unit receives excitatory input from one of its connections, each of its other connections will either be excited or inhibited. If these connections represent the attributes of the data then the data may be recalled by any one of its attributes, not just those that are part of an indexing system. As these connections represent the content of the data, this type of memory is called content addressable memory. This type of memory has the advantage of allowing greater flexibility of recall and is more robust. This distributed memory is able to work its way around errors by reconstructing information that may have been lesioned from the system.

1. Bechtel, W., & Abrahamsen, A. (1991). Connectionism and the mind: An introduction to parallel processing in networks. Cambridge, MA: Blackwell.

Crystallized Intelligence

Crystallized intelligence can be defined as "the extent to which a person has absorbed the content of culture." (Belsky, 1990, p. 125) It is the store of knowledge or information that a given society has accumulated over time.

Crystallized intelligence is measured by most of the verbal subtests of the Wechsler Adult Intelligence Scale (WAIS).

Crystallized intelligence is important to psychologists as it relates to the study of aging. There is ongoing intense debate among psychologists as to whether or not intelligence declines with aging. Horn (1970) hypothesized that because crystallized intelligence is based on learning and experience, it remains relatively stable over time. He claims it may even increase "as the rate at which we acquire or learn new information in the course of living balances out or exceeds the rate at which we forget." (as cited in Belsky, 1990, p. 125) On the other side of the debate, Belsky (1990) claims crystallized intelligence in fact declines with age. Why? Because, "at a certain time of life the cumulative effect of losses - of job, of health, of relationships - cause disengagement from the culture, and so forgetting finally exceeds the rate at which knowledge is acquired." (p. 125)

1. Belsky, J. K. (1990). The psychology of aging theory, research, and interventions. Pacific Grove, CA: Brooks/Cole.
2. Horn, J. (1970). Organization of data on life-span development of human abilities. In R. Goulet and P.B. Baltes (Eds.). Life-span developmental psychology: Research and theory. New York: Academic Press.

Cued Recall

This is a component of a memory task in which the subject is asked to recall items that were presented to them on an initial training, or initial presentation list.

However, it is slightly different than the free recall task because the subject is given a hint, or a cue, about the items on the original list. For example, and experimenter may say: "Tell me all the words from the list that were animals".

Deductive Inference

Inferences are made when a person (or machine) goes beyond available evidence to form a conclusion. With a deductive inference, this conclusion always follows the stated premises. In other words, if the premises are true, then the conclusion is valid. Studies of human efficiency in deductive inference involves conditional reasoning problems which follow the "if A, then B" format.

The task of making deductions consists of three stages. First, a person must understand the meaning of the premises. Next they must be able to formulate a valid conclusion. Thirdly, a person should evaluate their conclusion to tests its validity. Although deductive inference is easy to test or model, the results of this type of inference never increase the semantic information above what is already stated in the premises

1. Eysenck, M.W. (Ed.). (1990). The Blackwell dictionary of cognitive psychology. Cambridge, MA: Basil Blackwell.
2. Johnson-Laird, P. N. (1993). Human and machine thinking. Hillsdale, NJ : Lawrence Erlbaum Associates.

Dementia

Dementia is a clinical state characterized by loss of function in multiple cognitive domains. The most commonly used criteria for diagnoses of dementia is the DSM-IV (Diagnostic and Statistical Manual for Mental Disorders, American Psychiatric Association). Diagnostic features include :

• memory impairment and at least one of the following: aphasia, apraxia, agnosia, disturbances in executive functioning.
• In addition, the cognitive impairments must be severe enough to cause impairment in social and occupational functioning.
• Importantly, the decline must represent a decline from a previously higher level of functioning.
• Finally, the diagnosis of dementia should NOT be made if the cognitive deficits occur exclusively during the course of a delirium.

1. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.
2. Canadian study of health and aging: Study methods and prevalence of dementia. (1994). Canadian Medical Association Journal, 150(6).

Discrete Processing

A model using discrete processing requires that information is passed from one stage to another only after the processing in the first stage is complete. Therefore, the processing time required in a discrete model is additive and equal to the sum of the time taken at each level of processing.

The advantage of this type of model is that it provides a convenient method of understanding the effects of different variables on the performance of a given task.

1. Eysenck, M.W. (Ed.). (1990). The Blackwell Dictionary of Cognitive Psychology. Cambridge, MA: Basil Blackwell.

The Disjunction Problem

Any theory of the content of a representation must be able to explain how a representation can misrepresent --how it can represent an object as being something it is not, or as having properties it does not have-- basically how its content can be false of the object represented.

The difficulty is that we need to explain --in a principled, non-circular way-- how the representation can correctly represent some things which cause its activation, yet misrepresent other things which cause its activation. For instance, we9d like to be able to say that my kangaroo representation represents kangaroos. If so, then if a wallaby causes the activation of that representation, then the wallaby is misrepresented; the representation9s content that9s a kangaroo is false of the wallaby.

Unfortunately, to Fodor (1987, 1990) this doesn9t work. The problem is that if the wallaby can also cause the activation of my kangaroo representation, then we seem to have no principled reason for saying that the content of the representation is simply that9s a kangaroo rather than the disjunctive content that9s either a kangaroo or a wallaby. If this is so, then when a wallaby activates my kangaroo representation, this representation doesn9t represent the wallaby as something it is not. This representation has the (disjunctive) content that9s either a kangaroo or it9s a wallaby which, of course, is true of the wallaby.

This content might better be described as 3unspecific2, rather than 3disjunctive2. That is, perhaps the content is something like an unspecific description which applies correctly to all the things which can activate it, such as that9s a large animal with a long tail that gets about by hopping on its hind legs. So to say that some things which activate the representation are correctly represented and others are misrepresented doesn9t work. Even if I9ve only ever seen kangaroos, and have never met a wallaby, the wallaby can be correctly represented by this representation, because the wallaby is also a large animal with a long tail that gets about by hopping on its hind legs.

This is especially a problem for theories which explain content in terms of covariance: some sort of reliable, law-like, connection between tokenings of the representation and the occurrence of certain types of thing in the world. Such theories have to be able to justify describing the representation9s content 3conservatively2 as Cummins (1990) calls it, rather than 3liberally2; as that9s a kangaroo rather than that9s a large animal with a long tail that gets about by hopping on its hind legs. Cummins summarizes various attempts to do this, arguing that covariance theories don9t explain content in a way that allows representations to misrepresent.

Fodor (1990) claims that any theory which purports to account for the content of a representation must solve the disjunction problem. Such an account must be able to explain misrepresentation, by showing what a representation9s content is--exactly-- and also how a representation can be caused to be activated by something to which that content does not apply.

1. Cummins, R. (1989). Meaning and Mental Representation. Cambridge, Mass: MIT Press. A Bradford Book.
2. Fodor, J. (1987). 3Meaning and the World Order2. In Psychosemantics (pp. 97-133). Cambridge Mass.: MIT Press. A Bradford Book.
3. Fodor, J. (1990). 3A Theory of Content I: The Problem2. In A Theory of Content and Other Essays. (pp. 51-88). Cambridge, Massachusetts: MIT Press. A Bradford Book.

Elaborative Rehearsal

Elaborative rehearsal is a type of rehearsal proposed by Craik and Lockhart (1972) in their Levels of Processing model of memory. In contrast to maintenance rehearsal, which involves simple rote repetition, elaborative rehearsal involves deep semantic processing of a to-be-remembered item resulting in the production of durable memories.

For example, if you were presented with a list of digits for later recall (4920975), grouping the digits together to form a phone number transforms the stimuli from a meaningless string of digits to something that has meaning.

1. Craik, F.I.M., & Lockhart, R.S. (1972). Levels of processing. A framework for memory research. Journal of Verbal Learning and Verbal behavior , 11, 671-684.

Enactment

Weick (1988) describes the term enactment as representing the notion that when people act they bring structures and events into existence and set them in action. The process of enactment involves two steps. First, preconceptions are used to set aside portions of the field of experience for further attention, that is, perception is focused on predetermined stimuli. Second, people act within the context of these portions of experience guided by preconceptions in such a way as to reinforce these preconceptions. Hence, attention to certain stimuli will guide subsequent action so that those stimuli are confirmed as important. The result of the process of enactment is the enacted environment (Weick, 1988). This enacted environment comprises "real" objects but the significance, meaning and content of these objects will vary. These objects are not significant unless they are acted upon and incorporated into events, situations and explanations. In this way the enacted environment is a direct result of the preconceptions held by the social actor. An enacted environment is internalized by social actors as the way in which actions have led to certain consequences; it is therefore analogous to the concept of schema and is the source of expectations for future action (Weick, 1988) . An enacted environment is "a map of if-then assertions in which actions are related outcomes" that in turn serve as expectations for future action and focus perception in such way that these preconceived relationships will be supported.

The importance of the notion of enactment is that it provides a direct link between individual cognitive processes and environments. By showing how preconceptions can shape the nature of the environment this concept allows one to argue the importance of schema in the sense making process. Schema guide both perception and inference (Fiske & Taylor, 1991) and so will 'enact' environment by assigning significance, meaning and content to objects perceived in the environment.

1. Fiske, S.T., & Taylor, S.E. (1991). Social cognition (2nd ed.). New York: McGraw-Hill.
2. Weick, K. E. (1988). Enacted sensemaking in crisis situations. Journal of Management Studies, 24(4).

Encoding

Encoding refers to the processes of how items are placed into memory.

Encoding Specificity

The encoding specificity principle of memory (Tulving & Thomson, 1973) provides an general theoretical framework for understanding how contextual information affects memory. Specifically, the principle states that memory is improved when information available at encoding is also available at retrieval. For example, the encoding specificity principle would predict that recall for information would be better if subjects were tested in the same room they had studied in versus having studied in one room and tested in a different room (see S.M. Smith, Glenberg, & Bjork, 1978).

1. Smith, S.M., Glenberg, A.M., & Bjork, R.A. (1978). Environmental contest and human memory. Memory and Cognition, 6, 342-353.
2. Tulving, E., & Thomson, D.M. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80, 352-373.

Equilibration

According to Piaget, development is driven by the process of equilibration. Equilibration encompasses assimilation (i.e., people transform incoming information so that it fits within their existing thinking) and accommodation (i.e, people adapt their thinking to incoming information). Piaget suggested that equilibration takes place in three phases.

First children are satisfied with their mode of thought and therefore are in a state of equilibrium.

Then, they become aware of the shortcomings in their existing thinking and are dissatisfied (i.e., are in a state of dis-equilibrium and experience cognitive conflict).

Last, they adopt a more sophisticated mode of thought that eliminates the shortcomings of the old one (i.e., reach a more stable equilibrium).

Error Analysis

One of the key goals of cognitive science is to develop theories that are strongly equivalent with respect to to-be-explained systems. This requires that evidence be collected to defend the claim that the model and the to-be-explained system are carrying out the same procedures to compute a function.

One kind of information that could be used to examine this claim is called error analysis. In an error analysis, one could (for two different systems) rank order problems in terms of their difficulty, as revealed by their likelihood to produce mistakes. This is an example of relative complexity evidence. A more detailed approach would be to classify the nature of the errors that each system made. In either case, if the two systems were strongly equivalent, then we would expect them to produce the same rank orderings of difficulty, and to also produce the same qualitative patterns of errors.

1. Pylyshyn, Z.W. (1984). Computation and cognition. Cambridge, MA: MIT Press.

Extension

The extension of the term 'cat' is the class of 'cat'.

What a term means has two components: i) the referent of the term--this is 'class' talk, and is the component of meaning to which 'extension' applies; and ii) the sense of the term, i.e., all of the psychological associations that one has with that term--this is 'concept' talk. This second sense is referred to as the 'intension' of the term.

Examples of the two components follow. The referent of the term 'cat' is all the cats; the sense of the term is related to your experience of cats, their history, their attributes, etc. A classic example is 'the morning star' and 'the evening star'; both of which refer to the same thing, the planet 'Venus', but the sense of 'morning star' and 'evening star' is not the same. You cannot change the terms in a statement including one of them and retain the same truth value.

Other words sometimes used to pick out the distinctions between 'extension' and 'intension' are 'denotation' and 'connotation', respectively. Note the following definition by Cohen and Nagel:

A term [an element of a proposition] may be viewed in two ways, either as a class of objects (which may have only one member), or as a set of attributes or characteristics which determine the objects. The first phase or aspect is called the denotation or extension of the term, while the second is called the connotation or intension. The extension of the term 'philosopher' is 'Socrates', 'Plato', 'Thales', and the like; its intension is 'lover of wisdom', 'intelligent', and so on. (31)

1. Cohen, M. R. and Nagel, E. (1993). An Introduction to Logic. Indianapolis, Indiana: Hackett Publishing Company.

Fluid Intelligence

Fluid intelligence is tied to biology. It is defined as our "on-the-spot reasoning ability, a skill not basically dependant on our experience." (Belsky, 1990, p. 125) Belsky (1990) indicates this type of intelligence is active when the central nervous system (CNS) is at its physiological peak.

Fluid intelligence is measured by the performance subtasks on the Wechsler Adult Intelligence Scale (WAIS).

Fluid intelligence is important to psychologists as it relates to the study of aging. There is ongoing intense debate among psychologists as to whether or not intelligence declines with aging. Belsky (1990) claims fluid intelligence "reaches a peak in early adulthood and then regularly declines." (p. 125) This is because of the physiological changes that accompany aging. "The development of CNS structures is exceeded by the rate of CNS breakdown." (Horn, 1970 as quoted in Belsky, 1990, p. 125)

1. Belsky, J. K. (1990). The psychology of aging theory, research, and interventions. Pacific Grove, CA: Brooks/Cole.
2. Horn, J. (1970). Organization of data on life-span development of human abilities. In R. Goulet and P.B. Baltes (Eds.). Life-span developmental psychology: Research and theory. New York: Academic Press.

Formality Condition 

The semantic properties of a representation are the properties it has due to its relationship with the world; properties such as being true, of being a representation of something, of saying something about some object. On the other hand, the properties that the representation has in itself, are its formal properties. Fodor (1980) defines a representation9s formal properties negatively, by specifying what they are not: 3Formal properties are the ones that can be specified without reference to such semantic properties as, for example, truth reference, and meaning.2 (p.227) Fodor stresses that formal properties are not syntactic properties. A representation can have formal properties, and a process can operate on those formal properties, without that representation having a syntax (p227); rotating an image on a screen, for instance this operation is performed on the image9s formal properties, but the image doesn9t even have a syntax..

The point for a computational theory of mind, which takes mental processes to be formal operations on representations, (and thus, to Fodor, taking the mind to be a 3kind of computer2) is that such processes only have access to a representation9s formal properties. Computational processes do not have any access to semantic properties; that is, to a representation's relationships with the world.

Thus the processes that operate on representations cannot operate on the basis of what this is a representation of, or whether it represents that thing correctly or not, but only on the character of the representation itself, its 3shape2 as it were. Thus the Formality Condition incurs what Putnam (1975) calls Methodological Solipsism.

3If mental processes are formal, then they have access only to the formal properties of such representations of the environment as the senses provide. Hence, they have no access to the semantic properties of such representations, including the property of being true, of having referents, or, indeed, the property of being representations of the environment.2 (Fodor (1980), p231, Fodor9s emphasis)

1. Fodor, J. (1980). Methodological Solipsism Considered as a Research Strategy in Cognitive Psychology. In Representations (pp. 225-253). Cambridge, Massachusetts: MIT Press. A Bradford Book.
2. Putnam, H. (1975). 3The Meaning of Meaning2. In K. Gunderson (Ed.), Minnesota Studies in the Philosophy of Science (pp. 131-193). Minneapolis: University of Minnesota Press.

Free Recall

Free recall is a basic paradigm used to study human memory. In a free recall task, a subject is presented a list of to-be-remembered items, one at at time. For example, an experimenter might read a list of 20 words aloud, presenting a new word to the subject every 4 seconds. At the end of the presentation of the list, the subject is asked to recall the items (e.g., by writing down as many items from the list as possible). It is called a free recall task because the subject is free to recall the items in any order that he or she desires.

The free recall task is of interest to cognitive science because it provided some of the basic information used to decompose the mental state term "memory" into simpler sub-functions ("primary memory", "secondary memory"). This is because the results of a free recall task were typically plotted as a serial position curve. This curve exhibited a recency effect and a primacy effect. The behavior of these two effects provided support to the hypothesis that the free recall task called upon both a short-term and a long-term memory.

1. Fodor, J. (1980). Methodological Solipsism Considered as a Research Strategy in Cognitive Psychology. In Representations (pp. 225-253). Cambridge, Massachusetts: MIT Press. A Bradford Book.
2. Putnam, H. (1975). 3The Meaning of Meaning2. In K. Gunderson (Ed.), Minnesota Studies in the Philosophy of Science (pp. 131-193). Minneapolis: University of Minnesota Press.

Functional Analysis

Functional analysis is a methodology that is used to explain the workings of a complex system. The basic idea is that the system is viewed as computing a function (or, more generally, as solving an information processing problem). Functional analysis assumes that such processing can be explained by decomposing this complex function into a set of simpler functions that are computed by an organized system of sub-processors. The hope is that when this type of decomposition is performed, the sub-functions that are defined will be simpler than the original function, and as a result will be easier to explain.

A very detailed treatment of functional analysis is provided by Cummins (1983). He proposes a three-stage methodology that defines functional analysis. In the first stage, the to-be-explained function is defined. In the second stage, analysis is performed. The to-be-explained function is decomposed into an organized set of simpler functions. This analysis can proceed recursively by decomposing some (or all) of the sub-functions into sub-sub-functions. In the third stage, analysis is stopped by subsuming the bottom level of functions. This means that the operation of each of these operation is explained by appealing to natural laws (e.g., mechanical or biological principles). If functional analysis is applied to an information processing system, then the level of subsumed functions defines the functional architecture for that information processor.

Functional analysis is important to cognitive science because it offers a natural methodology for explaining how information processing is being carried out. For instance, any "black box diagram" offered as a model or theory by a cognitive psychologist represents the result of carrying out the analytic stage of functional analysis. Any proposal about what constitutes the cognitive architecture can be viewed as a hypothesis about the nature of cognitive functions at the level at which these functions are subsumed.

1. Cummins, R. (1983). The nature of psychological explanation. Cambridge, MA: MIT Press.

Functional Architecture

The functional architecture can be viewed as the set of basic information processing capabilities available to an information processing system.

"Specifying the functional architecture of a system is like providing a manual that defines some programming language. Indeed, defining a programming language is equivalent to specifying the functional architecture of a virtual machine" (Pylyshyn, 1984, p. 92).

1. Pylyshyn, Z.W. (1984). Computation and cognition. Cambridge, MA: MIT Press.

Generalization

Klahr & Wallace (1982) felt that Piaget's theory of adaptation was not enough to explain cognitive development. They therefore developed a new theory, and posited that the mechanism behind development was generalization.

Klahr and Wallace divided generalization into three more specific categories: the time line, regularity detection, and redundancy elimination (Siegler, 1991). These three categories are described below.

The Time Line

1. Klahr, D. (1982). Nonmonotone assessment of monotone development: An information processing analysis. In S. Strauss (Ed.), U-shaped behavioral growth. New York: Academic Press.
2. Siegler, R. (1991). Children's thinking. Englewood Cliffs, NJ: Prentice-Hall.
3. Vasta, R., Haith, M. M., & Miller, S. A. (1995). Child psychology: The modern science. New York, NY: Wiley.

Graceful Degradation

In a symbolic system removing part of the system will result in a clear degradation of performance. Removing a symbol token will result in the loss of the information stored in that token. The loss of an operating procedure destroys the systems ability to perform the missing process. The fall in performance is sudden and clearly defined. In a connectionist system performance does fall sharply with either damage to the system or erroneous inputs. Instead, the performance will decline gradually, depending on the nature of the loss and the architecture of the system. This property means that connectionist models still function relatively error free when the system has damage to its connections or units or when the input stimuli is incomplete.

1. Bechtel, W., & Abrahamsen, A. (1991). Connectionism and the mind: An introduction to parallel processing in networks. Cambridge, MA: Blackwell.

Hebbian Learning Rule

The Hebbian Learning Rule is a learning rule that specifies how much the weight of the connection between two units should be increased or decreased in proportion to the product of their activation. The rule builds on Hebbs's 1949 learning rule which states that the connections between two neurons might be strengthened if the neurons fire simultaneously. The Hebbian Rule works well as long as all the input patterns are orthogonal or uncorrelated. The requirement of orthogonality places serious limitations on the Hebbian Learning Rule. A more powerful learning rule is the delta rule, which utilizes the discrepancy between the desired and actual output of each output unit to change the weights feeding into it.

1. Bechtel, W., & Abrahamsen, A. (1993). Connectionism and the mind: An introduction to parallel processing in networks. Oxford, UK: Blackwell.
2. Hebb, D.O. (1949). The organization of behavior. New York: Wiley.
3. Rumelhart, D.E., & McClelland, J. L.(1986). Parallel distributed processing: Explorations in the microstructure of cognition, vol. 1: Foundations. Cambridge, MA: MIT Press.

Humor

There are many reasons why people find something humorous, which are reflected in the large number of theories on the subject. Humor has been related to aggression, incongruity, and surprise. The cognitive psychologist's interest in the subject is usually related to the notion that humor stems from a resolution of incongruity.

For example, consider this joke by W.C. Field. "Do you believe in clubs for children?" "Only when kindness fails". Schultz(1974) offered a three step theory of processing. In the first stage, the listener notices the incorrect interpretation of the ambiguous element (clubs = social groups). In the second step, the incorrect element of incongruity is processed ( "only when kindness fails"). In the final stage the hidden meaning of the ambiguous element is perceived (clubs = sticks). The incongruity resolution theory explains the fact that a joke previously encountered will seem less funny on subsequent exposure.

Similarly, Freud (1905, in Minsky 1985) suggested that humorous stories are a way of fooling our internal censors. A joke's power comes from a description that fits two different frames at once. The first meaning must be transparent and innocent, while the second meaning is disguised and reprehensible.

Although most cognitive psychologists have not extended their theorizing to humor, it does have an important cognitive aspect. In particular, cognitive theory helps provide an explanation of why verbal jokes are found amusing by looking at the comprehension processes involved.

1. Kristal, L. (Ed.). (1981). ABC of psychology. London: Multimedia Publications.
2. Minsky, M. (1985). The society of mind. New York, NY: Simon & Schuster.
3. Schultz, T.R. (1974). Order and processing in humor appreciation. Canadian Journal of Psychology, 28, 409-420.

Imagery Debate

The imagery debate centers around the problem of what can be viewed as the primitives of cognition. Primitives serve as the foundation of the algorithmic level of the computational hierarchy. Presumably, it is these primitives which are implemented in the physical substrate of the brain.

The central question related to the imagery debate then is: Do images form the basis of all our higher cognition? If not, what does? Could propositions serve that function? Or both images and propositions? Or something altogether different?

1. Kosslyn, S. M., Pinker, S., Smith, G., & Shwartz, S. P. (1979). On the demystification of mental imagery. The Behavioral and Brain Sciences, 2, 535-581.

2. 
   Pylyshyn, Z. W. (1981). The imagery debate: Analogue media versus tacit knowledge. Psychological Review, 88, 16-45.

3. 
   Anderson, J. R. (1978). Arguments concerning representations for mental imagery. Psychological Review, 85, 249-277.

Incidental Learning

The incidental learning paradigm is an experimental paradigm used to investigate learning without intent. Using this paradigm, several groups of subjects are presented with the same list of items (e.g., 20 words) and are instructed to process them in different ways (different orienting conditions), with each group asked to perform a different activity or orienting task with the list. For example,

• count the number of letters in each word (shallow processing)
• name a rhyming word for each item (again, shallow processing, but deeper than #1
• form an image of each word and rate the vividness of each image (deep processing).

1. Eysenck, M.W. (1974). Age differences in incidental learning. Developmental Psychology, 10, 936-941.

Induction Learning

Inductive learning is essentially learning by example. The process itself ideally implies some method for drawing conclusions about previously unseen examples once learning is complete. More formally, one might state: Given a set of training examples, develop a hypothesis that is as consistent as possible with the provided data [1]. It is worthy of note that this is an imperfect technique. As Chalmers points out, "an inductive inference with true premises [can] lead to false conclusions" [2]. The example set may be an incomplete representation of the true population, or correct but inappropriate rules may be derived which apply only to the example set.

A simple demonstration of this type of learning is to consider the following set of bit-strings (each digit can only take on the value 0 or 1), each noted as either a positive or negative example of some concept. The task is to infer from this data (or "induce") a rule to account for the given classification:

Techniques for modeling the inductive learning process include: Quinlan's decision trees (results from information theory are used to partition data based on maximizing "information content" of a given sub-classification) [3], connectionism (most neural network models rely on training techniques that seek to infer a relationship from examples) and decision list techniques [4], among others.

1. Adapted from lectures in a graduate course in representation & reasoning given by Dr. Peter van Beek, Department of Computing Science, University of Alberta.
2. A.F. Chalmers. What is this thing called science?. University of Queensland Press, Australia, 1976.
3. J.R. Quinlan. C4.5: Programs for Machine Learning. Morgan Kaufmann, San Mateo, 1993.
4. R.L. Rivest. Learning decision lists. Machine Learning. 2(3):229-246, 1987.

Inductive Inference

Inferences are made when a person (or machine) goes beyond available evidence to form a conclusion. An inductive inference is one which is likely to be true because of the state of the world. Unlike deductive inferences, inductive inferences do yield conclusions that increase the semantic information over and above that found in the initial premises.

However, in the case of inductive inferences, we cannot be sure that our conclusion is a logical result of the premises, but we may be able to assign a likelihood to each conclusion.

Similar to deductive inference, induction can be broken down into three stages. The first stage is to understand the observation or stated information. The second is to form a hypothesis that attempts to describe the above information in relation to t person's general knowledge. The resulting conclusion goes beyond initial information by incorporating one's general knowledge in the result. The third step is to evaluate the validity of the conclusion that was reached.

1. Eysenck, M.W. (Ed.). (1990). The Blackwell dictionary of cognitive psychology. Cambridge, MA: Basil Blackwell.
2. Johnson-Laird, P. N. (1993). Human and machine thinking. Hillsdale, NJ : Lawrence Erlbaum Associates.

Intension

What a term means has two components: i) the referent of the term--this is 'class' talk, and is the component of meaning to which 'extension' applies; and ii) the sense of the term, i.e., all of the psychological associations that one has with that term--this is 'concept' talk. This second sense is referred to as the 'intension' of the term.

Examples of the two components follow. The referent of the term 'cat' is all the cats; the sense of the term is related to your experience of cats, their history, their attributes, etc. A classic example is 'the morning star' and 'the evening star'; both of which refer to the same thing, the planet 'Venus', but the sense of 'morning star' and 'evening star' is not the same. You cannot change the terms in a statement including one of them and retain the same truth value.

Other words sometimes used to pick out the distinctions between 'extension' and 'intension' are 'denotation' and 'connotation', respectively. Note the following definition by Cohen and Nagel:

A term [an element of a proposition] may be viewed in two ways, either as a class of objects (which may have only one member), or as a set of attributes or characteristics which determine the objects. The first phase or aspect is called the denotation or extension of the term, while the second is called the connotation or intension. The extension of the term 'philosopher' is 'Socrates', 'Plato', 'Thales', and the like; its intension is 'lover of wisdom', 'intelligent', and so on. (31)

1. Cohen, M. R. and Nagel, E. (1993). An Introduction to Logic. Indianapolis, Indiana: Hackett Publishing Company.

Intention

Intentionality refers to "aboutness." Beings having intentionality have propositional attitudes, they have beliefs, knowledge, hopes, dreams, desires, etc. about things. Whenever we come across "that" in an utterance or piece of writing, we know that we are dealing with something intentional. (Notice the intentionality of the preceding statement.) If we hear someone say "ouch," "oops," "hey," etc., these expressions do not reveal what sets humans apart from the rest of the animals. Intentionality does; it is considered by most to be a singularly human feature.

This issue is important to the extent that any theory of consciousness, or mind, must answer as to how intentionality is possible.

'Intentional' is not to be confused with 'intensional' spelled with an 's', the latter of which refers to the meaning of a term, (along with 'extensional'). Intentional, intensional, and extensional can be paired loosely in the following way: intentional/propositional, intensional/conceptual, and extensional/perceptual.

Intentional Stance

An intentional stance refers to the treating of a system as if it has intentions, irrespective of whether it does. By treating a system as if it is a rational agent one is able to predict the system's behavior . First, one ascribes beliefs to the system as those the system ought to have given its abilities, history and context. Then one attributes desires to the system as the system ought to have given its survival needs and means of fulfilling them. One can then predict the systems behavior  as that a rational system would undertake to further its goals given its beliefs. Dennett argues for three main reasons for taking an intentional stance. First, it fits well with our understandings of the processes of natural selection and evolution in complex environments. Second, it has been shown to be an accurate method of predicting behavior . Third, it is consistent with our folk psychology of behavior .

1. Dennett, D.C. (1987). The Intentional Stance Cambridge MA, MIT Press

Intermediate State Evidence

One of the key goals of cognitive science is to develop theories that are strongly equivalent with respect to to-be-explained systems. This requires that evidence be collected to defend the claim that the model and the to-be-explained system are carrying out the same procedures to compute a function.

One type of evidence that can be used to support this claim is intermediate state evidence. This involves observations of the intermediate steps, and/or the intermediate states of knowledge, that the two systems pass through as they move from being given a problem to providing an answer.

For example, if one was using a Turing machine as a model, then an immediate source of intermediate state evidence would be what the machine does to its tape with each processing step.

In studying human subjects, intermediate state evidence is not directly available. However, one method that might provide some evidence about these intermediate states is protocol analysis.

1. Pylyshyn, Z.W. (1984). Computation and cognition. Cambridge, MA: MIT Press.

Intrusion Errors

In a recall portion of a memory task, these are errors that occur when the subject includes items that were not on the original list.

Learning Rule

Learning rules, for a connectionist system, are algorithms or equations which govern changes in the weights of the connections in a network. One of the simplest learning procedures for two- layer networks is the Hebbian Learning Rule, which is based on a rule initially proposed by Hebb in 1949. Hebb's rule states that the simultaneous excitation of two neuron results in a strengthening of the connections between them. More powerful learning rules are learning rules which incorporate an error reduction procedure or error correction procedure (e.g., delta rule, generalized delta rule, back propagation). Learning rules incorporating an error reduction procedure utilize the discrepancy between the desired output pattern and an actual output pattern to change (improve) its weights during training. The learning rule is typically applied repeatedly to the same set of training inputs across a large number of epochs or training loops with error gradually reduced across epochs as the weights are fine-tuned.

1. Bechtel, W., & Abrahamsen, A. (1993). Connectionism and the mind: An introduction to parallel processing in networks. Oxford, UK: Blackwell.
2. Hebb, D.O. (1949). The organization of behavior. New York: Wiley.
3. Rumelhart, D.E., & McClelland, J. L.(1986). Parallel distributed processing: Explorations in the microstructure of cognition, vol. 1: Foundations. Cambridge, MA: MIT Press.

Levels of Processing

Levels of Processing - an influential theory of memory proposed by Craik and Lockhart (1972) which rejected the idea of the dual store model of memory. This popular model postulated that characteristics of a memory are determined by it's "location" (ie, fragile memory trace in short term store [STS] and a more durable memory trace in the long term store [LTS]. Instead, Craik and Lockhart proposed that information could be processed in a number of different ways and the durability or strength of the memory trace was a direct function of the depth of processing involved. Moreover, depth of processing was postulated to fall on a shallow to deep continuum.

Shallow processing (e.g., processing words based on their phonemic and orthographic components) leads to a fragile memory trace that is susceptible to rapid forgetting. On the other had, deep processing (e.g., semantic or meaning based processing) results in a more durable memory trace.

A typical paradigm employed to investigate the Levels of Processing theory is the incidental learning paradigm. Results reveal superior recall for items processed deeply compared to those items processed at the more shallow level (Eysenck, 1974: Hyde & Jenkins, 1969).

Craik and Lockhart also distinguished between two kinds of rehearsal, maintenance and elaborative rehearsal. Of the two, elaborative rehearsal is the most effective in producing a more durable memory trace.

1. Craik, F.I.M., & Lockhart, R.S. (1972). Levels of processing. A framework for memory research. Journal of Verbal Learning and Verbal behavior , 11, 671-684.
2. Eysenck, M.W. (1974). Age differences in incidental learning. Developmental Psychology, 10, 936-941.
3. Hyde, T.S., & Jenkins, J.J. (1969). Differential effects of incidental tasks on the organization of recall of a list of highly associated words. Journal of Experimental Psychology, 82, 472-481.

Linguistic Determination

Linguistic determination is the argument that language directly effects that way that people think about and see the world. Linguistic determination is also known as the Whorfian hypothesis or the Sapir-Whorf hypothesis (Sapir, 1968; Whorf, 1956). Whorf provides the example of the Eskimo words for snow. The Eskimo people are inhabitants of the Arctic. Whereas in the English language there is only one word for snow the Eskimo language has many words for snow. Whorf argues that this language for snow allows the Eskimo people to "see" snow differently than speakers of other languages who do not have as many words for snow. That is, Eskimo people see subtle differences in snow that other people do not.

Researchers have studied color perception across different linguistic groups to find support for the Whorfian hypothesis (Berlin & Kay, 1969; Heider, 1972; Heider & Oliver, 1973; Miller & Johnson-Laird, 1976; Rosch, 1974). The evidence indicates that people of all cultures perceive colour in the same way. The tentative conclusion is that language does not determine the way that people think. It is possible that language, whiule not determining the way that people think may influence the way that people think. Exactly how language might influence thought is yet unclear.

Long Term Potentiation

The enduring facilitation of synaptic transmission that occurs following the activation of a synapse by high-frequency stimulation of the presynaptic neuron. (Pinel, 1993, p.515)

1. Pinel, J. (1993). Biopsychology (2nd ed.). Toronto: Allyn & Bacon.

Machine Learning

The acquisition and application of knowledge plays a central role in describing learning. For the most part, human beings perform this task quite well (for better or worse). It is under the banner of machine learning that researchers, particularly within artificial intelligence, attempt to develop methods for accomplishing this task algorithmically (i.e. on computers).

Dietterich differentiates between three types of learning a system can exhibit [1]:

• Speed-up learning occurs when a system becomes more efficient at a task over time without external input.
• Learning by being told occurs when a system acquires new knowledge explicitly from an external source.
• Inductive learning occurs when a system acquires new knowledge that was neither explicitly nor implicitly available previously.

1. T.G. Dietterich. Machine learning. Annual Review of Computer Science. Vol. 4, Spring 1990.
2. L.G. Valient. A theory of the learnable. Communications of the ACM. 27:1134-1142, 1984.

Maintenance Rehearsal

Maintenance rehearsal is a type of rehearsal proposed by Craik and Lockhart (1972) in their Levels of Processing Model of memory. Maintenance rehearsal involves rote repetition of an item's auditory representation. In contrast to elaborative rehearsal, this type of rehearsal does not lead to stronger or more durable memories.

1. Craik, F.I.M., & Lockhart, R.S. (1972). Levels of processing. A framework for memory research. Journal of Verbal Learning and Verbal behavior , 11, 671-684.

Mandelbrot Set

A Mandelbrot set is an intricate geometric shape, where if any region of the set is magnified, new and intricate details appear. Every time you focus further on one section, more detail shows up. This will continue ad infinitum, as you investigate further. It was originally postulated to help explain fractals.

Another way of looking at this is as follows. When "simple" laws govern systems with large numbers of variables, the underlying order may become obscured by our inability to track every component. Simple rules can produce incredibly complex effects. Mandelbrot sets relate philosophically to the study of cognitive science, in that some theories in the field may need to be more complex in order to be fully validated, while other topics may be simpler than they first appear. This seems to be the case in the study of groups of agencies and agents in Minsky's (1985) The Society of Mind.

1. Cohen J., & Stewart, I. (1994). The collapse of chaos. New York: Viking Press.
2. Minsky, M. (1985). The society of mind. New York, NY: Simon & Schuster.

Memory Span

Memory span refers to the number of items (usually words or digits) that a person can hold in working memory. Tests of memory span are often used to measure working memory capacity. A typical test of memory span involves having an examiner read a list of random digits (digit span) or words (word span) aloud at the rate of one per second. At the end of a sequence, subjects are asked to recall the items in order. The average span for normal adults is 7 (Miller, 1956).

1. Miller, G.A. (1956). The magical number seven plus or minus two. Some limits on our capacity for processing information. Psychological Review, 63, 81-97.

Metaphor

Metaphor is the use of a word or phrase to label an object or concept that it does not literally denote, suggesting a comparison of that concept to the phrase's denoted object. There are many nuances in the meanings of metaphors. Mark Johnson and George Lakoff discuss preconceptual elements (which include: general human purposes, cultural institutions and practices, theoretical paradigms, individual traits and values, and personality traits). They claim that it is only because of these preconceptions that metaphor is able to affect our thinking, emotions and language. Earl Mac Cormac writes that the way in which we explain things influences how we understand them. While this relationship may initially appear backwards, the circularity can easily be withdrawn when one realizes that after the original clumsy description is given, we start trying to make the thing we are describing fit the model, which is only eliminated if it does not fit.

Misrepresentation

A Representation represents, or is about, a certain object or state of affairs (the representation9s object) and says something about that object (the representation9s content). Misrepresentation happens when what that content says about the object isn9t true of the object. For instance my cow representation has a certain content; suppose that this content is something like that9s a four-legged mammal that gives milk, goes 3moo2, and eats grass. Anything this representation 3is about2 will be represented as something that description applies to. So if my cow representation is activated by--and thus refers to--a short fat muddy horse seen from a distance, that horse is misrepresented, because it9s represented as a four-legged mammal that gives milk, goes 3moo2, and eats grass, which is false of the horse.

Theories of content, which attempt to explain how representations correctly represent their objects have a tremendous amount of trouble explaining how they can also sometimes misrepresent their objects. Jerry Fodor9s (1990) disjunction problem points out the difficulty here. A representation9s content can9t be such that the representation represents whatever causes its activation. A representation with content construed in this way can9t misrepresent.

Modularity

Jerry Fodor (1983) is the strongest proponent of a modular theory of cognition. Fodor argues that certain psychological processes are self contained--or modular. This is in contrast to "New look" or Modern Cognitivist positions which hold that nearly all psychological processes are interconnected, and freely exchange information.

Fodor proposes a three tiered cognitive system. The first level of the system, the transducer level, transforms environmental signals into a form that can be used by the cognizing organism. The second level, the input systems level, performs basic recognition and description functions. In Fodor's model input systems are modular. The third level of the system, higher level cognitive functions, performs complex operations on the output of the input systems. An example of a higher level process is analogous thinking.

Fodor holds that input systems are modular and that higher level cognitive processes are non-modular. This means that all of the information necessary for performing their tasks of recognition and description is contained within the input systems. For example, object perception might be modular, in which case the object perception module need not reference language modules, or music modules, or mathematics modules in order to perform its operations. In contrast, higher level processes have access to all information contained within the cognitive system when performing a given operation. Fodor provides the example of scientific reasoning (a higher level cognitive process). Potentially, when solving a scientific problem, the scientist can reference any knowledge that he or she has about the world to help in solving this problem. As such, if necessary, knowledge about botany can be referenced in order to understand problems in mathematics.

Modular systems have the following properties:

1. They are domain specific--they operate on, and have a computational architecture that is unique to certain stimuli.
2. Their operation is mandatory, or they are cognitively impenetrable--beliefs cannot affect the operations of modules, we cannot help seeing, or hearing the world in a certain way.
3. Modules are fast--modular processes are among the fastest psychological processes, this is because modules are self-contained and need not spend time referencing information outside of the module to complete their tasks.
4. Modules are informationally encapsulated--they need not reference any other psycholgical systems in order to perform their operations.
5. Modules have shallow outputs--the output of modules is very basic, more complex representations follow after higher level computation.

1. Fodor, J.A. (1983). The modularity of mind. Cambridge, MA: MIT Press.
2. Fodor, J.A. (1985). Precis on The Modularity of Mind. Behavioral and Brain Sciences, 8, 1-42.

Neurocognition

The study of the relationships between neuroscience and cognitive psychology.

The goal is to look for specific neurophysiological correlates of cognitive functions. This is based on the assumption that specific brain regions are responsible for mediating certain aspects of cognitive function.

1. Pinel, J. (1993). Biopsychology (2nd ed.). Toronton: Allyn & Bacon.

Neuron

These are the specialized, functional cells of the nervous system that conduct neural information.

There were originally 2 basic hypotheses about the structure and function of the nervous system (Kolb & Whishaw, 1985, p.317):

1. Neuron Hypothesis: the nervous system is composed of discrete, autonomous cells, or units, that can interact but are not physically connected.
2. Nerve Net Hypothesis: the nervous system is composed of a continuous network of interconnected fibres.

The parallel distributed processing model, is a good example of a model that has attempted to account for the basic neural properties.

1. Kolb, B., & Whishaw, I. (1985). Fundamentals of human neuropsychology (2nd ed.). New York: W.H. Freeman & Co.
2. Pinel, J. (1993). Biopsychology (2nd ed.). Toronto: Allyn & Bacon.

Neuroscience

Neuroscience is the study of the nervous system and has many different branches, such as:

• Biopsychology,
• Developmental Neurobiology,
• Neuroanatomy,
• Neurochemistry,
• Neuroendocrinology,
• Neuroethology,
• Neuropharmacology,
• Neurophysiology, and
• Neuropsychology.

Occam's Razor

The simplest definition of Occam's Razor is "Don't make unnecessarily complicated assumptions". It can be used as a philosophical way of sorting the simple theories from the complicated ones. When scientists select theories, they don't just use the criterion of agreement or disagreement with observations. They also have aesthetic principles, and a desire for an elegant, universal theory. They use these aesthetic principles to remove the cloud of trivially competing theories that necessarily surround every theory. Occam's razor is a working rule of thumb, not the ultimate answer.

1. Cohen J., & Stewart, I. (1994). The collapse of chaos. New York: Viking Press

Paradigm

The Oxford English Dictionary defines a paradigm simply as an "example or pattern". Within the scientific community however, the notion of paradigm is a far more significant issue. It typically defines what a given individual is willing to accept of his or her field, and how they perform their own work within it---whether they are conscious of it or not. It is here in fact that the more formal concept of a paradigm is realized.

Chalmers [2], in a discussion of Kuhn's writings about what constitutes a shift in paradigm [3], loosely characterizes it as a framework of beliefs and standard which defines legitimate work within the science for which it applies. He states further that defining "paradigm" rigorously is inherently problematic. He does however offer some suggestions for what, at least in part, characterizes a paradigm; although worded with science in mind, some of these can be seen to apply to the concept of a paradigm in general.

A paradigm (from Chalmers [2]):

• is composed of "explicitly stated laws and theoretical assumptions".
• includes "standard ways of applying the fundamental laws to a variety of types of situations".
• possess "instrumentation and instrumental techniques necessary for bringing the laws of the paradigm to bear on the real world".
• "consists of some very general, metaphysical principles that guide work within the paradigm".
• "contains some very general methodological prescriptions".

1. D. Bohm and F.D. Peat. Science, Order, and Creativity. Bantam Books, New York, 1987.
2. A.F. Chalmers. What is this thing called science?. University of Queensland Press, Australia, 1976.
3. T.S. Kuhn. The Structure of Scientific Revolutions. University of Chicago Press, Chicago, 1970.

Parallel Distributed Processing Models

Parallel Distributed Processing (PDP) models are a class of neurally inspired information processing models that attempt to model information processing the way it actually takes place in the brain.

This model was developed because of findings that a system of neural connections appeared to be distributed in a parallel array in addition to serial pathways. As such, different types of mental processing are considered to be distributed throughout a highly complex neuronetwork.

The PDP model has 3 basic principles:

1. the representation of information is distributed (not local)
2. memory and knowledge for specific things are not stored explicitly, but stored in the connections between units.
3. learning can occur with gradual changes in connection strength by experience.

1. Rumelhart, D.E., Hinton, G.E., & McClelland, J.L. (1986). A general framework for parallel distributed processing. 

2. D. E. Rumelhart, J. L. McClelland, and the PDP Research Group (Eds.). Parallel distributed processing: Explorations in the microstructure of cognition. Vol. 1: Foundations. Cambridge, MA: MIT Press.

Parallel Search

Serial Search

Perseveration Errors

On a recall portion of a memory task, these are errors that occur when a subject repeats items that they have already said on that same recall trial.

Philosophy of Mind

The philosophy of mind has emerged as a field of philosophy in its own right, due to the convergence of issues raised in more traditional areas of philosophy, such as metaphysics, epistemology, and ethics.

Some questions asked by philosophers of mind reveal these origins. One might ask: Are mind and body one substance?; Does mind depend on the body?; Is 'mind' identical with 'body'? These questions may lead to others: Do humans actually make free choices, or are all human acts physically determined? As well as physical states, we have mental states and many of the latter relate to each other. For example, individuals have beliefs, desires, and feelings about other mental states, i.e., about concepts. When talk turns to such intentional states or propositional attitudes, further questions arise. Do only humans have intentionality? Must any account which attempts to explain our actions consider intentionality? Or can physical events (brain and body processes in interraction with the physical environment) wholly explain our actions?

Because of the nature of these questions, it becomes apparent why the philosophy of mind might cross over into cognitive science. Cognitive science, after all, tries to answer many of these same questions.

Piaget's Stage Theory of Development

Piaget was among other things, a psychologist who was interested in cognitive development. After observation of many children, he posited that children progress through 4 stages and that they all do so in the same order. These four stages are described below.

The Sensori-motor Period (birth to 2 years)

During this time, Piaget said that a child's cognitive system is limited to motor reflexes at birth, but the child builds on these reflexes to develop more sophisticated procedures. They learn to generalize their activities to a wider range of situations and coordinate them into increasingly lengthy chains of behavior .

Pre-Operational Thought (2 to 6/7 years)

At this age, according to Piaget, children acquire representational skills in the areas mental imagery, and especially language. They are very self-oriented, and have an egocentric view; that is, preoperational children can use these representational skills only to view the world from their own perspective.

Concrete Operations (6/7 to 11/12 years)

As opposed to Preoperational children, children in the concrete operations stage are able to take another's point of view and take into account more than one perspective simultaneously. They can also represent transformations as well as static situations. Although they can understand concrete problems, Piaget would argue that they cannot yet perform on abstract problems, and that they do not consider all of the logically possible outcomes.

Formal Operations (11/12 to adult)

Children who attain the formal operation stage are capable of thinking logically and abstractly. They can also reason theoretically. Piaget considered this the ultimate stage of development, and stated that although the children would still have to revise their knowledge base, their way of thinking was as powerful as it would get.

1. Santrock, J.W. (1995). Children. Dubuque, IA: Brown & Benchmark.
2. Siegler, R. (1991). Children's thinking. Englewood Cliffs, NJ: Prentice-Hall.
3. Vasta, R., Haith, M.M., & Miller, S.A. (1995). Child psychology: The modern science. New York, NY: Wiley.

Primacy Effect

The primacy effect is found when the results of a free recall task are plotted in the form of a serial position curve. Generally, this curve is U-shaped, and the primacy effect corresponds to the tail of the U on the left. This tail indicates that words presented at the start of a list of to-be-remembered items are better remembered than words presented in the middle of this list. It is called the primacy effect because these items were the ones presented first to the subject in the memory experiment.

The primacy effect appears to be the result of subjects recalling items directly from a semantic memory. This is because the primacy effect can be sharply attenuated by performing manipulations that adversely affect this system -- such as using fast presentation of items (which does not permit much elaborative rehearsal to transfer memories from short-term to long-term stores), or by using list items that have similar meanings (and thereby producing semantic confusions).

The primacy effect was important to cognitive science because it provided empirical evidence for the decomposition of memory into an organized set of subsystems, which is required by functional analysis.

Priming

Priming is discussed in the context of the activation theory. It is assumed that concepts that have some relation to each other are connected in some mental network, so that if one concept is activated, then concepts related to it are also activated.

Priming is a phenomenon related to this concept. It can be shown in the following example:

A subject is shown the word nurse. Presumably the subject will then think of other words related to the word nurse. If the subject is then shown either the word doctor or the word butter, the subject should be able to read the former word more quickly than the latter word because doctor is related to nurse and therefore has been recently accessed, and so more familiar to the subject.

The word nurse then serves to "prime" the second word, doctor.

Primitive

A primitive is a basic building block of a system. Complex systems can be decomposed into simpler things, but primitives -- by definition -- cannot.

To provide an example that gives a nice intuition about what a primitive is, consider teaching a child the meanings of different words. If a child asks us "What does `bachelor' mean?", we might break "bachelor" down into other meanings ("`Bachelor' means that someone is a `man' who is `not married'"). However, if a child asks us "What does `red' mean?", we are not likely to do this, because it is difficult to decompose such a basic term. Instead, we are more likely to point to different things that are `red'. In this sense, `red' represents something that we might call a semantic primitive (a basic meaning), while `bachelor' does not.

Primitives are important in cognitive science because of its tendency to view information processors functionally instead of physically. Because of this view, researchers use a methodology called functional analysis to decompose a complex information processor into simpler, functional components. However, if this decomposition is not stopped, the functional analysis goes on indefinitely and falls prey to Ryle's Regress. This means that the functional analysis is not explanatory. Researchers try to escape Ryle's regress by identifying a set of primitive functions which cannot be further decomposed. This set of functions is the functional architecture for cognition.

Production

A production system is program that comprises a series of conditional statements that specify what action is to be taken under certain circumstances. These 'If ... then ...' statements are known as productions. Each production has a condition and an action. If the condition is found to be true by the system then the action will be performed. For example, a production system for a thermostat may contain a production such as the following.

1. temperature > 70 and temperature < 72 ----> stop

Information from the environment is compared to the conditions of the production. If the condition to the left of the arrow is true then the process to the right of the arrow will be performed. In the above example will the thermostat will stop as long as the temperature remains within the range of 70 and 72 degrees. If the temperature is outside that range then a different production will be activated and the system will change behavior .

1. Newell, A., & Simon, H.A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice-Hall.

Production System

A production system is program that comprises a series of conditional statements that specify what action is to be taken under certain circumstances. These 'If ... then ...' statements are known as productions. For example, a production system for a cricket batsman may comprise a series of productions such as the following.

1. ball outside off stump ------> no action
2. ball pitched on wicket and good length ------> forward defensive stroke
3. ball pitched short on leg side and fast------> duck
4. ball pitched short on leg side and slow------> hook

1. Newell, A., & Simon, H.A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice-Hall.

Proposition

The proposition is a concept borrowed by cognitive psychologists from linguists and logicians. The proposition is the most basic unit of meaning in a representation. It is the smallest statement that can be judged either true or false. Anderson (1990) gives the following example of a sentance divided up into its constituent propositions:

"Nixon gave a beautiful Cadillac to Brezhnev, who was the leader of the USSR."

This sentence can be divided into three propositions:

1. Nixon gave a Cadillac to Brezhnev.
2. The Cadillac was beautiful.
3. Brezhnev was the leader of the USSR.

1. Anderson, J.R. (1990). Cognitive psychology and its implications (3rd ed.). New York: W. H. Freeman.

Protocol Analysis

Protocol analysis is one experimental method that can be used to gather intermediate state evidence concerning the procedures used by a system to compute a function. In protocol analysis, subjects are trained to think aloud as they solve a problem, and their verbal behavior  forms the basic data to be analyzed. The first step of a protocol analysis is to obtain, and then transcribe, a verbal protocol. The next step is to take the protocol and use it to infer the subject's problem space (i.e., infer the rules being used, as well as various knowledge states concerning the problem). The third step is to create a problem behavior  graph, which reflects state transitions as subjects search through the problem space in their attempt to solve the problem. Finally, the problem behavior graph is used to create a computer simulation (typically created as a production system) that will solve the problem. By comparing, in detail, the behavior  of the simulation to the verbal protocol, one can validate the assumptions that led to the program's creation. In turn, the program provides a rich description of an individual's processing steps, and transitions in knowledge,during the problem-solving process.

1. Ericsson, K.A., & Simon, H.A. (1984). Protocol analysis: Verbal reports as data. Cambridge, MA: MIT Press.
2. Newell, A., & Simon, H.A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice-Hall.

Recency Effect

The recency effect is found when the results of a free recall task are plotted in the form of a serial position curve. Generally, this curve is U-shaped, and the recency effect corresponds to the tail of the U on the right. This tail indicates that words presented at the end of a list of to-be-remembered items are better remembered than words presented in the middle of this list. It is called the recency effect because these items were the ones presented most recently to the subject in the memory experiment.

The recency effect appears to be the result of subjects recalling items directly from the maintenance rehearsal loop used to keep items in primary memory. In other words, it reflects short-term memory for items. This is because the recency effect can be sharply attenuated by performing manipulations that adversely affect such rehearsal -- such as delaying recall of list items with a distractor task, or by using list items that have similar sounds.

The recency effect was important to cognitive science because it provided empirical evidence for the decomposition of memory into an organized set of subsystems, which is required by functional analysis.

Recognition Recall

This is a variation of the recall portion of a memory task. The subject is not required to explicitly state the items, but instead, they must simply identify which items (from a larger group of items) were on the original list.

For instance, the subject may be read a large list of items and be asked to say "YES" if the item was on the list, and say "NO" if it was not on the list.

This task is slightly easier than the cued or free recall task. The answers provided by the subject fall into 4 categories:

1. HITS: These are the responses that correctly identify items as being from the original list when they actually are.
2. CORRECT NEGATIVES: These are the responses that correctly state an item as not being on the original list when it actually was not.
3. MISSES: These are the responses that fail to identify a word as being from the original list when it was.
4. FALSE POSITIVES: These are responses that incorrectly identify items as being from the original list when they were not on that list.

Recursive Decomposition

Recursive decomposition (Palmer & Kimchi, 1986) refers to the process whereby any complex informational event at one level of description can be specified more fully at a lower level of description by decomposing the event into:

• a number of components and
• processes that specifiy the relations among these components

1. Medin, D.L., & Ross, B.H. (1992). Cognitive psychology. Fort Worth, TX: Harcourt Brace Jovanovich.
2. Palmer, S. & Kimchi, R. (1986). The information approach to cognition. In T. Knapp, & L. Robertson (Eds.), Approaches to cognition. Hillsdale NJ: Erlbaum.

Relative Complexity Evidence

One of the key goals of cognitive science is to develop theories that are strongly equivalent with respect to to-be-explained systems. This requires that evidence be collected to defend the claim that the model and the to-be-explained system are carrying out the same procedures to compute a function.

One type of evidence that can be used to defend this claim is called relative complexity evidence. Imagine that someone is proposing that a Turing machine is a strongly equivalent model of how children do mental arithmetic. To collect relative complexity evidence concerning this claim, we could present a number of different addition problems to the Turing machine, and then rank order them in terms of the number of processing steps that each problem required. We could then present the same problems to a group of children, and rank order the difficulty they caused the children on the basis of reaction time taken to solve the problems. If the two systems are strongly equivalent, then we would expect the same rank-orderings to be obtained for both the Turing machine and the children. If they are not strongly equivalent (as we would expect in this example), then different rank-orderings would emerge because different procedures are used to solve the problems.

1. Pylyshyn, Z.W. (1984). Computation and cognition. Cambridge, MA: MIT Press.

Retrieval

Retrieval refers to the processes through which we recover items from memory.

Ryle's Regress

Ryle's Regress is a classic argument against cognitivist theories, and concludes that such theories cannot be scientific. The philosopher Gilbert Ryle (1949) was concerned with critiquing what he called the intellectualist legend, which required intelligent acts to be the product of the conscious application of mental rules. Ryle (p. 31) argued that the intellectualist legend results in an infinite regress of thought:

According to the legend, whenever an agent does anything intelligently, his act is preceded and steered by another internal act of considering a regulative proposition appropriate to his practical problem. [...] Must we then say that for the hero's reflections how to act to be intelligent he must first reflect how best to reflect how to act? The endlessness of this implied regress shows that the aplication of the appropriateness does not entail the occurrence of a process of considering this criterion.

Cognitive scientists must be constantly aware of Ryle's Regress as a potential problem with their theories, and must ensure that their theories include a principled account of how the (potentially) infinite regress that emerges from functional analysis can be stopped. This is why the identification of the functional architecture is one of the fundamental goals of cognitive science.

1. Guthrie, E.R. (1935). The psychology of learning. New York: Harper
2. Ryle, G. (1949). The concept of mind. London: Hutchinson & Company.
3. Tolman, E.C. (1932). Purposive behavior in animals. New York: Century Books.
4. Tolman, E.C. (1948). Cognitive maps in rats and men. Psychological Review, 55, 189-208.

Sapir-Whorf Hypothesis

See Linguistic Determination

Schemas

A schema representation is a way of capturing the insight that concepts are defined by a configuration of features, and each of these features involves specifying a value the object has on some attribute. The schema represents a concept by pairing a class of attribute with a particular value, and stringing all the attributes together. They are a way of encoding regularities in categories, whether these regularities are propositional or perceptual. They are also general, rather than specific, so that they can be used in many situations.

1. Anderson, J.R. (1990). Cognitive psychology and its implications. New York, NY: Freeman.

Semantics

Semantics deals with the relationship between representations and the world. Anything which can said to be a representation--which could be said to stand for, represent, point to, indicate, mean, refer to, or in some way be about something else--has semantic relations to that something else. Semantics is what makes the word Coffee9 mean that smelly muddy brown hot liquid that people drink.

A representation's semantic properties are those properties the representation has in virtue of the sort of relationship the representation has with a part of the world. So when we talk about what object (the thing in the world) represents, or whether the representation is a true representation of its object or whether it's a highly inaccurate representation of that object, or whether it misrepresents that object, we're talking about the representation's semantic properties.

The problem is that if cognitive scientists define the essence of cognition as processes operating on representations, then any process which operates on a representation has no access to that representation's semantic properties. Fodor9s (1990) Formality Condition maintains that any process which operates on a representation can only operate on the representation's non-semantic or formal properties.

The idea, then, is that if a process which operates on a representation is to be sensitive to the semantic properties of the representation, such as what object it represents, then that representation9s semantic properties must somehow be mirrored in the representation's syntactic properties. So my cow representation must be fairly complex, and somehow 3contain2 formal descriptions of all the properties I ascribe to cows, so that processes which operate on this representation (such as those which allow me to utter 3Cows give milk,2) can operate on those properties.

But whether the properties I ascribe to cows in such formal descriptions are true of cows is inaccessible to those processes. Whether what I believe is true or not is a semantic property of that representation9s relationship with the world. And semantic properties like truth are transparent to the processes that operate on my representations. Perhaps the best we can hope is that the formal properties of all my representations are consistent, and form a coherent network of beliefs that facilitate my acting successfully in my environment. Whether these are true or not is inaccessible to the brain-processes which operate on those representations. (Hence what Fodor (1980) calls 3Methodological Solipsism2)

1. Fodor, J. (1980). 3Methodological Solipsism Considered as a Research Strategy in Cognitive Psychology2. behavior  and Brain Sciences, 3(1), 63-109.
2. Fodor, J. (1978). 3Tom Swift's Procedural Grand-mother2. Cognition, 6.

Serial Position Curve

The serial position curve is used to plot the results of a free recall experiment. The x-axis of this curve indicates the serial position of to-be-remembered items in the list (e.g., the first item, the second item, the third item, and so on). The y-axis of this curve indicates the probability of recall for the item, which is typically obtained by averaging across a number of subjects

The serial position curve is important to cognitive science because it revealed two effects, the recency effect and the primacy effect, which were fundamentally important pieces of evidence for the functional decomposition of "memory" into an organized set of subsystems.

Serial Search

A type of memory search in which information is retrieved one piece after another. Serial searches are represented by a linear function. That is, when retrieval time is plotted against the number of items to be retrieved the slope of the graph is constant, and is equivalent to the amount of time that it takes to retrieve a single piece of information.

Serial memory search is often contrasted with parallel memory search in which a number of pieces of information are retrieved at the same time. Graphically, the slope of the line representing parallel search is zero. That is, as the number of items to be retrieved increases the amount of time that it takes to retrieve these items remains constant.

Sternberg (1966, 1969a, 1969b, 1975) argued that retreivel from short term memory relies upon serial type searches, whereas retrieval from long term memory relies upon parallel type searches.

Short Term Memory

Generally cognitive psychologists divide memory into three stores: sensory store, short-term store, and long-term store. After entering the sensory store, some information proceeds into the short-term store. This short-term store is commonly referred to as short-term memory.

Short-term memory has two important characteristics. First, short-term memory can contain at any one time seven, plus or minus two, "chunks" of information. Second, items remain in short-term memory around twenty seconds. These unique characteristics, among others, suggested to researchers that short-term memory was autonomous from sensory and long-term memory stores

Craik and Lockhart (1972) argued short-term memory was not autonomous from the other memory systems. They suggested that short-term memory and long-term memory were different manifestations of a single, underlying memory system.

As an alternative to short-term memory Baddely and Hitch have proposed the concept of a working memory. As in traditional models of short-term memory, working memory is limited in the amount of information that it can store, and the length of time that it can store information.

Spontaneous Generalization

Connectionist networks may be designed so that they can retrieve information from cues that are too vague to match a particular memory and provide a generalized picture of what is common to the memories that match the cues. Thus the network has the ability to generalize about classes of memories as part of its architecture.

1. Bechtel, W., & Abrahamsen, A. (1991). Connectionism and the mind: An introduction to parallel processing in networks. Cambridge, MA: Blackwell.

Strong Equivalence

Strong equivalence is a stronger condition for model validation than is weak equivalence. If two systems are strongly equivalent then

1. they compute the same function (i.e., they are weakly equivalent),
2. they use the same program to compute this function, and
3. this program is written in the same programming language (i.e., the two systems have the same functional architecture).

As far as "algorithmic" approaches to cognitive science are concerned (e.g., experimental psychology, psycholinguistics), the aim of the discipline is to generate strongly equivalent theories of people. This requires collecting evidence to support the claim that a simulation uses the same procedures to solve a problem as do human subjects, as well as evidence to support the claim that a proposed architecture is primitive. It is not surprising, then, that the search for strongly equivalent theories is a formidable (but necessary) challenge for cognitive scientists.

1. Pylyshyn, Z.W. (1984). Computation and cognition. Cambridge, MA: MIT Press.

Sustained Attention

Sustained attention is "the ability to direct and focus cognitive activity on specific stimuli." In order to complete any cognitively planned activity, any sequenced action, or any thought one must use sustained attention. An example is the act of reading a newspaper article. One must be able to focus on the activity of reading long enough to complete the task. Problems occur when a distraction arises. A distraction can interrupt and consequently interfere in sustained attention.

DeGangi and Porges (1990) indicate there are 3 stages to sustained attention which include:

1. attention getting,
2. attention holding, and
3. attention releasing.

1. DeGangi, G., & Porges, S. (1990). Neuroscience foundations of human performance. Rockville, MD: American Occupational Therapy Association

Symbolic Architecture

Symbolic architecture refers to the classical view of the architecture of the mind. In this approach the mind is viewed as a process in which symbols are manipulated. Symbols are moved between memory stores such as long term and short term memory and are acted upon by an explicit set of rules in a particular sequence. The symbolic architecture is the manner in which memory stores are related and the set of rules applied to the system.

The symbolic architecture approach has been widely applied and formed the basis of influential work such as Newell & Simon's Human Problem Solving. More recently, this approach to cognitive architecture has been challenged by the connectionist architecture approach.

1. Collins, A. & Smith, E.E. (Eds.). (1988). Readings in cognitive science: A perspective from psychology and artificial intelligence. San Mateo, CA: Morgan Kaufman.

Top-Down Processing

The cognitive system is organized hierarchically. The most basic perceptual systems are located at the bottom of the hierarchy, and the most complex cogntive (e.g. memory, problem solving) systems are located at the top of the hierarchy.

Information can flow both from the bottom of the system to the top of the system and from the top of the system to the bottom of the system. When information flows from the top of the sytstem to the bottom of the system this is called "top-down processing".

The implications of this top to bottom flow of information is that information coming into the system (perceptually) can be influenced by what the individual already knows about the information that is coming into the system (as information about past experiences are stored in the higher levels of the system).

Extreme versions of top-down processing argue that all information coming into the system is affected by what is already known about the world. An alternative version is offered by Jerry Fodor (1983). In his theory of modularity, Fodor argues that top-down processing occurs only in some parts of the cognitive system at certain times. Fodor rejects the idea that all stored information can potentially effect all incoming information.

1. Fodor, J.A. (1983). The modularity of mind. Cambridge, MA: MIT Press.

Turing Equivalence

Turing equivalence is another term for describing weak equivalence.

Turing Test

The Turing test is a behavior al approach to determining whether or not a system is intelligent. It was originally proposed by mathematician Alan Turing, one of the founding figures in computing. Turing argued in a 1950 paper that conversation was the key to judging intelligence. In the Turing test, a judge has conversations (via teletype) with two systems, one human, the other a machine. The conversations can be about anything, and proceed for a set period of time (e.g., an hour). If, at the end of this time, the judge cannot distinguish the machine from the human on the basis of the conversation, then Turing argued that we would have to say that the machine was intelligent.

There are a number of different views about the utility of the Turing test in cognitive science. Some researchers argue that it is the benchmark test of what Searle calls strong AI, and as a result is crucial to defining intelligence. Other researchers take the position that the Turing test is too weak to be useful in this way, because many different systems can generate correct behavior s for incorrect (i.e., unintelligent) reasons. Famous examples of this are Weizenbaum's ELIZA program and Colby's PARRY program. Indeed, the general acceptance of ELIZA as being "intelligent" so appalled Weizenbaum that he withdrew from mainstream AI research, which he attacked in his landmark 1976 book.

1. Colby, K.M. et al. (1972) Artificial paranoia. Artificial Intelligence, 2, 1-26.
2. Colby, K.M. et al. (1973) Turing-like undistinguishability tests for the validation of a computer simulation of paranoid processes. Artificial Intelligence, 3, 47-51.
3. Turing, A.M. (1950). Computing machinery and intelligence. Mind, 59, 433-560.
4. Weizenbaum, J. (1976). Computer power and human reason. San Francisco, CA: W.H. Freeman.

Veridicality

Veridicality is the extent to which a knowledge structure accurately reflects the information environment it represents. This is a construct of interest as our understanding of the relationship between knowledge structures and information environments is weak. In particular, the optimal level of veridicality is problematic. The value of a knowledge structure lies in its ability to simplify an environment, yet simplification increases the probability of a false characterisation and hence error. The study of veridicality is concerned with investigating the consequences of this trade off between accuracy and efficiency.

1. Walsh, J.P., Henderson, C.M. & Deighton,J. (1988). Negotiated belief structures and decision performance: An empirical investigationOrganizational Behavior and Human Decision Processes. 42, 194-216

Visuospatial Perception

This is one component of cognitive functioning and it refers to our ability to process and interpret visual information about where objects are in space.

This is an important aspect of cognitive functioning because it is responsible for a wide range of activities of daily living.

For instance, it underlies our ability to move around in an environment and orient ourselves appropriately. Visuospatial perception is also involved in our ability to accurately reach for objects in our visual field and our ability to shift our gaze to different points in space.

The association areas of the visual cortex are separated into two major component pathways, and are believed to mediate different aspects of visual cognition. In humans, the parieto-occipital region is believed to process visuospatial and visual motion types of information. Conversely, the inferotemporal region of the brain is believed to mediate our ability to process visual information about the form and color of objects.

1. Kolb, B., & Whishaw, I. (1985). Fundamentals of human neuropsychology (2nd ed.). New York: W.H. Freeman.
2. Pinel, J. (1993). Biopsychology (2nd ed.). Toronto: Allyn & Bacon.

Visuospatial Sketchpad

The visuospatial sketchpad or scratchpad (VSSP) is one of two passive slave systems in Baddeley's (1986) model of working memory. The VSSP is responsible for the manipulation and temporary storage of visual and spatial information. To date, more is known about the second slave system, the articulatory loop, than about visual coding in memory.

1. Baddeley, A. (1986). Working memory. Oxford: Clarendon Press.

WAIS

The Wechsler Adult Intelligence Scale (WAIS) was developed by Wechsler in 1955. An updated version of the scale (WAIS-R) was developed in 1981. WAIS measures global or general intelligence and is commonly used by psychologists. It is divided into two parts: the verbal scale and the performance scale. Each of these two parts is further divided into subtests, each of which taps a specific verbal or nonverbal skill. Each subtest has items ranging from easy to increasingly more difficult.

Verbal subtests measure "our store of knowledge" (Belsky, 1990, p. 120). They focus on

learned or absorbed knowledge [testing] knowledge of historical, literary or biological facts; knowledge relating to competent functioning in the world; knowledge of mathematics; knowledge of the meaning of specific words.

Belsky further asks if the dramatic age decline is confined mainly to particular subtests. Would we see the same age loss if we looked at data other than the cross-sectional studies used to determine the norms? (p. 121).

1. Belsky, J.K. (1990). The psychology of aging theory, research, and interventions. Pacific Grove, CA: Brooks/Cole.

Weak Equivalence

Weak equivalence is a relationship between the outputs of two systems that are being compared. If these systems are only weakly equivalent, then we can say that they are computing the same function (or generating the same external behavior), but that they are using different procedures to do so. For example, human chess players and computer chess players are weakly equivalent, in the sense that they both play the game of chess, but use very different procedures to decide which move to make next in a game. (Computer chess players usually use some form of intensive search, which is beyond the memory capacity of human players. Indeed, an interesting question is how humans can play chess so well given that they do not use brute force search methods!)

Weak equivalence is important in cognitive science in two respects. First, it is the kind of comparison that the Turing test offers, which is why it is also sometimes called Turing equivalence. Second, although weak equivalence is necessary for validating theories in cognitive science, it is not sufficient. This is because while it is required of theories or simulations in cognitive science that they compute the same functions as the to-be-explained system, it is also crucial that they compute these functions in the same way. This later requirement is called strong equivalence.

1. Pylyshyn, Z.W. (1984). Computation and cognition. Cambridge, MA: MIT Press.

Wernicke's Area

Named for Carl Wernicke who first described it in 1874, Werenicke's area appears to be crucial for language comprehension. People who suffer from neurophysiological damage to this area (called Wernicke's aphasia or fluent aphasia) are unable to understand the content words while listening, and unable to produce meaningful sentences; their speech has grammatical structure but no meaning.

Auditory and speech information is transported from the auditory area to Wernicke's area for evaluation of significance of content words, then to Broca's area for analysis of syntax. In speech production, content words are selected by neural systems in Wernicke's area, grammatical refinements are added by neural systems in Broca's area, and then the information is sent to the motor cortex, which sets up the muscle movements for speaking.

1. Gray, Peter. (1994). Psychology. New York, NY: Worth Publishing.

Working Memory

Working memory, the more contemporary term for short-term memory, is conceptualized as an active system for temporarily storing and manipulating information needed in the execution of complex cognitive tasks (e.g., learning, reasoning, and comprehension). There are two types of components: storage and central executive functions (see Baddeley, 1986 for a review). The two storage systems within the model (the articulatory loop [AL] and the visuospatial sketchpad or scratchpad [VSSP] are seen as relatively passive slave systems primarily responsible for the temporary storage of verbal and visual information (respectively).

The most important, and least understood, aspect of Working Memory is the central executive, which is conceptualized as very active and responsible for the selection, initiation, and termination of processing routines (e.g., encoding, storing, and retrieving).

1. Baddeley, A. (1986). Working memory. Oxford: Clarendon Press.

Z Lens

The Z Lens is a sophisticated piece of apparatus developed by Roger Sperry and his associates in 1955 to enable them to project visual stimuli onto the retina of the eye so that they are interpreted either by the left or right hemisphere of the brain, not both at once. Sperry, a pioneer of the split brain operation, used it to demonstrate that split brain patients had two separate visual inner worlds. If the picture of an object was presented to the left hemisphere, the patient recognized it when it was presented again to the same hemisphere. However, if the same object was presented to the other half of the visual field, the patient had no recollection of having seen it before.

1. Kristal, L. (Ed.). (1981). ABC of psychology. London: Multimedia Publications.