Friday, March 30, 2012

Learning Concepts (Part III)


Productions
Production rules are a primary component of many contemporary computer models of cognition (e.g., ACT, GPS, Soar). A production has the form: If THEN When the current state of memory matches the side of the rule, the specified is carried out. The action could be any form of mental processing. Productions can also generate new productions giving rise to new cognitive rules (c.f., creativity).
Flow of control in a production system goes through the set of productions sequentially until a condition is matched. After executing the action, the system continues with the next production or returns to the beginning of the set. This sequence is repeated until a terminal goal condition is satisfied. Thus, production systems require no executive level of control; all control is determined by the productions. Clearly, order of productions in the set is important since it determines which actions are satisfied first.
It is possible to add constraints to productions that alter the strict sequential order and hence introduce some form of higher level control. For example, preference can be given to conditions according to recency or frequency of occurence. Productions can be limited to firing only once for a given condition (rule of refractoriness). Or, goal symbols can be added to the conditions that must be satisfied in order for the production to be satisfied.
Productions map very closely onto the notion of rules found in many cognitive theories and hence are a natural representation to use when building computer models of such theories. They also resemble the S-R associations of behavioral theories, except that production rules do not normally encompass any notion of strength; they are all or none. However, some theorists have allowed individual production rules to have probabilities of executing based upon frequency of use or characteristics of the conditions.
References:
Klahr, D., Langley, P. & Neches, R. (1987). Production System Models of Learning and Development. Cambridge, MA: MIT Press

Feedback/Reinforcement

Feedback and reinforcement are two of the most pivotal concepts in learning. Feedback involves providing learners with information about their responses whereas reinforcement affects the tendency to make a specific response again. Feedback can be positive, negative or neutral; reinforcement is either positive (increases the response) or negative (decreases the response). Feedback is almost always considered external while reinforcement can be external or instrinsic (i.e., generated by the individual).
Information processing theories tend to emphasize the importance of feedback to learning since knowledge of results is necessary to correct mistakes and develop new plans. On the other hand, behavioral theories such as Hull, Guthrie, Thorndike, and Skinner focus on the role of reinforcement in motivating the individual to behave in certain ways. One of the critical variables in both cases is the length of time between the response and the feedback or reinforcement. In general, the more immediate the feedback or reinforcement, the more learning is facilitated.
The nature of the feedback or reinforcement provided was the basis for many early instructional principles, especially in the context of programmed instruction (e.g., Deterline, 1962; Markle, 1964). For example, the use of "prompting" (i.e., providing hints) was recommended in order to "shape" (i.e., selectively reinforce) the correct responses. Other principles concerned the choice of an appropriate "step size" (i.e., how much information to present at once) and how often feedback or reinforcement should be provided.
References:
Deterline, W.A. (1962). An Introduction to Programmed Instruction. New York: Prentice-Hall.
Markle, S.R. (1964). Good Frames and Bad. New York: Wiley.

Schema

Bartlett (1932, 1958) is credited with first proposing the concept of schema (plural: schemata). He arrived at the concept from studies of memory he conducted in which subjects recalled details of stories that were not actually there. He suggested that memory takes the form of schema which provide a mental framework for understanding and remembering information.
Mandler (1984) and Rumelhart (1980) have further developed the schema concept. Schema have received significant empirical support from studies in psycholinguistics. For example, the experiments of Bransford & Franks (1971) involved showing people pictures and asking them questions about what the story depicted; people would remember different details depending upon the nature of the picture. Schema are also considered to be important components of cultural differences in cognition (e.g., Quinn & Holland, 1987). Research on novice versus expert performance (e.g., Chi et al., 1988) suggests that the nature of expertise is largely due to the possession of schemas that guide perception and problem-solving.
Schema-like constructs also form the basis of many theories of cognition including: Schank (scripts), AC  (productions), Soar (episodic memory), Piaget, and Rumelhart & Norman (modes) as well as some instructional theories such as Bruner, Reigeluth, Spiro and Sweller .
References:
Bartlett, F.C. (1932). Remembering: An Experimental and Social Study. Cambridge: Cambridge University Press.
Bartlett, F.C. (1958). Thinking. New York: Basic Books.
Bransford, J.D. & Franks, J.J. (1971). The abstraction of linguistic ideas. Cognitive Psychology, 2, 331-350.
Chi, M., Glaser, R. & Farr, M. (1988). The Nature of Expertise. Hillsdale, NJ: Erlbaum.
Mandler, J. (1984). Stories, Scripts, and Scenes: Aspects of Schema Theory. Hillsdale, NJ: Erlbaum.
Quinn, N. & Holland, D. (1987). Cultural Models of Language and Thought. New York: Cambridge University Press.
Rumelhart, D.E. (1980). Schemata: The building blocks of cognition. In R.J. Spiro, B.Bruce, & W.F. Brewer (eds.), Theoretical Issues in Reading and Comprehension. Hillsdale, NJ: Erlbaum

Sequencing of Instruction

One of the most important issues in the application of learning theory is sequencing of instruction. The order and organization of learning activities affects the way information is processed and retained (Glynn & DiVesta, 1977; Lorch & Lorch, 1985; Van Patten, Chao, & Reigeluth, 1986)
A number of theories (e.g., Bruner, Reigeluth, Scandura) suggest a simple-to-complex sequence. Landa's algo-heuristic theory prescribes a cumulative strategy. According to Gagne's Conditions of Learning theory, sequence is dictated by pre-requisite skills and the level of cognitive processing involved. Criterion Referenced Instruction (Mager) allows the learner the freedom to choose their own learning sequence based upon mastery of pre-requisite lessons. Component Display Theory (Merrill) also proposes that the learner select their own learning sequence based upon the instructional components available.
Theories that emphasize the goal-directed nature of behavior such as Tolman or Newell & Simon would specify that the sequence of instruction be based upon the goals/subgoals to be achieved. Gestalt theories, which emphasize understanding the structure of a subject domain, would prescribe learning activities that result in a broad rather than detailed knowledge for a particular domain.
On the other hand, behavioral (S-R) theories of learning such as connectionism, drive reduction or operant conditioning, would tend to support a linear sequence of instruction. From the behavioral perspective, learning amounts to S-R pairings and mastery of a complex subject matter or task involves the development of a chain or repetoire of such connections. Indeed, a fundamental principle of Skinnerian programmed learning was the "shaping" of such S-R chains.
Theories of adult learning such as adragogy orminimalism emphasize the importance of adapting instruction to the experience or interests of learners. According to these theories , there is no optimal sequence of instruction apart from the learner. A similar position based upon abilities would be espoused by theories of individual differences (e.g., Guilford, Cronbach & Snow, Sternberg) and supported by research on cognitive styles.
References:
Glynn, S.M. & DiVesta, F.J. (1977). Outline and hierarchical organization for study and retrieval. Journal of Educational Psychology, 69(1), 69-95.
Lorch, R.F. Jr., & Lorch, E.P. (1985). Topic structure representation and text recall. Journal of Educational Psychology, 77(2), 137-148.
Van Patten, J., Chao, C.I. & Reigeluth, C.M. (1986). A review of strategies for sequencing and synthesizing instruction. Review of Educational Research, 56(4), 437-471.

Taxonomies
Following the 1948 Convention of the American Psychological Association, Benjamin Bloom took a lead in formulating a classification of "the goals of the educational process". Bloom headed a group of educational psychologists who developed a classification of levels of intellectual behavior important in learning. This became a taxonomy including three overlapping domains; the cognitive, psychomotor, and affective (see Anderson & Krathwohl, 2001; Bloom & Krathwhol, 1956, Gronlund, 1970).

Cognitive learning consisted of 6 levels: knowledge, comprehension, application, analysis, synthesis, and evaluation. For each level, specific learning behaviors were defined as well as appropriate descriptive verbs that could be used for writing instructional objectives. For example:
1.Knowledge: arrange, define, duplicate, label, list, memorize, name, order, recognize, reproduce state.
2.Comprehension: classify, describe, discuss, explain, express, identify, indicate, locate, recognize, report, restate, review, select, translate,
3.Application: apply, choose, demonstrate, dramatize, employ, illustrate, interpret, operate, practice, schedule, sketch, solve, use, write.
4.Analysis: analyze, appraise, calculate, categorize, compare, contrast, criticize, differentiate, discriminate, distinguish, examine, experiment, question, test.
5.Synthesis: arrange, assemble, collect, compose, construct, create, design, develop, formulate, manage, organize, plan, prepare, propose, set up, write.
6.Evaluation: appraise, argue, assess, attach, choose compare, defend estimate, judge, predict, rate, core, select, support, value, evaluate.
The Affective domain (e.g., Krathwhol, Bloom & Masia, 1964) consisted of behaviors corresponding to: attitudes of awareness, interest, attention, concern, and responsibility, ability to listen and respond in interactions with others, and ability to demonstrate those attitudinal characteristics or values which are appropriate to the test situation and the field of study. This domain relates to emotions, attitudes, appreciations, and values, such as enjoying, conserving, respecting, and supporting.
Although not part of the original work by Bloom, others went on to complete the definition of psychomotor taxonomies. For example, Harrow (1972) proposed these six levels: Reflex (objectives not usually written at this "low" level), Fundamental movements - applicable mostly to young children (crawl, run, jump, reach, change direction), Perceptual abilities (catch, write, balance, distinguish, manipulate), Physical abilities (stop, increase, move quickly, change, react), Skilled movements (play, hit, swim, dive, use), and Non-discursive communication (express, create, mime, design, interpret).
The significance of the work of Bloom and others on taxonomies was that it was the first attempt to classify learning behaviors and provide concrete measures for identifying different levels of learning. The development of taxonomies is closed related to the use of instructional objectives and the systematic design of instructional programs (see Gagne, Merrill or Mager ).
References:
Anderson, L. & Krathwohl, D. (2001). A Taxonomy for Learning, Teaching and Assessing: A Revision of Bloom's Taxonomy of Educational Objectives. New York<: Longman.
Bloom Benjamin S. and David R. Krathwohl, (1956). Taxonomy of Educational Objectives: The Classification of Educational Goals, by a committee of college and university examiners. Handbook I: Cognitive Domain. New York<: Longman, Green.
Gronlund, Norman E. (1970). Stating Behavioral Objectives for Classroom Instruction. New York<: Macmillan.
Harrow, A. (1972). A Taxonomy of the Psychomotor Domain. A guide for Developing Behavioral Objectives. New York<: McKay.
Krathwohl, David R., Benjamin S. Bloom, and Bertram B. Masia. (1964). Taxonomy of Educational Objectives: The Classification of Educational Goals. Handbook II: Affective Domain. New York<: David McKay Co., Inc.
Note: Thanks to Kevin C. Lawrence for his suggestion to include this entry and the links he provided. Thanks to Geoff Issacs for the reference to the SOLO taxonomy.