Behaviourism

Behaviourism

Introduction

Behaviourism, which was the predominant instructional paradigm for the first half of this century, is firmly rooted in the positivist, objectivist tradition. These views stem from 'hard' science and maintain that "knowledge is hard, objective and tangible" (Cohen and Manion 1989, 7). Jonassen (1990, 32) relates this to instructional design by saying that:

IST has its intellectual roots in an objectivist epistemology that holds that knowledge is based on some reality that is external to the learner and that all learners can come to know that reality in pretty much the same way.

This behaviourist approach holds that learning is evidenced by a changing of behaviour "so that if we control the behaviour of the learner, we also control his or her knowledge acquisition" and that "IST buys into this epistemology by insisting that knowledge results from the instructional interventions conceived by designers ... that mediate learning" (Jonassen 1990, 32). While Jonassen admits that this is in fact true, he questions what learning opportunities are lost "in the pedagogical constraints imposed by our interventions".

Behavioural learning theory "is based on the premise that learning results from the pairing of responses with stimuli" (Hannafin and Peck 1988, 46). The advantage of this approach is that it does not complicate the learning situation by taking into account the complexity of the brain. It also means that the process of teaching or instructional design is all important as it is by this that the learning process succeeds or fails. This makes the teacher the font of all knowledge in the learning situation and this knowledge is dispensed in a didactic fashion.

It is argued, however, by Jonassen (1991, 29) that a behavioural approach can be warranted in some circumstances, usually in the case of a beginning learners in a specific curriculum area with other more appropriate strategies being employed as the learners acquire more knowledge. He also differentiates between the instructional requirements for learning and training. He states that:

Most training is almost, by definition, convergent and objectivistic, because it supports explicit, company-generated performance goals (1991, 30).

While Tobias (1991, 41) claims that "... behaviourism has run its course and the cognitive paradigm is in command of most psychological provinces", the situation is not as clear cut as this would seem to suggest. In 1981, Burton, despite stating that "behaviourism is no longer a 'popular' orientation in educational or instructional design", maintained that "... behavioural technology will remain prevalent in both the classroom and instructional design shops" (1981, 25). A text on instructional design published in 1990 by Seels and Glasgow which, according to Jonassen (1990, 32) "while well written and constructed, remain[s] tied to the objectivist conception of controlled learning". Merrill et al. confirm this situation when they state that "In a recent meeting a colleague suggested that all the instructional design theory in current use was at least 20 years old and firmly rooted in behavioural psychology" (1990, 7).

Jonassen (1992, 32) argues that while "Behaviourism has been largely replaced by cognitive psychology in descriptions of how learning occurs" much of instructional design practice remains behaviourist in nature. He states that behaviouristic laws are not valid, in fact, although "behaviourism works ... to a degree" most people agree that it cannot "explain the complexities of human learning, especially knowledge-based learning".

As other areas of education have moved away from a behaviourist approach, it could be asked why instructional designers have been reluctant to adopt different paradigms. One reason would appear to be that 'old habits die hard'. An example of this is a system described by DeJoy and Mills (1989, 39) that has been developed for evaluating interactive software to be used with adult learners in the 'Personal Adult Learning Lab' at the University of Georgia's Centre for Continuing Education. Information from the clients of this laboratory have been used to determine a set of criteria for evaluation. DeJoy and Mills state that while these criteria have been presented for the purpose of evaluation, if they are considered to be factors that are found in good software, they should be considered by the instructional designer. They claim that:

According to our findings, the instructional strategies appear to impact the actual process of learning for each adult interacting with the material, as Burham and Seamons (1987) have suggested. Current thinking describes the process of learning as unique to each individual, influenced, in part, by experience, intelligence, motivation, and age (Evans 1987) as well as by individual ways of perceiving and processing information (Dixon, 1985; Korhonen and McCall, 1986). Self-instructional educational technologies must take these individual differences into account ... [and] learners continue to impose their own learning "styles" on the material and required opportunities to back-up, adjust presentation speed and sequence, and skip around the material as their interests dictated. When the instructional technology did not permit these learning strategies, adult learners became frustrated with the interaction (DeJoy and Mills 1989, 40).

Good non-behavioural stuff. However, when describing the behavioural characteristics required, their first statement is that "Behavioural objectives are clearly stated". How can such behavioural objectives be expected to be achieved when the users are given a high degree of freedom in the way in which they use the information? They also describe how user records are to be maintained. One could question the validity of collecting records of people who attend the laboratory on a drop in basis, but, as has been said, old habits die hard.

It could also be suggested that behaviourism is more comfortable than other paradigms. Behaviourism with its positivist outlook promises that there are some truths in instructional design. It also suggest that there is a real world that students can learn about. The content expert has 'knowledge' that he or she can pass on to the learner. This is a much more comfortable situation than one in which the 'teacher' is the facilitator of learning and the students construct their own knowledge. Using the behaviourist paradigm, as it is believed that there are truths in instructional design, prescriptions can be written. For example, Gagné, Wager and Rojas (1981, 22) state that there are nine events of instruction that should be attended to by authors of computer assisted learning materials. These are, Gaining attention; Informing learner of lesson objective; Stimulating recall of prior learning; Presenting stimuli with distinctive features; Guiding learning; Eliciting performance; Providing informative feedback; Assessing performance; Enhancing retention and learner transfer. There is no room for deviation from this prescription as Gagné et al. warn that "The major point made by these guidelines is that designers of CAI instruction need to attend to all nine of these events, rather than simply two or three" (1981, 21).

Using Gagné's guidelines, instructional design for CAI is easy, simply mould the content to adhere to the prescription, and as Merrill et al. (1990, 7) state, "The most widely applied instructional design theory is based largely on the work of Robert M. Gagné and his associates at Florida State University". Merrill refers to this instructional design theory as ID1 and he lists a number of limitations to it. These include:

ID1 content analysis does not integrate wholes which are essential for understanding complex and dynamic phenomena.

ID1 has limited prescriptions for knowledge acquisition.

D1 has limited prescriptions for course organisation.

ID1 theories are essentially closed systems.

ID1 fails to integrate the phases of instructional development.

ID1 teaches pieces but not integrated wholes.

ID1 instruction is often passive rather than interactive.

ID1 presentations must be constructed from small components (Merrill et al. 1990, 7-10).

Unfortunately, Merrill's alternative, which he calls ID2, appears to be an attempt to convert knowledge into a series of algorithms and the instructional design advice prescribed by a computer program (see Jones, Li and Merrill, 1990; Merrill, Li and Jones, 1991; Merrill, Jones and Li, 1992). Merrill is not alone in his desire to produce a computer program which acts as an instructional design adviser, but, as this concept is somewhat foreign (to put it nicely) to me, it will not be considered further.

Conclusions

Behaviourism has only been dealt with briefly here as it is not considered to be a paradigm that will enable the construction of 'mental models' or is suitable for use in a hypermedia environment. A behaviourist approach is prescriptive in that it dictates what knowledge the students will learn, in what order they will learn it and how they will learn it. In addition, the use of behavioural objectives ensures that learners concentrate on key points rather that considering the information as a whole. As links are not normally established between these objectives, knowledge gained would be fragmented which will also prevent the formation of an adequate mental model.

Server: http://scs.une.edu.au
Author: Tony Brown

Created: 25.6.97

© The University of New England, NSW, Australia

	Behaviourism

	Introduction

	Behaviourism, which was the predominant instructional paradigm for the first half of this century, is firmly rooted in the positivist, objectivist tradition. These views stem from 'hard' science and maintain that "knowledge is hard, objective and tangible" (Cohen and Manion 1989, 7). Jonassen (1990, 32) relates this to instructional design by saying that: IST has its intellectual roots in an objectivist epistemology that holds that knowledge is based on some reality that is external to the learner and that all learners can come to know that reality in pretty much the same way. This behaviourist approach holds that learning is evidenced by a changing of behaviour "so that if we control the behaviour of the learner, we also control his or her knowledge acquisition" and that "IST buys into this epistemology by insisting that knowledge results from the instructional interventions conceived by designers ... that mediate learning" (Jonassen 1990, 32). While Jonassen admits that this is in fact true, he questions what learning opportunities are lost "in the pedagogical constraints imposed by our interventions". Behavioural learning theory "is based on the premise that learning results from the pairing of responses with stimuli" (Hannafin and Peck 1988, 46). The advantage of this approach is that it does not complicate the learning situation by taking into account the complexity of the brain. It also means that the process of teaching or instructional design is all important as it is by this that the learning process succeeds or fails. This makes the teacher the font of all knowledge in the learning situation and this knowledge is dispensed in a didactic fashion. It is argued, however, by Jonassen (1991, 29) that a behavioural approach can be warranted in some circumstances, usually in the case of a beginning learners in a specific curriculum area with other more appropriate strategies being employed as the learners acquire more knowledge. He also differentiates between the instructional requirements for learning and training. He states that: Most training is almost, by definition, convergent and objectivistic, because it supports explicit, company-generated performance goals (1991, 30). While Tobias (1991, 41) claims that "... behaviourism has run its course and the cognitive paradigm is in command of most psychological provinces", the situation is not as clear cut as this would seem to suggest. In 1981, Burton, despite stating that "behaviourism is no longer a 'popular' orientation in educational or instructional design", maintained that "... behavioural technology will remain prevalent in both the classroom and instructional design shops" (1981, 25). A text on instructional design published in 1990 by Seels and Glasgow which, according to Jonassen (1990, 32) "while well written and constructed, remain[s] tied to the objectivist conception of controlled learning". Merrill et al. confirm this situation when they state that "In a recent meeting a colleague suggested that all the instructional design theory in current use was at least 20 years old and firmly rooted in behavioural psychology" (1990, 7). Jonassen (1992, 32) argues that while "Behaviourism has been largely replaced by cognitive psychology in descriptions of how learning occurs" much of instructional design practice remains behaviourist in nature. He states that behaviouristic laws are not valid, in fact, although "behaviourism works ... to a degree" most people agree that it cannot "explain the complexities of human learning, especially knowledge-based learning". As other areas of education have moved away from a behaviourist approach, it could be asked why instructional designers have been reluctant to adopt different paradigms. One reason would appear to be that 'old habits die hard'. An example of this is a system described by DeJoy and Mills (1989, 39) that has been developed for evaluating interactive software to be used with adult learners in the 'Personal Adult Learning Lab' at the University of Georgia's Centre for Continuing Education. Information from the clients of this laboratory have been used to determine a set of criteria for evaluation. DeJoy and Mills state that while these criteria have been presented for the purpose of evaluation, if they are considered to be factors that are found in good software, they should be considered by the instructional designer. They claim that: According to our findings, the instructional strategies appear to impact the actual process of learning for each adult interacting with the material, as Burham and Seamons (1987) have suggested. Current thinking describes the process of learning as unique to each individual, influenced, in part, by experience, intelligence, motivation, and age (Evans 1987) as well as by individual ways of perceiving and processing information (Dixon, 1985; Korhonen and McCall, 1986). Self-instructional educational technologies must take these individual differences into account ... [and] learners continue to impose their own learning "styles" on the material and required opportunities to back-up, adjust presentation speed and sequence, and skip around the material as their interests dictated. When the instructional technology did not permit these learning strategies, adult learners became frustrated with the interaction (DeJoy and Mills 1989, 40). Good non-behavioural stuff. However, when describing the behavioural characteristics required, their first statement is that "Behavioural objectives are clearly stated". How can such behavioural objectives be expected to be achieved when the users are given a high degree of freedom in the way in which they use the information? They also describe how user records are to be maintained. One could question the validity of collecting records of people who attend the laboratory on a drop in basis, but, as has been said, old habits die hard. It could also be suggested that behaviourism is more comfortable than other paradigms. Behaviourism with its positivist outlook promises that there are some truths in instructional design. It also suggest that there is a real world that students can learn about. The content expert has 'knowledge' that he or she can pass on to the learner. This is a much more comfortable situation than one in which the 'teacher' is the facilitator of learning and the students construct their own knowledge. Using the behaviourist paradigm, as it is believed that there are truths in instructional design, prescriptions can be written. For example, Gagné, Wager and Rojas (1981, 22) state that there are nine events of instruction that should be attended to by authors of computer assisted learning materials. These are, Gaining attention; Informing learner of lesson objective; Stimulating recall of prior learning; Presenting stimuli with distinctive features; Guiding learning; Eliciting performance; Providing informative feedback; Assessing performance; Enhancing retention and learner transfer. There is no room for deviation from this prescription as Gagné et al. warn that "The major point made by these guidelines is that designers of CAI instruction need to attend to all nine of these events, rather than simply two or three" (1981, 21). Using Gagné's guidelines, instructional design for CAI is easy, simply mould the content to adhere to the prescription, and as Merrill et al. (1990, 7) state, "The most widely applied instructional design theory is based largely on the work of Robert M. Gagné and his associates at Florida State University". Merrill refers to this instructional design theory as ID1 and he lists a number of limitations to it. These include: ID1 content analysis does not integrate wholes which are essential for understanding complex and dynamic phenomena. ID1 has limited prescriptions for knowledge acquisition. D1 has limited prescriptions for course organisation. ID1 theories are essentially closed systems. ID1 fails to integrate the phases of instructional development. ID1 teaches pieces but not integrated wholes. ID1 instruction is often passive rather than interactive. ID1 presentations must be constructed from small components (Merrill et al. 1990, 7-10). Unfortunately, Merrill's alternative, which he calls ID2, appears to be an attempt to convert knowledge into a series of algorithms and the instructional design advice prescribed by a computer program (see Jones, Li and Merrill, 1990; Merrill, Li and Jones, 1991; Merrill, Jones and Li, 1992). Merrill is not alone in his desire to produce a computer program which acts as an instructional design adviser, but, as this concept is somewhat foreign (to put it nicely) to me, it will not be considered further.

	Conclusions

	Behaviourism has only been dealt with briefly here as it is not considered to be a paradigm that will enable the construction of 'mental models' or is suitable for use in a hypermedia environment. A behaviourist approach is prescriptive in that it dictates what knowledge the students will learn, in what order they will learn it and how they will learn it. In addition, the use of behavioural objectives ensures that learners concentrate on key points rather that considering the information as a whole. As links are not normally established between these objectives, knowledge gained would be fragmented which will also prevent the formation of an adequate mental model.


	Server: http://scs.une.edu.au Author: Tony Brown Created: 25.6.97 © The University of New England, NSW, Australia