Conversation Theory: A Constructivist, Dialogical Approach to Educational Technology
Bernard Scott
Learning Environments and Technology Unit
University of the Highlands and Islands Millennium Institute
Lews Castle College
Stornoway
Isle of Lewis
HS1 2SD
UK
Scott, B. (2001). “Conversation theory: a dialogic, constructivist approach to educational technology”, Cybernetics and Human Knowing, 8, 4, pp. 25-46.
Abstract
This paper overviews conversation theory, as developed over three decades by Pask, Scott and others, with particular emphasis on its application to the field of educational technology. Topics covered include models for learning and teaching, individual differences in approaches to learning, CASTE (Course Assembly System and Tutorial Environment) and associated principles for course design and tutorial strategies, knowledge and task analysis and knowledge representation for course design. The paper begins with a brief biographical note on the life and work of Gordon Pask and ends with some examples of current applications and some thoughts about the role of conversation theory in future developments in educational technology.
Keywords
Conversation theory, Gordon Pask, holist, serialist, comprehension learning, operation learning, entailment structure, task structure, entailment mesh
Introduction
Conversation theory, as developed over three decades by Pask, Scott and others, can serve as a conceptual framework that brings order and sense to the evolving and complex area of application known as educational technology. Conversation theory provides clearly formulated pedagogic principles that help show how to deploy technologies to meet particular educational objectives. It also provides a set of concepts and principles that address more general issues concerning human systems and their organisation which, going beyond the strictly pedagogic, can help educational technologists and others appreciate and address the administrative, political and cultural implications of the new technologies.
It is not possible in a short paper to convey all that conversation theory has to offer. Rather, the paper is selective and summary, focussing on three main themes with, hopefully, enough coverage to inspire the interested to consult more substantial primary sources. The three themes are:
1. conversation theory as a theory of learning and teaching, particularly suited for understanding how to support effective student-centred learning when using learning technologies for the on-line delivery of resource based learning;
2. conversation theory as a source of principles of course design and the design of learning environments;
3. conversation theory, with its subtheory of conversational domains, as a source for methodologies of knowledge and task analysis, particularly suited for eliciting and representing the structure of course content in an on-line hypertext environment.
The paper is organised as follows. First there is a brief biographical note about the life and work of the late Gordon Pask. There is then a summary of conversation theory interpreted as a theory of learning and teaching. There is also a brief mention of some variants of conversation theory to be found in the educational technology and teaching and learning literature. Next, there is a description of CASTE (Course Assembly System and Tutorial Environment), developed by Pask and Scott to be, in Pask’s phrase, an embodiment of conversation theory. CASTE includes a formally specified set of tutorial heuristics that exemplify the design of tutorial strategies for ensuring that learners achieve mastery of a particular knowledge domain. Reference is made to work carried out on individual differences in styles and strategies of learning. CASTE is useful as an exemplifier of a principled approach to course design and for the design of student-centred learning environments. Reference is also made to the Thoughtsticker system, a more sophisticated course design and knowledge elicitation and representation tool developed subsequent to CASTE, which exemplifies methodologies for knowledge and task analysis and knowledge representation. There is a brief overview of the conversation theory subtheory of conversational domains and narrative structures from which the methodologies were derived. Finally, there is a discussion of the relevance of conversation theory for educational technology today with examples of current work where conversation theory is a major influence.
What I have tried to do here and in related papers is to provide non-technical accounts of conversation theory. Pask’s own writing are notoriously difficult, chiefly because he uses specialised notation schemes and technical vocabularies, both of which evolved over time, but also because he had a polymath’s eye for detail and possibly contentious issues whose relevance might not be obvious to a single discipline specialist. He was aware of leading edge developments in logic and theories of computation, in cognitive psychology and neuroscience and in work on artificial intelligence and robotics. In addition, he saw conversation theory as a useful framework for adumbrating a wide range of other work on human communication and human systems and on the design of man-machine systems and interactive environments.
Some of the material gathered together here in summary form can be found in more elaborated versions elsewhere (Scott, 1999, 2000, 2001a, 2001b, in press a).
About Gordon Pask
Gordon Pask (1928-1996) was a UK scientist and inventor, one of the founding fathers of cybernetics, the study of control and communication in complex natural and artificial systems. See Pask (1961) and Scott (1980, 1982, 1993, 2001c) for more on Pask’s approach to cybernetics. In his regular column in this journal, Ranulph Glanville has also frequently written about Gordon Pask.
Pask’s many interests included work as an experimental psychologist studying styles and strategies of learning and as a pioneer educational technologist developing adaptive teaching machines. CASTE (Course Assembly System and Tutorial Environment) was developed in the early 1970s. CASTE built on many lessons learned from the design and development of adaptive systems for the teaching of perceptual-motor skills, applying them, with refinements, to the teaching of complex intellectual subject matter. Much of the pre-CASTE work is described in Pask (1975a).
Pask’s ideas about learning and teaching were articulated as conversation theory. In Pask’s phrase, CASTE was an embodiment of conversation theory. Conversation theory includes a subtheory of “conversational domains” and associated knowledge elicitation methodologies for carrying out knowledge and task analysis (Pask and Scott, 1973; Pask, Scott and Kallikourdis, 1973; Pask, Kallikourdis and Scott, 1975; Pask, 1975b, 1976).
Others have developed variants of conversation theory, with applications to learning and teaching, learning to learn and the design of educational systems, e.g., Harri-Augstein and Thomas (1991), Laurillard (1993), Boyd (1993).
Gordon Pask is rightly considered as one of the fathers of both first order and second order cybernetics. The latter refers to the novel epistemological paradigm where the observer, rather than be a purely external observer to the systems he studies, is invited to acknowledge that he, too, is a system, an observing system. Conversation theory, with its accounts of observers in conversation and interaction, was a major contribution to the emergence of this paradigm. That grand old man of cybernetics, Heinz von Foerster (whom Pask considered his mentor), calls Pask “the cybernetician’s cybernetician” and talks of him, in some awe, as a genius (von Foerster, 2001).
Overview of Conversation Theory
Although conversation theory can be elaborated as a general theory of human communication and social interaction, here, for the sake of brevity, it is interpreted as a theory of learning and teaching, in which one participant (the teacher) wishes to expound a body of knowledge to a second participant (the learner). Parts of the body of knowledge are referred to as “topics”; the term “concept” is reserved for the mental procedures that indicate understanding of a topic. Particular instantiations or models of topics are referred to as “relations”, which may be defined canonically with respect to a particular universe of discourse and associated modelling facility (see figure 1).
Underlying assumptions of conversation theory include the following. The brain/body system is a dynamic self-organising, “variety eating”, adaptive and habituating system, subject to boredom and fatigue. As Pask often put it, “Man is a system that needs to learn”, thus the problem of motivation is not “that we learn” it is rather what is learned and why. The basic mechanisms that support learning and adaptation are the various forms of conditioning that take place in associative networks (parallel distributed systems), with attentional systems subject to sensory-motor feedback (including proprioception and kinaesthesia) and algedonic (pain, pleasure) feedback. For humans, learning is also about the construction of symbolic representations, subject to constraints of logical coherence, acquired through the medium of dialogic, conversational interaction and the inner dialogic processes of strategic and tactical attention directing. In conversation, narrative forms are constructed and exchanged (Scott, 1999, Laurillard, Stratfold et al 1999, Bruner 1996). What is memorable is that which can be “taught back” (Pask and Scott, 1972).
Habitual forms of behaving and thinking evolve by proceduralisation. Proceduralisation may be guided and monitored by learning and teaching strategies. In conversation theory, remembering is understood as a dynamic process of reconstruction that is always contextualised and social (minimally, with no other person present, a psychological individual remembers with herself). If we do use the metaphor of memories being “stored” in our brains and bodies, we should recognise they are also “stored in our environments including the brains and bodies of other people”. This is a Pask aphorism that predates the more recently articulated concepts of “situated” and “distributed” cognition.
Pask argues that the distinctions required to characterise the cognition of an isolated psyche are the same as those made by the external observer of a conversation (cf. Ryle, 1971). In the “outer conversations” that constitute social institutions, the participants agree and disagree and negotiate shared descriptions, explanations and justifications. In her “inner conversation”, a person explains and justifies herself to herself.
In order to clearly conceptualise the idea that cognition is conversational in form, conversation theory distinguishes a type of organisation, the psychological (p-) individual, that is distinct from the biological and which is applicable both to persons and the social systems that they form. The participants in a conversation are p-individuals. The conversation is itself a p-individual.
A p-individual is formally defined as a stable, organisationally closed systemic whole, a “selfreplicating system of memories and concepts”, where:
1. a concept is a procedure that “recognises, reproduces or maintains a relation”, e.g., in context, riding a bicycle, performing a calculation;
2. a memory is a metacognitive procedure that “recognises, reproduces or maintains concepts”, for example, in context, justifying a method or providing a chain of explanation showing how the understanding of particular topic is derived from or entails the prior understanding of other topics;
3. a description of a concept is a “task structure” that says “what may be done”;
4. a description of a memory is an “entailment mesh” that says “what may be known” (see also below).
Pask refers to a biological individual in general terms as a mechanical (m-) individual. Note the power of the distinction between the two kinds of individual: m and p-individuals are not necessarily in one to one correspondence. One m may house several p; one p may be housed by several m’s.
Minimally in a conversation the participating p-individuals distinguish and learn about each other. Where there is a particular topic under discussion, participants construct models of each other’s models of that topic. The particular conceptions and misconceptions (kinds of understanding) that participants have of the topic in question can, following Aristotle and others, be broadly classified as “knowing why” and “knowing how”. In conversation participants may share both kinds of knowledge, participants exchange theories and present evidence in support of those theories, in this sense conversation theory is a theory of theory construction and elucidation.
The basic conversation theory model is shown in figure 1. Pask refers to this model as the “skeleton of a conversation”. It shows a “snapshot” view of two participants in conversation about a topic. The model distinguishes verbal, “provocative” interaction (questions and answers) from behavioural interaction via a shared modelling facility or “micro-world”.
The horizontal connections represent the verbal exchanges. Pask argues that all such exchanges have, as a minimum, two logical levels. In the figure these are shown as the two levels: “how” and “why”. The “how” level is concerned with descriptions of how to “do” a topic: how to recognise it, construct it, maintain it and so on; the “why” level is concerned with explaining or justifying what a topic means in terms of other topics. These exchanges are “provocative” in that they serve to provoke participants to construct understandings of each other’s conceptions and (possibly) misconceptions of topics and the relations between them. This is the essential aspect that makes conversation theory constructivist and dialogical in approach and clearly distinguishes it from other approaches that see teaching as the transmission of knowledge from teacher to learner.
The vertical connections represent causal connections with feedback, a hierarchy of processes that control or produce other processes. At the lowest level in the control hierarchy there is a canonical world, a “universe of discourse” or “modelling facility” where the teacher (or computer-based surrogate, as in CASTE, below) may instantiate or exemplify the topic by providing non-verbal demonstrations. Typically, such demonstrations are accompanied by expository narrative about “how” and “why”, the provocative interactions of questions and answers referred to above. Note that the form of what constitutes a canonical “world” for construction and demonstration may itself be a topic for negotiation and agreement.
In turn, the learner uses the modelling facility to solve problems and carry out tasks set. He or she may also provide narrative commentary about “how” and “why”. In a computer-based environment these may be elicited using computer aided assessment tools with a variety of different question styles.
Figure 1. The “skeleton of a conversation” (after Pask).
Pask refers to learning about “why” as comprehension learning and learning about “how” as operation learning and conceives them both as being complementary aspects of effective learning.
The distinction between “how” and “why” allows for a formal definition of what it means to understand a topic. In conversation theory, understanding a topic means that the learner can “teachback” the topic by providing both non-verbal demonstrations and verbal explanations of “how” and “why”.