Figure 1: Engagement-Control
Interactivity
Abstract
A range of interactive concepts can be identified which relate to different tasks performed during an instructional event or transaction. To provide a context for these concepts, a three dimensional model of interactivity (engagement, control, concept), describing the interaction between a user and the system is introduced. The importance of this model lies in the multi-layered approach to educational software, focusing on instructional design, graphic design and communication design to implement interactions created to motivate and engage the learner. To provide a framework for this approach, the current paper extends this Engagement-Control representation through a re-evaluation of existing learning theories and the ways in which their interpretations might lead to different approaches to interactivity in an educational courseware environment. To achieve this, the discussion documents selected learning theories and their differences, identifies potential prescriptions for interactivity resulting from critical theoretical perspectives and extends the existing principles of instructional design to incorporate new theoretical factors to provide a context for interactivity.
Interactivity in learning is "a necessary and fundamental mechanism forknowledge acquisition and the development of both cognitive and physical skills" (Barker, 1994:1). It is no longer adequate to see our field of practice being limited to products where interactivity is trivialised to simple menu selection, clickable objects or linear sequencing. Interaction is intrinsic to successful, effective instructional practice as well as individual discovery (Sims, in press). This approach is supported by Spector (1995:531) who contends that:
making automated learning environments highly interactive is a multi-disciplinary art ... however, the level of interactivity as measured on anyone's scale does not approach the level of interactivity in a human tutoring situation
The challenge is therefore to make best use of the technology, not to replicate human behaviour and communication, but to enhance human-computer communications through a better understanding of the use and implementation of interactive events.
In an analysis of various levels or modes of interactivity (Sims, in press), it is suggested that the notion of interactivity is not just a descriptor for any multi-media product, but a complex activity which is critical to the effectiveness of educational software. The individual concepts associated with an interactive environments therefore relate to the different tasks which might be performed during any instructional event or transaction.
Consequently, a three-dimensional model of interactivity (illustrated in Figure 1 below) is proposed, which takes into account instructional intent, user control and learner attributes.
Figure 1: Engagement-Control
InteractivityThe first dimension, engagement, refers to interactivity which is either navigational (where the user moves from one location in the application to another) or instructional (where the user is involved with the content in a way designed to facilitate learning). The second dimension, control, refers to the extent to which the system (program control) or user (learner control) is making the instructional or navigational decisions. The third dimension, interactive concept, provides an indication of the type of interaction which might be expected under the varying conditions defined by the model.
The development of interactive instructional resources is therefore dependent not only on the accurate structuring and representation of content, but also the access provided by the application to the user, which is manifested through the in-built controls and interactivity. However, the original premise for the Engagement-Control model was based on extensive practical experience with the development of instructional software products. The next phase is to take a different approach and the revisit some of the thoeries relevant to learning and instruction which in an attempt to validate the original model as well as identify additional or alternate factors which may be linked to the effective application of interactivity to educational software.
The development of educational resources, especially those designed for computer-based delivery, have often been the result of projects which have conformed to one or more learning theories as well as adopting specific instructional design strategies to ensure their integrity. It is the contention of the author that there is no one true path by which effective educational software can be developed, but rather a vast and complex network from which developers must make selections based on the particular environment in which they are working. Of particular interest is the notion of interactivity, and the extent to which different theoretical perspectives can be applied to develop a more advanced notion of interactivity.
The following analysis examines, chronologically, a selection of learning theories which the writer believes are significant to the debate of interactivity. The purpose of this analysis is to provide a context in which the notion of interactive environments for teaching and learning can be better understood.
Early this century, the connectionist approach of Thorndike (1913) proposed that learning requires both practice and rewards with consequent transfer of learning resulting from previously encountered situations. One of the interesting aspects was that the ideas were focused on animal as well human (mathematics and spelling) learning. As such, it is to be noted that this, and many other of the theories of learning, are typically domain or learner specific rather than generic.
An well-known exemplar of this is the work of Piaget (1929), who concluded that children provide varying explanations of reality at different stages of cognitive development. Consequently, Piaget (1929) determined that learning materials and activities should involve the appropriate level of motor or mental operations for a child of given age and that teaching methods should actively involve students and present challenges.
While Thorndike's (1913) propositions covered all forms of learning, including animal, child and adult, Piaget (1929) was more focused on the developing person (from early childhood). As such, any application of a particular theory or approach must be considered in terms of the target audience to which it has been applied as well as its socio-environmental impact.
The Sign Learning theory of Tolman (1932) predicted that learning is purposive and goal-directed, often involves the use of environmental factors to achieve a goal and that organisms will select the shortest or easiest path to achieve a goal. While Tolman's (1932) work was the result of work with rats in mazes, the idea that learner's will develop a set of strategies to complete a task is important, as any interactive support (or scaffolding) will be potentially used by the learner to achieve the particular goal. In fact, It is here we see the links between some of the current emphases placed on learning such as situational (Sternberg, 1977; Lave, 1988) and minimalism (Carroll, 1990).
One of the major influences on education and training has been the behaviourist notion, and the original work of Skinner (1953) focused on operant conditioning (behaviourism), where it was demonstrated through extensive laboratory experiments with animals that positively reinforced behaviour will re-occur, with intermittent reinforcement particularly effective. Consequently Skinner (1953) concluded that with learning environments, information should be presented in small amounts to reinforce or shape an individual's response.
Another important distinction to make in terms of interactivity is whether it is based on an instructor's prescription or the individual pursuits of the learner. The very notion of reinforcement assumes that there are correct responses for which
The impact of Skinner's (1953) work have been significant for educational software developers as the early strategies implemented through authoring tools and systems were based on a tutorial model, which mirrored the earlier paper-based programmed instruction modules, which in turn were designed according to the behaviourist principles espoused through operant conditioning.
In contrast to a behaviourist approach, Wertheimer (1959) proposed a Gestalt Theory, whereby it was suggested that learners be encouraged to discover the essential framework of content material topic or problem (i.e. the relationship among the elements). In addition, it was suggested that gaps, incongruities, or disturbances are important stimulii for learning. These notions pre-date the later constructivist movement (Bruner, 1966), but provide a contrast between a learner centred or teacher centred philosophy, which is discussed in the following section.
However, in developing the role of the instructor (which is consistent with the emergence of Instructional Systems Design methodologies) Gagne (1962) proposed the existence of conditions of learning, with the premise that different instruction is required for different learning outcomes. This concept was later extended by Merrill (1980), and provided a substantial argument for an instructivist approach to teaching and learning.
The work of Gagne (1962) had a major effect on the design of instructional materials, and continues to be used to define guidelines for the development of instructional software. However, it is important to note that much of the early work on learning and instruction was based on the needs of the U.S. military - quite a different environment from a university or school.
In addition, the philosophy and prescriptions defined by the conditions of learning are very much based on the notion of an Instructor-Student relationship - where the knowledgeable instructor is providing the naive student with appropriate knowledge to achieve a particular instructional outcome. While perhaps appropriate for a structured training classroom, its frequent adoption by courseware developers cannot necessarily be justified, especially when the design relationship may focus on the student acquiring knowledge through exploration and discovery rather than through
In terms of interaction, the work of Gagne provides only for the student to respond to questions posed by the instructor. There is little room in this schema for students (or learners) to use interactive options as a cognitive extension to other learning activities and resources.
As if to counter the instructivist philosophy, Bruner (1966) proposed the notion of constructivism, with more emphasis on the learner although focusing on instruction being related to experiences and contexts that encourage student engagement, involvement and motivation. The more specific focus on the student implied that any instructional event should be structured to enable easy comprehension by the student, as well as encouraging discovery, extrapolation and generalisation.
Additional credence to the learner-centred approach was also provided by De Bono (1967), with the lateral thinking approach which simply proposed that solutions to problems might best be achieved by obtaining new perspectives on a problem, by decomposing and restructuring the elements. However, it should be noted that the environment for De Bono's (1967) ideas were the meeting room in a business operation - and therefore potentially specific to a particular target audience and context.
The contextualisation of learning was reinforced by the experiential learning work by Rogers (1969). In this instance, it was suggested that significant learning takes place when the subject matter is relevant to the personal interests of the student, with the extension that self-initiated learning is the most lasting and pervasive. This particular philosophy introduces the notion of the learner having control over the learning environment and, with educational software applications, identifying the need to create interactive structures by which the user (or learner) has control on both access to subject matter resources as well as navigation between content areas.
One of the more influential approaches was the Dual Coding Theory of Paivio (1971). In this instance, different types of processing were identified (representational, the direct activation of verbal or non-verbal representations, referential, the activation of the verbal system by the nonverbal system or vice-versa, and associative processing, the activation of representations within the same verbal or nonverbal system). More importantly, Paivio (1971) concluded that recall and/or recognition is enhanced by presenting information in both visual and verbal form - which provides support for the premise that multimedia, per se, is a more effective presentational mode. From another perspective, the Dual Coding Theory also highlights that different stimulii are used by learners and that their inclusion in any learning materials must be subject to careful analysis.
However, to confound the potential for technology to enable learning, there are also the social learning theorists (e.g. Bandura, 1971; Vygotsky, 1978) who suggest that he highest level of observational learning is achieved by first organizing and rehearsing the modeled behavior symbolically and then enacting it overtly. Coding modeled behavior into words, labels or images results in better retention than simply observing. In addition, cognitive development is limited to a certain range at any given age and full cognitive development requires social interaction. This latter point is significant in that it develops the notion that learning in a solitary environment (as many computer applications) is not positive.
A different perspective, but one which supported the notion that ability of learners to access relevant information was introduced by Craik & Lockert (1971). Using the notion o f levels of processing they suggested that increased processing of information during learning will contribute to retention and memory. And depending on the particular target audience, different elements could be applied to processing at various levels. For example, Knowles (1975) focused on adult learning (androgogy) and suggested that this particular group needed to be involved in the planning and evaluation of their instruction. In addition, experience (including mistakes) would provide the basis for learning activities and that adults are most interested in learning subjects that have immediate relevance to their job or personal life, with the actual learning process being problem-centered rather than content-oriented.
The emphasis on adult learning was also adopted by Cross (1976), who emphasised that adult learning programs should capitalize on the experience of participants, that adult learning programs should adapt to the aging limitations of the participants and that adults should be challenged to move to increasingly advanced stages of personal development. Finally, Cross (1976) suggested that adults should have as much choice as possible in the availability and organization of learning programs. From this particular approach, certain elements of interactivity such as learner control and choice are clearly identified. However, this must also be taken in terms of the possibility that adults are restricted through aging - and therefore potentially averse to using the full interactive capabilities of the technology. It would seem that many of the theories have been applied in some small way to the development of interactive applications for education and training - but which, if any, is the right one?
As the specifics of learning and instruction came to the fore, Sticht (1975) defined a Functional Context approach in which instruction should be made as meaningful as possible to the learner in terms of their prior knowledge, and that material and equipment should reflect that which would be used after the training. Sticht (1975) also emphasised the need for valid assessment through context/content specific measurement.
Another focus on how people may best learn was introduced by Cronbach & Snow (1977) in terms of aptitude treatment - aptitudes and instructional treatments interact in complex patterns and are influenced by task and situation variables. Their conclusion that highly structured instructional environments tend to be most successful with students of lower ability; conversely, low structure environments may result in better learning for high ability students has had a direct influence on developing the extensive research on learner control.
The seeming intertwining of approaches was captured by Sternberg (1977), with the Triarchic Theory which determined that training of intellectual performance must be socio-culturally relevant to the individual, and that training programs should provide links between the training and real-world behaviour. In addition, training programs should be designed with explicit instruction in strategies for coping with novel tasks/situations and that training programs should provide explicit instruction in both executive and non-executive information processing and interactions between the two. Finally, training programs should actively encourage individuals to manifest their differences in strategies and styles.
A more specific design was introduced by Salomon (1979), with the notion of Symbol Systems, whereby the symbolic coding elements of particular media require different mental transformations and hence affect the mastery of specific skills and that the level of knowledge and skill that an individual possesses will affect the impact of specific media sequences. More importantly, the nature of the learning/information processing tasks can affect the impact of specific media sequences, the social context of media presentations can influence what message is perceived and that there is a reciprocal relationship between media and learner; each can influence the other.
In terms of the educational technology field, the work of Merrill (1983) in developing the Component Display Theory has had a major impact. The essential factors of the theory predict that instruction will be more effective if all three primary performance forms (remember, use, generality) are present. While Merrill (1983) suggests that these primary forms can be presented by either explanatory or inquisitory strategies and that student control is an essential feature, it is clear from later writings (Merrill, 1996) that the ideas are grounded in instructivist philosophy.
As technology began to become more capable of engaging presentations, so the work of educational researchers began to extend the field. Spiro, Feltovitch & Coulson (1988) defined a Cognitive Flexibility Theory which suggested that learning activities must provide multiple representations of content, and that instructional materials should avoid oversimplifying the content domain and support context-dependent knowledge. Taking the earlier work on contextual learning, Spiro et al (1988) also suggested that instruction should be case-based and emphasise knowledge construction, rather than the transmission of information, with knowledge sources being interconnected rather than compartmentalized.
To support this the notion of Situated Learning Lave (1988) asserted that knowledge needs to presented in an authentic context, i.e., settings and applications that would normally involve that knowledge. In addition, it was also determined that learning requires social interaction and collaboration. More recently, an alternative form of theory on learning was proposed by Carroll (1990). The minimalist approach prescribed that learners should be allowed to start immediately on meaningful tasks, that instruction should minimise the amount of reading and other passive forms of training by allowing users to fill in the gaps themselves. While specific to the instruction of software applications, the minimalist approach also suggested the inclusion error recognition and recovery activities in the instruction and proposed that all learning activities be self-contained and independent of sequence.
One of the major difficulties with presenting such a wide range of theoretical perspectives is that the basic principles attributed to each theory has general merits. Perhaps the problem is that various developers have taken one particular theory to be generic and promoted it to the wider community of practising professionals. While academics might easily differentiate the various theoretical positions, it is not as easy for the lay person. Therefore the challenge from this brief analysis is to suggest that it is essential for any developer to consider the contextual and situational characteristics of the target audience in determining which instructional or learning strategy to employ in educational software.
The majority of the theories presented focus on the ways in which learning can be enhanced through instruction or teaching - with the underlying assumption that it is the teacher's structuring and delivery of content which is significant.
However, there are selected propositions which deal more with the tasks which the learner must undertake (Craik & Lockert, 1972; Salomon, 1979; Carroll, 1990). It is apparent then that there are too many learning theories and instructional design methodologies. For the uninitiated, it is recommended that the focus of any educational development be founded on the specific target audience being addressed (e.g. adult, child, experienced, novice) and the particular content-material being considered (e.g. technical, interpersonal, computer-based, socially relevant, attitudinal) rather than attempting to fit those factors into an existing methodology. However, once those factors have been determined, then a methodology or approach which fits that environment (e.g. Component Display, Lateral Thinking, Situational) can then be applied to ensure a level of educational integrity.
In essence, this brief analysis of theoretical approaches only confounds the problems which face educational developers - that there is still much to be understood about the ways in which people learn, the ways in which educational materials can be constructed and the way in which the delivery of those materials is interpreted by the target audience.
This paper has examined a range of options for interactivity, based on the desire to continue to develop our understanding of implementing effective instructional technology applications. The author contends that appropriate levels of theoretically based interactivity must be based on an eclectic theoretical position rather than one level of the truth. By focusing on instructional design, graphic design and communication design to implement interactions which will motivate and engage the learner, the on-going success of functional and effective interactive instructional applications is assured.
Bandura, A. (1971). Social Learning Theory. New York: General Learning Press.
Barker, P. (1994). Designing Interactive Learning, in T. de Jong & L. Sarti (Eds), Design and Production of Multimedia and Simulation-based Learning Material. Dordrecht: Kluwer Academic Publishers.
Bruner, J. (1966). Toward a Theory of Instruction. Cambridge, MA: Harvard University Press.
Carroll, J.M. (1990). The Nurnberg Funnel . Cambridge, MA: MIT Press.
Craik, F. & Lockhart, R. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning & Verbal Behavior, 11, 671-684.
Cronbach, L. & Snow, R. (1977). Aptitudes and Instuctional Methods: A Handbook for Research on Interactions. New York: Irvington.
Cross, K.P. (1976). Accent on Learning. San Francisco: Jossey-Bass.
DeBono, E. (1967). New Think: The Use of Lateral Thinking in the Generation of New Ideas. New York: Basic Books.
Gagne, R. (1962). Military training and principles of learning. American Psychologist, 17, 263-276.
Knowles, M. (1975). Self-Directed Learning. Chicago: Follet.
Lave, J. (1988). Cognition in Practice: Mind, mathematics, and culture in everyday life. Cambridge, UK: Cambridge University Press.
Merrill, M.D. (1983). Component Display Theory. In C. Reigeluth (ed.), Instructional DesignTheories and Models. Hillsdale, NJ: Erlbaum Associates.
Merrill, M.D. (1986). Reclaiming Instructional Design. Paper presented at the 1996 AECt Convention. Indianapolis.
Paivio, A. (1971). Imagery and Verbal Processes. New York: Holt, Rinehart & Winston.
Piaget, J. (1929). The Child's Conception of the World. NY: Harcourt, Brace Jovanovich.
Rogers, C.R. (1969). Freedom to Learn. Columbus, OH: Merrill.
Salomon, G. (1979). Interaction of Media, Cognition, and Learning. San Francisco: Jossey-Bass.
Schwier, R.A. & Misanchuk, E. (1993). Interactive Multimedia Instruction. Englewood CLiffs, NJ: Educational Technology Publications.
Sims (in press). Interactivity: A Forgotten Art? Computers and Human Behaviour.
Skinner, B.F. (1953). Science and Human Behavior. New York: Macmillan.
Spector, M.J. (1995). Integrating and Humanizing the Process of Automating Instructional Design, in R.D. Tennyson & A.E. Barron (Eds), Automating Instructional design: Computer Based Development and Delivery Tools. Berlin: Springer-Verlag.
Spiro, R.J., Coulson, R.L., Feltovich, P.J., & Anderson, D. (1988). Cognitive flexibility theory: Advanced knowledge acquisition in ill-structured domains. In V. Patel (ed.), Proceedings of the10th Annual Conference of the Cognitive Science Society. Hillsdale, NJ: Erlbaum.
Sternberg, R.J. (1977). Intelligence, Information Processing, and Analogical Reasoning. Hillsdale, NJ: Erlbaum.
Sticht, T.G. (1976). Comprehending reading at work. In M. Just & P. Carpenter (eds.), Cognitive Processes in Comprehension. Hillsdale, NJ: Erlbaum.
Thorndike, E. (1913). Educational Psychology: The Psychology of Learning. New York: Teachers College Press.
Tolman, E.C. (1932). Purposive Behavior in Animals and Men. New York:Appleton-Century-Crofts.
Vygotsky, L.S. (1978). Mind in Society. Cambridge, MA: Harvard University Press.
Waterworth, J.A. (1992). Multimedia Interaction with Computers: Human Factors Issues. Chichester, Sussex: Ellis Horwood Ltd.
Wertheimer, M. (1959). Productive Thinking.
New York: Harper & Row.
The author assigns to ASCILITE and educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The author also grants a non-exclusive licence to ASCILITE to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the ASCILITE 96 conference papers, and for the document to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the author.