Full Paper

Barry Harper

Barry_Harper@uow.edu.au

Graduate School of Education, Faculty of Education

John Hedberg

John_Hedberg@uow.edu.au

Interactive Multimedia Learning Laboratory http://www.immll.uow.edu.au/

University of Wollongong,

In today's society greater demands are being placed on education systems at all levels to produce citizens who can use knowledge in new domains and different situations. Members of society at every level are being asked to demonstrate advanced levels of problem-solving skills to retain their level of employment. Learning to think critically, to analyse and synthesis information to solve problems in a variety of contexts and to work effectively in teams are crucial skills for modern employees, and yet there is little evidence that our education systems are developing these skills in our children (Bransford, Goldman and Vye, 1991).

Schank and Cleary (1995) have argued strongly that our educational systems are not developing these essential skills, the Cognition and Technology Group at Vanderbilt(CTGV) (CTGV, 1993) have outlined what they claim are the flaws in the conventional approaches to schooling and teaching, based on Whitehead's (1929) notion of "inert" as opposed to "active" knowledge, and Berryman (1991) has described the erroneous assumptions that much of the teaching and schooling process is based on. Calls for the restructuring of the schooling process have also come from institutions such as the American Association for the Advancement of Science (1989), the National Council of Teacher's of Mathematics (1989) and have been called for in Australian reports such as the Discipline Review of Mathematics and Science (DEET, 1989).

However, not all calls are for development of problem-solving skills and higher order thinking. The recent political storm in Australia over basic literacy skills has again raised the difficult relationship that continues to exist between educational reformists and political agendas where children, teachers and schools are again defending themselves against questionable comparisons to past eras and political fads.

Recent curriculum documents in many western countries, and in particular in Australia, emphasise the skills of investigation, reflection and analysis to generate or refine knowledge. The appeal of cognitive process training to support this development is obvious, and it seems far more efficient to provide the student with general-purpose problem solving than instruction on specific solutions to specific problems. Jonassen and Tessmer (Jonassen et al, 1996-7) have also questioned the commonly used taxonomies of learning that are the basis of our curriculum documents, proposing that engaging in a greater range of learning outcomes than isolated intellectual skills is essential for meaningful learning.

There are, however, significant efforts being made to develop and implement alternative frameworks for learning and much of this current effort at renewal is based on a class of theories referred to as constructivism. Fundamentally, constructivism asserts that we learn through a continual process of constructing, interpreting, and modifying our own representations of reality based on our experiences with reality (Jonassen, 1994)

In parallel with this push for renewal, the integration of technologies which allow the representation of ideas in many different media forms through information and telecommunication technology, and specifically computers, into the educative process offers instructors unique opportunities to individualise instruction, place learners in open ended student-centred investigations, and for instructors to shift from their traditional instructor role to the role of mentor and co-learner.

However, the technology is not universally seen as a panacea for our current educational systems woes. As long ago as 1984 Cuban warned "there should be a page in the Guinness Book of Records on failed classroom reforms, for few ever seem to have been incorporated into teachers' repertoires". Like past revolutions in education, it will go the way of previous technologies unless there are changes to the educative framework. Alfred Bork (1995) has argued in his critical review of the failure of computers in schools and universities that the effective use of new instructional paradigms requires:

1. a shift in teachers' pedagogical approaches
1. software that supports modes of instruction with alternative frameworks

The development of software that supports modes of instruction with alternative frameworks is an issue that is now being addressed by researchers, but may not be being realised by developers of educational software. In this paper the question of how we implement the class of theories referred to as constructivism as learning environments for use in classrooms is addressed.

The starting point for the development of instructional technologies was with the belief that the instructor was designing a learning experience for a group of students. All the construction of meaning and the best way to represent concepts and ideas was undertaken by the instructor or designer. Essentially the models of communication employed were teacher to student and prescribed sets of activities for the students in which they practiced the required concept. The rise in cognitive approaches to learning during the seventies has re-focused this relationship so that we now view the learning environment as something the learner has a major impact upon, the process has to include the learner as an active participant. Thus we can place the learner in a variety of contexts and provided them with a number of different views of the learning process through which they come to know about the world.

As part of this reassessment of the range of resources and how they contribute to understanding, the conceptions of learning which predominate amongst instructors are also being re examined. Consider the five views about learning which were uncovered in a study by Säljö (1979)

1. Learning is acquiring information or 'knowing a lot'.
< >Learning is storing information that can be reproduced.
2. Learning is acquiring facts, skills and methods that can be retained and used as necessary.
3. Learning is making sense or abstracting meaning, relating parts of the subject matter to each other and to the real world.
4. Learning is interpreting and understanding reality in a different way. Learning involves comprehending the world by reinterpreting knowledge.

There is a qualitative difference in these views. They move from the sponge view to a view which requires the learner to work actively to understand the evidence they receive from the world and be aware of any framework or process which they apply to its interpretation. Studies have now consistently found that higher order thinking skills are not acquired through didactic approaches, but rather through learner's active involvement with information (Collins. Brown & Newman, 1989, Resnick, 1987). This latter view has been linked with theoretical views such as constructivism to produce design guidelines which challenge the framework of instructional technologies.

There have been a number of writers who have taken on this challenge of how we answer the question of implementing the class of theories referred to as constructivism as technology supported learning environments. Initial support for designers has come from proposed sets of design guidelines and instructional design goals that attempt to incorporate various views of, in particular, social constuctivist frameworks.

For example, Cunningham, Duffy and Knuth (1993) devised seven pedagogical goals for designers of constructivist learning environments:

1. Provide experience with the knowledge construction process
2. Provide experience in and appreciation for multiple perspectives
3. Embed learning in realistic and relevant contexts
4. Encourage ownership and voice in the learning process
5. Embed learning in social experience
6. Encourage the use of multiple modes of representation
7. Encourage self-awareness of the knowledge construction process

Later, Savery & Duffy (1995) described four principles that should be applied to modern technology-based learning environments based on constructivist views. These principles are:

1. Learning is an active and engaged process. "learners are actively engaged in working at tasks and activities that are authentic to the environment in which they would be used." (Savery & Duffy, 1995, p.37).
2. Learning is a process of constructing knowledge.
3. Learners function at a metacognitive level. Learning is focused on thinking skills rather than working on the "right answer the teacher wants". Students generate their own strategies for defining the problem and working out a solution. Student can gain wisdom through reflection.
4. Learning involves "social negotiation". Students are able to challenge their thoughts, beliefs, perceptions and existing knowledge by collaborating with other students thus assisting their cognitive development process.

More recently, Duffy and Cunningham (1996) have reinterpreted their original list of pedagogical to develop a list of seven "metaphors we teach by" as basic assumptions for design.

1. All knowledge is constructed: All learning is a process of construction
2. Many world views can be constructed: hence there will be multiple perspectives
3. Knowledge is context dependent, so learning should occur in contexts to which it is relevant
4. Learning is mediated by tools and signs
5. Learning is an inherently social-dialogical activity
6. Learners are distributed, multi dimensional participants in a sociocultural process
7. Knowing how we know is the ultimate human accomplishment

These basic assumptions have taken account of the sociocultural interaction of constructivist theory, incorporated learner ownership and voice in leaning into the construction process. A new element "learning is mediated by tools and signs" has been added to imply that the tools (technology) and signs (semiotic tools) we use change the form, structure and character of activities and thus our knowledge.

For designers to adopt these frameworks, they will also need to reassess the instructional design paradigms they are using. It is no longer sufficient to use the traditional hierarchical, prerequisite sequences (ie learning taxonomies) that concentrate on recall and application of knowledge. Jonassen and Tessmer (1996/7) have described a taxonomy that "elaborates structural and higher order cognitive, metacognitive and motivational leaning outcomes that are not included in the currently-used taxonomies of learning outcomes."

That is, if we are to base learning environments on the framework that Duffy and Cunningham have proposed, appropriate learning outcomes associated with this view of learning will need to be devised, and no longer should we rely on the behavioural bias of current instructional design models. Jonassen and Tessmer (1996/7) have proposed that we need to develop strategies that support:-

1. Active learners to engage in interaction with and manipulation of the exploration environments that we construct.
2. Learners to explore and strategically search through these environments
3. Intentional learners willingly trying to achieve cognitive objectives
4. Conversational learners engaged in dialogue with other learners and with instructional systems
5. Reflective learners articulating what they have learned and reflecting on the processes and decisions that were included in the process
6. Ampliative learners who generate assumptions, attributes and implications of what they learn

Many writers have sought to develop guidelines for developers of software that will support new modes of learning. By examining educational products based on these frameworks we can review the effectiveness of the application of new schema for categorising learning outcomes.

With an understanding of the shortcomings of much of the commercially generated learning packages available for new approaches to learning in the early 90's, we sought to combine the ideas of constructivist learning environments, situated learning and problem-based learning in rich information landscapes to form the basis for effective design.

There has been considerable controversy, which will no doubt continue to simmer, over the interpretation of the term constructivism. Phillips (1997, p85) claims that

Using the distinction proposed by Grandy (1997, p44) we are referring to cognitive constructivism (as opposed to epistemic or metaphysical constructivism) when proposing the idea of constructivist learning environments. Within this framework we have been working with what Geelan (1997) has categorised as an emphasis on a social focus as opposed to a personal one and an emphasis on an objectivist epistemology as opposed to a relativistic view. Within this social constructivist framework we support Ogborn's (1997) argument for an insistence on active learning, a respect for learners own thinking and on a high priority given to ideas taught to students making sense.

Earlier we have proposed (Hedberg, Harper & Brown, 1994) that learning outcomes from our packages depend on the starting points of ;the learning environment;the learner's view of the purpose of the task; and the motivation of the learner. The process of learning involves the construction of meanings by the learner from what is said or demonstrated or experienced. The role of the teacher is one of facilitating the development of understanding by selecting appropriate experiences and then allowing students to reflect on these experiences.

To the learner, the constructivist learning experience may not look welcoming. It may seem daunting and complex to those who feel ill-prepared for such creative freedom and choice of direction. Often constructivist learning situations suddenly throw students on their own management resources and many fend poorly in the high cognitive complexity of the learning environment. Cognitive support tools and the explicit acknowledgment of the double agenda of metacognitive self-management and learning can help. The scaffolding and coaching of the cognitive apprenticeship model offer another solution.

A number of multimedia design models have been developed which illustrate the combination of complex learning environments and which also give students their own real control over their learning environment. Our approach is based on a more organic and iterative approach than traditional instructional systems design. Figure 1 outlines our approach to design, based on three distinct phases with specific outcomes from each phase. The first phase takes the basic information derived from a needs assessment and converts it into a description of the Project space-the learning outcomes, the information which is to be included in the materials, how it is structured, what the target audience understands about the information and how it might be structured for the audience. A possible structuring device might be a concept map of the ideas and links that are to be included in the project.

The second phase reviews the basic description and seeks to link the elements through an appropriate instructional or presentation strategy. It also seeks to identify metaphors which help both the design team and the final presentation of the information structure. The outcome of the second phase would be a formal description such as a design brief. The detail would enable the reader to understand the underlying knowledge structures and the ways it is proposed to link them conceptually and intuitively.

The third phase is a third pass at the same material, this time with the express goal of linking the design ideas into a potential interaction structure. One output of this phase would be an interactive mock-up of the interactive materials using an authoring tool to illustrate not only static display of information but also the graphical and visual metaphors used to create understandable links. The information included in this prototype may include visual, motion, static graphics, sound and data landscapes as appropriate to the concept under development.

Each interaction consists of a node point which forms the basis of the interaction, a set of options which provide links to other nodes or additional information attached to the current node. One of the links must relate to earlier travelled or preferred paths through the materials, and each choice must inform the user about what is likely to occur as a result of a choice. These can translate into the traditional concept of results (correct or incorrect) or information feedback choice, but should also include simple feedback elements such as confirmation of choice (feedback that a button has been selected) or performance support enhancement such as suggested hints, or revision of the underlying concept/principle which might be employed to make the choice. Depending on the instructional strategy chosen another element might include the concept of duration, either time or the limit of options based upon previous choices or paths taken. What constitutes each of these functions and what they create in terms of cognitive skill development for the user are determined by their physical manifestation in terms of navigation options.

The complex integration now possible with a variety of hardware and software combinations raises problems for the user in that multiple paths are possible to the same or different end-points. Learners are faced with the need to understand what learning possibilities might be available from where they are in a multimedia learning environment. When a student can branch down multiple paths and rapidly change the direction and focus of the learning sequence, there is possible interference with effective learning through the inappropriate application of information by the learner to their internal schemata. Other concerns include disorientation with location, cognitive overload when following several trails or trying to remain oriented on their goal, flagging commitment and a poor presentation rhetoric or metaphor.

In moving toward such goals, we were faced with a variety of technologies and strategies. The challenge was to devise innovative and motivating applications regardless of any specific hardware format and to incorporate appropriate learning strategies and tactics. It should not matter whether the learning environment be CD-ROM or Web based, it has to be designed to enable both learners and instructors to function in a number of roles. Consider the options available within a networked learning environment. At one extreme we have the typical classroom, where the teacher and learner share the same space at the same time, and learners may work individually or in groups. At the other extreme, the teacher and learner can be at different venues, communicate asynchronously, and learners may or may not congregate to share their experiences or collaborate/cooperate with learning tasks.

The multitude of ways the teacher and learner can communicate, and the time and feedback quality of those communications largely determine the success of the teacher/learner relationship and the learning outcomes. With developments in educational software and the proliferation of both bounded interactive multimedia titles on CD-ROM and unbounded resources available through the Web, the learner usually occupies the role of software user. When the activities of the learner are regarded as the central focus in education (Schank and Cleary, 1995) and emphasis is placed on what they are actually doing when using the software, the question can then be asked: How can we best support knowledge construction?

As a software user, their actions may encompass the full range of activities offered by software designers, from passive guided direction in prescriptive environments through to simulations and open active gathering and re-construction of multimedia resources. What they learn and how widely they use that knowledge, skill or strategy is a function of the context of program use- the learner extracts from a program what sense they make of it, not what use the designer intended.

If the learner is a software producer, the next question to ask is: Why is the learner producing interactive multimedia software? If their focus is on the development of an interactive multimedia product - whether a CD ROM title or a Web page, then the emphasis will be on learning about interactive multimedia production as a body of knowledge with an accompanying set of skills- a situated and authentic activity which synchronises learning and doing, yet an activity in which the acquisition of content knowledge is a fringe benefit. This equates with designer as learner (Jonassen and Reeves, 1996). In this context, the cognitive load required to effectively use many production tools may be very high and require substantial time to "polish" the outcome.

If their emphasis is on the learning which occurs through the process of interactive multimedia construction- learner as designer (Jonassen and Reeves, 1996), then the nature of the product is far less important than the knowledge construction process which the learner experiences along the way. Less emphasis is placed on the refinement of production skills and more emphasis is placed on student initiated design and development with just-in-time skill support. When the focus is upon the process, the cognitive load of the construction tool(s) should be minimal to permit the learner to focus on knowledge construction.

Learning theory has influenced the structure of interactive multimedia CD-ROMs along an instructivist to constructivist continuum (Figure 2). There are many fine examples of software at points on this scale which when viewed collectively indicate a number of trends. For the software user, as you move towards the constructivist end of the spectrum, there is an increase in the potential for group interaction as the nature of tasks becomes more complex and learner generated.

The range and extent of user interaction with the data in the software increases, as the user is given more freedom to navigate, access, determine the format of information representation and manipulate the data using cognitive and metacognitive tools. All this freedom to access and manipulate data can be presented in an information landscape which provides context and supports structures. The landscape can be quite extensive and extensible if the aim is to cater for users with a broad range of background knowledge .

Constructivist software need not be used by a group, however, the individual user in this more democratic environment needs to display the motivation and metacognitive skills of a self-regulated learner to gain maximum benefit from the software without peer support. The group provides a discussion forum for suggestions, ideas and debate, a multitude of learning and problem solving strategies to share, and immediate personal feedback on all communication channels (auditory, visual, body language). Such group benefits are only achieved once group members have acknowledged the need to refine such skills as negotiation and collaboration.

As part of the design process described in figure 1, there are a number of key design issues that need to be specifically addressed. During the various phases of our design process there is a constant review of the influence of a series of design issues that are in effect constraints on the overall design of a project.

Information in a multimedia world can be presented in new and different ways. This may seem simplistic but it is an important consideration when designing or evaluating products and should not be overlooked. Designers of what users would describe as resource rich learning environments need to invest considerable time in the structuring and designing of representations of information. Renewed interest in visual literacy and the power of the visual representation of information and ideas is evidenced from the number of publications on these topics that are now becoming a crucial part of the multimedia literature (Olsen, 1997).

Viewing a map of the land or a spatial layout may provide an understanding of problems which are not evident in a straight text-based description of the issue. Providing access to information in different ways for different types of learners requires developers to employ a variety of devices such as metaphor, cognitive tools and search engines.

Creating access to information, especially using "hyper" links, can create new meanings not previously considered possible. Using interactive CD-ROM multimedia to model the knowledge base and to give the user freedom to interact with it, gives autonomy back to the user. Rather than provide a set of pre-designed sequences that assume one learning model, a more interactive approach can be developed by giving the user a bounded information landscape and the tools necessary to explore and investigate the information. Package designers have used a variety of techniques to help users around such bounded information packages. Unbounded learning environments, such as those developed for the World Wide Web bring with them a new series of problems to solve with cognitive load a key issue to resolve in developing navigation options.

Clear information representation and access facilitate the user's ability to find and manipulate much of the available information. Many proponents of the use of interactive multimedia talk about the interactivity involved in the exciting dynamic programs. However, this can sometimes be a trap. Allowing the user to simply choose between a set of options or turn pages of cute animations is not interactivity. Nor can it be claimed that it is user control. It is important that the user is required to think before a response is possible. Consider a typical arcade video game- only a few control buttons are provided, but the user can make a character jump, flail their sword, etc. This means that the choice and its consequence are part of the interactivity and intrigue of the game. It is the "stuff" which creates high engagement. However, adventure games often proceed along a time and movement axis which allows the user very little control over the direction the adventure might take. While this may be appropriate for a game, this type of movement through a learning environment might be very constricting and frustrate rather than engage learners.

Used effectively, the technology can allow users to interact in ways that the designers of the system did not plan and well designed interactive multimedia materials make it unnecessary to structure materials in advance for the user. Effective student use of unstructured materials however, will depend on access, understanding and available tools. Flexible access to the information caters for a broad range of users. Clear understanding of the metaphors used to structure information permits the user to clearly identify the mental model of the information provider, and thus glean deeper meaning from that structure. The use of cognitive tools may permit the user to extract, create, organise and orchestrate information their own way when solving personally meaningful tasks.

There has been a renewed interest in motivation with the current emphasis on authentic learning outcomes (Brown et. al., 1989) with strong support for the intrinsic form (integral to the learning environment) as opposed to extrinsic motivation (outside the learning environment). Motivation tends now to be perceived as a process more initiated by the learner than offered by external instructional events ( McCombs, 1994). Designers need not only to consider the power of authentic learning environments in enhancing the motivation of learners but also to understand that learners will have or are going to need self management and reflection skills.

As a strategy, problem-based learning has been raised to motivate learners and generate high quality learning outcomes. Engel wrote: "Problem-based learning is thus particularly suited to assist students towards mastery in a range of generalisable competencies and to support ... learning in the cognitive and affective aspects of a course." (Engel in Boud & Feletti, 1991, p. 29). This is quite a pivotal statement for our understanding and appreciation of how "learning" takes place and how problem-based learning as an instructional strategy can facilitate such learning. This statement is based on the premise that one learns more by actually doing something than by being told how to do it. As a pedagogical approach originally discussed in the mid 1950s problem-based learning is based on the premise that students learn more effectively when they are presented with a problem to solve rather than just being given instruction. As Stepien & Gallagher (1993) state: "Problem-based learning turns instruction topsy-turvy. Students meet an 'ill-structured problem' before they receive any instruction". In this way, students themselves "identify, and search for the knowledge that they need to obtain in order to approach the problem." This turns the normal approach to problem solving found in university and college programs on its head.

Perhaps the most institutionalised form of problem-based learning is the model developed by Howard Barrows for use within medical curriculum. (Barrows and Tamblyn, 1980 cited by Ross, 1991, p. 34). Barrows defines PBL as "...the learning which results from the process of working towards the understanding of, or resolution of, a problem." The key stages of the Barrows model can be summarised as: Problem Analysis; Information Gathering; Synthesis; Abstraction; and Reflection. The five phases can be implemented in a variety of ways and over various lengths of time. Savery & Duffy (1995) state that in their own implementation of PBL one problem presented carries through for the whole semester. Boud & Feletti (1991, p.21) claim: "Problem-based learning is an approach to structuring the curriculum which involves confronting students with problems from practice which provide a stimulus for learning."

Our examples are drawn from three multimedia packages which have been developed to allow learners to participate in communities of practice through immersion in authentic activities. The activities are embedded in realistic and relevant contexts and are not only visually accurate representations of real-world environments, but are also rich in real-world data and related information.

Investigating Lake Iluka is based around an ecology simulation and employs a number of different interface metaphors in presenting the materials to the user. The package is based on the concept of an information landscape that incorporates the biological, chemical and physical components of a range of ecosystems that make up a coastal lake environment. The representation of the information landscape attempts to situate learning in an authentic, relevant and realistic context. This representation has been designed to address the qualitative issues, in that the cognitive skills required are authentic, as well as the physical issues in that the graphic and metaphorical representations are believable. The user is given some problem solving strategies to investigate this information in a variety of ways using the range of measuring tools provided.

The package provides experience in and appreciation for multiple perspectives by allowing the user to access and collect media information and 'construct' their own understanding of the basic ecology concepts embedded in the package by using a text based Notebook (Figure 3). This facility provides the means for students to participate in the construction process and offers experiences to learners in knowledge construction. Hedberg and Alexander (1994), have argued contexts of this sort can provide a useful context for the development of what Lave and Wenger (1991) call legitimate peripheral participation. This participation refers to the engagement of a novice in a socially-based practice in which they can perform the same range of skills as an expert. Investigating lake Iluka is a structured environment which allows the novice to work with problems and learning situations which are some "distance" (peripheral) from the core of the expert's world. As the novice begins to practice more as a full practitioner the skills and shared experiences overlap more with those who are acknowledged as expert.

Inquiry and problem-solving techniques have been embedded in the package through case studies of ecological scenarios. These are presented to the user via surrogate media reports of the problems. Each scenario can be investigated using a range of measurement tools and selection of resources from specially written and organised information resources which have been designed to represent a variety of views. In this form the package is offering students the tools to develop knowledge which Rorty (1991, in Duffy and Cunningham, 1996) describes as "a consensus of beliefs, a consensus open to continual negotiation". However, knowledge is also a construction supported by participants in a community that simultaneously transforms and is transformed by such construction It is the interaction with others that creates the awareness of multiple perspectives. This process of engagement is not specifically supported in this package. We have encouraged the process by incorporating advise to teachers on using collaborative groups and reporting views, but we have little information on how the package is being used by the wider community of teachers at this point.

It is expected that users will develop a broad array of scientific investigation skills using this realistic simulation and the facilities will provide experience with the knowledge construction process.

Within constructivist frameworks, cognitive tools can help learners organise, restructure and represent what they know. It was considered that a series of cognitive and possibly metacognitive tools could be developed to support the perceived needs of the learners and incorporated into our design processes.

Jonassen and Reeves (1996) have summarised the foundations of cognitive tools research and have identified the following key principles in the context of multimedia design:

• Cognitive tools will have their greatest effectiveness when they are applied to constructivist learning environments.
• Cognitive tools empower learners to design their own representations of knowledge rather than absorbing knowledge representations preconceived by others.
• Ideally, tasks or problems for the application of cognitive tools should be situated in realistic contexts with results that are personally meaningful for learners. (p. 698)

Evaluation of Investigating Lake Iluka and use of the package in classrooms gave the researchers further insights into the use of a fuller range of what Duffy and Cunningham (1996) have called the 'metaphors we teach by" in a constructivist learning environment (Hedberg, Harper, Brown & Corderoy, 1994). Students working with the package expressed a need to:-

• have more access to the information resources to construct their own knowledge (ie access to the multimedia components)
• be presented with a challenge that was offered up-front and puts the challenge in context
• have access to the information in multiple forms and modes,
• have more support for the tasks undertaken and
• have access to what they called "suck it and see" tools- ie what if scenarios or simulations.

Additionally cognitive tools to support the user have been shown by Jonassen and Reeves (1996) and others to enhance the learning process and to support the users' investigations. If students are to truly create their own meanings and understandings of the phenomena they encounter, designers need to not only incorporate user tools which will enable them to present their findings using the full array of resources contained in the packages, but also support their investigations with powerful cognitive tools.

Schank and Cleary (1995) and Korcuska (1996) have described a set of innovative learning architectures based on their conceptualisation of realistic learning situations. They have created powerful example implementations of cognitive tools where different cognitive learning strategies are built into software and the learner is encouraged to explore their ideas and solutions with differing degrees of support and advice.

The innovative use of cognitive tools in interactive multimedia learning environments has also been reported by Lajoie and Greer (1995). The package Bio-World is an interactive learning environment designed to support the acquisition of scientific reasoning skills in high school students and integrates a variety of cognitive tools to assist in scaffolding scientific reasoning activity. Users of this package are engaged in explicitly justifying hypotheses with evidence; organising, categorising, and rating evidence; and constructing a final summary argument on the topic of bacterial and viral infections.

With a clearer understanding of the needs of learners, supported by the range of constructivist environments being reported in the literature, a further ecology package was developed.

Exploring the Nardoo provided a richer information landscape of resources based on a geographic metaphor which incorporated a Water Research Centre and a navigable river environment. The problem-solving challenge for students to become active participants in the learning process is presented on entry to the environment, the data collection facilities allow collection of a full range of media forms and simulators allow the user to ask questions and investigate possible answers to those questions . Exploring the Nardoo provides the student with a flexible set of tools made available through a personal digital assistant (PDA), Figure 4, to assist in the investigation process.

By providing a metaphor relating to the real world, students are encouraged to apply scientific concepts and techniques in new and relevant situations in this ecology-based application, throughout the problem-solving process. In so doing, the learner is likely to become more interested in developing questions, ideas and hypotheses about the learning experiences encountered. As an alternative teaching/learning strategy in the development of inquiry and problem solving techniques, this package incorporates high quality visual materials in the form of graphics, sound, text and motion video together with scientific measuring tools to aid in the construction of understanding.

The process of using source material within the package in support of an investigation has been enhanced to allow the student to:

• Decide precisely on the quantity and selection of text to be copied into their notes. This is either through making a selection and then 'grabbing' it into the PDA or by using a 'drag and drop' technique where the target text is selected or highlighted and 'dragged' into the notes module of the PDA .
• Use marker buttons as pointers to video, audio or picture information which can be displayed within the PDA's viewer along with any linked information.
• Manipulate marker buttons and text within the notes areas, via 'drag and drop', to facilitate the re-ordering of ideas in the process of building an investigative response in the form of a report, explanation, procedure etc.
• Use text style tools, within editable text notes, providing the opportunity to use font colour, style and size as organising criteria within the notes.

The joint combination of note book and viewer equips students to view and then critically evaluate or compare different representations of the same information concept. By collecting different media representations of the same topic and 'flipping' between these representations at their discretion, the student has the opportunity to establish cognitive links between different media forms which complement each other and support a central theme or information focus.

The package also provides the ability to record thoughts and impressions 'on-the-fly' whilst examining media stories. This provides the potential for students to reorganise or revise their thoughts to better 'make sense' of what they see and hear. Students are able to document their emerging ideas in support of an investigation or problem solving exercise whilst viewing different media. This provides support in the formulation of new schemata in the process of accommodating the new information.

Successful problem solving activities are reliant on numerous individual, social and environmental factors. From a technique perspective, Exploring the Nardoo endeavours to assist students by providing cognitive tools, or templates, upon which they can build their note taking or response writing activities. These are in the form of writing genre templates. Students may access the book containing these templates (as well as other organisational help on note taking, presenting and filing) from within the Water Research Centre - a metaphor within the information landscape of the package. Genre descriptions can be viewed and a genre template can be copied into the notes and used as a scaffold upon which to build or fill-in relevant information found whilst exploring the package.

To facilitate the re-ordering or re-prioritising of information Exploring the Nardoo provides a separate, expanded form of the notebook. This device has been termed a 'text tablet'. It provides the editing facilities offered by the PDA as well as other features to assist with the restructuring of notes into a form more suited to small group presentation or a particular genre style. The text tablet provides a larger expanse of editable screen/document space into which student notes may be copied to/from the PDA notes module. In addition we have sought to support social- dialogical activity by incorporating a presentation mode for students to present their solutions and ideas to other co-researchers and the whole class.

A writing genre template can also be loaded directly into the text tablet into which portions of the student's notes may be copied or dragged. Notes from prior sessions can be loaded into the text tablet and used in support of current investigations. Being able to store and report thoughts and impressions derived from media experiences by using the media itself (actual video/audio and pictures, not just text representations of the media) provides a more powerful means of 'reformulating' (Schroeder & Kenny, 1994, p 965) ideas. We are aware however, as Schroeder and Kenny (1994) point out "learner's not accustomed to this technique and multimedia facilities will require instruction in its use" before they become proficient with the technique but once accustomed to it the student has a powerful process at their disposal to gather, organise and illustrate their ideas. Support for teachers and students in the use of these features has been modelled using walk-through movies made available through the help system and also detailed in support notes available in reference books within the package.

For students to 'test' ideas, three simulators have been incorporated: a Personal, Home Water Use simulator (Figure 5), a 'whole catchment level' water demand Dam Management simulator and a Blue Green Algal Bloom simulator. Each of these simulators is a powerful exploratory tool, which provides support for the solution to one of the embedded problems by mimicking a 'real world process'. They enrich the 'quality' of the problem solving process for the user by providing unhindered access to act and become immersed in a 'real' situated process, manipulating the various causal parameters and testing hypotheses without a 'real' consequence or risk and in a time frame which is convenient and manageable for them "and enabling the learner to ground their cognitive understanding in their action in a situation" (Laurillard, 1996).

More able users are provided with the facility to solve problems at a deeper level through the testing of their own "what if" scenarios. This can, during the course of solving problems... "facilitate more detailed exploration and learning by;

• allowing the user to take readings at a sight and study the changes as the simulation runs,
• allowing the monitoring of all parameters while the simulation is running, with the aim of exploring the relationships between them" (Corderoy et al. 1993)

The incorporation of multiple forms of feedback in these simulators has added to their power as tools for learning. Data input is simply achieved and the values are displayed in windows during the simulation, allowing the user to check them against the outcome of the simulation. Likewise output data from the simulation is also visible at all times and the format in which it is displayed is user controlled. Pure numerical data is displayed in the output data windows, while the main screen of the simulator may be toggled between a graphical display or an animation of the process.

The ability to directly compare input data with output data in various forms simultaneously is a powerful feature of each of these simulators and helps the user in making connections and associations and forming an understanding of the interrelationships between 'cause and effect'.

As an example the Personal Home Water Use Simulator (Figure 5), is designed to provide the user with an indication of how efficiently they are using water around the home by comparing the use in their home with that of the national average for a home with the same number of occupants.

The user is able to set the 'levels of consumption' in various parts of the simulated house by using the 'up' and 'down' arrow buttons on the numerical input windows. Having entered the data the simulator provides a total consumption in kilolitres per quarter for the household compared to the 'National Average'. It also provides a cost for this water which is based on the cost per kilolitre, set by the user.

As an approach to offering students a facility for sharing ideas, we incorporated a presentation mode into the text tablet, with support tools for constructing of presentations in offered in the presentation guide. The tool, which facilitates student reflecting on their ideas with group or class interactions, still does not support students sharing ideas throughout the knowledge construction stage through social diaological processes.

The Nardoo project enabled the placement of resources within a spatial framework which was largely coded through a geographical metaphor, thus it reduced the learning and searching requirements. The Stage Pass project sought to move away from the narrow scientific concept of data and work with the construction of multiple meanings in a field that many would argue is highly subjective and open to numerous interpretations. Stage Pass introduces the learner to the world of Australian performing arts by exploring a performing arts venue (the Sydney Opera House) which showcases contemporary companies' performances, processes and people, and provides theatres and "tools" with which to direct scenes.

Many students have a narrow perception of how a theatrical performance is devised due to the lack of exposure or opportunity to view or take part in the performing arts. By extending the boundaries of interactivity in the context of a virtual world, we have provided learners with opportunities to express their own cultural interpretations and understandings. The project sought to:

• reveal the diversity and range of the performing arts in Australia
• inspire with excerpts of Australian productions
• encourage participation in the performing arts
• discover theatrical creativity through the creation of a scene
• explore and learn about backstage elements and processes
• explore the roles and career paths available in the performing arts

In this theatrical journey, we had the advantage of working with many visual metaphors. The world of theatre, opera, music theatre, dance and contemporary performance styles can be explored through devices such as "The Green Room" where the user can interrogate a database of the contemporary world of performing arts. The "Stage" space provides the opportunity to view sample scenes from productions which have been created by professional directors, using the same resources as are available to the user, and, more importantly, to personally direct and design scenes. In this process individual users explore processes of visual design, sound development, and the concept of direction and motivation. In this project the construction tools have also been extended to enable the user not only to collect from a defined set of resources, but also to construct their own resources based upon combinations of sets, costumes and performers.

The Stage space (Figure 6)with its database of vocal performances and movement animations and design elements provide the user with a rich set of tools enabling them to create performances with a strong feeling of stage depth, a rich sound accompaniment and effective, realistic animated performer movements. The creative key to the activity of directing a performance is to provide a wide range of choices, to this end we have developed a variety of video-based animations of recorded movement and multiple audio recordings of each line of dialogue. The user can identify the actor's intention behind making a movement or saying a line and this is used as the basis for selecting the recorded movement or speech. This approach is designed to move the user's exploration beyond the simplistic attachment of a line of dialogue or an action to a performer towards experimenting with the more realistic, complex issues associated with directing and delivering a performance.

Stage Pass enables the user to save their individually or group directed performance for comparison with others solutions to the task of directing a scene. In support of the free-form creativity available within the stage space, the user is asked to construct an appropriate set design for their performance. After trying their hand at the task they might ask how an "expert" in this theatrical area might have attempted the same task. Users are then presented with the professional designer's solution and they can listen to the logic of how it was developed. This modelling helps that novice make similar decisions to theatrical professional.

Stage Pass thus brings together several of the Duffy and Cunningham (1996) pedagogical goals. While working with the application, learners will be working within an environment which mirrors the world of theatre and supports the theatrical outcomes of many interpretations of each scene. Furthermore, each constructed performance can be compared with experts or other students, and learning can occur through the resolution of multiple responses to the same task. Key to the communication of the experience within this application, is the facility to save, share files between learners or re-present your constructed performance and interpretation to others located within the same classroom or across the Internet with other learners from different cultural backgrounds. This last act offers the potential to create a range of resources that are not bounded by the storage capacity of a prepackaged CD-ROM but are collected and shared from an ever growing unbounded Internet learning environment.

Learners using bounded or unbounded resources need to acquire a common set of skills in information appraisal, selection, organisation, structuring and communication of ideas in the solution of meaningful tasks. The nature (instructivist to constructivist) and source of these tasks (whether teacher or learner generated) help determine their relevance, complexity and ability to promote a deeper orientation to learning.

Learners as knowledge constructors share some fundamental features, whether working individually or in groups, on traditional screen based structures or within more of a web page hypertext architecture. They develop diverse skills in data/media selection, generation, organisation, orchestration and re-structuring. The opportunity to share the results of construction labour with others permits these active learners to gain feedback as to whether their level of understanding is sufficient to facilitate meaningful communication of ideas to others.

The means of information representation is as unique as the individual or group. Interactivity resides in both the recursive construction process and within interpersonal communications. It no longer relates to the ability of a user to manipulate data within a fixed structure. The ultimate in interactivity and motivation is the process of personal knowledge construction. The additional skills demanded of the learner as producer deal with resource management and increased metacognitive awareness of what is largely a self-regulated process.

Developing the packages described has challenged our traditional view of learning outcomes which we have tended to view as hierarchical prerequisite sequences. Jonassen and Tessmer (1997) have proposed that we can now engage a greater range of learning outcomes that those currently used which have been based on an instructivist view of instructional design. They have proposed a new taxonomy that elaborates structural and higher order cognitive, metacognitive and motivational learning outcomes that are not included in current taxonomies.

We now have the tools that will allow us to construct technology supported learning environments that can incorporate the type of learning outcomes that Jonassen and Tessmer (1997) have proposed such as hypothesising solutions, identifying/ defining problems, generating new interpretations, analogising, inferencing and semantic mapping/ conceptual networking.

Bounded learning environments can be designed to encourage student-centred investigations and knowledge constructions. However, the socio-cultural processes and social-dialogical aspects of constructivist learning frameworks are difficult to offer within these constraints. Table 1 outlines the aspects of the three packages we have developed that attempt to support Duffy & Cunningham's seven pedagogical goals.

(This project is a joint collaborative development with Professor Susan Metros from the University of Tennessee and the Applied Cognitive Science Research Group at McGill University.)

With a view to investigate how we could implement socio-cultural processes and social-dialogical aspects of constructivist learning frameworks within technology supported learning environments and being cognisant of the taxonomy of learning outcomes (Jonassen and Tessmer, 1997), we have developed some initial designs of a project that will attempt to take advantage of the strengths of both bounded and unbounded learning environments. The project is based on a virtual environment that will incorporate challenges for learners and information sources based on the metaphor of a Mall which will contain various locations that young people typically frequent. Spaces within the virtual environment will focus on specific challenges for learners and will include, clothing shops to choose, try on, purchase and wear clothes, card shop to design and send cards to friends and others on line, a food hall to choose both fast food and healthy alternatives, magazine rack to read articles about topics relevant to body image, a music shop, an appliance shop with rows of televisions with different media messages, a video arcade with game simulations relevant to specific issues, etc

Challenges embeded in the package will be randomly generated or selectable by the user and will address such issues as acne, pregnancy, body image, exercise, drugs, etc. Users will be able to select a persona or design their own and will be visually represented in the virtual space much the same as a in graphic MUD or MOO.

In order to develop the social-dialogical interactions users will be able to join the Mall via a school intranet and interact with students in their own school, or venture further to a World Wide Web based Mall and join learners from across the world. It is planned to incorporate a series of cognitive tools to support such learning outcomes as hypothesising solutions, identifying/ defining problems, generating new interpretations, analogising, inferencing and semantic mapping/ conceptual networking. The learning objectives will include learning skills to manage life situations, empowering students to adopt healthy lifestyles and values and attitudes such as a sense of worth and dignity as individuals, respect for values and attitudes of others and a sense of responsibility for personal and community health.

We believe that as we have progressed through our learning environments we have developed a greater range of tools which can support learners in their knowledge construction. For the future, we can see some exciting developments which, when combined with recent developments in technologies, can offer more comprehensive learning environments which in turn will more than meet the goals that Cunningham and Duffy have proposed.

The research direction our future products will focus upon will include:

The fundamental difference between the bounded and unbounded resource banks resides in the essential nature of learning activities. The design of interactive multimedia learning environments will increasingly include authentic activities which can be understood by less dexterous learners with specific cognitive tools, and different forms of representation. The development of the Web has enabled more components of a constructivist pedagogy to be included in the experience for the learner in any place and in any cultural milieu. The challenge for researchers is to demonstrate for developers how to capture these opportunities and support the intrinsic motivation of learners to explore their own world and the variety of viewpoints within it.

We wish to acknowledge the contributions to the ideas presented in this paper by our colleagues, Christine Brown, Robert Corderoy, Grant Farr, Brian Ferry and Robert Wright.

(c) Barry Harper, John Hedberg

The author(s) assign 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(s) also grant 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 97 conference papers, and for the documents to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the authors.