Collaborative Innovation as a Process for Cognitive Development.

In this article we propose a methodology for collaborative innovation, which leads to cognitive development. Motivation for innovation could be an effective means for efficient and meaningful learning. Meaningful learning contributes to cognitive development. Therefore, in professional education the aim is to maintain the continuous cognitive development through lifelong learning.

Distributed cognition of people and artifacts are integrated during the process of collaborative innovation. This, in turn, enriches the individuals' cognition, metacognitive abilities, techniques of interpersonal communication, and reflective thinking skills. A mechanism for the process of innovation is discussed to identify the possible stages of interaction for collaboration. The guidelines for interaction are identified to develop an intelligent support system. This support system guides participants in carrying out innovative activities. Based on participants' personal portfolios, the network can be used to perform an intelligent compatibility search on the World Wide Web (Web) for locating companions for collaboration. The same tool can also be incorporated for other ontological searches in connection with the process of innovation such as suggesting possible problems to be explored and providing relevant information on available resources.

In this methodology, the main intention is to continue the cognitive development process where the innovative activities are used to sustain the motivation of the learner. Any other outcome from these innovative activities were considered incidental. In future work, we intend to develop a measure for assessing cognitive changes for providing periodic feedback to participants about their progress. An intelligent tool would be developed for periodic assessment and reporting cognitive changes to the participants to help in sustaining their motivation.

We started with exploring the various ways of using the Web for educational purposes. In this particular piece of research, we used the Internet as a virtual communication channel for lifelong learning. The Internet provided us with a huge database of available information, which could be shared between "n" numbers of people, and the knowledge space is available for any one person at any given time.

In the past decade, information science researchers have started to develop technology that can support a group collaboration process. Integrated collaborative research systems (ICRS) typically support activities related to a research cycle: assembling information from a wide variety of sources, organizing and manipulating that information, analyze the information, and synthesize new information. The projects involved may cover extended time periods and employ multiple researchers who collaborate on a series of related topics. Systematic and relevant information collection, indexing, and management is crucial to the success of research collaboration.

In this article, our purpose is to develop a rational basis of expected benefits from collaborative innovation for achieving cognitive development. Our prpose is also to identify the required tools and techniques to support such collaborative activities across the boundary of academic specialization. The system at which we are aiming is meant for people with interest in similar subjects and topics, but with different major(s) or specialization(s). Therefore, using the Internet, the participants can communicate, share, and collaborate through various, new forms of web-based activities.

We claim that collaborative innovations lead to cognitive development. In support, we introduce the three theories of guided learning as discussed by Brown and Palincsar (1989): zone of proximal development, adapted from Vygotsky's theory; expert scaffolding, commonly associated with Wood and Bruner; and Socratic dialogues.

The learning theories underlying the related notions of a zone of proximal development, expert scaffolding, and Socratic dialogues all rely heavily upon the mechanism of internalization to fuel individual conceptual development. Thus, activities between the participating individuals, that is, partners during the process of collaborative innovation, pursue cognitive development.

The article is structured as follows. We have already mentioned the theories of learning which are the basic platforms for our system. Next, we discuss how to incorporate the system's approach of the teaching-learning- evaluation cycle into the lifelong learning process and the positive aspects of working in collaboration. Then we provide the concept of distributed cognition and its relation to collaborative activities. Our analysis then gives a justification of how the motivation for collaborative innovation helps the learning partners to sustain the motivation for lifelong learning. Thereafter, we explain the process of creativity and where and why we plan interactions. We spell out the objectives and design of the proposed system. This leads to the role of the Internet in collaborative innovation, in particular: a) the generation of a checklist to clarify partners' desires and expectations in the collaborative venture, b) possible provision of intelligence searches for different purposes, c) and applicati on of the concept of fuzzy set for qualitative assessment of cognitive developmental stages of the participants in different domains of learning. Finally, we discuss the implications of this work.

SYSTEM'S APPROACH TO LIFELONG LEARNING

According to the system theoretic point of view (Bertalanffy, 1968, Ruben, & Kim, 1975), to guide or control a system to achieve and maintain a desired outcome, two sets of parameters are essential. The first is a clearly specified goal, that is, setting the desired outcome(s). Next, to stabilize the system performance against any changes in the system parameters over time and/or against the unpredictable inputs from the external environment, it is essential to have a feedback mechanism to make the system "aware" of the actual outcome. The comparison of the desired outcome and the actual outcome makes it possible to adaptively adjust the required action on the system to sustain a desired performance.

In the present context of leaning systems, each system is comprised of an intelligent and motivated individual having an infrastructure and resources available for learning. The rest of the society may be considered as the environment of the system, which provides unpredictable inputs both favorable as well as distracting to the system. In formal learning situations, feedback mechanism is incorporated in the form of tests, exams, and competitions to make the system robust against external disturbance stimuli. These mechanisms are expected to evaluate different manifestations of the outcome of the process of learning and feed the information back in the form of results. A schematic diagram of the model of a sustainable learning system is shown in Figure 1.

In the case of lifelong learning efforts in informal settings, both the objective setting and feedback mechanism leave room for improvements. While the degree of clarity of the desired objective varies from individual to individual, no reliable feedback mechanism can be guaranteed to all the individuals in for the presently available infrastructures. Though human beings are capable of self-assessment to a certain degree, an external feedback, especially from other like-minded individuals, is known to have better appeal for consideration. In the present work, we intend to offer a preliminary analysis and discussion of a possible option to provide meet the above requirements.

In our design of learning, activities among the collaborating partners is somewhat like an expert-novice interaction. In the process of learning, the expert-novice role is interchangeable since our presumption is that every individual who would be taking part in this program is expert in a field in terms of knowledge and skill.

Working in companion groups has many pedagogical advantages:

* It provides socializing of people in the context of work and increases their motivation by giving social (mutual) recognition for their knowledge and helpfulness.

* Peer help is deeply situated in a shared context and can, therefore, provide a stronger learning experience for the partners.

* It is a way to make learning happen when the problem or drive arises.

* It promotes processes of self-explanation (Chi, de Leeuw, Chiu, & La Vancher, 1994) and reflection within the partners and, in this way, "reciprocal" learning takes place.

* It is cost effective, since it uses the learners, themselves, as teaching resources.

* The drive for collaborative innovation helps in sustaining motivation for lifelong learning.

* It facilitates social interaction in a group of learners and helps to create knowledge-anchored personal relationships among them.

DISTRIBUTED COGNITION VERSUS COLLABORATIVE INNOVATION ACTIVITIES

It is impossible to account for developments of distributed cognitions without reference to individuals' changed cognitions (Salomon, 1993). However, it has also been stated by Salomon, Perkins, and Globerson (1991) that it becomes observable, if not patently evident, that the collaboration of individuals and computers is often characterized by intellectually superior performances that cannot easily be accounted for by individuals' cognitions alone.

Interaction in the process of collaborative innovation would mean that, while the joint products of a cognitively distributed systems (e.g., a jointly designed plan of action) cannot be accounted for by the operation of its isolated components (the inputs from the members of a planning team), each partner can still be seen as having qualities of his or her own, some of which (e.g., unique expertise) enter the distributed partnership and are affected by it reciprocally, while other qualities may not be so influenced.

It has already been discussed by Steiner (1998) that management and innovation skills are important for success in an engineering career. Therefore, if partners, companions, or team members are engaged in innovation activities, then in addition to advantages mentioned in the introduction section of this paper, this process would provide engineers with competencies essential for success in the global marketplace, for career enrichment, for advancement in government and industry and, perhaps, assist them in realizing their full potential.

A number of published models of innovation are discussed by Rutherford (1992). From these, he points out five key aspects of innovation that require attention to get the best out of any process. These are: (a) role (or task), (b) responsibility, (c) resources, (d) recognition (or feedback), and (e) rewards. In our system, we plan to provide support in all these aspects of innovation processes through the interaction of human and computer systems. Using the Rutherford's framework, Jarvis and Quick (1995) ran a project on innovation for an undergraduate course. They reported that the majority of students stated that working in teams was the most rewarding experience. Thus, collaborative innovation activities support the implications of distributed cognition, resulting in superior cognitions of participating individuals.

MOTIVATION FOR LIFELONG LEARNING

Motivation can be defined as a concept used to describe the factors within an individual (or society) that arise, maintain, and channel (direct) behaviour towards goals. Maslow (1954) developed a theory of motivation based on a hierarchy of needs. Maslow believed that each person has five categories of needs, namely: (a) physiological needs, (b) safety needs, (c) social needs, (d) ego needs, and (e) self-actualization needs. Maslow believed that the lower needs (physiology and safety, etc.) needed to be at least partially covered before higher needs (ego and self-actualization) can be activated. Maslow suggests that once we have activated these higher needs, we often view them as more important than the lower needs. In the context of lifelong learning, the initial source of motivation usually comes from the higher needs. Our proposed process of collaborative innovation would help the participant to sustain motivation for continued satisfaction of more important higher needs. Factors such as accountability, p seudo deadlines for time-bound learning, mutual appreciation and recognition, and immediate feedback on performance involved in the process of collaborative innovation help in maintaining motivation for lifelong learning. This can become impossible in the case of self learning due to other distractions in life.

Figure 2 shows the major components and activities involved in the process of learning through self study and through collaborative innovation process. As shown, the motivation component is reinforced and sustained by the process of innovation in the latter case, unlike in the case of self study only.

PROCESS OF CREATIVITY AND POSSIBILITIES OF INTERACTION

Creativity involves producing something novel and appropriate (Martindale, 1981). Findlay and Lumsden (1988) further define "discovery" as the product of the creative process, and "innovation" as a discovery that is accepted by the creator's society. We have designed our system of collaborative innovation based on this definition where society represents the whole world. The novelty and appropriateness of production of an idea should not be culture specific.

On the basis of H. Poincare's explanation of creativity, G. Wallas formulated the analysis of creative thinking. As described by Partridge and Rowe) 1994), according to Wallas, there are four stages to be considered: (a) preparation, (b) incubation, (c) illumination, and (d) verification.

* Preparation is the stage of concentrated work, accumulating data and background information, and trying out various approaches. Occasionally this gives rise to a solution but often, as Poincare noted, nothing is achieved.

* Incubation is the time when the mind is relaxed or working on a different problem. This corresponds to Poincare's notion of ideas floating about the unconscious, occasionally interlocking, while consciousness is focused elsewhere.

* Illumination is the moment of insight when the solution found is made conscious. This stage seems impossible to predict or control.

* Verification consists of a further period of conscious work when results can be checked, for example, against new data.

In our system, the computer can help one to identify the partner(s) and to provide support during the process of innovation. It can also help in searching for relevant information. All these would be very helpful at the preparation stage.

In addition, at the stage of incubation, computers may periodically remind the partners to continue interacting even when the interaction frequency falls below a certain level. Incubation and illumination are the stages where the collaborating partners reflect on newly acquired knowledge achieved through interaction between them during the stage of preparation. After passing through these stages, the partner may independently come up with a solution. At the stage of verification, partners would evaluate their own or their companions' solution(s). Also, they perform self-assessment and peer-assessment for cognitive development.

OBJECTIVES OF THE PROPOSED SYSTEM

* Makes lifelong learning more goal oriented.

* Provides a mechanism of feedback to stabilize the learning process against external disturbances.

* Helps cognitive development through collaborative innovation efforts.

* Creates a database of resources to support innovation and creativity.

* Links people with similar interests to facilitate joint projects, idea sharing, and benchmarking.

* Encourages innovation efforts for lifelong learning.

ROLE OF THE INTERNET IN COLLABORATIVE INNOVATION

Computer technology offers several approaches that facilitate and provide the necessary personalized resources for help that can be made available to mass audiences. There are many Computer Supported Collaborative Work (CSCW) tools that facilitate communication among peers. However, they rarely provide personalized help on demand. In our system, we plan to introduce companions for collaborative innovation. The system gives support in several ways by providing:

* Appropriate partners who are knowledgeable and are likely to collaborate on a similar type of problem or innovative idea.

* A shared workspace for the collaborating partners where they can share the context of the problem.

* An appropriate "discussion facility" where people can ask study-related questions and give answers, and discuss problems of common interest either synchronously (like a chat facility) or asynchronously (like a bulletin board facility).

* Intelligent, ontological searches on various aspects to help carry out the collaborative innovation process.

* Assessment of cognitive development in the process of collaborative innovation.

Description of the System Modules

The essential system components are shown in Figure 3. The purpose of the net search manager is to initiate different search engines over the Internet, to search for relevant information, and to intelligently filter, categorize, and store "links to the web sites" relevant to the subject and topics offered to the users. The information is stored in the resource database. The innovation support manager provides an intelligent interface to the problem database, companion database, and resource database. The selection of a subject of study, identification of appropriate problems, selection of the appropriate companion, and accessing the resource database are performed through this module. The addition and updating of the resource and problem database by the experts are also done through this module. The assessment manager implements the adaptive assessment of the users at various Levels of the Cognitive Domain of Learning (LCDLs) and maintains the results in the assessment database.

Intelligent Features and Sequence of Interaction With the System

In the past few decades, the availability of cost-effective and efficient storage devices and information systems resulted in a rapid growth of available databases all over the world. Though information collection and storage has become easier, the effort required for retrieving relevant information has become significantly greater. We have provided the outline of some of the intelligent features, including intelligent ontological searches (Chen, Schuffels, & Orwig, 1996) planned for the proposed system.

The intelligent features required to support the various stages of the process of collaborative innovation are summarized in Figure 4 and are discussed briefly as follows:

* Problem or topic identification: A participant can assess the system with a problem in hand or may ask for a problem in order to learn a new subject or topic. The problem could be obtained from the problem-database available in the system as per the subject to be learned. The problem-database is assumed to be created by the experts in different areas. The problems in the database are prepared to be of an interdisciplinary and innovative nature, which would require knowledge of more than one knowledge domain. Selection of the problem automatically gives a tentative list of subjects, which provides the required subjects or topics to be covered for attempting to solve the problem.

* Companion identification: The participants may either have pre-decided partners for collaboration or they may employ companion search engines to locate companion(s) having similar interests and desired Level of academic background and technical expertise. In order to select a companion one should supply information regarding:

-Subject specialization.

-Highest degree obtained.

-Other fields of interest.

-Degree of expertise in different fields (like novice, expert etc.).

-Number of years in the profession.

-Field(s) of professional activities.

-Number of free hours available per week for other activities.

-Internet access (hours per day).

-Library access.

To initiate a search a similar form has to be filled up for the desired companion.

* Supporting material search: A database will be maintained containing the tentative course-outline (from novice level to expert level) of different subjects. This database will also contain information on course related web sites, books, and journals, etc. These course outlines help the participants identify their knowledge in any subject area and optimize the learning process by selecting the appropriate entry level.

* Computer assessment: A question bank is to be developed by the experts for all the major subject areas. These questions would be put under the appropriate level in the cognitive domain of learning to provide assessments in different LCDLs. Through automated evaluation, only the first three levels can be assessed, since the higher levels are expected to have non-unique, subjective results. An appropriate set of questions will be presented to the participants periodically for assessment. The computer assessment module is to be adaptive to the cognitive level attained by the individual under test. The scores will be stored in the assessment database for periodic analysis of progress and cognitive development.

Example of Interaction

When the partners are chosen for collaboration, then a checklist of questions can give new points of view to explore regarding the process of collaboration. This can also provide a clearer picture of what each partner wants to do. Any such list is left up to the partners to pursue each question actively, to restate the selected problem to meet their mutual interests.

A typical problem for innovation can be the following: "Design an adjustable and movable chair, such that a handicapped person can use it for sitting, relaxing, watching TV, and sleeping at will indoors and outdoors." This problem involves learning, thinking, and working across several disciplines. The problem is ill-defined and needs many interactions among the partners to explore it. In order to clarify the problem, statement one of the partners may fill up the checklist as shown in Figure 5.

In the actual system, there are supposed to be a large number of such occasions for interactions. Through interactions, collaborating partners will learn from each other more efficiently and effectively. In this way they will develop reflective thinking abilities and communication abilities at a distance.

Application of Fuzzy Set Theory for Assessment of Cognitive Development

A triangular measure (Bhattacharya, 1994) of cognitive development assessment procedure has been envisaged for the present system. In addition to the periodic assessment by the computer as described previously, each participant will be assessed by a partner and by oneself. The data obtained from these assessments provides an approximate measure of the cognitive development of the participants achieved through transfer, sharing, and acquisition of new knowledge. Since it is not possible to measure the absolute degree of cognition, we tried to assess differential changes in the LCDLs.

A fuzzy subjective measure (Zadeb, 1975; Klir & Folger, 1988) is proposed for evaluating the changes in the degree of different LCDLs. These levels are: knowledge, comprehension, application, analysis, synthesis, and evaluation (Bloom, Englehart, Furst, Hill, & Krathwohl, 1956) in the increasing order of cognitive ability from knowledge to evaluation. These levels are not discontinuous since they also overlap each other. This does not mean that the levels have to develop in a definite order, one after the other, as stated. Here, we may mention that, as our proposed system is targeted for engineering professionals, we expect that one can reach a higher LCDL by spending less time in searching and comprehending information for learning a new topic. This is possible because these professionals are able to realize the gap in their cognition and, therefore, can reach a higher LCDL by knowing and comprehending the information from the new resources to fill the existing gap in their cognition. The purpose of assessm ent is two-fold. First, it is expected to provide a basis for getting feedback from the other companion(s). Second, the process of self /peer assessment by considering the different aspects of cognitive domain in detail helps in developing metacognitive and reflective abilities. Although a proper measure of cognitive change is difficult, if not impossible, to assess, a pseudo measure helps in providing a common vocabulary among the companions to convey their perceptions of cognitive changes in themselves and others.

The steps of the proposed process of assessment are as follows:

1. Initialization Stage:

a. Learning about the characteristics of the cognitive domain using a tutorial incorporated within the system.

b. Initial assessment of self and peer status of levels in cognitive domain of learning in a topic at a time.

2. Interaction (through question-answer sessions and/or discussions).

3. Perceived qualitative changes in each of the LCDL.

4. Assigning the perceived changes in different LCDL with a degree of belonging (membership value) to five fuzzy sets defined by linguistic levels as No Improvement (NI), Less than Expected (LE), As Expected (AE), More than Expected (ME) and Excellent (EX).

This measure is subjective since the definition of the fuzzy sets used are not universal and the individuals' assigned meaning about the set definitions are relied. As a guideline for consistency in assigning the membership values, use of an arbitrary reference of set definitions has been visualized. Typical definitions of the fuzzy sets are shown in Figure 6. Since the fuzzy variable in the present case is "change in LCDL"--no standardized scale and unit for the universe of discourse is available. Thus, the actual mapping of the perceived cognitive change to the fuzzy set membership values depends on individual experience and expectations. The shape and relative difference between the set definitions referred to during assignment could be different than that shown in Figure 6. As long as the order of the set definitions is appropriate and the same definition is always referred, the consistency of evaluation is acceptable. The use of fuzziness in the assessment of self and companion provides the necessary to lerance in the measure of an otherwise non-measurable variable. The use of only computer-generated objective tests and their scores are not sufficient to provide proper assessment of diverse cognitive abilities. Therefore, in addition to computer-based assessment, it is desirable to employ subjective assessment of participants using multiple communication modes and channels and various forms of interactions.

Engaging, interested people in a process of collaborative innovation could maintain motivation for lifelong learning. Especially in engineering education, development of innovation skills are required to keep pace with swift technological developments. Therefore, by going through our proposed system, the participants would develop the skills of collaborative learning which would be helpful in their professional lives. The success in the process and the learning experience gained through it would be a strong motivating factor for further continuation of the process. In the formal education system, we get feedback on our learning from external sources in the form of formative and summative evaluations. However, after one enters professional life, there is no system available to provide such feedback on one's learning. Therefore, learning possibly gets lower priority as compared to other priorities of life such as job perspectives or promotions, social status, family constraints, or stimulants. Whereas in the c ase of collaborative learning, the feedback mechanism tries to keep the learning system stable and ongoing.

Learning transpires through distributed cognition or through redistribution of knowledge. In such an approach, one idea merges with other ideas to create a third idea. New knowledge is connected to the cultural artifacts of a situation, artifacts that include tools and people. When artifacts are shared with a learner at an appropriate learning moment, the artifacts assist the learner in gaining new knowledge. The learner is not able to obtain new knowledge without the assistance of the artifact at a teachable moment. The system theoretic view of "the whole is greater than the sum of the parts" implies that the learner can gain new knowledge by sharing ideas with people and tools, like computers. The culture of the environment becomes collaborative when a common effort is required to complete a task. The zone of proximal development is the region of activity that learners can navigate with aid from a supporting context. Technology allows the zone of proximal development to increase the depth of learning. It c an be a facilitator of learning because tools like computers hold meaning. The distributed knowledge inherent in computers allows the learner to enter the teachable moment by having usable information or funds of knowledge available to manipulate as needed. We intend to develop such a system where learning activities with the artifacts and with people would lead to cognitive development. In addition, the system with its existing question bank, automatic search, and assessment techniques, along with the help from the participants with regard to peer-assessment and self-assessment would provide the learners with a measure of the degree of change in different levels of cognitive domain of learning. The training in the analysis of oneself and the peer would help the partners to develop their metacognitive and reflective skills as well.

CONCLUSION AND FUTURE WORK

In this article, we tried to visualize the possibility of collaborative innovation for training professionals for lifelong learning. This is the ideational scaffolding for designing and developing an intelligent system for collaborative innovation. Theoretically, we gave justifications for the conceptual design of the proposed system. This stage of conceptual design is necessary before we arrive at a more instructional and implementation stage. Development of a system requires a detailed instructional design which, in turn, is based on the conceptual design. Implementation and experimentation come next, which is then followed by evaluation. We have presented and discussed a framework for the development of an intelligent system for lifelong learning through collaborative innovation. This framework needs a lot more brainstorming before we could state something about its effectiveness in quantitative terms. Introduction of fuzzy logic for the assessment of cognitive development is new, as there is no alternati ve method available for assessment of cognitive development. Assessment of cognitive development is necessary for providing feedback to continue the learning process through enhancement of motivation and drive for learning.

In order to develop a prototype of the system, we initially plan to design and implement a reduced version of the system. As described in the proposed framework, the supporting resource data necessary for each subject and topic considered is quite large and requires experts from a large number of fields. Thus, implementation of a reduced set of knowledge domains of a selected number of fields is expected to provide a test platform and assess the effectiveness of the system before we leap to the development of the complete system.

Acknowledgements

This contribution has gained substantially from discussion with Tanmoy Bhattachaya, University College of London. We want to thank Richiro Mizoguchi, Piet Kommers, and anonymous peer reviewers for their valuable suggestions and comments on earlier version of this paper.

References

Bhattacharya, M. (1994). Effectiveness of faculty development programme for science education. Unpublished doctoral dissertation, University of Delhi, Delhi.

Bertalanffy, L. von (1968). General system theory: Foundation, development, applications. New York: George Braziller.

Bloom, B.S., Englehart, M.D., Furst, E.J., Hill, W.H., & Krathwohl, D.R. (1956). A taxonomy of educational objectives: Handbook 1. The cognitive domain. New York: McKay.

Brown, A.L., & Palincsar, A.S. (1989). Guided, cooperative learning and individual knowledge acquisition. In L.B. Resnick (Eds.), Knowing, Learning, and Instruction (pp.393-452). NJ: Lawrence Erlbaum.

Chen, H., Schuffels, C., & Orwig, R. (1996). Internet categorization and search: A machine learning approach. Journal of Visual Communication and Image Representation. 7, 88-102. [Online]. Available: http://ai.bpa.arizona.edu/papers/som95/som95.html

Chi, M.T.H., de Leeuw, N., Chiu, M H., & La Vancher, C. (1994). Eliciting self-explanations improves understanding. Cognitive Science, 18, 439-477.

Findlay, C.S., & Lumsden, C.J. (1988). The creative mind: Toward an evolutionary theory of discovery and innovation. Journal of Social and Biological Structures, 11, 3-55.

Jarvis, P., & Quick, N. (1995). Innovation in engineering education: The "PAMS" project. Studies in Higher Education, 20, 173-185.

Klir, G.J., & Folger, T.A. (1988). Fuzzy sets, uncertainty, and information. NJ: Prentice Hall.

Martindale, C. (1981). Cognition and consciousness. Dorsey.

Maslow, A.H. (1954). Motivation and personality. Berlin: Harper & Row.

Mogee, M.E. (1993). Educating innovation managers: Strategic issues for business and higher education. IEEE Trans. Eng. Manag., 40, 410-417.

Partridge, D., & Rowe, J. (1994). Computers and creativity (1st ed.). London: Intellect Books.

Ruben B.D., & Kim, J.Y. (Eds.) (1975). General systems theory and human communication. NJ: Hayden Book.

Rutherford, D. (1992). Appraisal in action: A case study of innovation and leadership. Studies in Higher Education, 17, 201-210.

Salomon, G. (1993). No distributions without individuals' cognition: A dynamic interactional view. In G. Salomon (Eds.), Distributed cognitions: Psychological and educational considerations (pp. 111-138). New York: Cambridge University.

Salomon, G., Perkins, D.N., & Globerson, T. (1991). Partners in cognition: Extending human intelligence with intelligent technologies. Educational Research, 20, 10-16.

Steiner, C.J. (1998). Educating the innovation and management: The engineering educators' dilemma. IEEE Transactions on Education, 41, 1-7.

Zadeh, L.A. (1975). The concept of linguistic variable and its application to approximate reasoning. Information Science, Part I: 8, 99-249.