In designing computer based educational systems our primary concern ought not to be with a dazzling new technology nor should we be misguided by such romantically unrealistic goals and expectations as replacing teachers, textbooks, or even the physical and social learning activities of students through learner machine interactions.
Instead the main object in the design of computational media as a new form of “intellectual bootstrapping” ought to be its functional connection to a pedagogical and didactical philosophy. The design must take into account the proper use and integration of the system into the comprehensive range of learning and teaching activities.
Educational software should make sense from a pedagogical point of view. Hence, four crucial considerations should govern the design of machine supported contexts:
(a) A cognitive and instructionally efficient model of the task or the domain the system is designed
(b) A sound conception of the general and content specific learning processes associated with the domain
(c) A domain appropriate social cognitive concept of teaching balancing dimensions such as explicit instruction versus discovery learning or solo learning versus collaborative learning
(d) A view of the active nature of the learner.
Often though cognitive researchers analyse meaning structures and processes on a conceptual level, using formats that are neither translatable into instructionally efficient models of domain and tasks nor allow inference to any normative principles of instruction. On the other hand designers of textbooks and computational media as well as teachers are often not successful in performing microstructural cognitive task analysis. Yet such analysis would be beneficial in uncovering the properties of the representational and operative “tacit” knowledge inherent in the performance of a task.
This leads to following viewpoints:
1.Design and use computer based tools pedagogically that is as cognitive instructional tools for mindful for teachers and learners in a culture of problem solving.
2.Extend and empower the minds of intentional learners.
3.Provide learners with some guidance according to the “principle of minimal help”.
4.Have students construct and externalize their mental models.
5.Provide students with intelligible and effective representational tools of thought and of communication.
6.Promote the use of comprehension related strategies. Together with representational formats, general and domain specific strategies are the cognitive tools of thinking and problem solving.
7.Encourage reflective and self-directed learning.
8.Extend the use of computer based instructional tools into a supportive classroom culture of collaborative learning.
The catchword “intelligent tutoring systems” (ITS) has come to mean that a computer functions as an intelligent, dynamically adaptive substitute for human teacher, who is capable of performing sensitive cognitive diagnoses, which means to infer, on the basis of a constantly retuned student model, a person’s cognitive states – what the person knows. How she/he thinks and learns – on the basis of her interaction with the system.
There are good reason to be skeptical about the feasibility – and in part even the desirability – of intelligent systems that are based on full system control and deep student modelling. ITS in which a machine tailors its instruction to an individual student on the basis of an inferred, constantly updated, fine grained mental model, may be seen as a long term goal. But given the current state of the art, machine tutoring based on cognitive simulation of the student is not possible across a full range of open ended tasks and domains, where fuzzy language and qualitative world knowledge based reasoning are required.
Making errors or getting stuck is an inevitable part of learning. Since learners can become highly confused and demoralized by undetected errors some feedback must be provided – either immediate, delayed or on request. An intelligent tutoring system would leave it to student, to use or seek help or feedback from the system as per need. Feedback during problem solving makes it possible to determine when and how the observed knowledge construction activity of a particular student deviates.
ITS as a tool for learning through reflection – the unique power of the system to keep track of the actions used to carry out a task, to display thinking paths and to allow students to focus and reflect on the why’s and how’s of their own problem solving – all at their individual pace and learning path.
One should therefore not conceive intelligent systems in education primarily as substitute for intelligent teachers but as tools aimed at cultivating the intelligence of the user i.e. student as didactic mechanism directed to the greatest possible extent, at fostering learner autonomy and self-regulation.
- Anderson, J.A., Boyle, C.F., et al (1984). Cognitive principles in the design of computer tutors. Proceedings of the sixth annual conference of the Cognitive Science Society. Colorado.
- Brown, A.L. & Palincsar, A.S. (1989). Guided, cooperative learning and individual knowledge acquisition.
- Bruner, J (1961). Actual minds possible worlds. Cambridge, M A: Harvard University Press.
- Bruner, J.S., Olver, R. R., & Greenfield, P.M. (1966). Studies in cognitive growth, New York: Wiley.
- Clancey, W. (1989). Situated cognition and intelligent tutoring system. In H. Maurer (Ed.), Computer assisted learning. New York: Springs
- Collins, A., & Brown, J.S. (1988). The computer as a tool for learning through reflection. In H.Mandl & A. Lesgold (Eds.), Learning issue for intelligent tutoring systems. New York: Springs
- Glaser, R. (1987). The study of cognition and instructional design: Mutual nurturance, Behavioral and Brain Sciences, 10, 483-84.