The research
direction of educational field is going towards the teaching or development of
the cognitive skills or cognitive aspect of the subjects matter. This motivates
Intelligent Tutoring Systems (ITS) to progress in the
same direction. Current developments of ITS had successfully incorporated the
teaching of specific cognitive skills for example mathematical or science
analytical skills, programming and trouble-shooting skills. The research
direction that ITS is moving towards the generic cognitive skills. There are few
developments of ITS that develops generic cognitive skills. This research brings
ITS one step further in that direction with the proposal of an ITS shell that
develops critical reasoning skills. This research also examines the issues
involved in the development of cognitive skills development ITS and proposes
solutions to overcome these issues. The system that is developed by this
research is named as EpiList.
EpiList is an ITS that develops classification, generalisation and
comparison skills, as well as the reasoning skills involved in
classifying items. It is developed based on the List-making Game
proposed by Prof. Collins. It guides the student in generating lists of items in
a bottom-up environment. The tutorials of EpiList incorporate the teaching model
by Eggen and Kauchak. The models use sequences of generalisation and comparison
skills to highlight the mistakes made in the classification. Hence, EpiList
implicitly develops the generalisation and comparison skills while explicitly
develops the classification skill.
The formal definition methodology proposed by this research uses simple
sets and set operations in defining an ITS attempts to improve the
understandability and readability of ITS. This formal definition methodology is
used on EpiList. The domain and platform independent characteristic of the
formal definition make EpiList to become an ITS shell that is both domain and
platform independent.
One issue of ITS is the ad-hoc design and development of an ITS that makes
analysis of the system tedious and incomplete. On top of that, the conventional
approach for analysis may not be applicable to cognitive teaching ITS. This
research proposed a design and analytical methodology appropriate for
cognitive teaching ITS such as EpiList. This methodology is based on the
sound system theories. The architecture of EpiList is modeled into the
paradigm of system theory. This enables the analysis of EpiList in term of
system theories which presents the limitations of EpiList. Similarly, the
solution is also obtained based on theories and design of system.
EpiList I
EpiList I is the first version or prototype of the research. EpiList I guides the student to perform classification in a bottom-up learning environment. In the bottom-up environment, the items are first presented to the student and subsequently the student determines the categories to use to classify these items. The main interface (picture below) presents the categories the student selected and allows the student to perform their classification of the items.

The student click in the "Begin Checking" to activate the rules and algorithms to check the classification. The rules are grouped into intra-scheme and inter-scheme rules. The formal rules ensure the student classify the items correctly while the latter ensure the student obtain the correct classification scheme. The tutorials of the rules use sequences of generalisation and comparison to tutor the student. Hence, EpiList I implicitly develops the cognitive skills of the student while explicitly tutors the content of the system.
EpiList II
EpiList II is the second version of EpiList. The cognitive rules of EpiList II would explicitly monitors the usage of the generalisation and comparison skills in EpiList I. Whenever the system detects that the student is not able to perform the cognitive skills, EpiList II would explicitly teaches the student how to perform the cognitive skills. The teaching of cognitive skills in the tutorials of EpiList II is based on the teaching model "Cognitive Apprenticeship". According to the model, the tutorial of EpiList II explicitly guides and assists the student to perform the cognitive skill processes step by step. Subsequently, the tutorial would assess the student to ensure that they have understood the skill or more practice of the cognitive skills is necessary.

The picture above is the main interface of EpiList II. The interfaces of EpiList II are enhanced and improved with colours and icons to be more appealing to the children.
A-EpiList
A-EpiList is the short for Adaptive-EpiList. A-EpiList is the third version of the prototype EpiList. The tutorials and rules of EpiList I and EpiList II are independent of one another educating different aspect of the student. The adaptive rules of A-EpiList attempts to them up. A-EpiList uses information obtained from EpiList II, namely the competence of cognitive skills, to adapt the tutorials of EpiList I to individual student.
Technical Reports
Goh Ghee Ming, (2003), Student Evaluation of EpiList, Technical Report No. ISL-TR-01/03, Intelligent System Lab, School of Computer Engineering, Nanyang Technological University.
Conferences
Goh Ghee Ming, (2004), EpiList: An ITS Shell to Develop Generic Cognitive Skills for Bottom-up Knowledge Construction, International Conference on Computer in Education (ICCE2004).
Journals
Goh Ghee Ming, (2004), EpiList: An Intelligent Tutoring System Shell for Implicit Development of Generic Cognitive Skills that Support Bottom-up Knowledge Construction, under review by IEEE transaction in Systems, Man and Cybermatics.
The above information is copyright to Goh Ghee Ming, SCE, NTU, Singapore.
Date: November 2004