RuleWorks

A Framework for Model-Based Adaptive Training

Types of Learning

ITSIE types of learning are based on the mapping between available knowledge and goal expertise. The knowledge categories are:

  • Procedural Knowledge (K1);
  • Associative Knowledge (K2); and,
  • Principled Knowledge (K3).

These types of knowledge categories are a central component of the training system architecture (see Figure 1-1). Each of the three distinct types of knowledge models are presented in greater detail in the MOBAT framework. As shown in the previous section, the goal expertise categories are:

  • Skill-based Expertise (E1);
  • Rule-based Expertise (E2); and,
  • Model-based Expertise (E3).

The ITSIE types of learning are determined from the following Knowledge and Expertise pair sets:

  1. Inductive Learning mode => K1 & E2, K2 & E3;
  2. Rote Learning mode => K1 & E1, K2 & E2, K3 & E3 ; and,
  3. Deductive Learning mode => K2 & E1, K3 & E2.

See Figure 3-2 Types of Learning Styles (Learning Modes).

Learning Types

Figure 3-2 Types of Learning Styles (Learning Modes)

These three learning styles promote different types of learning. If knowledge is taught at the level above the required level of performance, then the trainee must deduce the relevant knowledge for performance, through a process of deductive learning. For example, a trainee may be taught the relevant principles (K3), but be expected to perform at the rule-based level (E2); this requires the trainee to reason deductively in order to generate the appropriate rules to perform the task. The presentation of knowledge, starting with the main points and leading to detailed examples and illustrations, is a typical deductive training sequence. This deductive sequence has the advantage of promoting a gradual understanding that can be built on in future learning scenarios.

A general (principled) representation might not be available. Perhaps only examples of knowledge at a level below the required trainee performance is available. In this case an inductive training sequence can be used in which the trainee is encouraged to induce more general relationships from examples. For example, a trainee may be taught the relevant procedures (K1), but be expected to perform at the (more general) rule-based level (E2), this requires the trainee to reason inductively in order to understand the appropriate rules to perform the task.

The level of knowledge required for a given level of performance may also be directly taught at the same level which is called the rote learning mode. For example, a trainee may be provided with an operating procedure (K1) and through memorisation be expected to recall the procedure and perform at the skill-based level (E1).

The classification of learning styles provides a principled approach to mapping expected trainee goal expertise level to the representation of domain knowledge.
*
Flower Show
  *  
© RuleWorks.co.uk | | Sitemap