Student Modeling for Personalized Education: A Review of the Literature - Advances in Personalized Web-Based Education - page 16

Table 1.2 Student modeling approaches in relation to student's errors and misconceptions
Perturbation Constraint-

based model

Stereotypes Cognitive

theories

Fuzzy
techniques

Bayesian
networks

35.71 % 21.43 % 14.29 % 14.29 % 7.14 % 21.43 %

Errors/
misconceptions

1.3.3 Cognitive Features Other Than Knowledge Level

Student's cognitive features are among the most sophisticated student character-
istics that are described in a student model. These features refer to aspects such
as attention, ability to learn and understand, memory, perception, concentration,
collaborative skills, abilities to solve problems and making decisions, analyzing
abilities, critical thinking, learning style and preferences.
Learning style refers to individual skills and preferences that affect how a stu-
dent perceives, gathers and processes learning materials (Jonassen and Grabowski
1993). Some learners prefer graphical representations, others prefer audio materials
and others prefer text representation of the learning material, some students prefer
to work in groups and others learn better alone Popescu (2009). Adapting courses
to the learning preferences of the students has a positive effect on the learning pro-
cess, leading to an increased efficiency, effectiveness and/or learner satisfaction
(Popescu et al 2010). A proposal for modeling learning styles, which are adopted
by many ITSs, is the Felder-Silverman learning style (FSLSM). FSLSM classifies
students in four dimensions: active/reflective, sensing/intuitive, visual/verbal, and
sequential/global (Felder and Silverman 1988; Felder and Soloman 2003). Another
method for modeling learning styles is the Myers-Briggs Type Indicator (MBTI)
(Bishop and Wheeler 1994), which identifies the following eight categories of
learning styles: extrovert, introvert, sensing, intuitive, thinking, feeling, judging,
perceiving.
Many researchers have modeled the student's learning style and preferences.
Most of them have been used stereotypes for modeling the particular cogni-
tive features. For example, the stereotypes of the student model of INSPIRE
(Grigoriadou et al. 2002; Papanikolaou et al. 2003) provides information about
the learning style of the learner. Furthermore, Surjono and Maltby (2003) have
used stereotypes to model the student's preferences (i.e. font, colour, illustration)
and learning styles (i.e. competitive, collaborative, avoidant, participant, depend-
ent, independent). Also, Glushkova (2008) has modeled the student's preferences,
habits and behaviors during the learning process by using stereotypes. Moreover,
Carmona et al. (2008) have used a student model that classifies students in four
stereotypes according to their learning styles. In WELSA (Popescu et al. 2009)
the courses are adapted to the learning preferences of each student applying
stereotyping.
In addition, Salim and Haron (2006) used a combination of fuzzy logic
with a stereotype-like mechanism to model the student's personality factor
MBTI (Myers-Briggs Type Indicator). AHA! (Stash et al. 2006) and TANGOW