The role of metacognitive skills in learning to solve problems

16
learners should be able to articulate their knowledge which directly supports
understanding of knowledge and conceptual change.
The self-regulation principle requires students in a constructivist learning
environment to regulate their learning, they should be able to actively
monitor and control their learning behaviour. This use of metacognitive
skills is directly related to transfer of knowledge. The possibility
of monitoring one’s progress while solving a particular problem in a
particular learning environment, for instance, is a type of regulation of
learning that is strongly advocated.
The reflection principle consists of the ability to reflect on learning
processes from an abstract viewpoint in order to gain insight into what
went wrong and what went well. Reflection is the overarching principle
over metacognitive skills. A specific debriefing or reflection phase during
the learning process would support this principle.
In conclusion, learners should gain experience with the material to be
learned in order to incorporate it in their existing knowledge structures.
Authentic and realistic learning material containing ill-defined problems
is deemed important in this context. Also, learners are responsible for
their own learning process. It is believed that learners who are capable
of viewing their learning from a broader, higher standpoint - learners
that are able to reflect on what they are doing - are better learners than
those who cannot or do not. Constructivist learning environments are
thought to initiate meaningful learning and the use of metacognitive
skills.
2.3.2 Metacognition in collaborative settings
Since collaborative learning is a key feature of constructivism it is of
importance to review the role of metacognitive skills in such a setting.
The use of metacognitive skills in collaborative learning settings is a
relatively unexplored area. A perspective from the field of AI is given by
Hoppe and Ploetzner (1999). They describe a computational simulation
model for the collaborative problem solver. A metacognitive level is
explicitly part of their model. Figure 2.2 depicts their model.
The domain model contains domain-specific knowledge about classical
mechanics problems. When a problem solver cannot solve a problem
at the cognitive level, the metacognitive level detects the impasse. It
identifies where an impasse is reached and which information is needed
to overcome the impasse. The communication interpreter reformulates
the impasse into a question that is posed to the other problem solver.
The other problem solver sees the questions as new problems and tries
to solve them. This model suggests that monitoring and controlling the
learning process is an essential characteristic of collaborative learning.