Modelling Integrated Learning Scripts

Notes from Pierre and Patrick’s article on modelling integrated learning scripts

Each phase specifies:
- the task that students have to perform
- the composition of the group
- the way that the task is distributed within and among groups
- the mode of interaction
- the timing of the phase

The effectiveness of collaborative learning depends upon:
- group composition (size, age, gender, hetegogeneity, …)
- the task features
- communication media

Complexity make it not possible to guarantee learning effects. Scholars attempt to directly influence the interactions by:
- augmenting the frequency of conflicts
- fostering elaborated explanations
- supporting mutual understanding

-> scripts are ways for enhancing the probability that productive interactions occure during collaborative leanring

Integrated learning
Group activities gain from being integrated with other classroom activities. The notion of integration is multidimensional
- spatial integration refers to scripts combining activities, both face-to-face and computer-mediated, that occur in a variety of places
- pedagogical integration (socio-constructivism)
- functional integration (data flow between multiple activities)
- teacher integration, the teacher remains the script driver (even if the role changes)

The syntax proposed in 2002 does not convey the spirit of collaborative learning. The challenge is not simply describing a sequence of events but formalizing the core mechanisms by which the sequence has been constructed, why it is expected to generate learning. IMS-LD could be expended, but currently it is more at the descriptive level than at the modeling level.
(groups are not defined explicitly but inderectly by assigning roles. This prevents for instance in building a Jigsaw script where team members have different
roles within each team, or where roles could rotate among group members.
The social sctructure of a script should be explicitly modeled

- A script materializes a distributed system (a script defines how the task is distributed among the different system componenent (namely people and tools)
- the cognitive distributed system is arbitararily defined by the task (->the task defines the reference plane or social plane zero)
- learning results from over-compensating the drawbacks of task distribution

Intra-psychological plane (individual)
Inter-psychological plane (group as long as team members maintain some representation of their teammates cognition)
social plane (individual representation disappear behind the culture that they jointly constructed

in SWISH
- individual, group, class, community, world planes

Functional integration of scripts refers to the dataflow activities at different planes.
Dataflow could be described by a small number of operators from moving up and down the plane structures, whatever the level is
Upward operators: aggregate, list, differentiate, …
Downward operators

A major challenge for the design of an abstract scripting language is defining of generic operators, independtly from the data structure to be handled

A distributed system activity is defined by 3 dimensions:
- the task structure
- the time structure
- the social structure