,
21 tweets,
8 min read
Read on Twitter
Reading through this now. Absolutely fascinating how set of ideas about how Cognitive Load Theory can predict when collaborative learning is beneficial. 
Collaborative learning is likely to be beneficial with the right types of dialogue, particularly peer critique, debate and seeking synthesis - this moves towards a shared mental model and helps distribute cognitive load
Effective collaborative learning requires a number of types of coordinating dialogue among group members
Of course, these coordinating conversations and tasks impose a cost. @P_A_Kirschner uses the term "transaction costs" here which seems very helpful.
For collaborative learning to be effective we need to reflect if the gain from sharing cognitive load between group members is outweighed by these transaction costs
Fascinatingly, when collaborative learning is pursued virtually instead of in person, the loss of in-person communicative cues (eg body language, intonation) may make group learning transaction costs even higher as we can often only rely on asynchronous communication via text
A key point in this piece by @P_A_Kirschner is that we've evolved to collaboratively learn biologically primary knowledge (e.g. through play, I guess?) but we've not evolved to collaboratively learn biologically secondary knowledge so it needs v careful thought & preparation
A key point here about the extraneous cognitive load from collaborative learning. We need learning design to reduce errors, conflicts and duplication (strikes me this is also true in individual learning? In fact task design & environment should perhaps reduce negative stress?)
This point about clarity in group members' roles reminds me of a point made by @dylanwiliam.
Collaborative learning could not only lead to more efficient learning but also could teach group members to learn better in groups in future.
Collaborative learning can lead to better shared understanding even if for some group members it's less efficient. The efficiency of collaborative learning is lower in groups not experienced in engaging in it.
Ooh, we’re on to reflections on designing collaborative tasks. *rubs hands together*
Careful collaborative learning design is even more important than instructional design for individual learning. Guidance needs be simple enough to process and could include e.g. worked examples and guidance on group roles
Ooh! Is @P_A_Kirschner suggesting that collaborative learning could have greater potential for novices than experts? That would be fascinating, given my understanding of previous literature suggesting that collaborative learning seems to suit experts better.
Four factors to consider when pondering creating a collaborative learning group.
The bigger the learning group, the greater the potential gains from shared working memory BUT the greater the transaction costs to coordinate the group AND the greater the risk of loafing, free-riding and failure. (Certainly this feels intuitively right!)
Team role clarity can reduce extraneous load (and presumably minimise social conflict?)
Aha, here we go, more nuance around that novice-expert angle. Novices benefit from the presence of experts. Experts tend not to benefit if they already have the knowledge to solve the problem on their own.
For really effective collaborative learning we ideally want group members to have existing collaborative experience and to be familiar with each other.
.@P_A_Kirschner summarises the potential of Collaborative Cognitive Load Theory.
@threadreaderapp unroll
