Co-learning

From DFM Wiki
Revision as of 13:19, 18 April 2022 by EricThrift (talk | contribs) (Uploaded some notes for the e-book synthesis chapter.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

We have approached the preparation of this e-book as a co-learning activity.

What is co-learning?

What is “co-learning”? The term is used in a variety of different ways in the academic literature.

  • Governance: egalitarian problem-solving process across heterogeneous/asymmetric stakeholder groups, including urban design [1]
  • Disciplinary knowledge generation: classroom learning as furthering academic knowledge (e.g., joint reflection and discussion on texts in a seminar) [2]
  • Collaborative learning (among classmates) as pedagogy [3]
  • Experiential learning (in the community) as pedagogy [4] - e.g., entrepreneurship work that involves the educators and community partners)
  • Indigenous research as reconciliation; two-eyed seeing [5]
  • Mutual training of algorithms in machine learning: cyclic co-learning, democratic co-learning, etc. [6]
  • Applied, procedural learning among members of a team (“community of practice”, “situated learning”), which may be goal-driven or institutionalized
  • Informal learning within a community forum: Twitter/social networks [7]
  • Synthesis or dialogue between different narratives [8]

What do all these forms have in common? The learning process itself is social/interactive – not just individual learning in/about a social environment

  1. Shared purpose: defining a problem or solution, understanding one another, achieving consensus…
  2. Participants are individuals with their own goals, knowledge, skill levels, and approaches; it is a heterogeneous group rather than homogeneous collectivity
  3. Participants have something they want to learn from others (and from the shared activity)
  4. Participants have information and perspectives to share with the others

We can see these points at operation in the DFM project. The Partnership Grants scheme is intended to foster mutual co-operation and sharing of intellectual leadership, but also has an emphasis on the formalization of partnerships in which collaborative learning can occur across different institutions.

TABLE: Four examples of co-learning in DFM
Experience Interactivity (shared goal) Individual learning goals Individual sharing goals
Dried fish stories video Meetings to discuss the storyboard; discussion of significance; notes Learn why dried fish matters in other countries across the region Communicate research findings; share value of dried fish in own country
Tastes and smells Storytelling on Zoom Compare own experience to others’; understand what governs taste for dried fish, what is common and what is unique Share personal experiences about taste and smell
Computer-assisted research Participation in a literature survey project; interaction via Zotero, Word comments, etc. (online annotations) Learn which dried fish literature gaps / knowledge gaps can be addressed by own research Contribute tools (Eric); contribute domain-specific expertise
Researching the researchers Project planning meetings and communications; reflection on the project Find better ways of receiving and communicating information that work for everyone Communicate problem areas

Discussion draft

Co-learning a term that was used by TBTI as part of the transdisciplinary co-production framework, defined by Polk as a form of problem-based learning:

TD co-production is a research approach targeting real life problem solving. Knowledge is co-produced through the combination of scientific perspectives with other types of relevant perspectives and experience from real world practice including policy-making, administration, business and community life. Co-production occurs through practitioners and researchers participating in the entire knowledge production process including joint problem formulation, knowledge generation, application in both scientific and real world contexts, and mutual quality control of scientific rigor, social robustness and effectiveness.

We also have the idea of “co-learning” in an Indigenous context, two-eyed seeing.

The implication is that we are seeking collaboration not just within a group of equal colleagues, but among people who come from different communities and possibly very different standpoints.

There is an implication of using co-learning to overcome inequalities. From that perspective, co-learning signals a departure from the model of “participation” in a study or research project initiated by a controlling actor, toward a more collaborative process in which participants are equal partners. It can also imply hybrid learning methods, that incorporate elements familiar to different participating communities – bringing together academic analysis and storytelling, for example – so that all groups are put on an even footing. This also implies that the actors have something to learn from one another: it is not a directed form of learning as teaching, nor an extractive mining of knowledge as data collection, but a teaching one another and sharing knowledge with one another and learning to work within one another’s knowledge protocols.

We similarly have some descriptions of “co-learning” within university settings, associated with calls for two-way or multidirectional flows of knowledge between the institution and its community setting, or between teaching and research. For example, geographers le Heron et al. define co-learning as:

coordinated and targeted approaches to maximizing the synergetic relationships between research and teaching such that their symbiotic development capitalizes on prior learning and experiences of all involved and feeds back positively on the nature and quality of both research and teaching environments.

Their analysis is motivated by the argument that co-learning is just as valuable as academic research in furthering knowledge production and the future of the discipline.

We also have a parallel use of the term co-learning in machine learning applications, where different algorithms work together to classify a dataset and are trained by adapting to the results of the other algorithms.