Tuesday, July 10, 2012

A review of 20 years of research in the Learning Sciences

Janet Kolodner
Below are my notes from the keynote talk given at the 10th bi-annual international conference of the Learning Sciences by Janet Kolodner, one of the founders of the field of Learning Sciences.

The Learning Sciences started in the late 1980s with the foundation of the "Journal of the Learning Sciences" (JLS). The Learning Sciences (LS) are a multi-disciplinary field that aims to develop models of learning in real-world situations. LS combines constructs and methods from a variety of fields, such as cognitive science, computer science, educational psychology, anthropology, science education research, and linguistics. Findings from the Learning Sciences appeal to multiple audiences and aim to have an impact on education.

Basic observations:
-In the 21st century, everyone needs to continue learning to fully participate in the workforce and citizenry (because available tools are changing fast).
-Education may be the way to close the equity gap.
-Schools (in the dominant industrial-model as we know them) can't do it all.

After 20 years of Learning Sciences research, what do we know about learning?
1) Learning deeply (as a process of mental model building) requires sustained and long-term effort and requires a lot of help (facilitation, scaffolding): Helping learners build mental models.
a) From a cognitive perspective, learning is a process of iteratively constructing, revising, and connecting concepts (mental models, ideas, etc.): Factual knowledge and procedural knowledge.
-Becoming fluid at reasoning skills is an iterative process of composing and debugging sequences of procedural knowledge ("How-to") [Schank & Abelson; Anderson; Newell, Rosenbloom & Laird]
b) We can only learn on the edges of what we already know (Zone of proximal development) [Vygotsky]
-We realise a need for learning if we can't do something we want to do, something turns out differently than expected, or something happens that we can't explain [Schank & Abelson; Anderson]
c) Revising one's mental models requires a great deal of reflection and interpretation [Bransford; Collins; Schank]
-Framing first, details later [Collins]
-Repeated deliberate practice: Help learners to repeatedly experience the result of their decisions and interpret and reflect on those results to debug their reasoning.
-Promote asking questions and wonder to identify holes in one's mental models.
-Tell only when learners are ready (A time for telling); Give learners a role in telling their peers.
-Give learners learning goals that are relevant to them to put in the mental energy to revise mental models.
-Help learners to connect ideas (mental models) by helping them recognise and revisit prior ideas [Linn]

2) Promoting and sustaining active engagement is key to learning: Intrinsic motivation
a) Personal and epistemological connections promote engagement [Papert; Resnick]
-Learners connect when they can connect to prior knowledge.
-Learners connect when they are interested, e.g. by using authentic real-world settings such as real audiences.
b) "Communities of Practice" [Lave, Wenger]
-Encouraging community (affinity groups, interest groups, communities of practice) encourages engagement.
-Community can be created in formal and informal learning environments by creating activities that promote shared values and ways of doing [Kolodner]
-Community engagement gives learners opportunities to imagine what they might be.
c) "Thick authenticity" for maintaining integrity [Schaffer and Resnick]
-Solving authentic problems relevant to the world learners live in by having available the same kinds of tools and resources professionals would have
-Big questions or challenges promote personal connection [Krajcik; Edelson; Reiser; Kolodner]
-Assessment should be authentic, embedded and meaningful to learners.
d) Turning over agency to learners allows learners to decide for themselves to participate [Holland]\
-Knowledge building with a community results in collective cognitive responsibility, mental model building and knowledge integration [Scardamalia; Bereiter; Linn].
e) Provide scaffolds that promote success (including productive failure) and self-assessment

3) The one-size-fits-all approach does not work, as everyone's interests and Zone of Proximal Development (ZPD) are different (This also applies to teachers as learners).

A vision for learning environments of the future:
-Less emphasis on "lessons" and more emphasis on "challenges": Lessons are about doing something, challenges are about achieving something.
-Lectures will be for purposes identified by learners. Lectures will be often short, interactive, and impromptu. Lectures should be recorded to allow revisiting.
-Reading will be for purposes identified by learners to answer questions they have raised.
-Classrooms will be places where students and teachers address challenges together. Solving design challenges provide rich learning opportunities. Challenges can be of local relevance. Challenges can be done in groups (both within class and external).
-Learning will be purposeful: Skills and practices for living a life; for stewarding a planet; for health and well-being; for someday joining a workforce; for actively participating in civic life.
-Assessment will be purposeful: Assessment not only for accountability (e.g. current standardized assessments) but for the purpose of scaffolding and promoting self-monitoring.
-Learning technologies will be integrated in purposeful ways, e.g. models (show hidden mechanisms); simulations (allow to understand what happens when); communications to connect with communities, resources, and experts; videos; visual interfaces to access and analyse data; sharing data with real audiences; embedded assessments. Needed: Technologies should not be stand-alone but integrated across disciplines, grades and time (learning trajectories); used for learning and assessment; used by teachers and students; used in and out of school. Learning often happens during the interaction between community members while using technology - make that part of the design.

Some good examples:
-Students learn about chemistry through the challenge of improving the air quality in their community.
-Students learn about the process of science by improving cooking recipes.

How can Learning Sciences research gain more influence?
1) Practice what we preach! e.g. in teacher education courses, curriculum design
2) Building on each others' work to focus on integration: Work on coordinated projects. Integrate findings.
3) Shift towards implementing research in the community: Work with interpreters and integrators (who take the best of our ideas and products and hold them into integrated platforms and packages). Not every researcher has to take research to practice, but all should be willing to make their findings and products available.
4) In order to successfully implement technology in the classroom, teachers need to know how to use it.
5) We need to evaluate under what circumstances technology and instructional strategies are effective: First, identify why and under what circumstances technology works effectively in small trials (understand affordances and challenges); second, large controlled random trials; third, implement your innovative product (scale up, but not before rigorous testing). To make implementations successful, they need appropriate teacher professional development.
6) Work on what's important: Start by identifying challenges of national or international importance.
7) Learning science research should draw from at least three different literatures (cross- and inter-disciplinary), otherwise your work is not important enough.

1 comment: