AERA: TUesday 1

Richard Lesh, Indiana University "Complex Systems Overview"

• • Almost everything we study is a complex system (an ill-defined term covering a broad range of stuff).
• • The Tao that can be told is not the eternal Tao
• • The action is in the interaction.
• • Feedback cycles
• • (Double Pendulum example -- if you control for the inital order, you will not be controlling it
• • in cooking, you take fresh things and taste them. Don't just follow rules.
• • I like this guy!!!
• • A unit is an undefined term. It is a property of a system, not a property of a piece of a system.
• • Virutally everything we want to describe are systems
• • What we're interested in are emergent properties, not initial properties.

takeaway: Emergence!!! Yes!!

Michael Jacobson, Nanyang Institute of Education (singapore) "Complex systems in Education: A universal Acid"

• • Why complex systems in education? An emerging science that integrate ideas and methods from many disclipines.
• • Integrarive conceptual perspectives, systems evolve, self-organization., interactions with each other
• • Traditional subject areas are not about complex systems -- just in higher level graduate education.
• • 20-30 year gap before it gets into mainstream system. How do we shorten it?
• • (kids can and will understand them, but their parents resist it)
• • not very good at conveying his ideas and points
• • there's a bias toward "top-down" explanations, and against "bottom-up" ones
• • Darwin's Dangerous Idea by Daniel Dennett
• • http://mjjacobson.net

takeaway: Wow! He tried way too hard to get WAY too much in this slideshow -- mostly just a lot of names, not much else effectively conveyed.

Uri Wilensky, Northwestern University
http://ccl.northwestern.edu

• • The time has come to have a widespread adaption of complex-systems persepctives, through agent-based modeling to reformulate school content, as an experimental methodss for evaluating policy, and as a method to build and assess learning theory
• • Instead of finding better ways to understand roman numerals, shifting to arabic numbers (a different form of representation) offers a significant advange
• • Structurations and restructurations
• • Agent based models built from computer software, each modeling their own state and ways of being (role-playing -- the stuff of Katie Clinton's work with dolphins, kitties, and the King of Persia) -- simulations
• • Also very similar to what Will Wright is doing in Spore -- where you can evolve your character and set them free to interact with their environment with the affordances that they have.
• • Educational Policy agent-based model (on website)

takeaway: good points, well-conveyed. We need to change the foundations of our system -- but is there a way to do it without causing too much chaos? (methinks: Start by teaching that chaos and uncertainty are *good* things!!!!)

Cindy Hmelo-Silver and Roger Azevedo

• • Complex systems: hierarchical nature, heterogeneous components, emergent or causal -- depending on one's perspective.
• • Problem is that people only tend to think about what they can see, and many of these systems are both counter-intuitive or hidden, or that they operate at frequencies that we don't or can't observe, makes them hard to address.
• • SRL-- self-regulated learning
• • To learn about circulatory system takes: Classroom 2-3 weeks; lab 40 minutes;
• • Lack prior knowledge, or cognitive skills, or metacognitive skills,. or skills oro sustain motivational level, or skills to assess dynamic components of the task and learning context
• • Lack certain scientific reasoning skills
• • Using CBLEs necessitates scaffolding

takeaway: A good argument for SRL, and I'm thinking "via video games" here...

Jay Lemke, U Michigan Nora Sabelli, SRI International

• • Key concepts from CSTheory: emergent patterning, self-organization, dynamic attractors and repellers, multi-scale hierarchical organization, self-regulation, coupling matrix, auto and cross-analysis, information flows and constraints, information generation, system-environment interactions, developmental...
• • Get a very small USB web cam to take pics of slides! Cool idea in use by a guy in front of me!!
• • Causal loops (non-linear networks) vs. causal chains produce new info
• • Integrate systems vs. isolable units of analysis
• • Dynamic models and simulations vs. input-output modeling
• • Unique vs. generic systems
• • Emergence vs. determinacy
• • Tinkering near the threshold of collapse (the Active Adaptive Management paradox) (systems are always near collapse and are not nearly as robust as we often think they are)
• • Re-engineering whole systems vs. iterative reform
• • Support vs. control: reversing accountability
• • Interdependence with other systems
• • Emergence always happens in the middle -- between possibilities and constraints
• Nora
• • How is learning organized? (cool graphic of four-legged stool of sytem options)

1. o What people learn - content
2. o Which people learn what - equity
3. o What people learn - cognition
4. o What people learn - context

takeaway: messing about could be dangerous because "systems are always near collapse and are not nearly as robust as we often think they are" -- but then isn't just about anything worth doing somewhat dangerous?

James Bransford (??)

• • We are arguably built to understand causal chain reasoning
• (look for the person /thing who did it -- if you can't find it, make up a supernatural entity)
• • Intelligent Design and the belief of Irreducible complexity is the result of a failure to understand complex systems and emergence.
• • He says we need a developmental system for teaching complex systems. if we try to jump too far ahhead, it results in a degraded learning
• • He calls for knowledge-building

takeaway: I should read more of him...

John (??) Summary

• • LIFE - Learning in Informal and Formal Environments http://life-slc.org
• • Things that Make us Smart by Donald Norman
• • Doesa the topic make me think different? (better?)
• • 24 rabbits introduced to a farm in Australia, resulting in the great rabbit infestation (Rabbit-proof Fence)
• • would agent-based modeling have saved the day? yes
• • Redesigning education (cool chart)
• • What brings out the best in humans, and why? -- disasters.
• • General issues; models for understanding nature; understanding human design

takeaway: tongue-in-cheek and effective.