Learning networks

Christensen’s 5th chapter proposes a valuable (but ultimately incorrect) three-part business model lens through which he proposes we consider education: consulting (services); value-chain (manufacturing); and user networks (black market). The parenthetical examples are mine: Christensen claims that telecommunications is a user network when in fact it’s a service (access to “wires” owned by a telco) as well as a value-chain (resale of bandwidth); consulting could also be viewed as experts providing a service within a user network rather than a distinct type. However, the metaphor of current public schools as a value-chain model is accurate, as is the view of special education as consultative and unscaleable one-to-one education.

The dismal evaluation of and outlook for textbooks is well-supported (although his terms are inaccurate: commercial systems are actually delivery mechanisms; “high fixed costs” are actually “sunk costs” because a business can have continuing high fixed costs whereas sunk costs such as the investment to create a book are one-time). His argument falls apart, though, in the claim tha,t “people will assemble them [learning kernels] together into entire courses.” If this were possible, libraries would have precluded the need for schools. Learners don’t know how to structure the learning they need because they don’t know the end goal. Learning opportunities or situations or problems must be constructed by experts, although not necessarily subject experts who often make unexplainable leaps in problem-solving.

The attempt to equate Web 2.0 technologies with the need for educational reform also falls short. QuickBase is not a replacement for SAPs’s ERP software; it’s an online service from a software company seeking to change its value-chain distribution model. Second Life is not a 3D world “‘created entirely by its residents;'” it’s a hosted software application whose creators charge real dollars for the service afforded by a virtual space. And finally, the idea that learners can self-educate smacks of self-medication and the potential for uninformed abuse. At the same time, the vision of public school education replaced by user networks guided by experts is enticing.


Christensen’s Disrupting Class is a good read–the stories seem real. The first chapter starts by revisiting Gardner’s multiple intelligences but adds a couple layers I missed before: inside each intelligence are different learning styles (VARK design types) and within each style are different paces (time on task). Good additions.

Designating only two types of interface (modular and interdependent) seems overly binary; interfaces are more like a continuum, and even if you can segment them, I think there are more than two (for example, a modular interface can be unpredictable when humans form part of the chain). I’m also not sure there are only 4 interdependencies (temporal, lateral, physical, and hierarchical) in public schools. However, the argument that we know we should provide customized education but cannot do so because (in part) of these interdependencies is compelling.


I especially appreciated this article as it clearly delineated between education and learning; while acknowledging the role of the latter in the former, the author emphasizes that education is “a process of changing the behavior patterns of people.” Although written 60 years ago, the distinction between “needs” as a gap between a current set of behaviors and a desirable norm (TAKS), and “needs” as an equilibrium state was eye-opening.

The author discusses three means for determining educational objectives:

  1. learners
  2. culture (contemporary life)
  3. subject matter experts

In a forward-looking manner, he advocates using culture as the primary basis for determining objectives. He also distinguishes between education for things that are important today versus things that are important for conditions that will be encountered in the future.

The discussion of subject matter experts could have been more prescriptive when the author asked the experts, “What can your subject contribute..?” The admonition from Wiggins to discover the central questions of a discipline was more useful.

The coverage of learning theory, constructed as a dichotomy between S-R behavioral learning and Thorndike’s “generalizable” learning, was just as useful (although perhaps not as research-based) as the tripartite theory models explored earlier; constructivism and cognition both fall into the generalizable category. However, the two-dimensional chart was the real contribution of the article; not only did it provide a specific application of the ideas advanced in the article, but it also made clear the definition of objectives covering both the “kind of behavior” and the “content (context) or area of life in which this behavior is to operate.”

Ten Core Insights–But Not Much New

While the 4-element principle–Learner, Mentor, Knowledge, and Environment–was a welcome reduction from 8+ part models, it seems to add nothing new (apologies to Judith), and LeMKE could just as easily have been KEEN (Knowledge, Environment, Expert, Novice).The idea that tools shape our learning was familiar but I’d hoped for some solid research into how new tools are changing how we learn (i.e., does multi-tasking produce better learning outcomes?).

The changing role of instructors (“all teaching functions no longer need to be embodied in one person”) was well-articulated. However, the most useful section was the discussion (although too brief) on research into the proliferation of receptor nodes caused by multiple new knowledge items (pattern-matching meets case-based learning?).

The ZPD discussion seemed to miss what I thought was the key point with Vygotsky (or maybe I’ve misread Vygotsky): the ZPD exists just beyond what a learner knows; the statement that students are outside their ZPD when they are totally lost rings hollow–the students may simply have no context to match against because the appropriate historically-relevant (historical in the individual sense) case has not been presented. Or, they could just be lost. I expected to see the idea that faculty are challenged to individualize instruction (to match ZPD) paired with the idea of community support.

A gem was Freeman’s description of “meaning as a process of successive approximations.” However, the denigration of learning the vocabulary of a discipline, while accurate in its caution against over-reliance on this method, ignores research into the value of declarative rules for novices.

Many aspects of the article were routine:

  • plan assessments simultaneously with instruction
  • learners bring their own history
  • concentrate on the core concepts

At the same time, the metaphor of content as an onion with the big idea at the center works effectively, especially in terms of the learning goal being a slice that drills to the core but includes outer contexts unique to each student. The final insight–that time on task produces more learning with all else being equal–is not supported although it seems to make common sense. What seems suspect is the statement that content chunking is, “one reason why games and role-playing scenarios are popular and valuable.” Games are popular and effective because they are fun. Maybe even addictive.

PBL – pure and simple

While PBL is the topic, the article actually focuses on PBL as an exemplar implementation of constructivism which proposes:

  • Understanding occurs in our interaction with the environment (distributed cognition).
  • Cognitive conflict (again, the Wiggins’ idea of misperception) is the learning stimulus and determines the organization of what is learned (this latter concept is never explained); ideas are tested against alternative views and contexts in a collaborative community of practice.
  • Knowledge evolves through social negotiation which evaluates the viability of individual understandings (the community determines if a particular answer is viable).

Constructivist strategies include collaboration, personal autonomy, generation , and reflection, all of which are embedded in the original PBL model created by Barrows.

GBS aka CBR aka PBL

While Schank’s work in scenario-based learning is well-known, this article expands his approach to encompass a broader design theory he calls goal-based scenarios and a learning theory he terms case-based reasoning; both seem derived from problem-based learning. CBR is postulated as the way in which experts solve problems and is essentially learning from prior experience via analogy: a case is a memory, and experts have large libraries upon which they draw. CBR enables reasoning across contexts, and while it seems obvious that experts organize their libraries through indexing (labeling and filing), little practical design advice is offered. The general suggestion to design roles and goals to create motivational and sensible contexts is tied to Schank’s narrative approach. More generally applicable are the ideas that goals produce expectations, and that expectation failures demand explanations; the necessity of failure as a primer for learning (reminiscent of Wiggins’ misperceptions) suggests building learning experiences with high probability of non-optimal solutions. The most useful aspect of the article came in the editorial comment (not from Schank) that teaching is the transition from learning theory to design theory.

The 7 components of GBS include:

  1. Goal – process knowledge or content knowledge learning goal
  2. Mission – performance goal provided for initial motivation
  3. Story – narrative for immersion and context leading to motivation
  4. Role
  5. Operations – activities with decision points leading to consequences
  6. Resources – well-organized stories to compare with cases in memory
  7. Feedback – consequences, coaching, and domain expert stories

Expertise: a long and winding road

The idea that experts tackle problems that increase their expertise seems supportive of the self-efficacy behind ACT-R:

  • Reinvestment – the  motivational aspect
    • conserving resources to have energy to put back into new problems
  • Progressive problem-solving – the cognitive aspect
    • tackling more difficult problems AND tackling more complex representations of recurrent problems
    • represents working at the edge of competence (ZPD)

Pattern learning, which occurs without extensive effort, involves choosing the right patterns and recognizing when no pattern fits. This seems to equate with imagaic memory which is efficient for spatial and temporal data.

Procedural learning starts as step-by-step problem solving which become “chunked” into a single procedure; while this automaticity frees resources, it becomes a handicap in the improvement of performance. However, automaticity does not inevitably lead to inflexibility if automated skills are building blocks to new skills that are not automated.

However, learning is also pattern and procedural learning; what distinguishes expertise is the seeking of complexity. Complexity is described as a matter of the number of constraints. Because most real-world problems are not reducible to a step-by-step economy, we use simplified representations (akin to simulations). The class of problems that are endlessly complex are the constitutive problems of a domain.

Experts are motivated by:

  • flow:  total absorption and feeling of control, loss of normal time; becomes addictive to the point that problems are invented
  • second-order environment: the expert sub-culture where your recognition as an expert matters to you (not useful in fostering early development of expertise)
  • heroism: effort disproportionate to rewards

Competitive environments foster expertise. However, so do expert sub-cultures which may not always involve competition but always involve recognition of success and help/cooperation leading to success. The expert environment constantly changes as the experts become more expert; the reason experts help each other is to help that environment conttinue to get more difficult (i.e., inventing problems). Expert teams are one example: everyone more or less knows how to do everything so the focus is on the goal, not on individual achievement.