coherence paper about framing and transfer

April 13, 2007

I am continuing taking some notes for the refinement of my paper.

In general, it seems that the ideas which i had in writing my paper are quite similar with the ideas below, and i can refine some parts using their terminology.
I like the context dependent pattern of activation idea.

Resources, framing, and transfer
David Hammer, Andrew Elby, Rachel E. Scherr, Edward F. Redish
In J. Mestre (Ed.), Transfer of Learning: Research and Perspectives /working title/.
Greenwich, CT: Information Age Publishing.

We often focus on whether and how students apply what they know in one context to their reasoning in another. But we do not speak in terms of transfer. The term connotes to us a unitary view of knowledge as a thing that is acquired in one context and carried (or not) to another. We speak, rather, in terms of activating resources, a language with an explicitly manifold view of cognitive structure.

The literature on transfer

Barnett and Ceci (2002) have developed a taxonomy, listing phenomenological aspects of where and when to look for transfer, such as the knowledge domain, the physical setting, and the time.
Their taxonomy also distinguishes what kinds of knowledge might be transferred (procedures, representations, or principles) along with observable features of performance (speed, accuracy, approach).

In the transfer literature, discussions of cognitive structures and mechanism have focused on:
(1) the nature of the knowledge or skill hoped to be transferred and
(2) the role of metacognition or metacognitive scaffolding in supporting transfer.

The standard references cited in the literature pay little direct attention to the ontology of cognitive structure. What elements of cognitive structure do researchers attribute to the knowledge or ability they are investigating, to see whether it transfers?

The tacit, default stance, as Greeno (1997) identified, is to think of the knowledge or ability as a thing that an individual acquires in one context and may or may not bring to another. We describe this as a unitary ontology (Hammer, 2004b), thinking of the particular piece of knowledge as an intact cognitive unit, in close correspondence with the observable idea or behavior, be it a principle, fact, or procedure.

Cognitive science research has been pursuing an assortment of particular models (diSessa, 1993; Minsky, 1986; Thagard, 1989). Here we proceed from the most general assumptions about manifold structure and use the generic term resources to refer to components in that structure.

Our framework ascribes cognitive objects to individual minds, but at a finer grain-size than concepts or abilities as people experience them. In this view, knowledge and experience are emergent, analogous to other emergent phenomena in complex systems, in which the things we see—traffic jams, birds flocking, and so on—emerge from many small agents acting in local concert. In other words, we need to be alert to the tendency to thingify experience (Minsky, 1986; Wilensky and Resnick, 1999).

In a resources-based framework, we can view learning an idea not as the acquisition or formation of a cognitive object, but rather as a cognitive state the learner enters or forms at the moment, involving the activation of multiple resources.
Transfer would then be understood in terms of the learner entering or forming a similar state later in a different context.

The activation of finer-grained cognitive resources should often depend on the social and physical environment such that the resulting knowledge can coherently be attributed to the overall system (people + environment).

The difficulties of drawing boundaries around the concept of transfer stem from an ontology of cognitive structure that is both tacit and unitary.

Theoretical framework of resources and framing

We start with a review of this manifold ontology of mind, contrasting the notion of conceptual and epistemological resources with unitary views of (mis)conceptions and epistemological beliefs.
We then review research from linguistics and cognitive science on framing, which provides a complimentary perspective to research on transfer in that the core phenomena it describes are the context-dependent coherences of individual’s interpretations of social or natural phenomena.In our ontology of cognitive structure, framing corresponds to locally coherent activations of resources.

A view of manifold resources provides a more generative basis for curriculum and instruction focused on student knowledge and reasoning.

Research on misconceptions posits conceptions as cognitive units. The conception is the basic unit of cognitive structure, and an incorrect conception impedes progress toward expert
understanding. In some cases, attributing robust conceptions is appropriate, but resources-based accounts afford the alternative of understanding the conception as a local or momentary activation of another sort of cognitive structure.

diSessa’s accounts of phenomenological primitives (diSessa, 1993) and coordination classes (diSessa and Sherin, 1998), for example, attribute cognitive structures at other levels, as minigeneralizations from experience whose activation depends sensitively on context. DiSessa and Sherin (1998) discuss coordination classes as internally coherent networks of primitives and readout strategies.

Thagard’s (1989) model of explanatory coherence among propositions models locally coherent networks of propositions.

In the misconceptions view, the students’ explanation is assumed to stem from pre-compiled knowledge that is simply wrong, in contrast, according to our resources-based interpretation, the student compiles her explanation in real time from conceptual resources that are neither right nor wrong.

In a conceptions framework, it is difficult to account for why students would so quickly and easily drop a robust conception. In contrast, a resources framework readily explains — and even predicts — these kinds of shifts.

The term frame generically to refer to a locally coherent set of activations.Frame is an individual’s interpretation of What is it that’s going on here?

We posit the existence of numerous metacognitive and epistemological resources, including ones for understanding the source of knowledge (Knowledge as transmitted stuff, Knowledge as fabricated stuff, Knowledge as free creation, and others); forms of knowledge (Story, Rule, Fact, Game, and others); knowledge-related activities (Accumulation, Formation, Checking, and others); and stances toward knowledge (Acceptance, Understanding, Puzzlement, and others). Preliminary empirical work suggests that, perhaps to a greater extent than conceptual resources, epistemological resources tend to become activated in locally coherent sets.

Thinking in terms of manifold cognitive elements allows models of mind that can respond differently in different moments.

Along with Tannen (1993), we seek evidence of framing mainly in speech and other communicative acts. By a frame we mean, phenomenologically, a set of expectations an individual has about the situation in which she finds herself that affect what she notices and how she thinks to act. An individual’s or group’s framing of a situation that can have many aspects, including social (Whom do I expect to interact with here and how?), affective (How do I expect to feel about it?), epistemological (What do I expect to use to answer questions and build new knowledge?), and others.
Turning back to ontology, we take framing as the activation of a locally coherent set of resources, where by locally coherent we mean that in the moment at hand the activations are mutually consistent and reinforcing.
The phenomenology of framing corresponds, in this model, to a distributed encoding—the interpretation is distributed across a network of cognitive elements rather than located in any particular one.Frames can often shift easily.

In our theoretical perspective, framing generally involves the activation of numerous low-inertia cognitive resources rather than a single, high-inertia cognitive unit. Therefore, the resulting cognitive and behavioral stabilities are local to the moment. However, as we noted, when the same locally coherent set of resources becomes activated again and again, it can eventually become sufficiently established to act as a unit.

We do not rely entirely on passive reframing, in which contextual cues cause reframing to just happen in our students. We also appeal to active reframing, encouraging students to monitor actively their approach to learning. In other words, we try to get students to take an intentional stance toward epistemological framing.

We discuss three mechanisms by which a set of resource activations becomes stable, that is, reliably mutually activated and locally coherent within a given context.

One mechanism is structural: If resources have become compiled into a unit, their mutual activation and coherence are built into the cognitive structure itself. diSessa and Sherin (1998) and diSessa and Wagner (this volume) discuss such a unit as a coordination class. A set of resource activations can also be locally coherent for non-structural reasons.

Learner must actively monitor her thinking to ensure that her resource activations are globally coherent, rather than relying on features of the context to enforce that stability.

We are identifying three mechanisms for stability in a set of resources. One is contextual, a passive activation based on the situation, where by passive we mean that the pattern forms and persists without metacognitive resources playing any role.

A second mechanism is deliberate, meaning that it involves epistemological /metacognitive resources. To reason in a manner consistent with the Newtonian definition of force, a learner generally needs to monitor what conceptual resources she is activating and how.

The third mechanism for stability, again, is structural. With reuse, a set of activations can become established to the point that it becomes a kind of cognitive unit, and so a kind of resource in its own right. The cognitive unit
can have its own activation conditions, passive or deliberate. But once activated, the internal coherence in the resource activations is automatic.

By context, for an individual with respect to a set of resources, we mean the circumstances for passive but reliable activation.

If learning X in some context means that students reliably show that knowledge or ability in that context, there is no reason to expect that X exists other than as a pattern of activation in that context.
If its activation and stability as a set depend on features of that context, as is the case in passive activation, then the knowledge is not well attributed to the individual; it is distributed across the individual and the context, and therefore cannot be viewed as a thing the subject could move (transfer) from one context to another.

Tannen, D. (1993). Framing in discourse. New York: Oxford University Press.
Barnett, S. M., Ceci, S. J. (2002). When and where do we apply what we learn? A taxonomy for far transfer. Psychological Bulletin, 128(4), 612-637.
diSessa, A. A., Sherin, B. L. (1998). What changes in conceptual change? International Journal of Science Education, 20(10), 1155-1191.


One comment

  1. […] concepts. Then we try to tag the new abstract symbol (eg. photosynthesis) to these embodiments and relate these sensory-motor experiences to the new elements from the environment. It is about migrating the […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: