What is reflection's role within Kolb's Experiential Cycle? Viewing the writing process as a form of Kolb's Experiential Cycle, what is reflection's role within this cycle or process?

Research Approach

Quantitative/ Mixed Method

Knowledge Claims
(methodology)

Postpositivistic

Strategies of Inquiry

pre-experimental

Methods

Talk aloud protocol, text analysis

Sample/Sampling

Convenience sample of 10-20 students (most likely from my class, maybe more?)
Writing process artifacts, drafts, peer responses, reflections
Talk aloud transcripts for just activity of reflection for one writing cycle (with three drafts and three reflections)
Possible survey of student attitudes and impressions as well
Possible interviews with select students (perhaps 4) to investigate conclusions from transcript analysis (member checking)

Data Analysis

Close text analysis (possibly Content Analysis if I develop a coding system) of transcripts, interrater reliability check, statistical analysis of coding results

Possible Results

This study does two things—1) it seeks to affirm the Experiential Learning Model as a good explanation of the Writing Process, 2) it seeks to be a step in "successive approximation" in diagramming in more detail what happens inside the "black box" called "Reflective Observation" (and we can include Abstract Conceptualization).  The idea is to detail out the model more. 

Assessment

The danger in this study is that I see the Kolb Model as descriptive. I'm making nearly the same mistake as Flower and Hayes.  North says, "Formalist will set out to discover its [phenomenona's] underlying logic—the how or why or by what mechanism. …The eventual potency of any model we construct would depend on the degree of correspondence between its output data and the output data of the writing process, texts." Right now, I still don't see how to avoid the trap of representing the model as a representation/description of reality.
This study would also be logistically hard to carry out. Just to do the speak aloud protocols would be hard—I'd have to purchase recording devices or acquire them somehow. The advantage is that today these devices are very portable and less expensive and digitize. Transcription would be a pain also. Lots of time and work. (It would be cool if the audio could be automatically converted to text.) Also, the sample size and population seem problematic to me. Who? How much?
On the other side, since I have become more deeply interested in reflection I have basically been interested in a Formalist inquiry; I've has a model for where and how reflection fit within a model (that I considered a description) of the writing process/experience.

Permutations

I could investigate or triangulate to other models
--Flower's and Hayes Social Cognitive Model of the Writing Process
--Moon's Learning Stages Model
--Hayes' Model of Revision (which places reflection prominently)
--even Mezirow's Transformational Learning theory has a model that lurks beneath its assumptions
In a sense, all these models blend together for me?