Individual Similarity v Individual Differences

Yea, looks like the approach of my study is really an individual differences approach. That's the slant to my story? Universal theories of reasoning aspire to explain how we humans as a whole reason. This is an individual similarities approach, intending to collapse differences and focus on understanding our shared mental architectures.

In contrast, by throwing in WM capacity, we are interested in how a proposed general rule might vary across individuals.

Perhaps this might help in my writing of the introduction. Or perhaps not. :p

Senior Seminar: Lessons Learned (Pt 1)

1. Start early. START EARLY. START. EARLY.
   
2. Draft, draft, draft. Don't stop or get bogged down with details. Blah out everything that you can, process it, and then repeat the process. We've got to keep the ideas flowing.
   
3. Carve out significant blocks of time (at least 1 hour, preferably 2-3 hrs) to sit down and JUST WORK ON THE STUFF.
4. Divide and conquer. Break up the large chapters of writing into small writing assignments.
5. ORGANIZE. Related to point 4.
6. Related to point 2: PROTOTYPE. Use pilot studies. Try and fail. Don't fall into the trap of trying to run everything in your head.
   
7. Read up literature on experimental tasks if they are available. It'll help you get a sense of what the instrument is designed to measure, and what conditions are optimal for getting "clean" data.
   

The Revenge of the Strategy Theorists

Previously, I had written off the strategy theories of reasoning (e.g., the reasoning research group at Plymouth) as irrelevant to my research problem. Today, I re-read Bacon, Handley, and McDonald's (2007) paper on reasoning and dyslexia, and traced a citation back to Bacon, Handley, and Newstead's (2004) book chapter that described protocol and behavioral data on verbal and spatial strategies in transitive reasoning.

It looks like I need to reframe my theoretical thinking. This empirical battle may not be between mental models and imagery theories of reasoning, but rather between the domain-general vs domain-specific views on reasoning competence. Mental models theory falls squarely in the domain-general camp, and the logic of the VI-hypothesis rests partially on the assumption that we (i.e., normal, healthy adults) are "not supposed" to reason with visual images. The strategy theorists, however, argue against universal theories of reasoning, preferring to allow for individual variation in the algorithm classes that are employed in reasoning, each as valid as the other but differing in "fit" for various problem contexts (Bacon et al, 2004).

This has enormous implications for my data interpretation and discussion. It would be scientifically untenable to merely ignore the arguments of the strategy camp. In fact, my data have the potential to actually support their position: if I do find the VI effect, one possible reinterpretation of it would be as evidence for difference between "verbal" and "spatial" reasoners, as identified by Bacon et al (2007). In any case, the ardent call of the strategy camp to consider their data cannot be ignored in my case.

I must read a little more from the strategy literature and perhaps reframe the structure of my review of literature. There seems to be less and less impetus for devoting special sections to "syntactic" and "semantic" classes of reasoning theories; rather, it appears that a fruitful strategy for lively intellectual discussion would be to pit universal theories (spearheaded by the mental models theory) against the domain-specific views. Perhaps a novel synthesis might be reached using this dialectical approach. :)

The Bacon et al (2004) paper is in Roberts & Newton (2005) Methods of Thought: Individual Differences in Reasoning Strategies.