Remember to use the blog comment box to identify which one of these you will read - and try to "pair up" so you have someone to talk through the paper with.
Examples:
- Huang - generating idea sketches through neural network systems
- Ball and Christensen - analogical reasoning
- Kim et al - creativity and limited commitment mode control strategy (e.g., Goel and Pirolli)
- Jin and Cusilp - mental iteration
- Kokovitch - mind mapping thinking tools
- Purcell and Gero - fixation
Key questions for discussion:
- In general, what can cognitive theories tell us about how people learn?
- What can they help us understand about how and why people design in the ways they do?
- What are some implications for understanding and/or facilitating design learning?
Oh yeah - the paper for the analysis we looked at in Week 9 (ping pong vs. streetcrossing) is in Blackboard for Week 9 (Adams-DTRS6).
15 comments:
I'm going to read the Huang (Investigating the cognitive behavior of generating idea sketches through neural network systems) paper.
I'm going to read Huang as well.
I think I'm going with Jin and Chusilp.
- Anne
I will go with Purcell and Gero (fixation)
Jin and Chusilp
Kokovitch
Jin and Chusilp
Comments on Svinicki:
While reading about the learning styles, I started to wonder about how a student’s learning style changes over time. For instance, the cognitive-metacognition model sounded much more like graduate school, however I can not imagine this approach would work for younger students. Are there any studies that look at how a person’s learning style changes with time?
I also was curious on if there was a way to evaluate professor’s given their teaching methods. For instance, it might be interesting to record multiple professors’ lectures and analyze their wording with regards to which learning style they best represent. Would a professor with higher end-of-year teaching scores show a propensity for a certain teaching style or a balance of styles?
On Huang:
For those that read Huang, I am kind of confused by the purpose of the study. Obviously if you draw a base a certain way, then that limits your options of what the rest of the pattern will be, do you really need a neural network? Plus, what happens if you don’t draw the base first? I know when I sketch, I have a picture pre-determined in my mind and I will start with the piece that is most important with regards to scale or one that is easiest to draw. The NN then really wouldn’t be analyzing what ideas I am generating given what I sketched, but would just be looking at common drawing components given what I sketched. I like the use of neural networks with design but did not like the example used in this paper.
Anne, since you read the Jin/Chusilp paper, you might be better able to answer some questions I have.
It was refreshing to get another look into the value of iterations. It was obvious to me that iterations depends on the design problem and constraints at hand. I liked the idea of the PR loop, since it allows the coevolution of the design to happen. It seemed like the IS loop focused more on the would-be results of the design process. So why did it seem that the paper alluded to the fact that IS loops allow for more creative designs? I would think that the coevolution of the design within the PR loop would allow for a more creative design.
I found it fascinating that more constraints lead to less frequent iterations. Does that mean that designers who are overburdened with constraints focus on trying to fit within the realm of the constraints and nothing more? If the designers can only see the constraints in their way and not the new possibilities that the constriants allow for, what does that mean for educators when we focus on the constraints of the problem? Perhaps constraints inhibit a design if the design is only focusing on the generate, compose, and evaluate portion of the design process. If the designer was comfortable with coevolution then I think that constraints will be less of a hinderence. Perhaps the key to good design education is allowing the students to take part in reflection of the total project so that constraints are satisfied and they are able to take part in design exploration.
I'm doing Kokovitch too
Tiago-
The idea of fewer PR loops and more IS loops in creative design made sense to me... PR loops (I think of them as spiraling down, narrowing your options) are about revising the problem definition. Maybe you re-read the problem and realize you missed a constraint in the problem statement, or maybe one constraint implies another constraint you didn't think about the first time, so you're slowly narrowing your design down to something less and less creative based on those constraints. IS loops (I think of them as spiraling up, broadening your options) open up new, unique, and potentially creative ideas that you didn't think of the first time you read the definition. Maybe in reading the problem the first time you subconsciously applied a constraint or brought up examples from your repertoire, but in thinking about it further you realized you could branch out into something less constrained and unique.
The idea that more constraints = fewer iterations made sense because with constraints you're already being led toward the solution without being overwhelmed with a lot of options which would need to be evaluated and eliminated later.
I hope you're able to follow my logic.
Oct 28, 2009/10/28 Design Cognition
In the Svinicki paper, a lot of what she mentioned about the metacognition and leaner-centered cognitive models sounds like design. Strategic learning =~ the design process, social constructivism =~ design as a social process, situated cognition=~ design as situated, motivation under cognitive paradign=~ “surprise”. Therefore the learning strategies Svinicki suggested can lend their hand on design education. The exception might be the “observe expert model”, since experts might operate on too much tacit knowledge that’s not apparent during observation.
The adaptive expertise description in the HPL article definitely fits how we discussed design expertise is like. Instead of talking about the mental models in links, associations, are there other kinds of representation of adaptive expertise that would fit with design?
Trying to think about how people learn, I think that these two papers compliment each other. The first paper took as from the very first behavioral approaches to today's views about the learner-centered class that also focuses on metacognition, classroom management and group work, while the second paper elaborated more not on the method used to learn.. but more i would say on what skills have got developed when you have become an expert. Thinking about design, i think i can map how we learn design more on the HPL chapter. We start as novices by randomly searching for information and solutions, while as time goes by (working alone, or with groups, practicing more trial&error strategies, and reflecting on our past experience) we end up on having developed a lot of design solutions per problem category, various working /data collecting/ problem solving strategies, and big chunks of information/solution categories. I have to admit though that (probably this is my bias since i have been taught design in the recent years) when trying to imagine how design would be taught in the 60's, I could not come up with any clear behavioristic model to teach that, since i can not very clearly identify parts to be purely memorized, except from a possibly linear design process.
While I enjoyed the chapter from How People Learn, it is hard for me to find "new information" from it. I feel like I have read this article before, numerous times. I struggle to find a conclusion or result about the differences between novices and experts that I did not already know.
I do however, especially like the first line of the last paragraph on pabe 49-"Expertise in an area does not guarantee that one can effectively teach others about that area" I feel this is an issue in engineering education that plagues the technical field. I have yet to understand why a college professor is not required to take education courses, but an elementary teachers spends years learning how to teach. I agree with the NAS and I hope that through this new field of ENE, we can help develop better engineering teaching skills.
The idea of "Adaptive Expertise" (HPL 45) really resonated with me. In considering the relationship between the two sushi experts, I wonder how each was trained. Personally, if to compare the two, I would claim the creative chef to be more of an expert than the one who can follow a recipe, but it this true?
From a teaching perspective, the recipe reader could probably easily teach his methods, with some additional commentary on style and form. The creative cook would have a difficult time teaching his methods because there is no order to the madness. I relate to this because I consider myself a creative cook. When asked for a recipe, I spout off a list of somewhat random ingredients and spices with no real concept of quantity or order. While I can't articulate my "training" as a common creative kitchen chef, I can't teach it simply either.
The recipe follower must be trained by the behaviorist model. If you mix these 8 items in a certain way, you will be rewarded with X item. Though this may help the person to become an expert in following recipes, where is the flexibility to stray and be creative? Is this important? What jobs/industries is this necessary? So it appears to me that the idea of adaptive expertise stems from cognitive models of learning (including metacognition and learner-centered models), with perhaps some foundations in the behaviorist model. Perhaps I could teach creative cooking by a cognitive apprenticeship and authentic problem solving applications.
So my first question was: which sushi expert is better? recipe follower or the creative? I guess it depends. If you order a Las Vegas roll, you expect a certain thing, so creativity isn't necessary. The creativity is best used to innovate and create (or DESIGN) something new. So if we are intending to teach design, and yield future engineers, creativity and innovation are key traits we should wish to instill in students. Therefore finding ways to integrate cognitive learning models into the somewhat necessary behaviorist model of teaching undergraduate engineering courses really is imperative. But then again, perhaps that's what we are all here for.
Post a Comment