Wednesday, October 8, 2008

Week 8 - Cognition and Design

The readings for this week (Hunt and HPL) give an overview of cognition and early approaches to articulating design as cognition (Simon Chp5). This is a fair amount of reading, but they are written for a broad audience and so should be "easier" to read than say, Goel and Pirolli. The first chapter of Simon is provided as a "for your information" and provides an in depth analysis of what he means by a "science of the artificial".

Simply put - as you read these consider:
(1) What is cognition? What does it have to do with how people process information, store and retrieve from memory, organize their knowledge, and reason?
(2) How can you relate these ideas to how people design and what they know about design? Feel free trying to connect these ideas to what we did this last week!


celia said...

Cognition is a word which carries rich meanings. In general,it can be interpreted as "the process of understanding and making sense of the world". Reseach in cognition should include but not limit in how people process information, store and retrieve from memory,etc.
Studying cognition in the context of engineering design will offer us insights into how people acquire design knowledge and how they approach design. Particularly, the research in expert designers and expertise development gives us a chance to know what a successful learning process and an excellent performance look like.
Though computer programs is claimed to be used as simulations of human thoughts, I doubt whether they can really simulate. It seems that they follow certain rules like "pattern-action" "searching through problem solving states". My questions: Can these machines/programs generate new ideas? How can they simulate design as social process?

§adieLovingtonNibblesworth said...

I love Celia's comment here.
My more formal training in cognition comes from the Italian school, which is heavily concerned with deep questions in ontology, and therefor is steeped in semantics. The main issue of cognition in my training is the portability of the definition to artificial systems, thus we negotiate cognitive process as not only branches of mental processes, but as negotiations of STATES of mind, (aka mental states play a role in cognition).

As for the writings, there is a rich traditions in the MODES of cognitive thought and how those came to be realized, as well as the way cognitive processes depend on other mental faculties. Hunt in his paper directly references the dependence of cognitive thought on other functions, and most importantly emphasizes the role of memory being not at all divorced from the any cognitive state-of-mind. In fact, one could say that the way cognitive processes pull from and contribute to memory *is* still the quintessential problem in theory of mind today.

My interpretation of the HPL work is that the expert and novice have different ways of cognitively processing any given state of the world. A twist on that is that the novice and expert have different states of mind when cognitively evaluating a world situation. From this lemma, the work of HPL is a rich platform for interpreting cognitive processes in design.

From last week's work, we have seen that some designers cognitively react to snapshots of a design process quite differently than others. Other designers are unable, cognitively, to step back from the situation and evaluate readiness/"finishedness" effectively. We have seen examples in class of the "good enough" evaluation, and some students of design are capable of cognitively taking a step back, assessing the state of the situation, and accurately determining when it is "finished." Not only is the ability to judge readiness or "finishedness" a specific, cognitive ability, but the ability to step back and realize when that is useful is a valuable, cognitive tool.

Aidsa said...

In general terms, cognition refers to a faculty for the processing of information, applying knowledge, and changing preferences. In the paper by Hunt, cognition was presented related to abstract concepts such as mind and learning; the capabilities of the mind and expected properties of an artificial “mind” by the use of computer programs (“artificial intelligence”). It was considered an abstract property of advanced living organisms and studied as a direct property of a brain at symbolic levels. Which I believe have some relationship to the ideas presented in the chapters of HPL about expert and novices mental models. The authors provided six principles of expertise which differentiate experts from novices. Computers programs that have been developed to mimic the human mind and mental processes, such as the GPS, intend to simulate mental models followed by experts.

Five of the six principles of expertise could be related to the design process. First, when solving a problem, designers seek for patterns of information based on individual perception (social aspect). Experts develop patterns of meaningful information (schemas) while novices lack of the ability to encode meaningful information. Secondly, ideas get organized according to the domain. Experts’ thinking appears to be more organized around big ideas according to their expertise; they seek to understand the problem around these ideas. As a contrast, novices approach problems trying to fit a formula or procedure they are familiar with.
The third aspect relates to the access to knowledge pertinent to the problem. Experts retrieve information relevant to the problem stated. This requires for conditionalized knowledge instead of inert knowledge. They need to know when, where, and why to use the knowledge. Similarly, the next step involves for the knowledge retrieved to be effortless or fluent for the designer. The final principle relates to adaptive expertise. Expert designer should be metacognitive; thus they should not only be able to solve a problem, but to consider weather it is the best way to design.

CJ said...

I found the readings for this week very interesting, and a great introduction to cognition. I especially feel like I have a better understanding of memory, based on that chapter by Hunt. The human mind/brain is fascinating, and there are so many factors to consider when we talk about how people think (situations, past experiences, etc.).

From HPL we get a bit of a broader idea about how novices and experts differ, not just novice and expert designers. I think Aidsa’s comments are interesting, and I actually think that all six of the key principles are relevant to design. The last one of the key points that I thought was relevant is “though experts know their disciplines thoroughly, this does not guarantee that they are able to teach others.” We have also found that this seems to be true of expert designers, that although they know how they would approach certain problems, they may not be able to classify each and explain them to others. I think this could also be true of decision making in design, experts may easily be able to choose between different alternatives, but may not easily be able to explain to a novice how to choose.

Also, although the article states that “expert teachers know the kinds of difficulties that students are likely to face” and how they can utilize the students’ existing knowledge in order to make new information meaningful, I do not necessarily think that all teachers do. However, I think it may be a difficult task, and also largely depends on the motivation of the students and their willingness to participate, learn, and try to understand.

little-T truth said...

This week's readings were great for thinking about how engineers/designers/individuals go about thinking about the/a "process."

Cognition is generally as sense of understanding or awareness in which thought takes place. There is some debate about the relationship between the idea of the mind, as an abstract thinking engine, and the brain, as a biologically functioning organ. There is some relationship, in my opinion, between the two in how they function, i.e. a healthy brain contributes to a stronger mind in SOME subtle way.

In relation to actual thought processes, cognition varies by individual based upon the structures they are able to contruct as they learn. This process of constructing learning, especially "chunks" of information which can further increase one's thinking process, is not only a learned characteristic but a developed (trained) one. People, in my opinion, instinctively learn how to learn up to a point. There is a necessity, however, to be taught (or to learn) by organized education to start making connections in relations to ideas that may not (in my case I think it is not) possible without the educational challenege. Examples of this would be schemas that experts are able to develop over time is based on their early experiences as novices where they were "forced" to repeat a process approach to problem-solving (of well-structured problems)./ This structuring provided the beginning of a new branch of mental models they would use to approach the process of solving all problems. As the individuals gained experience within their domain and hrough practice, they modified their mental models, including the creation of shortcuts ("chunks"), that allows them to approach the problem-solving process in a way that is perceived as expert today, yet was based partly in the training of earlier education.

People engage in design based on their wordview and processview of the world. These constructs are unique to the individual, particularly in how they have been combined to form the mental model utilized for solving problems. I use myself as an example in this case. When confronted with a paper engineering problem, such as the coating is dusting off the paper in the supercalenders (supercalender is a stack of rolls that use mechanical action to impart smoothness/gloss to coated papers for magazines, etc..), i immediately begin to think about the concepts of coefficient-of-friction and coating structure mobility. If I am being asked to assemble a piece of engineered furniture, because I do not have the best mechanical prowess, I return to my basic problem-solving approach of laying everything out, what do i have, what do i need, how do the instructions say I should do this, etc... essentially I feel like I make less jumps using "chunks" but can still utilize a gained knowledge of approach o solve the problem.

Thus cognition and how individuals engage in their thinking relates to how they would go about being a designer and engaging in design projects.