Meeting #02 took place on 26th April 2017 at 9 o'clock at the Institute of Experimental Psychology in Bratislava. NG and MG attended. The meeting lasted till 11:30.
1) The first topic was understanding of the term program-solving perspectives. This term is apparently very confusing, as it is a custom term, generated to serve the needs of this project. MG proposed to drop this term, as it is not a good practice to define your own terms, especially if you are inexperienced to properly define them in a way they cannot be disputed. Considering this, we will go and replace all occurrences of the term perspectives with other terms (mostly related to Simon's mental models and mental representations).
2) Second topic was about experimental tasks.We have two available tasks for our experiment. One is regarding matchsticks-moves and the other regarding programming tasks. <Insert links here> Both are concerned viable options, so in the next week they have to be thoroughly examined and described. NG will provide MG with descriptions and definitions of arrays and lists and prepare a first draft of the project experiment.
Hello Nejc,
during the last week, I was thinking about your project and searching for the research problem that could fit your interests. Let me describe you "where is my mind" right now, think about it and let me know whether such research problem (or some parts of it) interests you or not.
The main idea i was considering was that a simple "switch" of a perspective could lead to a better performance in a certain problem solving tasks. However when researching many previous studies I found that we should be more specific and it will require to go further (deeper) in defining a research problem in this area. Many previous findings show that although our first idea is correct, in real world many people fail to recognize more than one/two perspectives of how to solve the situation. Moreover, when one recognizes a perspective, he tends to use it in other similar or isomorphic problem situations although there could be a better or simpler/optimal alternative. Authors describe this phenomenon as an "Einstellung effect". This is a good research area that could be used as a theoretical basis of our own study. Another field that is very related to the einstellung effect and a negative transfer of previous perspectives is a functional fixidness (although this is more a case of an insight problem solving). When considering all these research areas we could (in a very simple way) conclude that people often fail in recognizing other perspectives, they tend to fixate on a previous mental models that were helpful and when solving a new task they tend to search for similar patterns -some sort of confirmation bias. When defining our research problem, we could test these previous findings and also go further in researching how these "biases" are strong and how they interfere in solving another similar/isomorphic problem solving tasks.
These findings arose some questions in my mind that could be a basement of our project:
Q. 1: Do people tend to use one learned perspective?
Q. 2: To what extend are people rigid in using this perspective (although there is a better one)?
Q. 3: Are some perspectives more common to use?
Q. 4: What do people tend to do in a situation when their previous perspective is useless (how much time does it take to recognize another perspective)?
Q. 5: Does learning a different perspectives lead to better performance in an isomorphic tasks?
I was also thinking about the methodology of this research. I will try to describe this but I am pretty sure it will be quite confusing.
These questions could be answered by an experimental investigation and it would require to create three different groups that will solve a certain set of problem solving tasks but before that they will be shown and learned a different perspectives how to solve them. Group A will be learned perspective X, they will solve a certain number of tasks that are solvable only using this one perspective. Group B will do the same but only with a perspective Y learned and used. After that, both groups will solve TASK T that is solvable by both of these perspectives and both of them can lead to an optimal solution (testing Question 1). Further, they will solve another set of tasks that are solvable by learned perspective, but using a different perspective could lead to a more optimal solution (Question 2). Another task will require participants to recognize another perspective because the learned one will not lead to any solution (Question 4). Finally, there will be a Group C that will be learned both of the perspectives. They will also solve TASK T, and all the following tasks like other groups. This could answer whether they will tend to use one of these perspectives more frequently and whether they will be better in solving the last tasks that require switching perspectives (Q 3 and 5). Please find attached the document called "NG_MG_methodology" that could give you some kind of graphic illustration of experimental conditions.
This is just an idea of how the research problem could be formulated. Please let me know whether there is something that interests you, i think there is still a plenty of work on it, there is a wide range of how this plan can be widen or be more specified. However, when taking it globally I think that focusing on how we can improve/learn (or even whether it is able to learn) the ability to recognize and use different perspectives is more specified and deeper research problem than just comparing two different perspectives in a particular tasks. Naturally, it has also some disadvantages, mainly when searching for some problem solving tasks that could fit this methodology (I have found only one - the problem of matchsticks). I am not sure whether we can find a certain programming tasks that could fit this even partially. I was also studying the alternative you gave me (trees, arrays and lists) but it is very difficult for me to understand some things in this field and also to see how it can be used in a certain research problem. Finding a research problem that could fit the method is a reverse process and it is quite difficult even when you are not balancing in the middle of two science disciplines:) I think that for now we should decide whether we want an interdisciplinary (maybe more simple) research problem, or we will go deeper and use more common methods. Please let me know what you think about all of this.
Thank you very much!
First meeting took place on 10th of April 2017 at 9 o'clock at the Institute of Experimental Psychology in Bratislava. NG and MG attended.
NG explained the idea behind this research and took extra time to thoroughly explain 1) background research, 2) the gist of the new project and 3) his ideas how the project should look like. This was the main point of this meeting - to bring MG up to speed regarding our project.
Final conclusion was that MG will find and research the relevant articles to fully understand what the project is all about.
NG:
Due to my work schedule, I can only manage meetings in person in afternoons or during weekend.
Other times I would like to avoid if possible.
As you said, you have time only on Mondays and Wednesdays at 10:00 - 15:00, it is possible (though not favorable) that I come around next week on Monday.
Would that be OK for you?
In the meantime, a Skype call is still possible during the weekend. On Friday, I have still some work to finish.
What do you think?
Message me when you finish your first overview over the the wiki.