fredag 2 november 2012

Theme 2


What is theory? Gregor says that this question is not treated very much in the history. After reading the article by Sutton and Staw I can’t agree with that. Kaplan and Merton has for example said that theory is “answers to queries of why”. A little bit unclear and big answer, but still you can use that definition as a start. It’s about why things happens more or less, according to these guys. These thing can be anything from thoughts to events. Sutton and Staw says that “Strong theory, in our view,
delves into underlying processes so as to understand the systematic reasons for a particular occurrence or nonoccurrence.”.

So, if theory is about reasons for why things happen, what is theory not? Sutton and Staw makes it easy for us and puts all their points up, one after the other. They say that theory is not: References (they say that it’s used as a smoke screen to hide that you doesn’t have theory), Data, List of variables, Diagrams, Hypotheses. So, what are these things then? I think it more of tools to create your theory, to answer the why.

My paper this week is “Feature detection and letter identification” from the journal Vision Research with an IF of 2.414. This articles main goal is to look at how people recognize objects and in this case how they identify letters. The paper tries to look at how computation works when people are doing visual identification of familiar objects. One of the theory they base this paper on is signal detect ability for identifying one-of-many known signals in white noise by Van Trees, 1968. The theory type here is EP I think, since it is a mathematical theory which says what happens, why and especially what number you will get in which situations.

The benefit with this type of theory is that it is easy to duplicate their work since the theory often is good described in not only their work but also other people previous work.
Limitations is just that, that you can be limited by the theory and it’s guidelines of this theory. If you have guidelines on every step of the theory you must be critical to this theory and also to if it’s the right theory for you.

They have some other theories describes in this article, where on is Wertheimers (1923) theory which says that “a good object is more readily perceived as a whole”. This is more of the type Analysis with just a very short description. This is not limited as much as the above theory, but it doesn’t give you much to work on either.

1 kommentar:

  1. I agree that blindly following a certain theory is bad practice, which you claim as well. Theories are there to explain and provide a framework for understanding, but also to be criticized and questioned. This is sort of like applying “what does not kill you only makes you stronger” on theory, which opens up prior definitions for improvement.

    I must say I did not fully understand the Wertheimer’s theory, would you like to explain it shortly with an example?

    SvaraRadera