Theme 2: Theory

Theory

1. What is Theory? Theory is the underlying logical explanation of for example one or more hypothesis. It’s the “why” in an answer to a question, its an explanation to an observed pattern. Theory is for explaining “why” something is.

But what is not Theory? As we read in the text “What theory is not” we find that it’s quite a lot actually. The 5 main points are data, references, list of variables, diagrams and hypotheses. Data is not theory as it’s merely observations of occurrences and not explanations as to why these things are occurring. References is not theory either as they themselves do not contain the so important explanation to an occurrence as well, its just points to where someone else might have had a theory but without the possibility to absorb the explanation first hand in for example a paper. Hypotheses are mere assumptions or predictions and can not be theory without logical arguments as to why this hypotheses came true. etc
However these 5 might be parts to help creating a theory, but them by their selves are not theory.

2. I choose the paper “A Computational model of “Active Vision” for Visual Search in Human-Computer interaction” (http://www.tandfonline.com/doi/full/10.1080/07370024.2011.625237)
from the journal HUMAN-COMPUTER INTERACTION that has an impact factor of 1.476

The goal of the paper is to make way for a model that is suited for active-vision interfaces. The paper has been made around 4 central questions regarding just that. When do the eyes move? Where do the eyes move? What can be perceived in a fixation? and what information is integrated between eye movements? The paper answers all these questions using already know models to find the best one to predict eye movement. I think its fairly hard to categorise this as I’ve never done it before, but if i have to i think it might be a nr 1 in the table provided by the text from Gregor, and that would be Analysis as this paper uses existing models and analyses them i search for the beast match for a “future” active-vision model. It does however answer some questions using these models, and that one might say is some kind of prediction or explanation. But as it mainly focuses on testing other models i categorised it as Analysis.


3. The benefits are that using the Analysis model is that its maybe easier to understand as a reader since it’s step by step based with description and analysis of the subject. And by doing just that you can combine more models or theories as in my text where they analyse quite a few different models to find the most suitable for the task.
The limitations of this theory model is that it doesn’t really explain or predict anything. It just takes it for what it is and repacks it and thus not generating that much new information. Another limitation connected to the paper might be that since they focused around these 4 questions it was also “only” what they did. There might have been a lot more to the subject that they could have learned, but by focusing so narrowly they might have missed things.