Argument Structure to a Position Paper

A position paper takes similar shape to a research paper. A position paper focuses on the theory and analysis, while a research paper focuses proposed novel methodologies and results. The structure in common is the argumentation style. You want your reader to be convinced of your argument or proposal.

Your reader is a Cognitive Science researcher who may or may not know all the details of the work discussed in your paper, but is familiar with the topics and issues involved. Your job is to identify a relevant Cognitive Science problem, develop a position on a specific topic, gather all the necessary evidence on that topic, and present everything coherently in a position paper to your reader.

Below are suggestions to structuring your essay paper. Bare in mind that these are suggestions and you do not need to follow them word for word.

Sections

Although explicit sections are not necessary, you should be able to separate your paper as follows:

Introduction
Background
Analysis
Discussion
References

Details to consider for "Introduction"

Introduction: what is the problem you are investigating?

provide motivation for the problem
- why is it important?
- does it come up often?
- what is the impact of solving the problem?
define the problem
- provide necessary terminology and framework for the reader
- what's been done? identify related research and methods that have tackled this problem (if there are none - and you should be > 99% certain of this - then identify most closely related areas and suggest why they may not have been extended to this problem.)
- what's missing? identify research holes
present your argument
- state very precisely what you are arguing for in this paper
- outline the steps in the argument

Details to consider for "Background"

Background: what are the related frameworks or models?

General comments: Present the work clearly. Cite the appropriate sources -- please distinguish between primary and secondary sources. Be consistent with the terminology. Do not confuse notation or terminology between the model and an implementation of it. Do not confuse the hypotheses in the model and what researchers have (mis)interpreted.

Details to consider for "Analysis"

Analysis: what are the evidence to your argument?

General comments: Use research results as evidence to your argument. Be clear what it is that the authors did and claimed, and be clear what your interpretation of it is. Be critical to the research you read -- in every paper you read, there will be advantages and disadvantages to any approach (otherwise the problem would be solved already and there wouldn't be any research on it!). Be objective and try and see if the results can be interpreted in a way different from what the authors claimed. (This is what the Professor means by finding a "counterexample".)

Apply the same strategy to your own paper -- given the set of evidence and the problem you have identified, can the evidence together be interpreted in a way different from what is stated in your argument?

Details to consider for "Discussion"

Discussion: what should the reader take away from this paper?

How realistic are the positive results? Given the specific scope of the problems discussed, can we do better? Where might we expect limitations?

Negative results that were brought up before can be discussed here. Why were negative results observed? Are such negative results realistic? If so, what necessary changes are needed in the model/framework/etc. in order to incorporate these negative findings? If not, what can be changed in the research to avoid such patterns?

What are the major challenges observed? Are these obstacles commonly known? Is there a repeated pattern? Why are they hard?

This section may serve as a summary or conclusion, and a way to add comments on your own argument (e.g., justification).

Details to consider for "References"

References:

General comments: Be selective about your references. Make sure the references you choose are reliable.

You will need references for the background material. These may be standard references that are often cited in the literature.

You will need references for your evidence. These are the references that the Professor refers to when he says "recent papers". Look for conferences or journals for this purpose.