PSYC 579 – Design Reports


A report is due from each student at the time of the final exam.  The purpose of this report is to examine a visual interface (of a program, website, etc), and (i) determine the extent to which it is optimally designed from the point of view of perceptual mechanisms, (ii) give some suggestions as to what be done to further improve its design.


The Project


Imagine that you have been hired by Company X as a consultant to evaluate their latest system / website.  Your job is to thoroughly analyze its visual/interface design and prepare a report to show them three things: (i) what works, and why, (ii) what does not work, and why, and (iii) ways that its visual design can be improved, either by modifying its existing functionality or by adding new functionality outright.


You can select any target system you like.  But it should be complex enough to be analyzed in terms of several different psychological mechanisms.




1.    Introduction.  The report should begin with a brief introductory section describing what this system does. Assume that the reader doesn't know what the system does, or how a typical user would interact with it.

[At least 300 words; screenshots are useful]


2.    Five perception components to investigate.  Chosen from the first set of lectures for each week (Tuesdays).  For each component, provide:

a)    Background.  A review of the current state of psychological research on this component, including the basic mechanisms involved.  This should have the form of a brief summary aimed at an uninformed (but intelligent) reader, with emphasis on those aspects discussed in points (b)-(d).  It should include relevant sections of Ware and of Palmer, as well as any other relevant sources (including papers on design).  Each citation should be from one or more of the following (in descending order of priority): Ware, Palmer, Required Readings, Suggested Readings, anything else that is relevant.  However, it should not make direct mention of the course notes themselves.

[At least 450 words.]

b)    Design Strengths.  Describe 3 different aspects of the design that make good use of this component.  For each, explain why, drawing on relevant information in the review in part (a).  [At least 3 x 50 = 150 words.]

c)    Design Weaknesses.  Describe 3 different aspects of the design that make poor use of this component.  For each, explain why, drawing on relevant information in the review in part (a).  [At least 3 x 50 = 150 words.]

d)    Recommended Improvements.  A list of 3 recommendations about how to improve the design of the system while not interfering with existing functionality.  These can be either a fix to a problem described above or a completely new function.  For each explain why it would improve performance, drawing on relevant information in the review in part (a).  [At least 3 x 50 = 150 words.]


3.    General Conclusions.  Summarizes the main highlights of the sections above.

[At least 200 words.]


The report should total at least 300 + 5 * (450 + 450) + 200 = 5000 words.


The document should be a pdf file.  Please email to me by the due date.




At c. 300 words a page, this report will be over 15 pages long; it will therefore take a while to do.  A good approach is to try for one component per week on average, starting around the midterm break.  You should then have most of it done by the end of classes, and can use the week after that to polish it up some more.  It is a bad idea to try to do everything during the last two weeks of class.


Writing the report at a rate of one or two components per week will allow you to think about your system at the time that component is discussed in class.  When you write the review part of each perceptual component, your understanding of its nature will still be relatively clear, and so it should be relatively easy to do this at that time.  (If it's not, this may motivate you to ask relevant questions in class.)  Also, applying the principles discussed in class immediately to your target system will help make clear how the perceptual and design aspects regarding that component interact.


Frequently Asked Questions


  1. Q: Suppose there are several reasons for recommending something (e.g., changing the colour of the background).  Is this counted as just one recommendation?

A: Not necessarily.  If the reasons for doing something involve different perceptual mechanisms, these can be counted as distinct.  Similarly, if the same perceptual mechanism is used to justify different design recommendations, these can also be counted as distinct.


  1. Q: Do the recommendations have to be entirely positive?  What if they result in a potential problem for some other perceptual component?

A: Recommendations that cause potential problems for other perceptual components are okay.  But the potential problems should be pointed out, and the possible trade-offs discussed briefly.


  1. Q: What if the system isn't really such that it can be usefully analyzed in terms of five different perceptual components?

A: The system should be such that it can at least be analyzed from three or four.  If this is not possible, you should consider another candidate system.  If the fifth component is only somewhat relevant, be a bit imaginative—part of the value of this exercise is to get you used to applying perceptual components to visual design.  Plus, if you've done the other four components well, you'll still get a 90 percent mark even if the fifth component is only halfway relevant.


  1. Q: What if I want to discuss some aspect of perception not covered in the lectures (e.g. transparency)?

A: This is perfectly fine.  The lectures are intended only as an introduction to visual perception and visual design—there is a lot more out there.  Most of the "basic components" of perception have been covered in this course, and chances are that what you're interested in is part of an existing component (e.g., transparency is a further development of the component on lightness and brightness.)  Take a look at e.g., Palmer, and see if it is a natural extension of something we've covered.  If in doubt, contact me.


  1. Q: What about aspects of perception that overlap several components?  (E.g. the colour of an illusory surface resulting from grouping, or depth defined by motion.)

A: Pick the way that works best for you.  One way of doing this is to look at whether the issue is one of "means" or "ends".  For example, depth due to motion parallax can be analyzed in terms of motion (i.e., properties of the motions that create depth).  It could also be analyzed in terms of depth (i.e., properties of the depth profile that results, without much consideration of how it was generated).  If in doubt, contact me.


PSYC 579 home page