HCI Course Exercise: Level 11

This exercise requires a critical, directed survey of recent work in HCI, selecting a particular area of work, and isolating the most promising paper(s), so as to design a new piece of empirical work that would help take the field forward.

You are free to select the area of work of interest. However, it is recommended that you perhaps use the topic areas which organise the CHI Conferences to narrow down and help define your area. For instance, the area could be HCI for ubiquitous computing, or HCI for interactive artistic work, or HCI for embodied conversational agents.

For instance, suppose you choose to consider embodied conversational agents. An interesting set of questions arise as to what the correct ways of controlling the facial expressions and head gestures which correspond with particular language acts. In particular, should we base the face's expressive repertoire on one individual's facial expressions and head gestures, to ensure consistency? If so, which individual should we choose? And how important is it to get the expressions and gestures to line up properly with the spoken descriptions? In surveying this area, you would want to pick up on the most relevant articles, and then argue which one (or ones) is/are most promising, what interface design recommendations derive from them, and why at least one interface design recommendation is important. You would then need to suggest a new investigation or experiment to resolve an outstanding question concerning that design recommendation.

Materials Provided

Here are some initial pointers:


  1. Use web resources, web search (e.g., Google Scholar), and the Library's access to electronic journals, to start from the given references, and locate relevant recent work which bears on an interface design issue of interest.
  2. Design a simple empirical investigation. Justify the details of the proposed method, and explain what the results would indicate.
  3. Write up a report.

To select a promising direction from recent research, you might look for a cluster of closely related papers, or a paper by a prominent author you believe is likely to be influential. The IRM course will be useful to you here.

The Report

You should write a report about 2000-3000 words in length (advisory word limit for guidance only). The report must cover the following aspects of the exercise:

  1. It should contain a critical summary of the relevant literature, emphasising any significant recommendations for interface design.
  2. It should also contain a report on the design for an investigation testing an interface design recommendation (resulting from the survey), and could contain pilot results and simple analysis.

Don't forget to put your name and matriculation number on the front cover.

Getting Started

Parts 1 and 2 of Dix, Finlay, Abowd & Beale address many of the basic issues that you may encounter in the course of the exercise. Several chapters in Part 3 and 4 go into more depth in particular topics, and could help provide background depending on what you choose to investigate. For example, Chapter 20 provides an overview of Ubiquitous Computing.


The deadline for submission of reports will be 4pm, Friday 7th December (that is, the end of week 12 of Semester 1).

Submission should be an electronic document in pdf format, submitted via the on-line submit system as follows

submit msc hci-5 1 report.pdf

The first page should give your name and matriculation number. Please use the suggested file name for our convenience.


Grading is based upon the following basic scheme:

Detailed assessment draws upon the following points:

Please note that only a small percentage of marks are allocated for the bonus; we expect that most people will not attempt it.