Models of interaction, part 1

on

Recently, I’ve been involved in a lot of discussions about exploratory search on this blog and in comments on The Noisy Channel. One way to look at exploratory search (and there are many others!) is to separate issues of interaction from issues of retrieval. The two are complementary: for example, recently Daniel Tunkelang posted about using sets rather than ranked lists as a way of representing search results. This has implications on one hand for how the retrieval engine identifies promising documents, and on the other for how results are to be communicated to the user, and how the user should interact with them.

Many theories and models of information seeking have been published in the literature. These deal with various factors: affective (Kuhlthau, 1988), process-oriented (Marchionini, 1995), behavioral (Ellis, 1989), tactical decisions (Bates, 1989), searchers’ goals and intentions (Belkin, 1980), etc. These models are¬† mostly high-level descriptions that don’t address issues of interaction. The specific details of interaction are left unexplored, with the tacit assumption that the user “just follows a link” or looks through a ranked list of postings.

But given that sometimes the actual mechanics of interaction can affect the effectiveness of the overall system, can we design interfaces¬† effectively without a good model of what’s going on? Waterworth and Chignell’s (1990) paper titled “A Model of information exploration” proposed an interesting model that tried to capture some aspects of interaction. It described three dimensions: structural responsibility (who is in control of the process) , target orientation (how much expressiveness the user has in specifying the required information), and interaction method (the nature of interactions that mediate exploration).

I’ve been thinking about this model and its implications, and don’t think it quite gets it right. In particular, the interaction dimension distinguishes between descriptive (i.e., query-based) and referential (i.e., link-based) extremes, but as I argued in an earlier post, this is a false dichotomy. Over the next few posts [part 2], I would like to explore this topic in more detail, with the goal of deriving an interaction model for information seeking that can be useful for design.

I would love comments and feedback on my ideas (as I get a chance to write about them) as I’ve been sitting on this for several years, and would finally like to get it off my chest!

References

  1. Bates, M.J. (1989) The Design of Browsing and Berrypicking Techniques for the Online Search Interface. Online Review, 13, pp. 407-424.
  2. Belkin, N. J. (1980). Anomalous states of knowledge as a basis for information retrieval. The Canadian Journal of Information Science, 5, 133-143.
  3. Ellis , D.A. (1989) A behavioral approach to information system design. Journal of Documentation, 45, pp. 171-212
  4. Kulhthau, C.C. (1988) Developing a Model of the Library Search Process: Cognitive and Affective Aspects. Reference Quarterly 28, pp. 232-42.
  5. Marchionini, G. (1995) Information Seeking in Electronic Environments, Cambridge: Cambridge University Press.
  6. Waterworth, J.A. and Chignell, M.H. (1991) A Model for Information Exploration, Hypermedia, 3 (1), pp. 35-58.

2 Comments

  1. Nice post–I’m looking forward to the sequels! As for my push for representing search results as sets, my hope is that it would be a first step in the right direction–and one that Cranfield-addled IR researchers could relate to. Indeed, it might even help address Ellen Voorhees’s objection that all of the interactive IR evaluation ideas are impractical.

    http://thenoisychannel.com/2008/04/17/ellen-voorhees-defends-cranfield/

  2. […] an earlier post, I described Waterworth and Chignell’s model of information exploration, and distinguished in […]

Comments are closed.