Thin Slicing in Information Space
Andrew Dillon, for the ASIST Bulletin, April/May 2005
This article originally appeared in the Bulletin of the American Society for Information Science and Technology, Vol. 31, No. 4, April/May 2005, a publication of the American Society for Information Science and Technology, 1320 Fenwick Lane, Suite 510, Silver Spring, MD 20910 USA. Phone 1 (301) 495-0900 http://www.asis.org
If term frequency is anything to go by, thin slicing is becoming the phrase de jour in my information world. Digital or analog, I cannot seem to browse anything at the moment without being hit over the head by Malcolm Gladwell's latest tome, Blink, which pertains to the power of thinking without thinking. The essence of this book is that snap judgments are often based on fairly deep knowledge, freed from the constraints imposed by consideration of too much information. No need to get started on the data versus information argument here - psychologists never cared much for that distinction anyhow, and we all surely know by now that an antelope is not a data point, it is an information "thing."
What is cute about thin slicing, apart from the phrase, is its justification for the cognitive miser in us all. (That equally cute phrase comes from BJ Fogg, a keynoter at this year's IA Summit). In a world of overwhelming perceptual stimulation, it seems that reducing the data and allowing intuition to guide us may be a useful coping strategy - it was surely such a blink reaction of my own that led me to avoid Gladwell's earlier book, The Tipping Point. Blink is a wonderful collection of anecdotes exemplifying this natural cognitive response in multiple situations and contexts. (My favorite is the expert who knew instantly upon seeing a new acquisition that the museum had sunk millions of dollars into a fake sculpture, despite its having conducted myriad tests of the object's authenticity). I have the same experience with many websites - just one look and I get an uneasy feeling.
While Gladwell does not mention the digital world much in his book, there are fairly well established findings in the user experience world that suggest people are themselves pretty responsive to very quick impressions of a resource or application. I've mentioned aesthetics before in this column, and it is clear from research on this topic that, despite our protests to the contrary, we really are quickly influenced to think positively or negatively about an application or a website on the basis of its initial look. Indeed, in my own research I saw how people inferred usability from just a quick look, and that subsequent testing revealed these guesses to be just that, guesses. Perhaps even more worrying, evidence of poor usability gained by interacting with the design did not initially change the beliefs of these users about the application's quality.
Snap judgments about information are everywhere. Recently, I came across a media report of iPod users who complained that the random function on their players was not really selecting tunes at random. Their belief was based solely on the occurrence of a couple of songs that popped up more than they considered appropriate for truly random play. Apparently this myth has reached such epic proportions that Steve Jobs asked the engineers at Apple to check and verify the algorithm. Those with a taste for conspiracy won't like (or even believe) the outcome, but it seems that random is random, and it really is possible to get 10 consecutive heads on 10 coin tosses. I have bewildered students repeatedly in my research class with this fact until they stop to think about it properly, that is, to realize that getting 10 heads in row is exactly as likely as getting any other specific sequence of heads and tails you care to list in advance. Blink, and you'll miss that one. The problem with those iPod users was they never let a random play sequence run its course, so to speak - which could take a week of continuous listening, so let's not be too accusatory here. Hitting the random function before the last one ends results in a new random sequence being generated, with the logical chance that the same old country song you once thought endearing will pop up as often as if it was on rotation on MTV.
This is the culture of quick: quick perceptions, quick reactions and quick conclusions. Blink, the book, legitimizes this response. Gladwell says he wants people to take rapid cognition seriously, and I agree with him. For better or worse, he is pointing to something in the human response to information spaces that should be of concern to any thoughtful information architect.
Comments and reactions to this column can be emailed to firstname.lastname@example.org