They have ran a small survey using Amazon Mechanical Turk (which is a pretty cool concept for itself) asking ~150 users to describe their search experience.

IMHO, their data analysis resulted in a very intriguing social search model.

“As we outlined through the model, social inputs may help users throughout the search process. Before searching, social interactions may help establish the requirements for the actual search task. During search, especially for self-motivated informational searches, users may talk to others for advice, feedback, and brainstorming to improve their search schema and query keyword selections. After search, users may still wish to engage with others to collect additional feedback or to share knowledge gained during the search.”
(Brynn M. Evans and Ed H. Chi – Towards a Model of Understanding Social Search | 2008)

P.S. It seems like we are on the right direction… :)

__________________

Quick (a.k.a. too late) Y.A.S.B.P. (a.k.a. Yet Another Shameless Blog Plagiarism) update: I had a strange (yet familiar) feeling that I have read about this research somewhere before but I didn’t remember where so after I submitted this post, I have found that Ofer had picked it up way before me and as always, wrote a much better post about it…