This research study involved a systematic investigation into the relationships among: (1) the techniques used by search analysts during preliminary interviews with users before engaging in online retrieval of bibliographic citations; (2) the amount of new information gained by the user as a result of the search; and (3) the user's ultimate satisfaction with the quality of the items retrieved. A series of controlled experiments, which involved two search analysts (Canadian university librarians) and 150 users, were conducted to explore the effects of two interview techniques: the conscious use of "open" and "closed" questions, and the use of pauses of different lengths by search analysts during the online negotiation interviews. Analytical techniques included two-way analysis of variance and path analysis of data. Among the findings were the following: the asking of open and closed questions had a modest effect on the amount learned by users; the type of pause did have a significant effect on the amount clients learned; the average user's satisfaction was higher when open questions were asked; overall satisfaction was lower when moderate pauses were used; those learning most about their topic were, overall, more satisfied than those who learned less; and those placing high importance on the information obtained tended to have lower satisfaction scores. (Author/ESR)