Post‐decision consolidation processes were studied in a two‐stage lottery design. School children (1) rated the attractiveness of audiotape head sets, (2) decided for which of two head sets they would like to get a lottery ticket, (3) threw a dice to determine if they got a lottery ticket in the preferred lottery or not, and (4)drew a ticket in the final lottery. Before and after each of these stages ratings were obtained of the attractiveness of the alternatives on four attributes. The results were interpreted in terms of Differentiation and Consolidation Theory. Post‐decision consolidation appeared immediately after the outcome of the dice throw. Subjects who won a lottery ticket consolidated their prior decision to a greater degree than those who lost. The effect was so strong for the former group that a pre‐decision attractiveness disadvantage was restructured to a post‐decision advantage. Folowing the decision, consolidation remained at approximately the level reached immediately after the dice throw for losers and winners, respectively, but with an increasing trend over time for winners. Even following a final loss of an audiotape headset the level of consolidation was maintained.

We investigated a recently introduced office automation system at Swedish Telecom. In a first study, where 275 users answered an inventory, the users indicated that they found the system useful, but that they would like better user support. Paper and on-line support were most often used at the syntax level, but were not regarded to be very satisfactory. Human advisors were found to be consulted most often, as well as being reported the most satisfactory means of support at the task and conceptual levels. In an interview study concerning the electronic mail system, the results from 35 users showed that the users were satisfied with the system, but that they knew fairly little about it, particularly at the semantic level. A third study showed that the system support personnel knew their users rather well. When asked to describe the system, system support personnel mainly described the system in computer terms, whereas many other users described it non-informatively. It is concluded that knowledgeable human advisors are needed lo support the efficient use of a system by illuminating task and semantic aspects. At the same time, easy-to-use manuals are needed to support the reminding about syntactic and interaction details.