Head-mounted displays, as popularized by virtual reality systems, offer the opportunity to immerse a user in a synthetically generated environment. While there is much anecdotal evidence that this is a qualitative jump in the user interface, there is little quantitative data to establish that emersion improves task performance. The authors present the results of a user study: users performing a generic search task decrease task performance time by roughly half (42% reduction) when they change from a stationary display to a head-mounted display with identical properties (resolution, field-of-view, etc.). A second result is that users who practice with the head-mounted display reduce task completion time by 23% in later trials with the stationary display, suggesting a transfer effect.<>