In order to attract the attention of users current online shopping environments often involve the use of dynamic display changes such as animation, flashing or rotating text, and pop-up boxes. While there is some experimental evidence that supports the effectiveness of such changes for the purpose of capturing attention, such changes might also distract the user and hinder task performance. Thus, a key usability challenge for the designers of such environments involves balancing the need to attract attention with the desire to minimize any interruption of the performance of users. To date there is very little systematic investigation of the influence of display changes on consumers¡¦ attention and decisions during the browsing of online shopping web sites. The main goal of the present experiments was to research factors that might be important in determining the influence of such display changes. To accomplish that experimental tasks were created that resembled some aspects of the visual decision process and dynamic changes that occur during online shopping. However, in order to permit greater experimental control, the information environment used in the present tasks was substantially simplified as compared to online environments. Specifically, in 3 experiments, participants¡¦ eye movements were monitored while they chose between 2 alternatives, each represented by a set of visual images, with one set placed on the top of the screen and the other on the bottom. Immediately prior to this decision participants performed one or more binary decisions based on subsets of these images. On some trials, images were spatially swapped in the display presented during the final decision as compared to a prior presentation. Across experiments, we manipulated the relevance of the change to task performance as well as the recency of the information that was modified. By analyzing behavioral and eye movement measures, we documented evidence that participants flexibly and effectively accommodated to a variety of display changes. However, there was cost associated with display changes in the form of longer viewing times. In addition, task relevant changes and changes to recently viewed information were generally more disruptive. We discuss the implications of the present findings for the design of online shopping web sites and for future research that would attempt to generalize the present findings to more complex and more realistic online environments.