This paper describes a study conducted to understand, in part, the effects of web interface features (image size, fidelity, and motion) on responses such as attention, and memory. The increasing proliferation of B2C web sites and their attempts to enhance the experience of customers shopping on-line has made the work of Reeves and Nass on the psychological responses elicited by interactions with media relevant to the electronic commerce domain. This study is an attempt to validate the claims of Reeves and Nass and extend their theory to web-based media. We have conducted a laboratory experiment to test the influence of three web design features—image size, fidelity (clarity of an image), and motion—for an experimental electronic commerce website. Subjects were instructed to search for information on the web, and given attention and memory tasks that were then used to measure the impact of these three web design features. Results indicated that, at the early stages of a subject’s interaction with a web site: (1) higher visual fidelity images on a web interface lead to greater user attention to the product examined than lower visual fidelity images; (2) motion on a dynamic web interface demands greater user attention than a static web interface; and (3) an interface with higher fidelity and motion leads to greater attention span in comparison to one associated with only one feature manipulated. In addition, compared to smaller images, larger images on a web interface enhance user memory performance for images. In terms of practical applications, the study indicates that interface features, such as fidelity and motion, which are instrumental in keeping customers at one’s web site longer, are important and may lead to an eventual purchase. Second, it is becoming evident that a key role of the web site is not only to lead to the purchase of a company’s product over the web, but also to lead customers to visit one’s physical store, and eventually to an “off-site” purchase. The results of this study show that size is an important variable that influences customers to remember the image aspects of a product, and this might lead to a higher likelihood of off-line buying. Overall, this study confirms the relevance of Reeves and Nass’ studies in the area of human–media interaction. Also, it sheds new light on the application of their work to the electronic commerce context. It also contributes knowledge to the research community with a relatively new paradigm of studying interface and human–computer interaction.