由眼動行為透析購物網頁的表現性美感體驗

Abstract

The experiential component of user behavior has recently become a focus in human-computer interaction research (Hassenzahl & Tractinsky, 2006; Moshagena & Thielsch, 2010) and adapted in ISO 9241-210. Visual aesthetics constitutes an important aspect of user experience as evidenced in the over-cited quote “Attractive things work better” (Norman, 2003). Currently, however, we still know very little on how beauty and use are associated since less research efforts were directed toward understanding the underlying process or mechanism of one’s aesthetic judgment than the measurement and description of such judgments. Purpose of the Study. This study examines the effects of webpage layout and mode of use on users’ aesthetic responses and explains the underlying mechanism using eye tracking metrics. It is hypothesized that webpages with the waterfall layouts created stronger expressive aesthetics perception (i.e., creative, original and usable) (Lavie & Tractinsky, 2004) than the grid layout. This effect was smaller when the viewer searched through the webpage with a specific goal than when they simply viewed it leisurely (H1). The closely neighboring and similarly sized grids in the grid-layout webpages are expected to create strong attentional capturing effects against current processing, resulting in greater numbers of fixations and regressive viewing for the grid than the waterfall layouts (H2). Method. Sixty participants took part in the layout (2) x mode of use (2) mixed design experiment, with layout (grid vs. waterfall layout) as a within-subject factor and the mode of use (viewing leisurely vs. looking for a product that one likes) as a between-subject factor. Eye movements were recorded as participants viewed and scrolled up/down webpages of 48 different clothing in each, after which they made evaluative responses towards the webpage. Results. The waterfall layout outperformed the grid layout on the expressive aesthetic measure but not the classical aesthetic measure. This layout effect was larger when the user was engaged in the action mode of use (i.e., leisure viewing) than the target mode of use (i.e., find the best one) (Hassenzahl & Ullrich, 2007), supporting H1. Number of fixations as well as object-based recurrence (Anderson et al., 2013) were both larger for the grid layout than the waterfall layout, supporting H2. Eye movement metrics including total fixation duration, recurrence and mean fixation duration provided accounts of expressive (not classical) aesthetic judgment in multiple regression models. These findings suggested that expressive aesthetic judgments are related to the degree to which one’s viewing behavior constrained by the visual interface is conducive to the current processing goal