Moderating Variables

Linkages between perceived beauty and various outcomes or between design attributes and aesthetic perceptions, even if backed up by solid research evidence, common sense, or philosophical arguments, should not be considered universal or deterministic (Sutcliffe, 2010). First, as mentioned in the previous subsection, against studies that empirically found associations between aesthetic evaluations and evaluations of other perceived system attributes, such as usability, there are studies that found weaker or no such associations, indicating that at least under certain circumstances they do not hold. Second, in social settings, where research on the “beautiful is good” phenomenon accumulated evidence earlier and for much longer than in our field, findings suggest that the associations between attractiveness and perceptions of other human attributes are not unqualified (Eagly et al., 1991). Third, for all we know about socio-technical phenomena, it makes little sense that such deterministic relationships exist in a complex reality that involves individual, social and technological forces. Thus, adopting a contingent approach to the study of visual aesthetics would probably be more productive in describing if and how aesthetic evaluations mediate between various antecedents and consequences.

The challenge is then, to identify and examine how various factors serve to alter or moderate the aesthetic process. In Tractinsky (2006) I have provided a partial list of such potential moderators.� The list included the type of system used (a typology that can span multiple dimensions such as consumer product vs. a computer application; small vs. large display; personal vs. public; hedonic vs. utilitarian, etc.), the use context (e.g., work vs. entertainment), cultural differences (national, sub-cultural, idiological), and so on. Individual differences constitute an interesting group of potential moderators, because people vary greatly in their sensitivity to aesthetic stimuli and in their aesthetic preferences (e.g., Bloch et al., 2003; Hoyer & Stokburger-Sauer, 2011). Jacobsen (2004) study found consistent intra-individual aesthetic judgments but strong inter-individual differences in beauty judgments. In addition, the group model of aesthetic judgment misrepresented about half of the study’s participants. Pandir & Knight (2006) also found disagreement on aesthetic preferences in a study of different websites.

Contextual factors, such as domain and type of task are mentioned by Norman (2004) as important considerations for the type of aesthetic design required for users’ performance and satisfaction. He argues that in certain domains (e.g., control rooms) attractive design may not necessarily be desired. Ben-Bassat et al (2006) found that people weighed more usability over aesthetic factors when faced with a performance-oriented task, and Van Schaik and Ling (2008) demonstrated that attractiveness ratings were affected by providing context for the evaluation task. In online shopping environments Cai and Xu (2011) found that the effect of expressive aesthetics on shopping enjoyment was stronger when shopping for hedonic products compared to utilitarian products.�

Individual factors may also affect how anteceding variables (e.g., objective design attributes) are perceived differently by people with different aesthetic tastes (Hoyer and Stokburger-Sauer, 2011). In the domain of web-site design, Park et al., (2004) found that variability in user tastes is associated with aesthetic fidelity (i.e., the degree to which users felt the target impressions intended by designers). Individual differences were also found to affect the relative importance of aesthetics in people’s preference of web-sites (Hartmann et al., 2008).

Attributes of the choice process were found to moderate the relation between aesthetic evaluation and product choice, especially when users are required to trade-off aesthetic for other system qualities. For example, Ben-Bassat et al (2006) found that system preference or choice were affected by aesthetics under ordinary conditions (e.g., questionnaires) but not when the participants had to bid for a system with which they will perform competitive tasks. Diefenbach and Hassenzahl (2007) showed that under a beauty-usability trade-off, although people may prefer more beautiful products to more usable ones, they choose the more usable product if they cannot justify choosing the more beautiful one.

Cross cultural studies have shown that national and professional cultures affect various relationships between aesthetic evaluations, their antecedents and their consequences. Several studies have demonstrated this moderating effect in the context of websites. For example, Cyr (2008) found effects of visual design on trust in China but not in Canada or Germany, and Cyr et al (2010) found different reactions to web-site color appeal in Canada, Germany and Japan. Hartmann et al (2007) found that the aesthetic evaluations and the importance of aesthetics are contingent on users’ background (design vs. technical; Western vs. Asian).

The contingent nature of the aesthetic process is exemplified by Moshagen et al’s finding that high visual aesthetics improved performance under poor usability but had no effect under high usability. Consequently, they quoted Liu’s (Liu 2003) principle that �“... ergo-aesthetic design does not imply that workplace or product designers should only use designs that are pleasing or attractive. On the contrary, ergo-aesthetic design advocates the careful and proper selection of aesthetic levels of design to fit the needs and characteristics of the intended use” (p. 1298).

Дата добавления: 2015-11-16