Abstract
The fluency theory of aesthetics proposes that people prefer stimuli that are easier to process (Reber et al., 2004). This was based, in part, on Martindale and Moore's (1988) evidence that prototypical colors were preferred to non-prototypical colors. However, their color set (from Rosch, 1975) confounded prototypicality with saturation—most of the prototypes were highly saturated and the non-prototypes were less saturated. Recent evidence demonstrated that verbal fluency—operationalized as naming response time (RT)—predicted color preferences of dichromatic and trichromatic males (Álvaro, et al., 2015). However, saturation is known to be related to color preference (Guilford & Smith, 1959; Palmer & Schloss, 2010) and it is possible that saturated colors are both preferred and more easily named, without fluency directly influencing color preferences. We tested whether fluency predicted color preferences when saturation was controlled. We included approximations of the eight prototypes (saturated) and eight non-prototypes (desaturated) from Martindale and Moore (1988). We also included +/-1 and +/-2 hue steps in Munsell space from those initial colors, while controlling value and chroma within each color category. Participants completed a color preference task and color naming task (counterbalanced). In the color preference task participants rated how much they liked each color. In the naming task they named each color as fast as they could. We calculated naming RT as the time between stimulus onset and naming vocalization. Color preferences were related to naming RT in the full dataset (r=-0.27, p< 0.05), but the relation was eliminated when chroma was partialed out (r=-0.13, p =0 .263). Within the saturated set where prototypically only varied by hue, participants were faster at naming colors that were closer to the prototype (r=0.33, p< 0.05), but there was no such relation between prototypicality and color preferences (r=-0.06, p=0.714). These findings suggest fluency may not causally influence color preferences.
Meeting abstract presented at VSS 2017