Semantic Effects on Aesthetic Preference for Color Harmony in Visual Displays
Previous research on color pair preferences has found that people generally prefer harmonious color combinations (Schloss & Palmer, 2011). Research in aesthetics of spatial composition has shown that although there are default preferences (e.g., center and inward biases) (Palmer, Gardner, & Wickens, 2008), such preferences can be violated when contextual cues are introduced (Sammartino & Palmer, submitted). This experiment aimed to investigate whether default color pair preferences could also be violated by changing semantic context. Participants were instructed to imagine that they were graphic designers who were asked to choose which color pairs they would prefer for the main colors of an album cover. There were four bands, two of which had harmonious sounding names ('Peace' and 'Unity') and two of which had dissonant sounding names ('Friction' and 'Chaos'). To emphasize this distinction between the two types of bands, participants were presented with two 15-second samples of music by each band. We found that for harmonious bands, participants preferred a harmonious color combination, but for dissonant bands, participants preferred a dissonant color combination. These results suggest that, like spatial composition preferences, people's color preferences change with context to fit the images' intended meaning.
Cross-modal relations between emotional content and preference for harmony
Previous research has shown that individuals differ in the degree to which they prefer harmonious color combinations, as measured by the correlation between ratings of preference and ratings of harmony for figure-ground color pairs (Schloss & Palmer, VSS-2007), Further research shows that this tendency is also correlated with preference for harmony in music and preference for figural goodness in spatial images for Berkeley undergraduates majoring in Psychology, Art, and Music, although specific training in a relevant domain tends to decrease a person's preference for harmony in the domains of training (Griscom & Palmer, VSS-2010). In the present study, we investigated the relationship between preference for harmony and the emotional associations of these stimuli. Participants were asked to rate various stimuli for consistency with the emotions of happy-sad and angry-calm. The stimuli were 56 color pairs, and 14 30-second clips of classical piano music, 35 images of a single dot at one of 35 positions inside a rectangular frame, and 22 Garner-type 9-dot configurations that spanned a wide range of "harmony" ratings. We found that there were strong correlations between ratings of harmony and ratings of positive emotional associations for music and color pairs: e.g., the music that was judged to be harmonious tended strongly to be judged as happy rather than sad and calm rather than angry. The same was also true of positive emotions and ratings preference for music and color pairs. In previous research music and color pairs showed the highest cross-domain correlations in preference for harmony (r = .64). These findings suggest that consistent cross-domain preferences for harmony may reflect, in part, a preference for the positive emotional associations evoked by harmonious stimuli.
Individual Differences in Preference for Harmony
Previous research has shown that individuals differ in the degree to which they prefer harmonious color pairs, as measured by the correlation between their ratings of preference for figure-ground color pairs and their ratings of harmony for the same color pairs (Schloss & Palmer, VSS-2007). In this study, we investigated whether individual preference for visually "harmonious" or internally coherent stimuli is consistent across different stimulus types. The stimuli used were: 35 images of a single dot at one of 35 positions inside a rectangular frame, 22 Garner-type 9-dot configurations, and 16 color pairs. All displays were chosen to span the full range of internal coherence possible within the given stimulus type. Twenty subjects were asked to rate aesthetic preferences for each stimulus on a computerized line-mark scale, and were later asked to rate the internal coherence of the same stimuli ("harmony" for color pairs, "goodness of fit" for dot-in-a-frame images, and "simplicity" for Garner dot patterns) using the same method. Subjects also completed the 44-question Big Five Inventory and the 40-question Sensation Seeking Scale. We found that individual subjects' preference for internally coherent stimuli (i.e., harmonious/good-fitting/simple displays) was strongly correlated across different stimulus types: r=.46 for color pairs and dot-in-a-frame images, r=.71 for color pairs and Garner dot patterns, and r=.49 for dot-in-a-frame images and Garner dot patterns). Somewhat surprisingly, the personality measures we examined were not significantly related to preference for harmonious stimuli. These results may indicate that there may be an underlying factor (aesthetic style?) connecting preference for harmony across visual stimulus types, and we are currently engaged in an expanded follow-up study using a larger number of color-pairs, quasi-randomly generated polygons that differ in the number of sides and degree of symmetry, and harmonious-to-dissonant solo piano music.
The Relationship between Color and Form in Judgments of Preference and Harmony
In this study we investigated how the shapes and colors of lines influenced both preference and harmony ratings. We used two types of lines, irregularly jagged and smoothly curved, which were each presented in isolation as well as in side-by-side pairs: both jagged, both curved, and jagged + curved. Colors included eight saturated hues (unique-red, orange, unique-yellow, chartreuse, unique-green, cyan, unique-blue and purple). Single lines appeared in each of the eight colors. Pairs of lines appeared in one of three color relationships: same hue (identity), adjacent hues (analogous), or opposite hues (complementary). Average preference ratings were only affected by line type: curved single lines were preferred to jagged single lines, and pairs containing same-shaped lines (both curved or both jagged ) were preferred to pairs that contained one line of each type. In contrast, harmony ratings were strongly influenced by both line type and color relationship. Single curved lines were judged more harmonious than single jagged lines, curved pairs were more harmonious than jagged pairs, and pairs with different lines types were judged least harmonious. For all line pair types, harmony ratings increased as a function of hue similarity, consistent with Schloss and Palmer's (VSS-07) findings that color harmony for figure ground pairs is largely driven by hue similarity.
All of the above research is funded in part by NSF Grant BCS-0745820 to Stephen E. Palmer and by a gift from Google, Inc.