Thomas Langlois

Color, Music, and Emotion

The original color-music experiment used 18 recordings of orchestral selections as musical stimuli. 1. Will the color-music associations replicate with more tightly-controlled musical stimuli? 2. Which musical dimensions are most in uential in the relation between colors and music? The present study used single-line unaccompanied piano theme-and-variation melodies to vary: mode (major vs. minor key) note-density (quarter-notes vs. eighth-notes), tempo (fast vs. slow speed) register (high vs. low) while controlling for (i.e., no variation in): dynamics (soft vs. loud) orchestration (multiple instruments playing together) timbre (tone color) harmonic mode (harmony of simultaneous notes)

Effects of Implied Motion and Facing Direction on Positional Preferences in Single-Object Pictures

Palmer, Gardner, and Wickens studied aesthetic preferences for pictures of single objects and found a strong inward bias: Right-facing objects were preferred left-of-center and left-facing objects right-of-center. They found no effect of object motion (people and cars showed the same inward bias as chairs and teapots), but the objects were not depicted as moving. Here we measured analogous inward biases with objects depicted as moving with an implied direction and speed by having participants drag-and-drop target objects into the most aesthetically pleasing position. In Experiment 1, human figures were shown diving or falling while moving forward or backward. Aesthetic biases were evident for both inward-facing and inward-moving figures, but the motion-based bias dominated so strongly that backward divers or fallers were preferred moving inward but facing outward. Experiment 2 investigated implied speed effects using images of humans, horses, and cars moving at different speeds (e.g., standing, walking, trotting, and galloping horses). Inward motion or facing biases were again present, and differences in their magnitude due to speed were evident. Unexpectedly, faster moving objects were generally preferred closer to frame center than slower moving objects. These results are discussed in terms of the combined effects of prospective, future-oriented biases, and retrospective, past-oriented biases.

Texture-Color Associations in Non-synesthetes

Previous work has revealed that non-synesthetes exhibit cross-modal music-to-color associations that appear to be mediated by emotion: e.g., people chose happy-looking colors as going best with happy-sounding music and angry-looking colors as going best with angry-sounding music (Palmer et al., 2013). A series of further studies revealed that music-to-texture associations (Langlois et al., VSS-2014; Peterson et al., VSS-2014) and shape-to-color associations (Malfatti et al., VSS-2014) also appear to be mediated by emotion, although non-emotional mediators are present as well. Here we show that emotional mediation generalizes to cross-modal associations from visual texture to color and from color to visual texture. We used the 37 colors of the Berkeley Color Project: saturated, desaturated, light, and dark shades of 8 hues (red/orange/yellow/chartreuse/green/cyan/blue/purple), plus white, black, and 3 grays. The textures consisted of the same 28 simple line-based textures generated by Langlois et al. (VSS-2014). In the color-to-texture condition, participants viewed each color individually and selected the three textures that were most consistent with it and then the three textures that were least consistent. Participants then rated each texture and each color individually on 5 emotional dimensions (e.g., happy/sad, angry/not-angry) and 5 non-emotional dimensions (e.g., safe/harmful, sharp/smooth). For each dimension, we computed an index of the color-texture associations (CTAs) as a weighted average of the relevant emotional (e.g. happy/sad) or non-emotional (e.g., sharp/smooth) ratings of the 6 textures chosen as going best/worst with each color. In the texture-to-color condition, each texture was shown individually, and participants selected the three colors that were most (and later, least) consistent with it, from which we computed an index of the texture-color associations (TCAs) in an analogous manner to the CTAs. Consistent with previous results, participants in both conditions reliably associated colors and textures that have similar emotional content for most of the emotional dimensions. Meeting abstract presented at VSS 2015.

Relations among Visual Texture, Musical Features, and Emotion

Previous research indicates that systematic music-to-color cross-modal associations in non-synesthetes are mediated by emotion (e.g., Palmer et al., 2013; Langlois et al., under review; Whiteford et al., VSS-2013). The present research asks whether specific musical features mediate cross-modal associations from music-to-texture by using single-line melodies that vary along highly controlled musical dimensions: register (low/high), mode (major/minor), note-rate (slow/medium/fast), and timbre (piano-sound/cello-sound). We also investigated whether these associations are mediated by emotional and/or geometric factors. First, 46 non-synesthetic participants picked the 3 most-consistent (and later 3 least-consistent) textures for each of 32 variations on a synthesized melody from a 4x7 array of black-and-white textures. Next, they rated each melody, and later each texture along 5 emotional dimensions (Happy/Sad, Angry/Not Angry, Agitated/Calm, Weak/Strong, Harmonious/Disharmonious), and a series of geometric dimensions (e.g., Simple/Complex, Sharp/Smooth, Granular/Fibrous, Curved/Straight, Separate/Connected). For each dimension, we computed Music-Texture-Associations (MTAs) as a weighted average of the ratings of the 3 textures chosen as going best/worst with each melody. Results indicated that cross-modal melody-to-texture associations were emotionally mediated, because the correlation between the emotional ratings of the music and the emotional MTAs of the chosen textures were so high (Angry/Not-Angry=.79, Calm/Agitated=.91, Active/Passive=.76, Harmonious/Disharmonious=.64). Unlike music-to-color associations, the Happy/Sad correlation (r=.31). was not significant. Melodies-to-texture associations were also mediated by shared geometric features (e.g., Sharp/Smooth=.96, .Curved/Straight=.92, Simple/Complex=.89, Granular/Fibrous=.80). Crucially, the cross-modal melody-to-texture associations corresponded to specific musical features. k-means clustering analyses revealed strong timbre and note-rate effects: cello melodies were paired with straight/sharp/fibrous textures, and piano melodies with curved/smooth/granular textures. Different textures were also chosen for different note-rates, with more granular/separate textures being chosen with slow note-rate melodies, and more fibrous/connected textures being chosen with faster note-rate melodies. These clusters show that the note-rate and timbre of a melody can be inferred reliably from visual texture correspondences alone. Meeting abstract presented at VSS 2015.