ArticlePDF Available

Color preferences change after experience with liked/disliked colored objects



How are color preferences formed, and can they be changed by affective experiences with correspondingly colored objects? We examined these questions by testing whether affectively polarized experiences with images of colored objects would cause changes in color preferences. Such changes are implied by the ecological valence theory (EVT), which posits that color preferences are determined by people's average affective responses to correspondingly colored objects (Palmer & Schloss, Proceedings of the National Academy of Sciences, 107, 8877-8882, 2010). Seeing images of strongly liked (and disliked) red and green objects, therefore, should lead to increased (and decreased) preferences for correspondingly colored red and green color patches. Experiment 1 showed that this crossover interaction did occur, but only if participants were required to evaluate their preferences for the colored objects when they saw them. Experiment 2 showed that these overall changes decreased substantially over a 24-h delay, but the degree to which the effect lasted for individuals covaried with the magnitude of the effects immediately after object exposure. Experiment 3 demonstrated a similar, but weaker, effect of affectively biased changes in color preferences when participants did not see, but only imagined, the colored objects. The overall pattern of results indicated that color preferences are not fixed, but rather are shaped by affective experiences with colored objects. Possible explanations for the observed changes in color preferences were considered in terms of associative learning through evaluative conditioning and/or priming of prior knowledge in memory.
Color preferences change after experience with liked/disliked colored objects
Eli D. Strauss1, Karen B. Schloss2, and Stephen E. Palmer2
1Department of Zoology, Michigan State University
2Department of Psychology, University of California, Berkeley
Corresponding author:
Karen B. Schloss
3210 Tolman Hall
Berkeley, CA 94720
How are color preferences formed, and can they be changed by affective experiences with
correspondingly colored objects? We examined these questions by testing whether affectively
polarized experiences with images of colored objects cause changes in color preferences. Such
changes are implied by the ecological valence theory (EVT), which posits that color preferences
are determined by people’s average affective responses to correspondingly colored objects
(Palmer & Schloss, 2010). Seeing images of strongly liked (and disliked) red and green objects,
therefore, should lead to increased (and decreased) preference for correspondingly colored red
and green color patches. Experiment 1 shows that this cross-over interaction does occur, but
only if participants were required to evaluate their preferences for the colored objects when they
saw them. Experiment 2 shows that these overall changes decreased substantially over a 24-hour
delay, but the degree to which the effect lasted for individuals co-varied with the magnitude of
their effects immediately after object exposure. Experiment 3 demonstrates a similar, but weaker
effect of affectively biased changes in color preferences when participants did not see, but only
imagined the colored objects. The overall pattern of results indicates that color preferences are
not fixed, but rather are shaped by affective experiences with colored objects. Possible
explanations for the observed changes in color preferences are considered in terms of associative
learning through evaluative conditioning and/or priming of prior knowledge in memory.
Color preferences change after experiences with liked/disliked colored objects
What determines aesthetic preferences for colors? Recent theories have suggested several
possibilities. The cone-contrast theory posits that color preferences arise from hardwiring in
early visual processing , the color-emotion theory suggests that they arise from the emotional
content of colors , and the ecological valence theory (EVT) hypothesizes that they result from
people’s combined liking/disliking reactions (valences) to all correspondingly-colored ecological
objects .
Consistent with the EVT, 80% of the variance in adult Americans’ average preferences
for 32 chromatic colors could be predicted from the weighted affective valence estimates
(WAVEs) for those colors (Palmer & Schloss, 2010). The WAVE of a color was defined as the
average of the liking/disliking ratings of all things associated with that color, weighted by the
similarity between the color of the object and the color patch with which it was associated.
Color preferences of British and Japanese participants also showed strong positive correlations
with preferences for correspondingly colored objects.1 With only correlational evidence,
however, the direction of causality is unclear. Do affective experiences with colored objects
drive color preferences, as the EVT proposes, do color preferences determine people’s affect
about objects, or both? Clearly, colors can influence object preferences, as when someone
purchases one T-shirt among many that differ only in color. The primary focus of this article,
however, is whether experiences with colored objects causally influence color preferences, as the
EVT predicts.
1 Taylor and Franklin (2012) found that the number of associated objects for each color explain as much variance in
color preferences as WAVEs do, with more preferred colors being associated with fewer objects. However, a
subsequent study of individual differences shows that individuals’ color preferences are significantly related to
object valences but not to the number of objects they associated with the colors .
There is indirect evidence that people’s preferences for specific color-associated entities
can change their general color preferences. Schloss, Poggesi, and Palmer compared students’
color preferences at two rival universities with strong color associations: the University of
California, Berkeley (blue/gold) and Stanford University (red/white). As the EVT predicts,
students at both universities liked their own school’s colors better than their rivals did, and the
degree to which they did so was correlated with their self-reported amount of “school spirit”
(positive affect) for their university. This pattern supports the idea that preferences for color-
associated things influence preferences for the associated color, because it is highly implausible
that students choose their university and develop their level of school spirit based on how much
they like that university’s colors and dislike its rival’s.
The experiments described below directly tested the hypothesis that color preferences
change as an individual processes positive/negative experiences with correspondingly colored
objects. Each experiment included three phases. First, participants rated their preferences for a
set of colors. Second, as part of “a separate study,” they either saw or imagined 40 colored
objects while performing various tasks. One group saw/imagined positive red objects (e.g.,
strawberries and cherries) and negative green objects (e.g., mold and pond scum), whereas the
other group saw/imagined positive green objects (e.g., ripe kiwifruit and healthy trees) and
negative red objects (e.g., open wounds and sores). Finally, participants rated their preferences
for the same colors again. We assessed reliable increases/decreases in color preference due to
experience with positive/negative objects.
Experiment 1: Effects of Object Exposure on Color Preferences
We attempted to alter color preferences by showing participants affectively biased
images of 40 colored objects in a three-phase experiment. After rating their preferences for 37
colors (to provide a baseline), participants were randomly assigned to one of two groups. The
G+R- group saw 10 images of positive green objects (e.g., ripe kiwi, spring foliage), 10 images
of negative red objects (e.g., open wounds, an infected eye), and 20 images of relatively neutral
objects of other colors (e.g., a screwdriver, a ladder). The R+G- group saw 10 images of positive
red objects (e.g., raspberries, roses), 10 images of negative green objects (e.g., pond scum, moldy
food), and the same 20 images of other-colored neutral objects. Finally, all participants rated
their color preferences again to measure changes caused by viewing the colored objects.
Participants were told that the first and third parts were research on color aesthetics, whereas the
second part (as well as a previous initial study) was research on spatial aesthetics. Different
experimenters ran the color and spatial studies to reinforce their separation.
In Experiment 1A, participants saw each image of each colored object in four “spatial”
tasks: verbal-label fit, center-localization, complexity judgment, and object preference rating. In
Experiment 1B, they completed the first three “spatial” tasks but did not rate object preferences.
By comparing the results we could determine whether affective judgments were necessary to
elicit changes in preference.
Participants. Results are reported for 46 participants (28 females) in Experiment 1A and
46 in Experiment 1B (34 females), divided into two equal groups for the image-exposure phase.
Participants in all three experiments had normal color vision (Dvorine Pseudo-Isochromatic
Plates) and gave informed consent. The UC Berkeley Committee for the Protection of Human
Subjects approved the protocol.
Design, Displays, and Procedure. To make the experiment seem like an assortment of
different experiments, subjects first completed a short, unrelated spatial aesthetics task. They
then completed the following three phases of the present experiment.
Phase 1: Initial Color Preference. The colors consisted of eight hues
(red/orange/yellow/chartreuse/green/cyan/blue/purple) at four different saturation/lightness
levels (saturated/light/muted/dark), plus five achromatic colors (white/light-gray/medium-
gray/dark-gray/black) (see Table A1 for CIE 1931 xyY and Munsell coordinates). The colors
were presented on a gray background (CIE x=0.312/y=0.318/Y=19.26) and were viewed from a
distance of approximately 60cm. The iMac computer monitor (1680x1050 px, 39.58°x25.36°)
was calibrated using a Minolta CS100 Chroma Meter.
The colored squares (100x100px, 2.5°x2.5°) were presented singly in random order,
centered on the monitor. Participants indicated how much they liked each color on a line-mark
response-scale by sliding the cursor to the appropriate position and clicking the mouse. The 400-
px response-scale at the bottom of the screen was labeled “not at all” at the left end and “very
much” at the right end, with a tick-mark at the center indicating neutrality. Responses were
rescaled from –100 to +100.
Participants first completed an “anchoring task” (Palmer & Schloss, 2010) to scale their
preferences in the present context. They viewed all 37 colors simultaneously and were asked to
point to the colors they liked most/least. They were instructed to rate the color they liked most as
“very-much” and the color they liked least as “not-at-all.”
Participants then rated all colors once in each of two blocks. Each color was presented
until a response was made, and the next trial began 500-ms later. We calculated the correlation
between their ratings for corresponding colors in Blocks 1 and 2. If it was less than +.70, their
performance on the color preference task was considered insufficiently reliable. . Thirteen
participants from Experiment 1A and 19 from Experiment 1B were thereby excluded and
provided no data on the tasks described below, but still received full credit for participating. The
cutoff value of +.70 was chosen (somewhat arbitrarily) to ensure that participants had
sufficiently stable baseline preferences to make detection of post-treatment changes likely. Only
including participants with stable preferences helped ensure that observed changes in preference
in the experiment are due to the experimental treatment.
Phase 2: Image Exposure. Participants completed several “spatial tasks” to experience
the colored objects. These tasks, along with the initial spatial aesthetics task (see above), were
administered by a second experimenter to reinforce the cover story that they were unrelated to
the color preference task. All participants in both Experiments 1A and 1B completed the
following judgments: verbal-label fit, center-localization, and complexity. Only Experiment 1A
participants also completed the final object preference task.
In each task, participants saw 40 images of objects in random order: 10 green, 10 red, and
20 objects of other colors (Figure B1). Only the red and green objects were specific to the object
exposure group (R+G- or G+R-).
In the verbal-label fit task, images were presented with labels describing the content of
the image. Subjects were asked to categorize each image as either “well labeled” or “poorly
labeled” using the right and left arrow keys (respectively). This task was designed to inform
participants indirectly of what object(s) were depicted in the images. The labels were always
reasonably good for the critical red and green objects (e.g., “strawberries” for strawberries and
“pond scum” for pond scum). They were sometimes non-specific for the other colored objects
(e.g., “tool” for a screwdriver), but were never misleading. In the center-localization task,
participants were asked to move the cursor to the center of the focal object(s) and click the
mouse. In the complexity judgment task, participants were asked to rate each image’s complexity
on a line-mark scale from “very-simple” (left end point) to “very-complex” (right end point).
Participants in Experiment 1A then completed the object preference task in which they
saw each image and were asked to rate how much they liked the object(s) on a line-mark scale
from “very-little” (left endpoint) to and “very-much” (right endpoint). This task was excluded
from Experiment 1B to determine its influence on preference changes for corresponding colors.
Phase-3: Post-Exposure Color Preference. Phase-3 was the same as phase-1 except
participants were told that some colors were different. Two additional colors (bright-orange and
lavender) were added.
After the experiment was completed, participants were asked to guess its purpose. If they
mentioned changing color preferences by exposing them to positive/negative colored objects,
their data was excluded from the analyses. Two participants did so in Experiment 1A and five in
Experiment 1B, none whose data are included in the results reported below.
Results and Discussion
An initial pilot study was conducted to determine which colors to examine for changes in
preference after image exposure. Five other participants were shown each of the 40 object
images (Figure B1) and asked to pick the color most similar to the dominant color in the image
from among the 37 colors. Saturated Red (SR) and Dark Red (DR) (henceforth referred to as
“reds”) were most frequently selected for the red images and Saturated Chartreuse (SH) and
Dark Chartreuse (DH) (henceforth referred to as “greens”) were selected for the green images.
We measured how participants’ color preferences changed after image exposure by subtracting
their average preferences for reds and greens in phase-1 from their average preferences in phase-
3. The average change in preference for the other 33 colors were considered controls.
Figure 1. Changes in preference for reds (circles) and greens (squares) and the average of
the other colors (triangles) as function of object exposure group in Experiment 1A after
rating obejct preferences (A), and those in Experiment 1B, who did not rate object
preferences (B). Error bars represent the standard errors of the means (SEMs) of the
difference scores.
The results (Figure 1) show a three-way interaction between change in preference
(red/green), image-exposure group (R+G-/G+R-), and inclusion of the object preference task
(Experiment-1A/Experiment-1B) (F(1,88)=5.13, p<.05, η2=.06). The participants who
completed the object preference task (Experiment 1A) showed a significant interaction2 between
image-exposure group (R+G-/G+R-) and preference change (red/green) (F(1,44)=7.81, p<.01,
η2=.15) (Figure 1A). Those in the R+G- group showed an increase in preference for red over
2 All subsequent statistics are one tailed (unless otherwise specified) because the EVT predicts the direction of the
relevant effects.
green (F(1,22)=3.25, p<.05, η2=.13), and those in the G+R- group showed an increase in
preference for green over red (F(1,22)=7.17, p<.01, η2=.25). Tests against zero (no change) and
against averages of all other colors indicated that preference for the colors of positive objects
reliably increased (R+G-: F(1,22)=2.97, 3.48, p<.05, η2=.12, .14, respectively; G+R-:
F(1,22)=6.07, 4.71, p< .05, η2=.22, .18). The predicted negative trends in preference for colors of
negative objects were present, but not statistically reliable (Fs<1). Although the cause of this
asymmetry is unclear, it is not due to a floor effect in red/green preferences preventing decreases.
In the initial color preference task, mean preference for the reds was greater than for the greens
(reds=24.8, greens=-22.9, t(54)=6.32, p<.001, two tailed), but both were much more preferred
than the least-liked colors (e.g., dark-yellow=-52.87, dark-orange=-63.94). Participants may have
attended less to the unpleasant images in the experiment to minimize negative emotions, but such
an interpretation requires further study.
The EVT also predicts that the degree to which an individual’s color preferences will
increase/decrease due to object exposure will depend on how strongly he/she likes/dislikes the
objects. We therefore correlated the difference between individuals’ preference ratings in
Experiment 1A for the red/green objects (Ro-Go) with the difference between their change in
preferences for the relevant red/green colors (Rc-Gc). This reliable correlation (r=+.41, p<.01)
further supports the EVT, which predicts that the more strongly people prefer the liked red
objects (R+) to the disliked green objects (G-), the more their preference for red should increase
relative to green, and vice versa.
Participants who did not rate object preferences (Experiment 1B) (Figure 1B) did not
show the color x image exposure group interaction (F<1) of Experiment 1A. Perhaps when not
explicitly evaluating the objects, participants were not affectively engaged with the stimuli, in
which case the EVT implies no change in preference for the color of the object. Another
possibility is that the absence of the preference task simply reduced the number of exposures. It
seems unlikely that one additional viewing of each image would have such a dramatic effect,
however, especially given that a similar cross-over interaction arises in Experiment 3 when
objects were imagined only twice.
In Experiment 1B participants did show an unexpected main effect in preference change
for red vs. green (F(1,44)=5.32, p<.05, two-tailed, η2=.11, ), but this effect was not reliable
within either of the object exposure groups alone (R+G-: F(1,22)=2.57, p=.11, η2=.11; G+R-:
F(1,22)=2.80, p=.11, η2=.11, two-tailed). Given that red was initially more preferred than green,
the main effect of color may be due to mere exposure effects, where additional exposures to an
affectively polarized stimulus further polarizes attitudes towards it . The main effect of group in
Figure 1B was not significant (F(1,44)=1.91, p > .05 η2=.04).
Experiment 2: Effects of a Post-exposure 24-hour Delay on Color Preferences
In Experiment 1 people’s preferences for specific colors changed after experiencing and
rating preference for affectively biased images of correspondingly colored objects, when there
was minimal delay between the end of the image-exposure phase and the start of the post-
exposure color-preference task. Here we measured the temporal duration of this change in color
preference by testing both immediately and 24 hours later.
Participants. Results are reported from 67 participants (48 female), with 34 in the R+G
and 33 in the G+R- group. An additional seventeen participants tested in the initial color-
preference task did not complete the experiment because their Block-1/Block-2 correlation in the
initial color-preference task was less than .7. The data from two additional participants were
excluded because they correctly discerned the purpose of the manipulation.
Design, Displays, and Procedure. All methods were identical to Experiment 1A with
the addition of a second post-exposure color-preference task approximately 24 hours after the
initial one.
Results and Discussion
The immediate retest replicated the interaction in Experiment 1A between preference
change and object-exposure group (F(1,65)=5.18, p<.05, η2=.07). We replicated the within-group
differences between changes in red and green preferences in the R+G- group (F(1,33)=3.75,
p<.05, η2=.10), but the corresponding difference in the G+R- group was only marginally reliable
(F(1,32)=1.67 p = .10, η2=.05) (see Figure 2A). After a 24-hour delay, however, there was no
corresponding interaction (F<1) or within-group differences (F<1 for both groups), indicating
that the overall effect of image exposures is short-lived (Figure 2B). However, an analysis of
individual differences showed that the degree to which individuals changed their preferences on
the immediate test was predictive of the degree to which the effect was evident at the longer
delay (Table 1), suggesting that stronger effects of object exposure are more likely to last longer.
Females show this pattern more reliably than males, but it is possible that this effect is influenced
by differences in sample sizes (48 females, 19 males).
Figure 2. Changes in preferences for reds (circles) and greens (squares) and the average
of the other colors (triangles) as function of object-exposure group after no-delay (A) and
a 24-hour delay (B). Error bars represent the standard errors of the means (SEMs).
Table 1. Correlations between individuals’ changes in preference for red/green after no-
delay vs. a 24-hour delay, separated by group and gender.
Red Green
R+G- .58*** .55***
G+R- .69*** .56***
Male .58** .41
Female .60*** .55***
**p<.01, ***p<.001, two tailed
Experiment 3: Effects of Imagining Objects on Color Preferences
In Experiments 1 and 2 people’s color preferences changed systematically in the direction
predicted by the EVT after viewing an affectively biased series of photographs of colored
objects. In Experiment 3 we studied whether similar changes can be induced by people
imagining similarly biased sets of colored objects. The theoretical question is whether stimulus-
driven sensory color experiences are necessary for affect about the colored objects to change
color preferences, or whether merely activating preexisting affective knowledge of perceptual
properties is sufficient.
Participants. Results are reported from 56 participants (42 female), with 28 in each
object exposure group. Seventeen participants did not complete the experiment because their
Block-1/Block-2 correlation in the initial color-preference task was less than .7. The data from
six additional participants were excluded from the analyses because they discerned the purpose
of the manipulation.
Design, Display, Procedure. The design, displays, and procedures were the same as in
Experiment 1A, except that participants completed the mental imagery tasks described below in
Phase-2: Mental Imagery. Participants completed three mental imagery tasks: color
matching, imagery strength, and object preference. In the color-matching task, participants were
presented with group-specific object descriptions (Table B2) displayed as black text on the gray
background of the monitor. They had 2-sec to form a mental image of the object, after which the
array of 37 colors appeared. Their task was to click on the color that best matched the color in
their mental image using the mouse. This task required participants to attend to the characteristic
color of the described objects. The imagery-strength and object-preference tasks were then
administered as one experiment. Participants imagined each object for 2-sec, then rated the
strength of their mental image on a line-mark scale from “very-vague” to “very-vivid,” and how
much they liked the object in their mental image on a scale from “not-at-all” to “very-much.”
Results and Discussion
As in Experiments 1A and 2, there was a reliable interaction between color and object-
exposure group (F(1,54)=7.66 p<.01, η2=.12). R+G- participants showed a significant increase
in preference for red over green (F(1,27)=5.64, p<.05, η2=.17), and G+R- participants showed an
analogous trend that was nearly significant (F(1,27) = 2.11, p =.08, η2=.07) (Figure 3). These
results are comparable to those from Experiment 1A, as indicated by no experiment x color x
exposure-group interaction (F<1) and similar color x exposure group interaction effect sizes
(Experiment 1A: η2=.15; Experiment 3: η2=.12).
Figure 3. Changes in preferences for reds (circles) and greens (squares) and the average
of the other colors (triangles) as function of imagined object exposure group. Error bars
represent the standard errors of the means (SEMs).
These findings suggest that strongly color associated mental images, memories, feelings,
and concepts can influence preferences for colors, much like sensory experiences of depicted
objects do. Such changes would not occur unless people had previous knowledge of the objects’
colors, so the most relevant factor is likely the temporally contiguous activation of affective
responses and color associations to the same objects. This is the case in Experiments 1A, 2, and
3, but not in Experiment 1B where affective information may not have been activated without the
explicit affective rating task.
General Discussion
The present results support the Ecological Valence Theory’s (EVT) proposal that color
preferences are determined, at least in part, by object preferences. Experimentally controlled
exposure of pictures of colored objects influenced people’s preference for abstract patches of
color provided that they simultaneously evaluated their affect for those objects. In particular,
exposure to liked/disliked red and green objects led to corresponding changes in preferences for
red and green color patches, and the strength of these changes was correlated with the magnitude
of differences in preferences for the corresponding objects (Experiment 1A). This interaction
was absent when participants did not evaluate their object preferences during the exposure phase
(Experiment 1B), suggesting that affective evaluation of the colored objects is crucial for object
exposure to influence color preference in this paradigm. Experiment 2 replicated the interaction
from Experiment 1A in immediate testing, but not after a 24 hour delay. However, the degree to
which an individual’s change in color preference lasted overnight reliably co-varied with the
strength of the change following immediate image exposure. In Experiment 3 a similar
interaction was evident in color-preference changes after forming mental images of affectively
biased sets of colored objects. Thus, activating preexisting knowledge of perceptual and affective
properties of liked/disliked objects, without sensory experience, is sufficient to change color
How are these effects to be understood? One possibility is that viewing/imagining
colored objects produces incremental strengthening of existing associations. This learning
account is analogous to evaluative conditioning, which may be a form of classical (Pavlovian)
conditioning in which the conditioned response is a preference judgment rather than an explicit
behavior . In the context of the present study, preferences for non-color attributes (such as the
positive/negative taste) of colored objects (unconditioned stimuli) would be transferred to the
colors (conditioned stimuli) associated with those objects to produce a change in color
preference (conditioned response). Because our participants already knew both the object-color
associations (e.g., ripe strawberries are red) and the corresponding affective judgments (e.g., “I
love ripe strawberries”), it is implausible that the results are due to newly learned associations.
Rather, color preferences may be updated after new experiences with previously known colored
objects by strengthening pre-existing associations. This account suggests that such preference
changes should be relatively durable, however, contrary to their 24-hour extinction in
Experiment 2, unless one also assumes that recent updates to evaluative associations decay over
Another possibility is that experiences of viewing/imagining colored objects while
considering their affective values activate (i.e., prime) this pre-existing knowledge. Then, when
asked to rate color preferences, people compute them “on-the-fly” by weighting the relevant
information (e.g., object preferences) according to their current memory strength. In this case,
no long-term effects would be expected, because increased strength of the relevant knowledge in
memory would presumably subside rapidly after its recent activation, consistent with the results
of Experiment 2. The fact that comparable changes in color preferences were obtained following
the mental imagery tasks in Experiment 3, which presumably activated people’s prior knowledge
about object colors and valences, is also consistent with this priming account. Even if the present
effects are due to priming, however, the initial learning of the associations may be due to some
form of evaluative conditioning. Further investigation will be required to determine the
mechanisms involved and the circumstances under which they operate.
In any case, the present results establish a causal connection between people’s
preferences for an abstract patch of color and their preferences for objects that are
characteristically that color. This finding represents further support for the EVT, beyond prior
correlational evidence (e.g., Palmer & Schloss, 2010; Schloss, Poggesi & Palmer, 2011), but it
leaves open several important questions about how preferences are shaped by experience.
Among the most important is whether different objects are weighted differentially in determining
one’s preference for a particular color. Potentially relevant factors include an object’s
importance in one’s life, its perceptual salience, the frequency with which it is experienced, and
the strength of the affective response it causes. Substantial additional research will be necessary
to understand the precise mechanisms that link people’s experiences of colored objects to their
preferences for the corresponding colors.
The authors thank two anonymous reviewers and Rosa Poggesi, Thomas Langlois, Will
Griscom, Mieke Leyssen, Sarah Linsen, Ania Jaroszewicz, Jessica Jimenez, Lily Lin,
Chistopher Lau, Mathilde Heinemann, Madison Zeller, Arielle Younger, Meghna Dholakia,
Jackson Jewett, Stephen Guo, Kelly Whiteford, Saki Wang, and Madeline McComb. The project
was supported by National Science Foundation Grants Nos. 1059088 and 0745820 and a Google
Gift to S.E.P. Any opinions, findings, and conclusions or recommendations expressed in this
material are those of the author(s) and do not necessarily reflect the views of the National
Science Foundation.
Brickman, P., Redfield, J., Harrison, A. A., & Crandall, R. (1972). Drive and predisposition as
factors in the attitudinal effects of mere exposure. Journal of Experimental Social
Psychology, 8(1), 31-44.
De Houwer, J., Baeyens, F., & Field, A. P. (2005). Associative learning of likes and dislikes:
Some current controversies and possible ways forward. Cognition & Emotion, 19(2),
De Houwer, J., Thomas, S., & Baeyens, F. (2001). Association learning of likes and dislikes: A
review of 25 years of research on human evaluative conditioning. Psychological Bulletin,
127(6), 853.
Hofmann, W., De Houwer, J., Perugini, M., Baeyens, F., & Crombez, G. (2010). Evaluative
conditioning in humans: a meta-analysis. Psychological Bulletin, 136(3), 390- 421.
Hurlbert, A. C., & Ling, Y. (2007). Biological components of sex differences in color preference.
Current Biology, 17(16), 623-625. doi: 10.1016/j.cub.2007.06.022
Ling, Y., Hurlbert, A. C., & Robinson, L. (2006). Sex differences in colour prefernce. In N. J.
Pitchford & C. P. Biggam (Eds.), Progress in Colour Studies 2: Cognition. Amsterdam:
John Benjamins.
Ou, L.-C., Luo, M. R., Woodcock, A., & Wright, A. (2004). A study of colour emotion and
colour preference. Part III: Colour preference modeling. Color Research & Application,
29(5), 381-389.
Palmer, S. E., & Schloss, K. B. (2010). An ecological valence theory of human color preference.
Proceedings of the National Academy of Sciences, 107(19), 8877-8882.
Palmer, S. E., Schloss, K. B., & Hawthorne-Madell, D. (under review). Ecological influences on
individual differences in color preference.
Schloss, K. B., Poggesi, R. M., & Palmer, S. E. (2011). Effects of university affiliation and
"school spirit" on color preferences: Berkeley versus Stanford. Psychonomic Bulletin &
Review, 18, 498-504.
Schloss, K. B., Strauss, E. D., & Palmer, S. E. (2012). Object color preferences. Color Research
& Application.
Taylor, C., & Franklin, A. (2012). The relationship between color–object associations and color
preference: Further investigation of ecological valence theory. Psychonomic bulletin &
review, 19(2), 190-197. doi: 10.3758/s13423-012-0222-1
Yokosawa, K., Schloss, K. B., Asano, M., & Palmer, S. E. (in preparation). Cross-cultural
studies of color preferences: US and Japan.
Appendix A
Table A1. CIE 1931 values and Munsell values for the 32 chromatic colors and CIE
1931 values for the five achromatic colors (CIE Illuminant C), table from .
Color x y Y Hue Value/Chroma
Saturated 0.549 0.31
322.93 5 R 5/15
Light 0.407 0.32
649.95 5 R 7/8
Muted 0.441 0.32
422.93 5 R 5/8
Dark 0.506 0.31
17.60 5 R 3/8
Saturated 0.513 0.41
249.95 5 YR 7/13
Light 0.399 0.36
668.56 5 YR 8/6
Muted 0.423 0.37
534.86 5 YR 6/6
Dark 0.481 0.38
810.76 5 YR 3.5/6
Saturated 0.446 0.47
291.25 5 Y 9/12
Light 0.391 0.41
391.25 5 Y 9/6.5
Muted 0.407 0.42
649.95 5 Y 7/6.5
Dark 0.437 0.45
018.43 5 Y 5/6.5
Saturated 0.387 0.50
468.56 5 GY 8/11
Light 0.357 0.42
079.90 5 GY 8.5/6
Muted 0.360 0.43
642.40 5 GY 6.5/6
Dark 0.369 0.47
318.43 5 GY 4.5/6
Green Saturated 0.254 0.44
942.40 3.75 G 6.5/11.5
Light 0.288 0.38 63.90 3.75 G 7.75/6.25
Muted 0.281 0.39
234.86 3.75 G 6/6.25
Dark 0.261 0.41
912.34 3.75 G 3.75/6.25
Saturated 0.226 0.33
549.95 5 BG 7/9
Light 0.267 0.33
068.56 5 BG 8/5
Muted 0.254 0.32
834.86 5 BG 6/5
Dark 0.233 0.32
413.92 5 BG 4/5
Saturated 0.200 0.23
034.86 10 B 6/10
Light 0.255 0.27
859.25 10 B 7.5/5.5
Muted 0.241 0.26
528.90 10 B 5.5/5.5
Dark 0.212 0.23
610.76 10 B 3.5/5.5
Saturated 0.272 0.15
618.43 5 P 4.5/17
Light 0.290 0.24
249.95 5 P 7/9
Muted 0.287 0.22
222.93 5 P 5/9
Dark 0.280 0.18
17.60 5 P 3/9
Black 0.310 0.31
Dark gray 0.310 0.31
Med Gray 0.310 0.31
Light Gray 0.310 0.31
White 0.310 0.31
Table B1. Verbal object descriptions provided for the R+G- G+R- object exposure
groups in Experiment 3.
R+G- Group: G+R- Group: Both Groups:
R+ Objects G+ Obejcts Neutral Objects
strawberries parakeet power outlet
salsa margarita school bus
raspberries mint chocolate chip ice cream denim
pomegranate kiwi grape soda
ruby granny smith apple tires
rose four-leaved clover eggplant
cranberries pine tree Barney the dinosaur
grapefruit forest pencil
cherries $100 bill printer paper
wine avacado post-it notes
G- Objects R- Objects paper clip
booger big zit Navy uniform
mold rug burn concrete
pond scum surgery cardboard
slime deep cut desert
puke pink-eye bulldozer
snot rotten tomato caution sign
pus bloody nose taxi
bile chicken-pox band-aid
radioactive waste bloody tampon
weeds scab
Figure B1. Colored images shown to the R+G- and G+R- object exposure groups in
Experiments 1 and 2. The control images were shown to both groups.
... Another theory suggests that color preferences are likely to be associated with one's mood at that moment, regardless of the object (10). Despite publications claiming that color preference is a systemic response originating from a specific part of the brain, numerous studies revealed that people's preference for certain colors reflect their moods at the moment in question (2,(17)(18)(19)(20)(21). This provided the foundation for subsequent studies investigating the effects of color preferences on cognitive and emotional judgments in humans, and the origins of such effects. ...
... Despite the remaining open questions of which subjective (top-down) and objective (bottom-up) features exactly drive (interindividual) differences in empirical aesthetics, consistent response patterns were found and attributed to certain aesthetic primitives. Stimulus properties such as contour shape (Bar and Neta, 2007;Vartanian et al., 2013), color Strauss et al., 2013;Elliot and Maier, 2014), as well as symmetry (Tyler, 2003;Bertamini et al., 2018Bertamini et al., , 2019, order, complexity (Nadal et al., 2010;Van Geert and Wagemans, 2021), and global image properties (e.g., fractality) were proposed as objective predictors of aesthetic preference (Chamberlain, 2022). However, other approaches stress the idiosyncrasies of preferences, demonstrating a stronger shared taste for natural or naturally inspired aesthetic domains as opposed to artifacts of human culture (Vessel et al., 2018). ...
Full-text available
The interest in the response to contours has recently re-emerged, with various studies suggesting a universal preference for curved over angular stimuli. Although no consensus has yet been reached on the reasons for this preference, similar effects have been proposed in interior environments. However, the scarcely available research primarily depends on schematic or unmatched stimuli and faces heterogeneity in the reported results. In a within-subject design, we investigated the claimed contour effect in photo-realistic indoor environments using stimulus material previously tested in virtual reality (VR). A total of 198 online participants rated 20 living room images, exclusively manipulated on the contours (angular vs. curved) and style (modern vs. classic) levels. The scales represented aesthetic (beauty and liking) and stress (rest and stress) responses. Beyond our main focus on contours, we additionally examined style and sex effects to account for potential interactions. Results revealed a significant main effect of contours on both aesthetic (η 2 g = 1-2%) and stress (η 2 g = 8-12%) ratings. As expected, images of curved (vs. angular) contours scored higher on beauty, liking, and rest scales, and lower on stress. Regarding interactions with style, curvature was aesthetically preferred over angularity only within images depicting modern interiors, however, its positive effect on stress responses remained significant irrespective of style. Furthermore, we observed sex differences in aesthetic but not in stress evaluations, with curvature preference only found in participants who indicated female as their sex. In sum, our study primarily confirms positive effects of curvature, however, with multiple layers. First, the impact on aesthetic preference seems to be influenced by individual and contextual factors. Second, in terms of stress responses, which might be especially relevant for designs intended to promote mental-health, the consistent effects suggest a more generalizable, potentially biophilic characteristic of curves. To the best of our knowledge, this is the first study to demonstrate these effects in fully-matched, photo-realistic, and multi-perspective interior design stimuli. From the background of a previous VR trial from our research group, whereby the same rooms did not elicit any differences, our findings propose that static vs. immersive presentations might yield different results in the response to contours.
... These differences might reflect differences in the anthropological, ethnic, and religious identity, which are extremely difficult to grasp empirically and are rarely taken into account in the experimental protocols (e.g., the influence of Confucianism in China vs capitalism/individualism in north America). Moreover, preference has been regarded as a problematic concept to be empirically investigated in the aesthetic domain, likely resulting from a complex combination of several universal and individual factors, such as aesthetic pleasure, aesthetic interest, ease of processing, affective mediation, and familiarity with the stimuli (e.g., [85,218]; see also the recent paper by Chen et al. [48], on "taste typicality"). ...
Revealed more than two millennia ago by Pythagoras, consonance and dissonance (C/D) are foundational concepts in music theory, perception, and aesthetics. The search for the biological, acoustical, and cultural factors that affect C/D perception has resulted in descriptive accounts inspired by arithmetic, musicological, psychoacoustical or neurobiological frameworks without reaching a consensus. Here, we review the key historical sources and modern multidisciplinary findings on C/D and integrate them into three main hypotheses: the vocal similarity hypothesis (VSH), the psychocultural hypothesis (PH), and the sensorimotor hypothesis (SH). By illustrating the hypotheses-related findings, we highlight their major conceptual, methodological, and terminological shortcomings. Trying to provide a unitary framework for C/D understanding, we put together multidisciplinary research on human and animal vocalizations, which converges to suggest that auditory roughness is associated with distress/danger and, therefore, elicits defensive behavioral reactions and neural responses that indicate aversion. We therefore stress the primacy of vocality and roughness as key factors in the explanation of C/D phenomenon, and we explore the (neuro)biological underpinnings of the attraction-aversion mechanisms that are triggered by C/D stimuli. Based on the reviewed evidence, while the aversive nature of dissonance appears as solidly rooted in the multidisciplinary findings, the attractive nature of consonance remains a somewhat speculative claim that needs further investigation. Finally, we outline future directions for empirical research in C/D, especially regarding cross-modal and cross-cultural approaches.
... People have associations between colours and concepts, which influence a wide variety of judgments in visual cognition. Colour-concept associations affect object recognition (Macario, 1991;Nagai & Yokosawa, 2003;Ostergaard & Davidoff, 1985;Tanaka & Presnell, 1999;Wurm et al., 1993), colour perception (Delk & Fillenbaum, 1965;Hansen et al., 2006;Olkkonen et al., 2008;Witzel, 2016;see Valenti & Firestone, 2019 for contrary evidence), perceptual experiences in other modalities (e.g., flavour) (Piqueras-Fiszman & Spence, 2012;Velasco et al., 2014), colour preferences (Palmer & Schloss, 2010;Schloss & Palmer, 2017;Strauss et al., 2013;Taylor & Franklin, 2012), visual reasoning with information visualizations (Lin et al., 2013;Schloss et al., 2018;Schloss et al., 2019), and interpretations of other people's emotions (Thorstenson et al., 2018). Thus, to fully understand visual cognition, it is necessary to understand the nature of colour-concept associations. ...
Full-text available
Colour-concept associations influence fundamental processes in cognition and perception, including object recognition and visual reasoning. To understand these effects, it is necessary to understand how colour-concept associations are formed. It is assumed that colour-concept associations are learned through experiences, but questions remain concerning how association formation is influenced by properties of the input and cognitive factors. We addressed these questions by first exposing participants to colour-concept co-occurrences for novel concepts (“Filk” and “Slub” alien species) using a category learning task. We then assessed colour-concept associations using an association rating task. During alien category learning, colour was a noisy cue and shape was 100% diagnostic of category membership, so participants could ignore colour to complete the task. Nonetheless, participants learned systematic colour-concept associations for “seen” colours during alien category learning and generalized to “unseen” colours as a function of colour distance from the seen colours (Experiment 1). Association formation not only depended on colour-concept co-occurrences during alien category learning, but also on cognitive structure of colour categories (e.g., degree to which an observed red colour is typical of the colour category “red”) (Experiment 2). Thus, environmental and cognitive factors combine to influence colour-concept associations formed from experiences in the world.
Human males and females show average gender/sex differences for certain psychological phenomena. Multiple factors may contribute to these differences, including sex chromosomes, exposure to gonadal hormones, and socialization or learning. This study investigated potential hormonal and socialization/learning influences on gender/sex differences in childhood preferences for color, specifically pink and red vs. blues, and for toys. Children (aged 4 to 11 years) with congenital adrenal hyperplasia (CAH, n = 43 girls and 37 boys), marked by elevated prenatal adrenal androgen exposure, and without CAH (n = 41 girls and 31 boys) were studied. Prior research indicates girls with CAH are masculinized for certain behaviors, such as toy choices, while boys with CAH generally do not differ from boys without CAH. In the current study, children indicated preferences for stereotyped hues of pink vs. blue as well as two control color pairs. They also indicated their preference between gender/sex-typed toys (doll vs. car) presented in black and white, in gender/sex-congruent colors (pink doll vs. blue car) and in gender/sex-incongruent colors (pink car vs. blue doll). Color findings: Control girls preferred stereotyped pink over blue more than boys or girls with CAH did; the latter two groups did not differ in their color preferences. No preference differences occurred for other color pairs. Toy findings: In black/white or gender/sex-congruent colors, boys preferred the car more than control girls or girls with CAH did, while girls with CAH preferred the car more than control girls did. In gender/sex-incongruent colors (pink car vs. blue doll), boys still preferred the car, while girls with CAH showed reduced and control girls showed increased preferences for the pink car compared to the car preferences in black/white. Results support learning theories of color preferences, perhaps also influenced by pre-existing toy preferences which may occur for other reasons, including early androgen exposure. Specifically, girls with CAH may have learned they do not enjoy stereotypical toys for girls, often colored pink, and pink coloring may subsequently diminish their preference for a car. Our results highlight the importance of gonadal hormones and learning in the development of childhood toy and color preferences.
Full-text available
Understanding the colour preference in microgravity environments will enable better design of future spacecraft and extra-terrestrial environments. In this study, a space station’s crew cabin was simulated and evaluated in 33 different colours by 55 participants using a standard body position change methodology in controlled conditions. Three body positions were tested, and included normal sitting position (SP), to reflect terrestrial conditions; − 15◦ head-down (HD) bed rest, to simulate a microgravity state; and 9.6◦ head-up tilt (HU) bed rest, to simulate a lunar gravity state. VR devices were worn by participants to ensure an immersive environment in which to evaluate the different coloured environments across the three different body positions. The results show that in all colour environments, there was no significant difference in the colour preference between SP and HU, but there was a significant change in the colour preference in HD compared to SP and HU. In the three positions, the participants appeared to prefer lighter colours rather than darker ones, warmer colours in ST and HU, and cool colours in HD. In this short-term simulation study, no evidence of gender differences in colour preference by body position was found, but in each body position, men and women had different preference levels for basic hue and chroma brightness.
According to the ecological valence theory (Palmer & Schloss, 2010), color preferences derive from the emotional valences of color-related objects. Color, translucency, and sweetness are regarded as sensory cues that aid in the identification of ripe fruit. After conducting a literature review that demonstrated humans’ innate preference for sweetness and sugar, we posited that the preference for transparent and translucent colors might be associated with the identification of ripe fruits as sources of sugar. A total of 200 respondents provided preference and appetite ratings for 18 stimuli comprising six colors with three different levels of transparency. The respondents identified the tastes they expected the stimuli to have. The results were recorded for both men (29.5%) and women (70.5%). Overall, we found that the transparent and translucent color stimuli received higher average preference and appetite ratings than the opaque stimuli. Notably, our analysis showed correlations between respondents’ preferences and their appetite ratings for translucent red, yellow-green, and purple, the colors the respondents associated sweetness with. This study highlights the potential association between preferences for translucent colors and identification of sweet-tasting foods.
The main perceptual-cognitive limitations of CDOs (Colour Deficient Observers) are analysed, along with the uses and limitations of tools that either transform images so that CNOs (Colour Normal Observers) see them as CDOs (simulation) or transform images so that CDOs can use them as CNOs (daltonization). The four main uses of colour (comparative, denotative, connotative, and aesthetic) are analysed, along with their relation to, alternatively, the ability to discriminate colour stimuli or to categorize colours. These uses of colour are applied to analyse the possible effects of daltonization tools.
With the population aging in Taiwan, it is projected that elder care robots incorporating smart technologies will play an essential role in ambient assisted living. This research has two purposes: (1) to investigate whether older adults’ color emotion associations and color preferences for robot appearance affect their perceptual judgments, and (2) to explore gender differences in this judgment. Phase I of this research uses a questionnaire to investigate 91 participants’ preferences for robot style and their emotional trigger words for the role of robots. Phase II experiments on another 60 older adults to identify whether their color emotion associations and color preferences affect their perceptual judgments. The research results show that, regardless of gender difference, participating older adults prefer a robot with non‐human‐like features. The results also show that there is no significant difference between males and females in terms of the effect of color emotion association on a robot’s appearance. Older adults tend to associate warm colors with emotional semantics, such as friendly, comfortable, reassuring, gentle, and lively. Preferred colors include red, white, and yellow. Black and grey are almost not preferred by older adults. There are significant differences in the preference for gender for colors white and purple. Older females prefer purple more while it is white for older males. For the rest colors, there exist no significant differences between males and females. Color attributes do not have any effect on color emotion association, whereas color preference is highly positively correlated with b*.
Full-text available
In this study three colour preference models for single colours were developed. The first model was developed on the basis of the colour emotions, clean–dirty, tense–relaxed, and heavy–light. In this model colour preference was found affected most by the emotional feeling “clean.” The second model was developed on the basis of the three colour-emotion factors identified in Part I, colour activity, colour weight, and colour heat. By combining this model with the colour-science-based formulae of these three factors, which have been developed in Part I, one can predict colour preference of a test colour from its colour-appearance attributes. The third colour preference model was directly developed from colour-appearance attributes. In this model colour preference is determined by the colour difference between a test colour and the reference colour (L*, a*, b*) = (50, −8, 30). The above approaches to modeling single-colour preference were also adopted in modeling colour preference for colour combinations. The results show that it was difficult to predict colour-combination preference by colour emotions only. This study also clarifies the relationship between colour preference and colour harmony. The results show that although colour preference is strongly correlated with colour harmony, there are still colours of which the two scales disagree with each other. © 2004 Wiley Periodicals, Inc. Col Res Appl, 29, 381–389, 2004; Published online in Wiley InterScience ( DOI 10.1002/col.20047
Full-text available
Evaluative conditioning (EC) is one of the terms that is used to refer to associatively induced changes in liking. Many controversies have arisen in the literature on EC. Do associatively induced changes in liking actually exist? Does EC depend on awareness of the fact that stimuli are associated? Is EC resistant to extinction? Does attention help or hinder EC? As an introduction to this special issue, we will discuss the extent to which the papers that are published in this issue help to resolve some of the controversies that surround EC. We also speculate about possible boundary conditions of EC and attempt to reconcile conflicting results on the functional properties of EC.
Full-text available
Ecological valence theory (EVT; Palmer & Schloss, Proceedings of the National Academy of Sciences 107:8877-8882, 2010) proposes that color preferences are due to affective responses to color-associated objects: That is, people generally like colors to the degree that they like the objects associated with those colors. Palmer and Schloss found that the average valence of objects associated with a color, when weighted by how well the objects matched the color (weighted affective valence estimates: WAVE) explained 80% of the variation in preference across colors. Here, we replicated and extended Palmer and Schloss's investigation to establish whether color-object associations can account for sex differences in color preference and whether the relationship between associated objects and color preference is equally strong for males and females. We found some degree of sex specificity to the WAVEs, but the relationship between WAVE and color preference was significantly stronger for males than for females (74% shared variance for males, 45% for females). Furthermore, analyses identified a significant inverse relationship between the number of objects associated with a color and preference for the color. Participants generally liked colors associated with few objects and disliked colors associated with many objects. For the sample overall and for females alone, this association was not significantly weaker than the association of the WAVE and preference. The success of the WAVE at capturing color preference was partly due to the relationship between the number of associated objects and color preference. The findings identify constraints of EVT in its current form, but they also provide general support for the link between color preference and color-object associations.
Full-text available
The ecological valence theory (EVT) posits that preference for a color is determined by people's average affective response to everything associated with it (Palmer & Schloss, Proceedings of the National Academy of Sciences, 107, 8877-8882, 2010). The EVT thus implies the existence of sociocultural effects: Color preference should increase with positive feelings (or decrease with negative feelings) toward an institution strongly associated with a color. We tested this prediction by measuring undergraduates' color preferences at two rival universities, Berkeley and Stanford, to determine whether students liked their university's colors better than their rivals did. Students not only preferred their own colors more than their rivals did, but the degree of their preference increased with self-rated positive affect ("school spirit") for their university. These results support the EVT's claim that color preference is caused by learned affective responses to associated objects and institutions, because it is unlikely that students choose their university or develop their degree of school spirit on the basis of preexisting color preferences.
Full-text available
This article presents a meta-analysis of research on evaluative conditioning (EC), defined as a change in the liking of a stimulus (conditioned stimulus; CS) that results from pairing that stimulus with other positive or negative stimuli (unconditioned stimulus; US). Across a total of 214 studies included in the main sample, the mean EC effect was d = .52, with a 95% confidence interval of .466-.582. As estimated from a random-effects model, about 70% of the variance in effect sizes were attributable to true systematic variation rather than sampling error. Moderator analyses were conducted to partially explain this variation, both as a function of concrete aspects of the procedural implementation and as a function of the abstract aspects of the relation between CS and US. Among a range of other findings, EC effects were stronger for high than for low contingency awareness, for supraliminal than for subliminal US presentation, for postacquisition than for postextinction effects, and for self-report than for implicit measures. These findings are discussed with regard to the procedural boundary conditions of EC and theoretical accounts about the mental processes underlying EC.
Evaluative conditioning refers to changes in the liking of a stimulus that are due to the fact that the stimulus has been paired with other, positive or negative stimuli. Although evaluative conditioning appears to be subjected to certain boundary conditions, significant evaluative conditioning effects have been obtained using a large variety of stimuli and procedures. Some data suggest that evaluative conditioning can occur under conditions that do not support other forms of Pavlovian conditioning, and several models have been proposed to account for these differences. In the present article, the authors summarize the available literature, draw conclusions where possible, and provide suggestions for future research.
In this article, we investigate how context influences color preferences by comparing preferences for “contextless” colored squares with preferences for colors of a variety of objects (e.g., walls, couches, and T‐shirts). In experiment 1, we find that hue preferences for contextless squares generalize relatively well to hue preferences for imagined objects, with the substantial differences being in the saturation and lightness dimensions. In experiments 2 and 3, we find that object color preferences are relatively invariant when the objects are (a) imagined to be the color that is presented as a small square, (b) depicted as colored images of objects, and (c) viewed as actual physical objects. In experiment 4, we investigate the possibility that object color preferences are related to the degree to which colors help objects fulfill particular functions or outcomes. We also discuss relations between our results and previous theories of color preference. © 2012 Wiley Periodicals, Inc. Col Res Appl, 38, 393–411, 2013
Consistent with Schloss and Palmer's (VSS-2009) Ecological Valence Theory (EVT) of color preference, 80% of the variance in average American preferences for 32 chromatic colors was explained by the Weighted Affective Valence Estimate (WAVE) of American preferences for the objects that are characteristically those colors. To test predictions of the EVT cross-culturally, corresponding color preferences and ecological WAVE measures were collected in Japan and Mexico for the same 32 colors. American participants showed a broad preference for cool over warm hues, an aversion to dark orange (brown) and dark yellow (olive), and greater preference for more saturated than less saturated colors. Japanese participants showed similar preferences for cool over warm colors, dislike for brown and olive, and high preference for saturated colors, but a greater preference for light colors (pastels) and a lesser preference for dark colors relative to Americans. Mexican participants showed the same aversion to brown and olive, but liked warm and cool colors about equally and tended to like both light and saturated colors less than American and Japanese participants. The WAVEs in each culture were computed from the results of the same three-part procedure: eliciting object descriptions for each of the 32 colors, rating the similarity of the presented color to the colors of the described objects, and rating the affective valence (degree of liking) of each described object. The WAVE for each color is the average valence over objects weighted by the average similarity of the given color to the described object. American WAVEs predict American preferences (r=.89) better than Japanese (r=.77) or Mexican preferences (r=.54). Similarly, Japanese WAVEs predict Japanese color preferences (r=.66) better than American preferences (r=.55) or Mexican preferences (r=.29). These findings are consistent with the EVT, which predicts that culturally specific WAVEs should predict within-culture preferences better than between-culture preferences.
From the response competition interpretation of the attitude-enhancing effects of exposure it was hypothesized that (1) arousal during exposure would decrease affective ratings whereas (2) arousal during rating would decrease the ratings of low frequency stimuli but enhance the ratings of high frequency stimuli. Experiment I provided weak support for the first hypothesis while Experiment II provided clear support for the second hypothesis. Experiment III, exploring an unexpected finding from Experiment I, suggested that exposure will lead to more favorable ratings only if a stimulus is initially neutral or positive.