Color and Psychological

Functioning
Andrew J. Elliot
1
and Markus A. Maier
2
1
University of Rochester and
2
University of Munich, Munich, Germany
ABSTRACTColor is a ubiquitous perceptual experience,
yet little scientic information about the inuence of color
on affect, cognition, and behavior is available. Accord-
ingly, we have developed a general model of color and
psychological functioning, which we present in this article.
We also describe a hypothesis derived from this model re-
garding the inuence of red in achievement contexts. In
addition, we report a series of experiments demonstrating
that a brief glimpse of red evokes avoidance motivation
and undermines intellectual performance, and that it has
these effects without conscious awareness or intention. We
close with thoughts on the need for rigorous scientic work
on color psychology.
KEYWORDScolor; red; avoidance; approach; motivation
Every visual stimulus processed by the human perceptual system
contains color information. Given the prevalence of color, one
would expect color psychology to be a well-developed area.
Surprisingly, little theoretical or empirical work has been con-
ducted to date on the inuence of color on psychological func-
tioning, and the work that has been done has been driven mostly
by practical concerns, not scientic rigor. As such, although the
popular and applied literatures are replete with statements re-
garding the content of color associations and their presumed
impact on behavior (e.g., Green is peaceful and helps people
relax), the lack of theory and carefully controlled experimenta-
tion makes clear conclusions about color associations and their
implications elusive (Levy, 1984; Whiteld & Wiltshire, 1990).
Given the disparity between the ubiquity of color stimuli and
the dearth of extant theory and research on color psychology, we
have developed a general model of color and psychological
functioning. In this article, we set a conceptual and empirical
context for our model, present the model, and describe one main
hypothesis derived from it. Then, we overview a research pro-
gram designed to test various aspects of this hypothesis. Finally,
we briey describe a second hypothesis generated from our
model, and close with thoughts on the need for rigorous scientic
work on color psychology.
EXTANT THEORETICAL AND EMPIRICALWORK
Most existing work on color and psychological functioning is
applied, as opposed to theoretically based. The questions that
drive this type of research include: What colors inuence retail
behavior? What colors inuence food preference? What colors
inuence worker mood and productivity? What colors inuence
physical health and aggressive behavior? What color prefer-
ences are associated with different personality types? Such re-
search simply seeks to establish relations between color stimuli
and affect, cognition, or behavior for pragmatic purposes; it
seeks neither to explain why such relations occur nor to test
basic principles regarding psychological functioning.
Of the existing research that is theoretically based, most has
been loosely guided by Goldsteins (1942) proposal that red and
yellow are naturally experienced as stimulating and disagree-
able, that these colors focus people on the outward environment,
and that they produce forceful, expansive behavior, whereas
green and blue are experienced as quieting and agreeable, focus
people inward, and produce reserved, stable behavior. Subse-
quent researchers have tended to interpret Goldsteins proposal
in terms of wavelength and arousal: Longer-wavelength colors
like red are experienced as arousing, and shorter-wavelength
colors like green are experienced as calming (e.g., Stone &
English, 1998).
Aside from Goldsteins proposal and its derivatives, most
theoretical statements about color rely on general associations.
Different colors are presumed to have different associations, and
viewing a color is thought to trigger psychological responses
consistent with these associations. For example, Frank and
Gilovich (1988) posited that black is associated with evil and
death and, therefore, leads to aggressive behavior. Likewise,
Soldat, Sinclair, and Mark (1997) proposed that red and blue
are associated with happiness and sadness, respectively, and
therefore lead to cognitive processing and behavior consistent
Address correspondence to AndrewElliot, Department of Clinical and
Social Sciences in Psychology, University of Rochester, Rochester,
NY, 14227; e-mail: [email protected].
CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE
250 Volume 16Number 5 Copyright r2007 Association for Psychological Science
with those emotions. Such models tend to focus on one or two
colors/associations and typically propose general links between
colors and functioning across situations.
Existing research on these proposals tends to be sparse and
spotty, occasionally supporting some hypotheses but not others.
Although the popular and even scientic literatures commonly
state as fact that long-wavelength colors are arousing and short-
wavelength colors are calming, the actual data simply are not
supportive. Frank and Gilovichs (1988) proposal is supported
by some data, but that proffered by Soldat et al. (1997) is not.
Furthermore, the extant research on color and psychological
functioning in general is plagued by several weaknesses. First,
perhaps due to the applied nature of the work, many studies have
neglected to follow basic experimental procedures such as ex-
perimenter blindness to hypothesis and condition. Second, many
of the manipulations in these studies have been uncontrolled
(e.g., presenting color on an ofce wall for 4 days) or have altered
participants typical perceptual experience (e.g., presenting
color via overhead lights). Third, and most important, almost no
extant research has examined the effect of hue while controlling
for lightness (similar to brightness) and chroma (similar to sat-
uration), despite the fact that these other color attributes can
themselves inuence psychological functioning.
A MODEL OF COLOR AND PSYCHOLOGICAL
FUNCTIONING AND A HYPOTHESIS DERIVED FROM
THE MODEL
We (Elliot, Maier, Moller, Friedman, & Meinhardt, 2007) have
developed a general model of color and psychological func-
tioning, the core premises of which are stated in the following.
First, colors can carry specic meanings. Color is not just about
aestheticsit also communicates specic information. Second,
color meanings are grounded in two basic sources: learned
associations that develop from repeated pairings of colors with
particular messages, concepts, or experiences; and biologically
based proclivities to respond to particular colors in particular
ways in particular situations. Some color associations may
emerge from learning alone, but color theorists suspect that
many such associations emerge from evolutionarily ingrained
responses to color stimuli (Mollon, 1989). Research indicates
that colors often serve a signal function for nonhuman animals
(e.g., the redness of fruit signals readiness for eating), thereby
facilitating tness-relevant behavior (Hutchings, 1997). If, as we
suspect, humans are prepared to respond to color stimuli in a
similar fashion, then at least some color associations may rep-
resent a cognitive reinforcing or shaping of biologically based
response tendencies. Third, the mere perception of color evokes
evaluative processes. Color computations occur at an early level
within the visual system, and evaluative processes are so fun-
damental that they are present, at least in rudimentary form, in
all animate life (Schneirla, 1959). By evaluative processes we
mean basic mechanisms that discern whether a stimulus is
hostile or hospitable (Elliot & Covington, 2001). Fourth, the
evaluative processes evoked by color stimuli produce motivated
behavior. Color stimuli that carry a positive meaning produce
approach responses, whereas those that carry a negative
meaning produce avoidance responses. Fifth, color typically
exerts its inuence on psychological functioning in an automatic
fashion; the full process from evaluation of the color stimulus to
activation and operation of motivated behavior typically takes
place without conscious intention or awareness. Given that the
inuence of color tends to be nonconscious in nature, color ef-
fects tend to persist, even when they are deleterious. Sixth, color
meanings and effects are contextual. A given color has different
implications for feelings, thoughts, and behaviors in different
contexts (e.g., achievement contexts, relational contexts).
Our research to date has focused primarily on the color red in
achievement contexts. Our hypothesis is that red carries the
meaning of danger in such contexts, specically the psycholog-
ical danger of failure (Elliot, Maier, Moller, et al., 2007). One
source of this reddanger link is presumed to be teachers use of
red ink to mark students mistakes and errors. This specic as-
sociation is likely grounded in a more general societal association
between red and danger where negative possibilities are salient,
such as stop signs and warning signals. These learned associa-
tions may be bolstered by or even derived from an evolutionarily
ingrained predisposition across species to interpret red as a signal
of danger in competitive contexts. For example, in primates, red
on the chest or face (due to a testosterone surge) signals the high
status, and thus danger, of an opponent; Setchell & Wickings,
2005). Thus, through both specic and general associative pro-
cesses that may themselves emerge from biologically based pro-
clivities, red carries the meaning of failure in achievement
contexts, warning that a dangerous possibility is at hand. This
warning signal is posited to produce avoidance-based motivation
that primarily has negative implications for achievement out-
comes. The inuence of red in achievement contexts is presumed
to take place outside of individuals conscious awareness.
OUR EMPIRICALWORK ON RED IN ACHIEVEMENT
CONTEXTS
We began our empirical work with four experiments designed to
test the effect of red on intellectual performance (Elliot, Maier,
Moller, et al., 2007, Experiments 14). In the rst experiment,
participants completed an anagram test that contained a red,
green, or black subject number in the upper right-hand corner.
Green provided a chromatic contrast to red, its opposite in several
color models, and green has some general associations with ap-
proachmotivation. Black, an achromatic color, served as a neutral
control. At the end of this and all experiments in this research
program, participants received a careful debrieng that probed
their awareness of the purpose of the experiment. Results indi-
cated that participants shown red solved fewer anagrams than
those shown green or black; those shown green or black did not
Volume 16Number 5 251
Andrew J. Elliot and Markus A. Maier
differ. Participants were unaware of the purpose of the experi-
ment. Additional experiments replicated this nding using
different achromatic controls (white, gray), a different method
of presenting color (on a test cover; see Figs. 1 & 2), and
color stimuli equated on all color parameters except hue (this was
true in all experiments reported below). In some of these
subsequent experiments, participants motivation and perceived
competence were assessed with self-report measures; null results
were obtained on these measures, indicating that participants
were unaware of the effect color had on their motivation and
performance.
Given that our initial experiments showed no effect of red on
conscious reports of avoidance motivation, we conducted two
additional experiments to examine the effect of red on noncon-
scious avoidance motivation (Elliot, Maier, Moller, et al., 2007,
Experiments 56). Both of these experiments manipulated color
using the test-cover procedure described above. In one experi-
ment, after the color manipulation and before (ostensibly) taking
a test, participants selected the number of easy and moderately
difcult items they wanted on the test; selection of easy items is
a classic indicator of avoidance motivation. Results indicated
that participants shown red selected more easy items than those
shown green or gray; participants shown green or gray did not
differ. In the other experiment, after the color manipulation and
prior to (ostensibly) taking a test, participants prefrontal cortical
activity was assessed using electroencephalography (EEG);
right (relative to left) prefrontal cortical activity indicates that
avoidance motivation has been activated in the brain. Results
indicated that participants shown red evidenced more right
prefrontal cortical activity than those shown green or gray;
participants shown green and gray did not differ.
In a separate set of experiments, we examined the effect of red
on physically enacted avoidance behavior (Elliot, Maier, Binser,
Friedman, & Pekrun, 2007). In a rst experiment, participants
were shown red or green on the cover of an analogies test that
they would (ostensibly) take in an adjacent lab. Participants
shown red, relative to those shown green, knocked fewer times on
the door of the adjacent lab as they anticipated taking the test. In
a second experiment, participants were shown red, green, or gray
on the cover of an IQ test that they would (ostensibly) take. A
sensor was placed on participants to assess their body movement
upon presentation of the test cover. Participants shown red
moved their bodies away from the test cover to a greater degree
than did those shown green or gray; those shown greenor gray did
not differ (see Fig. 3). Debrieng indicated that the effect of red
in these experiments occurred without participants awareness.
Fig. 1. An example test cover used in our experiments.
6
6.5
7
7.5
8
8.5
Red Green Gray
N
o
.

o
f

c
o
r
r
e
c
t

i
t
e
m
s
Fig. 2. The effect of color on IQtest performance in Elliot, Maier, Moller,
et al. (2007, Experiment 4): Mean number of correctly solved items by
color on the cover of the test (means are adjusted for general ability,
premanipulation performance, and gender). Condence intervals (95%)
are indicatedbyvertical lines. Red participants performedsignicantly
worse than green participants and gray participants, who did not
differ from each other.
Fig. 3. The effect of color on body movement in Elliot, Maier, Binser,
Friedman, and Pekrun (2007, Experiment 2). Mean inclination of the
upper body over time in milli g (mg) units as related to the color on the
cover of the IQ test. Negative mg values indicate angle adjustment away
from the test cover.
252 Volume 16Number 5
Color Psychology
Finally, we conducted four experiments designed to test
whether nonconscious avoidance motivation mediates the dele-
terious effect of red on intellectual performance (Maier, Elliot, &
Lichtenfeld, 2007). In the nal experiment in this series, partic-
ipants were shown red or gray on the cover of an IQ test and then
completed a visual-matching task assessing local (relative to
global) processing of stimuli. Local processing represents an often
rigid constricting of attention to the trees as opposed to the
forest and is a well-established indicator of avoidance motiva-
tion. After the visual-matching task, participants completed an IQ
test. Results indicated that participants shown red performed
worse and evidenced more local processing than did those shown
gray. Furthermore, local processing was shown to mediate the
direct effect of red on performancethat is, red led to more local
processing, which in turn undermined performance.
OTHER CONTEXTS AND COLORS
All of the research that we have overviewed has beenconductedin
achievement contexts, but we are currently examining a second
hypothesis that focuses on the color red in relational contexts
(Elliot & Niesta, 2007). We posit that in relational situations,
specically those involving sexual attraction, red carries the
meaning of love, passion, and sexual readiness. These associa-
tions are likely grounded in the use of red hearts to symbolize
romance on Valentines Day; the use of red lipstick, rouge, and
lingerie to heighten attractiveness; and the use of red light to
signal sexual availability in brothels. These learned associations
may be bolstered by or even derived from the biologically in-
grained use of red to attract mating partners during estrus in many
nonhuman female mammals (Mollon, 1989). Thus, through asso-
ciative processes that may themselves emerge fromevolutionarily
based proclivities, red signals love, passion, and sexual readiness,
and the perception of red is presumed to produce approach-
motivated behavior outside of individuals conscious awareness.
Color effects on psychological functioning are not thought to
be constrained to red. Other colors undoubtedly impact affect,
cognition, and behavior as well, and research to examine such
possibilities is needed. Acore premise of our model is that color
effects are context specic, and it will be important to attend
carefully to this issue in subsequent research. Thus, in the
United States, green may be linked with money and facilitate
spending in consumer contexts, but green (especially blue-
green) may be associated with mold and quash ones appetite in
culinary contexts. Likewise, in the United States, black may be
linked with evil/death and lead to aggression in zero-sum com-
petitive contexts, but black may be associatedwith eroticismand
enhance arousal in sexual contexts. Furthermore, although we
believe that some color meanings and effects (such as those that
are the focus of our work thus far) are biologically based and
pancultural, it is likely that at least some color meanings and
effects are entirely learned and vary by culture (e.g., black has
negative associations in the United States that are not present in
other countries lacking a history of prejudice against African
Americans). As such, context must be considered not only in
terms of domain but also in terms of culture.
CONCLUSION
Our research both provides a conceptual framework to guide
research in the neglected area of color psychology and illustrates
how rigorous empirical work in this area may be conducted. We
think that this is a highly promising research area in which many
pressing questions await empirical consideration (e.g., How do
color associations develop, and howdoes this development differ
when biologically based predispositions are present versus ab-
sent? How potent are color effects in real-world contexts con-
taining a wide variety of color stimuli? What is the duration of
color priming?). The scientic study of color and psychological
functioning is not an easy enterprise, as it requires careful as-
sessment and calibration of lightness and chroma, as well as
hue. However, we believe such efforts pale in comparison to the
benets of documenting the inuence of a ubiquitous feature of
the perceived social environment on important affective, cog-
nitive, and behavioral processes outside of conscious awareness.
Social-cognitive research on priming focuses extensively on the
effects of lexical, contextual, and relational stimuli on psycho-
logical functioning; we think the time has come to broaden this
focus to include color stimuli.
Recommended Reading
Elliot, A.J., Maier, M.A., Moller, A.C., Friedman, R., & Meinhardt,
J. (2007). (See References)
Fehrman, K.R., & Fehrman, C. (2004). Color: The secret inuence (2nd
ed.). Upper Saddle River, NJ: Prentice Hall.
Whiteld, T.W., & Wiltshire, T.J. (1990). Color psychology: A critical
review. Genetic, Social & General Psychology Monographs, 116,
387412.
REFERENCES
Elliot, A.J., & Covington, M.V. (2001). Approach and avoidance moti-
vation. Educational Psychology Review, 13, 7392.
Elliot, A.J., Maier, M.A., Moller, A.C., Friedman, R., & Meinhardt, J.
(2007). Color and psychological functioning: The effect of red on
performance attainment. Journal of Experimental Psychology:
General, 136, 154168.
Elliot, A.J., Maier, M.A., Binser, M.J., Friedman, R., & Pekrun, R.
(2007). The effect of red on avoidance behavior in achievement
contexts. Manuscript submitted for publication.
Elliot, A.J., &Niesta, D. (2007). [The effect of red on sexual attraction].
Unpublished raw data.
Frank, M.G., & Gilovich, T. (1988). The dark side of self and social
perception: Black uniforms and aggression in professional sports.
Journal of Personality and Social Psychology, 54, 7485.
Volume 16Number 5 253
Andrew J. Elliot and Markus A. Maier
Goldstein, K. (1942). Some experimental observations concerning the
inuence of colors on the function of the organism. Occupational
Therapy and Rehabilitation, 21, 147151.
Hutchings, J. (1997). Color in plants, animals, and man. In K. Nassau
(Ed.), Color for Science, Art, and Technology (pp. 222246). Am-
sterdam: Elsevier.
Levy, I.B. (1984). Research into the psychological meaning of color.
American Journal of Art Therapy, 23, 5862.
Maier, M.A., Elliot, A.J., & Lichtenfeld, S. (2007) Nonconscious
avoidance motivation mediates the negative effect of red on in-
tellectual performance. Manuscript submitted for publication.
Mollon, J.D. (1989). Tho she kneeld in that place where they grow. . .
Journal of Experimental Biology, 146, 2138.
Schneirla, T. (1959). An evolutionary and developmental theory of
biphasic processes underlying approach and withdrawal. In M.
Jones (Ed.), Nebraska Symposium on Motivation (pp. 142). Lin-
coln: University of Nebraska Press.
Setchell, J.M., &Wickings, E.J. (2005). Dominance, status signals, and
coloration in male mandrills (Mandrillus sphinx). Ethology, 111,
2550.
Soldat, A.S., Sinclair, R.C., & Mark, M.M. (1997). Color as an envi-
ronmental processing cue: External affective cues can directly
affect processing strategy without affecting mood. Social Cogni-
tion, 15, 5571.
Stone, N.J., & English, A.J. (1998). Task type, posters, and workspace
color on mood, satisfaction, and performance. Journal of Envi-
ronmental Psychology, 18, 175185.
Whiteld, T.W., & Wiltshire, T.J. (1990). Color psychology: A critical
review. Genetic, Social & General Psychology Monographs, 116,
387412.
254 Volume 16Number 5
Color Psychology