Colour psychology: real or myth? What the evidence actually shows

Colour psychology is half real. The real half is colour associations: people in dozens of countries agree that red goes with love and anger, and that yellow goes with joy. Whereas the myth half is colour effects. Those are the famous claims that a colour changes how you feel, how you perform or what you spend, and most of them shrank or vanished when bigger studies tested them again.

What's the difference between a colour association and a colour effect?

Most arguments about colour psychology go wrong right here, because they treat two different things as one. A colour association is a meaning people link to a colour, like red with love or danger. A colour effect is a claim that seeing the colour actually changes you: your mood, your pulse, your test score or what you buy. The evidence for associations and the evidence for effects look very different.

Associations are the well-mapped part. Researchers have charted which meanings people give colours in more than 30 countries, and people mostly agree. Whether seeing a colour makes your heart race is a separate question, though. And for colour, that second question is where most of the famous findings fell over.

The biggest review of the field lands in the same place. Jonauskaite and Mohr pulled together 128 years of research (opens in a new tab), which came to 132 studies, 42,266 people and 64 countries. Most of those studies measured associations. Put simply, their conclusion is that colours carry emotional meaning, but we still don't know whether we actually feel it.

Do people around the world agree on what colours mean?

Yes, and to a striking degree. A 2024 study asked 7,393 people in 31 countries, and 69% linked red with love while 55% linked red with anger. Yellow was linked with joy by 56%, and black with sadness by 54%. People aged 16 to 88 answered in much the same way. One detail matters, though: in that study people saw colour names, not colour patches.

The wider research shows the same broad pattern. Light colours tend to go with good feelings and dark colours with bad ones. Red goes with strong feelings of both kinds. In 2020 a study across 30 nations found the colour and emotion links were very alike everywhere, and most alike between countries that share a language or a region.

So where do colour associations come from? Mostly from the world around us. Yellow is tied to joy more strongly in rainy countries far from the equator. Even favourite colours follow experience. In one US study, how much people liked the things a colour reminds them of, such as clear skies or rotting food, explained about 80% of which colours they liked. Blue came top there. But blue isn't the world's favourite, because a group in rural Namibia showed none of the supposedly universal patterns.

Red means love and anger at the same time

In that same 2024 study across 31 countries, red was the top colour for love and also one of the top colours for anger. That's not a contradiction. A colour's meaning comes from where you see it, not from the hue alone.

Red on a Valentine's card says love. Red on a cartoon face says anger. And red on a warning sign says danger. It's the same red every time, and the setting does the talking.

Colours can also carry specific meanings in particular cultures. So before you rely on a colour to say something, check how your own audience reads it. How nearby colours change the way a colour looks is a different kind of context, and that's covered in The same colour looks different in different places.

What happened when the famous colour effects were tested again?

Most of them shrank to almost nothing, or disappeared. The early studies were often small, and small studies can throw up big results by chance. When researchers ran larger studies, pooled many studies together, or registered their plans in advance, the effects faded.

Andrew Elliot is the field's leading researcher on red, and he's candid about this. Writing in 2015 (opens in a new tab), he said it was too early for bold theories, and that colour findings tend to be "provocative and media friendly". On red and attraction, his own co-authors went further. In their 2018 review, two of the three thought the effect might not exist at all, while Elliot thought it was small and depended on the situation.

So if a colour claim sounds surprising and comes from one small study, treat it as a maybe.

Did wearing red really help athletes win?

Red did seem to help once, but the scientists who found it have since shown the advantage is gone. Back in 2005, Hill and Barton published a paper in Nature showing that athletes in red won more combat bouts at the Athens Olympics. Before 2005, red won 56.8% of close bouts.

Then in 2024 the same two scientists joined colleagues to pool 6,589 contests (opens in a new tab) from Olympic Games and world boxing championships since 1996. Red won 50.5%. That's no better than a coin toss.

Where red did seem to win, the advantage may have lived in the judges' eyes. In one study, referees were shown the same fights with the colours digitally swapped, and they gave more points to red. And when taekwondo moved to electronic scoring, the colour effect vanished. A link between red shirts and success in English football also failed when it was reanalysed with six more European leagues.

Does pink really calm people down?

The evidence says no, and the man who named the colour helped show it. "Baker-Miller pink" started with a 1979 paper by a researcher named Schauss. The paper measured strength by pushing down on a person's outstretched arm while they looked at a pink or blue card. It claimed the effect began within 2.7 seconds, and that it worked even on colour-blind people.

Two years later, in 1981, Schauss co-authored a proper test. A Californian jail painted a holding room pink and counted the incidents for a year. There were 228, against 237 the year before. Incidents halved in the first three months, then rose above the previous year's level by the end. The authors ended up calling the original arm-push method "dubious".

Warm colours and warm rooms

In a study published in 2026, eight laboratories in six countries ran the same experiment with warm and cool lighting. How warm people felt followed the actual temperature, not the colour of the light. People strongly link warm colours with heat, but warm lighting doesn't seem to make them feel any warmer.

One small virtual reality study did find orange daylight felt slightly warmer. Keep in mind these tests used light, not paint, and the tests of painted rooms are older and smaller. Which of two colours looks warmer is a different question again, and it's answered in Warm vs cool, and why it's relative.

A side note: light on your eyes is not colour on a page. Bright, blue-rich light does make you more alert and shifts your body clock, through a light sensor in the eye. But that depends on the amount and spectrum of the light, not the colour you see, so two lamps that look identical can differ. Morning sunlight wakes you up. A photo of a sunrise doesn't, and neither does a blue logo.

Where the colour psychology rules came from

Most of the rules trace back to anecdotes that lost their caveats each time someone retold them. In Theory of Colours (1810), Goethe (opens in a new tab) described colours stirring particular feelings. His evidence was his own impressions and stories, such as a Frenchman whose conversation changed after a room was redone from blue to crimson. Even Goethe admitted some meanings are learned, though. Green has to be taught before it means hope.

Then in 1939 Goldstein, a neurologist, described brain-injured patients whose outstretched arms drifted differently in front of coloured sheets. There was no comparison group. From the 1940s the colour consultant Faber Birren turned observations like these into rules for business. His 1945 book Selling with Color gives figures with no citations, such as more salads sold on green plates.

Every retelling dropped a qualifier. Brain-injured patients became "people". A few observations became "studies show".

Why did the red button beat the green one?

Most likely because it stood out, not because it was red. The famous "red beats green by 21%" test ran on one page in May 2010, on the Performable blog, with "over 2,000 visits" and no published click counts. Its own author said the result couldn't be generalised. His guess was that red won because it contrasted with the page's green colour scheme.

It's like a worker in a hi-vis vest. They stand out on a building site because nobody else is wearing bright orange. Put everyone in hi-vis and no one stands out.

The thing is, hue alone is a fragile way to stand out. Take an example red (#D32F2F) and green (#388E3C). They differ strongly in hue but hardly at all in lightness, so the contrast ratio is just 1.21:1, where 1:1 means no contrast. About 8% of men of European ancestry have red-green colour blindness, and in my simulation the difference between the two colours fell to about a tenth for them. In other words, for those visitors the button can look nearly the same as the page. Can everyone actually read your design? covers contrast in full.

Worth knowing: a 21% "win" can be pure chance. I simulated two identical buttons, 1,000 visits each, 200,000 times. With a 5% click rate, one button beat the other by 21% or more about one time in six. At 2%, it was more than one time in four. Nobody knows the real click rate in the famous 2010 red-button test, so it can't be judged either way. One small test proves very little.

Where do the popular colour statistics come from?

I traced each of the famous colour statistics as far back as it goes. None of them leads to a study that measured what the claim says.

Does colour change what people buy?

Colour shapes first impressions and expectations, but there's little evidence that a hue changes buying on its own. In lab studies with invented logos, red read as exciting, blue as competent and brown as rugged. Green didn't link to any personality trait at all. Those were mostly US students rating made-up brands, though, so they measured impressions, not sales.

Lightness and saturation matter as much as hue. Brighter, more saturated versions of a colour read as more exciting. And in a 1997 study of adverts, lightness and saturation drove liking more than hue did.

Food is the best-supported area. Colour strongly sets what people expect a food or drink to taste like, which is why people misname a drink's flavour when it's the "wrong" colour. Packaging colour shapes expectations before tasting, but it has little effect after.

Everywhere else, the effects are small. Across 30 years of store experiments, warm or cool store colours didn't change whether people intended to buy (opens in a new tab). In one web study, 65% of students picked a blue online shop over an identical green one, yet the difference in trust was about 0.05 on a 7-point scale. What does help is fit. People judge a logo colour more appropriate when it suits the kind of product.

Can your brand own a colour?

Only with a lot of consistency, and even then colour is a weak signal on its own. Australian researchers at the University of South Australia's Ehrenberg-Bass Institute compared over a thousand brand assets. People linked a brand's colour to the brand just 12% of the time (opens in a new tab), against 40% for shapes such as logos and packs.

Part of the problem is that lots of brands share the same colours. Plenty of them are blue, so blue alone can't point back to you. A shape is far more distinctive. So use your colours the same way everywhere, and pair them with a shape people can recognise. Building a brand palette that works covers how to give each colour a job.

What should you do differently in your next design?

Use colour's meanings on purpose, and stop expecting a hue to do the persuading for you.

1. Use associations as a shared vocabulary. Use them deliberately, and then check them against your product, your audience and your competitors.
2. Choose for fit, not from a meaning chart. Ask whether the colour suits what you sell and who buys it.
3. Make the important thing stand out by contrast. A button should be the most noticeable thing nearby, in lightness as well as hue.
4. Pick lightness and saturation as carefully as hue. They carry much of a colour's mood.
5. Never let colour alone carry a message. Errors, success and "sale" need an icon or a word as well. Designing for colour blindness explains why.
6. Be consistent, and pair colour with a shape. A colour becomes yours through repetition, and your logo gives people something stronger to recognise.
7. Be sceptical of small tests and confident statistics. If a colour choice really matters, test it on your own page with enough traffic.

Frequently asked questions

Is colour psychology real or pseudoscience?

Partly real. People around the world link colours with emotions in similar ways, such as red with love and anger, and yellow with joy, and those shared associations are well measured. What's weak is the claim that seeing a colour changes how you feel or behave. Famous examples, like red lowering test scores or pink calming prisoners, mostly disappeared when larger studies re-tested them. The field's leading researchers say it's too early for bold claims.

Do red and yellow make people hungry?

There's no good evidence that they do. The claim goes back to colour consultants' anecdotes, not experiments. The studies that tested red with food point in different directions: one found red plates and cups reduced snacking, while another found red and black plates increased eating compared with white. Fast-food brands probably use red and yellow because they're bright, easy to spot and, by now, expected in the category, not because they cause hunger.

Does blue really make people trust a brand?

Not on its own. People do associate blue with competence, and in lab studies invented logos in blue were rated more competent than the same logos in other colours. But that's an association, not proof that blue makes customers trust you or buy more. Blue is also common in some industries, so it can signal that a brand "belongs in this category". Fit with the product matters more than the hue itself.

What colour button converts best?

No single colour converts best. The famous test where a red button beat a green one by 21% was one page with about 2,000 visits, and its author said the result couldn't be generalised. Red probably won because it stood out against the site's green colour scheme. What actually helps is contrast. Make the button the most noticeable thing nearby, in lightness as well as hue, so it works for colour-blind visitors too.

Do colours mean the same thing in every culture?

Mostly, yes, but with real differences. A study of over 4,500 people in 30 countries found colour and emotion associations were strongly alike everywhere, and more alike between countries sharing a language or region. Experience shapes them too, which is why yellow is linked with joy more strongly in rainy countries far from the equator. Colours can also carry specific cultural meanings, so check how your audience reads a colour before relying on it.

For the curious

The fading effects are a replication problem. The early studies were small (the median sample in the red-and-test-scores research was 76), and null results often went unpublished, which is called publication bias. A meta-analysis pools studies together and can correct for it. Pre-registration means publishing the plan before collecting any data. For red and attraction, pre-registered studies gave an effect size of d = 0.10, which the authors put as r² = .002. The warm-room idea is known as the hue-heat hypothesis. And the idea that we like colours for the things they remind us of is the ecological valence theory (Palmer and Schloss, 2010).

Light's alerting effect runs through melanopsin, a light-sensitive pigment in the eye, and it's measured under CIE S 026:2018.

The button figures use the WCAG 2.2 relative luminance contrast formula and the Oklab colour difference (ΔE 0.297, falling to 0.028 under simulated deuteranopia, using the Machado, Oliveira and Fernandes 2009 model). The hex values are illustrative, not taken from any brand. Both calculations were checked against colour-science 0.4.7. The chance figures come from a binomial simulation, with 200,000 runs per case.

Sources (37)

• D. Jonauskaite and C. Mohr, "Do we feel colours? A systematic review of 128 years of psychological research linking colours and emotions" (opens in a new tab), Psychonomic Bulletin & Review 32(4), 2025
• D. Jonauskaite et al., "A comparative analysis of colour–emotion associations in 16–88-year-old adults from 31 countries" (opens in a new tab), British Journal of Psychology 115(2), 2024
• D. Jonauskaite et al., "Universal patterns in color-emotion associations are further shaped by linguistic and geographic proximity", Psychological Science 31(10), 2020
• D. Jonauskaite et al., "The sun is no fun without rain", Journal of Environmental Psychology 66, 2019
• S. E. Palmer and K. B. Schloss, "An ecological valence theory of human color preference" (opens in a new tab), PNAS 107(19), 2010
• C. Taylor, A. Clifford and A. Franklin, "Color preferences are not universal", Journal of Experimental Psychology: General 142(4), 2013
• A. J. Elliot, "Color and psychological functioning: a review of theoretical and empirical work" (opens in a new tab), Frontiers in Psychology 6, 2015; A. J. Elliot and M. A. Maier, "Color psychology", Annual Review of Psychology 65, 2014
• A. J. Elliot et al., "Color and psychological functioning: the effect of red on performance attainment", Journal of Experimental Psychology: General 136(1), 2007
• T. Gnambs, "Limited evidence for the effect of red color on cognitive performance: a meta-analysis" (opens in a new tab), Psychonomic Bulletin & Review 27(6), 2020; T. Gnambs, C. Kovacs and B. Stiglbauer, "Processing the word red and intellectual performance", Collabra: Psychology 6(1), 2020
• A. J. Elliot and D. Niesta, "Romantic red", Journal of Personality and Social Psychology 95(5), 2008
• G. K. Lehmann, A. J. Elliot and R. J. Calin-Jageman, "Meta-analysis of the effect of red on perceived attractiveness" (opens in a new tab), Evolutionary Psychology 16(4), 2018
• R. Mehta and R. Zhu, "Blue or red? Exploring the effect of color on cognitive task performances", Science 323, 2009; K. M. Steele, "Failure to replicate the Mehta and Zhu (2009) color-priming effect on anagram solution times" (opens in a new tab), Psychonomic Bulletin & Review 21(3), 2014
• R. A. Hill and R. A. Barton, "Red enhances human performance in contests", Nature 435, 2005
• L. S. Peperkoorn, R. A. Hill, R. A. Barton and T. V. Pollet, "Meta-analysis of the red advantage in combat sports" (opens in a new tab), Scientific Reports 14, 2024
• N. Hagemann, B. Strauss and J. Leißing, "When the referee sees red…", Psychological Science 19(8), 2008; G. Apollaro and C. Falcó, "When taekwondo referees see red, but it is an electronic system that gives the points" (opens in a new tab), Frontiers in Psychology 12, 2021
• N. Goldschmied, P. Furley et al., "Red shirt color has no effect on winning in European soccer" (opens in a new tab), Psychology of Sport and Exercise 58, 2022
• A. G. Schauss, "Tranquilizing effect of color reduces aggressive behavior and potential violence" (opens in a new tab) (the original 1979 paper), Journal of Orthomolecular Psychiatry 8(4), 1979
• R. J. Pellegrini, A. G. Schauss and M. E. Miller, "Room color and aggression in a criminal detention holding cell" (opens in a new tab) (the 1981 jail test), Journal of Orthomolecular Psychiatry 10(3), 1981
• R. M. Gerard, Differential effects of colored lights on psychophysiological functions, PhD thesis, UCLA, 1958 (as described in later sources); S. Lee and S. Westland, "Does colour really affect pulse rate and blood pressure?" (opens in a new tab), AIC 2015
• M. Bavaresco et al., "Beyond hue and heat" (opens in a new tab), Building and Environment 292, 2026; G. Chinazzo et al., "Temperature–color interaction", Human Factors 63(3), 2021
• CIE, TN 015:2023 (opens in a new tab); T. M. Brown et al., "Recommendations for daytime, evening, and nighttime indoor light exposure", PLOS Biology 20(3), 2022
• J. W. von Goethe, Theory of Colours (opens in a new tab), trans. Eastlake, 1840 (orig. 1810), Part VI
• K. Goldstein, The Organism (opens in a new tab), 1939
• F. Birren, Selling with Color (opens in a new tab), McGraw-Hill, 1945
• J. Porter, "Button Color Test: Red beats Green" (opens in a new tab), Performable blog, 25 May 2010 (archived), and "Red vs. Green Buttons Redux" (opens in a new tab), 1 June 2010
• D. Bowman, "Goodbye, Google" (opens in a new tab), 2009; A. Hern, "Why Google has 200m reasons to put engineers over designers" (opens in a new tab), The Guardian, 2014
• L. I. Labrecque and G. R. Milne, "Exciting red and competent blue: the importance of color in marketing", Journal of the Academy of Marketing Science 40(5), 2012
• G. J. Gorn et al., "Effects of color as an executional cue in advertising: they're in the shade", Management Science 43(10), 1997
• C. Spence, "On the psychological impact of food colour" (opens in a new tab), Flavour 4, 2015; O. Droulers, A. Le Hen and S. Lacoste-Badie, "Exploring the effects of packaging color on product evaluation" (opens in a new tab), Food Quality and Preference 144, 2026
• H. Roschk, S. M. C. Loureiro and J. Breitsohl, "Calibrating 30 years of experimental research" (opens in a new tab), Journal of Retailing 93(2), 2017
• S. Lee and V. S. Rao, "Color and store choice in electronic commerce" (opens in a new tab), Journal of Electronic Commerce Research 11(2), 2010
• P. A. Bottomley and J. R. Doyle, "The interactive effects of colors and products on perceptions of brand logo appropriateness", Marketing Theory 6(1), 2006
• P. Phua, L. Bali, Z. Anesbury and B. Sharp, "Shape-based assets are strongest" (opens in a new tab), International Journal of Advertising, 2026; E. Ward, S. Yang, J. Romaniuk and V. Beal, "Building a unique brand identity" (opens in a new tab), Journal of Brand Management 27, 2020
• O. Genschow, L. Reutner and M. Wänke, Appetite 58, 2012; A. Akyol et al., Nutrition Journal 17, 2018; T. E. Dunne et al., Clinical Nutrition 23, 2004
• J. Birch, "Worldwide prevalence of red-green color deficiency" (opens in a new tab), JOSA A 29(3), 2012
• W3C, WCAG 2.2 (opens in a new tab), Success Criteria 1.4.1 and 1.4.3
• Machado, Oliveira and Fernandes, colour vision deficiency simulation model, 2009, as implemented in colour-science 0.4.7