Quick! Think of a color. Now think of a number between one and 10. What did you choose?

If a common factoid is to be believed, the answer is most likely “blue” and “seven”. And we know what you’re thinking: that there must be some kind of confirmation bias at play here; that huge swathes of people simply can’t be so predictable; that the law of large numbers must surely even everything up over a big enough population. 

But in fact, there’s some bona fide evidence to support the so-called “blue seven phenomenon” – even if almost all of it comes from the same few researchers from the 1970s. So what’s going on?

The evidence

It started in the second year of that decade, when William Simon, a researcher at Southampton College – the establishment no longer exists, having been absorbed into Stony Brook University in 2006 – surveyed some 490 college students with the same questions we posed at the beginning of this article.

“The number seven and the color blue were by far the most frequently written,” Simon reported, with close to one in three picking the former, and more than two in five choosing the latter. That’s well outside what you’d expect from random chance, and definitely intriguing enough for Simon to extend his investigation – which is precisely what he did the next year, moving from college students down to elementary and junior high school students.

The results, though, were the same: “533 elementary and junior high school children were requested to write down a number between 0 and 9 and the name of a color,” he reported in 1972, and “the number seven and the color blue were by far the most frequently written (p < .001 in both cases).”

Further investigations by other researchers found similar results. A smaller 1976 study among male university students in Kenya seemed to reproduce the effect pretty strongly; the same thing happened in Australia in 1978 even after subjects were informed of the bias. Clearly, this was a phenomenon that transcended cultural boundaries, potentially innate in humans from birth. 

Or was it? As time went on, some researchers questioned these initial surveys. Were they looking at a “blue seven phenomenon”, or two unrelated effects that had been unnaturally thrown together? Did the phrasing of the questions make a difference? And even if the phenomenon was real, what would that mean?

In short: is blue-seven real? And if it is… then why?

Blue?

It’s one of the first questions little kids learn to ask to get to know each other: “What is your favorite color?” And, perhaps surprisingly, it’s also one of the first things psychologists ever asked: “Surveys on color preference can be found among the very first psychological experiments,” explained Miho Saito, an educational and social psychologist from Waseda University, in 2015. 

“Some studies have been carried out on the preference for colors associated with particular objects,” Saito wrote. “Many, however, have investigated the affective appeal of color, not in combination, but separately, so as to evaluate single colors themselves without the influence of other variables.”

As a result, we have quite a lot of data about favorite colors around the world – and that blue comes out on top is fairly undeniable, to be honest. Oh sure, there are some significant cultural variances: in Asia, for example, people love white and hate grey; in West Africa, surveys found brown and black to be favored, while red took the top spot in Europe and Central Africa. But in all these regions, blue was still high up on the list – and equally importantly, it was the only color not connected to a taboo in any researched culture.

Other studies have found more evidence for blue’s popularity. In 1981, for example, researchers asked more than 400 people from across nine different cultural groups to choose their favorite and least favorite colors. “Results showed that vivid blue was the only color that was commonly preferred highly by all groups,” Saito reported.

In other words, blue may not be everyone’s favorite color worldwide, but it is a pretty safe choice. There’s not all that many people out there who hate blue, relatively speaking; when surveyed, people tend to associate it with a feeling of “pleasantness” rather than the opposite.

Now, why exactly that happens is a question yet to be answered. Whether or not we like a color is thought to be a multi-layered thing, with our base reaction to it at the center, deeper connections to the color above that, and shallower connections around the edge – so, for example, witnessing your entire family getting eating by a tiger might have more to do with your hatred of orange than the fact that Vogue says it’s out of fashion this season. 

But the near universality of blue makes it interesting. “Further studies are necessary to clarify other factors which may influence this phenomenon,” Saito wrote, “because color preference is such a fundamental human trait.”

Seven?

On to seven, then, and this one is perhaps easier to explain. Seven comes with two very big advantages when it comes to being picked “randomly”: first of all, we’re all very familiar with it – think “seven deadly sins”, “seven wonders of the world”, “seven days of the week”, and myriad other septets we’re brought up knowing about – and second of all, it… well, it feels random.

No, we know that doesn’t technically make sense, but you know what we mean, right? “The numbers one and 10 don’t feel random enough, neither does two, nor the other even numbers, nor five, which is right in the middle,” explained mathematics communicator and author Alex Bellos back in 2014. “So we quickly eliminate all the numbers, leaving us with seven, since seven is the only number that cannot be divided or multiplied within the first 10.” 

“Our response is determined by arithmetic,” Bellos wrote. “Seven ‘feels’ more random. It feels different from the others, more special, because – arithmetically speaking – it is.”

And, unlike our preference for blue, it seems like this really is it. We’re not particularly drawn towards seven because we love it – although, apparently, we do love it – but because we’re trying to fulfill that “random” or “arbitrary” criteria set forth by the question. We know this, because when you change the question slightly – rewording it so that seven is still a legitimate answer, but not necessarily the “most random” one using the logic described above – the frequency with which it turns up drops dramatically.

For example: pick a number between six and 15. Seven is a perfectly valid choice here, but you’re much less likely to go for it – according to one 1976 survey of 237 college students, using these 10 options instead of the standard zero-to-nine version drops seven as a response from almost one in three to just barely over one in six. Similarly, phrasing the question in a way that suggests seven might be an “obvious” answer drops its popularity significantly.

In short, the researchers concluded, “the predominance of seven is due to a choice […] motivated by the desire to appear to comply with the request of the experimenter, even when there is no utilitarian reason to do so.”

“The experimenter’s request is essentially a request for a spontaneous response, and the subject is placed in a paradoxical situation – only if he does not try to comply can he comply,” they reasoned. “But then his response might not appear to be in compliance because of its commonness or obviousness. So, if he wishes to appear to comply, the subject must carefully select his response and thus fail to comply. We believe that this is what subjects do.”

Phenomenon?

So, it seems like there is at least a “blue phenomenon” and a “seven phenomenon”. But a “blue seven phenomenon”? Maybe not so much.

What do we mean? Well, in a few of those early studies, the “blue seven phenomenon” was presented as something of an all-or-nothing: “the preference and favorite conditions were categorized as conditional choices in a two-by-two contingency table,” explained one 1977 investigation into the effect. 

When this idea was actually interrogated, though, it was found to be misleading. When the standard pair of questions were posed to more than 100 respondents, the results were as expected: “The results replicate[d] the earlier reports of a preference for blue and seven,” the study admitted, and “generally blue was chosen as the respondents’ favorite color.”

But crucially, one being true had pretty much no bearing on whether the other would be so. “No conditional relationship between blue and seven was observed,” the study concluded.

In other words? Yeah, chances are pretty good you said “blue” up at the top of the article. Sure, you most likely said “seven”. But the chances that you said both… well, it’s less of a given after all.