Take a look at the picture and honestly tell me which shape you would match to the word ‘bouba' and which you would match to the word ‘kiki'.
If you associated ‘bouba' with the rounded shape and ‘kiki' with the pointed one, you'd be among the 95 percent of people to do so. Neither answer is wrong, since neither of these shapes has ever been officially named. Even if they had been, there's no expectation that any test subject would know their names. The naming of the objects isn't about identification, it's just about onomatopoeiac word association.
This particular word association is common across languages. The first version of the experiment was conducted by Wolfgang Kohler, a psychologist. He rowed himself out to the Canary Islands, and stopped on Tenerife, where the primary spoken language is Spanish. There he showed people a rounded shape and a spiky shape, and asked them to identify one as ‘baluba' and one as ‘takete'. Overwhelmingly, people thought the sounds of the ‘t' and ‘k' denoted sharpness, while the ‘b' and ‘l' seemed softer.
The Bouba-Kiki Effect, one of the best-named effects ever, was named after a derivation of Kohler's experiment. Two more psychologists Vilayanur S. Ramachandran and Edward Hubbard went to India, and asked American college students and India Tamil speakers to pick a name for a shape. This time they used the words ‘bouba' and ‘kiki'. Up to 98 percent of the people questioned picked ‘kiki' as the spiked shape and ‘bouba' as the round one.
There was some argument that the written word was an influence, at least on English speakers. The letter ‘k' and the lower case ‘i' were straight and sharp letters, while every letter in ‘bouba' is rounded. This might cause people to match the shape of the letter with the shape of the object. Recently, though, it was shown that The Bouba-Kiki Effect was in full force for children as young as two and half, far too young to read. It's the sound of the words that makes people associate them with one shape or another.
Hubbard and Ramachandran argued that The Bouba-Kiki Effect is actually an example of synesthesia; the conflation of one sense with another. People with certain kinds of synesthia would see colors when they heard music or taste something when they saw the wrong word or image. Most cases of synesthesia are personal, and hard to verify. It's also generally rare. Some studies estimate synesthesia as occurring in one out of every 20,000 people. The Bouba-Kiki Effect might be a kind of synesthesia, the association of certain sounds with certain images, that's so mild and so massive that it goes unnoticed until psychologists wave flashcards in people's faces.