Knowledge Engine

Do people really find golden rectangles the most beautiful?

·no read data

The idea

The golden sectiongolden section The proportion of about 1.618 to 1 that comes from splitting a line so the whole is to the longer part as the longer part is to the shorter. is the proportion you get when a line is split so that the whole is to the longer part as the longer part is to the shorter, a ratio of about 1.618 to 1. Popular books often claim that rectangles with this shape look most pleasing, and that artists and architects have used it for that reason. The psychological claim, that ordinary people prefer it, has a specific origin: some of the first experiments in empirical aesthetics.

In the 1860s and 1870s the German physicist and psychologist Gustav Fechner laid out ten white rectangles, from a square up to a ratio of 2.5 to 1, and asked people without artistic training which was most and least pleasing. According to the account given by Vico-Prieto and colleagues in 2016, his participants gave 347 responses, and 35 percent favoured the golden rectangle, more than any other shape. None chose it as least pleasing. The rectangle that sits at 1.618 was seventh of the ten in order of ratio, so it was not at the edge of the display. On its face, this looks like a clear vote.

The concrete example that shows why this did not settle the matter comes from Holger Höge in 1995. He repeated Fechner’s two procedures, drawing rectangles (the method of production) and picking from a set (the method of choice), but changed the verbal criterion people were given. Different criteria produced different proportions, and in his words, under both conditions “no preference for the golden section was found”.

Why it might change a decision

The finding bears on anyone choosing a format because of a claimed aesthetic advantage: a designer laying out a web page, a publisher sizing a book, a photographer cropping an image. The testable behaviour is this: if the golden section has a special pull, then in a paired comparisonpaired comparison A test in which people see two options at a time and pick one, repeated across many pairs to build up an order of preference. between a golden-ratio layout and nearby alternatives such as 1.5 or 1.78, people should pick the golden version clearly more often than chance. If they pick it only over very lopsided options, the pull is about avoiding extremes, not about 1.618.

That distinction is visible in a 2013 study by Tractinsky, David and Krupnik, who asked 91 students to compare wireframe layouts. For mobile-device shapes the golden ratio scored highest, but a 1.78 ratio was statistically indistinguishable from it. For web pages the golden ratio sat in the middle of the pack, and shapes closer to square did better. The authors describe their results as mixed.

How solid is this?

The early positive results come from small, uncontrolled studies by today’s standards, and the Fechner numbers above are as reported secondhand in a 2016 paper, not checked here against his original text. Later work did not converge. John Benjafield’s 1985 review concluded that measurement techniques were too imprecise to yield a definitive answer. Frans Boselie’s 1984 study in Perception found that in seven of twelve pairs of line drawings, the version built on a simple 1:1 ratio was preferred significantly.

The most direct modern test in this set is by Vico-Prieto, Cagigas, Rosas and Callejas-Aguilera (2016). They turned five Mondrian paintings into black-and-white versions with golden-ratio, half (1/2) and one-sixth (1/6) divisions, and gave psychology undergraduates 1.5 seconds to pick the more beautiful of a pair: 36 students in a within-subjectswithin-subjects A study design in which the same people take part in every condition being compared, rather than different people in each. experiment and 68 in a between-subjectsbetween-subjects A study design in which different groups of people each take part in only one of the conditions being compared. one. Golden versions beat the lopsided 1/6 versions, with an effect sizeeffect size How large an effect is, as opposed to whether it exists at all. A result can be statistically reliable and still be too small to matter. (η²) of 0.225 in the second experiment’s 1/6 group, but did not beat the symmetrical 1/2 versions. The samples were small, three quarters male and from one university, and the authors limit their conclusion to a Western cultural perspective.

The strongest competing explanation is not that preferences are random. McManus and colleagues, writing in 2010 in Psychology of Aesthetics, Creativity, and the Arts, report that preferences for rectangles are weak at the level of the population but strong and varied in individuals. On that reading, an average across a crowd can show a mild bump near 1.6 while hiding people who reliably prefer squares and others who prefer long thin shapes. Method matters too: which shapes are on offer, how they are arranged and what people are told to judge all shift the answer, as Höge’s results show.

The popular claim, that the golden ratio is a universal key to beauty, outruns this evidence. What the data support is narrower: a modest and method-dependent advantage over extreme proportions in some studies, no special advantage over nearby or symmetrical proportions in several others, and large differences between people.

Sources

  • Höge, H. (1995). “Fechner’s Experimental Aesthetics and the Golden Section Hypothesis Today.” Empirical Studies of the Arts, 13(2), 131–148. doi.org/10.2190/uhtq-cfvd-cau2-wy1c (abstract)
  • Vico-Prieto, A. F., Cagigas, A., Rosas, J. M., and Callejas-Aguilera, J. E. (2016). “Experimental approach to the study of beauty: The role of golden proportion.” Psicológica, 37, 187–207. uv.es
  • Tractinsky, N., David, E., and Krupnik, A. (2013). “Exploring the Aesthetic Effects of the Golden Ratio in the Design of Interactive Products.” Proceedings of the Nineteenth Americas Conference on Information Systems, Chicago. core.ac.uk
  • Boselie, F. (1984). “Complex and Simple Proportions and the Aesthetic Attractivity of Visual Patterns.” Perception, 13(2), 91–96. doi.org/10.1068/p130091 (abstract)
  • Benjafield, J. (1985). “A Review of Recent Research on the Golden Section.” Empirical Studies of the Arts, 3(2), 117–134. doi.org/10.2190/p9cq-hyla-3792-l8h7 (abstract)
  • McManus, I. C., and colleagues (2010). “Beyond the Golden Section and normative aesthetics: Why do individuals differ so much in their aesthetic preferences for rectangles?” Psychology of Aesthetics, Creativity, and the Arts. doi.org/10.1037/a0017316 (abstract)

Check yourself

Answers are checked in your browser. Nothing is recorded or sent anywhere.

  1. According to the account in Vico-Prieto and colleagues (2016), roughly what share of Fechner's responses picked the golden rectangle as most pleasing?

  2. What did Höge (1995) find when he gave people different verbal instructions in Fechner-style production and choice tasks?

  3. In the 2016 Mondrian experiments, what did the golden-ratio versions beat?

  4. Which statement best fits the evidence described in the post?

Questions without answers

  1. If you wanted to test whether the golden ratio matters in a product you care about, such as a book cover or a phone screen, what would you compare it against so that 'not lopsided' and 'golden' could be told apart?
  2. If individual preferences are strong but point in different directions, what does an average preference across a crowd actually tell a designer?

Back to the archive