Species of serendipity

Like most ideas, this one arrived in the shower. I needed to write a post for this week, but my list of topics was wearing thin and the weather is finally starting to get nice enough to distract me. Sure, I had a few promising ideas lined up, but they all need more time to develop. Plus I had a DVD to watch: a Nature of Things episode on serendipity in science due back at the library. Then it hit me – of course! I’ll watch the episode and then write about that.

Serendipity – supposedly one of the top ten most untranslatable words in the English language – was coined in the 1700s by Horace Walpole as a play on the tile of a Persian fairy tale. The Three Princes of Serendip takes place in Sri Lanka. It follows the adventures of three brothers exiled from the island by their father the king, in hopes that his sons might achieve a more worldly education. In the course of their travels, the princes go on to solve many mysteries – like unintentionally tracking down a lost camel on scant evidence – thanks to their sagacity and a series of lucky accidents.

Since Walpole, the word has taken on a close association with Eureka moments in science, starting with Archimedes’ famous bath. Supposedly, the ancient Greek mathematician solved the problem of measuring the volume of irregular objects after noticing how his own body displaced water in the tub.

Scientists have taken a great interest in tracking serendipity, perhaps because it seems to play a role in research success. Wikipedia has an extensive list of celebrated examples, from Viagra to chocolate chip cookies. Many have looked for ways to encourage this kind of scholarly luck. For instance, after his Nobel prize winning work on viruses, the molecular biologist Max Delbrück is perhaps best known for coming up with the principle of limited sloppiness: researchers should be careless enough that unexpected things can happen, but not so sloppy that they can’t reproduce them when they do. Alexander Fleming had this advantage when he discovered penicillin. He first noticed its antibiotic effects in a stack of dirty culture dishes that he hadn’t bothered to clean before leaving for summer vacation.

So how do people study something that is by definition rare and unusual? Psychology Today has summed up some of the latest research on luck, most of it based on surveys of people who claim to be especially serendipitous1. Not surprisingly, they are more competent, confident and willing to take risks than the rest of us. They are also more extroverted and less neurotic than most. Being born in the summer apparently helps as well – especially May.

Other advice might be more practical.

Richard Wiseman has spent years comparing the self-reported lucky ones to regular folks in all kinds of psychological field experiments, and his work suggests that the balance between attention and focus has something to do with it. When Wiseman gave subjects the task of counting all of the photographs in a newspaper, the lucky ones were a lot faster. The reason? They were more likely to notice a glaring hint that he had printed on page 2: “Stop counting – there are 43 photographs in this newspaper.” People who described themselves as unlucky missed the hint, and they inevitably failed to notice a similar message later in the paper stating that they could win $250 just by telling Wiseman about it2.

What causes this tunnel vision? Psychologists have long known that anxiety can be part of it, since negative emotions tend to narrow our focus3. On the flip side, recent experiments show that putting yourself in a good mood – even if it’s induced by playing some positive music – can make you much more likely to notice small details4. We also know that not all motivating factors are equal: financial incentives might actually decrease your luck. In a classic experiment, college students of equal ability were given a series of math problems in which they had to use different-sized jars to measure out certain quantities of water5. The first 9 problems could be solved in an analogous way, whereas the last one was different. However, this last problem had an even simpler solution than the rest because only two of the jars were needed. When the students were offered a reward, they were able to solve these puzzles faster except for that last one. The extra motivation actually decreased their performance on the novel problem – and all it took to create this choke effect was a measly $1.50 reward!

The latest work on serendipity looks at its effect on knowledge-based work – especially in the context of internet resources and search. These kinds of happy accidents are difficult to study because they are personal; as Elaine Toms puts it, “My serendipitous experience may not be yours.”6 A professor at Dalhousie University, Toms triggers luck in the lab by adding unexpected content to search tasks, and monitoring how people respond. Her research demonstrates that exploratory search and social networking can lead to new knowledge in a serendipitous way, especially if the information has time to percolate7. This incubation effect is consistent with other results in the psychological literature: extra time to digest information unconsciously can improve performance on all kinds of tasks, from word games to creative writing8-9. And curiously, complete mental rest is not the best strategy. If you want your unconscious mind to get things done, try setting your conscious brain to work on other low-demand cognitive exercises9.

This brings up another question about serendipity: if there are things we can do make it happen, how much does chance have to do with it anyway?

Archimedes’ Eureka moment is not truly serendipitous the sense that Walpole had in mind. The problem of volume was, after all, part of his life’s work – Archimedes figured out the volume equations for spheres and cylinders among many other shapes. No word on whether a steamy bathroom helps, but he definitely benefited from the incubation effect – and to be truly serendipitous you have to find something that you weren’t searching for in the first place. According to the chemist Royston Roberts, Archimedes’ bath was pseudoserendipitous10. Roberts came up with this concept because he wanted to make sure that credit for hard scientific work was given where due. Take the American entrepreneur Charles Goodyear – he spent years experimenting with different ways to harden rubber before discovering vulcanization in a lab accident. Some scholars have even developed a set of logical equations to distinguish pseudo examples like Goodyear and Archimedes from the real thing; these equations also define several different species of serendipity, like when an incorrect assumption leads to new knowledge11.

Royston Roberts points to George de Mestral as a case of true serendipity10. While out with his dog in the woods, de Mestral, a Swiss engineer, picked up some burdock burrs. Curious about how they worked, he had a look at the burrs under the microscope. De Mestral noticed the fine hook structures that help bind the seeds to fur, clothing or anything fibrous – and he was inspired to try making his own version. The reversible fabric adhesive he came up with is ingenious. Any movement – whether pressing the two surfaces together or tugging them gently apart – only helps Velcro bind more tightly.

But Roberts is wrong – this isn’t true serendipity either. De Mestral may have brought the burrs home accidentally, but his goal of figuring out how they worked was an intentional one. More importantly, he wasn’t first to solve this particular problem – the burrs were. De Mestral just found a clever way to copy their design by heat-treating nylon. And though nature certainly came up the design by chance, the burdock plants were, in a sense, looking for it. Seed dispersal is a universal problem, and the hook-covered burr is a common solution that has evolved in many species.

Is there any real serendipity? Biologists can list many examples of traits that originally evolved for one purpose, but eventually led to entirely different advantages. The classic example is feathers. These first appeared on dinosaurs as insulation, but by exploiting their aerodynamic advantages for flight, birds were able to expand around the world. In the last few years paleontologists have even turned up dinosaurs with brightly-coloured feathers and elongated peacock-like plumes, suggesting that plumage was also used for communication and display before feathered flight evolved12-13.

As for science and everyday life, I suspect luck plays another subtle role that often goes unnoticed. A recent study found that our minds wander away from the situation at hand nearly 50% of the time14. The subjects of this one, people around the world who had their thoughts tracked with the help of an iPhone application, also reported that they were slightly less happy whenever this occurred. Unpleasant, maybe, but mind wandering is also the stuff of new ideas, which aren’t born from scratch so much as chance associations of past experiences. I’ve noticed that you can often track these down if you try. A good mood, a keen eye and plenty of incubation time are useful, but serendipity also depends on the framework of knowledge already in place.


  1. Webber, R. 2010. Psychology Today. 1 May 2010.
  2. Wiseman, R. 2003. Skeptical Inquirer. May 2003.
  3. Easterbrook, J. A. 1959. Psychological Review 66: 183-201.
  4. Rowe, G. et al. 2007. PNAS 104: 383-388.
  5. McGraw, K. O. and McCullers, J. C. 1979. Journal of Experimental Social Psychology 15: 285-294.
  6. “Gone Sideways” The Nature of Things. 16 December 2009.
  7. McCay-Peet, L. and Toms, E. G. 2010. Proceedings of the IIiX’10.
  8. Zhong, C. et al. 2008. Psychological Science 19: 912-918.
  9. Sio, U. N. and Ormerod, T. C. 2009. Psychological Bulletin 135: 94-120.
  10. Roberts, R. M. 1989. Serendipity: Accidental Discoveries in Science. Wiley.
  11. Dias de Figueiredo, A. and Campos, J. 2001. Proceedings of the Workshop Program of ICCBR.
  12. Li, Q. et al. 2010. Science 327: 1369-1372.
  13. Zhang, F. et al. 2008. Nature 455: 1105-1108.
  14. Killingsworth, M. A. and Gilbert, D. T. 2010. Science 330: 932.