Synaesthesia: The Strangest Thing

  • John Harrison
Oxford University Press: 2001. 250 pp. £16.99, $35
Credit: DAVID NEWTON

Of all the neurological disorders known to modern science, synaesthesia may be the only one that routinely inspires envy in those who study it experimentally. The experience of the synaesthetic seems somehow richer and more compelling; after all, who wouldn't want to experience music in full, glorious colour? Most 'normal' humans experience occasional cross-modal associations, perhaps with the assistance of Mahler or LSD, but this seems thin compared with the inner world of the synaesthetic painter who always experiences the spoken word 'Moscow' as “darkish grey, with spinach-green and pale blue in places”. How can one not envy the woman who experiences shapes accompanying smells, so that the smell of wintergreen evokes “ragged edges”, and peanut butter smells like “things falling down and backward”?

For the past decade, John Harrison has been studying such people, and his book provides an accessible survey of what is known about this most intriguing of disorders. Although he wrote the book alone, most of the work he reports was done in collaboration with Simon Baron-Cohen, and the theory that they jointly developed forms the basis for much of Harrison's approach.

This theory begins with the well-established fact that all humans are born with substantial interconnections among the different sensory pathways in the brain. These normally die off within the first six months of life, but if a pathway's normal source of input is cut off, as occurs in congenital blindness, it may retain its connections to other pathways. Harrison draws the reasonable inference that in synaesthesia some of these pathways are somehow preserved into adulthood. He supports this hypothesis by pointing out that synaesthetics usually say they have had the condition for as long as they can remember, and that (according to an analysis of the limited pedigree data available) the most common form of the condition appears to be genetic — specifically, an X-linked dominant trait that is partially lethal in males.

Unfortunately, definitive tests for neuroanatomical connectivity are impossible in humans, as they require the use of injected tracers, followed immediately by post-mortem sectioning and staining of the brain. Harrison does present neuroimaging data showing that, when colour–word synaesthetics listen to words while blindfolded, some of the brain areas activated are those that have been implicated in colour processing. But most of these areas are known to be driven by stimuli other than just colour even in normal people, and some of the areas activated in Harrison's imaging studies are not sensory at all. Even the clearest result, which shows activity in a highly specialized colour area, cannot identify the source of the activity. So it remains a possibility that the visual pathway is being driven not by cross-wiring but by something more like visual imagination.

Overall, Harrison's theoretical discussions are relatively superficial; where the book really shines is in discussing the data. Reading this book is like looking over a detective's shoulder at the collection of case-files and newspaper clippings spread out on his desk. Harrison has done all the arduous work of gathering the raw data and sorting through historical reports; now he invites the reader to join him in sifting through it for patterns and meaning.

And a fascinating deskful it is. Harrison calculates that the overall population prevalence of synaesthesia is at least 1 in 2,000; most of the synaesthetics he has found are word–colour synaesthetics, and about 85% of these are women — though, perhaps unsurprisingly, all of the famous historical candidates for synaesthesia that Harrison found are men.

It turns out that the colour of a word is usually determined by its first letter, and that each synaesthetic's list of associations is idiosyncratic, with no discernible commonalities even among identical twins. But any one person's associations remain very stable over time. This is the basis of the now-standard 'test for genuineness', which checks the constancy of a subject's associations for a given list of words — non-synaesthetics tend to match fewer than 40% of their previous associations, whereas synaesthetics average more than 90%.

The organization of the book is somewhat frustrating. At times it's structured like an autobiographical narrative, at others it's more like a standard scientific monograph, while in many places it becomes an introductory text in scientific methods. Harrison's pedagogical digressions are sometimes important, but more often baffling. For example, the very brief discussion of neuroimaging data is preceded by more than 20 pages on the general history of neuroimaging. Ultimately, though, he doles out enough fascinating titbits to keep the reader's interest, and the unanswered questions are tantalizing. I found the book especially good 'afternoon coffee-break' reading — although doubtless some synaesthetics would find that an unconscionable clash of tastes.