About the SA Blog Network

Bering in Mind

Bering in Mind

A research psychologist's curious look at human behavior
Bering in Mind Home

Autistic Savants: Geniuses of Obscure Devotions

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Color postcard. “I.C.R.R. Through University, Miss.” Published by University Store, University, Miss. Circa 1920.

(The following is a companion piece to the Slate article, “Eugene Hoskins Is His Name: The long-forgotten story of a black autistic man in Oxford, Miss., who crossed paths with William Faulkner.” You can read that story by clicking here.)

When Professor Hiram Byrd opened up the autistic savant Eugene Hoskins’ private notebook back in Oxford, Mississippi in 1920, here’s what he described seeing:

“Mississippi Division, Jackson District.” Then follows the names of stations leading from Jackson, Tenn – (Just as they were spelled out in his book) Bemis, Melases, Medon, Teag, Toone, Shandy, Bolivar, Hickery Valley, Temper, Grand Junction, Michigan City, Hudsonville, Holly Springs, etc. He has two pages of these names, aggregating 34, which he took great delight in naming off to me. Then comes a page with the following words on it. “Engine Number is 1746 Run from Frogmoor to Mounds Engine Number is 795 919 914 906 851 945 887 Run from Water Valley to Frogmoor 1 Miles to Jackson.” Given a start on this page, he reels off everything from it. Then comes another page which starts “Northern Line Passenger Engine Number is 1140 1139 1008 1065 1080 1141. Runs from Champaign to Centralia 130 Miles Illinois Division Champaign District.”

This was two decades before autism was identified as a distinct neurocognitive disorder, so at the time, Byrd could only scratch his head over Eugene’s unusual preoccupation with train schedules. They riveted his attention. Today, we know that such an obsession isn’t altogether unusual among autistic savants. Fast-forward to 1989 and cross the Atlantic Ocean to London, England, for example, where the psychologists Neil O’Connor and Beate Hermelin were studying a group of British autistic savants who, just like Hoskins, were infatuated with public transport. For them, life was all about getting a handle on bus schedules rather than train schedules.

“They were of low general intelligence,” explained the authors of their sample of six, “but had an unusual knowledge of bus routes, bus numbers and bus timetables.” A series of experiments comparing this group with non-autistic control subjects revealed that their astonishing memory for bus-related items failed to extend to non-bus topics, such as lists of vegetables. In fact, when trying to memorize random digit pairs, the autistics shoehorned this novel numerical information into their bus mnemonics, spontaneously seeing all digits as bus routes and bus numbers. They even tried to lump together the random number pairs presented to them by appealing to their knowledge of home garages from which their trusty London-based buses were run.

Autistic savants don’t appear to be gifted with a general memory, in other words; instead, their phenomenal skills are limited to niche areas or subject matters, or, in the parlance of today’s autism research, “circumscribed interests.” Not all autistic individuals are savants, which implies a mental or sensory disability but an outstanding capacity in a narrow domain of intellectual or artistic function. Nor are all savants autistic. But these so-called “islets of abilities” are much more common among autistic people (9.8 percent) than among those who are mentally retarded (0.6 percent). In addition to numbers, dates and places, and often linked to them, researchers have identified several other frequent categories of circumscribed interests, unusual preoccupations that manifest at very young ages in autistic children. Things like trains, vehicles, planes, blocks, home electronics, computer equipment, road signs and sporting equipment are all subjects of “high autism interest.”

Children without autism can have passionate interests and hobbies as well, of course, but as psychologist Lauren Turner-Brown and her colleagues showed last year in a study with autistic children and their “typically developing” peers, the circumscribed interests of the former are uniquely non-social in nature. A recent study led by psychologist Noah Sasson revealed that even two-year-old autistic children exhibit greater attention to visual depictions of “high autism interest” objects over faces or photos of more commonplace items, such as clothing. Some psychologists, such as the Cambridge University autism researcher Simon Baron Cohen, believe that while high-functioning autistics fail to understand—and to be even less interested in—the psychologically nuanced dramas and everyday soap operas appealing to the rest of us, they often possess heightened abilities in the areas of physics and mechanical systems. Another way to say this is that autistics gravitate to the question of how things work (including how other people behave) instead of the question of why they do so: theirs is a worldview based on the observable principles of movement rather than a philosophy of abstract hidden causes.

One especially revealing example of how a high-functioning autistic individual can get by in the real world by exploiting their heightened knowledge of surface-level behaviors, without having to think very deeply about the confusing mental states underlying other people’s actions, is provided by University of Sheffield psychologist Digby Tantum. This is the case a woman with Asperger’s syndrome trying to navigate her way around the use of a crowded ATM machine:

She had observed that when people lined up, they left a gap between themselves and the person in front, and that this gap was substantially larger in the case of men standing behind women. She used this information to jump lines, looking for this combination and pushing in behind the woman nearest the front who was followed by a man.

This woman’s understanding of the way people work was motivated by a desire to learn how they behaved in this particular social setting, not their mental reasons for doing so. Only by assessing and becoming extraordinarily sensitive to the way routines and conventional social rules intersect with people’s overt behavior could she enter the social environment, albeit inappropriately in this instance—after all, she couldn’t understand why those waiting patiently in line behind her got so angry.

What first captured Byrd’s attention and prompted him to look deeper into Eugene Hoskins’ case were his calendar-calculating feats, meaning he could swiftly name the weekday of any given date. It’s only in recent years that scientists have examined the phenomenon of human calendar calculation, and what they’ve discovered about modern chronologic prodigies aligns remarkably well with Hoskins’ impressive performances in Oxford. According to the behavioral neurobiologist Anna Dubischar-Krivec, calendar-calculator skills tend to develop between the ages of 8 and 15. Often other savant talents, such as musical, drawing, or navigational virtuosity, accompany them. In a 2009 study comparing three savant calendar calculators with “normal” controls, Dubischar-Krivec found that unlike expert mathematicians who could also calculate dates rapidly, the autistic sample was unable to articulate exactly how they were able to pull off their astonishing feats. And the same went for Hoskins. “Asked how he does it, Eugene says he can’t tell you—that he doesn’t know himself.”

A talented mathematician, on the other hand, might explain the formulae of perpetual calendars, such as the 28-year rule (the same calendars are shared by two years 28 years apart within the same century) and Pope Gregory’s exception to this rule, in which he declared in 1582 that century years are not leap years unless they’re divisible by 400. Although some questions remain over savant calendar calculation, scientists realize it’s not quite as supernatural as it was considered in Hoskins’ day. After conducting a series of experiments meant to probe the techniques of perhaps the fastest living autistic-savant calendar calculator (a 21-year-old man named “Donny” with an IQ of 71 who tested at a 98 percent level accuracy in naming the weekday of any date between the year 1 and the year 9999, at speeds up to 700 ms), Marc Thioux and his colleagues from the Yale School of Medicine conclude that the starting point for such an awe-inspiring capacity is the autistic child’s preoccupation with numbers and dates, in combination with a penchant for repetition and ritual. Parents or others in the community marvelling over and praising the child for their unique numerical abilities are also rewarding the autistic person with social acceptance, further motivating their skill development in this rather non-practical domain.

Similarly, Dubischar-Krivec argues that calendar calculation in autistic savants involves a mix of rote memory for previous dates and, eventually (at least for some, like Donny), an implicit understanding of the underlying algorithms driving the Gregorian calendar, such as the fact that if a non-leap year starts on a Sunday, the following year will begin on a Monday. Like most autistic calendar calculators, Hoskins had his limits: “He can’t go back beyond 1901,” Byrd says, “and can’t go forward beyond 1924. But during these 24 years success is 100%.” Most telling, perhaps, is that Dubischar-Krivec’s handful of autistic-savant calendar calculators often failed to solve basic arithmetic problems in which memorization failed them.

In high school, I had a classmate who was a calendar calculator, who walked on his tiptoes and seemed altogether indifferent to the ridiculous social dramas of adolescence that burdened the rest of us. Looking back, of course, it’s clear as day that he was an autistic savant. I only wish I knew then what I know now. But research into both autism and savantism has grown at an astonishing rate over the past few decades. And these findings are slowly trickling into the public consciousness, so that we are becoming more aware of—and more importantly, more sensitive to—issues of neurodiversity and the signature characteristics of those falling along the autistic spectrum.

Jesse Bering About the Author: Jesse Bering is Associate Professor of Science Communication at the University of Otago in New Zealand. He is the author of The Belief Instinct (2011), Why Is the Penis Shaped Like That? (2012) and Perv (2013). To learn more about Jesse's work, visit or add him on Facebook ( Follow on Twitter @JesseBering.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 3 Comments

Add Comment
  1. 1. Rae Ray 6:41 pm 02/17/2012

    Calendar calculation may not be a very practical skill, but how useful would it be to block out “ridiculous social dramas”! I suppose it all boils down to where you place value, on human relationships or a rapid ability to calculate, in other words, crosswords or sudoku?
    Do you think children may develop their autism as a defense mechanism to some traumatic social drama?

    Link to this
  2. 2. fsilber 1:54 pm 04/12/2012

    No need to presume any traumatic social drama. Speaking as a person with MILD Asperger’s Syndrome (which in turn is mild autism), I am weird because my brain developed that way on its own, probably due to some unknown combination of genes, in utero hormones, and viral infections. My lack of intuition about the ways of people came just as naturally as my precocious childhood understanding of basic first-order logic even before I had the words to express the principles that seemed intuitively obvious to me.

    Link to this
  3. 3. Dever 2:24 am 06/1/2012

    I enjoyed reading your article!

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Scientific American Dinosaurs

Get Total Access to our Digital Anthology

1,200 Articles

Order Now - Just $39! >


Email this Article