Non-algorithmic access to calendar information in a calendar calculator with autism.
Calendar savants may tap flexible memory, not one secret math rule.
01Research in Context
What this study did
One teen with autism could name the day of the week for any given date. The team gave him forward dates, backward dates, and surprise questions to see how he did it.
They watched for patterns in his errors and timing. The goal was to test if he used a hidden math rule or something else.
What they found
His mistakes were scattered, not the steady slips a formula would make. He could answer reversed questions just as fast, which rules out a single-step math trick.
The authors say his skill looks like flexible, non-formula memory retrieval instead of a fixed calendar code.
How this fits with other research
De Marco et al. (2016) later watched another savant and found two gears: rules for future dates, memory for past dates. That study extends the 2006 idea by showing the brain can swap strategies depending on time direction.
Allen (1981) timed two psychiatric patients and saw steady reaction-time patterns tied to month and year. That early hint of structure seems to clash with the random errors seen here, but the difference is method: M used timing data, L et al. used error scatter and reversed questions.
Together the papers paint a fuller picture: calendar savants may pull from several mental toolkits, not one master rule.
Why it matters
If a client shows savant calendar skills, do not assume they use a single math hack. Probe with mixed-direction questions and watch for unstable errors; you may uncover flexible memory instead. This view can guide you toward teaching general date skills rather than drilling one formula.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Ask your calendar-savvy learner a few backward date questions and note any scattered errors.
02At a glance
03Original abstract
The possible use of a calendar algorithm was assessed in DBC, an autistic "savant" of normal measured intelligence. Testing of all the dates in a year revealed a random distribution of errors. Re-testing DBC on the same dates one year later shows that his errors were not stable across time. Finally, DBC was able to answer "reversed" questions that cannot be solved by a classical algorithm. These findings favor a non-algorithmic retrieval of calendar information. It is proposed that multidirectional, non-hierarchical retrieval of information, and solving problems in a non-algorithmic way, are involved in savant performances. The possible role of a functional rededication of low-level perceptual systems to the processing of symbolic information in savants is discussed.
Journal of autism and developmental disorders, 2006 · doi:10.1007/s10803-005-0059-9