Something you guys should know (savant skills)

Post Reply New Topic

I've been fascinated with autism and mental process for some time now. I guess I could say it's a special interest of mine. I've done some searching on WP for "savant" and "splinter skills" and what I've read is all over the place. The predominate view on this forum seems to be that to be a savant one must have lower than average IQ. There is also some confusion as to the age of onset of the skill as well as what constitutes a savant skill.

Recently I have listened to a series of presentations organized by The Royal Society which included many of the most respected researchers in the field, the topic of which was savant skills in ASD. Particularly interesting was the presentation titled "Prevalence And Types of Talents In the Autism Spectrum" by Patricia Howlin. The study her team performed involved 137 autistic individuals and spanned over 3 decades. This is very significant because most of what the general population know about savant skills is based on old research of individual cases or a very small sample. The individuals in Howlin's study had a range of FSIQ of 28-135, so both LFA and HFA/AS were studied. The definition that was set for investigating skills was subtest scores exceeding 1 SD of the norm and 2 SD of the individuals subtests mean score (the avg. for subtests is a score of 10 with an SD of 3. So, an individual with an FSIQ of 100 would need to score 16+ on a subtest to qualify for this standard). I will give you guys some key points that surprised me and will probably surprise you.

1. There were no savants among individuals with an FSIQ below 60.
2. A large proportion of savants had an IQ above 100, they tended to be high on a range of subtests but particularly VIQ.
3. The Block Design Task (BDT) was the most common subtest for qualified scores. The Digit Span and Arithmetic subtests were also common areas.
4. The rate of savant skills in the sample was approx. 25% (compared to the 10% I had previously believed).
5. Many individuals developed the skills as adults. One interesting point was that Temple Grandin was there and spoke during the Q&A portion. Her point was that it took her 3 years to develop her skill.
6. Most of the skills were of the calculation variety (calendar calculation is the most common reported skill).

These were some of the points I found most interesting but there were many more. The audio presentation is 41 minutes long but I would urge any of you who are interested in this topic to listen to it. There were many other presentations made on different aspects of this subject that should be of interest to you but I have neither the time nor energy to write about all of them. One note, the presentations are in audio only format so unfortunately I couldn't see the graphic representations. Below I will link the source page with all of the presentations. The one by Howlin, mentioned above, is the 3rd down. I would like to hear your thoughts on this.


www.royalsociety.org/2008-Talent-and-Autism/ <--- link here

I took an IQ test when I was in highschool. I am not going to give my exact IQ because I don't see it to very relevant. However, I did do very well in working memory (top 2%). I also did pretty well in the block design subsection. Apparently all my tetris skills had paid off.

In highschool I became fascinated with multiplying large numbers in my head. I could multiply 8 digit numbers if I was given sufficient time (about an hour or so). I see this as an act of extreme patience and stubbornness combined with some innate working memory ability which may be indicative of Asperger's.

I am not a savant though. A true savant would be able to multiply those numbers in less than a minute. They would do it without being able to explain the cognitive processes behind it. To me that is what true genius is. To just "see" the answer without having an explainable process behind it.

Something you're not taking into consideration , OP: only NTs consider those "savant" skills. We work at them relentlessly and are never fully satisfied with the results . It's hard to just consider yourself a "savant " in a field when you feel like you've cheated : you worked at it day and night for 3 years , and NTs are impressed because they haven't , but YOU know they could have . So basically , complimentary comments make me feel like a fraud and I hide my "talents ". There could be more than 25%.

Their definition of savant seems far more inclusive than I would have expected. It is of course going to be difficult/impossible to find a specific, measurable definition that everyone can agree on, but I've always envisioned a savant as being capable of something incredibly difficult, something thought near impossible by the general population, whereas their definition seems to allow someone to be called a savant if they perform one of these subtests at a level one standard deviation better than the general population. One standard deviation is impressive, particularly for people who on other subtests perform far more poorly, but doesn't quite qualify as what I had previously thought a savant skill to be.

The existence of Daniel Tamment's savant abilities should make it perfectly clear that one need not have an IQ below 70 to qualify as a savant. It's just that most cases of prodigious savantism DO occur in those with an IQ below 70, which says many things about the underlying neurobiology.

I agree.

Is Daniel Tammet an acquired savant ? Are his abilities due to a series of severe childhood seizures or one particular grand mal seizure that affected his brain ?


TheSunAlsoRises

The point of my post was that the study was large scale, and that it turned what I thought about savant skills upside down. What you're suggesting is if anyone worked relentlessly for 3 years anyone could develop savant skills. I disagree, this study was conducted by some of the best minds in the field, some of whom have relentlessly studied these skills over a over a period of 40 years, it's their lifes work. Their goal is to learn how to unlock this potential in everyone. If these people could develop the skills they would gladly do so. But they haven't.

When was the last time you met a NT math professor that could calculate 3,748,982 x 8,340,724 in less than 5 sec? I haven't, but the HFA kid that counts floor tiles in the math lab at my college can. One would think that a math professor who has devoted their life to math would develop the skill, but the vast majority can't. Although, one could safely say they've been performing multiplication for more than 3 years. I'm willing to bet that most would like to have the skill.

I do agree that some people hide their skills but it would be a small percentage. The one thing everyone agrees on is that it happens much more often in autism than any other condition or lack thereof.

Not everyone who met the SD criteria was found to be a savant. Also, it is generally accepted that there are 3 divisions of savant skills: Splinter skill savant, talented savant, and prodigious savant. When most people think of savant they think of the prodigious category but it is exceedingly rare. It is thought that 50-100 people in the world have this ability. But if you think about it even the most common splinter skill, calendar calculation, is very rare among the general population.

Go here:

http://www.timeanddate.com/calendar/?ye ... &country=1

Watch the 1st of January, then skip forward a year at a time and watch what day of the week Monday falls, then go back to 1900 and watch the relationship between the 1st of jan and 1st of feb. the calendar moves in very predictable ways.

With enough obsessive study of calendars and perhaps assigning a colour to each year depending on what day Monday starts on and perhaps fuzzing the leap years, you'd be able to figure this out pretty quickly. There's probably even a simplistic way to calculate this, but memorising using a colour or other nmemonic would be faster to recall.

Now if I spent all my free time for a year obsessively examining calendars, I'd probably be able to do the calendar trick. There is a lot to remember, but it is not memorising 100 different calendars of 365 days.

Impressive nonetheless.

Jason

Often-times, It's the juxtaposition of..... a person having a profound disability present along side an extraordinary ability such as calender calculating. Granted calender calculating can be learned, with moderate success, if someone practiced and put forth the effort. BUT, ahhhhhhhh, the question THAT neuro-scientists are puzzling over IS how can a person who can't read, write, spell, count, remember your name after a 5 minute conversation, and has severe learning disabilities possess the ability to calender calculate over hundreds of years and do complex square roots in their mind (instantaneously) without having the fundamental tools used for neuro-typical learning and problem-solving ??

Something else is going on here THAT science can't explain......yet.

TheSunAlsRises

Epilepsy, so yes. It was several events rather than one big seizure.

I think there are two different kinds of savant skills. Those with an IQ below 70. I think they'd be more extreme savant skills. And the other kind would be of someone with a higher than average IQ, but a milder form.

Daniel Tammet might possibly be an outlier because of the seizures that he suffered in childhood. He might be an Autist who is an acquired savant (which could explain 'why' he has both right brain and left brain phenomenal abilities). He is one of the few people i have seen that possess a wide range of prodigious savant abilities that one would normally find in those with an IQ purported to be below 70.

TheSunAlsoRises

Last edited by TheSunAlsoRises on 06 Mar 2012, 12:07 am, edited 1 time in total.

I think the reason why the results of this study seem surprising is simply because of the way it defines savant skills, I do not believe most people use that precise of a a definition when they define a savant.

Yes, it is a lack of differentiation.

"Savant" is an umbrella term.

I've always seen the definition as has a skill that is:
- At least 1 standard deviation beyond the average population
-At least two standard deviations beyond their average abilities.

That seems to be what they're saying here too.

Dealing with IQ subtests like that seems to be talking about is something I've not seen before though.

You're absolutely right Jason. But issue based on this system is that you would need to extrapolate random dates of the year based on the Jan/feb day and that requires what considerable calculation for most people. I do agree with you that it can be learned but how many people can learn it and give the correct day for May 17, 2783 in under 2 seconds? That's really what the difference is here. The difference is that of a peewee league baseball player compared to a major league player. I don't know rugby and cricket terminology but I'm sure you know what I'm getting at with the baseball reference.