Something you guys should know (savant skills)

Post Reply New Topic

I'm not too familiar with these subtypes of savant skills, but I think that most people have a certain idea of what it means to be a savant, and that the standard used in this study is significantly lower than this commonly held idea. Saying "25% of people with autism are savants" without qualification is likely to mislead most people.

Sorry for using the baseball analogy again but... I think of it this way- The people in the study who displayed splinter skills are like avg. major league baseball players. People who are 'talented savants' are like all star major league players. 'prodigious savants' are the babe ruths, hank arrons, and cy youngs. You can see the vast difference between a babe ruth and an average player but think about this; just getting signed to a major league team is a indicates extraordinary skill.

I'm not sure what you consider qualification. If you feel that a study with a sample of 137 individuals that spans 30+ years, with clearly defined standards, is "without qualification" or merit I won't even attempt to convince you otherwise. IMO perfect pitch, the ability to multiply numbers in the millions instantly, and to give a weekday for a date hundreds of years in a past or future in 2 seconds are all extraordinary skills. You might think they're trivial, I don't know.

The idea of what constitutes savant among the general public is typically 'Rainman' or old studies of usually less than 5 individuals. Like usual, the public is 20-30 years behind the scientific community. Look at the general population's definition of what autism is; again, rainman, ret*d, psychopath, etc etc (30 years behind)... Hell, just look at the situation in France, they still believe autism is caused by bad parenting (100 years behind).

But anything outside the numbers is just my opinion Ellintonia. I appreciate the comments.

I would definitely consider this study to be 'qualification', I'm just concerned that its result might get repeated out of context, e.g. with people telling each other "I heard about this study the other day which found that 1 in 4 autistic people are savants" without concerning themselves with the details. Not many would actually bother to read the study, or listen to that 40 minute speech. To use your analogy, I think most people would hear this and think that 1 in 4 autistics are like Babe Ruth, when this is clearly not the case.

I also don't consider any of the skills you mentioned are trivial, and I think that anyone with these abilities (with the possible exception of perfect pitch) would qualify as a savant. But I'm not convinced that all the people called savants in this study have skills like that, though I could be wrong. The speaker mentioned that someone could be called a savant if they performed one SD higher than the general population in a subtest. One SD better than the average in mental calculation wouldn't be anywhere close to multiplying 6 digit numbers almost instantly, and (back to your analogy) one SD better than the average at baseball wouldn't mean you are good enough to be a major league player.

Someone who has only one spike could be considered savant while another person who has 3 spikes with heights that are consequently smaller than 2 SD due to raised average "floor" couldn't? I don't really get this definition. I guess such definition can only indicate that the person may have some unusual skill, so 25% may be an overstatement.

For example, I scored pretty high at block design and matrix reasoning (17 each) but my "floor" is above average, so it stands out only 1 SD. And what would you say when a spike appears downwards? My largest "negative" spike exceeds 2 SD (symbol search at 1 SD below average).

People with uneven abilities indicated by outstanding subtest scatter usually develop coping strategies to overcome their weaknesses thus they less likely to test low at their weak areas. However, I can imagine that for whatever reason sometimes they don't, so we'll see an unusually spiky scatter. That's when savant skills might form?

Genetic memory..........


TheSunAlsoRises

Ok, leme rephrase mysef. When most people think of a savant they think of a stereotype, not someone who technically has a savant skill by the definition of a savant skill.

Ah, that's why it's a trick, your assuming that to give an impressive demonstration a savant would need to know all the way up to the year 3000, all they really need is 100 years for the trick to work. It's still impressive either way, but this somewhat dims the magic of it.

Jason

I think the "under 2 seconds" part is key. That tells me that this isn't a math trick that comes from obsessive study of calender pattern calculation. For it to be that fast, it needs to be done subconsciously. Conscious calculation is just too slow, even if you have discovered a wonderful algorithm.

A while back in another thread somebody was calling the NT ability to subconsciously and immediately read body language a "savant skill". Thus ensued a debate over the definition of savant. The common definition is a subconscious skill that is rare. The subconsciousness is what differentiates it from practiced expertise (learning a calender algorithm). But he brought up a good point. If you must study and practice reading body language while others just "read" it with no conscious calculation at all, they will seem to have a savant skill even if it technically isn't because it's a common skill.

Perhaps the reason savant skills show up more often with autism is that it's normal for a human brain to be able to subconsciously read patterns without having gained conscious learned expertise and the difference is just which skills this applies to in a particular person. It seems shocking when a human brain can make calendar calculations in 2 seconds but ordinary when a different human brain can make body language calculations just as fast. Perhaps the only real difference is which subconscious calculations a particular brain is wired to make.

Can a person learn to make conscious calculations very, very quickly through the use of algorithms? Yes. Doing it for calendar calculation has been discussed in this thread and doing it for reading body language gets discussed in lots of other threads. But a conscious calculation, however expert and perfectly algorithmed, is never going to be as fast as an unconscious calculation. So I don't think it's a matter of these calender calculating savants (or whatever the skill) studying it very, very intently. That makes an expert but I don't think it makes a savant. I really think it's wiring. But in a weird way, we all have this wiring. It's just that it gets used for different unconscious calculations in different people. And so far nobody has figured out a way to tap into what really is a ubiquitous cognitive power so that people can choose to "add on" a subconscious calculation, even though the researchers do say it should be theoretically possible.

Practice is involved in reading body language via socialization. IF a Neuro-typical had very little exposure to the socialization process YET could read body language on a prodigious level (like an observational genius) this would be equivalent to the calendar calculating ability found in Autists with savant syndrome. It's not enough to just be able to read body language...... a person would have to be able to do it subconsciously on an extraordinary level without having had the social interaction typically needed inorder to perform such a function.

I agree it's wiring, however; i think it might be cross-wiring that is allowing certain people access to......innate capabilities present in us all.

TheSunAlsoRises

Maybe it is NT social cognition itself that prevents savantism in NTs. In the literature, the researchers are always writing about "top-down modulation" of lower-level sensory processing, e.g. NTs automatically filtering out a lot of physical details to enhance perception of social signals - speech over background noise, faces over clumps of dirt on the ground, states of mind in facial eggspressions over physical details of parts of faces, top-down modulation for automatic and obligatory social cognition. No one knows how savant skills work, but I am guessing that filtering out large amounts of physical details and passing everything through a social filter is not going to be conducive to the development of savant or splinter skills. Autistics don't have as much filtering or top-down modulation, so maybe there is less of a block or barrier to the development of these skills.

I think what you say has a lot of validity. It reminds me of the story Dr. Grandin tells of the yakkety yaks around the camp socializing while the non-social person is in a corner somewhere developing fire.

TheSunAlsoRises

OK Ojani thanks for this post because now I see where I didn't make this clear enough and people are getting the wrong impression. If you look up to where you quoted me I've bolded the words "investigating skills". I should have done a much better job of clarifying this. The 1 SD/2 SD criteria is not how savant was defined. This criteria was only used to select the sample of people. In other words, of the people who scored the '1 SD/2 SD' 25% had at least 1 'splinter skill'. The only purpose of the subtests was for the selection of the people to be investigated. Parent reported skills were the other criteria for inclusion in the study. They had to use these criteria for inclusion to keep the study from being unwieldy. This criteria for cognitive skills would preclude people with an FSIQ of 140+ (the 2 SD personal subtest score) because that in itself is considered unusual and it sets a floor of 13 on any subtest (the 1 SD from norm). Furthermore, they weren't looking to match any particular subtest with a particular skill, they were only looking for unusual deviation.

In your own case your scores do not mean you can't have savant skills, it just means you wouldn't have been included in the study unless you fell in the category of parent reported skills. My original post was meant to point out highlights to look for in the study. In hindsight I shouldn't have expected people to listen to the linked audio presentation, let alone read the actual study.

I like what you wrote here. --->

Great post. I think you're right, the unconscious part is a common theme in many savant skills and differentiates a savant skill from a trivial skill.

I wish I could reply to all the posts here but I just don't have the time.

This was a great post too. I was imagining what it would be like if if the AS/NT ratio in of the general population were reversed, would NTs seem like mind readers? Wouldn't the ability to read body language intuitively be an extraordinary skill in an AS world?

No, because NTs are useless at reading autistic body language.

It could be that they suck at it because they are so rarely exposed to it.