Interesting duality - implicit learning

[Azul]

So I've been reading to my project in grad course and noticed something interesting.

Laurent Mottron says in his 2017 article:
Should we change targets and methods of early intervention in autism, in favor of a strengths-based education? - European Child & Adolescent Psychiatry

Implicit learning has repeatedly been demonstrated in autistic individuals [7, 46] and is actually slowed down by explicit instructions [29]. Autistic children, and not only savants, can spontaneously learn large amounts of complex information. The immediate value of the information they choose to learn may not be obvious, but autistic children resist conventional ways of learning, precisely because they learn by themselves, rather than because they are “incapable of learning” (see Dawson et al., 2008 for an informed discussion).

(On his prediction that lateral tutorship would be better than face-to-face intervention):

Last, lateral tutorship exposes the child to finalized, completed, and contextualized actions, whereas AIBI learning techniques consist of prompting, shaping, fading, chaining, differential reinforcement discrimination training, and errorless learning of mostly fragmented actions.

Then I remembered a thread here, where I had commented, on the need some have for clarifying underlying concepts against superficial information:
First principles reasoning and asking too many quesions
(This is just anecdotal evidence and is a bit of brainstorming)

While "implicit learning" doesn't map to "first principles reasoning" exactly, I would say their contraries, (explicit, fragmented and algorithmic learning) do sound very similar to (superficial explanations, often formulaic, without context and blindly sequential). Both of them offer bare bones and non contextualized information that is superficially structured.

Note that sequential explanations, when well organized and deep, can in my experience facilitate learning, but that's on the merit of the underlying understanding, since the format alone could represent either a good or bad explanation.

But then there's a constant on papers I'm reading (on art therapy and autism) on the need to simplify and fragment, algorithm-like, instructions for more severe cases. Like drawing steps.

Does the association between "implicit learning" and "first principles reasoning" hold some relationship?
Or this shows some kind of imprecision in my comparison?
What do you think of the common cited method of simplifying, fragmenting and "algorithmizing" teaching for autistics?*
Could this step fragmentation often cited (or its effectiveness) have more to do with verbal/written communication than learning itself?**

*I want to put a disclaimer that I suspect I might be on the spectrum, but my test results sometimes are borderline, I don't have a lot of experiences others have and I'm fairly ignorant on the matter. If the questions sounds silly, that's the reason.
**(This seems to be in line with the rest of the article, where he talks about verbal development).

(It appears to work differently for social implicit learning.)

 

[Shevek]

Implicit is not a term I'd try to use. Things that are considered obviously implied in social situations I clearly miss, while almost nobody else sees the logical corollaries that are obvious to me.

 

[Azul]

The definition given on article 46 is this:

Implicit learning is defined as the acquisition of information or motor skill without conscious access to what was learned or even to the fact that learning occurred

This seem an all purpose term that doesn't explicitly single out social learning. But I agree with you, it sounds broad because of this.

At first glance it doesn't seem the same as the other concept, but their contraries have similarities. And it also apparently goes against the common discourse I've read.

I wonder how much of this is (my) imprecision of concepts and how much reveals something.

 

[Neonatal RRT]

My learning has been some combination of many types,...although when it comes to "special interests", it seems to me that implicit learning has been most important when it comes to learning about a topic. Then, once that foundation has been set, and I wish to apply that knowledge and perhaps coming up with new ideas, then the "first principles" learning takes over,...as often times discoveries can be made when things are reverse engineered. Keep asking "why", working your way back "upstream" until you find those first principles. Once you do that then you can understand things at the molecular or even atomic level.

Elon Musk, a fellow "Aspie" has done the same things. Most of his self-learning (implicit) has then led him to first principles thinking. From a business perspective, he has been extremely successful, understanding the value in vertical integration, materials processing, eliminating supply chain issues, bringing production up to scale, and ultimately making products and providing services that are sometimes 10X less costly than his nearest competitor,...and make a profit where others have failed.

 

[Darkkin]

There is a difference between clear, effective communication and implicit infantalization (stupid proofing directions) because of a preconceived assumption of autistic comprehension processes.

I have a hyperspecific language comprehension matrix (99.8th percentile all standardized language tests, written, verbal, multiple choice, etc, thoughout my education) yet I am a dysphonentic reader, who started preschool able to read independently and knowing numerous multisyllabic words on sight, with definition(s) and spelling, but I cannot sound a two syllable word out. According to a first principles approach (phonetic sounds, I should never have made it passed kindergarten).

If something doesn't make sense, seems vague, or appears illogical, e.g. directions for a project, I will say something. I will actively seek to define my working parameters.

One area of interest I have is poetry, even though I have zero comprehension of metre (iambic in particular is one of the circles of hell, but I'm weirdly effective with tercet based poetry forms that styme even the most seasoned writers (specifically villanelle and its derivatives).

Redundancies and overwriting are two of the most common issues. One of the skills writing balanced poetry hones is effective use of language and implicit carry over of direct objects in order to reduce an overabundance of weighty pronouns and synonyms.

The flipside is writing so generic it is as unique as a street sign.

e.g.

the cat sat there

on the flipside of generic is subtle, but implicit that plays deliberately on assumptions...

There, the cat, sat.

Most people will assume that There is capitalized as the start of a sentence, treating the word as a preposition. And that is incorrect.

The name of the cat is There and There sat. There, the cat.

It comes down to the onboard hardware of the individual, but also to the approaches and abilities of those conveying the information.

It is the drawback of having a standardized system. Any atypical learners/thinkers are going to be shoved through a round hole no matter the number of corners they possess because realistically accommodation is not workable in a widespread medium (standard classroom) due to limited resources. Moreover autistics diagnosed and identified by education systems tend to be ASD2, ASD3, and ASD1 (gender bias toward males 4:1).

Implicit social norms require that female ASD1s assimilate younger and more completely than our ASD1 male counterparts. Basically we are taught to mask and seek social intercourse whether we want to or not.

ASD1 females have an implicit tendency toward language skills and hypersensitive emotions, often the counterpoint of ASD1 males.

Implicit learning is a hardwired skill that comes without conscious effort.

e.g. a proclivity for numbers, language, music, athleticism, empathy, etc.

First order concepts are the most generic terms a subject can be reduced to. Lowest common denominators, if you will or the phonetic sound of a letter.

Depending on the inherent skill sets of the individual and the topics under discussion, first order concepts being used as milestones to define a learner's progress can prove to be a crippling blow to atypical learners.

e.g. the ability to sound out words and one's reading comprehension.

But show them the whole idea and many atypicals will be able to field strip the concept in very little time because of hardwire workarounds that have come built in.

One major challenge with a lateral tutoring approach is standardization itself. Out of the box thinkers are stymied and criticized so often and at such an early age, they can often withdraw from the learning process because they learn 'wrong' working under the assumption that they are either stupid or intellectually inferior because of HOW they learn.

There are any number of polymorphic intelligences found throughout the human populations, all with their own unique learning profiles.

Neurodivergents as a whole (roughly 10 - 12% of the general population (based on rate occurence of ADHD and autism within the general population as they are the most common NDs), tend to have very jagged learning profiles often centered around strengths and interests.

The goal of education (reasonably and realistically speaking) is to cater to the needs of the many, not the few, as education is the basis for the advancement of civilization.

Neurodivergents account for a huge faction of notable movers and shakers through history simply because they think uniquely and are not defined by the homogenized products of public education. Many finding their stride later in life once they had more autonomy over their time and choices.

Frame of reference for my neck of the woods:

54% of Americans have a 6th grade reading comprehension.

70% of the population has an average IQ of between 85 and 115. Less than 5% are above 125.

But again, this is based on the homogenized testing method.

You can be a genius with tools, carpentry, construction, or as a mechanic and still fail basic math, with a preference to read manga or graphic novels.