Why “Struggling” in Math Isn’t About Intelligence

[GHA]

Let me start with this — struggling in advanced math has nothing to do with how smart you are. Early math is linear — 1 + 2 = 3. But as it gets more advanced, math becomes multi-dimensional. You’re holding different formulas, switching between concepts, tracking how they connect — all at once. That’s not just problem-solving, that’s multi-tasking under pressure.

For some minds, that constant juggling is exhausting. It’s not about lack of understanding — it’s the mental load of keeping so many parts active in your head at the same time.

And this isn’t unique to neurodivergent thinkers — plenty of NTs hit the same wall. Some brains are simply wired for deep focus, not constant switching. In a math exam, that might feel like a disadvantage. But in the real world, that same wiring gives you the edge to see patterns others miss, go deeper into problems, and create solutions that no one else can.

It’s not a weakness. It’s your hidden strength.

 

[Neonatal RRT]

An interesting perspective.

I hit my wall at advanced calculus and physics. At that level of calculus, you are more or less, creating your own formulas to solve problems. However where I struggled was looking at the problem and deciding which pathway to use to solve the problem. I could watch the instructor walk through the problem and once he/she did it, the "lightbulb" came on and I could finish, but, for me, just to look at the problem and come up with the correct pathway was very difficult.

I ran across this in advanced physics, as well. Get me into the lab, present me with the experiment, and I could tell you pretty accurately what the end results would be...but the mathematical proof is where I would stumble.

 

[Hypnalis]

Let me start with this — struggling in advanced math has nothing to do with how smart you are. Early math is linear — 1 + 2 = 3. But as it gets more advanced, math becomes multi-dimensional. You’re holding different formulas, switching between concepts, tracking how they connect — all at once. That’s not just problem-solving, that’s multi-tasking under pressure.

For some minds, that constant juggling is exhausting. It’s not about lack of understanding — it’s the mental load of keeping so many parts active in your head at the same time.

And this isn’t unique to neurodivergent thinkers — plenty of NTs hit the same wall. Some brains are simply wired for deep focus, not constant switching. In a math exam, that might feel like a disadvantage. But in the real world, that same wiring gives you the edge to see patterns others miss, go deeper into problems, and create solutions that no one else can.

It’s not a weakness. It’s your hidden strength.

This seems to be how it feels to people who have reached a bottleneck in math at some point. (**)

But there's are two similar, arguably better partial explanations:

1. There are multiple points in studying math where you must understand something that the next stage of learning absolutely requires. This can happen due to a lack of capacity, poor teaching, or laziness, or misfortune (off school during a critical time, and poor handling of "catch-up" teaching, etc
But if it happens, the path forward can be very difficult
2. There's a distinct step from "counting" to applying a kind of abstract thinking that doesn't come naturally to everyone, and is possible, but quite difficult to learn if it doesn't come naturally.
The principle in (1) applies at this level too of course - it's a hierarchy of knowledge.

(2) is tested in IQ tests, but at a relatively low level. Mostly it's pattern matching, which is not the same thing at all (our visual system uses pattern matching that a lot, so we're wired naturally for it).

If (2) dominated IQ testing, a lot of people would "fail". But it doesn't.

The particular skill/talent doesn't require multi-tasking. But it does require a good understanding of a lot of abstract principles - understanding meaning you can apply them readily, which of course implies that they become part of your autonomic memory (compare with learning the "moves" for some physical activity, where you act without thinking (walking, running, most sports).
When you really know how to do something, you don't have to think about doing it. Note that in math

(**)
Just for context, I've always found doing "math stuff" easy. This is convenient of course, but it also gives me a context when comparing my intelligence and skill set with people who aren't good at math.
It's handy IRL of course, especially in IT. But I'd have happily traded some of it for other innate skills so I had a more balanced set of talents and abilities.

 

[mr prosser]

Might be more of an ADHD thing, but the textbooks and the way it tends to be taught are both incredibly boring, generally speaking.

 

[Angular Chap]

Dyscalculia is a fairly common autism co-morbid condition, but it doesn’t seem to receive anywhere near as much attention as dyslexia for some reason. I attended a top-tier webinar just after my diagnosis and dyscalculia was even missing from that. They went into every other co-morbid such as ADHD, dyslexia and dyspraxia, but no dyscalculia.

I have dyscalculia myself (although no official diagnosis) and no one seemed to notice, I guess I masked it well enough. I was a straight A student at school, except in maths where I barely scraped a C as long as I was allowed to use a calculator.

 

[Crossbreed]

I was at the top of my class in math until the end of 11th grade. That was when we were introduced to matrix math. Suddenly, they were giving random rules about functions between two matrices with no context on how they applied to real-world problems. (I had a sense of that in all prior math classes.)

Not wanting the extra homework load, I did not take Trigonometry in 12th grade. (I did, however, extract some circular trig from my 12th grade physics class.)

 

[GHA]

This seems to be how it feels to people who have reached a bottleneck in math at some point. (**)

But there's are two similar, arguably better partial explanations:

1. There are multiple points in studying math where you must understand something that the next stage of learning absolutely requires. This can happen due to a lack of capacity, poor teaching, or laziness, or misfortune (off school during a critical time, and poor handling of "catch-up" teaching, etc
But if it happens, the path forward can be very difficult
2. There's a distinct step from "counting" to applying a kind of abstract thinking that doesn't come naturally to everyone, and is possible, but quite difficult to learn if it doesn't come naturally.
The principle in (1) applies at this level too of course - it's a hierarchy of knowledge.

(2) is tested in IQ tests, but at a relatively low level. Mostly it's pattern matching, which is not the same thing at all (our visual system uses pattern matching that a lot, so we're wired naturally for it).

If (2) dominated IQ testing, a lot of people would "fail". But it doesn't.

The particular skill/talent doesn't require multi-tasking. But it does require a good understanding of a lot of abstract principles - understanding meaning you can apply them readily, which of course implies that they become part of your autonomic memory (compare with learning the "moves" for some physical activity, where you act without thinking (walking, running, most sports).
When you really know how to do something, you don't have to think about doing it. Note that in math

(**)
Just for context, I've always found doing "math stuff" easy. This is convenient of course, but it also gives me a context when comparing my intelligence and skill set with people who aren't good at math.
It's handy IRL of course, especially in IT. But I'd have happily traded some of it for other innate skills so I had a more balanced set of talents and abilities.

Thank you for sharing this — your breakdown of (1) and (2) is clear and adds depth to the discussion.

I fully agree that missing a single foundational step at a critical stage can make the road ahead in math disproportionately hard. I’ve seen this happen even to people who are otherwise exceptionally talented in their fields. And yes, abstract thinking in math is a distinct leap — one that, if it doesn’t come naturally, can take immense effort to internalise.

Where my own observations come in is from watching how people process this leap differently. For some, especially those with a more linear or highly focused thinking style, advanced math begins to demand simultaneous juggling of multiple abstract layers — not just one idea at a time, but the interplay of several, all held in working memory while moving towards a solution. This isn’t “multi-tasking” in the casual sense, but it does draw on similar cognitive bandwidth, and if that bandwidth is already optimised for depth over breadth, the strain shows.

That’s why you can have someone who can see the why behind a problem instantly, or predict experimental outcomes with precision, yet still find the formal proof process exhausting — because the proof requires holding every moving part in active play at once.

In other words, I think our perspectives meet in the middle — the bottlenecks you describe, and the processing style I’ve seen, both point to the same outcome: brilliance in one dimension, friction in another.

 

[Hypnalis]

I was at the top of my class in math until the end of 11th grade. That was when we were introduced to matrix math. Suddenly, they were giving random rules about functions between two matrices with no context on how they applied to real-world problems. (I had a sense of that in all prior math classes.)

This might be an Aspie thing.

The only obvious use for matrix math when it's first taught is solving systems of simultaneous equations, which sounds abstract at the time. It's actually good for a lot of other stuff, but that's not obvious at first.

Wanting to know the reason for arbitrary stuff (like cultural conventions for appearance and behavior) seems to be is fairly common among ASD, so you may have been caught a "speedbump" that an NT would not have been affected by.

Fun fact: there's a matrix math "trick" that allows very efficient storage of modifications to a matrix that stores an "archetype".
Long ago I read an article (**) that claimed that is what our brains use for facial recognition - relevant because we're generally good at parallel processing (as per my other posts in this thread - interesting coincidence), but generally bad at accurate storage of data ,

So for facial recognition, we store 20 or so "model faces", and one set of storage-efficient modification factors for those per person. IIRC there's a bit more to it than that, but just "implementation details".
Also recognition doesn't use the entire face - the actual "system" is computationally efficient as well as storage efficient.

Which shows up when you ask people with no training to draw faces: there are some typical, quite large errors (like top and back of the head way too small) because people have to learn how to draw the parts that memory doesn't use for facial recognition.

(**)
The theory may have changed since then - it was "late 20th century", and a lot has been learned since then.
But the Matrix/Eigenvector thing is math - it may have been refined, but it can't have changed.

 

[GHA]

This seems to be how it feels to people who have reached a bottleneck in math at some point. (**)

But there's are two similar, arguably better partial explanations:

1. There are multiple points in studying math where you must understand something that the next stage of learning absolutely requires. This can happen due to a lack of capacity, poor teaching, or laziness, or misfortune (off school during a critical time, and poor handling of "catch-up" teaching, etc
But if it happens, the path forward can be very difficult
2. There's a distinct step from "counting" to applying a kind of abstract thinking that doesn't come naturally to everyone, and is possible, but quite difficult to learn if it doesn't come naturally.
The principle in (1) applies at this level too of course - it's a hierarchy of knowledge.

(2) is tested in IQ tests, but at a relatively low level. Mostly it's pattern matching, which is not the same thing at all (our visual system uses pattern matching that a lot, so we're wired naturally for it).

If (2) dominated IQ testing, a lot of people would "fail". But it doesn't.

The particular skill/talent doesn't require multi-tasking. But it does require a good understanding of a lot of abstract principles - understanding meaning you can apply them readily, which of course implies that they become part of your autonomic memory (compare with learning the "moves" for some physical activity, where you act without thinking (walking, running, most sports).
When you really know how to do something, you don't have to think about doing it. Note that in math

(**)
Just for context, I've always found doing "math stuff" easy. This is convenient of course, but it also gives me a context when comparing my intelligence and skill set with people who aren't good at math.
It's handy IRL of course, especially in IT. But I'd have happily traded some of it for other innate skills so I had a more balanced set of talents and abilities.

Thank you for sharing that — it really captures what I was trying to explain.





It’s not the understanding that’s missing — it’s that the brain works best when it can see the process, connect it to something tangible, and then run with it. Once that pathway is lit, everything flows. But expecting someone to pull that pathway instantly from a static problem on paper can be like asking a sprinter to start mid-stride without a warm-up.





In your case, the fact that you could predict experimental results so accurately tells me you had the core conceptual understanding — the why. The difficulty came from translating that deep, intuitive grasp into the rigid, symbolic “proof” the system demands.





And in my view, that’s not a flaw — it’s just a difference in how the brilliance is expressed.

 

[Hypnalis]

Thank you for sharing this — your breakdown of (1) and (2) is clear and adds depth to the discussion.

I fully agree that missing a single foundational step at a critical stage can make the road ahead in math disproportionately hard. I’ve seen this happen even to people who are otherwise exceptionally talented in their fields. And yes, abstract thinking in math is a distinct leap — one that, if it doesn’t come naturally, can take immense effort to internalise.

Where my own observations come in is from watching how people process this leap differently. For some, especially those with a more linear or highly focused thinking style, advanced math begins to demand simultaneous juggling of multiple abstract layers — not just one idea at a time, but the interplay of several, all held in working memory while moving towards a solution. This isn’t “multi-tasking” in the casual sense, but it does draw on similar cognitive bandwidth, and if that bandwidth is already optimised for depth over breadth, the strain shows.

That’s why you can have someone who can see the why behind a problem instantly, or predict experimental outcomes with precision, yet still find the formal proof process exhausting — because the proof requires holding every moving part in active play at once.

In other words, I think our perspectives meet in the middle — the bottlenecks you describe, and the processing style I’ve seen, both point to the same outcome: brilliance in one dimension, friction in another.

<intro>
It looks like we've both edited our posts at the same time. Luckily we're tuning rather than flaming, so it doesn't matter
I'll leave my reply in its current form, so there are parts (like the reference to "juggling" that's aren't technically correct now, but do no harm).
</intro>

Agree.
We don't disagree much - mostly that I don't agree on the "simultaneous" aspect of of applying "college-level" math.

OFC there are "multiple abstract layers" at that level, but it's not "juggling" if you understand them all properly.
The working memory idea is valid - it's probably the same as my "proper understanding.

But to stay with the IT metaphor: there's a big difference in complexity between multi-tasking with a singe-core CPU (i.e. time-slicing) and multitasking one program on every core of a multi-core CPU (multi-threading with suitably clever handling of shared dynamic data).

Perhaps you might look at it this way: both proving non-trivial things in math, and solving complex applied math problems, require selection of the best of a number of possibly applicable methods, and potentially (at the "high end" of both pure and applied match) developing problem-specific methods.

The selection process is qualitatively different from the application of a known principle or method.

This is a fairly large step from basic math though, and even then it doesn't load your ability to handle many things at once.

BTW last I looked (pre-AI) there was still a lot of discussion going on about whether the "logic" part of a human brain truly processes in parallel or not.
The brain as a whole certainly does, & it's certainly a multi-layered processing engine (needed for heartbeats etc).
But back then "they" weren't certain consciousness actually works in parallel (i.e. the "single-core time-slicing" vs "multi-core, multiple parallel threads" example above).

AFAIK there have been significant advancements due to both AI research and much more sensitive ways of measuring electrical activity in the brain. Perhaps that discussion has been resolved.