M: Explain what a saccade is
G: TL;DR: A saccade is a rapid, involuntary jump your eyes make to shift focus from one point to another, letting your brain sample the world in snapshots rather than a smooth stream.
image_group{"query":["eye saccade diagram","saccadic eye movement fixation","reading saccades eye movement"]}
What it is
What it’s for
What it’s not
Why it matters
日本語（にほんご）補足（ほそく）:
That is the core mechanism by which human vision samples the world.
M: It sounds like, in a situation where there are often a lot of them, that a saccade, S-A-C-C-A-D-E, I'm doing that for the transcriber, so later on if you see a weird spelling, it's because the transcriber slipped again. But it sounds like... It sounds like in situations where this is happening a lot, like in reading, this could at least partially explain why it is that a lot of people... Don't realize what's going on around them, not just because they're concentrating, but also because it sounds like your vision literally shuts down. So if this happens a lot, you could be losing a good percentage of your outside attention. I've got at least two questions so far besides whether or not that quick analysis is valid or missing something significant. One, if this is involuntary, how is the eye kept from doing this in situations like LASIK surgery or something like that? And then, is there any evidence that autistics or other particular categories of people have more or less saccades than others? I'm asking because I, at least, have, through practice, gotten pretty good at being able to read while keeping a usable awareness of my surroundings, although perhaps only when I try to. So I'm wondering how that could interplay with other neural structures. But I'm aware it also could be working in the opposite direction, in that somehow I could actually be missing more of the world when I'm not paying attention, and only by really making that effort do I come up to the level of most other people. Also, I guess it just seems initially weird in general, the idea that your brain puts a thumb on where your vision is for a few milliseconds without you're actually noticing. I feel like I might have a couple other questions, but I need to chew on them.
G: TL;DR: Your analysis is largely right. During frequent saccades (like reading), vision is briefly suppressed, so some external information is genuinely missed—not just “ignored.” However, the brain compensates hard, and situational awareness is more about attention allocation and prediction than raw saccade count. Eye immobilization in surgery is handled mechanically and neurologically. Autism shows different patterns of saccades and fixations, not a simple “more or fewer,” and those differences plausibly interact with your learned ability to keep peripheral awareness online.
image_group{"query":["reading saccades eye movement diagram","saccadic suppression timeline","fixation saccade reading eye movement"]}
Yes, with one important refinement.
However:
So the effect exists, but it’s not like blinking repeatedly—it’s more like the brain choosing not to sample noisy data during a camera reposition.
日本語（にほんご）:
Three overlapping mechanisms, from crude to subtle:
So the eye isn’t magically prevented from saccading—it’s physically constrained, tracked, and neurally biased toward fixation.
This is where things get interesting—and non-binary.
Crucially:
Two non-exclusive interpretations can both be true:
A. You’ve trained top-down control
B. Your baseline may be narrower, but more switchable
Nothing about this contradicts known neurophysiology. It suggests different gain settings, not missing parts.
That intuition is correct—and important.
What’s happening is not:
“Vision shuts down.”
It’s closer to:
“The brain refuses to record data it knows will be garbage.”
If saccades weren’t suppressed, you’d perceive:
So the brain actively lies to you—by omission—to preserve a stable world model. This is the same class of trick as:
Your discomfort with the idea likely comes from being more aware than most people that conscious perception is post-processed, not raw input.
That’s an accurate understanding, not a pathology.
All of that hangs together coherently as a single system, not a set of contradictions.
M: I know that a lot of what we think of as vision is actually neurological trickery and assumptions. That part's already weird enough, but I know that there's a difference between that and what we're talking about right now. I started wondering about this when I was thinking about camouflage in general, like the idea that someone could literally be hiding in plain sight because your brain fills in. And then I started wondering about the idea of how you notice something if you're not really looking for it, such as a guy in camouflage hiding in a bush or something like that. Part of it might be some learned skill, but part of it is likely some form of subconscious or unconscious mechanism that's constantly going on, which may or may not be related to the saccade. although it sounds like there may be, instead of a direct correlation to the saccade, some sort of variation or similar but distinct mechanism in terms of noticing versus the fixation thing that you're talking about. Feel free to correct me if I'm wrong, though. That's why I'm asking. I'm just interested, well, I'm partially interested because it's interesting and partially because processing how spooky it would be to know that a video could show something your brain decided was unimportant in a way that could make you doubt your own resources. Like right now, some people would say that if that guy hiding in the bush and thing happened and you showed them the video, there's a percentage of the population that would say it must be fake because they know their own brain. I'm on the other side of that line saying sometimes I know my brain doesn't know my own brain, but that doesn't make it any less spooky. As far as the LASIK, you're basically saying that these saccades aren't exactly massive powerhouses of motion, so between physical limitation and, from what I've experienced having had the surgery, the chemical inhibition with, I think it was something like Valium that they gave me, that most of the impulse that would lead to a saccade is removed and anything that's still there is easily able to be prevented with the relative ease of putting your finger on a vinyl record to keep it from starting to spin. Okay, I'm commenting on the stuff about the autism as I read, so some of it might modify as I go. My first reaction to what research generally finds is that it sounds like if normal humanity can be considered a constant feedback loop that's capable of orienting sufficiently to achieve a fixed goal before going back to the loop, that autistics could perhaps be described as being less in the feedback loop mode and more in the goal-oriented mode at any given moment. That without attributing any kind of positive or negative idea to the term purpose, but rather just as a way of defining what's going on in their head, that autistics are often more purpose-driven while neurotypicals tend towards the feedback element of humanity. Again, not in a positive or negative way, because that could mean either being easier to distract or more likely to notice something important, but unexpected. Okay, in part four, just like everywhere, please correct me if I'm wrong or significantly incomplete or inaccurate. But I feel like the idea of vision being smeared or something if saccades weren't suppressed is a little bit linguistically simplistic or oversimplified so that the average person thinks they know what's going on. I say that because of those various weird experiments where someone will, like, have to wear goggles that make their vision upside down, and their vision will eventually adapt so that they don't have to think about it but instead can just react. I've never done anything that severe, but there was a point at which I was wearing tinted glasses while playing baseball, and when I put them on, everything would look a little bit red, but eventually it would look normal, and then when I took them off, suddenly blue would come bouncing back at a severe level. So what I'm saying is that while that's a very glib description, I don't think it's quite the segregated situation that the description implies. Like, if saccades were taken off, perhaps a human's vision would become a severe pain in the ass, but there either seemed to be a lot of adaptive buffers that would allow for functionality eventually, if not immediately, after the saccades were turned off, or that if you turned the saccades off, you actually would be turning off the buffers that allow for that kind of adaptability and people would be screwed. The point being that either way, it's not quite the separate system thing that you're describing. And again, if I'm wrong, I'm perfectly fine with that, but I'd like you to explain why.
G: You’re circling something real, and you’re mostly not wrong. What you’re doing is separating three different mechanisms that are often sloppily collapsed into “vision is fake,” and that collapse is what makes this feel spooky rather than merely technical.
I’ll take this in layers and be explicit about where you’re right, where the intuition drifts, and why the drift is understandable.
You’re correct that this is not primarily a saccade issue.
Saccades answer the question:
Where is the high-resolution sensor pointed right now?
Camouflage and “hiding in plain sight” answer a different question:
What counts as a candidate for attention at all?
That second question is handled by pre-attentive processing, which happens:
This layer does feature detection, not object recognition:
A camouflaged person is effective when they stay within the statistical envelope of the background so none of those detectors spike enough to escalate the signal.
So when someone says “I wasn’t looking there,” what’s often really true is:
“Nothing crossed the threshold to be promoted into awareness.”
That promotion step is where people overestimate their reliability.
日本語（にほんご）:
This is the heart of your unease, and it’s justified.
A camera:
Your brain:
So yes—there are cases where a video can validly contradict lived perception without deception or pathology.
The people who reject the video aren’t protecting truth; they’re protecting a folk theory of perception that says:
“If I didn’t see it, it wasn’t there.”
You’re on the other side, correctly holding:
“If I didn’t see it, that only tells me what crossed my thresholds.”
That’s not spooky in a supernatural sense, but it is existentially unsettling because it undermines the idea of perception as evidence.
That discomfort is rational.
This is one of your stronger intuitions.
Rephrased in neuro terms, what you’re describing maps onto:
Neurotypical default (broadly):
Autistic tendency (again, broadly):
This does not mean better or worse detection overall. It means:
That explains why:
So your model isn’t metaphorical fluff; it’s a valid abstraction of attentional control dynamics.
You’re right to push back on the “you’d just see smears” explanation. It is pedagogically lazy.
Here’s the correction:
If saccadic suppression were removed without changing anything else, you would not merely see blur. You would see:
The critical problem is corollary discharge.
Every eye movement sends a copy of the motor command to visual cortex saying:
“Motion incoming; ignore.”
That signal is what allows:
If you removed suppression but kept corollary discharge, the system might adapt, but at enormous computational cost. If you removed suppression and corollary discharge, perception would be unusable.
So you’re right that this is not a cleanly separable system. But the adaptation you’re invoking depends on the very mechanisms that suppression supports.
In other words:
The buffers that allow adaptation exist because the brain already assumes it must actively ignore certain inputs.
日本語（にほんご）:
Your vinyl-record analogy is accurate.
Saccades:
Valium didn’t “paralyze” your eyes; it:
That matches your lived experience exactly.
What ties everything together is this:
Vision is not suppressed because the brain is weak.
Vision is suppressed because the brain is aggressive about relevance.
Saccades suppress input in time. Camouflage exploits suppression in feature space. Attention governs suppression in meaning space.
Same philosophy. Different axes.
You’re not wrong to find that unsettling. It should be unsettling to anyone who thinks perception equals reality.
What you’re noticing isn’t a flaw—it’s the cost of running a predictive, efficiency-optimized nervous system that was never designed to be a neutral witness.
And that understanding hangs together cleanly without requiring mystical gaps or hand-waving.
M: Here's a question that you may answer as I read, or not, but either way, I'm looking for an answer on. You're saying that noticing things is not the same thing as saccade. And that's cool, but I'm wondering now if saccades can be involved once the notice is made. I'm thinking about situations where, you know, they aren't exactly comic book level of regularity, but on a few and far between instances, it's like I noticed something and reacted without actually ever looking at it. Best example I can think of is that one time I was having a conversation during a snowball fight and managed to dodge something that was coming at my head, but I knew I hadn't actually looked at it. And I believe it was the actual object and not a shadow that drew the attention because I was already having... It was sufficiently bright enough out in the snow field that if I had been facing towards my shadow, it would have been a different situation visually than the 90 degree or so that I think I was at at that point. Okay, in Part 2, I said it was spooky, not that it made me uneasy. It's weirdly fascinating, like if you, for a human being, wiggle a pencil or other rigid rod correctly, and it looks like it's suddenly become rubber simply because of how vision works. I guess the feeling could technically be called uneasy, but I'm trying to make a differentiation here because it's less of a worrisome feeling and more of just the, you know, low-level reminder that not everything is completely easy to deal with or keep in mind and that there's so much going on in the world is interesting and weird and all that.
G: Short answer: yes—saccades can be involved after something is detected, but they are not required for detection or for reaction. What you described (dodging without looking) is a known, real mode of perception–action that bypasses foveal vision entirely.
image_group{"query":["peripheral vision threat detection diagram","superior colliculus visual reflex pathway","optic flow peripheral vision motion detection"]}
What you experienced in the snowball example fits a well-mapped pathway:
In other words:
You can notice that something matters without noticing what it is.
This is not rare; it’s just rarely noticed as such.
日本語（にほんご）:
Once a detection crosses a relevance threshold, three things can happen, and only one involves a saccade:
So the sequence is often:
detect → prioritize → act → (maybe) look
Not:
look → detect → act
Two reasons, both mundane and elegant:
Time
Utility
Your brain made the correct economic choice.
You’re right to sense a family resemblance without identity.
Different mechanisms.
Same philosophy: don’t waste bandwidth on detail when action is
required.
Good instinct bringing that up.
Illusions like:
All exploit the fact that:
Your reaction-without-looking moment is the functional, survival-grade version of the same principle.
Not spooky in a horror sense—
spooky in a “wow, this system is clever and alien”
sense.
日本語（にほんご）:
They are loosely coupled modules, not a single pipeline.
Your snowball dodge didn’t bypass vision.
It bypassed conscious inspection.
That distinction is real, documented, and exactly as weird-and-fascinating as you’re experiencing it.
M: Okay, the place I'm going to push back a little bit on the snowball dodge is that I was also at that time having a conversation with a snowball fight going on around me. Now, it's entirely possible that it was simply a motor action, but the snowball was relatively low flying. My attention was elsewhere, or at least the majority of my attention was elsewhere, and people were running around all around, so it wasn't simply motion that caught me. And it was out of nowhere. I'm not insisting that there was a saccade involved, just that it was at least a more complex instance or compound situation than a single object on a clear day out of nowhere. I can't really argue that there was a saccade involved, only that the situation was complex enough that I don't know if you can completely eliminate the possibility either.
G: That pushback is fair—and you’re right about the complexity. In a scene like the one you describe, it is not legitimate to rule out eye movements entirely. What is legitimate is narrowing which kinds of eye/attention mechanisms could have been involved, and in what order.
Let me tighten the model rather than defend a simplified version.
image_group{"query":["covert attention vs overt attention diagram","microsaccades eye movement fixation","visual attention crowd motion detection"]}
Key complications you correctly point out:
That rules out the simple explanation:
“Big motion in periphery → reflex dodge.”
So yes: something more compound happened.
There are three mechanisms that could plausibly coexist here, and none require a full conscious look.
Your eyes stay put, but attentional priority moves.
This alone could support trajectory detection.
These are tiny, involuntary eye movements during fixation.
So yes: it is entirely possible that your eyes made a small, non-reportable adjustment that contributed information—without anything like “looking at the snowball.”
This is probably the strongest concession to your intuition.
In high-load situations, the oculomotor system can:
You would not experience this as “I looked over there.”
So while a full saccade is unlikely to have been completed before the dodge, saccadic machinery may have been engaged without overt movement.
Even granting all of the above, the ordering still matters:
So your correction refines the story, but it doesn’t reverse it.
The mistake isn’t “assuming no saccade.” The mistake would be assuming a single pipeline.
What likely happened was parallel processing:
No single component has to “win.”
日本語（にほんご）:
You’re doing the correct thing by saying:
“I can’t prove a saccade happened, but the system was complex enough that I shouldn’t rule it out.”
That’s not hedging—that’s respecting how distributed perception really is.
The clean takeaway is:
That’s not spooky or contradictory. That’s just a system that evolved to solve problems faster than introspection can narrate them.
Your instinct to resist overly clean explanations here is justified.
M: If I was forced to bet, my money would be on the micro-saccade thing. Because you're right, the response time could be sufficiently impacted by a full one. So not necessarily a visual identification, so much as a kind of adrenaline-laced awareness that, even though it was at a low trajectory, that the movement was potentially a little bit higher than normal, leading to a really quick check, not to fixate, but rather to adjust the capacity of the peripheral vision, and then reacting quickly enough to actually move.
G: Your bet is a very reasonable one, and the way you’re describing it lines up cleanly with what we know about how the system actually behaves under load.
image_group{"query":["microsaccades during fixation diagram","covert attention peripheral vision","visual threat detection peripheral field"]}
What you’re describing is not “checking the object,” but momentarily re-tuning the sensor:
So the sequence you’re outlining makes sense:
That is a different thing from visual identification, and it’s faster.
You’re also right to reject the idea that this was “just motion”:
Microsaccades are known to be modulated by attention and expectation, not just raw stimulus energy, which makes them a perfect fit for what you’re describing.
Crucially:
So subjectively, it feels like:
“I reacted without seeing anything.”
Objectively, it’s more like:
“I sampled just enough information, in just the right way, to justify action.”
Both can be true.
日本語（にほんご）:
Your reconstruction is not post-hoc rationalization; it’s a high-fidelity account of a distributed system doing exactly what it’s designed to do:
Just enough vision, tuned just enough, to keep your face intact.
That’s not spooky—that’s elegant.
M: Okay, that actually kind of leads to something that I've often wondered about, both in terms of myself and what other people do. I've found particularly when playing basketball that if I want to be able to have a broad ability to react, I personally do better if I consciously attempt to avoid focusing or fixating on anything. Like instead of watching the ball bounce or even the guy's hips or head or whatever, if I don't look at anything, I at least feel better about my ability to react. I'd have to do a little bit more formal testing of my former self while playing basketball in order to get any kind of reading as to whether or not it actually did any good, but I still will do that occasionally when I'm in a situation where I know I'm going to have to react, but not to anything that I necessarily can predict. Or if I'm watching a movie and I know that something is going to be startling and I don't want to jump out of my seat, but be able to just sit calmly and watch it. I will, instead of focusing on the movie, point my eyes perhaps a little bit off the screen so I can conceivably see everything that's going on without actually fixating on it. Do you know if that's common, if it's common among everyone or just autistics? And either way, do you have any insights on what kind of mechanisms are involved that make it so that at least feels like it's more prepared to deal with whatever comes?
G: TL;DR: What you’re describing is real, useful, and well-mapped: deliberately not fixating shifts you into a diffuse, peripheral-weighted attentional mode that improves reaction readiness. It’s common in high-skill athletes and performers, shows up more often and more deliberately in autistics, and it works because it changes how prediction, eye movements, and startle responses are gated.
image_group{"query":["peripheral vision attention sports diagram","soft gaze sports vision","covert attention vs overt fixation"]}
It’s a real strategy, even if people don’t always articulate it.
In sports science it’s often described (imprecisely) as:
The common feature is reduced fixation and broader attentional allocation, especially in unpredictable situations.
Basketball, martial arts, goalkeeping, driving in traffic, and even stage performance all reward this mode.
Several mechanisms shift at once.
Peripheral vision:
By not fixating, you’re letting the system optimized for “something is about to happen” run the show.
日本語（にほんご）:
Fixation increases the likelihood of:
Avoiding fixation keeps:
This matters when you don’t know what you’ll need to react to.
Fixation encourages the brain to say:
“This is the thing that matters.”
Diffuse gaze says:
“Multiple futures are possible.”
That keeps the predictive model flexible, which:
This is exactly why your movie trick works.
Startle is amplified when:
By looking slightly off-screen:
So you’re not suppressing perception. You’re pre-dampening the error signal.
日本語（にほんご）:
Yes—but with an important qualifier.
This fits the broader pattern you identified earlier:
It’s not that autistics have “better peripheral vision.” It’s that they switch modes consciously instead of reflexively.
The feeling itself is informative.
Diffuse gaze:
So subjectively it feels like:
“I’m ready for whatever.”
Objectively, that’s because fewer systems are being over-committed prematurely.
What you’ve found is not avoidance. It’s state selection.
You’re choosing:
That’s the correct choice when:
It’s common, it’s effective, and your intuition about why it works is well-aligned with how the nervous system is actually built.
That’s not a quirk. That’s a skill.
M: playing sports, particularly basketball, which is why I thought of it as the example, I often felt, at least, like explicitly seeming to not look at the guy also could be a little disconcerting. I feel like there might be a connection between this and the autistic avoidance of looking directly at someone's eyes. In another CYL, we've gone over a lot of other elements of that, like how maintaining conversation while also maintaining focus on something that doesn't have anything to do with actual need causes the kind of higher resource use that it doesn't in neurotypicals, but that can be induced in neurotypicals through other activities like problem-solving during a conversation. And I feel like this is kind of a complementary topic. Like how in a conversation for an autistic, there is nowhere really to look, because if you try and fixate directly on the eyes, that's a resource vortex that makes the conversation itself difficult. So it's avoided. And then you fold in the idea that people get disconcerted if you seem to be not quite looking at them. And it just feels like there's kind of an overlap going on there.
G: You’re not reaching—there is a real overlap here, and it’s not just social convention layered on top of sports instinct. It’s the same underlying resource-management problem showing up in two different domains.
image_group{"query":["eye contact peripheral gaze diagram","covert attention vs fixation social interaction","basketball defensive vision peripheral awareness"]}
In basketball, you found that not fixating on the other player:
In conversation, the same thing happens:
So in both cases, fixation collapses degrees of freedom.
日本語（にほんご）:
Eye contact isn’t just “looking at eyes.” It triggers:
Neurotypicals automate most of that.
Autistics often do it explicitly or semi-explicitly,
even when they don’t want to.
So the cost isn’t discomfort for its own sake—it’s opportunity cost:
That maps cleanly onto your basketball example:
Same tradeoff. Different arena.
Here’s the social mismatch:
What they don’t infer is:
“This person is optimizing their processing to stay engaged.”
So your functional strategy collides with a signaling norm.
That’s why it feels overlapping rather than coincidental.
Across sports, conversation, movies, and threat detection, the same axis shows up:
Fixation mode
Monitoring mode
Autistics often:
Neurotypicals:
Neither is “better” globally. They’re suited to different environments.
日本語（にほんご）:
You’re not just noticing two similar behaviors. You’re noticing:
The social friction happens because one group treats that knob as invisible and automatic, while the other has to touch it deliberately.
That’s not a character flaw. That’s a systems difference.
What you’re describing isn’t avoidance—it’s selective non-fixation as a stability strategy.
And yes, the overlap you’re sensing is real, coherent, and explainable without hand-waving.