Smart people/Scientists, weigh in on this

SkyHog

SkyHog

Touchdown! Greaser!
Joined
Feb 23, 2005
Messages
18,431
Location
Castle Rock, CO
Display Name
Display name:
Everything Offends Me
I was driving, and almost got into an accident when something occurred to me. The way I understand our eyesight to work is that light is projected through the hole thingie in our eyes and is received by nerves inside the eye, which the brain processes into what we see.

That means it can't possibly be realtime. The delay must be so miniscule to almost never matter, but there must be times where something goes wrong...

Is it plausible for the eyes to see something as being a miss, only to have it hit? For example, a car accident, a baseball to the face, being punched, etc.?
 
SkyHog said:
I was driving, and almost got into an accident when something occurred to me. The way I understand our eyesight to work is that light is projected through the hole thingie in our eyes and is received by nerves inside the eye, which the brain processes into what we see.

That means it can't possibly be realtime. The delay must be so miniscule to almost never matter, but there must be times where something goes wrong...

Is it plausible for the eyes to see something as being a miss, only to have it hit? For example, a car accident, a baseball to the face, being punched, etc.?
Click to expand...

Yes, there is a delay for processing, and it isn't exactly minuscule. More accurately we experience a phenom known as "persistence of vision". Sight is analogue, not digital. But, since it takes time to process the image, the image exists for a fraction of a second longer than the event. Throw another image at it before the previous image is gone and it appears to be continuous. It is the entire basis of the movie industry. I don't remember the exact number but it is somewhere around a 16th of a second(?). I think 16 frames per second is a minimum for movie projection.

Now if you are required to DO something besides just process an image then the delay can be as much a second or so for a mature non-athlete.
 
Last edited:
What I've read is visual processing time is ~ 0.30 seconds from image capture to recognition.
 
SkyHog said:
Is it plausible for the eyes to see something as being a miss, only to have it hit? For example, a car accident, a baseball to the face, being punched, etc.?
Click to expand...

You will have to alter the title for me to answer, but YES the eyes are not perfect; actually its not the ocular mechanism or the optic nerves...I am pretty sure it is the CPU that screws things up. Ever hear someone say they saw something which is widely accepted to be impossible? I think the brain pieces together info and sometimes comes to the wrong conclusion....and then presents it to the person as 'the best fit', even confusing the person about what they actually saw.
Also I don't think the processing time is much of a factor here.
Glad your cpu didn't mess up so much that you can't tell us about it!
 
SkyHog said:
Is it plausible for the eyes to see something as being a miss, only to have it hit? For example, a car accident, a baseball to the face, being punched, etc.?
Click to expand...

A delay in recieving a photo won't make the photo wrong, if you want to take the analogy of the eye as a camera passing photos up to the brain. The brain is getting (slightly) old news.

As others have said, the much bigger issue is that the brain takes in a huge amount of visual data then interprets it into a picture of what's going on (hence all the visual illusions). It's not a camera at all, but a data interpreter/simplifier that can be fooled all to easily.
 
Certainly the system can be fooled- that is the basis for optical illusions. Also the basis for camouflage- break up an outline in the right way and we miss seeing the ship or plane.

The system isn't exactly simple. Rods detect light intensity while cones detect color and these receptors are not uniformly distributed in the eye (remember your lessons for night flight).

The "refresh rate" is on the order of 15-16 "frames per second" so it is possible for a fast moving object to escape our notice, especially if something is "strobing" around us such as a light on 50-60 HZ AC or the refresh on a TV set.

As others have mentioned, there is an awful lot of data processing going on and a number of simplifications are being made to pick out patterns.
 
Any NHRA fans out there? Isn't there a minimum time limit between when the light goes green and you leave the line where it's considered a false-start because there's 'no way' that anybody's reaction time is that quick?

(Unless Johnny Fever is in the house)
 
Steve said:
What I've read is visual processing time is ~ 0.30 seconds from image capture to recognition.
Click to expand...
I was just watching a show about visual processing and they were saying that under normal condition we process at about 30 frames a second (.033 seconds) and that under stress that rate can increase greatly which is why when we are involved in great stress like an accident we seem to perceive time differently.
 
SkyHog said:
Is it plausible for the eyes to see something as being a miss, only to have it hit? For example, a car accident, a baseball to the face, being punched, etc.?
Click to expand...
Only if, during that delay time, someone rushes out there to perform a "do-over" so quickly that the eye doesn't catch the do-over.
 
That may be true for visual acquistion, but actual response takes longer. The signal speed along the neural paths is finite.

smigaldi said:
I was just watching a show about visual processing and they were saying that under normal condition we process at about 30 frames a second (.033 seconds) and that under stress that rate can increase greatly which is why when we are involved in great stress like an accident we seem to perceive time differently.
Click to expand...
 
Dart said:
Yes, there is a delay for processing, and it isn't exactly minuscule. More accurately we experience a phenom known as "persistence of vision". Sight is analogue, not digital. But, since it takes time to process the image, the image exists for a fraction of a second longer than the event. Throw another image at it before the previous image is gone and it appears to be continuous. It is the entire basis of the movie industry. I don't remember the exact number but it is somewhere around a 16th of a second(?). I think 16 frames per second is a minimum for movie projection.

Now if you are required to DO something besides just process an image then the delay can be as much a second or so for a mature non-athlete.
Click to expand...


24fps. All in all, it takes on average about 1/3 of a second between viewing and reaction, though this can be considerably shortened in extreme situations.
 
Matthew said:
Any NHRA fans out there? Isn't there a minimum time limit between when the light goes green and you leave the line where it's considered a false-start because there's 'no way' that anybody's reaction time is that quick?

(Unless Johnny Fever is in the house)
Click to expand...


Negative sir, if the green hits .00000000001 before you break beam, it's good. That's part of staging technique, how shallow/deep into the staging beam you put your wheel. Personally, I like staging very shallow and stepping off as a reaction to the last yellow. That gives me a .2 second inertial advantage over a guy with the same reaction time as I but staged deep and waiting for the green, and there have been times where the race was won or lost by less.
 
We talked about this quite a bit when I worked at the Big Yellow Box (the insiders name for Eastman Kodak). Like most physical limitations it's also dependent on the person. Some people are actually able to detect 50hz at least in peripheral views. This can be quite disturbing in Europe where the power grid is at 50Hz instread of the 60Hz here in the US. We are also effected in a physical way by some flicker rates and mixtures of rates. There was a folk lore story about one of the photo scientists I worked with doing a demonstration to upper management when 3D films were first beginning to be shown. His point was that there is a physical danger if it is done incorrectly. I don't remember the exact percentage but MANY of the execs left the demonstration early due to ...."accute nausia"...

Vision also acts in a quite non-linear way. Intensity, contrast, motion and even spacial frequency all effect what we perceive. Some day when your driving down the road and you go under a bridge with fencing on both sides if you look closely you can see a lower frequency "pattern" of the fence embedded in the view. This is easily explained in the mathematics and is called aliasing.

Then there are general Optics issues. We all have some degree of imperfection in the lens and shape of our eyes. Visual distortion due to these imperfections can cause a number of "errors" in the visual input. limited sight range, limited accuity etc.

The long and the short of it for me is that our vision is nothing short of a miracle. Even with it's imperfections it offers us an interface to the world around us with enormous information bandwidth. That coupled with the brains information algorithms we can actually train for specific purposes allow us to overcome some of the limitations by being able to predict future events and therefore overcome some of those limitations.

I've rattled on enough. I'm just an electrical guy, maybe there's an optics guru among us with some more interesting insights.
 
MikeG said:
We talked about this quite a bit when I worked at the Big Yellow Box (the insiders name for Eastman Kodak). Like most physical limitations it's also dependent on the person. Some people are actually able to detect 50hz at least in peripheral views. This can be quite disturbing in Europe where the power grid is at 50Hz instread of the 60Hz here in the US. We are also effected in a physical way by some flicker rates and mixtures of rates.
Click to expand...

I have a problem with this and flourecent lights even at 60hz, they really annoy me as it's like always being under a low grade strobe.
 
Henning said:
Negative sir, if the green hits .00000000001 before you break beam, it's good. That's part of staging technique, how shallow/deep into the staging beam you put your wheel. Personally, I like staging very shallow and stepping off as a reaction to the last yellow. That gives me a .2 second inertial advantage over a guy with the same reaction time as I but staged deep and waiting for the green, and there have been times where the race was won or lost by less.
Click to expand...

Yep... green means go. The way you win the race is not by actually waiting until the green, but launching at that exact instant before the green so that you pass 0.0000000000000001 after the greens are lit up. There are even practice trees you can buy to get your reaction time better, but they involve a hand-held button, which is not exactly like letting off the brake/hitting the gas. None of my friends who drag race use them.

Alternately, you could be racing against me, in which case you'll probably win because I'm pretty bad at hole shots. Give me a road course, however, and it's another matter entirely...
 
Henning said:
I have a problem with this and flourecent lights even at 60hz, they really annoy me as it's like always being under a low grade strobe.
Click to expand...

Most industrial fluorescents actually run at 20 KHz. Even the old style "flashes" at twice the line frequency (e.g. 120 Hz in the US). I'm pretty sure there's no way you can detect either directly, but if you move the lamp (or your head) rapidly you will see multiple images of it.
 
Henning said:
I have a problem with this and flourecent lights even at 60hz, they really annoy me as it's like always being under a low grade strobe.
Click to expand...

Incandescent lights ought to be bothersome as well since they will strobe at double the line frequency as well. Spinning toys can demonstrate this effect although flourescent lights seem to show it better.
 
Ted DuPuis said:
Yep... green means go. The way you win the race is not by actually waiting until the green, but launching at that exact instant before the green so that you pass 0.0000000000000001 after the greens are lit up. There are even practice trees you can buy to get your reaction time better, but they involve a hand-held button, which is not exactly like letting off the brake/hitting the gas. None of my friends who drag race use them.

Alternately, you could be racing against me, in which case you'll probably win because I'm pretty bad at hole shots. Give me a road course, however, and it's another matter entirely...
Click to expand...

Depends, if you run with an automatic type transmission, you'll probably have a tranny brake, which is basically a solenoid which opens valves for 1st gear and reverse to hold the car in place with the throttle mashed, let go of the button, it releases the reverse clutches and you launch.
 
Henning said:
Depends, if you run with an automatic type transmission, you'll probably have a tranny brake, which is basically a solenoid which opens valves for 1st gear and reverse to hold the car in place with the throttle mashed, let go of the button, it releases the reverse clutches and you launch.
Click to expand...

Good point. I always run manual (part of why my holeshots aren't so good), and so do my friends. Unless I wanted to take the Excursion and drag race it. I forgot about the fact that many people who drag race will run automatics (shows how much attention I pay to it).

Of course, the other thing that the trainer tree doesn't account for is the delay in your car. My understanding was that's always what people are going for - that launch just enough before that you won't get a red light, and each car/bike/whatever will have some sort of independent delayed response. I think the trainer tree just sees how close past the green you can do.

Once again, I prefer road racing and only have a few friends who drag race, so I'm no authority on this stuff. :)
 
smigaldi said:
I was just watching a show about visual processing and they were saying that under normal condition we process at about 30 frames a second (.033 seconds) and that under stress that rate can increase greatly which is why when we are involved in great stress like an accident we seem to perceive time differently.
Click to expand...

Ahh, very interesting. That actually makes sense.
 
THIS is sort of interesting - apparently our brain "fills in" and generates an image of what's going to happen 1/10 sec into the future, so we're prepared for it!
 
You must log in or register to reply here.
Back
Top