One cool reason for this is that auditory input lasts longer in your mind than visual. I don’t know if we’ve tested touch, smell, or taste, but maybe someone else has those research articles ready.
Vision lasts for something under a second. You can test by flashing a grid of numbers, say 5x5, up on a screen for a very brief period of time, and then cue the subject to recall a specific line after a period of time. There are too many numbers to instantly memorize, too many (more than the 7 +/-2 limit we typically see) to rehearse to remember all of them, and the subject doesn’t know which line to rehearse anyway, so they can’t repeat all the numbers to ‘win.’ I remember that the time until decay was something like a second. Up to that point, a subject could repeat all the numbers in a line if cued.
Auditory was tested in a similar manner, but I can’t quite remember the details. Maybe something like a high/medium/low voice, and then being cued? Anyway, time for decay for the sensory input was something like 4-8 seconds. Most people have probably had an experience where they were working on a task, a coworker or roommate came up and asked them what they wanted for lunch, and the reply was, “hang on a second, let me finish this,” they finish the (very brief) task, turn to the other person and say, “what did you say?” while screwing up their face in memory and then answer the question without it being repeated.
As an interesting aside, chimpanzees have amazing visual recall from brief cues. I remember videos of them being able to memorize a sequence of 20+ squares to press on a grid after the correct sequence had been flashed up for a split second.
… and my subsequent hypothesis would be that… the extent to, and manner in which that has occured is very far from uniform in the entire currently existing human species.
See my other comments around 50% - 70% of the population just not having an inner monologue, not actively using that complex language processing internally in a conscious manner, seemingly because for them, those abilities are… underdeveloped? not prioritized by the brain? maybe not practiced, learned, taught?
Like, am I wrong here, or … are you not really sapient if you don’t/can’t do that?
Sentient, of course, but… sapient?
How do you ‘make decisions’ if… you’re basically just acting instinctually, not actually ‘figuring it out’ in some kind of actual semantic/logical sense?
Aren’t you basically just … the sum of your experiences at that point, not being actually capable of metacognition?
Like, one large problem with trying to make ‘AI’ right now… is that it can emulate human-ish discourse fairly well… but it can’t maintain consistency, it can’t metacognate, it has a process for how to formulate a response… but it can’t evaluate the process by which it formulates responses, consider modifications to that process, potentially make those modifications.
Its about as ‘honest and consistent’ as a person with no internal monologue, no self evaluation, just responds to inputs based on its version of experience and instinct (data and trained maps of weights for symbolic associations).
The human brain also generally responds faster, with lower latency, to audio input than it does to visual input.
This is why actual pro gamers will prioritize a positionally accurate, low latency audio setup at least as much as, if not more than a ludicrous visual framerate that goes far beyond the human ability to perceive.
One cool reason for this is that auditory input lasts longer in your mind than visual. I don’t know if we’ve tested touch, smell, or taste, but maybe someone else has those research articles ready.
Vision lasts for something under a second. You can test by flashing a grid of numbers, say 5x5, up on a screen for a very brief period of time, and then cue the subject to recall a specific line after a period of time. There are too many numbers to instantly memorize, too many (more than the 7 +/-2 limit we typically see) to rehearse to remember all of them, and the subject doesn’t know which line to rehearse anyway, so they can’t repeat all the numbers to ‘win.’ I remember that the time until decay was something like a second. Up to that point, a subject could repeat all the numbers in a line if cued.
Auditory was tested in a similar manner, but I can’t quite remember the details. Maybe something like a high/medium/low voice, and then being cued? Anyway, time for decay for the sensory input was something like 4-8 seconds. Most people have probably had an experience where they were working on a task, a coworker or roommate came up and asked them what they wanted for lunch, and the reply was, “hang on a second, let me finish this,” they finish the (very brief) task, turn to the other person and say, “what did you say?” while screwing up their face in memory and then answer the question without it being repeated.
As an interesting aside, chimpanzees have amazing visual recall from brief cues. I remember videos of them being able to memorize a sequence of 20+ squares to press on a grid after the correct sequence had been flashed up for a split second.
And the Cognitive Tradeoff Hypothesis would say we gave up that ability to get complex language.
… and my subsequent hypothesis would be that… the extent to, and manner in which that has occured is very far from uniform in the entire currently existing human species.
See my other comments around 50% - 70% of the population just not having an inner monologue, not actively using that complex language processing internally in a conscious manner, seemingly because for them, those abilities are… underdeveloped? not prioritized by the brain? maybe not practiced, learned, taught?
Like, am I wrong here, or … are you not really sapient if you don’t/can’t do that?
Sentient, of course, but… sapient?
How do you ‘make decisions’ if… you’re basically just acting instinctually, not actually ‘figuring it out’ in some kind of actual semantic/logical sense?
Aren’t you basically just … the sum of your experiences at that point, not being actually capable of metacognition?
Like, one large problem with trying to make ‘AI’ right now… is that it can emulate human-ish discourse fairly well… but it can’t maintain consistency, it can’t metacognate, it has a process for how to formulate a response… but it can’t evaluate the process by which it formulates responses, consider modifications to that process, potentially make those modifications.
Its about as ‘honest and consistent’ as a person with no internal monologue, no self evaluation, just responds to inputs based on its version of experience and instinct (data and trained maps of weights for symbolic associations).
The human brain also generally responds faster, with lower latency, to audio input than it does to visual input.
This is why actual pro gamers will prioritize a positionally accurate, low latency audio setup at least as much as, if not more than a ludicrous visual framerate that goes far beyond the human ability to perceive.