Norma Graham makes an interesting point, which I've seen quoted many times, in her book on spatial vision. She notes that it is as if the brain is transparent to what is happening in the early levels of visual processing, and that this is curious. It's curious, but it's a typical stance for someone who studies spatial vision; we assume that discrimination or identification or detection of signal strength is mediated almost entirely by the filters that are transducing the signal, not those that understand it, or respond (overtly) to it.
Whether or not this transparency holds when complex images are viewed is, I think, totally unknown. It may be that the percept elicited by a complex scene really is the sum of its parts, and that it simply provokes additional sensations of meaning, identity, extension, etc., which are tied to spatial locations within the scene. So, the visible scene that we are conscious of is indeed an object of spatial vision. This is the point of view I generally adopt, and I think it is common.
Another view is that the percept is entirely inference. Boundaries, surfaces, colors, textures, etc., are qualities in themselves, inferred from particular organization of spatial structure, and these then are organized in such a way that objects and identities and meanings can be inferred in successive stages. These inferences are what is seen consciously; perhaps inferences, and the evidence for them (the matter of spatial vision), are experienced simultaneously, but the substantive inferences, being the important elements of experience, are what dominate consciousness. So, only a small part of the phenomenal scene is actually constituted by e.g. luminance contrasts, and much more of it is constituted by higher level inferences. I think this point of view is also common, maybe especially in the current generation of visual neuroscientists.
The latter view is not exclusive of Graham's observation. If the patterns that are viewed are simple enough, they will not form objects, and will not have meaning. Or, they will be interpreted only as what they are, which doesn't require much inference, or only circular inference (which isn't a bad thing necessarily, when you really do want to conclude that a thing is itself, e.g. a gaussian blob of light, on the basis of its being a blob of light; usually, you want to infer that there is a letter on a page on the basis of a particular arrangement of blobs of light).
So, in the experiment that I'm currently analyzing to death, I am clearly taking the first view, in which case I think my conclusions are solid. If the second view is more accurate, what does the result mean? It could mean that inferences about image strength are based on higher frequencies just because they are the more susceptible to loss in a weak signal. If I'm asking subjects to judge image contrast, they could easily interpret this as judging image strength, and then their judgments would be biased towards the most delicate parts of the image, but they would still take everything into account.
This latter interpretation is still interesting, but it doesn't require "suppression". It is worth mentioning and I should at least include it in the manuscript, although the FVM talk probably will not have space... already there's barely space for the default story.
Friday, August 17, 2012
Thursday, August 16, 2012
nausea, photophobia, headache
woke up slightly late today, started to feel sick on the train; slightly nauseated all day long (mainly manifesting as excessive salivation, gross), a bit of photophobia. might be eyestrain or due to my crooked (broken, $10) glasses. briefly on tuesday night inverted the contrast on firefox, but then switched it back. at the time it was because my glasses seemed smudgy and i just couldn't get them clear, thought, "ah, it's a sign!", but decided no, it's the glasses. right now definitely a slight headache. as usual in my forehead, just above my eyes, maybe a bit left of center.
quick note
Something I've been meaning to write a note on for a while. Even as I write this, I don't have the idea quite in my head.
What is it about apparent contrast that is interesting, as opposed to sensitivity? By most measures, they're the same thing. If I ask someone to do an experiment where they have to discriminate between one contrast and another, I'm assuming that they are doing some mental comparison between two apparent contrasts, or memories of two apparent contrasts. By varying the physical difference between the two contrasts, I can then quantify the person's performance on this task, e.g. the difference at which they can no longer measurably discriminate. These are the sorts of measurements that are usually made in psychophysics.
These performance measures are understood to reflect the subjective, phenomenal properties of interest. But they involve a nearly unsolvable confound: we can't tell what part of the discrimination is due to relative internal response strength and what part is due to internal noise.
So to me, measuring apparent contrast is a way at getting around this problem. You get to measure, directly, the internal response to the stimulus. The new problem, then, is in quantifying what you've measured. The reverse correlation experiment that I did, I realize now, was cognizant of all this, but it was subliminal for me. The experiment is not measuring performance, but it is very similar to an experiment that would be measuring discrimination performance. In this experiment, the stimuli are always easily discriminable, so there are no limits to measure. The subject is asked to discriminate between the strengths of the two stimuli, but I measure no interval or reliability of this discrimination. This is because there is no objective stimulus strength.
The purpose of the experiment is to find out what constitutes stimulus strength when strength is defined, to an observer, as luminance contrast. What I get back is (no matter how I measure it) a description of what components count more or less than other components in making decisions about stimulus strength. I then test a bunch of plausible models to see whether they might also count components in similar ways. Lucky for me, only a particular type of model works, so I can make a sort of conclusion from the study.
So, apparent contrast is the way things look, and then behaviors can be carried out on the basis of how things look. Most visual psychophysics directly analyzes the behaviors that are based on appearances. I've tried to directly analyze the appearances themselves. Did I succeed?
What is it about apparent contrast that is interesting, as opposed to sensitivity? By most measures, they're the same thing. If I ask someone to do an experiment where they have to discriminate between one contrast and another, I'm assuming that they are doing some mental comparison between two apparent contrasts, or memories of two apparent contrasts. By varying the physical difference between the two contrasts, I can then quantify the person's performance on this task, e.g. the difference at which they can no longer measurably discriminate. These are the sorts of measurements that are usually made in psychophysics.
These performance measures are understood to reflect the subjective, phenomenal properties of interest. But they involve a nearly unsolvable confound: we can't tell what part of the discrimination is due to relative internal response strength and what part is due to internal noise.
So to me, measuring apparent contrast is a way at getting around this problem. You get to measure, directly, the internal response to the stimulus. The new problem, then, is in quantifying what you've measured. The reverse correlation experiment that I did, I realize now, was cognizant of all this, but it was subliminal for me. The experiment is not measuring performance, but it is very similar to an experiment that would be measuring discrimination performance. In this experiment, the stimuli are always easily discriminable, so there are no limits to measure. The subject is asked to discriminate between the strengths of the two stimuli, but I measure no interval or reliability of this discrimination. This is because there is no objective stimulus strength.
The purpose of the experiment is to find out what constitutes stimulus strength when strength is defined, to an observer, as luminance contrast. What I get back is (no matter how I measure it) a description of what components count more or less than other components in making decisions about stimulus strength. I then test a bunch of plausible models to see whether they might also count components in similar ways. Lucky for me, only a particular type of model works, so I can make a sort of conclusion from the study.
So, apparent contrast is the way things look, and then behaviors can be carried out on the basis of how things look. Most visual psychophysics directly analyzes the behaviors that are based on appearances. I've tried to directly analyze the appearances themselves. Did I succeed?
Monday, August 13, 2012
unifications of china 1
Random idea from this weekend: create a set of spatiotemporal maps illustrating the unifying conquests of China. For fun. Let's make a list:
1. Qin: Ying Zheng and Guanzhong
Qin was one of many Warring States in the centuries leading up to the first true unification of China in 9780HE. Qin was based in the area around and to the west of Xi'an, which is protected by mountain ranges and accessible only through narrow passes (I traveled through the Hangu pass to visit Xi'an in 12010HE): hence the region's name of Guanzhong, "within the passes". The conquest has an ill-defined starting point, since the different states had been in contention for centuries. However, it was with Ying Zheng's rule that most of the work was done: between 9771 and 9780, China proper went from seven states to one. This period, the Qin Unification War, could be taken as the first.
2. Han: Liu Bang and Guanzhong
Qin didn't long outlive Ying Zheng, who died in 9791. Soon after his death, Qin was overthrown and broken up into a number of kingdoms, united in theory by the Emperor of Chu, who was in fact a puppet of the warlord Xiang Yu. Xiang Yu's confederation quickly disintegrated into civil war between Xiang's Chu state and Liu Bang's Han state, which lasted from 9795 to 9799, when Chu was finally defeated and absorbed into Han. Han, by the way, based its power in the mountain-protected cities of Hanzhong and Chang'an, as Qin had done. Xiang Yu had placed his capital in Pengcheng, in eastern China. This war, and its result, was messy: at any given point in time, even for a decade after the war ended, it was unclear just who was in charge of particular regions, and a lot rested on the proclaimed allegiances of one or another warlord. However, there are standard interpretations of who was with who and when, that could be used to clarify an illustration.
3. Wei/Jin: Cao Cao and Guandong (east of the passes)
Han lasted for 400 years. When it finally collapsed around 10190, there were more than 20 years of war, followed by a half-century period of fracture into the Three Kingdoms of Wei, Wu, and Shu. The Wei state, based in the western edges of the central plains, just east of the mountain strongholds favored by Qin and early Han, eventually conquered Shu in 10263, was replaced in a coup by Jin in 10265, and finally conquered Wu in 10280. After this, Jin slowly fell apart, and China wouldn't be put together under one government again for another four centuries. The Wei/Jin unification was so slow, taking more than 80 years, that it can't really be considered a 'conquest'; it was a slow succession of local wars, with long spaces of quiet in between. I don't think this one would count.
I don't know much about the establishment of Sui - we'll wait until I've read a bit more on it before I continue.
1. Qin: Ying Zheng and Guanzhong
Qin was one of many Warring States in the centuries leading up to the first true unification of China in 9780HE. Qin was based in the area around and to the west of Xi'an, which is protected by mountain ranges and accessible only through narrow passes (I traveled through the Hangu pass to visit Xi'an in 12010HE): hence the region's name of Guanzhong, "within the passes". The conquest has an ill-defined starting point, since the different states had been in contention for centuries. However, it was with Ying Zheng's rule that most of the work was done: between 9771 and 9780, China proper went from seven states to one. This period, the Qin Unification War, could be taken as the first.
2. Han: Liu Bang and Guanzhong
Qin didn't long outlive Ying Zheng, who died in 9791. Soon after his death, Qin was overthrown and broken up into a number of kingdoms, united in theory by the Emperor of Chu, who was in fact a puppet of the warlord Xiang Yu. Xiang Yu's confederation quickly disintegrated into civil war between Xiang's Chu state and Liu Bang's Han state, which lasted from 9795 to 9799, when Chu was finally defeated and absorbed into Han. Han, by the way, based its power in the mountain-protected cities of Hanzhong and Chang'an, as Qin had done. Xiang Yu had placed his capital in Pengcheng, in eastern China. This war, and its result, was messy: at any given point in time, even for a decade after the war ended, it was unclear just who was in charge of particular regions, and a lot rested on the proclaimed allegiances of one or another warlord. However, there are standard interpretations of who was with who and when, that could be used to clarify an illustration.
3. Wei/Jin: Cao Cao and Guandong (east of the passes)
Han lasted for 400 years. When it finally collapsed around 10190, there were more than 20 years of war, followed by a half-century period of fracture into the Three Kingdoms of Wei, Wu, and Shu. The Wei state, based in the western edges of the central plains, just east of the mountain strongholds favored by Qin and early Han, eventually conquered Shu in 10263, was replaced in a coup by Jin in 10265, and finally conquered Wu in 10280. After this, Jin slowly fell apart, and China wouldn't be put together under one government again for another four centuries. The Wei/Jin unification was so slow, taking more than 80 years, that it can't really be considered a 'conquest'; it was a slow succession of local wars, with long spaces of quiet in between. I don't think this one would count.
I don't know much about the establishment of Sui - we'll wait until I've read a bit more on it before I continue.
Subscribe to:
Comments (Atom)