Hallucinations are easier to correct for? Hm. So, I start out with an input channel whose average throughput rate is T1, and whose reliability is R1. Case 1, I reduce that throughput to T2. Case 2, I reduce the reliability to R2.
A lot seems to depend on T2/T1 and R2/R1. From what I’ve gathered from talking to blind people, I’d estimate that T2/T1 in this case is ~.1. That is, sighted people have approximately an order of magnitude more input available to them than blind people. (This varies based on context, of course, but people have some control over their context in practice.) Hallucinations vary. If I take as my example the week I was in the ICU after my stroke, I’d estimate that R2/R1 is ~.1. That is, any given input was about ten times more likely to not actually correlate to what another observer would see than it usually is.
Both of these estimates are, of course, pulled out of my ass. I mention them only to get some precision around the hypothetical, not as an assertion about what blindness and hallucination are like in the real world. If you prefer other estimates, that’s fine.
Given those estimates… hm. Both of them suck. I think I would probably choose hallucination, in practice. I think I would probably be better off choosing blindness.
False information is definitely more damaging than non-information, because in the best case scenario you ignore the false information. In less-than-best-case scenarios, you fail to ignore the false information and are actively misled.
Suppose there are 10 boxes, one of which contains cash.
If you could open the boxes and see which one had cash, you’d be in great shape.
But if you can’t, you obviously should prefer leaving all the boxes closed (blindness), rather than somehow seeing cash in box #7 even when it isn’t there.
I think the only reason people would be tempted to choose hallucination is that hallucinations in real life are usually relatively mild and often correctible, whereas blindness can be total and intractable with present technology. So given the choice between schizophrenia and blindness, I probably would choose schizophrenia, because schizophrenia is treatable.
One reason I would be tempted to choose hallucination over blindness is that hallucinations feel like knowledge, and blindness feels like lack of knowledge, and I’m more comfortable with the feeling of knowledge than I am with the feeling of the lack of knowledge.
Depends on if you’re hallucinating everything or your vision has at least some bearing in the real world. I mean, I’d rather see spiders crawling on everything than be blind, since I could still see what they were crawling on.
Not 100% sure they would be for me: I’m not arachnophobic at all. (I would be willing to eat a spider for five dollars, if I was sure this couldn’t cause be to get sick.)
Thomas Hardy
Pretty sure most people would pick hallucinations over blindness. Easier to correct for.
Hallucinations are easier to correct for?
Hm.
So, I start out with an input channel whose average throughput rate is T1, and whose reliability is R1.
Case 1, I reduce that throughput to T2.
Case 2, I reduce the reliability to R2.
A lot seems to depend on T2/T1 and R2/R1.
From what I’ve gathered from talking to blind people, I’d estimate that T2/T1 in this case is ~.1. That is, sighted people have approximately an order of magnitude more input available to them than blind people. (This varies based on context, of course, but people have some control over their context in practice.)
Hallucinations vary. If I take as my example the week I was in the ICU after my stroke, I’d estimate that R2/R1 is ~.1. That is, any given input was about ten times more likely to not actually correlate to what another observer would see than it usually is.
Both of these estimates are, of course, pulled out of my ass. I mention them only to get some precision around the hypothetical, not as an assertion about what blindness and hallucination are like in the real world. If you prefer other estimates, that’s fine.
Given those estimates… hm.
Both of them suck.
I think I would probably choose hallucination, in practice.
I think I would probably be better off choosing blindness.
False information is definitely more damaging than non-information, because in the best case scenario you ignore the false information. In less-than-best-case scenarios, you fail to ignore the false information and are actively misled.
Suppose there are 10 boxes, one of which contains cash.
If you could open the boxes and see which one had cash, you’d be in great shape. But if you can’t, you obviously should prefer leaving all the boxes closed (blindness), rather than somehow seeing cash in box #7 even when it isn’t there.
I think the only reason people would be tempted to choose hallucination is that hallucinations in real life are usually relatively mild and often correctible, whereas blindness can be total and intractable with present technology. So given the choice between schizophrenia and blindness, I probably would choose schizophrenia, because schizophrenia is treatable.
One reason I would be tempted to choose hallucination over blindness is that hallucinations feel like knowledge, and blindness feels like lack of knowledge, and I’m more comfortable with the feeling of knowledge than I am with the feeling of the lack of knowledge.
Wrong input > no input? I’m not so sure.
Depends on if you’re hallucinating everything or your vision has at least some bearing in the real world. I mean, I’d rather see spiders crawling on everything than be blind, since I could still see what they were crawling on.
Some things (for instance, eating) would definitely be more enjoyable while blind rather than while hallucinating spiders.
Not 100% sure they would be for me: I’m not arachnophobic at all. (I would be willing to eat a spider for five dollars, if I was sure this couldn’t cause be to get sick.)
If the wrongness is so blatant that it’s easy to tell which parts of the input are likely wrong and disregard them...
A fair point. But then you’re no better off than with not having that part of the input at all.
Well, I would, but “pretty sure most people” sounds like blatant generalizing from one example to me.