Comment

Comment on Can a reasonable person genuinely believe in ghosts?

Thanks for your response. If you and I agree on anything it’s that we should do more science to understand this stuff better.

The scientific method involves looking at both the cases where it seems like something happened and the cases where nothing happened (e.g. someone said they had an experience but it clearly didn’t match reality). If you cherry pick just the events that “showed” what you want, that’s confirmation bias.

Confirmation bias is real, but this isn’t it. If I believe that all swans are white , and then I come across a black swan, should I just dismiss that data point because it would confirmation bias (perhaps people would accuse me of wanting this outcome)? No. Ignoring the black swan isn’t the way to go here. It wouldn’t be ridding ourselves of confirmation bias, it would be ridding ourselves of critical data that contradicts our starting hypothesis.

Similarly: even if supernatural stuff that is hard to explain happens in only a percentage of cases, discarding that data isn’t ridding ourselves of confirmation bias; it’s simply choosing to ignore critical data. That’s not good science.

I instead urge you to go read scientific papers on the topic, and specifically not just the ones that seem to suggest the outcome you want to hear.

This is what I started with, so for the longest time I was very skeptical, just like most people in this thread. It is my belief that anyone with an open mind who takes in all the information on this topic (including the studies that suggest supernatural outcomes and those that don’t; the first-hand accounts and the skeptical rebuttals) will inevitably come to the same conclusion that I have. That was my experience, anyway. This is not a conclusion I was looking for; I was really stubbornly against this stuff for the longest time, but I was forced to change my mind.

It’s also worth noting that the book talks about more than just near-death experiences; I just used them as an example.

source

Sort:hotnew top

clean_anion@programming.dev ⁨1⁩ ⁨hour⁩ ago
This can be verified by asking people who have had near-death experiences whether or not they experienced something correct in their near-death experiences. Obviously, such experiences are traumatic, and multiple studies show that people can hallucinate due to the release of various neurotransmitters associated with the same.

We want to calculate the probability that someone manifested as a ghost given that they had an interesting near-death experience. We assume that anyone having a true supernatural experience experiences visions that are absolutely true. For each person, there are two possibilities (we’ll calculate the probability of each later).

The first possibility is that a person, in fact, experienced hallucinations. The second possibility is that a person experienced a ghostly manifestation.

Now, we further give people an objective multiple-choice quiz about the positions of various objects in an environment. To generate this quiz, we ask each person to choose the environment they believe themselves to have manifested in. We verify that they have never been to this environment before and did not have any method of knowing about this environment (e.g., if a subject saw a person going into a room and later gave an exact description of the person in the given room, it will be disregarded). We only test people who believe that they experienced a supernatural event. All options are framed in an equivalent manner and are presented in a randomized order to remove cognitive biases and implement double-blind protocols. We further use questions with non-obvious answers such that they differ from previous implementations (e.g., a vision of a surgery table with an overhead light is obvious, and by itself, not indicative of supernatural phenomena).

If the subject hallucinated, we assume that they have a random chance of predicting the positions of various objects. We now repeat this quiz a large number of times in accordance with the law of large numbers. If, after many repetitions, we find a sufficient deviation from the expected result (e.g., if each question had one correct answer and three incorrect answers, with the observed rate of correct answers being 50% instead of 25%), then we would have evidence supporting the existence of ghosts.

If, however, the results show no sufficient deviation from the expected results, then we would find that the probability of a perceived encounter being supernatural is approximately zero.

In this way, we can use scientific methods to test claims of ghost-like phenomena.

NOTE: If we only focus on the 25% of the cases as mentioned in the above example, we find that we are not focusing on the remaining 75% of the cases. Presenting only 25% of the cases, without giving any thought to the remaining 75% of the cases is an incorrect method of analysis as explained above.

source
WolfLink@sh.itjust.works ⁨15⁩ ⁨hours⁩ ago
Confirmation bias is when the outcome could be adequately explained by luck.

In the topic of near death experiences, if there are 1,000,000 near death experiences and 100 involve someone “knowing something they shouldn’t be able to”, those 100 cases are more likely to be remembered or recorded as significant than the other 900,000 cases. This can lead to an apparent statistical significance in correctly knowing “unknowable” information, when really it’s just people “guessing” correctly.

The “black swan” scenario is a bit different but it would be something like if you are more likely to record a swan sighting if the swan is black, you will significantly overestimate the frequency of black swans.

Im not saying the cases of apparent supernatural effects should be ignored, I’m saying they need to be taken in the context of all similar events, including the mundane, to understand if there even is an effect (knowing something that shouldn’t be possible) or if it’s just a handful of lucky guesses.

source
- ageedizzle@piefed.ca ⁨11⁩ ⁨hours⁩ ago
  
  In the topic of near death experiences, if there are 1,000,000 near death experiences and 100 involve someone “knowing something they shouldn’t be able to”, those 100 cases are more likely to be remembered or recorded as significant than the other 900,000 cases
  
  These are nowhere near the real numbers. No one could realistically conduct a study on near death experiences that included 1,000,000 participants
  
  source
  - WolfLink@sh.itjust.works ⁨10⁩ ⁨hours⁩ ago
    That’s a rough estimate of mine on how many near death experiences might have happened, studied or not, and that’s why it’s such a problem to only focus on the anecdotal cases that get recorded because they are interesting.
    
    A proper study doesn’t need to include 1,000,000 cases, but it does need to ensure that it doesn’t have bias in the cases it does include.
    
    source
    ageedizzle@piefed.ca ⁨5⁩ ⁨hours⁩ ago
    Okay. Thanks for your comment. This discussion we’re having here is one of the few threads that hasn’t devolved into name-calling so I appreciate that.
    
    I have two responses to this.
    
    The first is that I’m still not so sure I agree with the framing here regarding cherry-picking or bias. Your concern seems to be (and correct me if I’m wrong) something like this: in most cases nothing out of the ordinary happens, so if we only focus on the few cases where something paranormal seems to have happened then we disregard the vast majority of the data and are only focusing on anecdotes. It’s more scientific to focus on the bulk of the data, where nothing interesting happens. (Again, please correct me if this is a misrepresentation.)
    
    I don’t agree with this characterization because important data is often few and far between. But we shouldn’t discount it simply because it is rare. For example, consider Hawking radiation. From what I hear it’s an important concept in theoretical physics. But Hawking radiation is very hard to observe. In fact, it’s only been observed once, and the observation wasn’t even in the wild; it was in a lab. This was an important observation; it provided experimental support for an important concept. Say I was a physicist and I was sceptical of Hawking radiation. What should I do with this information? Should I say “well, this data doesn’t matter, most of the time we can’t observe Hawking radiation anyway so this data is just anecdotal”? No, that would be an improper response. Sure, this data is rare, but that doesn’t mean I can just label it as anecdotal and reject it on that basis. Because the data, though rare, still is very hard to explain without the concept of Hawking radiation. Similarly: it is possible that interesting data regarding near-death experiences are rare. Does that mean that this data is anecdotal and should be ignored? No. So long as we have cases that are genuinely hard to explain without supernatural explanations (and, I believe, we do) then that data will be very important. Because we still have to explain what was going on in those cases.
    
    Another example of this is in the Earth sciences, where the large portion of the field is literally trying to create theories to explain one-time events. For example, the extinction of the dinosaurs. Should we reject the theory that they were killed by an asteroid or meteor or whatever simply because that event only happened once, and the event is therefore merely anecdotal? No. Even events that only occur once may require us to construct novel theories. So long as we cannot explain the event with current theoretical frameworks then it is our duty to invoke a new framework. As it is with dinosaurs so too with NDEs: even if there was just one, spectacular event that was difficult to explain with current frameworks, then it is our duty to invoke novel theoretical frameworks (so long as we actually want to know what’s going on). If the data leads us to theories that are paranormal in their character then, oh well, that’s just where we’ll have to go. If we want to follow the data to where it leads, then we cannot rule out certain destinations ahead of time.
    
    It’s also worth pointing out that focusing on single cases is common practice in psychology and medicine. Sure, it’s not a replacement for theoretical understanding or large-scale studies, but it is still informative (for the reasons mentioned above). When researchers document and discuss a single interesting case it is known as a case study.
    
    The second thing I wanted to say was regarding your estimates of total NDEs versus potentially paranormal NDEs. You seemed to be trying to aggregate over all the NDEs that have every happened and tried to find the ratio between the NDEs that are interesting versus the one that are amenable to mundane explanations. But I don’t know if this is super helpful. Because, for one thing, we’re largely left guessing at the numbers (how do we how many were interesting? how do we know how many were mundane? there’s literally no way to know). Even if we only look at all the data that we do have then we have to do that in a controlled manner, otherwise we’ll run into issues. If we only run thing haphazardly, back of the envelope style then we don’t know our scope (how many cases are we dealing with?) and we cannot control for any confounding variables (is this data interoperable?) or trace the data chain-of-custody (how did we even get this data to begin with, and how did that colour its presentation?). In short, it’s too messy.
    
    So what we need, instead, is something more controlled. Ideally for something like this we’d want to look at a meta-analysis. But unfortunately I don’t know if anything with the required scope exists (if you can find one though, let me know). So the second-best thing to look at is an individual study. You mentioned earlier that you were looking at some studies. If you found any that you thought were interesting then it would probably be more productive to poke holes in that study specifically. I would be happy to discuss the merits of any study of your choosing and then take things from there.
    
    But if we do that then I think the ratios you were discussing in your message would dissolve. This is because its practically impossible to conduct an NDE study with a large sample size (it’s hard to predict if/when/how/where someone will die, and the vast majority of those that do die don’t come back to talk about it). And with small studies even a single hard-to-explain NDE would be a relatively large percentage of the total sample (which should, I think, mitigate the concerns you expressed in your earlier message; but correct me if I’m wrong on that).
    
    source
    -> View More Comments