Interview With Steven Novella of Skeptics Guide

[DysonSphere]

Originally Posted by Venom

Except that they are purrely subjective?

Open_mind basically made my point for me, everything you are aware of right now is subjective. You don't know if there is an 'out there' at all.

What separates the 'real' from the 'perceived'? Vividness is part of what makes real, real. Peak experiences and OBEs are sometimes more rich or real feeling than our day to day experiences. The second aspect of real experience is cross-reference. If we both see a wall then its obviously not a hallucination. Again, mystical experiences meet this criteria, there are many parallels between them. Although there are differences also, I think its worth noting that a color-blind person sees the world differently, dogs which rely on scent probably perceive world differently but at its core we are experiencing the same reality. The same, again, could be said of perceptions that appear to transcend bodily limits.

The point is, is that there is no reason, other than simple prejudice, to not consider the reality of mystical experiences. Unless of course you buy into consciousness being some sort of illusion. Then you couldn't consider it because you're not here.

[mszlazak]

Originally Posted by David Bailey

Mszlazak,

This shows just how differently two brains can interpret identical data! I thought Alex was definitely winning all the way (I am not sure hat is quite the right way to approach these problems, however).

Steve and his colleagues were waffling about obscure statistical effects while discussing Sheldrake's work with dogs. This allowed Alex to point out the huge difference - 4% vs 26% (if memory serves me right) and their response was to simply ask how anybody could possibly believe in psychic dogs - in a fairly scornful way! To me, that was extremely revealing. It would be interesting to know how regular listeners to that show rated their performance.

I assume that both of us would like to resolve the question as to whether ψ exists in some sort of definitive way - at least to our own satisfaction. If you take the example of anticipatory dogs, that will only be done by revealing an alternative mechanism by which this happens - dog and cat owners have known about this phenomenon for ages, the question is not if it happens, but how. Steve guessing that maybe a dog can identify the owner's car at a distance of 1 kilometre will not cut it - he would have to demonstrate that that explanation works - say using taxis.

David

The 4% versus 26% isn't the point, it's whether the one experiment that got this big effect had a large enough sample size (not according to Steve), controlled for other sources of normal information transfer to the dog like sounds or smells picked up when the owner was close enough home (it didn't according to Steve) and whether this study has been independently repeated with similar size results (it hasn't). Alex didn't give an adequate response to any of this.

So, given the nature of the evidence so far and the notion that the dog was psychic, it is not only reasonable but should be expected that the response to this stuff is that it's likely due to experimental design problems or fraud on the part of the investigator.

Now, if these and other problems were resolve, would that mean psi? No, since a specific mechanism would need to be found to account for the results, otherwise you have AGAIN the problems associate with naive eliminative induction.

Alex was finally left with the implicit admission that the evidence isn't worthy and he invites skeptics to come along with him on his NEW experiment.

----------------------

[David Bailey]

Originally Posted by mszlazak

Alex was finally left with the implicit admission that the evidence isn't worthy and he invites skeptics to come along with him on his NEW experiment.

----------------------

Well, maybe Alex should answer that point - I certainly didn't hear one, and I did not think one was justified!

Can't we get away from such a point-scoring approach to all this - I mean, I think we both want to know the truth about all this - whatever it turns out to be.

Since I am pretty convinced that the dog experiment taps into something - i.e. it is not a statistical fluke - it would be good if researchers could nail what that is - obviously with ψ as the last option. Even Steve admitted that his dog does it.

You might argue - probably correctly - that a decent explanation would not end all discussion of ψ (!!), but it would be a far more persuasive blow than all these arguments about statistics regarding a phenomenon that is actually rather easy to observe.

David

[alextsakiris]

Originally Posted by David Bailey

Mszlazak,

This shows just how differently two brains can interpret identical data! I thought Alex was definitely winning all the way (I am not sure hat is quite the right way to approach these problems, however).

Steve and his colleagues were waffling about obscure statistical effects while discussing Sheldrake's work with dogs. This allowed Alex to point out the huge difference - 4% vs 26% (if memory serves me right) and their response was to simply ask how anybody could possibly believe in psychic dogs - in a fairly scornful way! To me, that was extremely revealing. It would be interesting to know how regular listeners to that show rated their performance.

I assume that both of us would like to resolve the question as to whether ψ exists in some sort of definitive way - at least to our own satisfaction. If you take the example of anticipatory dogs, that will only be done by revealing an alternative mechanism by which this happens - dog and cat owners have known about this phenomenon for ages, the question is not if it happens, but how. Steve guessing that maybe a dog can identify the owner's car at a distance of 1 kilometre will not cut it - he would have to demonstrate that that explanation works - say using taxis.

David

Sheldrake used taxis in his trials... the results were 4% versus 55% in one experiment and 1% versus 26% in the other. Skeptics would do well to READ and not just listen to Steve.

[David Bailey]

Originally Posted by alextsakiris

Sheldrake used taxis in his trials... the results were 4% versus 55% in one experiment and 1% versus 26% in the other. Skeptics would do well to READ and not just listen to Steve.

Thanks for that - I think the problem is that Steve uses the trick of discussing a wide variety of possible issues without first checking if these were excluded in the original experiment. This makes it very difficult for the interviewee to respond definitively - unless perhaps it was Sheldrake himself. I noticed in particular that he was still making comments about small statistical effects when he came to Rupert's work - where the effect sizes are huge.

I think the trick (conscious or unconscious) is that Steve gives the impression that he is making comments specifically related to a particular experiment, when in reality he is just reciting potential problems in ψ experiments in general.

David

[mszlazak]

Originally Posted by alextsakiris

Sheldrake used taxis in his trials... the results were 4% versus 55% in one experiment and 1% versus 26% in the other. Skeptics would do well to READ and not just listen to Steve.

This is a funny response coming from you! Why then would anyone want to listen to your interviews with Steve. You didn't have anything to say when he countered with his response to the other study! But now you bring up this study and blame him for your failure in not mentioning it! So, Steve was suppose to respond to something you will post or bring up in the future. I guess that would be par for the course around here.

Anyway, ETinKC in this thread The Skeptics' Guide to the Universe Forum :: View topic - Episode #125 had this to say about the study with the 4% to 55% results:

Joining the discussion late here.

Lots of talk about how to control for possible answers to the percentages in the dog study (4% vs. 55%) but I am not convinced that there is even an effect here based on a quick reading of the study and curious how hard people have looked at the data. Now I am not a stats person, but it seems to me that there may be a fatal flaw in the data.

First off, the statistical sample is way to small to mean anything here.

Second, the whole pre-beep time, where the dog is going to the window before the owners (or anyone if we are to believe the experiment) knows that they are coming home is almost to ridiculous to comment on, but interesting that they included it to make the discrepancy look bigger, because if it was part of the away time the percentages are nowhere near as impressive. However it is also telling in what I think is really going on here.

Now a true 4% to 55% comparison would be if the the dog never went to the window at all in the waiting time (or, in 100 trials - walked to the window only 4 times) and 55% of the time during the coming home time ( I guess the other 45% they were just having too much fun licking themselves to get up - also, once they go to the window during the home time - why not stay there until we are home. Lazy mutt.)

However, looking at the graphs what I see is a dog that is spending more time at the window the longer an owner is gone, nothing more.

An easy way to test this would be to have the coming home time vary randomly from 10 minutes after leaving to 10 hours over many many trials, and then group away verses coming home times for each total time away. If this was a real effect, the time at the window during the coming home time versus the time at the window during the away time would be consistent no matter the total time away, and my guess is that it is not. (I would hazard to guess that it isn't even with this study, but again the sample time is too small to tell you anything more then the dog had to pee.)

Eric

I don't know if what he says is valid since I haven't read this study.

---------

[Open Mind]

Originally Posted by mszlazak

I don't know if what he says is valid since I haven't read this study.

Sheldrake has countered those type of criticisms......

Dialogues and Controversies - Controversies - Richard Wiseman - Evidence on Dogs
[i] '...First, they suggest that Jaytee may simply have gone to the window more and more the longer Pam was out, and hence been there most when she was on the way home. But a comparison of Jaytee's behaviour during Pam's short, medium and long absences shows that this was not the case (Sheldrake, 1999b). Moreover, in control experiments in which Pam did not come home, Jaytee did not go to the window more and more as time went on (Sheldrake, 1999b, Figure B.2).

Second, they say that my experiments "appear to contain design problems (Blackmore, 1999)". Susan Blackmore's comments were made in an article in the Times Higher Education Supplement , which concluded: "There are better ways to spend precious research time than chasing after something that lots of people want to be true, but almost certainly is not." She thought she had spotted "design problems" in my experiments with Jaytee (Sheldrake 1999b) because "Pam was never away for less than an hour". (In Wiseman Smith & Milton's experiments Pam was likewise never away for less than an hour.)

This is why Blackmore thought there was a problem: "Sheldrake did 12 experiments in which he bleeped Pam at random times to tell her to return..... When Pam first leaves, Jaytee settles down and does not bother to go to the window. The longer she is away, the more often he goes to look.

[Y]et the comparison is made with the early period when the dog rarely gets up." But anybody who looks at the actual data (Sheldrake, 1999b, Figure B.4) can see for themselves that this is not true. In 5 out of the 12 experiments, Jaytee did not settle down immediately she left. In fact he went to the window more in the first hour than during the rest of Pam's absence, right up until she was on the way home, or just about to leave.

In the light of Blackmore's comments, I have reanalyzed the data from all 12 experiments excluding the first hour. The percentage of time that Jaytee spent by the window in the main period of Pam's absence was actually lower when the first hour was excluded (3.1%) than when it was included (3.7%). By contrast, Jaytee was at the window 55.2% of the time when she was on the way home. Taking Blackmore's objection into account strengthens rather than weakens the evidence for Jaytee knowing when his owner was coming home, and increases the statistical significance of the comparison. (Including the first 60 minutes of Pam's absence in the analysis, by the paired-sample t test, t=-5.72, p=0.0001; excluding the first 60 minutes, t=-5.99, p<0.0001.)

Blackmore's claim illustrates once again the need to treat what sceptics say with scepticism

[Open Mind]

Will SGU & NESS play fair with the suggested protocol?

If a medium was successful this would be very embarrassing to SGU, NESS and JREF. As a result, a protocol best to test for mediumship might not be their priority, will they instinctively try to change the protocol to maximize the odds of medium failure or raise bar to present any mild success as failure? I hope not.

During CSICOP's test of Natasha Demkina, this quote from Andrew Skolnick (CSMMH) revealed the attitude of many 'skeptics' during investigations. '.. I had a big fight with them [Hyman, Wiseman] at breakfast before the test. During the test, I just had to bite my tongue and "pray" that we weren't going to pay dearly for the numerous compromises that were permitted. We came awfully close to paying for those lapses'

So much for trying to create a 'psi conducive environment' .

[Paul C. Anagnostopoulos]

Originally Posted by Open Mind

If a medium was successful this would be very embarrassing to SGU, NESS and JREF. As a result, a protocol best to test for mediumship might not be their priority, will they instinctively try to change the protocol to maximize the odds of medium failure or raise bar to present any mild success as failure? I hope not.

How do you decide if they are changing the protocol to maximize the odds of failure or minimize the odds of leakage and cheating?

~~ Paul

[Phronk]

Originally Posted by Paul C. Anagnostopoulos

How do you decide if they are changing the protocol to maximize the odds of failure or minimize the odds of leakage and cheating?

~~ Paul

If you want to be like many psuedo-skeptics, you simply accuse them of it after the study has been conducted. No need for evidence; the accusation will do (and be cited by other pseudo-skeptics as evidence itself). Bonus points if the accusation is actually testable, but you haven't bothered testing it, or didn't like the results of the test so you ignore them.

But in all seriousness...minimizing cheating and maximizing odds of finding a true effect should be exactly the same goal. A study where cheating is easy is useless to parapsychologists and a study where a signal couldn't get through even if it existed is useless to skeptics. If people just make it their goal to find the truth rather than getting all hot-headed about supporting what they already believe, a well-designed study shouldn't be hard to come up with.