Sheldrake dissertation

There are several reasons why I'm inclined to reject what you say:

1)          Sheldrake used random numbers to determine the time when Pam would return home. This was done after Pam had left home. Thus the dog had no 'data' on which he could anticipate the return of his owner.

12 of the 107 recordings were with randomly selected returns. However, for whatever reason, it ended up that the period Pam was absent was almost the same amount for each experiment - 2 to 2-2/3 hours. Again, we have the problem that if the dog tends to start spending more time at the window after the owner has been absent for a few hours, a set of experiments in which Pam is asked to return during that period doesn't distinguish whether it's psi or ordinary behaviour.

2)          The records of the dog's behaviour were assessed blind by people who did not know the return time.

I'm not sure how that helps in this case.

3)          Sheldrake did one or two experiments in which Pam was instructed not to return home that day at all. These showed no sign of canine anticipation.

Sheldrake made 10 recordings during evenings when Pam planned, before she left, to not return or to return very late. He showed in the other trials that he tended to spend less time at the window during the evening, and seemed to do so in these cases as well (hard to know for sure since we are just given averages). He also did not spend time at the window during Pam's return in 41 of 80 other recorded trials.  

I think Sheldrake' experiments - though simple - are far more difficult to debunk than people seem to believe.

I'm not sure why you call it "debunking". My perspective is that of a researcher, so I'm just looking at what conclusions the data support. It's pretty common for those who perform the experiments to get ahead of themselves in that regard, so it is often the case that other scientists find that only lesser conclusions can be drawn, during critical examination of a research paper (I'm referring to all fields of science, not just parapsychology). If a pharmaceutical company tries to claim that their drug is fabulous, nobody finds it hard to believe when methodological issues lead physicians to be skeptical of their conclusions when the data is examined.

Linda

I agree with the author's basic premise that Sheldrake seemed to receive unusually harsh treatment. However, I think the author missed the point a bit on this issue - Wiseman does not owe Sheldrake an apology. The problem was that Wiseman chose to look for psi in a way which was valid - "was there a change in the dog's behaviour, especially a change which was distinct to her parents, when Pam was on her way home?" He did not find any change in behaviour specific to Pam's return. Sheldrake, on the other hand, chose to look at it in a way which could be ordinary behaviour as well as psi - "how much time did the dog spend at the window the 10 minute periods before and after Pam began her return home?" Sheldrake found that more time was spent at the window during the 10 minutes when Pam was returning than the preceding 10 minutes. However, since the dog tended to start spending more and more time at the window the longer Pam was gone, most 10 minute periods, regardless of whether they corresponded to Pam's return, would show the pattern Sheldrake found.

So Wiseman was correct to say that he did not find evidence of psi (because he actually measured for psi and could not find it) and to agree that he replicated Sheldrake's results that the dog spent more time at the window the longer Pam was gone, because it wasn't a measure for psi.

Linda

On the score of apology - I was referring more to the fact that, if there's a researcher conducting experiments, and that researcher gets into contact with you, and you decide to do your own experiments with assistance provided by the original researcher, and your results replicate the original researcher's despite you doing significantly fewer trials - to have your work published first, refuse to cite the original (and, at the time, on-going) research, take a decade to directly acknowledge your replication, and to frame your own experiments in your books and interviews in a way that diminishes (and sometimes leaves out altogether) the original research, is misleading to the public and grossly unfair to the original researcher. Whether Sheldrake's hypotheses are right or not is irrelevant to this issue as far as I'm concerned - Wiseman acted in appalling bad faith.

 Whether Sheldrake's hypotheses are right or not is irrelevant to this issue as far as I'm concerned...

Anyone else agree with this sentiment?

12 of the 107 recordings were with randomly selected returns. However, for whatever reason, it ended up that the period Pam was absent was almost the same amount for each experiment - 2 to 2-2/3 hours. Again, we have the problem that if the dog tends to start spending more time at the window after the owner has been absent for a few hours, a set of experiments in which Pam is asked to return during that period doesn't distinguish whether it's psi or ordinary behaviour.


I'm not sure how that helps in this case.


Sheldrake made 10 recordings during evenings when Pam planned, before she left, to not return or to return very late. He showed in the other trials that he tended to spend less time at the window during the evening, and seemed to do so in these cases as well (hard to know for sure since we are just given averages). He also did not spend time at the window during Pam's return in 41 of 80 other recorded trials.  


I'm not sure why you call it "debunking". My perspective is that of a researcher, so I'm just looking at what conclusions the data support. It's pretty common for those who perform the experiments to get ahead of themselves in that regard, so it is often the case that other scientists find that only lesser conclusions can be drawn, during critical examination of a research paper (I'm referring to all fields of science, not just parapsychology). If a pharmaceutical company tries to claim that their drug is fabulous, nobody finds it hard to believe when methodological issues lead physicians to be skeptical of their conclusions when the data is examined.

Linda

Well  you man have some relevant points, but I'd say that if anything they justify further research - which Wiseman could have done, but didn't. He also pretended that he had failed to replicate Sheldrake's experiment, when in fact he had changed it.

The point about only counting the dog's first visit to the window, is that it totally invalidates the experiment, because dogs will make odd trips to the window to investigate assorted noises etc.

The honourable thing to do here, would have been to acknowledge that he had reproduced Sheldrake's experiment, and then engage in a discussion as to how to deal with random trips to the window for other reasons.

Are you, in effect, saying that there is no way to test the assertion that dogs anticipate their owners' returns because dogs respond to other stimula?

The point about blinding the assessment of the videos was vital. As I understand it, the scorers watched segments of video and had to decide if the dog was or was not anticipating the return of his owner, without knowing if that segment of video was just before she would return, or at some earlier point.

Anyone else agree with this sentiment?

Well I suppose there are two questions, is Rupert Sheldrake treated fairly? Is his experiment valid?

It is worth remembering that no experiment is ever perfect - it is limited by all sorts of practicalities - in this case, obviously the endurance of the owner was only finite.

On the score of apology - I was referring more to the fact that, if there's a researcher conducting experiments, and that researcher gets into contact with you, and you decide to do your own experiments with assistance provided by the original researcher, and your results replicate the original researcher's despite you doing significantly fewer trials - to have your work published first, refuse to cite the original (and, at the time, on-going) research, take a decade to directly acknowledge your replication, and to frame your own experiments in your books and interviews in a way that diminishes (and sometimes leaves out altogether) the original research, is misleading to the public and grossly unfair to the original researcher. Whether Sheldrake's hypotheses are right or not is irrelevant to this issue as far as I'm concerned - Wiseman acted in appalling bad faith.

That doesn't seem to be what happened, though. And no researcher is under an obligation to perform poor research, just because the original researcher chose to do so.

Wiseman chose to do good research - he wanted to test the claim that there was something about the dog's behaviour which allowed the parents to tell when Pam was on her way home in a way which was valid. And remember, this was the claim made by Sheldrake, the parents, Pam, etc. Using a valid outcome measure, he was unable to find any behaviour (and he tried several different ways of trying to find it) which changed when Pam was one her way home. 

Contrast this with Sheldrake, who tested the claim that the dog spent more time at the window in the 10 minutes after Pam left for home than in the 10 minutes before she left. He discovered that even when Pam wasn't returning, the dog would spend more time at the window than he had in the preceding 10 minutes. So it would be no surprise (certainly not "psi") to discover that the dog also continued this behaviour right up until Pam's return. Yet he failed to account for this serious problem in his choice of outcome measure (what he used as a proxy measure for "psi"), and continues to this day to promote this invalid outcome measure as though it were valid.

Wiseman could not "cite" research which had not yet been published. But more importantly, Sheldrake did not present the data in a way which allowed for valid conclusions to be drawn about psi (as opposed to Wiseman who did), so there was no way for Wiseman to make use of Sheldrake's research (so why expect him to make much of a fuss over it?). Wiseman did suggest to Sheldrake that he could apply the valid outcome measure to Sheldrake's raw data, but Sheldrake did not do so.

Why is it "bad faith" that Wiseman focussed on the results of a good outcome measure, rather than the results of a poor outcome measure, when talking about the research? Why isn't it "appalling bad faith" that Sheldrake doesn't mention that he replicated Wiseman's negative results?

Linda

Well  you man have some relevant points, but I'd say that if anything they justify further research - which Wiseman could have done, but didn't. He also pretended that he had failed to replicate Sheldrake's experiment, when in fact he had changed it.

I don't think Wiseman ever said that he failed to replicate Sheldrake's experiment. He said that he failed to find any evidence that the dog was demonstrating psi, rather than ordinary behaviour. It was Sheldrake who claimed the dog's ordinary behaviour was a demonstration of psi. We don't know what discussions they had about outcome measures, so we don't know that Wiseman changed anything. For all we know, it was Sheldrake who changed his outcome measure, when he discovered that using Wiseman's valid outcome measure gave him unremarkable results.

The point about only counting the dog's first visit to the window, is that it totally invalidates the experiment, because dogs will make odd trips to the window to investigate assorted noises etc.

Have you read any of these papers? The times the dog went to the window to investigate something were marked and described and excluded as the signal. You can suggest anything you want as the signal (in order to get rid of this invalid criticism against Wiseman). You still won't find something which signals Pam's return (many attempts have been made to find something).

The honourable thing to do here, would have been to acknowledge that he had reproduced Sheldrake's experiment, and then engage in a discussion as to how to deal with random trips to the window for other reasons.

Both Wiseman and Sheldrake marked visits to the window for other reasons, and it doesn't play a part in either of their results. And why do you expect Wiseman to emphasize the ordinary effect which Sheldrake claims as a psi effect, when what Wiseman was looking for was a psi effect?

Are you, in effect, saying that there is no way to test the assertion that dogs anticipate their owners' returns because dogs respond to other stimula?

What's wrong with what Wiseman and Sheldrake did - exclude the visits where the dogs respond to other stimuli?

The point about blinding the assessment of the videos was vital. As I understand it, the scorers watched segments of video and had to decide if the dog was or was not anticipating the return of his owner, without knowing if that segment of video was just before she would return, or at some earlier point.

I agree that the blinding was vital. Both researchers had blind judges. I don't see how that mitigates Sheldrake's choice of an invalid outcome measure, though. Note that what you describe above is what Wiseman did. Sheldrake only had the judges mark when Jaytee went to the window, for how long, and if there was something outside the dog may be responding to. The judges did not decide if the dog was showing anticipatory behaviour, so Sheldrake did not look at whether the dog showed anticipatory behaviour. 

What's interesting in the discussions I've had over the years, is that most people expect that the way to look at this is to look for anticipatory behaviour in the videos (as you also suggest). Yet they don't seem to realize that it was Wiseman who did this and Sheldrake who did not.

Linda

That doesn't seem to be what happened, though. And no researcher is under an obligation to perform poor research, just because the original researcher chose to do so.

Wiseman chose to do good research - he wanted to test the claim that there was something about the dog's behaviour which allowed the parents to tell when Pam was on her way home in a way which was valid. And remember, this was the claim made by Sheldrake, the parents, Pam, etc. Using a valid outcome measure, he was unable to find any behaviour (and he tried several different ways of trying to find it) which changed when Pam was one her way home. 

Contrast this with Sheldrake, who tested the claim that the dog spent more time at the window in the 10 minutes after Pam left for home than in the 10 minutes before she left. He discovered that even when Pam wasn't returning, the dog would spend more time at the window than he had in the preceding 10 minutes. So it would be no surprise (certainly not "psi") to discover that the dog also continued this behaviour right up until Pam's return. Yet he failed to account for this serious problem in his choice of outcome measure (what he used as a proxy measure for "psi"), and continues to this day to promote this invalid outcome measure as though it were valid.

Wiseman could not "cite" research which had not yet been published. But more importantly, Sheldrake did not present the data in a way which allowed for valid conclusions to be drawn about psi (as opposed to Wiseman who did), so there was no way for Wiseman to make use of Sheldrake's research (so why expect him to make much of a fuss over it?). Wiseman did suggest to Sheldrake that he could apply the valid outcome measure to Sheldrake's raw data, but Sheldrake did not do so.

Why is it "bad faith" that Wiseman focussed on the results of a good outcome measure, rather than the results of a poor outcome measure, when talking about the research? Why isn't it "appalling bad faith" that Sheldrake doesn't mention that he replicated Wiseman's negative results?

Linda

That is exactly what happened, so far as I can see - Wiseman entered into experiments with Jaytee after communicating with Sheldrake, whose research was ongoing. He received assistance from Sheldrake in setting up his own trials - far fewer than Sheldrake's. Wiseman got his research - which was in response to Sheldrake's - published first, and while it's true he never claimed that he failed to replicate the pattern, it took him years to directly acknowledge that he had, and has in various books and interviews reframed the sequence of events to downplay or even eliminate Sheldrake's role in facilitating his, Wiseman's, own work.

This is bad faith. This is deceptive. And again, I'm discounting any question of the research itself, whether Sheldrake's hypothesis was right or not. Wiseman could have waited until the work he was responding to was published, or else he could've simply responded to it through comment and debate.

That is exactly what happened, so far as I can see - Wiseman entered into experiments with Jaytee after communicating with Sheldrake, whose research was ongoing. He received assistance from Sheldrake in setting up his own trials - far fewer than Sheldrake's. Wiseman got his research - which was in response to Sheldrake's - published first, and while it's true he never claimed that he failed to replicate the pattern, it took him years to directly acknowledge that he had, and has in various books and interviews reframed the sequence of events to downplay or even eliminate Sheldrake's role in facilitating his, Wiseman's, own work.

Sheldrake and Wiseman began their research around the same time. Wiseman finished his research within 6 months, while Sheldrake did not complete his research until around the time that Wiseman presented his results at the conference of the Parapsychological Association. Sheldrake never presented or published his results for an additional three years, when it appeared in abbreviated form in his book. This was well after Wiseman published his results, so there would have been no opportunity whatsoever for Wiseman to refer to Sheldrake's research or the pattern Sheldrake would use to claim success. Sheldrake finally published his research in 2000, which by this point was four years after Wiseman first presented his research, and two years after it was published.

Wiseman worked with Sheldrake and Smart to develop a methodology which would minimize the contribution of ordinary factors, and Sheldrake did not inform them that he would, instead, be looking for patterns in the data without minimizing the contribution of ordinary factors. Wiseman only learned of Sheldrake's intentions after Wiseman had already presented his research. And he did not regard Sheldrake's search for patterns compelling or valid, since ordinary behaviour would generate the pattern Sheldrake found, and the pattern search seemed to be post hoc given that it had never come up in discussion at the start of their research.

Wiseman openly describes Sheldrake's role in facilitating his work, and thanks him (as well as Smart, her family, and Jaytee) in his published paper. Wiseman was under no obligation to mention an analysis which he found to be invalid, when discussing his own work. And furthermore, how could he even be expected to discuss an invalid analysis which he wouldn't have a chance to evaluate for another four years? 

This is bad faith. This is deceptive. And again, I'm discounting any question of the research itself, whether Sheldrake's hypothesis was right or not. Wiseman could have waited until the work he was responding to was published, or else he could've simply responded to it through comment and debate.

Why on earth would Wiseman be expected to put a hold on presenting his findings until Sheldrake got around to publishing his own, especially when it took him overly long to do so? It was perfectly legitimate for Wiseman to publish his findings and discuss the implications of those results. Three years later, when Sheldrake finally got around to presenting an abbreviated version of his own results, Wiseman did respond to it through comment and debate.

None of this can be regarded as bad faith or deceptive.

Linda

Sheldrake and Wiseman began their research around the same time. Wiseman finished his research within 6 months, while Sheldrake did not complete his research until around the time that Wiseman presented his results at the conference of the Parapsychological Association. Sheldrake never presented or published his results for an additional three years, when it appeared in abbreviated form in his book. This was well after Wiseman published his results, so there would have been no opportunity whatsoever for Wiseman to refer to Sheldrake's research or the pattern Sheldrake would use to claim success. Sheldrake finally published his research in 2000, which by this point was four years after Wiseman first presented his research, and two years after it was published.

Wiseman worked with Sheldrake and Smart to develop a methodology which would minimize the contribution of ordinary factors, and Sheldrake did not inform them that he would, instead, be looking for patterns in the data without minimizing the contribution of ordinary factors. Wiseman only learned of Sheldrake's intentions after Wiseman had already presented his research. And he did not regard Sheldrake's search for patterns compelling or valid, since ordinary behaviour would generate the pattern Sheldrake found, and the pattern search seemed to be post hoc given that it had never come up in discussion at the start of their research.

Wiseman openly describes Sheldrake's role in facilitating his work, and thanks him (as well as Smart, her family, and Jaytee) in his published paper. Wiseman was under no obligation to mention an analysis which he found to be invalid, when discussing his own work. And furthermore, how could he even be expected to discuss an invalid analysis which he wouldn't have a chance to evaluate for another four years? 


Why on earth would Wiseman be expected to put a hold on presenting his findings until Sheldrake got around to publishing his own, especially when it took him overly long to do so? It was perfectly legitimate for Wiseman to publish his findings and discuss the implications of those results. Three years later, when Sheldrake finally got around to presenting an abbreviated version of his own results, Wiseman did respond to it through comment and debate.

None of this can be regarded as bad faith or deceptive.

Linda

You've got to be kidding me.

By Wiseman's own account (in his papers; one he hasn't always stuck to in the popular press) his research came about after television coverage of Sheldrake's experiments with the dog. It was in direct response to another person's research. To say that they began around the same time implies two independent projects (an implication Wiseman has made at times in the popular press) and completely misrepresents what happened.

And of course Wiseman didn't have to hold off on publishing his work until Sheldrake did his; there's no legal requirement; but I would think it's self-evident that a response study should follow the initial one, especially when the initial researcher was cooperating with the response study, done by a party he must have known wasn't sympathetic to some of his ideas.