[Chris Noble]

Originally Posted by David Bailey

This example is interesting for a number of reasons:

a) We both assumed that out pet sat waiting for us for hours because he was there waiting for us when we turned up.

b) I was totally skeptical when our neighbour told us that he sat in her house and asked to be let out shortly before we turned up. The extra time involved in all that makes it much less plausible that he could hear our car (a very common make) at least a couple of minutes away.

I suspect that most cases of "psychic" pets are explained by a).

If your cat only began to signal that it wanted to go out when you were a few minutes away I would think that the most likely explanation is that it heard the car. You really need to give a good reason why this explanation is less likely than "psi".

I don't think that this explanation played a role in the specific case of the Jaytee trials where the "signalling" began before the owner even got to her car. I do suspect that it explains a great number of cases for the general phenomena. I don't think you can expect a single explanation to cover all cases. In the Kane trials the dog only began "signalling" after the owener began the journey home. The owner always travelled in the same car. There was only one trial where the owner was a significant distance from home and the "signalling" in this case only started right at the end of the journey home.

I don't think Alex would have shown us this video if it were not reasonably typical - from what he says, he has been busy trawling for dogs that can do this to order. I know you think Alex has 'gone native' but I tend to think he is like me - he would really like to know the answer!

It would be good if he stated whether it was typical.

I think we basically agree that if the signal is strong enough, it is more important to concentrate on eliminating extraneous ways that the dog could know - or other possible artefacts.

On reflection I think the most reasonable strategy would be to do a number of "loose" trials without randomisation, without using a different car etc and then a series of "tight" trials with all of these extra controls. If the performance drops off then this suggests that at least part of the effect has perfectly natural explanations. It would be important to do enough trials to get some statistical significance and that all possible controls (within reason) are used in the "tight" trials.

I would say that a tentative 'theory of ?' (or at least much of it) is that some information can leak from individual to individual (human or otherwise) by non-physical means.

This is exactly the problem. You haven't described a scientific theory even a tentative one. All you have done is describe what "psi" isn't - physical.

It is true that we can't turn ? on and off - which makes experiments more difficult - but that is true in many other areas of science - vulcanology, astrophysics, etc. but that does not prevent progress.

It is true that you can't directly control distant stars but you can look at different regions of space where one factor differs and see what difference that makes. You can't do the same for "psi".

Clearly one thing that limits these experiments is the patience of the owners, and the time involved. Sheldrake may not have had the luxury to vary every possible factor exhaustively. This is why I think it would help if a little more money were available for this sort of research.

As I stated before a more reasonable strategy would be to compare a statistically significant number of "loose" trials and "tight" trials. This should be more reasonable for everybody involved.

Suppose that Alex arranged for a few trials in which someone else drove Jane's car home. If that produced a negative result, would you feel that the issue of car sound - which I agree is important - had been adequately dealt with? I think this would be a better test than Jane returning in a different car, because a clear response to the 'wrong' person driving home would be very informative.

That is a good idea. It would be important to look at whether the average time that the dog spends "signalling" is greater when the car is returning than during the non-return period. If there is no difference or if it only comes in less than the last 3 minutes then I would agree that this factor will have been adequately dealt with in this specific case. Again it would be important to do a statistically significant number of trials. It would also not rule out this explanation for the n% of dog owners who think their pets are psychic.

Since I guess different people drive the same car in different ways, another option would be to have a few trials in which Jane drove to within 500 yards, and then drove off. If the dog went into anticipation mode, this might be a bit ambiguous, but if he did not respond until later when she really came back, it would be pretty decisive.

I think this one would be to open to interpretations either way.

I would rather Alex's experiment failed to find a ? effect than that he 'found' a false one. In that, I think you, I and Alex are in complete agreement.

His statement on one of his podcasts that "dogs that know are real and we're gonna prove it" suggests that he ahs already made his mind up. In the youtube video he explicitly states that the dog is telepathically communicating with it's owner.

As I say, I strongly suspect there is a genuine effect
here - so prepare yourself for a possible change of world view!

I strongly suggest that this will all fizzle out. One advantage of Alex going for the "Pons and Fleischman" premature announcement is that this won't end up hidden in the filedrawer.