Haunted Place

[davidsmith73]

Errm, as I said. Of course, first parapsychology should really demonstrate that these anomolous methods of information transfer actually exist.

see the paper I linked

That's obvious.

And as Psiload points out the very paper you link to starts by categorically saying:

"the list is not offered as providing compelling evidence or “proof” of psi,"

So how you can quote this paper as anything that clearly shows evidence of Psi is beyond me.

Lets take a quote from the conclusion:

"It appears quite clear from the above review that irrespective what interpretation is given to specific research reports, the overall results of parapsychological experimentation are indicative of an anomalous process of information transfer, and they are not marginal and neither are they impossible to replicate. In the face of this, the critic who merely goes on asserting there is no evidence for psi is using a tactic reminiscent of Mohammed Saeed al-Sahaf, Iraq’s former information Minister, in blindly asserting there are no American troops in Baghdad."

Would you say that all the experiments listed in the paper can be accounted for by either methodological error, fraud or chance?

Pot? Kettle on line one for you.
The only games are being played by you. They are different fields and the difference is what they are studying (as should be self evident).

What parapsychology is studying is observable and is measured using behavioural protocols and physical correlates, just like conventional psychology. It is similar to other fields of science in this respect, contrary to your assertion Ashles. You keep saying they are different fields. I'm not arguing with that. I'm arguing against your assertion that parapsychology is unlike any other scientific discipline.

What parapsychology studies as its field (i.e. paranormal abilities) have not been shown to exist.

It appears that they do according to the paper I linked. You asked for evidence. You have access to it, but seem to ignore it. How far does the phrase "extraordinary claims need extraordinary evidence" have to do go to convince someone like you?

The fact that they use some techniques used by the field of psychology means nothing in terms of validating these claimed phenomena.
I can search for ghostly cold spots using a thermometer, but it doesn't mean that ghosts exists just because I use accepted physics to search for them.

Strawman argument again. Controlled scientific studies of psi do not attempt to measure cold spots in haunted houses. This would of course not be acceptable proof of ghosts because there is no methodology as to what anomaly is being measured. Parapsychology uses techniques used in conventional psychology and it also uses them in the context of properly controlled experiments, not silly made up examples like yours.

 

[davidsmith73]

Typical.

Why does every parapsychological paper read like a college kid's letter to Mom & Dad?

P.S.- Send money.

Right, so no other acedemic field is in the habit of trying to convince someone they are worth funding

 

[dharlow]

If, by anomolous cognition you are referring to unusual mental processing of information or stimulus (i.e. matters occurring entirely within the subject's mind) then this is the field of psychology. If parapsychologists are doing this then they are just duplicating the work that belongs in another field.

I'm not sure about that. Chris French, who is hardly a believer in the paranormal, has nevertheless been one of a few psychologists to branch away into the subfield of Anomalistic Psychology. In fact, French helped create his own department at Goldsmith University.

From his website:

"While psychology, neurology, and other scientific disciplines are rich with explanatory models for human experiences of many kinds, these models are rarely extrapolated to attempt to explain strange and unusual experiences."

A few psychology departments offer a class in anomalistic psychology, but only a very few. Much of what French and his unit does is similar to parapsychological research, although less emphasis is placed on eliciting "psi" and more on examining correlates of belief, etc... French is an active participant in parapsychology debates, and is an associate member of the Parapsychology Association. It is not terribly easy to differentiate what he does from what most parapsychologists do, although he is more skeptical toward psi than an average parapsychologist would be.

Hence, while one could perhaps do some work on anomalistics in a few psychology departments, it is more likely that you would go into Anomalistic Psychology or Parapsychology for such a focus, both of which are essentially studying the same things.

 

[Psiload]

Right, so no other acedemic field is in the habit of trying to convince someone they are worth funding

None other with so little justification.

Parapsychology is more like the double secret academic probation college student who keeps writing home asking for money.

In terms of a cost:benefit ratio, parapsychology is in a class by itself.

 

[davidsmith73]

In terms of a cost:benefit ratio, parapsychology is in a class by itself.

What is this ratio for parapsychology? How much does a typical psi experiment cost?

 

[Ashles]

Hence, while one could perhaps do some work on anomalistics in a few psychology departments, it is more likely that you would go into Anomalistic Psychology or Parapsychology for such a focus, both of which are essentially studying the same things.

This would still be a subset of Psychology.

When I was studying Experimental Psychology one of the courses within it was Experiential Psychology which looked at NDEs, OBEs, hallucinatory experiences, Shamans, belief systems etc.

This sounds similar to what Chris French is investigating. I don't see why this would not still clearly be part of the field of Psychology. It was where I studied it.

Would you say that all the experiments listed in the paper can be accounted for by either methodological error, fraud or chance?

Yes I would say they could be. Of course I can't say they are, but it is entirely possible, even probable.
The criticisms are themselves all contained as references within your paper so I am hardly alone in finding issue with the experiments.

Parapsychology uses techniques used in conventional psychology and it also uses them in the context of properly controlled experiments, not silly made up examples like yours.

The example was to illustrate a point. Obviously this passed you by.

How far does the phrase "extraordinary claims need extraordinary evidence" have to do go to convince someone like you?

I love the way language changes over the course of these debates. It reaches this point of "So I have completely and totally countered your argument definitively with excellent examples so you must have some kind of problem"
When of course you have done no such thing. You have mentioned the usual experiments and even (and I give you credit for this) provide them in a handy list together with some of the main criticisms of the work.
Your link, in essence, displays the problem. It is not cut and dry, no matter how much you would protest that it is, and your own link demonstrates this. There are experiments and there are criticisms of these experiments.

I think this quote from the paper sums it up:

Perhaps the most reliable and replicable finding in parapsychology is that of the experimenter effect (Parker, 1978).

Believers find results, sceptics don't.
Why?
Sceptics believe it is because believers pre-convince themselves of results, use poor protocols, "file drawer" effect etc.
Believers believe it is because of yet another psychic effect.

Who is right? At the moment it appears to depend on your point of view.

 

[Psiload]

What is this ratio for parapsychology? How much does a typical psi experiment cost?

- 1:0

- Too much.

Next question.

 

[dharlow]

This would still be a subset of Psychology.

When I was studying Experimental Psychology one of the courses within it was Experiential Psychology which looked at NDEs, OBEs, hallucinatory experiences, Shamans, belief systems etc.

This sounds similar to what Chris French is investigating. I don't see why this would not still clearly be part of the field of Psychology. It was where I studied it.

It may technically fall under the blanket definition of what psychology studies, but if no one in psychology is actively researching it, it really isn't a part of the field. You note that "one of the courses" in your discipline covered some of the subject material of parapsychology. One class in large and diverse field like psychology won't make for much growth for research. I took a meteorology class in undergraduate in Geology, the only one they offered. It gave a simple background to the field, but didn't give me any outlet for research or any real understanding of what the field involved. For that, I had to go to graduate school in Meteorology with other people engaged solely in that line of work. The same is true for anomalous experiences.

French (along to some degree with Wiseman and Blackmore among others) seems to be trying to bridge the gap between mainstream psychology and parapsychology...he may be successful, although I doubt it in the long run. Parapsychology only came into being in the first place because mainstream science did not attempt to tackle "paranormal" experiences in a scientific manner. Because of this lack of interest, parapsychology came into being, and while that gap has been filled to some extent by some serious researchers who follow the scientific method, it is also taken advantage of by those wishing to promote their own pre-conceived pro-psi views.

 

[Ashles]

One class in large and diverse field like psychology won't make for much growth for research.

Well that's certainly true, but when I was researching NDEs for a dissertation I discovered there was a huge amount of research on it (and other similar areas) in Psychology reference books.
But it obviously didn't deal very much with the actual Psi 'abilities', but the beliefs and belief perpetuation systems connected with them.

 

[davidsmith73]

Would you say that all the experiments listed in the paper can be accounted for by either methodological error, fraud or chance?
-----------------------------------------------------------------------
Yes I would say they could be. Of course I can't say they are, but it is entirely possible, even probable.

Is that what you want to believe? It's true that any experiments in science are prone to fraud. Method errors and chance? Now that's something you'll have to demonstrate.

The example was to illustrate a point. Obviously this passed you by.

The example was spectacularly misplaced.

I love the way language changes over the course of these debates. It reaches this point of "So I have completely and totally countered your argument definitively with excellent examples so you must have some kind of problem"
When of course you have done no such thing. You have mentioned the usual experiments and even (and I give you credit for this) provide them in a handy list together with some of the main criticisms of the work.
Your link, in essence, displays the problem. It is not cut and dry, no matter how much you would protest that it is, and your own link demonstrates this. There are experiments and there are criticisms of these experiments.

There are far fewer criticisms than tightly controlled experiments that offer no grounds for such criticism. If you take this into account, then yes there is evidence for psi.

I think this quote from the paper sums it up:

Believers find results, sceptics don't.
Why?
Sceptics believe it is because believers pre-convince themselves of results, use poor protocols, "file drawer" effect etc.
Believers believe it is because of yet another psychic effect.

Who is right? At the moment it appears to depend on your point of view.

No not really. As the paper also says:

"Perhaps the most reliable and replicable finding in parapsychology is that of the experimenter effect (Parker, 1978). It should be emphasized for the reader new to the subject that the effect cannot explain the psi-effect if all the controls are in place."

This is of course just common sense. If the disbeliever doesn't get results and the believer does but they perform identical experiments then it is unlikely due to poor experimentation on either part.

 

[Ashles]

No not really. As the paper also says:

"Perhaps the most reliable and replicable finding in parapsychology is that of the experimenter effect (Parker, 1978). It should be emphasized for the reader new to the subject that the effect cannot explain the psi-effect if all the controls are in place."

This is of course just common sense. If the disbeliever doesn't get results and the believer does but they perform identical experiments then it is unlikely due to poor experimentation on either part.

So exactly as I said then?
"the effect cannot explain the psi-effect if all the controls are in place" and the point is that the controls might not all be in place.

I am getting tired of refuting your claims. It is obvious that you wish to believe in the existence of Psi.
The experiments are not conclusive. This is beyond question. Why you keep claiming that Psi has been definitively exhibited when it blatantly hasn't is rather curious.
There are effects that may be genuine or may be experimenter bias, poor protocol etc. but to declare that psi has been demonstrated is just flat out incorrect.

Is that what you want to believe? It's true that any experiments in science are prone to fraud. Method errors and chance? Now that's something you'll have to demonstrate.

I don't have to personally demonstrate anything of the sort. I'll go with the scientific community as a whole who currently see no useful evidence of Psi. It is not an accepted scientific phenomenon.
It's actually up to you (or the parapsychologists) to conclusively demonstrate it's existence.

Of course until then you are entitled to believe it exists but please stop trying to convince us or yourself that it has been somehow 'proved'.

What people believe

A survey of 1,574 adults found that 60 percent agreed or strongly agreed that some people possess psychic powers or extrasensory perception, a premise that is generally discarded as unproved by most scientists.

A sad state of affairs.

 

[davidsmith73]

So exactly as I said then?
"the effect cannot explain the psi-effect if all the controls are in place" and the point is that the controls might not all be in place.

If I read a paper that describes adequate controls then I trust that they have been adhered to. Otherwise I would have to reject all scientific papers. You are selecting which papers to trust and which to reject, based on what?

 

[Ashles]

If I read a paper that describes adequate controls then I trust that they have been adhered to. Otherwise I would have to reject all scientific papers. You are selecting which papers to trust and which to reject, based on what?

Based on the fact that other researchers don't get the same effect result when following the same experimental protocol.

This means either

a) An additional psychic effect exists whereby the attitude of the Experimenter psychically effects the subjects

or

b) The original experiments were not carried out exactly as stringently as described

Until Psi is demonstrated to exist then I am going to assume b) is the case. You, of course, are welcome to assume a).

 

[davidsmith73]

Based on the fact that other researchers don't get the same effect result when following the same experimental protocol.

This means either

a) An additional psychic effect exists whereby the attitude of the Experimenter psychically effects the subjects

or

b) The original experiments were not carried out exactly as stringently as described

Until Psi is demonstrated to exist then I am going to assume b) is the case. You, of course, are welcome to assume a).

I assume a) because there is enough evidence for psi to justify this assumption.

However, there is the option for an experiment to be developed that bypasses the experimenter effect. See here:

http://www.dbem.ws/Precognitive Habituation.pdf

 

[Ashles]

Interesting - that would be the most useful type of experiment to show a Psi effect.

I have two quetions about that study:

1) Has it been replicated by sceptics? (I can't find details of Kenneth Savitsky's replication)

2) Once again it seems to suffer from overcomplexity of the protocol. For an experiment designed to be extremely simple, why is the data then subdivided?

The PH hypothesis is that the repeated exposures of the target can reach back in time to diminish the arousal it would otherwise produce, thereby rendering negatively arousing targets less negative and positively arousing targets less positive. Because the two pictures in each pair are matched for valence and arousal, participants are predicted to prefer the target-to-be on trials with negatively arousing pictures but the non-target on trials with positively arousing pictures (erotic pictures). Preferences on trials with non-arousing (“low-affect”) pictures were not expected to differ from chance.
To date, more than 400 men and women have participated in 9 variations of the PH experiment, including an independent replication by a skeptical investigator. Collectively the studies provide strong support for the two predicted effects. Across the six basic studies, the hit rate was significantly above 50% on negative trials (52.6%, t(259) = 3.17, p = .0008) and significantly
below 50% on erotic trials (48.0%, t(149) = -1.88, p = .031).

Is the habituation effect not simple to display? Why must the erotic and negative categories be introduced?
Why not just display the effect as simply as possible and then you can start showing subtle subdivisions.

Starting to make predictions about the affect of precognitive habituation on different types of stimuli seems to be putting the cart before the horse a bit.

These could all be ways of making the data look better after the fact (I'm not saying they are but why allow that possibility?)

As it stands you have a set of data half of which is slightly above chance and half of which is slightly below. Overall it adds up to chance. (I understand it isn't as simple as that but when you are trying to show something this ground breaking why muddy the waters at all?)

I think it is a good experiment, but I would like to see it carried out with one simple type of stimulus and subsequently replicated by others.

 

[davidsmith73]

Is the habituation effect not simple to display? Why must the erotic and negative categories be introduced?
Why not just display the effect as simply as possible and then you can start showing subtle subdivisions.

I do not see how the use of negative and positive pictures is complicating the experiment. In fact, the introduction of these different types of picture is a demonstration that parapsychology is not pseudoscience. The prediction tested is that subliminally displayed positive pictures will become less positive, leading to the participant chosing the alternative picture, and vice versa for negative pictures. In other words, previous scientific knowledge has been used and built upon. People claim that parapsychology does not do this. Not so.

If you want a simple demonstration then just look at one set of experiments! The fact that Bem continues to test further aspects of his hypothesis is in line with how science should be conducted.

Starting to make predictions about the affect of precognitive habituation on different types of stimuli seems to be putting the cart before the horse a bit.

Not at all. That is exactly how science should progress. You construct hypotheses and test them. Furthermore, Bem bases his hypothesis on previous knowledge about how normal psychology works.

These could all be ways of making the data look better after the fact (I'm not saying they are but why allow that possibility?)

How on earth do you conclude that? please explain

As it stands you have a set of data half of which is slightly above chance and half of which is slightly below. Overall it adds up to chance. (I understand it isn't as simple as that but when you are trying to show something this ground breaking why muddy the waters at all?)

The directions away from chance were in the predicted direction. Combining the data would be contradicting the predictions of the hypothesis

Consider the same experiment except that it is not a precognitive test but a normal test using subliminally presented images before the participant makes their choice, not after. The test would predict that they would prefer the negative targets and the positive non-targets, just like the precognitive experiment. The results might be that the hit rate was 60% on negative trials and 40% on positive trials. Now, if you suggested to a psychologist that combining the data adds up to chance and this shows that the results could be due to chance, he would laugh you out of his office.

I think it is a good experiment, but I would like to see it carried out with one simple type of stimulus and subsequently replicated by others.

Why a simple type of stimulus? In what way would this be a better experiment?

 

[Ashles]

To deal with several of your points at once, there could be a suspicion that the data was collected and, across the board, did not show anything particularly significant. However when the data was subdivided in a certain way then the two groups show individually significant results.

I'm not saying this is what actually happened, but, for a start, I wouldn't be able to raise these questions if he had just used a single stimulus type.

Instead of just trying to show the existence of precognitive ability (this would be amazing in itself!) he tries to go straight to showing the difference of precognitive habituation on different stimulus types. Why?
Show the effect simply and replicably and then you can start to investigate the phenomena further.

Why a simple type of stimulus? In what way would this be a better experiment?

Sorry I wasn't clear there. I meant one single type of stimulus e.g. valency grouped photos. Nothing erotic or otherwise influential - just pairs of similar photos.
Keep it simple and lets firstly just analyse the precognitive effect.

 

[davidsmith73]

To deal with several of your points at once, there could be a suspicion that the data was collected and, across the board, did not show anything particularly significant. However when the data was subdivided in a certain way then the two groups show individually significant results.

The experiment was clearly based on the predictions of normal psychology. What would be his hypothesis if he was intially conducting one big experiment using both types of picture? That subliminally presenting either type of picture would cause the participant to prefer it? That would be in direct contradiction with what we know about the reactions of normal psychology. To be frank, this is rather a desparate resort to bring the experiment into doubt and is not based on any logical criticism about the reality of the experimental method.

I'm not saying this is what actually happened, but, for a start, I wouldn't be able to raise these questions if he had just used a single stimulus type.

What if he first presented the results of the positive stimulus trials? Would you accept them as evidence or simply as an artifact of the procedure? "Artifact!" shouts Ashles.

If critics suggested the single stimulus results are artifactual, what if he then presented the rest of the results that show the predicted opposite effect? Would you then fall back on your original argument? "Data mining!" shouts Ashles.

Do you see how easy it is to construct a way to deny any results are showing a real effect.

And I thought the moon landing conspiracies were far fetched

Instead of just trying to show the existence of precognitive ability (this would be amazing in itself!) he tries to go straight to showing the difference of precognitive habituation on different stimulus types. Why?
Show the effect simply and replicably and then you can start to investigate the phenomena further.

What is complicated about including a different stimulus type that predicts the opposite effects??!!! There is absolutely no complications involved in that whatsoever. In fact, its good science because it tests different aspects of the hypothesis and thus strengthens the likelyhood that we are dealing with a real effect. I'm very perplexed by your views here.

Sorry I wasn't clear there. I meant one single type of stimulus e.g. valency grouped photos. Nothing erotic or otherwise influential - just pairs of similar photos.
Keep it simple and lets firstly just analyse the precognitive effect.

Bem's first experiment in his series was to find out if neutral stimuli could produce an effect compared to more arousing stimuli. Only the high arousal stimuli got a significant effect. So if we took your advice and performed an experiment just with neutral pictures we would conclude that there is nothing there and the effect would go undiscovered. A good scientists however would think about it and test to see if the level of arousal is a factor in psi.

 

[Ashles]

I'm very perplexed by your views here.

Obviously. A simpler experiment would make more sense for an initial demonstration of something as ground-breaking as Psi.

This is one area where the line between "good science" and a desperate need to show some kind of results can get a little blurry.

I guess we just need to wait for some replication of these results before we can comment much further.

As it stands this is a clearer result and less open to interpretation than the Ganzfeld. Overall this would appear to be the absolute best ever experiment for indicating Psi.
Why has it not been replicated more often?

And is 2% deviation from chance in a sample of 400 people really statistically significant? If it is then fair enough but it seems an awfully small amount.
But I did mechanics not statistics so I'll accept other people's information on this.

 

[davidsmith73]

Obviously. A simpler experiment would make more sense for an initial demonstration of something as ground-breaking as Psi.

Experiment 101 attempts to find out if any type of stimulus produces the effect. That's the simple experiment. Remember that it's now encouraged in parapsychology to publish the results of exploratory pilot studies as well as formal studies, to control for the file drawer problem. So in effect, the various experiments that are included in the paper are the result of an investigation into what type of stimulus and what type of participant gets the strongest effect.

This is one area where the line between "good science" and a desperate need to show some kind of results can get a little blurry.

What is specifically wrong with his experiments? Right now you have nothing to justify what you are saying. I think it is you who really wants Bem to have fiddled the results. Take an objective look at your beliefs.

I guess we just need to wait for some replication of these results before we can comment much further.

Here here.

Ashles, will you hereby agree that if the a successful replication of these experiments is performed, you will accept that psi is real?

Why has it not been replicated more often?

Probably for the same reasons that you or Randi have not tried it. I guess this is the first time you have read about this experiment? Perhaps if sceptics stopped concentrating on laughing at the Silvia Brownes of the world we might get much further in the debate.

And is 2% deviation from chance in a sample of 400 people really statistically significant? If it is then fair enough but it seems an awfully small amount.
But I did mechanics not statistics so I'll accept other people's information on this.

It's only 2% deviation if you look at the whole set of data for each stimulus type. Further down the paper the experiments get more interesting in that the high scoring individuals are the ones that find the stimuli most affective, according to various psychological scoring methods. This would indicate that further replications should screen participants to get a larger effect.