Vision From Feeling

[SezMe]

Why is it too big? Is it a PDF? If you want some reduction done I probably can help.
PM me and I'll try and assist in stripping out the text.

There's a limit on the size of post attachments. It is a doc file. I've zipped it up and will try to attach it to this post.

ETA: TaDa, it appears to have worked.

BTW: Black is the original IIG proposed protocol and the blue comments are by VFF.

 

[volatile]

Ah. OK. I'll put my green goggles on and go back to lurking.

Bonus points for the "green goggles" reference.

The ones that show up sperm, from the beginning of Basic Instinct?

 

[Gord_in_Toronto]

The ones that show up sperm, from the beginning of Basic Instinct?

I only get four hits for a Google on "green goggles" "Basic Instinct" so I think the answer is "No".

 

[JWideman]

Thanks, SezMe. VERY enlightening. So, as suspected, she wants to be able to cherry pick the data.
No, VFF, it's not a two-way mirror. It's a "pane of glass". Why would it be a two-way mirror?

I predict either she won't ultimately agree to the protocol, or she'll fail spectacularly and still claim she has some special ability.

 

[godofpie]

I am curious about this part VFF. I was hospitalized in '05 for stomach ulcers. There was never any pain involved. They were weeping ulcers. I went to the doctor because I started passing out from from loss of blood. Would you have been able to see or sense the ulcers since there was no pain? Also, I am a migraine suffer. Would I have to be experiencing a headache for you to diagnose my condition? Thanks.
VFF Quote from protocol negotiations-
" I can not detect pain that is not current. Can the IIG check with the subjects just before the test whether they are currently experiencing the pain? Pain may of course still be listed on their diagnosis form whether the pain is current or not. But for purposes of checking my answers, please ensure that your information on their ailments is current for the time of the test rather than being just current for the time when subjects answered their ailment form. "

 

[Locknar]

I predict either she won't ultimately agree to the protocol, or she'll fail spectacularly and still claim she has some special ability.

Agreed.

 

[nathan]

But she's told that her ability doesn't work always, but when it does it's accurate.

Didn't the same kind of claim come up with 'Edge' (the dowser)? I suspect it comes up in a lot of claims

The problem is how does she know when here ability is working? Is it a postdiction? Is it special pleading (oh, that's the wrong type of pregnancy)? Is she cluing from something else? (I am reminded of a classic Neural Net experiment to detect tanks hiding under trees, it worked perfectly on the test data, but failed spectacularly in real life. It had learned how to spot sunny and cloudy days, which just happened to correlate with the test data.)

My understanding is that's why the forced choice from a fixed pool is the way to go. It forces you to test the claim you make -- a non-forced choice is testing something else.

 

[Ocelot]

OK I've read the protocol and it looks like a good start.

I must say that some of the objections being raised are simply invalid. It is the role of the testing organisation to test the claims being made rather than to harrngue the claimant into making a more easily tested claim.

It is not Anita's claim that she will always see a health problem when it is present. It is her claim that when she does perceive information, it is uncannily accuarate.

Many paranormal claims are capricious in nature. That the claimant is unable to determine what factors influence whether they will recieve the benefit of their powers or not is not sufficent reason to dismiss the existance of the effect.

This is accommodated by the protocol by allowing Anita to pass on a subject. Such selection has been dismissed as cherry picking. It isn't. Cherry picking requires that the picker has access to the unblinded information and chooses only cases where their predictions conform to reality. Anita doesn't have access to that information when she makes the decission to pass. I would challenge those who suggest this is cherry picking to devise a way to game the test to produce an unwarrented pass by exploiting the ability to pass. I don't believe that it can be done.

Science is littered with test which can be positive or inconclusive but not negative. We are, nontheless still able to evaluate the reliability of these tests.

For this reason I question the presence of a healthy volunteer. I Anita claiming that she can see someone who is perfectly healthy and (occassionaly) declare them to be so. Or would her ability suffer from the same limitations of modern medicine in that it can only assume someoen is perfectly healthy so long as it can find nothing wrong?

Such selection can be problematic in another way. A claimant who wants to go along with the idea of testing but knows that they can't pass will often try to engineer an exit strategy. They can't pass but at the saem time they can ensure that they can't "fail" The responsibility for not passing is then palmed off onto the testers for being unable to accomodate the claimant's needs.

We know then that Anita is aware that her abilities are capricious, we have no need to prove that to her. We have no specific reason to doubt her sincerity in wanting to be tested. However past experinece of a great number of insincere claimants has understandably jaded some of the responses here.

Accomodating her request for more participants is a hurdle. Recruiting people with ailment that have no outwardly visible indications is not easy. I see that the IIG has made some effort to be more inclusive in thier selections by aiming to diguise some ailments. Sunglasses glove etc. No doubt they are aware that if one volunteer is to wear sunglasses to diguise an eye infection then all must do so or it is easy to discount those that are not wearing sunglasses as not suffering from that ailment. The more subjects, the more ailments involved, the greater the measures must be taken to disguise those ailments. The situation that must be avoided is that Anita is left with no uncovered skin to view, a stipulation she has every right to make if she believes it affects her abilities.

I might suggest an alternative form of viewing - if Anita is agreeable. The subject is sat in front a long viewing tube which restricts the field of vision from the other side. Their exposed back faces the tube. Anita on the other side will only be able to see and area of skin - say about a foot square.

Obviously the same precautions with regard to viewing the subject entering and exiting the viewing area should still be made.

It is the difficulties with procuring suitable subjects for this sort of test that will cause substantial delays in this testing. WHilst I do understand that Anita want to be tested using the application of her ability with whihc she has most experience she must recognise that replicating these experiences is a time consuming matter. Whereas replicating the detection of lactobaccillus in breakfast cereal is relatively straightforward. As such it would be easy for her to aquire a great deal of such straightforward experience prior to testing. I encourage her to report back with her findings in her own test that she's performing on this as it seems to me to be a very promissing avenue of investigation.

The difficulties in recruitment may be especially true if ailments such as intermittant pain cannot be detected. Once agian we must aim to test the claim. If there is a known cause to the pain that can be detected by Anita, then all well and good but we cannot fail her for not detecting intermittant pain that is not present at the time of testing if she does not claim to be able to do so.

I understand the potential need for excluding auditory prompts. If Anita envisages that this might cause difficulties it may be possible to for subjects to be sleceted with conditions that should not be through sound. After all - "I can hear skin gingivitis" would be a paranormal claim in it's own right. However if the worry remains in the minds of the investigators that sound is a problem then perhaps Anita would find noise cancelling headphones or her own choice of music less of a distraction.

There is provission for Anita to continue with the testing after having made a diagnosis on the minim of ten patients. I can see no circumstance when this would be of benifit to Anita's chances of scuess if the subject pool is limited to 15. With a pass mark of 90% if she has diagnosed 10 patients then additonal attempts can only preserve a pass or trun it into a fail. If she has only got 8/10 at that stage (the highest possible without passing) getting the remaining 5 right will still leave her with 13/15 or 87% - a fail.

I'm having difficulties working out the statistics for a chance pass.

Will Anita be presented with 15 possible diagnoses for 15 patients or is there a list of symptoms where each subject may have any number of said symptoms, might this include multiple patients with the same combinaiton of symptoms. With the necessary complication of Anita being able to pass the whole matter becomes somewhat tricky to analyse, with a statistical concurrence of certain symptoms meaning that an educated guess would perform significantly better than random ticking of boxes without any need for paranormal abilities.

How then was the passmark of 9/10 arrived at?

 

[JWideman]

My objections are regarding Anita's objections, not the protocol itself. Particularly her desire to have it not be counted as a failure if she is unable to diagnose any of the subjects. That's why I don't think she'll agree to any protocol.
How many subjects would she need in order to make 10 observations? 100? 1000? In 4 days she's documented only 2 observations on her website.
What I find very odd is that, after insisting so strongly upon this point, she states at the end that her only objection is to the music. Which suggests to me that if she does get to the point of being tested, she will claim the protocol wasn't what she agreed to.
To address your concern, each subject has one ailment.

 

[Locknar]

It is not Anita's claim that she will always see a health problem when it is present. It is her claim that when she does perceive information, it is uncannily accuarate.

Many paranormal claims are capricious in nature. That the claimant is unable to determine what factors influence whether they will recieve the benefit of their powers or not is not sufficent reason to dismiss the existance of the effect.

This is accommodated by the protocol by allowing Anita to pass on a subject. Such selection has been dismissed as cherry picking. It isn't. Cherry picking requires that the picker has access to the unblinded information and chooses only cases where their predictions conform to reality. Anita doesn't have access to that information when she makes the decission to pass. I would challenge those who suggest this is cherry picking to devise a way to game the test to produce an unwarrented pass by exploiting the ability to pass. I don't believe that it can be done.

<snip>

I diagree; you have too many outside influences to deal with to allow passing on subjects.

Being able to look at the "subject"...perhaps she passes on those she thinks (consciously or sub-consciously) look healthy/fit, look happy, look clean/well groomed, smell nice, etc.

By default, this would yield a selection pool of only those looking unhappy, unhealthy, etc. and thus increase the odds of a correct guess.

 

[Coveredinbeeees]

I put my hands up and admit that I haven't read the protocol yet but, from the comments of those who have I get the feeling that having a control group of guessers might help.

If a number of people with a similar level of biology / anatomy training to Anita were to take the same test they would be likely to pick out subjects with visual clues as to their ailments and the test could be run such that Anita has to out-perform the control group.

Just a thought,

'beeees

 

[nathan]

Being able to look at the "subject"...perhaps she passes on those she thinks (consciously or sub-consciously) look healthy/fit, look happy, look clean/well groomed, smell nice, etc.

Yeah, that's what I was trying to say in an abandoned posting. One ailment she claims to be able to diagnose is 'missing arm', (but note, not 'missing finger' -- a substantially smaller body part). Being able to sometimes tell when someone is missing an arm when you can observe them, and know when you are sure about it, is not really a paranormal skill is it?

I know the subject is supposed to be covered to some extent, but the lack of an arm may well be clear in some circumstances -- all you need is to pass when those circumstances don't arise.

There's a lack of definition of what constitutes 'missing arm' - below elbow amputee, above elbow amputee, shoulder amputee? Are they allowed to wear a prosthetic?

ETA: Here's the algorithm for passing the current protocol:

*) for all observations where it is impossible to determine that the subject is an amputee, pass -- your power is weak on that one.
*) for all other cases claim 'amputee'
*) repeat until 9 amputees spotted
*) success!

[replace 'amputee' with more specific amputee, if the protocol demands it, I can't recall if it specifies]

 

[Ocelot]

ETA: Here's the algorithm for passing the current protocol:

*) for all observations where it is impossible to determine that the subject is an amputee, pass -- your power is weak on that one.
*) for all other cases claim 'amputee'
*) repeat until 9 amputees spotted
*) success!

[replace 'amputee' with more specific amputee, if the protocol demands it, I can't recall if it specifies]

OK that wouldn't work for one particular reason. Not enough amputees in the pool. If 9 of the 15 subjects had the same diagnosis that in itself would be a major flaw in the protocol.

However you've stated that under certain conditions it would be possible to detect a stationary amputee despite their amputation being disguised. That would clearly be a spearate flaw in the protocol if those conditions were allowed. Can you be more specific?

 

[Pup]

How did the test go?

Anyone noticed we suddenly seem to be talking among ourselves? Where's VFF the last few days? I'm curious how the cereal self-test went.

 

[nathan]

OK that wouldn't work for one particular reason. Not enough amputees in the pool. If 9 of the 15 subjects had the same diagnosis that in itself would be a major flaw in the protocol.

If we run out of test subjects before reaching 9 claimed amputees, then we're possibly into swedish diary territory. Let me explain: If there was at least claimed one amputee (and I'm making the assumption that it is not a false positive, because we've carefully observed the subject before claiming the amputation), we'll have one hit. That would fail the protocol, because of the bounded number of test subjects, but would allow the claimant to say 'well, when I was able to make a judgement, I was 100% correct, it was only the lack of test subjects that stopped me succeeding the challenge'.

However you've stated that under certain conditions it would be possible to detect a stationary amputee despite their amputation being disguised. That would clearly be a spearate flaw in the protocol if those conditions were allowed. Can you be more specific?

I am claiming that the disguise will not be 100% perfect. Combined with the ability to reject particular subjects, that's all that's required to defeat the protocol by mundane means. IIRC the disguise mentioned is a shawl/cloak/gown -- have you tried to sit *perfectly* still for 10 minutes?

If the disguise is more like showing the face through a hole in a rigid body -- like the seaside amusing photo things -- then I think it is much better blinding for mundaneness.

 

[Ocelot]

My objections are regarding Anita's objections, not the protocol itself. Particularly her desire to have it not be counted as a failure if she is unable to diagnose any of the subjects. That's why I don't think she'll agree to any protocol.

To be fair any protocol for a capricious ability can only uphold the ability. If the claimant doesn't manage to prove their claim then no conclusive determination can be made. The results are presented and whoever reads them can draw their own subjective interpretation - they may lean towards the claimant having an off day or the claim being incorrect it's up to them.

Even if she does make the required ten diagnoses and all ten are incorrect it doesn't prove that the times she's claimed she was correct she wasn't obtaining information by paranormal means.

It would prove that her abilities weren't as consistent and reliable as she thought.

If you were testing the accuracy of facial recognition software in a simulated crowd scene you could take two measures. Lets say that there were X people in the crowd scene, Y people had visible faces all of which the software attempted to match and the software correctly matched Z faces. The ratios Z:Y would be a fair representation of the software's accuracy. The lower ratio Z:X would be a measure of the usefulness of the software in that scenario, however it would be unfair to count that lower ratio as a measure of the software's accuracy. One might as well include all the people in the world who didn't even appear in the scene as failures.

Diagnostic tests vary in their accuracy, different measures are taken: Predictive Value, Specificity, Sensitivity. Anita is not claiming that her ability kicks in each and every time there's something to see. Just like the purveyors of facial recognition software freely admit to being useless when presented with a picture of the back of someone's head. Instead she's claiming a low false positive rate that is all. So that's what must be tested.

If her ability doesn't kick in at all during the testing then her ability won't have been proven, the result will be inconclusive but we will have learned a little about how applicable her abilities are under such conditions.

My guess is that her ability works as a form a synaesthesia, but instead of these visual clues being linked to a sense of sound smell, or and idea of number, it's linked to a sense akin to a little Dr Gregory House inside her subconscious. I don't think it's a paranormal sense I think she's unaware that she's picking up many many clues and assimilating them into a diagnosis.

I have no reason to doubt that Anita is genuine in her desire to find out how her ability works or any reason to believe that she's dogmatically bonded to finding out that it must be paranormal.

How many subjects would she need in order to make 10 observations? 100? 1000? In 4 days she's documented only 2 observations on her website.

I think that's a very interesting question. It would appear that from Anita's own observations she's as yet unable to answer that question. She's aware that the people with whom she shares her observations tend to confirm them. As we both know there are many non-paranormal reasons why this may be the case. There are sources of information other than the paranormal that her synaesthesia may access, there are biases that may give the impression of greater accuracy that is actually being obtained. What she doesn't know is how many of the people that her peculiar senses ignore, are suffering from complaints that she's previously been able to identify in other circumstances.

If nothing else this test should give us some sort of indication of the answer to this question.

If she manages 8 diagnoses from 15 we know that the protocol would only need a little tweaking if it were attempted again.

If she only manages two diagnoses from 15 we would have some indication that a much larger pool would be required - quite possibly making the test impractical.

Incidentally would the test be called off early if she passed on 6 out of 15 meaning she could no longer reach 10 out of 15 diagnoses or maybe if she passed on 7 out of 15 meaning that she could no longer reach 9 correct diagnoses. What would happen if she could only diagnose 9 people but they turned out to be all correct - would that be a pass? It would appear to be more statistically more improbable than getting 9/10?

A game theorist with 9 diagnoses under her belt would take a guess on the 15th subject anyway even if they couldn't see anything.

The 2 recent observations on her site are not very indicative of the paranormal. In both instances there is the potential for other sources of information about the complaint. The reporting that there were no visual indications is entirely subjective with Anita clearly unable to objectively document what information was available to her subconscious.

They're not exactly ideal examples for the test material but we must persevere and do our best to give her the best chance of passing the test or else the test means nothing.

What I find very odd is that, after insisting so strongly upon this point, she states at the end that her only objection is to the music. Which suggests to me that if she does get to the point of being tested, she will claim the protocol wasn't what she agreed to.

She did request a larger pool if possible. I think it would be a shame if the test were inconclusive - and I have not problem using that term instead of failed - because Anita was just short of the required number of diagnoses.

I do see that from a preconception that the claimant is trying to engineer and exist strategy then her asking for a larger pool, requesting the test be considered inconclusive rather than a failure if the pool were not large enough but then agreeing to the original suggestion for the size of the pool all confirms such a strategy.

However these observations are not proof of that scenario. They are equally consistent with a person who believes that they have the powers that Anita claims and doesn't want the difficulties with raised by recruiting additional test subjects to delay the tests.

To address your concern, each subject has one ailment.

And I should assume their ailment is unique within the pool?

 

[nathan]

The claimant is (rightly) concerned that her ability doesn't work all the time. But that's a red herring. No ability works all the time, there's always some failure rate. All it does is change how you calculate what's needed to distinguish the ability from random chance. Having the claimant allowed to reject test subjects as unsuitable *during the test* is manipulating that threshold.

Here's an example:
Claim:I have a magic[*] screwdriver that can unscrew slot-headed screws.

Protocol: As selection of slot headed screws are provided, screwed into a number of different blocks of wood. Each one is presented in turn and I have 10 minutes to unscrew the screw. However, I am allowed to reject a screw/block combination as unsuitable during the test, if I believe my magic screwdriver fails on that one.

Pass Criteria. Of those that I claim I have unscrewed, I have to have been correct for 9/10 of them. I have to have unscrewed at least 9.

See the problem? It's obvious I can pass this test, if there are sufficient blocks of wood. And if there are insufficient blocks, I can still claim that for the cases where I claimed I had unscrewed the screw, I was right 100% of the time -- presuming I'm not an idiot who cannot tell when a screw's been unscrewed

There are several ways of fixing this. One is to allow me to reject at most N screw/block combinations out of the pool of M that I attempt. But that's just a fancy way of changing the pass threshold. We could reduce the threshold to 7/10 or something. Reducing the threshold makes it much clearer how far away one is from random chance. With a forced choice design, the random chance calculation is easy to do.

In VFF's case, the concern seems to be that out of 10 subjects VFF won't get the tingly feeling for all 10 of them. But for 15 subjects she thinks she will get the tingly feeling for at least 10 of them. When she gets the tingly feeling, she thinks it is right at least 90% of the time. If that's the case, then use all 15 test subjects and have a threshold of diagnosing at least 9 of them. Then the maths is much easier to figure out given the number of different aflictions and the number of subjects with each afliction.

[*] not very magic

 

[Ocelot]

If we run out of test subjects before reaching 9 claimed amputees, then we're possibly into swedish diary territory. Let me explain: If there was at least claimed one amputee (and I'm making the assumption that it is not a false positive, because we've carefully observed the subject before claiming the amputation), we'll have one hit. That would fail the protocol, because of the bounded number of test subjects, but would allow the claimant to say 'well, when I was able to make a judgement, I was 100% correct, it was only the lack of test subjects that stopped me succeeding the challenge'.

In which case the test wouldn't have been passed and we would ahve obtained information as to the applicability of the claimant's powers. Everybody's a winner. I'd have no objection to that.

I am claiming that the disguise will not be 100% perfect. Combined with the ability to reject particular subjects, that's all that's required to defeat the protocol by mundane means. IIRC the disguise mentioned is a shawl/cloak/gown -- have you tried to sit *perfectly* still for 10 minutes?

If the disguise is more like showing the face through a hole in a rigid body -- like the seaside amusing photo things -- then I think it is much better blinding for mundaneness.

Indeed, I too had concern about the disguises method. That's why I suggested a viewing tube where Anita could only see a portion of the subjects exposed back. Your cut out idea is simpler though now I come to think about it the face if rather more expressive than the back. I'm wondering about the possibility of some sort of Clever Hans scenario if the subject can see Anita's deliberations. As such the back owuld be preferable so that the subject couldn't see Anita and even if the subject did pick up infomration about Anita's deliberations they're less likely to give it away via their back. At the very least Anita shouldn't be castigated for asking if it would be a "two way mirror" anything that stops the subject from picking up where Anita is focusing her attentions is a good idea.

I update my suggestion to incorporate your method. The subject should sit with their back exposed to a rigid screen with a hole that makes a aportion of thier skin visible from the other side.

I have sat posed in art class but that was a long time ago when I was in the peak of physical fitness and able to choose a relaxed and comfortable pose. I guess that's kinda the point.

 

[Ocelot]

The claimant is (rightly) concerned that her ability doesn't work all the time. But that's a red herring. No ability works all the time, there's always some failure rate. All it does is change how you calculate what's needed to distinguish the ability from random chance. Having the claimant allowed to reject test subjects as unsuitable *during the test* is manipulating that threshold.

Here's an example:
Claim:I have a magic
[*] screwdriver that can unscrew slot-headed screws.

Protocol: As selection of slot headed screws are provided, screwed into a number of different blocks of wood. Each one is presented in turn and I have 10 minutes to unscrew the screw. However, I am allowed to reject a screw/block combination as unsuitable during the test, if I believe my magic screwdriver fails on that one.

Pass Criteria. Of those that I claim I have unscrewed, I have to have been correct for 9/10 of them. I have to have unscrewed at least 9.

See the problem? It's obvious I can pass this test, if there are sufficient blocks of wood. And if there are insufficient blocks, I can still claim that for the cases where I claimed I had unscrewed the screw, I was right 100% of the time -- presuming I'm not an idiot who cannot tell when a screw's been unscrewed

There are several ways of fixing this. One is to allow me to reject at most N screw/block combinations out of the pool of M that I attempt. But that's just a fancy way of changing the pass threshold. We could reduce the threshold to 7/10 or something. Reducing the threshold makes it much clearer how far away one is from random chance. With a forced choice design, the random chance calculation is easy to do.

Frankly that's just incorrect. I've just generated 100 random integers between 1 and 10. I don't care which ten you pick if you can predict 7 out of 10 of them correctly I'll believe you're not just guessing.

If the protocol is sucessfull in ensuring that Anita is not privvy to non-paranormal sources of information it doesn't matter that she can pass on a subject during the test. It's only if she can select which subjects are to be counted after receiving information that this becomes a problem.

The concern that the disguises may not be 100% perfect is indeed valid, as it does mean that she's making the decission when to pass after reciveing information from a non-paranormal source. So the ability to pass on a subject compounds that imperfection.

Since stopping her from being able to pass, completely invalidates the test the only solution is to fix the imperfection in the protocol which gives her access to non-paranormal sources of infomration since this would be a concern even if she didn't have the ability to pass on a subject.

In VFF's case, the concern seems to be that out of 10 subjects VFF won't get the tingly feeling for all 10 of them. But for 15 subjects she thinks she will get the tingly feeling for at least 10 of them. When she gets the tingly feeling, she thinks it is right at least 90% of the time. If that's the case, then use all 15 test subjects and have a threshold of diagnosing at least 9 of them. Then the maths is much easier to figure out given the number of different aflictions and the number of subjects with each afliction.
[*] not very magic

I'm afraid that's just wrong again. Firstly Anita hasn't said how often she'll get the "tingly feeling" as far as I can tell she hasn't done a test that would give her this information. For me, thats a secondary objective of the proposed test. She certainly hasn't said she'll get it 10 out of 15 times in fact she's asked that the number 15 be upped to increase the chances that she'll get at least 10 "tingly feelings"

More importantly your suggestion of changing the pass amrk based upon how often Anita feels she can get a result is open to abuse. As it stands getting 9 right out of 15 guesses is orders of magnitude easier than getting 9 right out of 10 guesses (even if those are the ten out of fifteen in which she feels most confident) By extrapolating the principle ad absudium it becomes more obvious. What if she were to say that she gets the tingly feeling only one time in a hundred thousand, would ten correct precitions out of a million uphold that claim?

 

[EHocking]

Anyone noticed we suddenly seem to be talking among ourselves? Where's VFF the last few days? I'm curious how the cereal self-test went.

I find it quite intriguing how VFF's claims have changed over a very short time.
From OP on:

So far my accuracy on health information has been excellent.

I would remember if I had been wrong, and I have not been wrong yet.

by using this ability I have obtained information about chemistry, materials, plants, animals, bacteria, foods and medicines ... I do not have a single example of when I would have been incorrect.

I haven't been recording my hits and misses, yet I seriously can not recall a single time when I would have been incorrect. I almost expect mistakes sometimes, ..yet there have been no mistakes yet.

I have had many confirmed hits. I would estimate it as at least a hundred, ...I can not recall a single miss.

I wouldn't take it personally if I made mistakes in my observations, it's just that there haven't been any.

I am very open to experiencing my first incorrect observation.
Yet, so far, always correct.

Yes I am open to becoming convinced that my information is inaccurate! ...Yet so far always correct.

Then suddenly, post IIG protocol interest...

My accuracy is good,

Luckily as I've just stated I do not detect each case in which an ailment occurs, nor have I ever claimed to.

There are actually plenty of cases where I do not detect things that are considered as being present, and this is the disclaimer for my ability that I have always outlined. I have never claimed to be able to detect all cases in which an ailment occurs, since I do not detect in all cases in which an ailment occurs.

I have however had to ask for an additional condition to the test. That is if I do not detect something in a total of ten persons, then I do not want that to constitute a failed test.

... and finally with no sense of irony at all,

I am thus not consistent with most other typical claimants.