posted 05-14-2012 05:09 PM            

Everything is the same inside, but it is a nicer package, about half the size of the LX4000.

I use the modular right now, because, well, I can, and because it is easy for me to snap in an additional EDA module, second cardio or 3 more movement channels.

Most people seem to like the nice neat single unit LX5000, and I would too if I were not always taking something apart or adding something to study it.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-15-2012 04:56 PM            

Ray – since you’ve mentioned that you like to take things apart and add things, would you like to comment on this image and the board that was “added” to replace the EDA circuit in the LX4000? Is this the same attention to detail Polylawman can look forward to in his LX5000, modular or all-in-one? And what is the explanation given to examiners regarding the EDA deficiencies in the LX4000?

posted 05-16-2012 12:47 PM            

One of the things that I really liked about this forum in the past was that it was free from advertising and marketing.

I would hate to see that change.

The value of this forum has been the opportunity to meet professionals and make friends from around the world, and to achieve a better dialog and understanding of polygraph practices and fieldwork among professionals in government service, law enforcement, and private practice.

Examiners were free to argue and discuss anything without being forced into market positions. Or we could just goof off and that too was actually valuable, though in a more social/professional sense.

In fact, we had a brief forum/group discussion at one time to achieve a consensus on whether or not instrument vendors would or should be permitted into this forum.

The hazard that comes with vendor participation in this forum is that, as customers and representatives, polygraph examiners can sometimes become easily polarized and can actually begin to start working against each other instead of working with each other to improve our knowledge, skills and professionalism.

Because instrument vendors are so heavily invested in things like market share as opposed to knowledge, skills and professionalism, there may be a tendency to engage in highly competitive, if not aggressive marketing.

In the worst cases, marketing efforts become negatively focused and begin to spread mis-information about competitors, instead of marketing and selling products based on their strengths and merits.

It is my belief that the difference between low-road marketing (i.e., attempting to sell products by criticizing competitors) and high-road marketing (i.e., marketing and selling products based on strengths and merits) will ultimately shape the character and experience of this forum, the profession, and perceptions of the polygraph profession both within the profession and on the part of the community.

Look now at the recent legislative efforts to win approval for voice stress as an alternative to polygraph in sex offender treatment and supervision programs in some states. Legislators have a tendency to perceive the argument between polygraph and voice stress as nothing more than a turf-war. They do not find it helpful.

The point is that negative advertising does NOT make anyone smarter and does not help anyone to make better decisions - any more than negative campaigning during elections can help voters make better informed decisions. Negative marketing, like negative campaigning, has the effect of creating division and adding confusion instead of providing good usable information.

Readers of this forum should be encouraged to understand the strengths and assets of all instruments available today. Readers should also understand that vendors have inherent interests to make there products look good while making the products of competitors look bad.

Some industries have had to develop ethical standards for the types and forms of advertising that are permitted. Tobacco companies, for example, are no longer permitted to use Joe-the-camel or images that look like cartoon characters likely to grab the attention of children. Pharmaceutical companies, for another example, will tend to avoid making misleading or negative statements about competing products unless they are citing research. Even then, research that is done independently, or reviewed independently, gives a much different impression to intelligent folks than research from within a company proving their products superior to the products of competing companies.

On the other hand, politicians and interest groups tend to operate with the ethics of alley-cats - and will tend do or say anything in order to achieve an objective.

The purpose of this forum was to promote discussion, connections, knowledge and more effective work. It will be a great loss if we transform this forum into nothing more than a way to turn polygraph examiners into customers for vendors.

Here is a link to an interesting thing about marketing: and the trend of change from a market economy to a market society.
http://www.nytimes.com/2012/05/13/opinion/sunday/friedman-this-column-is-not-sponsored-by-anyone.html

Lets take the high road here, and let the forum continue to provide a great benefit to the profession - without a lot of negative advertising from competing vendors.

Bottom line: if we pollute this forum with marketing nonesense, then it will be at risk for remaining polluted FOREVER!

This forum is too valuable to ruin.

If anyone has any questions about the Lafayette LX4000 or LX5000, then please contact Lafayette Instrument directly for accurate information about the merits and designed capabilities of the LX system.

As for this forum, I propose that you (David) and I allow others to continue this discussion of their experiences, questions, and impressions - and refrain from further comment unless someone has a question for us from the field.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-16-2012 01:04 PM         

Let me get this straight... You're advocating censorship for our own good?

posted 05-16-2012 01:46 PM            

Let's not pee in our own pool by turning this forum into a lot of advertising - or worse, fill it with toxic sludge in the form of negative marketing.

I'm advocating for restraint, in the form of self-control.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-16-2012 03:18 PM         

To paraphrase: Let them report, and we'll decide.

posted 05-16-2012 04:11 PM            

Well said, from the granite state (live free or die).

My answer...

Toxic sludge.

Ultimately it is the responsibility of each individual to not play in it, not go near it, and of course don't put yer mouth on it.

Do we really want a laisse-faire society? If so then anyone who owns a property can dump whatever they want onto it.

Advertisers can say whatever they want and make it look like journalism - with no requirement to state that it is advertisement.

Cigarette companies can market to children - because it is the parents who decide and influence the kids. Right? (And besides, there is research - funded by tobacco companies - showing that 80% of lung cancer deaths could be avoided through CT scans.)

Kellog, ConAgra, Sysco and Purina can put whatever they want into their products - because you can read the label. They can say anything is "natural" or "organic" with no meaning to the word - and that pink slime is "meat."

And, why try to micro-manage capitalism (and banking/investments) at all? People know the risks (they are told the risks in writing).

On the home front, why license polygraph examiners? Or doctors... It should be the responsibility of the consumer to check the credentials of the professionals. Why worry about voice stress vendors and the things they say in public and to legislators? (Anyone with a brain will seek out the facts and make the right choice.)

Now seriously: Do we really want to pollute this forum with negative advertising driven by business goals? I suggest no. Keep the discussion focused on productive and useful information related to conducting the effective and accurate polygraph examinations in the service of more effective risk assessment and risk management. Community safety, national security and individual justice depend, in part, on good polygraph.

The best way to be helpful will be to ask each vendor to market themselves based on the strengths and merits of their products - and to refrain from misinformation and negative advertising as a form of marketing.

Of course there will be the tendency to show only the best side of everything. It is ultimately the role of the customer to learn to see the bigger picture, and evaluate the different advantages. If you want accurate information comparing one instrument to another, then I think you should insist on independent research and evaluation, but even that will be frustrated by the lack of transparency of information. So we also need open source and standardized models for some things.

There is always a balance to achieve between free-market competitiveness - which leads to innovation and advancement, and transparency and standardization - which can lead to less waste, greater interoperability, less volatility/increased stability and more satisfaction in the field.

Anyone want to go back to the days when every cell phone had a different charging connector?

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-17-2012 07:12 AM         

Fearmongers who advocate the suppression of free speech "for the greater good" are far more dangerous than virtual toxic sludge.

posted 05-17-2012 10:05 AM            

I hope you’re not advocating that examiners ignore important information about their polygraph software and instrumentation. If software and instrumentation professionals bring information to this forum do you seriously think examiners should censor that dialog? David may have cast a stone at Goliath but he certainly didn’t try to sell a Limestone polygraph. Limestone Technologies has never abused this forum for marketing initiatives. Why all the smoke regarding competitive marketing and toxic sludge?

I’m a little disappointed that you neglected to provide any answers to the deficiency that’s displayed in the image. You present yourself as a Lafayette sales representative, a Lafayette school director in Mexico, an unbiased researcher and scientist, an APA director and private examiner. Why do you immediately assume Dave’s message to be marketing misinformation? You’re usually very vocal about the value of scientific peer review and how we’re all expected to elevate the perception of credibility assessment in the scientific community. Shouldn’t the software and instrumentation also be held to some standardized level of safety and reliability?

The polygraph instrument is a consumer electronic device that’s designed to measure precise physiological measurements. Do you think the replacement LX4000 EDA should be floating loosely inside the enclosure? Should it be protected with cardboard and electricians tape? What deficiency necessitated a complete replacement of the EDA circuit? It may be your opinion that the workmanship we see displayed in the LX4000 image is acceptable however this is clearly not your professional expertise. My career has been spent working in the scientific community and my vocation is Electronic Engineering Technology. My professional training and experience sees this as a red flag and I think we should all have some genuine concerns. This LX4000 hardware revision would not pass the scrutiny of an inspection by any professional consumer protection agency. Let’s all critically consider the impact of a defective EDA and how this would affect the final chart analysis. If you think that the image Dave posted is over sensationalizing the problem, please explain why?

Obviously Dave has misplaced his post on this topic. This would be a more appropriate thread for Skip’s new topic, Polygraph Mistakes & Corrections. Let’s all hope that examiners are never censored from learning more about genuine concerns regarding instrumentation.

Respectfully,

posted 05-17-2012 11:15 AM         

See what you have done? I hope you are happy! Seriously, you did not create ANY negativity here. You simply lit the fuse and now everybody is playing catch with the burning dynamite! Sit back and watch the fun!

The gloves frequently come off (or were never on) in our discussions which is what makes this forum so great. It is a good way to get more info and to learn something you might not have known yesterday.

"Ding-Ding-Round 2".

posted 05-17-2012 12:13 PM            

I will continue to provide any information I can to the polygraph profession and polygraph professionals. However, I cannot possibly have the answers to these engineering questions, and it therefore makes no sense to continue to ask these questions of me.

I would like to invite anyone with any questions to contact Lafayette directly for information about the design and development of the LX4000 and LX5000.

I will also continue to suggest that this forum will foster discussions of greater value to the profession if we try to avoid market-wrangling in here. It serves only to polarize people and damage relationships.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-17-2012 12:13 PM            

The photo went up here with a picture of a new EDA circuit, and the question had to do with how that new EDA implied a deficiency in the old one, but yet no evidence was offered to support that being the case. Maybe it is. I don't know. Maybe they couldn't purchase the old circuit due to technology changes. Maybe this one isn't as good as the other, but maybe it doesn't matter to the practicing examiner.

Think about this: there's a manufacturer out there who would (and has) said that this discussion is academic since that manufacturer is the only one to understand EDA making all others inferior (deficient). Where does this conversation end, and how do examiners assess the situation? If people really want to educate examiners, then engage in a discussion of the hows and whys along with the unknowns, assumptions, etc. In other words, educate examiners so they can make their own decisions as to what is best. The real question is whether any of this matters. We have confirmed cases with lots of data from different instruments, and it would appear results from those would therefore generalize to what examiners are doing in the field.

In my experience, one instrument had a lot of plunging EDAs. I didn't have that same experience with others. Did it matter as far as results are concerned? I doubt it. Was it a pain? That's a different question.

Since examiners have these discussions off-line, I think it's healthy to have them here. I think the manufacturers should try to respond to questions and issues, but any negative statements about somebody else and their product is going to raise suspicions, especially since some of the comparisons I've seen in the past are inaccurate.

I think a better approach would be to say what you offer and why you chose what you chose for options, circuits, etc. There are differences, and some are based on assumptions (opinions), and some are based on science.

Every time a manufacturer makes an improvement, we could ask how the prior version was deficient, and next we'll be hearing how a test was invalid because of the "deficiency" that really didn't matter anyhow. I don't want to have those senseless conversations, especially if lawyers are involved.

posted 05-17-2012 01:19 PM            

quote:

---

Fearmongers who advocate the suppression of free speech "for the greater good" are far more dangerous than virtual toxic sludge.

---

In general I agree that virtual toxic sludge is less scary than the idea of interfering with free speech. But there are limits to everything, including free speech.

But I wonder if this is an absolute position? Is it truly black or white with nothing in between? Because so few things in life are actually black and white (spanking children, for example: is it just bad always under any circumstance? because it is violent. just curious).

Oliver Wendell Holmes struggled with this very issue when he wrote this

quote:

---

The most stringent protection of free speech would not protect a man falsely shouting fire in a theater and causing a panic.

---

Its all complicated, and it is important to not get crazy over it. Discussions over natural and legal rights go the heart of the social contract philosophies of Locke and Rousseau and others who so clearly influenced the American Revolutionaries who fought for and founded the freedoms that are protected by our constitution and those who serve today.

Looking forward to seeing you again when we get to your part of the country in a couple of weeks. You can help me introduce my lovely bride to some of the practical details of our 1st and 2nd Amendments.

Peace,

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-18-2012 12:55 PM            

- Opposite EDA recordings depending on the manual versus automatic setting

- Static discharge problems that resulted in loss of chart data

- USB communication problems that are directly related to a short-cut in the design. The LX4000 is an RS232 serial device that employs inexpensive serial to USB flip chip transceiver technology.

Members of this forum should appreciate that we aren’t the only ones who have noticed the Lafayette EDA anomalies. There was a thread on this very forum started by Poly761 in May 2008 when he posted charts with both the EDA manual and detrended displayed. Here are some comments you may recall.

Barry C - Polygraph Place -May 26 2008
“It appears if one uses the "bigger is better" rule, you could get two different scores depending on whether the manual or
detrended signal was used. What is the detrended filtering, and which is correct? This looks like, for the most part, slow drops
(and you'll notice the non-detrended is constantly dropping) are filtered and put on a baseline.
In any event, I'm curious as to what it's doing. (I have a Lafayette, but I only used it a couple times since the new detrend
feature was added. It looks like a much better EDA than they had previously, but this jumped out at me.”

rnelson - Polygraph Place - May 29 2008
“Look at cht2c and you'll see what looks to me like a problem. The raw EDA on C4 is plunging, while the detrended EDA shows a
small ascending segment after answer, of almost 1/2 chart division. The next question, R10 shows both raw EDA and detrended EDA both show an ascending segment shortly after answer, of just over 1/2 chart division. The problem, for me, is that
ascending segments are interpreted as indicative of Sympathetic activation. The raw EDA at C4 does not indicate sympathetic
activation, and what looks like sympathetic activation in the detrended EDA is fictitious - it is a feature of the filter's math, and not the sympathetic nervous system. “

Let’s all realize that we’re not talking about perceived problems. These are very real problems and they’ve been kept a little too private since discovery in 2007. Can examiners really call the manufacturer and expect full disclosure now? The smoke show has gone on long enough and it’s time for some honest answers.

Lafayette presently claims that once an examiner has read up on their 6 new EDA modes they “will agree that Lafayette has the best EDA in the industry”. Lafayette is dancing on a very slippery slope by adding more digital filtering choices and this could prove to be counterproductive. The two modes Lafayette first introduced have presented enough complications for examiners and adding four more choices will no doubt compound the problem. Doesn’t the profession deserve an independent unbiased validation study to ensure the actual specificity and reliability of these new modes of operation?

FYI... I’m not marketing. I’m a technologist and don’t get to participate in the forum too often. Please appreciate that software and instrumentation are very important components of the polygraph and this is my area of expertise. I’ve always been committed to adding value to this profession and I’ll continue to propel the message that examiners need to look closely at the safety, efficacy, and quality of their polygraph software and instrumentation.

Sincerely,

posted 05-18-2012 01:15 PM            

When I posted that back several years ago, I think I found the problem on charts created on a Lafayette LX3000. I'm not sure I did on the LX4000, but I could have. I understand there was a fix since then. Any time you filter one way and then another, you're bound to get a conflict now and then. We see the same thing with algorithms. Usually, they agree, but sometimes they don't. We expect that given problems with any measurements....

One of my complaints was a plunging EDA with many examinees on the Lafayette, something I haven't experienced with a Stoelting or a Limestone. (Sorry Axciton users; I've never used one of those.)

posted 05-18-2012 09:58 PM            

Good point. This is a great forum for healthy discussion about important topics (and the occasional distracting sidebar into motorcycle or other topics).

Accurate discussion is excellent and makes us smarter and better to make informed decisions.

One source of information will be experiences from other professionals in the field. Experience is very useful to other professionals who can learn from it. Additionally, instrument developers and vendors need feedback in order to make the instrument better - unless we think that everything is already as perfect or as good as it can ever get.

Another source of information will be the instrument developers and vendors. The dark possibility here is that vendors give only flattering information about the products and leave out everything else (not saying that actually occurs, i'm just sayin'...). the other dark possibility is that vendors might devote a lot of effort to providing a long list unflattering information about the products of competitors. (Again, not saying this actually happens, just sayin'...).

The fallout within the profession is a reduced ability for field examiners and program managers to make good informed decisions.

In the end it damages the credibility of the profession because it can start to look like nothing more than a market turf war instead of providing information. This kind of damage to the profession is very difficult to undo.

What professionals outside the polygraph world sometimes observe is that polygraph professionals are sometimes extremely quick to indict each other and each other's stuff as "invalid" or incorrect. The impression this gives is that the polygraph is so flakey that nothing is correct, nothing is reliable, and that everything anyone does is criticizable as somehow wrong. In reality polygraph is science and works far more often than not, because it is about some simple forms of physiological data.

The real challenge then is this: how to prevent the discussion from degrading into negative marketing and turf wars that are intended not so much to educate our profession but directed only to increase market penetration.

My solution is not to censor anyone but to ask for restraint and intelligence, and some form of ethical decorum in advertising, marketing and discussion.

I'm not suggesting rules or enforcement. I'm suggesting that we talk about it. With time, and with discussion, the right course becomes clear and almost second nature to the profession. I'm suggesting that we take the high road and emphasize the strengths of our products and services.

I'm also suggesting that everyone should learn to ask smart questions, listen to what is said by whom, and also listen carefully to what is not said.

Of course, if things get ugly, someone should politely remind us to "buy an ad."

That said, I'll work on getting information to this forum.

The most important thing is, in my view, that we plan for a successful future for the profession - and that will mean that we avoid getting mired in discussions that aggravate the inevitable tension that exists between competitors in the marketplace. That will only damage relationships, alienate individuals, and polarize the profession around instrument vendor loyalties.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-19-2012 09:04 AM         

Naturally, a polygraph newbie is ga-ga over the whiz-bang computerized instrument's capabilities over those of inked systems -- we learned on analogs at Backster back then. But over the past 8 years I have often been very frustrated (in test) with my LX4000's sudden DAS disconnects, "sample size error" messages, "corrupt file" messages (wouldn't THAT be great in court) and, of course, the dreaded plunging EDAs. Post test, when in chart review mode, the chart magnifications jumped around willy-nilly. Maddening.

There have been many in-test moments that I wanted to abort the test and fling the whole LX4000 rig into the dumpster. But that's not an option. You have to persevere poker-faced and not let the examinee know your "instrument" is giving you fits.

I come from a background in high-tech industry and well know that hardware components can have glitches and software programs can have bugs. So I just endured.

A problem is that we become heavily inve$ted in our systems. I have a lot of bells and whistles on my LX4000: all the pads, PLE, etc. It really add$ up.

In high-tech parlance, my Lafayette has become a "legacy" system. (Legacy means that you're more or less stuck with it.) Changing to a different manufacturer would be very costly and require a learning curve.

Also, Lafayette is "the devil that I know."

Let me point out, that, over the years, everyone at Lafayette has been great, and the latest s/w for the LX4000 seems to be the best yet, BUT...would I buy another Lafayette? I'm not so sure.

I plan on upgrading next year. You can bet that I'll be giving ALL of the instruments a very close look.

This topic (instrumentation) has been conspicuous by its absence and is long overdue.

Dan

posted 05-24-2012 06:14 PM            

I am sure that qualified electronic engineers exist within the DOD whether at Defense Advanced Research Projects Agency (DARPA) or through The Military Standard (MIL –STD 202) “DEPARTMENT OF DEFENSE TEST METHOD FOR ELECTRONIC AND ELECTRICAL COMPONENT PARTS” (Defense Standardization Program (DSP).) Certainly, qualified personnel must exist within the DOD to create such standards in concert with the scientific staff at NCCA.

posted 05-26-2012 01:23 PM            

Ted:

quote:

---

We should eliminate the personal, or personnel part of this conversation and keep it technical. As you can see, the manufacturers really do listen. Our input will only help them to improve their product which is a win-win no matter what brand is sitting in your lab.

---

I have an old Axciton also. And I have kept an old DOS computer, with a DB9 connector, that I could run it on. At this point the 4 channel instrument is not compliant with 2012 requirements for activity sensors (activity - not movement - because the goal of cms is to engage in data-changing physical activity without engaging in visible movement).

Regardless, keeping the discussion focused on the facts and details will help a lot.

What will not be fun, and could be confusing and damaging to the profession, is a lot of arm-wrestling and personal
struggles.

Skip:

quote:

---

It would appear the National Center for Credibility Assessment ( NCCA) should develop minimum electronic standards for the collection and display of EDA as well as respiration and cardiovascular change. The latter two being simple movement sensors would require very little in the way of minimum standards.

---

Minimum standards seem like a good idea.

At present we do have ASTM standards, though I doubt they are adequate.

I am not sure that more government regulation is actually the best long term solution.

Before anything is standardized it would be a really great idea to make sure that polygraph professionals are provided with more accurate information about the issues inherent to data acquisition, signal processing and storage and security for PDD data.

EDA circuits are not actually complicated. Lykken and Venables showed us a long time ago how to make it work. The more complicated issues - and differences - will involve the signal processing issues when the analog electronic circuit data is turned into a digital signal.

quote:

---

At a minimum, it appears that to insure no loss of usable data, one should display both the raw and filtered data so the examiner can see if differences exist. Unfortunately, if the signal is faulty at collection and display, then dual display of faulty data is not the answer.

---

This is exactly why it is important for the profession to have access to good information regarding the issues and solutions surrounding the signal processing of EDA and other data. Most importantly, let's not try to pretend that digitized data is somehow not subject to signal processing. All digitized data are subject to signal processing - or it would never get to your computer screen. What the profession needs is accurate conceptual information and accurate details about how the data are handled.

At present there seems to be some mythological perceptions or misunderstandings about the facts and real issues.

For example:

quote:

---

At a minimum, ... one should display both the raw and filtered data so the examiner can see if differences exist.

---

Of course there will be some differences between the raw and filtered data. You cannot filter something and leave it unchanged at the same time. Mostly they will be in agreement. Sometimes they will differ. Whicgh one is correct is a matter for research not opinionizing. Our misconception that raw and automatic EDA data should never be different is based only on the fact that we have not historically been able to see both as you suggest. The real question is this: do we want to know?

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-27-2012 10:12 AM         

You said:

>>>What the profession needs is accurate conceptual information and accurate details about how the data are handled.<<<<

This raises an interesting question...

You have long advocated full disclosure of how scoring algorithms work (Polyscore vs. OSS3, for example). Are you advocating for a similarly transparent approach -- for software and hardware alike -- across the entire polygraph signal-acquisition-and-processing spectrum?

If so, how do you propose your call for transparency be reconciled with the realities of commercial enterprise's quest for developing the better mousetrap via their proprietary designs? How would manufacturers protect their hard-earned intellectual property -- and the potential financial rewards such R&D toil can bring -- from being ripped off? (Are patents the answer, or is it a lot more complicated than that? I don't know.)

From the end-user perspective, it seems we've been so fixated on techniques, that the h/w and s/w elements of the polygraph equation have escaped similar scrutiny. The examiner in the trenches takes for granted that his instrument "works", provided it passes muster with the obligatory functionality exercise.

Perhaps there's a lot more to it than that.

Maybe "transparency" makes sense. Otherwise, how do we know what it is we're looking at on the screen?

This is a thought-provoking and necessary conversation.

posted 05-27-2012 09:13 PM            

Nobody disputes the importance and value of commercial interests and intellectual property.

Money pays the 'lectric bill, makes the world go round, and can be converted into useful things like coffee, shoes, and shiney objects.

Commercial interests, and the ability to make money, are the incentive for work including development, and advancement that our profession needs.

There is, however, a balance to achieve between the interests of the profession and community/nation and commercial interests. Commercial interests are sometimes clear and simple - sell stuff, make money, control as much market share as possible. Community and national interests are also sometimes clear and simple (in concept) - safety and integrity. We, as a polygraph profession, have to be clear about which priorities are our highest priorities.

What is complicated is how exactly to use the results of a test that is known to be good but still imperfect. Effective and ethical decisions surrounding both the use of the polygraph and polygraph field practices are going to be made when good information is available, and when the scientific and ethical questions of critics and opponents can be answered and put to rest. This will require a certain degree of information to accomplish.

There are well established mechanisms to manage these complications: patents, IP laws, non-disclosure agreements. All of these allow people to protect their interests while achieving a needed degree of accountability.

There is nothing wrong with the developers of any algorithm choosing not to disclose the details. The result of not disclosing and not publishing is a static condition surrounding our profession's knowledge base. Some researchers did publish information, and it is immensely beneficial to us - when we pay attention. When we ignore knowledge it does nothing for us.

What is wrong is using an algorithm to make or influence decisions that affect the lives and futures of individuals and communities when we do not know the details. Even the NAS in the 2003 NRC report could not make much sense out of some algorithms.

Another problem involves the profession - that may be at risk for mythological expectations for perfection (not possible) when they are told to remain uninterested in the details. Related to this is the risk for misapplication of an event-specific decision model to multi-issue exams - because we are not well educated about the issues surrounding statistical decision models and decision theory. An even larger problem involves the confusion and chaos the results when different algorithms produce different results (which they occasionally do). It should not be a great surprise that different statistical models can produce occasionally or slightly different results. But it bothers our profession a lot when we have been encouraged to engage in black and while (childlike and concrete) thinking that says it should be simple and the results should always be the same. Then we engage in fictionalized opinions about the differences, and the arguments get slanted by egos, politics, and economic interests instead of facts. Because we have such a low level of information in the profession: we are at risk for algorithms and methods that make faulty, erroneous, and irresponsible application of statistical principles. Then scientific minded observers (critics) outside the polygraph profession notice quite clearly that we don't know what we are talking about.

We are the only profession in existence that is afraid to use computers to score and analyze our data. Our discussions about test validity have sometimes tended not to sound much at all like discussions about test validity. These discussions have sometimes sounded more like discussions about resumes and CVs - the practicioners and experts - instead of discussing the data.

The solution: openness. There is nothing to be gained by extinguishing private or proprietary interests. That is not the goal. Our solution was to develop a new and opensource algorithm (OSS-3) that is unencumbered by proprietary economic interests. There is then no reason that anyone cannot be given information on how the algorithm works and what it does. When we do observe the occasional differences between the algorithm and manual scoring we can determine the cause.

Thing about knowledge (real knowledge, not just memorized parrot talk) is that it is supposed to be auto-catalytic. That is, knowledge leads to more knowledge. Though we will never know everything, we can try to keep learning. One difference between pseudoscience and real science is that real science recognizes that all theories are ultimately incomplete and therefore wrong - and the thing to do is keep learning and keep improving our understanding and knowledge of the issues. Improvement = knowledge improvements that contribute to improved accuracy effectiveness of the test results or improved understanding of when and why it works and when and why it sometimes doesn't. Pseudoscience is characterized by knowledge that is both out of sync with (not catyzed by and does not catalyze) related professions. Knowledge is pseudoscience is static - and tends to come only from the founding source.

At present our profession is so far in the dark that we would impeach the validity of a perfectly good exam simply because someone did not complete the functionality exercise. This seems to imply that there is something very complicated and special and unknown that could affect the very outcome of the test if this exercise is not completed. This is a PERFECT example of why we need to get more information to the professionals in the field regarding what the instrument does.

In fact, a serious system problem would most likely be detected without a functionality exercise.

In fact, a minor sensor problem might not invalidate the test result - in a scientific sense.

In fact, system problems are random events that are just as likely to occur after the functionality exercise as before.

Although it may be important, at times, to show and verify that the instrument is functioning properly (and dispel baseless concerns) - it is not helpful to us to make polygraph professionals vulnerable over non-issues that do not actually affect the validity of the test results.

Similar mythological assumptions seem to surround the engineering issues of EDA. For example - look in the psychophysiological literature and you will find that descending raw/unfiltered EDA data is a known phenomena for some persons. This is the reason that engineers long ago began to add the option of automatic filters to the EDA channel.

We will have increasing legislative challenges as the purveyors and developers of voice stress technologies continue to try to achieve legislation that will allow them to participate and compete in the polygraph arena. It will do us no good to make ourselves unnecessarily vulnerable, and it will do us no good to not understand the science and technology behind what continues to be the best and only valid scientific method for instrumental detection of deception and credibility assessment.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-28-2012 12:19 PM            

Lykken and Venables advocated the use of conductance as the method of choice for measuring tonic skin conductance using a constant-voltage method. In the 1970s they tried to convince other scientists to accept conductance, constant-voltage, as the standardized method. Can you tell us when Lafayette began using conductance? Did Lafayette immediately take advantage of this academic knowledge?

You say that measuring EDA is not complicated and I respectfully believe that you're underestimating the complexity of the task. Does your employer, Lafayette, also believe that measuring EDA is easy? The LX4000 image that's disseminated at every opportunity shows that Lafayette provides a feature enhancement that completely replaces the LX4000 EDA . When Lafayette bypasses the original EDA and inserts a new EDA circuit doesn't that indicate a deficiency or problem? This wasn't an EPROM flash, or an added jumper. It wasn't some small change to a resistor or capacitor. The feature enhancement provided is a completely new EDA daughter board. Let's not get too concerned that the new board is floating loosely inside the LX4000 enclosure. Let's just focus on what necessitated the complete change. Can you provide any relevant information about this LX4000 feature enhancement?

Several days ago you said that you work on getting information to this forum and I'm eager to learn more about the feature enhancement that's displayed in the image that examiner, Dave Bethune, has posted. Lafayette promises lifetime software updates and support. I own several LX4000 polygraphs and I would sincerely appreciate an answer. I don’t think I’m the only person that genuinely cares about instrumentation problems that might result in opposite EDA results. I’m just saying...

posted 05-28-2012 03:31 PM            

1) As before, I am not an engineer, and cannot provide much detail on the circuit change, and 2) I don't think it would be a good precedent if I were to agree to make myself accountable to your questions in this manner. The continued suggestion that I explain this is starting to feel like mere provocation and marketing. That's not fun.

Nevertheless the topic is important, and it is appropriate to have conversations about general issues.

When I say that measuring EDA is not extremely complicated it is all relative...

EDA is itself a very complex phenomena - such that entire academic text books and chapters have been written on the subject (Bouscein 2011; Dawson et al 2000 in Cacioppo et al) and others (Prokasy & Raskin, 1973). For example, we have always tended to say that EDA is a function of sweat activity, but it has been known from Darow (1927) and others who studied EDA and sweat secretion, that phasic EDRs tend to occur about 1 second before sweat appears on the skin. Therefore EDA, as we know it, seems to have more to with activity of the sweat glands and enervating neurons than actual sweat. This is the reason that our EDA circuits work when the medium of conduction between the electrode and skin is a chemical paste and not actually sweat. If we read carefully we will notice that a descending EDA pattern is a well known phenomena for some persons.

Lykken's reasons for recommending conductance instead of resistance is not completely convincing to everyone. Read Boucsein, who criticizes some of Lykken assumptions.

Lykken's basic reason, according to Dawson et al, have to do with the fact that conductance is more linearly related to the number of active sweat pores which act like resistors in parallel. Resistance, as I'm sure you known, is a more complex function (reciprocal of the sum of the reciprocals). Ultimately it does not matter to a computer how complex the math is - computers will do it fast and correctly every time.

Because resistance and conductance are mathematically related one can be calculated from the other. Ohms law says that R = V/C where R = resistance, V = voltage and C = current. Therefore, if the current is held constant then resistance can be calculated by measuring the voltage change across the two electrodes. If the voltage is held constant then current can be measured. Remember that Boucsein questioned the assumptions of the parallel resistance/conductance assumptions pertaining to the bioelectric action of the sweat pores as uncertain. Remember that EDA is actually complicated

Of course there are a lot of engineering decisions and signal processing details that go into the design of a system that runs on USB power (5v, that is also used to power the rest of the device), conversion of the signal to digital, processing of the signal for display, incorporation of user convenience features such as sensitivity and centering adjustments, and everything else that goes into making the data acquisition device and the software work.

The information and schematics published online have sufficient detail that even I - a non-engineer - could probably go to Radio Shack, or order parts online, and figure out how to build an EDA circuit. It's not rocket science. Designing polygraph instruments for field examiners does not seem to involve the actual development of new types of EDA circuits.

Of course engineering and programming expertise are necessary to put it together and make all the data acquisition sensors, signal processing, and analysis tools work as a complete system.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-28-2012 04:46 PM         

Is the EDA mod in question something those of us with older LX4000s should get? What are we missing without it? Is there a charge for the upgrade?

posted 05-29-2012 07:26 AM            

I wrote you a response last night, and then accidentally deleted it before posting.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-29-2012 12:29 PM            

The medical profession clearly lays out standards such as the following examples:

"To reduce artifactual distortion of the ST segment, the 1990 AHA document recommended that the low-frequency cutoff be 0.05 Hz for routine filters but that this requirement could be relaxed to 0.67 Hz or below for linear digital filters with zero phase distortion.23 The ANSI/AAMI recommendations of 1991, affirmed in 2001, endorsed these relaxed limits for low-frequency cutoff for standard 12-lead ECGs, subject to maximum allowable errors for individual determinants of overall input signal reproduction.24 These standards continue to be recommended."

and this

The ANSI/AAMI standard of 1991, reaffirmed in 2001, recommended a high-frequency cutoff of at least 150 Hz for all standard 12-lead ECGs.24 The ANSI/AAMI document also details maximum allowable errors for individual determinants of overall input signal reproduction, which extend beyond the scope of the present report but are important guidelines for manufacturers.24 These most recent limits continue to be recommended for adolescents and for adults, with extension of the high-frequency cutoff to 250 Hz in children,35 subject to demonstration of fidelity testing by individual manufacturers according to standard methods.23 Electrocardiographs should automatically alert the user when a suboptimal high-frequency cutoff, such as 40 Hz, is used, and a proper high-frequency cutoff should automatically be restored between routine standard ECG recordings.

These appear to be something that the industry to which the industry must adhere if they want to sell digital EKG. This comes from a large statement and is shown here for exaqmple only as what I mean by NCCA publishing minimum standards to data aquisition.

posted 05-29-2012 02:27 PM            

quote:

---

3. By the 2013 APA Annual Seminar the Research and Development Committee Chair shall revisit and propose minimum standards for polygraph instrumentation for a Board vote. The proposal shall include recommendations for new sensors and data channels

---

posted 05-29-2012 05:22 PM            

Those are some rather specific recommendations for filtering data. It is interesting because they seem to recognize that signal processing is a reality - because it improves the usability of the data.

They are using both low-pass and high-pass filters. They have also thought about phase distortion when they do this. They want zero phase distortion, which means that their DSP filters have to be designed to achieve this. Phase distortion refers to subtle shifts in the time scale at which events are shown to occur. They result from some types of filtering. There are ways of correcting for phase distortion, and some types of filters are less prone to this problem.

In the other thread I posted four images of EDA data. The quality of the data differs substantially - indicating very different ways of processing the signal.

Perhaps it would be a good idea to have more clear standard guidelines for signal processing.

One complication is that we don't yet know which solution actually gives the most accurate results.

Some people are bothered by any signal processing, and advocate for manual/raw EDA data. How sure are we that manual/raw EDA data is actually raw data with no filtering. I dunno. One clue is that unfiltered EDA data will be a bit uglier (high frequency artifacts are not removed). Another clue is that manual/raw EDA data is known to exhibit a descending pattern for some individuals. A more rare phenomena is the pattern of persistently ascending EDA data.

Difficulties and dissatisfaction with manual EDA data is what lead to the addition of automatic EDA modes. Automatic means "filtered" using high-pass, low-pass or a combination of both. In the past this was done with old fashioned resistors and capacitor. My first polygraph instrument (computerized) even had a mechanical switch on the device to change EDA modes (that is a clue that filtering is done with hardware components and now software).

Today, signal processing is more likely to be accomplished with digital filters - which can be designed to achieve the same goals as analog filters, but also offer some additional capabilities. For example: a requirement for zero phase distortion may impose a requirement for digital filtering instead of analog, and may mean the difference between use of finite impulse response (FIR) filters and infinite impulse response (IIR) filters. FIR filters operate on the data for a fixed time period with no feedback. IIR filters also operate for a fixed period of time period but feedback some of the processed signal to each new segment of data. In this way, IIR filters continue to carry forward some small aspects of data that was processed/displayed since the onset of data collection.

The information that you provided suggests that EKG technicians have some control over the selection of filter cutoffs - and this is evidently audited so they will know how the signal was processed.

Do we know enough about how polygraph EDA data is processed? My answer is "no." I believe that examiners in the field should know more. Knowledge will dispell mythology and fictional expectations about what EDA data should do or should look like.

Bottom line is that strong opinions are not scientific evidence, and we do not actually know what is the best solution. Therefore, I will advocate for making the options available: raw/unprocessed EDA data, automatic EDA data, and other solutions as they are developed.

The alternative is to lock ourselves into simplistic thinking, and avoid dealing with the fact that EDA data can be processed in different ways that may occasionally lead to different scores. If we are childlike and want to pretend/assume perfection, then we will be opposed to knowing the facts, and will experience a lot of anxiety over normal differences that our childlike thinking tells us should not occur. Differences do exist. If we have the conversations and study the issues we will eventually understand the optimal solution that gives the correct answer for most people most of the time.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-30-2012 04:09 PM            

I would think that a series of test charts of digital data could be created/constructed by someone. This test series could have definitive changes at specific intervals at varying resistance/conductance changes. Such data could then be "fed" to different instruments or circuits to see how that instrument displays and differentiates that data. Certainly electrical interference and other common artifacts could be introduced at given intervals to see how a circuit handles the data. If a low or high pass filter drops or distorts "data" in its quest to remove the interference/artifact then it would not pass the test for use.

I understand that my meager attempt to explain myself is probably not helping my argument, but I still think that a component upon which we place so much emphasis should have sound technical underpinnings and not be open to speculation as to its accuracy.

My truck as a tire pressure system that alerts me to low tire pressure by displaying a warning light on the dash. It doesn't tell me which tire or tell me how low or what the pressure is in that low tire or what the pressure is in the other three. It sucks for that reason. My wife's car displays the individual PSI pressures of the four tires and alerts the driver to the specific tire that is low and what the pressure is in that tire.

It's not rocket science but its pretty damn impressive. I can put an analog pressure gauge on any one of those tires and check to see if the digital sensors are accurate, assuming my Wal-Mart tire gauge from China is accurate.

Finally that sensor inside each tire is spinning and bumping along the highway at
812 rpm at 60 mph (15 inch tire)over potholes and road kill and it still works. I doubt very seriously that it's taped inside the tire or attached with two thin wires like an afterthought.

I opened up a motion sensor pad a few years ago because it quit working. I assumed it was probably a broken wire and I was going to reconnect it. What I found was strips of silver colored duct tape with a dozen piezoelectric sensors stuck to it and thin wires connecting them in series to a Radio Shack plug with the worst soldering job I had ever seen! It looked like a third grade science project that didn't win honorable mention.

posted 05-30-2012 04:52 PM            

“The cheap electronic toys I buy for my grandchildren that are made in China have cleaner and more professional circuits in them.”

I agree with that statement and you certainly didn’t need a PhD in Electronics Engineering to make that observation.

Dr. Ray,

Sorry I’m asking such provocative questions. You’re a little thin skinned for a scientist.

posted 05-31-2012 11:29 AM            

I think you are adding a little bit of drama here.

I don't see any electrical tape in David's picture of the LX4000.

As for cardboard... it is my understanding that electrical insulation papers are well known engineering products for mechanical and electrical insulation.
http://en.wikipedia.org/wiki/Electrical_insulation_paper
http://en.wikipedia.org/wiki/Fish_paper
http://en.wikipedia.org/wiki/Vulcanized_fibre

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-31-2012 11:55 AM            

Ray,

Look closer... the electricians tape is wrapped around the .10 cent RJ45 communication jacks.

posted 05-31-2012 12:34 PM            

I've been re-wiring my old house. Most of the wiring is 90 years old. Some modern work was done, a few decades ago. When I open the receptacles to pull new wires, the old wires are screwed to the terminals and the devices are wrapped in tape.

Removing 90 year old post-and-tube wires is interesting too. Electrical taped splices splices everywhere in my basement, walls and attic.

To me it looks odd. But whomever wired my house and made those splices actually did a good job. The wires have been providing juice without fail for over 90 years.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)