Selected reader comments on "Identifying terrorists before they strike"

See Identifying terrorists before they strike

Dear Steve:

Your website article on 'Identifying terrorists before they strike' should be read by all members of the executive branch of governments and the INS along with the airline industry. This is the epitome of an ounce of prevention ! I applaud your insights and expect to see this propelled into the media asap. I'm doing my part right now.

My brothers in law are both in law enforcement and I've passed this info on to them and hopefully nothing even remotely like 9-11 will ever happen anywhere in the world again.

Gene Koch

Dear sir,

I very much enjoyed your article on brainwave fingerprinting. I recently posted a response which I include below. I have long wondered why more attention was not being paid to developments in brain computer interface systems (BCIs), a supercategory of brainwave fingerprinting. Now that it's getting attention, I want in. I am writing because I want to get involved in the commentary process. I am not quite as good as Larry Farwell, but you'll have a hard time finding someone who knows more about BCIs. I realize you aren't the main contact person, but you would know who to contact. You seem like a bright guy and a good writer. Perhaps we could work together. I am not looking for a full time job by any means, as I am busy with my PhD, but given the volume of responses, it is clear there is a demand for someone such as me with a solid knowledge of where BCIs are and where they are going. Thanks, -Brendan Allison

I am a graduate student at UC San Diego, and the work I am doing for my doctoral thesis involves the same brainwaves as those used by Dr. Farwell. I thus feel that I can offer a different perspective than many readers, as I have thought about the ethical and practical ramifications of this technology for many years. I am delighted to see this much attention being paid to this technology, but am concerned about a few glitches which I would like to correct. Before I continue, I wish to emphasize that, while I have been in contact with Dr. Farwell for a couple years and have great respect for his work, I do not work with him and he may or may not agree with all of my comments below.

First and foremost, the "brainwave fingerprinting approach" is not a universal lie detector. It works by determining whether or not a stimulus (such as a picture) has been previously seen by a subject. It could be very useful in determining whether someone is familiar with the inside of a terrorist training camp or cockpit, but you CANNOT simply ask any yes/no question and determine whether someone is lying.

Second, it is true that Dr. Farwell's system has been 100% accurate in numerous tests. However, this system would necessarily not work on 100% of the population. It is dependent on the P300 complex, a very well studied series of brainwave components. It is known that a few people do not exhibit a normal P300; individuals with brain damage, memory or attention deficits would exhibit abnormal P300s and thus the "brainwave fingerprinting" approach may not be able to be used with them. These people would not get false positives, they would just not be effectively screened by the system (so a backup security procedure would be necessary). As research into the brain continues and we learn more about the EEGs of these unique subject populations, the range of people who would not be good candidates for this system will diminish.

SK: Yes, I totally agree with the above two points. I've tried to communicate this accurately in my writeup.

Mr. Kirsch's article states: 

It's less intrusive because you don't have to answer any security questions. You just put on a cap and watch TV for 10 minutes. The profiling is just a yes/no profile of certain knowledge you have. It is not a psychological profile and the data gathered cannot be used for psychological profiling (see more information below)

he later states: 

If you don't want to answer the question, you can simply press a button or remove the headband at any time you see a question you judge is too personal. So there is no risk that this data can be used for psychological profiling.

It is true that someone who did not want to be screened by this system could simply close his or her eyes, remove the headband, etc. However, in the scenario described by Mr. Kirsch, great care would need to be taken in order to ensure that the information remained anonymous; otherwise it could indeed be used for psychological profiling. As this technology advances, it is likely that the "10 minute" estimate Mr. Kirsch provides could be reduced considerably.

SK: Selection of test material could be done in a public process or approved by a public review board. The fact you can take the test anonymously really removes the incentive to do psych profiling and use that data against specific individuals.

I also wish to comment on a later part of his essay, in which he states:

We do not (and in fact cannot) ask open-ended questions such as "what do you know about x?" We cannot determine how you "feel" about things. We cannot determining what you are planning to do in the future. We cannot tell whether you were speeding yesterday. We cannot tell whether you did drugs yesterday or while you were in high school We cannot "ask" you any questions that you do not want to answer.

Mr. Kirsch is correct in all of these except one. You can, in fact, gain some information about both a subject's current state of intoxication and overall substance use pattern. Marijuana, alcohol, caffeine, nicotine, and other drugs have been conclusively and repeatedly shown to affect the P300 complex. Furthermore, the chronic use of at least alcohol and marijuana is apparent in the P300 even if the subject is not intoxicated during the test. Finally, individuals at risk for alcoholism show differences in the P300, even if they have not tried alcohol. For anyone interested in this topic, Dr. John Polich has an excellent review of the P300 and clinical factors which affect it:

SK: Farwell's computerized test equipment does not analyze these patterns nor report on them. The system could be modified to do this if desired. However, that's not the objective of the screening and if it were used in this way, there would be a public outcry, in much the same way there is a public outcry if the police or government overstep their authority.

Polich, J. P300 clinical utility and control of variability. Journal of Clinical Neurophysiology 1998 Jan, 15(1):14-33.

In fact, Dr. Polich has many articles on clinical factors which can affect the P300. Just do a search through any good library for articles he published and you will find them.

One of the responses to this article, from someone only identified as "xxx xxx" and Posted At: 9:45 GMT 10/02/2001, also demands a reply. Here is what this individual states:

This is more of the same. It's no better than finding criminal behavior by counting bumps on someone's head. An intelligent terrorist (and most will be bright) can evade such precautions by avoiding being nervous (or being consistently nervous).

It is unfortunate that intelligent debate on such an important technology is clouded by idiots like this who clearly don't take the least time to learn about a topic before spewing their inane, knee jerk reaction to it. The approach of finding criminal behavior by reading bumps on one's head (called phrenology) was dead before the turn of the 20th century. It was a fad which did not stand up to even casual scientific testing. The fact that the P300 reflects memory, and thus can be used to determine whether someone previously saw an item, has been known for over 20 years and has been repeatedly verified in countless professional, peer reviewed journals. I have data from my work which shows this effect and stake my reputation on it, as hundreds of my colleagues have already done.

As to the notion that someone could fool the system by being nervous, or not being nervous: this system is very different from conventional polygraphy. It is not based on classic physiological signs of nervousness such as increased heartbeat or sweating. Dr. Farwell published an article in 1991 with Dr. Donchin which very clearly showed that his approach works even if subjects are trying to fool his system. Furthermore, had xxx xxx bothered to read Dr. Farwell's article which he published with the FBI, s/he would know that the system was tested extensively on FBI agents who were themselves trying to fool the system and failed.

In summary, Dr. Farwell has made tremendous progress toward a completely effective, non-intrusive lie detection system which could work on anyone. There are a few issues that need to be further researched before it should be relied upon as a sole means of detecting deception, but it is currently ready as a supplementary approach. Furthermore, the widespread application of such a system and related technologies do present ethical issues. As with any new technology, we as a society must decide through informed discussion and debate whether the benefits outweigh the costs. I urge those who are interested in this technology or concerned about its application to learn more about it through the links provided in Mr. Kirsch's excellent article before commenting on it.

I would love to talk more about this issue, but (ironically) have to attend a meeting at my lab to look over some of my P300 data. Thanks to Mr. Kirsch for drawing attention to this important topic.

Brendan Allison, MS, PhD student Cognitive Neuroscience Laboratory Dept. of Cognitive Science University of California, San Diego

Dear Steve,

This is an interesting idea. I am curious if the machines during profiling, could determine the difference between having read material on a given subject, and having actually performed the task. For instance, in the wake of the WTC tragedy, many of us have immersed ourselves with information ranging from self defense, to the various chemical agents that could be used in a bio attack. Many have felt threatened by unknown future attacks, and have perhaps bought a gun, gone to a shooting range, read articles on how to create simple weapons out of common household things, the list could go on and on. How would this computer screening know the difference?

Thanks for addressing this concern,

Terrie

SK: If you have a gun, you'll recognize the type you have. But you probably won't recognize an AK47. You won't recognize the code book for al qaeda. You won't recognize the al quada world HQ building. so we test you on stuff you aren't going to be familiar with.

Response to Terrie's e-mail from Brendan Allison:

Hello,

Steve is correct. The brainwaves involved (called the P300 complex) reflect your familiarity with specific objects more than general categories. Another aspect of them which has not been as widely presented is that they also reflect the object's importance to you. (The jargon used among scientists is that they index stimulus novelty and saliency.)

Thus, if you have a pistol or hunting rifle at home, and are shown a picture of an AK-47, you may get a slightly different response than someone who has never seen any gun, but you will certainly not get the same response as an individual who is very familiar with that weapon. True, someone who is a big fan of AK-47s and looks at pictures of them every day but has never actually used one may produce a false positive in the system. Here is how you could separate those people from individuals who are trained with that weapon. In addition to showing a picture of an AK-47 fully assembled, you could also show it in various states of disassembly which are meaningless to a casual observer. For example, you could show the weapon which has been disassembled for cleaning, or show it being unjammed. You can then mix in some bogus pictures of the AK in these two states, such as a disassembled weapon with some pieces missing, or extra parts, or the weapon in some weird configuration. Only someone who knows the weapon and how to use it would show a distinct brainwave response to the correct picture of it being cleaned or unjammed.

Similarly, while many people have seen pictures of shooting ranges, few would legitimately be familiar with the shooting range at a terrorist training camp. Many people have heard of concerns with crop dusters, and have seen pictures of them, but fewer people would recognize the inside of one or pages from the inside of a training manual. I would assume that the pictures which people would see would reflect a variety of different images only a terrorist would consider familiar. It's true that there are people who are experts with AK-47s and are not terrorists, but such people (if innocent) would not also recognize the inside of a training camp as well as pictures of senior members of a terrorist network as well as pictures of the Khyber Pass taken from specific locations inside Afghanistan held by the Taliban.

The other important issue is that of stimulus saliency- that is, how important something is to you personally. Even if two people are equally familiar with an AK-47, the P300 complex looks different for people who consider it very important to them. This could be used to discriminate fans of action movies from those who consider an AK a critical component of their lives.

In any case, I do not think that brainwave fingerprinting is ready to be deployed on a national or global scale a the sole means of security screening. It is developed enough to be an important component of a security package, and, with some more research, it could be improved to the point that it is reliable enough to stand on its own.
SK: That is why my approach allows for manual appeal of the security ranking to account for unusual cases. It is the false positive that is the issue, not the false negative. These false positives are not P300 errors, but basically poor "test materials." As Brendan points out, with good materials and a bit more research to further refine the technique (see next paragraph), we'll have fewer and fewer false positives over time. However, the vast majority of the population can be handled automatically with today's technology at considerable savings in cost and consumer convenience.

That's my answer for Teri's question, but I'd like to throw out another comment about the brain and memory which could help make the system much more efficient. We all have many different memory systems which use different areas of the brain. For example, we know that declarative memory- memory for objects, for example- is different from procedural memory- memory of how to do a task. The P300 complex reflects declarative memory. A brainwave fingerprinting system could be enhanced by utilizing multiple memory systems. People who have not only seen an AK but have also used it extensively will not only remember what it looks like, they will remember what it feels like and what their fingers have to do while using it. Thus, such people would show different activity in their primary motor cortex and primary somatosensory (touch) cortex. This would beautifully distinguish those who see pictures of an AK from those who have been trained with it.

Yet another form of memory is priming. If I say "king" to you, your brain shows a certain response. However, if I say "king" to you after saying a related word, such as "queen" or "throne" or "crown," you will show a different response. This is because your language system cannot help but think of words related to a certain word, even if you aren't conscious of it. Here is how this could be incorporated into a brainwave fingerprinting type of approach. Let's say that the director of a terrorist training camp is named Bob Smith (not a viable pseudonym in Afghanistan, but let's pretend). If I show an innocent person a picture of Bob Smith, and then a picture of bin Laden, s/he would recognize only the latter. A guilty person would recognize both. So far, this has nothing to do with priming; it's using the P300 complex I mentioned above for object familiarity. However, a guilty person would also have been "primed" for the bin Laden picture, since he would know that Bob Smith was related to bin Laden. So the guilty person would be different in 2 ways. First, he would recognize Bob Smith, and second, he would recognize that Bob Smith and Osama are related. You are thus tapping into 2 different systems which could back each other up.

Dr. Farwell is certainly aware of these different memory types. His deception detection devices which he has worked on for years all have names which include something like "Multifaceted Electroencephalographic Response." I am confident that his system will only become more flexible, effective, and easy to use as research continues.

Brendan Allison BCI Project Director Cognitive Neuroscience Laboratory Dept. of Cognitve Science UC San Diego

Dear Steve, Your anchor desk references are very timely in that this whole scenario has been in the procedural development as well as both the psychology of such a procedure and the accuracy of the evidence obtained. I am a former government employee and was a member of the Special Forces in the 1960-70's and know well the validity of this method of interrogation which is a more humanistic approach than what I am seeing take place as far as security is concerned at the moment. I wrote a Department of Justice opinion on this very subject as well as a book on economics and the affect technology will play in future decisions to be made for the security and safety of America and the World as a whole. I support further implementation of this type of Neuro testing and feel like you, that now is better than later or never to begin approaching these kinds of methods to assure who we allow to share our freedoms and not necessarily that this would institute jeopardy to those who would participate in this type of testing.

If we might share any knowledge in appreciation of this cause, I would be happy to reply to any inquiries you may have.

Sincerely,

Stan Metcalf, American -Retired

Hi Steve

I came across your website on "Identifying terrorists before they strike". It is interesting concept which combines two technologies with which I am familiar. I have just finished reviewing Farwell's patents for potential application in another arena and was also instrumental with the development and fielding of the iris identification hardware at Sensar.

The company I am now working for, Sarnoff Corporation, has a world wide reputation for developing technology and generating spin off companies. I have attached a brief write up on Sarnoff, our business unit, and our Design Review Process for your review.

I'd like to discuss my interests and Sarnoff's capabilities further with you.

Best regards,

Dr. Tom Chmielewski

Steve, I'm the COO of a biometric security company, so I'm gung ho for the majority of your scenario, but what are the images to be shown that would trigger positive brainwaves? As I understand it, there need be images to determine culpability. You can't just ask questions like "would you like to see Israel perish in a ball of fire?" and be sure of the result. I tried to read the document but got no sense of the imagery involved.

SK: Words and pictures, e.g., a picture of the inside of the al Queda corporate terrorist training HQ would only be known by a member of al Queda.

Eric Rugart COO OpenSociety Technologies 919 Conestoga Road, Suite 3-210 Bryn Mawr, PA 19010 [email protected]

PS: The national picture database you describe was attempted by the CIA four years ago through state DMVs, called the true-ID initiative. It set off a huge protest and endangered our own funding. I do think times will change.

I like the idea. It has probably been pointed out that if an attempt was made to beat the system in the same manner as a Polygraph, (persuading yourself that because of a certain point of view your answers are "true"), it would need to be at such a deep level that it would virtually ammount to a "cure" - changing the basic outlook that supports the need for violence.

As a foreigner already seeking a visa for lecturing I would welcome such a test if it speeded up the process.

Laurie Penman

SK: It's much lower tech than that. It only can determine what images are familiar to you. It doesn't change your personality.

My only fear is that this information won't get to the people who make the decisions regarding appropriate anti-terrorism response systems. I wish you luck in getting this system to the demonstration phase. I for one would gladly submit to screening, as your proposed system appears to offer all the benefits without unacceptable costs and a quantum leap forward in accuracy to boot. Even the proposed implementation plan sounds workable to me, as it targets both the highest risk and most likely to submit population first (those entering or returning to the US), while avoiding those most likely to resist (domestic travellers). Your idea of giving people who do not wish to be automatically screened the option to go the manual route would surely have the desired effect of encouraging automatic screening over time as public awareness of the benefits increases and the desire to avoid long lines overcomes resistance. Good luck.

Thank you for the thoughtfulness and constructive nature of this article. As an attorney, I am very impressed with the concept and analysis. Imagine how fortunate we are to live in a country where the freedom to explore and expound upon ideas that can allow us all to advance exists.

Thanks again.

Jim Towne

Mr. Kirsch-

I've read your article and find the conclusions as to the efficacy of such an approach indisputable. During the 1980's while working on my PhD in Neuroscience I developed software (with supplemental hardware) that would accomplish much the same result using 16 EEG electrodes, A-D and FFT software and a GW pupilometer. I copyrighted the software under the name of ACSCAN. During the extensive testing of this device we found that every subject (among thousand of psychiatric patients) had a unique, reproducible profile.

Since that time I'm sure there have been hundreds of more sophisticated devices invented. My point is that the technology exists, is not cost-prohibitive and could be rapidly deployed with a minimum of trained personnel necessary to insure it's continued accuracy.

Just my 2 cents worth of opinion.

Michael Molyn El Dorado Hills, CA

I read your article on MSN.com and it sounds fantastic. Maybe some people will look at is the same way we look at old shaman medicines, but by the way it sounds, you're completely right. I studied Computer Science in college, but I don't have even close to the knowledge that you have. I hope your idea becomes a reality and we can counter the terrorist actions that have overcome the world with your help. To me, you'll go down in history as a radical, but at the same time, a savior. Take care.

Nick M. Widowski

Steve, you have found a truly elegant testing solution. while it is true that an individule can not fake or misrepresent their reactions to the given questions, that reaction could be generated in any number of ways. for instance you could show a picture of a terrorist training camp, this could cause recognition by the person having actually having trained there, having traveled through simular terane, having been exposed to documentary information, having simular military training or any number of other sources. how do you propose to "grade" the responses? I clame no special insite, but the popular view that the use of a plane as a bomb hadn't been considered amazes me. the jump of intelect required to go from car bomb to (insert any mode trasportation here) bomb is not that far. I agree that there would be no false reactions to the questions presented, the technology imployed precludes that, I question the ability of the inquiziters to frame those questions in a non-discriminitory manner.

SK: Vince, it's your choice...who would you rather trust? the terrorists or your elected officials?

Wouldn't "Thought Pattern Recognition" or TPR be a better name than "Brain fingerprinting?"

Mr. Kirsch -

You make some interesting arguments, and to be fair, with all the limitations you've placed on the system it might be reasonably analogous to security precautions in place now--only grossly more accurate. But you don't seem to take into account any sort of mission-creep. What if a national database is set up to keep track of our "anonymous" identifiers--how is our given name even relevant any more? And sure, you can refuse to answer any of the questions--unless they become mandatory. And sure, it'll only relate to terrorist activity--unless the scope of the system is increased to include the war on drugs, or affiliation with questionable political groups. After all, terrorists move around inside the country, and many have ties to criminal or extremist political activities, so what better way to keep americans safe then by tracking their movements and subjecting them to an annual loyalty test? What would a law-abiding citizen have to hide?

Frankly, I don't see this taking off. Certain terribly inconvenient ammendments about bearing witness against oneself aside, if a system like this took off--a perfect knowledge detector, as it were--it would inevitably become a tool for the courts. And I suspect that there are more than a few politicians and high-ranking law officers who really, really wouldn't like an unquestionably accurate measure of what, exactly, they know.

Best regards,

Nathan Winant

SK: This instrument is just an automated version of doing what El Al does. Do you think because El Al asks security questions today that that will mean that Israel has a right to ask you about your sex life tomorrow? 

Any system can be abused. You're not forced to use the system. And you're right in that if the government adopts this today, it doesn't predict what will happen in the future. I know of no piece of legislation that does this now. But that's really a completely separate issue. Perhaps you can lead the charge for a Constitutional Amendment that would prohibit our government from ever doing anything bad in the future?

I read through the stuff. Interesting. I am still not clear if it measures propensity to do a future act based on anything other than a previous act. For example, most of the terrorists who were involved in the WTC attack appear to have been sleepers, not involved in previous assaults. From his description, I am not sure if it is able to flag those people.

SK: Intent has nothing to do with what is asked by the computer. The computer tests only your familiarity with something. Do you think those terrorists would have recognized a picture of a bin Laden terrorist training camp?

Steve,

Your brainwave fingerprinting idea sounds very interesting. Here is one way that Bin Laden could get around it:

(1) Recruit a new terrorist candidate 

(2) Get him to take the brainwave test 

(3) Then send him to terrorist school to "learn the ropes"

Since he wouldn't need to be re-fingerprinted again for a few years, then he could be trained and carry out an attack with his old fingerprint in the database -- thus getting around the system.

Regards, Lex

SK: The simple counter is to do 10 minutes every 3 years, but you have to get a "quick scan" every 3 months. A quick scan takes 2 minutes and just catches the obvious cases such as this one. In general, they don't just recruit people off the street like this because if they did, it would be easy to penetrate. Apparently, most members are related by blood, so by the time you are of useful age, you're already tainted. 

Other resources
Identifying terrorists before they strike
Appendix: (costs, configuration, list of objections)
Reader comments
FAQ on brain fingerprinting (written by a PhD student at UCSD)
List of endorsers

Press articles
ZDNet article
ZDNet article#2
Response to article in The Register