Friday, February 25, 2011

Confirmation Bias and the Scientific Method

Marc Hauser, a primate psychologist at Harvard and expert on the evolution of morality, was recently found guilty of eight counts of scientific misconduct. At issue was his interpretation of video tapes of rhesus monkeys performing tasks that test their ability to learn sound patterns. Hauser "saw" the behavior that supported his hypothesis and was convinced that he was right even when other members of his research group could not. Being an eminent scientist, his group members deferred to his authority, and his interpretation prevailed in the publications that followed.

Physicist Robert Park wrote in a web column that Hauser fudged the data, implying that it was premeditated and deliberate. Scott Lilienfeld, a psychologist, feels that this might be a simple case of confirmation bias, a psychological response of the brain to reinterpret the world by distorting the data to favor the believer's expectations - a phenomena that is commonly at play in strengthening religious faith.

I just returned from a trip to Wright Patterson Air Force Base, where I gave a seminar about our work on self-healing materials and fundamental limits. My visits to the materials lab are always gratifying because the interdisciplinary team of researchers there understand our work from a broad range of angles. There are chemists who understand structural subtleties of chromophores and how they aggregate as well as the role of the host polymer on the properties of the embedded chromphores. The physicists and quantum chemists, on the other hand appreciate the beauty and utility of our models of light-matter interactions. Each individual brings a unique perspective that enriches my understanding of materials and potential mechanisms of a variety of interesting phenomena that we can apply to interpreting our data.

The initial response of people who have just learned about our observations of self healing conclude that diffusion is responsible. Some of the air force scientists shared this concern. The idea is simple; the laser heats the molecules in the polymer, and the added kinetic energy causes them to move away from the laser. When the laser is turned off, the random walk associated with thermal jiggling causes the molecules to return. Thus, rather than the molecule burning (i.e. breaking into pieces) and then recovering (i.e. resembling), they simply move away and return.

When we first observed this phenomena, diffusion was the first hypothesis that we tested using optical absorption spectroscopy. All molecules absorb light at a set of discrete characteristic wavelengths. The DO11 molecule, our model system, has a big absorption peak centered in the middle of the visible part of the electromagnetic spectrum. If molecules move away from the beam, then the height of the peak will drop, but the shape will remain unchanged. If the molecular structure changes, as should happen in the photo-decomposition process, then a new peak forms that is characteristic of the "burnt" molecule. One can show that as one set of molecules is being converted into another set, then the spectrum evolves in a way where all the spectra cross at one point. This point is called an isobestic point.

When I am asked if self healing might be due to diffusion, I can confidently respond that we see an isobestic point in the linear spectrum. Everyone in the audience then usually thoughtfully shake their heads and acknowledge that this is strong evidence against the diffusion hypothesis.

However, upon reading the article about the Hauser case, I began wondering if I am not being deluded by confirmation bias. While we do see an isobestic point, the process of aligning the probe light (used to measure the spectrum) with the pump is difficult, so the measured absorption changes are not always clean. So, I started wondering if we were not being fooled by subconsciously dismissing data that does not meet our expectations. Upon my return, I met with my students and suggested that they try all sorts of other experiments that differentiate between the two mechanisms. I regret this extra burden that I place on my graduate students, but I would hate to be wrong.

On the computational front, we are also seeing weird results that may render some of our ideas invalid. In particular, we normally observe that when a quantum system is at the fundamental limit, only two excited states contribute to the nonlinear response. We call this the three-level ansatz. In recent Monte Carlo calculations, we are seeing rare outliers where more states contribute. Even more disturbing is that in these rare cases, the limit appears to be broken by a tad.

There are two interpretations to these results. First, there may be a flaw in the fundamental limit calculations. In our work using variational approaches of potential energy functions and vector potentials, we never see a system with a nonlinear response larger than 0.709 times the fundamental limit. In these cases, the three-level ansatz seems to always hold. That gives me some degree of confidence that the calculated limits are correct. The resolution to the problem might lie in the fact that the Monte Carlo work uses truncated sum rules. However, such truncated sum rules are also used to calculate the limits. Are we dismissing the effects of truncation when it suits our purposes?

It is likely that there are many subtle issues that we will need to consider to resolve our new observations. I fear that my brain may be driven by confirmation bias into believing in the fundamental limits and to blame counterexamples on the problem of truncation. We must consider the possibility that the limit calculation may be flawed, in which case it needs to be fixed. In the end, getting to the truth should be our top priority.

I learned about the quirky Monte Carlo results just before my trip to Dayton, Ohio. In addition to my preoccupation with my possible affliction with confirmation bias, I am overwhelmed with all sorts of other work, such as doing the annual reviews of our faculty, refereeing papers, reading dissertations, writing letters of recommendation, and writing new papers, as well as struggling to keep up with my lectures and homework solutions - all while fighting a nasty cold.

I look forward to weekends as a time to catch up with my work. Given my workload, I cannot realistically climb out of this hole; but, I take pleasure in my expectations that in the process of preparing for class and pondering our problems, I will learn new physics. Perhaps these activities will lead to the next new breakthrough on our understanding of the universe. I'll keep you posted.

Sunday, February 13, 2011

Advice to future academics

A colleague and old friend of mine was charged with running a session at an NSF-supported workshop on career choices for PhDs. He thought that it would be instructive to compile the key success factors for each type of job. So, he asked me for my thoughts. I provided the following subjective response that is based on my own limited experience. Many of you may have other ideas. If so, I would like to hear about them.

I believe that success in a physics department at a major university depends critically on the following factors:

1. An excellent grounding in the core physics areas, as taught in a rigorous set of doctoral-level classes. This lays the foundations upon which a researcher builds a deep understanding from which new ideas spring and puzzling results are interpreted. I believe that being required to pass a PhD qualifier exam helped me bring together my knowledge into a coherent form that serves me even today in many aspects of my research.

2. Experience doing research. There are many intangibles required of a researcher, which are not easily taught in the classroom. The best way of learning is doing. A key part of my development was working with other bright colleagues in the lab both in graduate school and early in my career. I think that it is easy to train a specialist to work in a narrow research area. The training of an academic scientist, on the other hand, requires the ability to sniff out new research directions and to take chances on new ideas that may be far outside our comfort zones. Again, the best experience is working with the brightest and most successful people.

3. Writing skills. Academic scientists are required to generate funding from external sources, which in addition to good ideas, demands the ability to write clearly. Our product is new knowledge, which appears in journals for future consumption. Since writing grant proposals and papers is a large part of what we do, success hinges on the ability to write clearly and effectively.

4. Communication skills. Presentations at conferences and other research institutions are a critical part of giving our work visibility. This requires the ability to present ideas clearly and at the appropriate level for the intended audience. Good Communication skills are also a crucial aspect of good teaching in the classroom and when interacting with students/colleagues in the lab.

5. Passion for learning. A passion for learning new things drives both excellence in teaching and creative research. An individual who is not continually being challenged is not being fully productive. An expert is a person who has learned everything the he or she needs to know to perform a specific task while a scientist is a person who is comfortable living in the unknown. An indicator that I am doing my job is the constant feeling of being stupid. Smug satisfaction in one's own expertise is a sign of mental stagnation. If you have become an expert, move on to something new.

Saturday, February 12, 2011

Cutting life in half

When our son was born in 1985, we purchased a Sony Betamax camcorder. Back then, the Sony Beta format was king. Its quality was clearly superior to the competing VHS format. However, Sony made the mistake of keeping their technology proprietary, so the VHS format eventually won out. These days, VHS tapes are passe, as is its 8mm successor.

While I was in graduate school, our Egyptian friend introduced us to tabouli, a Lebanese dish that my wife and I loved from the minute we took our first taste. Our friend Saleem graciously volunteered to teach us to make this wonderful but labor intensive dish at our home, and that's how our Friday-night tabouli and pasta dinner ritual started.

I introduced my new Betamax camcorder to our guests at one of these dinners by taping the event. Later, while we excitedly watched ourselves on a tiny TV, Saleem commented, "This camera cuts life in half. We spend half our time living and the other half watching ourselves on TV." I was reminded of these past good times while preparing my annual review materials. It made me realize how much time we waste documenting what we do.

To waste more of our time, our university implemented an ill-conceived notion to computerize the whole process with a clunky system. The system, called WORQS, was designed to save time; but, it becomes more cumbersome when the reported activities become nuanced and diverse, as is often the case for productive faculty members. My extreme dissatisfaction with this system lead me to coin the phrase WORQS SUQS.

I have already spent a large chunk of my Saturday afternoon updating my CV so that I can then spend another day entering it again into WORQS. My aggravation with the process reminds me of all sorts of new well-intentioned but time-wasting demands on researchers. For example, in an effort to prevent scientific fraud, the National Science Foundation now requires a data management plan. I agree with transparency as well as forcing overly-protective scientists to publicly share their data, especially when the work is paid for by our citizens. However, such rules enforce undue additional time demands on all the scientists.

I have many complaints about regulations that waste time with little in the way of overall returns, but I will leave that for another day. For now, I am attaching, below, an exchange that I had with the administration a few years ago. You may wonder why I wasted time doing so. My assumption was that a university, populated by members of the intelligentsia, would respond positively to a logical argument, resulting in increased efficiency. Instead, I was perceived a spoiled brat who refused to do what little was asked of me. Here is the exchange, with names removed:

Dear Administrator,

I am surprised at the university's attitude that we are to fill out our annual reviews online because WSU pays us and that's what we are expected to do - case closed. This goes against what I understand to be the principle of shared governance. The purpose of my email to you and the WORQS committee was to point out how some parts of the system are poorly designed from the point of view of the faculty member and the chair who does the annual reviews. I was not attempting to shirk responsibility, but rather, was bringing to your attention the features of WORQS that are inefficient, inferior to the present system, and do not fulfill - as far as I can see - the needs of the administration and the mission of the university.

I wholeheartedly embrace the idea of a computerized system that both collects important data for the university; and saves time of faculty members and chairs when preparing annual reviews. In particular, the data mining features are excellent. Not having to enter my lists of grants and proposals; but rather having WORQS lift this information from OGRD saves lots of time and effort. And having WORKS download a list of the classes that we teach by pulling data from the registrar is another good idea.

Unfortunately, there are other aspects of WORQS that are cumbersome and poorly designed. Is the point of WORQS to make it less time-consuming to enter information? Is it to provide better information for the chair in evaluating performance? Are there certain key metrics that the university needs to collect for the state legislature for purposes of accountability or to give us ammunition for showing how great we are? If so, portions of WORQS need a serious overhaul.

I focus my criticism on the scholarship menu as an example of bad design.

Scholarship Menu

Time Savings?

Since the WORQS system is purported to save time - or at least to not take much extra effort, I timed the publication-entry process in the "scholarship" menu item. It took me, a fast typist and computer geek, 50 minutes to enter my 2006 publications (had I not timed myself, I would have guessed 10 minutes) compared with 55 seconds to cut and paste the same ones. Net time wasted: 49 minutes.

The major problems from the faculty member's perspective:

a) Having to add journals in one menu area first then pulling down a journal name from the publications area takes time. Note: One of the new journals in which I publish was not on the list, which I had to add. After adding it (I needed to first find its ISSN), I got an email a few minutes later that it was on the system.

b) The method of adding coauthors is perhaps the worst feature. In one case, I had a paper with 13 coauthors, most of whom will probably never coauthor another paper with me in the future. Your system erroneously assumes that collaborators remain static. Since the coauthor field was not required, I just left it blank. Going through this process would have wasted lots more time. In fact, I left most of the non-required fields blank.

c) There is no field to add the year of a publication. There are cases where a paper appears in print in the year following the "publication date" (the volume number, etc. is not known until after annual reviews), so those papers fall through the cracks.

d) The present system allows for double counting by allowing a talk and a conference proceedings paper to be entered individually.

Usefulness to Chair?

With regards to publications, the information that is not required (like the coauthor list) is perhaps one of the most important since it gives a snapshot of collaborations and the students involved in the research. As chair, I would refuse to force my faculty to waste their time making such entries since for my purposes, seeing a CV is enough to judge their work.

Usefulness to WSU

Getting the names of the journals entered correctly and free from typos appears to be a priority to the administration and WORQS is clearly designed in a way that gets this information accurately. Other parts of the form leave plenty of room for inaccuracies.

This one menu item illustrates the vast number of the kinds of scholarly works produced by our faculty and the unique regard for journal articles. On the one hand, entry of journal publications enforces strict compliance on getting the journal name right, while we can add whatever we wish in all the other fields. This suggests that journal names are held in high esteem while other fields are not, so these other sections could and should therefore be left to free-form entry.

If WSU truly is interested in the most efficient system, then the following entry method would be superior:

1. Set up WORQS to import from online databases (such as Web of Science, for example) an individual's publications that appear in refereed journals.

2. Allow the faculty member to enter missing journal articles by wholesale cutting and pasting. Units whose research is published in journals that are not covered by databases should be permitted to use free-form entry.

3. Have WORQS scan the list of WSU coauthors and mark the ones that are students (grad or undergrad) from WSU's personnel/student database.

4. Since WORQS does not seem to care about the accuracy of conference proceedings, etc, allow faculty to cut and paste such "other" scholarly work freestyle.

The above would be a smart use of technology and would better embrace what I assume to be in the true spirit of WORQS.

Other Menu items

There are many other areas that can be improved. Similar to the method of getting grant activity directly from OGRD, why not do the same for committee service? At the level of department, college, and university, there are lists of committee memberships in place, so lift that info electronically in one swoop. The same thing can be done with M.S. and Ph.D. committees from lists compiled by the grad school. Is the WORQS committee analyzing the system and implementing such features to save time? If there are difficulties with implementing these kinds of recommendations, the response should not be to stick the faculty with the chore. Rather, you should call upon the faculty for ideas to improve the system. We all have a common vested interest in making WSU a better place.

As a general observation, it does not make sense to design a uniform online system when there are so many permutations in the diversity of work that we do. Even if you were to take into account all contingencies, there are just too many variables to enter efficiently. In fact, I would argue that if a single efficient form could be designed that takes into account all that we do, it would be an indicator that we are a mediocre institution. WORQS should be a system that is designed to save time and to augment annual reviews, not to drive a process that encumbers us in a web of minutia and forces us into conformity. Given the present implementation of WORQS, it emphasizes the latter.

In your email, you state, "we are relying on a faculty committee to work on this form and they simply don't agree with your argument." The issue is not open to opinion. Rather, we should turn to factual data to support our conclusions. In your trials of WORQS, have there been objective studies that compare it with the old-fashioned CV in terms of time and flexibility? Have you timed faculty who are using WORQS to determine the time it takes to fill out various parts of the form? Would a totally automated system that gathers the info needed by WSU, augmented with free form entry for the chair's evaluation, not make the most sense? If such a system is going to be required of all faculty, a cost-benefit analysis needs to be done; and, if any unit or group takes the brunt of the cost, then the system should not be imposed on anyone. I find it unacceptable that we are being told that WORQS is great when there is evidence to the contrary. Faculty should have more input before use of the system is required.

A university's top priority is to foster an atmosphere that nurtures a dynamic interplay of education and the generation of new knowledge. The administration's responsibility is to remove burdens on faculty so that we can fulfill our mission. I believe that this whole business with WORQS and how it is being imposed on us sends the opposite message. Perhaps I am naive in believing that our faculty is so diverse and creative that one form can not do justice without becoming cumbersome.

On a closing note, we need to ask ourselves the question, "How are we to think outside the box if we spend so much of our time filling out boxes?" Let's use technology to save time and make us more productive.


Mark G. Kuzyk
Professor and Associate Chair of Physics
and past Chair of the Materials Science Program

P.S. Most faculty in our department with whom I have discussed WORQS also found it quite cumbersome. I am attaching an email that I got from Faculty X. Also, you need to consider the reactions of the staff to this new system. They do not have the luxury of speaking up in the same way as faculty. Based on several conversations, my sense is that WORQS is demoralizing to the staff when most of the form is devoted to information that has no bearing on their work. Have you had broader input from staff beyond committee members?

My Original Email:

Thanks for your detailed response. I have been thinking about annual reviews from the point of view of a faculty member and as associate chair (I will be doing annual reviews this year in our department). I applaud the time saving features such as grants showing up automatically. I was credited with a $6 million dollar grant which I never got. So, that will make me look good! On the other hand, adding publications in the piecemeal way requested is very cumbersome. Even when the journal names are on a pull-down menu, it still takes lots of time to add the title, volume number, pages, type of pub, etc. - especially for productive faculty. Other parts of the form suffer from the same issues. The best solution is to give a choice of cutting and pasting items wholesale OR using the form.

The bottom line is that annual reviews are used to assess performance; and, the chair is best suited to do so. I can tell when one publication is of sufficient high quality that it trumps 10 others. So statistics are useless in my analysis of faculty performance. If WSU wants to gather statistics, then I would be happy to provide my CV for statistical analysis. But, as faculty members, we should not be REQUIRED to waste time entering information in a form that really has nothing to do with annual review, nor does it enhance the process in any way. In your email, you state, "…we can't do that if the system is to be useful." Useful to whom? It's certainly not useful to faculty nor is it to those of us who are doing evaluations.

Our time is chipped away to the extent that I am finding it impossible to do any research, so I do it in my times of "leisure." Yeh, this way of doing annual reviews may take only an extra ½ hour, but there are countless numbers of other such things that we are already doing that also don't make sense. Enough is enough!

My previous communication was sent to you through so and so because I was not aware of the committee working on this form. I hope that my comments are passed along to whatever group is working on this process and to whoever is making the decision to impose this online system.

At this rate, next we'll be requiring all teaching evaluations to be standardized. Though I would be a big beneficiary of such a system, this would be a huge mistake.

Your Response:

I understand your frustration about the many activities that faculty members are asked to do. Unfortunately, your comments have already been passed on to the faculty committee, and they disagree with you. WSU pays your salary all year. For that, the University only asks you to explain what you're doing once per year. Even if you have many, many, many publications and you are a completely incompetent typist, using the pulldown list will take you only minutes more than cutting and pasting. Your use of the pulldown list will allow the University to collect much better data. (So, for example, you won't get credited with the publication equivalent of a $6M grant that you didn't earn.)

I'm sorry that this issue is frustrating to you, but we are relying on a faculty committee to work on this form and they simply don't agree with your argument. So far, approximately half the colleges have decided to use WORQS for annual review this year even though it is still optional. To date, at least some of your colleagues believe that WORQS is a much needed effort saver for the entire University.

Tuesday, February 8, 2011

Good news comes in threes or fours (or fives!)

In the fall, I wrote of my despair when hearing about three papers that had been rejected, all in the same week. Finally, all of these papers are in various stages of the publication process.

First, our paper on photo-mechanical optical devices (which we call PODs), just appeared in Optics Communications this month, which you can view by clicking here. I have talked about this work at several conferences, and I am getting very positive feedback. I am hoping to get this work funded in the near future. I believe that breakthroughs in this area could lead to amazing new technologies that will excite even the most jaded techies.

Secondly, the paper that got the nasty review, which I described in my post The Good, the Bad, and the Nasty, just got accepted. The letter from the editor is appended to this post.

A recent paper that we submitted to Advanced Materials, a high-impact journal, just appeared in print. This is a comment that my former student and colleague Javier Perez-Morena and I wrote on the work of some of our colleagues from Australia. We showed that their results were even more important than they originally thought - a happy outcome for all!

I learned about all three good news items this morning.

Finally, our paper on imaging studies of self-healing will appear shortly in the Journal of the Optical Society of America B. In addition, we have several papers that are in various stages of preparation and under review. New trials and tribulations surely await us!

***Update*** After all of this good news, I was pleased to learn of a fifth item of good news. In a previous post, I had talked about our newest Monte Carlo work, which we submitted to JOSA B. I just heard from Shoresh that this paper was accepted for publication with optional minor revisions. For once, the reviewers and I agree!

Letter from Journal of Chemical Physics

February 8, 2011

Title: The effect of electron interactions on the universal properties of systems with optimized off-resonant intrinsic hyperpolarizability

Author(s): David Watkins and Mark Kuzyk

Professor Mark G. Kuzyk
Washington State University
Department of Physics and Astronomy
Post Office Box 642814
Pullman, WA 99164-2814

Dear Professor Kuzyk,

The above manuscript has been accepted for publication in the Journal of Chemical Physics. You may receive requests from our office to ensure that all manuscript files are complete and suitable for typesetting. Once the manuscript files are in an acceptable format, they will be forwarded to the American Institute of Physics publication office.

This e-mail is the only notification you will receive of the acceptance of your paper. If you have questions about the production of your manuscript, you may find contact information for AIP production staff at...

No revisions of the manuscript can be made before the galley proof stage.

Sincerely yours,

Letter from Advanced Materials

Dear Dr. Perez-Moreno,

We are pleased to inform you that your Comment

"A Correspondence on "Organometallic Complexes for Nonlinear Optics. 45.
Dispersion of the Third-Order Nonlinear Optical Properties of Triphenylamine-Cored Alkynylruthenium Dendrimers". Increasing the Nonlinear
Optical Response by Two Orders of Magnitude." by Javier Perez-Moreno, Javier Perez-Moreno
Mark G. Kuzyk
has now been published online.

Your article is available from

The citation data and abstract (if applicable) are available free of charge from the same link; access to the full text may require a subscription.
Please use the above-mentioned URL to link to the article from your institutional homepage, e.g., on publication lists.

A reference to your article is also available from your personal homepage by selecting "Author" and then "My Published Articles".

Best wishes,

Advanced Materials
(Editorial Office)