Dr. Tchavdar Todorov
Reader at the School of Mathematics and Physics, Atomistric Simulation Centre (ASC)
Queen's University Belfast
University Road, Belfast
BT7 1NN Northern Ireland, UK
An Interview with Tchavdar Todorov, Queen's University Belfast, Northern Ireland, UK
Beilstein Magazine 2017, 3, No. 8 doi:10.3762/bmag.8
published: 4 January 2017
Tchavdar Todorov: We live in an age when the conditions under which we function are evolving faster than the characteristic social adjustment period. Electronic communication, for example, has resulted in a fundamental change in our working patterns. From email to the ease of accessing publications online, we are exposed daily to very different quantities and types of materials as compared to just 10 years ago. If we define research integrity, in the broadest terms, as the hygiene (metaphorically) of our working habits, environment and relationships, it is easy to see that the ethical aspects of research are facing new challenges.
We all have an intuitive notion of research integrity, yet quantifying it is not necessarily easy. Most of us probably had our first inkling that research conforms to definite ethical norms as early as our PhD years, although nowadays this can happen even earlier – as undergraduates in research project work, for example.
To define it in a few words, research integrity means honesty, fairness and dedication. But exactly what that implies in practice will depend on the context, career stage, and our individual professional and collegial circumstances. We can identify some generic situations where our academic ethical instincts become “activated” and then seek to identify a common denominator and thus interpret this powerful, yet sometimes elusive, “voice of conscience”.
TT: There is a big difference. It is possible to write superbly well, yet side-step the ethics of writing. Research integrity in research articles means, above all, honesty with the results: reporting all that is relevant and not cherry-picking just the points that best fit our claims; making the discussion and conclusions as solid as possible; and – crucially – giving credit to relevant work by others. A related subject is probably the most serious academic offence – plagiarism. Plagiarism can even take place subconsciously, such is the “osmotic” ability of information we’ve seen, read or heard elsewhere to diffuse into our own minds and present itself as ours. One needs to have a stringent set of test questions to continually ask oneself in order to ensure that proper credit is given where it is due. Anything can be plagiarized, a conversation, a presentation, a sketch, not just written work. We run dedicated sessions on these issues for undergraduates as early as their first year.
TT: It is the collegial and professional environment and relations as a whole. For most of us this starts with student–supervisor relations. Imagine you are a PhD student. You have been fortunate to make a solid start and have taken ownership of the project early on. This scenario, if we are not careful, can generate delicate problems. The first is student–supervisor friction. It could spring partly from a tacit sense of rivalry and partly from an element of competition for leadership. The authorship of PhD papers with multiple authors can raise the question of how to protect the student’s place in the work being written up. The rule is to put the interests of the student first. It is their project and the gateway to the rest of their career. This does not automatically mean, however, that they should always be the first author on collaborative papers. That depends on the relative contributions of all other authors and on who has led the actual writing.
TT: There isn’t a what-to-do recipe, but there are clear don’ts. As supervisors we need to remember that the project, irrespective of who designed it or how it is funded, is the property of the student. Our involvement as supervisors is our job, not a dividend. We must teach the student to “drive safely” and independently as quickly as we can and then we step aside. Equally importantly, however, as students or supervisees at any level, we need to work hard and earn intellectual ownership, with the responsibility this implies, and in the spirit of collaboration, not competitiveness.
There are simple norms that can make a big difference in maintaining collegiality. One is the avoidance of gossip. Workplace gossip is never good. It is worse in an academic setting, because research and academic attainment are something so personal and delicate. Another is that in research, competitiveness can be a great force for advancement but only when combined with the right inner motivation. Ambition should be engaged to advance greater causes than just ourselves. I have been fortunate to work with students, colleagues and academic leaders with these qualities and can appreciate the difference they make.
TT: Peer review is an essential mechanism for enforcing research integrity. It is also a major test of research integrity. In peer review, the question of conflict of interest arises first. As a reviewer, whether for a grant application or a research paper, we must be committed to writing a fair and informed reference, before accepting the review invitation. Equally important is the challenge of doing a decent job when we do accept. A well-researched and carefully thought-out review for a paper takes days, not hours, to write. We should remember also that good reviews can be of crucial help to editorial offices as well as everyone else who reads them, such as authors and fellow reviewers, and can set standards. Peer review, for all its potential faults (as any human-led process), is probably the number one mechanism of “policing” research integrity.
A student asked me recently, “Who ensures that published work is correct?” A simple question that shows the enormity of the responsibility involved. Even undergraduates make use of published work, and if there are problems within the work, the detrimental effects start there. Such is the importance of peer review.
TT: Yes, postdoc–principal investigator (PI) relations and collaborative work are further examples. We live in a “publish-or-perish” age that is combined with the pressure of obtaining funding, where grant application success rates are sometimes only a few percent. A successful grant application followed by successful recruitment automatically places everyone under a spotlight. The postdoc needs to produce the best work possible, in order to increase their career prospects, and the PI must “deliver” the programme in the application, and as many published articles as possible. Such are the initial conditions for a grant-funded piece of research.
Set against this background is the reality of research. When research ventures into new territory, things can easily not go to plan. Unexpected challenges arise, methods may need re-formulating, time-consuming computations may need re-doing, bugs in codes may hold you back, or papers that you thought were finished may suddenly throw up a can of worms. Or the postdoc may receive an early offer elsewhere and, being in no position to decline, may announce their departure halfway through the grant. These pressures and unknowns can put working relationships under great strain.
TT: The above conditions can encourage – nearly enforce – unethical scientific behaviour: superficial and hasty work, premature publications, shortcuts and insufficient checks, and when it comes to the fundamental basis of science, that is, having time for quiet critical thinking, this can almost sound like a joke in high-pressure situations. Let’s pick a classic pathology that can arise in this scenario.
Progress has been made and a paper has been written and submitted. During the review, or let’s say just post-acceptance, an error comes to light. This could be a bug in a computer program, or faulty reasoning in the underlying theory, or something incorrect in the experimental setup. What do we do? The path of least resistance is to do nothing, and that is an example of scientific fraud. However inconvenient or unfortunate the moment, when a problem in a piece of research comes to light, we must act on it immediately. For example, write an erratum or clarification. If the paper has not been published yet, inform the editors, and make the changes needed, even if that means delays in publication. If it is beyond salvation, then we must retract. But the thing to never do, however tempting it may seem, is to sweep the problem under the carpet. Making a conscious compromise even only once can be the start of a roller coaster that may – with a fair probability – compromise the rest of our work in science.
Research nowadays often puts collegiality and “good practice” to the test. One needs to be flexible and accommodating. But two norms that we should keep to are not to compromise where research ethics are concerned, and not to allow situations to take root that create unfairness. Good work should be recognized, and all those involved should give, as well as expect, credit where credit is due.
Analogous considerations apply to funding applications and the all-important scientific support case. “Overselling” what one is promising to do to make it look more avant-garde, or not giving full credit to relevant achievements by others can ruin a project, as can “overcosting” the budget.
By now we see the emergence of a common thread: integrity in science and its opposite – scientific or academic misconduct – are fundamentally the same as in any walk of life, with the same basic misdemeanours (e.g. dishonesty). These norms arise similarly in our day-to-day lives. In science, they just become projected onto the research context and realities.
TT: Students, even in an undergraduate setting, are very receptive to questions and discussions about research integrity. That early stage in our careers is where good practice and clear research ethics can be instilled, through dedicated seminars and courses, as well as in one-on-one communication with students working on a project or assignment. Students quickly “tune in” to the subject and start asking all the right questions, opening the ground for further education.
Academic publishers and funding agencies have more of a “policing” role: by this stage it is probably too late to try to educate from scratch. The task then is to enforce scientific honesty. University administrations have a vital role to play in making research integrity and good practice a rewardable goal, alongside productivity and meeting targets.
TT: It is an uplifting experience to see how aware and responsive young people are to the challenges and questions of research integrity. Once a discussion has been started, about plagiarism for example, or about the responsibilities of research reporting in publications, students immediately pick up key themes and start asking incisive questions. They can do this in relation to their own work, or in relation to the wider research setting. Early education will always be a primary mechanism to address these challenges.
Another resource open to us all is the school of life: learning from our mistakes and from those of others.