Roshan Cools completed her undergraduate degree in Experimental Psychology, at the University of Groningen, in 1998. She then moved to Trevor Robbins’ lab at the University of Cambridge, UK, for an M Phil degree (1999), a PhD degree (2002), a St John’s College Junior Research Fellowship (2002 - 2006) and a Royal Society Dorothy Hodgkin Research Fellowship (2002 - 2006). She spent two post-doc years at UC Berkeley working with Mark D’Esposito from 2003, before moving back to Cambridge in 2005, where she obtained a Royal Society University Research Fellowship (2006 -2007). In November 2007 she returned to The Netherlands. She is now Principal Investigator at the Donders Institute for Brain, Cognition and Behaviour (Centre for Cognitive Neuroimaging) and Professor of Cognitive Neuropsychiatry at the Department of Psychiatry of the Radboud University Nijmegen Medical Centre. Her oration is planned in the last week of December this year. In an interview with Mariska Kret, she openly talks about her interesting research career.

One of your research goals is to understand the neural and neurochemical mechanisms underlying flexible behaviour. What is your definition of flexibility and in what way do you think researchers need to be flexible?

I think I would say now that one of my primary research goals is to understand the neurochemistry of cognitive control more generally rather than flexibility specifically. What is cognitive control? This is a somewhat poorly defined term,but generally refers to the ability to direct behaviour to currently relevant goals and critically depends on cognitive stability - our ability to maintain stable representations of current goals and our ability to resist temptation.This is a capacity that is more important today than ever before: We live in a consumer society that is characterized by an unprecedented abundance of information, choice options and temptations. Accordingly, we exercise cognitive control on a daily basis, e.g. when choosing between the healthy meal and the unhealthy snack, or when deciding between studying for an exam and partying until late.

However, adequate control requires more than the active maintenance and protection of current temptation-resistant goal representations (cognitive stability). It also depends on the ability to flexibly update those representations when circumstances change, i.e. cognitive flexibility.

One of the most fascinating aspects of our environment is that it is changing constantly. The ability to flexibly adapt to these changes is a capacity that humans are uniquely good at. Its importance is illustrated by the current turmoil in the financial markets. While a certain government might have previously voted against a particular plan, it might later completely reverse its behavior and vote in favor of this same plan, when changes in the environment have become sufficiently salient. Such flexible minds are essential for preventing disastrous outcomes, like collapsing banking systems.

Similarly, a flexible mind is critical for productive science. Most important discoveries represent incidental findings that go beyond the original goal for which the experiment was set up. This is why it is important to continue fundamental science rather than putting all our eggs in the basket of product-oriented research.

Is it always good to be flexible? Can you think of situations where tenacity is better than flexibility?

So how do our minds adapt to the changes around us? This is not a straightforward issue, because only some of the changes around us are relevant and require cognitive flexibility. Most other changes are irrelevant and should be ignored. In the latter case adaptive behavior depends on cognitive stability rather than cognitive flexibility. What we need is an ability to dynamically regulate the balance between these two processes (flexibility versus stability) depending on current task demands.

Some researchers criticize others for being too flexible in their definitions…

For a researcher, working from a strong theory has clear advantages. Other researchers will remember you and your theory. Moreover, it wouldn’t be good to be constantly distracted by different new input, new publications, new lectures. Besides, being too flexible makes it too easy, especially in fMRI, to interpret the results any way you want, which of course raises statistical issues…So, on the one hand, one needs to work from a clear research line, a theoretical framework. But on the other hand, as Trevor once advised me: “ never stick to your hypotheses!” A good balance between cognitive stability & sticking to the goal is as important as flexibility & letting go of your hypothesis in case of inconsistent data.

Has it ever happened to youthat you were completely wrong in your hypothesis and needed to step aside from it?

I do not work from one specific model, but I do work with clear hypotheses. I now have data that conflicts with the “Overdose hypothesis”, a theory in which I believed. The medication doses necessary to remedy dopaminelack in the dorsal striatum in PD patients may overdose the ventral striatum where dopamine levels are relatively intact. Consequently, PD patients become overly sensitive to reward and insensitive for punishment, after taking their medication. BUT, when we investigated whether this medication has similar effects in healthy people, we found contrasting results that were dependent on baseline levels. Whereas participants with low basal dopamine levels became more sensitive for reward and less for punishment, participants with high baseline levels showed exactly the opposite pattern. So this made me doubt about the Overdose hypothesis. We thought that the behavioral effects in PD patients after taking their medication, was the effect of an overdose on a healthy, intact limbic system. I have not yet published this but it is a very robust effect and I need to move forward with it.

What research methods do you use?

I started with behavioral experiments in different patient groups who exhibit impulsivity and/or compulsivity (Parkinson’s disease, ADHD, antisocial personality disorder, depression) and tested them on and off medication. Via psychopharmacological interventions such as controlled medication withdrawal procedures in patients with Parkinson’s diseaseand ADHD, administration of acute drug doses to healthy volunteers but also dietary depletion procedures, I gained more and more insight in cognitive control. This broadened my interest and I started to apply different research methods from experimental psychology such as fMRI and TMS.

What do you specifically investigate with TMS?

With TranscranialMagnetic Stimulation we want to investigate the role of dopamine in the basal ganglia on cognitive flexibility (this idea is based on the finding that PD patients have lower dopamine levels and are rigid in their movements, but also in their thinking).And than I found out about work from Canada, in which a TMS method was used that has very specific effects on dopamine in the striatum. By applying TMS to the frontal cortex, dopamine levels in the part of the striatum that is connected with the frontal cortex were influenced. What we did was the following. After applying TMS to the frontal cortex, we put participants in the MRI scanner and gave them a certain task on which the performance is normally influenced by dopamine. We found that TMS over the frontal cortex had very selective and specific effects on the striatum while executing this task. This way, we found a new technique to influence the striatum via the frontal cortex.These findings are particularly interesting in the light of treatment for PD but also more broadly, for cognitive neuroscience as a whole.

You started your career in Groningen. What brought you there? You moved from Groningen to Cambridge (UK), to Berkeley (USA) than went back to Cambridge and since 2007 you work in Nijmegen. Have you settled down?

I choose for Groningen because the study Psychology was one of the better ones in the Netherlands. Now I am back in the Netherlands and I am happy to be home. I was tired of all the travelling and yes, I settled down and bought a house in Nijmegen! I could have stayed in Cambridge for another 10 years. But I was there already for 8 years and the facilities were not so great asthe Donders offered me.

Could you describe the difference between the research climate at the Donders Institute and the other institutes you worked at?

My position in Nijmegen is very luxurious, as compared to both Berkeley and Cambridge. I hardly have to teach and there is plenty of support, so I don’t need to spend much time on administrative issues or solving technical problems. I can really focus on my research. And the facilities are much better here. In my office in Cambridge, there were 15 year oldspider webs hanging down from the high ceiling. The buildings were old, the scanners also...

In Berkeley and Cambridge, a lot of researchers (big names) are working on the same topic (in Berkeley the frontal lobes, in Cambridge psychopharmacology and fronto-striatal circuits). But even though at the Donders people are much more working on their own topics, there is more collaboration. I also needed more intellectual space. The danger is that in a place where multiple people have very similar ideas and study similar questions, people come up with the same research questions.

Did this research climate where a lot of researchers were working on the same problem also lead to more competition between researchers?

Yes, it did, and it also resulted in more stress. But actually, I never really suffered from it much.Of course, at the Donders we are stimulated to get grants and there is a lot of competition there. But topic-wise there is less competition.

You have a very impressive list of publications. Which one would you like every ‘psychonoom’ or ‘psychonomist’ to read?”

The most accessibleone is the chapter in The Paradoxical Brain.

You have a large lab with many PhD students and post-docs. At this point in your career, do you have time to analyze data yourself or do you spend most time supervising the others?

I am beginning to learn to let go of data collection and analysis. I found this very difficult, but I realize that my PhD students and post-docs are often better at it than me. So I leave it to them. But I do look at the results step by step with each and every student.

Roshan Cools is key-note lecturer at this year’s NVP conference. In her talk she will focus on the role of serotonin in impulsive behavior and depression.