'Integrated science': a better way of teaching science?

Summary: Princeton University currently offers a special course programme for undergraduates: integrated science. Would this be a good idea to stimulate scientific talent, or is it just a marketing ploy?

Quite some time ago, one of my former students pointed me towards a website of Princeton university, http://www.princeton.edu/integratedscience/ . His question to me, as beginning expert on science talent development, was what I thought of such a programme.

While this question may sound simple, it actually goes quite deep, and it is only now (with a lot more reading and thinking under my belt) that I'd venture to give an answer. But before that, I should probably explain what this 'integrated science' is all about.

'Integrated science' is an undergraduate programme at Princeton University, consisting of a number of related courses which cover some mathematics, physics, chemistry and biology as well as teaching students how to apply for example computer programs to science problems. Its aim seems to be to give students a broad yet solid science background, as well as preparing them for the sciences of the future.

For the students, integrated science would seem to offer multiple advantages.

First of all, integrated science may be less 'scary' to students than traditional monodisciplines such as offered at Dutch universities. Many students, when leaving highschool, are rather uncertain what discipline to pursue; it is a weighty decision, and despite open days at the universities students have often no idea what they get themselves into. This leads to large numbers of students abandoning their studies within the first year, or switching studies (only 20% of Dutch students get their BSc or BA in three years). Other students try to prevent burning bridges by following studies that are 'generally useful', like economics, management sciences, psychology or law. Natural science is generally considered too specialist, and therefore too scary to choose. By creating a course such as integrated science that students can choose while feeling that they can still go any direction they want (so not necessarily being condemned to a life of 'stuffy' labwork), the Princeton faculty makes it more attractive to choose science. And since it is, in my opinion, very good if more people have at least a decent understanding of science, and some people who may not think themselves to be much of a scientist to discover the joy of science, I can only applaud integrated science for that.

The second reason that I think integrated science may be useful has to do with the American system. While many European systems (such as the Dutch system) only allow students to choose from one of the many monodisciplines, the US system generally allows students to pick and choose what courses to follow. However, the freedom this offers is not without its price, for at least some students apparently wind up with a hodgepodge of courses that doesn't make them very marketable at all (as the Avenue-Q song goes, “What do you do with a B.A. in English?”). Having a solid course with a certificate may ensure that Princeton students don't shoot themselves in the foot (too much) and are better prepared for a career.

However, there may have been another undercurrent in my former student's question, and it may be well to discuss it here, since this, after all, aims to be a science talent blog. That question is: would integrated science make people better scientists than traditional (European) monodisciplinary studies?

The cop-out answer would be that we can only know for sure in twenty years (which is scientifically speaking true). However, I think we can make some decent predictions based on history.

Briefly, if one looks at the careers of famous scientists (such as Darwin, Einstein, Newton), most of them seem to have been very monodisciplinary. That does not mean that they did not have hobbies (Einstein played the violin and loved sailing), neither does it mean that they never read books outside their field (Darwin read about geology, as well as Malthus' book on economics and population growth). However, their main field was the subject of almost all their exertions. Even people who exhausted their university's courses in multiple fields (like Linus Pauling did) quickly specialized in one area. Globally, if one trusts Simonton's conclusions in his book “Origins of Genius”, there are basically two kinds of creative breakthroughs. The first and most common kind is by people who have thoroughly mastered a specific field over many years (such as Charles Darwin). The second kind is usually a once-in-a-lifetime spark of brilliance of someone doing work in one field (and having reached a decent level there) switching to another field; but then the multidisciplinarity could be considered more 'serial', stacking one field on top of another, instead of truly combining them. Creativity researchers have also found that most breakthroughs are a combination of deep domain-specific knowledge obtained by many years of intensive studies in a particular field and general heuristics that everyone possesses (see for example Weisberg's paper in the Cambridge Handbook of Expertise and Expert Performance). In short: if multidisciplinarity does make a great scientist, it is not readily apparent from history.

Of course, this is just history, and people could justly accuse me of not taking into account that in the future areas will pop up that combine multiple traditional fields. However, that has been happening for over a hundred years already – take for example biochemistry. Biochemistry slowly came into being during the 19th century – but not by people who had elaborate chemistry and biology background, but by chemists who were curious enough to investigate biological processes. In general, even authorities in 'new fields' commonly just start as specialists in one close field, who'discover' the adjacent field during their PhD or postdoc and make it their speciality. For example, one of the most well-known Dutch bioinformaticists started out as an 'ordinary' biochemist who discovered the possibilities of the computer, and one of my own former supervisors, a computational chemist, started out as an regular organic chemist who did an internship with a professor who was just discovering the possibilities of computer calculations for chemistry. In brief: it would be reasonable to expect that even the leaders of the new scientific areas of the future start out as good monodisciplinarists today, even though they probably would be monodisciplinarists with some curiosity and an open mind about developments outside the traditional confines of their field.

It IS of course possible that a specially combined programme such as integrated science would outperform traditional multidisciplinary science, or even monodisciplinary science. At the moment, though, based on what I know, this seems rather unlikely, as our knowledge of education is still so shaky that attempts to let students generalize beyond the specific knowledge taught in courses is generally doomed to failure; and integrated science may not be much more effective than any other 'multidisciplinary' university programme. It will probably teach students some tricks that they may be able to reproduce, but deep understanding and subsequent innovative thinking ('far transfer', as it's called in the training lingo) may be outside of the capabilities of even Nobel-prize-possessing faculty to just 'teach in the course'; students can probably only develop that during long subsequent self-study and specialisation.

This somewhat bleak conclusion, however, does not mean that interdisciplinary science is useless. If it dispels the fears that many young people have about being or becoming a scientist, if it helps young people to get a solid education to get a job (multidisciplinarity, though a bit awkward in academia, is often appreciated in industry as there one needs to collaborate to a much greater extent with people with different backgrounds), it is probably a valuable addition to the Princeton curriculum. And, as the young talented mathematicians and reseach neurologists in Bloom's study ('Developing Talent in Young People') have shown: it does not really matter if one tries out diverse areas during their bachelor studies; sooner or later they found 'their field', and their passion for it made them excel in it. 'Integrated science' may be a counterproductive (un)specialisation for a PhD-level young scientist. However, for freshmen, it may serve as an attractive springboard to eventually help them find the field they'll love.