Challenge Those Assumptions!

by Kevin Boehnke

Happily, we live in a time where challenging dogmatic ideas about the world is not punishable by being burnt at a stake. We know the earth is spherical, that the Earth revolves around the sun, and innumerable other facts about our tiny place in this immense universe.  These breakthroughs have come when thinkers push back at conventional wisdom with careful reasoning to build new frameworks of thought. Despite our vast knowledge, our understanding of the world remains incomplete. Luckily, we’re continuing to push at the boundaries of thought, although we operate under the restrictions of our current scientific assumptions, which are influenced by the social  and cultural milieu where science takes place.

Humans are fallible, even scientists who strive for objectivity. Our assumptions limit our interests, the questions that we ask, and the ways that we try to answer them. However flawed these assumptions may be, this process still leads to progress, with jerky leaps of discovery and jubilation at having found ‘the answer’ followed by shame and self-doubt upon realizing one or more of the following:

  1. An important factor wasn’t taken into consideration.
  2. The exciting results are actually spurious because of poor study design or execution.
  3. Science is really hard, and there is lots of failing involved.

Scientific discovery: The viewpoint from a 2nd year PhD student. Source: Kevin Boehnke

However, despite these challenges, researchers are keeping at it, elucidating the details of the magnificently complex puzzle that we call life. An important way they do this is by challenging the basic assumptions of their fields, which in turn pushes up scientific standards across the board. Some recent victories on this front in the realms of mice and men:

1. The National Institute of Health (NIH) enacted a policy that would eliminate sex bias in NIH funded research, effective October 2014. Historically, male subjects were overwhelmingly used in both animals and human clinical trials and basic science experiments. This occurred for several reasons, including: hormonal variation due to the menstrual cycle can lead to less easily interpretable results, exclusion of women from clinical and drugs trials due to fear about reproductive effects (remember Thalidomide?), and sexist attitudes – because more scientists were men, they decided to study men. However, it’s pretty clear that men and women function differently biologically (childbirth, anyone?), and that they would be affected differently by medications and other chemical exposures. Who would’ve thunk it!? (OK, probably most of you).

2. Lab mice are being kept too cold! Mice prefer temperatures at 30-31C, but are kept at 20-26C for the convenience of researchers. At 20-26C, it’s easier to work (much less sweat), mouse cages need to be cleaned less often, and mice eat and drink less. However, at colder temperatures mice appear to divert energy towards keeping warm and away from promoting an immune response, causing faster rates of tumor growth. This likely skews cancer rates in mouse studies, falsely masking (or amplifying) the effects of treatments and exposures.

If this room were at 30C (86F), these poor surgeons would be sweating like pigs. Source

3. Mice and rats become anxious and stressed out in the presence of men and masculine odors/pheromones. This may bias studies examining stress responses and pain, since stressed out animals have higher pain thresholds than those who aren’t stressed out.  While these results don’t mean that humans will behave in the same way, this may bias health outcomes in many animal studies, since stress is so intrinsically tied to different physiological responses.

A simple solution to some of these problems: document everything and make that information available. Reporting the temperatures that animals are kept at, the sex of test subjects, and the sex of research staff in animal studies doesn’t take much time, and then eliminates potential confounding. Obviously, there are many more factors at play that could also be reported. While this new set of data would be massive (and tedious to gather), we are in the age of big data, and could crunch these extra variables when analyzing results.

So what will change next? In my mind, some of the lowest hanging assumption fruit for human studies is race and ethnicity.  The people most likely to participate in research studies are white, male, middle-class,  and educated. Despite the fact that this population is clearly not representative of the general public, human trials rely on data from this group to identify safe drugs and treatments to use universally.

While fixing this would be an important step forward, it is not without controversy.  When BiDil (a heart-disease drug known to be effective in African Americans) was introduced on the market, it was a financial failure because of concerns over scientific racism: why was this single drug only tested on African Americans, when all other drug trials are overwhelmingly carried out on whites? It’s not an environment than engenders too much confidence, especially with the difficult past of the Tuskegee syphilis experiment and compulsory sterilizations. At the same time, recognizing that there are genetic differences between people of different descent is key to finding the most effective treatments. Trying to address this in an open, civil way is a must for starting to fill this gap in clinical studies.

As humans, we use assumptions to build a framework for how to think about the world. We would be totally lost (or extinct) without our ability to make snap judgments and think about ideas linearly, both of which are impossible without assumptions. However, it is worthwhile to step back and re-evaluate them. Otherwise we might still think that the sun revolved around the Earth.


Jazin, Elena, and Larry Cahill. “Sex differences in molecular neuroscience: from fruit flies to humans.” Nature Reviews Neuroscience11.1 (2010): 9-17. DOI

Kokolus, Kathleen M., et al. “Baseline tumor growth and immune control in laboratory mice are significantly influenced by subthermoneutral housing temperature.” Proceedings of the National Academy of Sciences110.50 (2013): 20176-20181. DOI

Sorge, Robert E., et al. “Olfactory exposure to males, including men, causes stress and related analgesia in rodents.” Nature methods (2014). DOI

Swanson, G. Marie, and Amy J. Ward. “Recruiting minorities into clinical trials toward a participant-friendly system.” Journal of the National Cancer Institute 87.23 (1995): 1747-1759. link