Can Theory Change What it is a Theory About?

In Beyond Freedom and Dignity B.F. Skinner writes, “no theory changes what it is a theory about; man remains what he has always been.” By this Skinner means that the underlying rules or processes that guide human behavior are constant, and that knowledge of these processes does not change their nature. However, throughout the social psychological literature we see suggestions of just the opposite—knowledge of a psychological process can change the psychological process. For example, Schmader (2010) provides evidence that simply teaching people about stereotype threat may “inoculate them against its effects.” The theory of social identity threat postulates that people are sensitive to contexts that threaten their identity, and when such a situation is detected people engage in ruminative conflict that can distract them enough to undermine their performance in that setting. Schmader is claiming that giving people knowledge of psychological processes predicted by theory changes the processes that unfold. This point raises several important questions: what is a psychological theory? Does psychological theory describe stable processes in the Skinnerian sense? Can we think of psychological theory in the same way that we think about theories of say physics or biology? If we believe theory must have some element of stability (e.g., if we believe light traveled at the same speed in the middle ages as it does today), and that theories exist out side of and are independent from our knowledge of their existence (e.g. the theory of special and general relativity existed before Einstein identified them, and his discovery did not change their quality), then can we classify social psychological theories as theories? My sense is no. Or maybe we need to modify our definition of what qualifies as a theory. Or perhaps our definition of stability in the processes that underlie phenomena and our belief that observation is independent from underlying processes needs modification.


Schmader, T. (2010). Stereotype Threat Deconstructed. Current Directions in Psychological Science, 19, 14–18. doi:10.1177/0963721409359292




Why Quants Don’t Know Everything

I’ve written previously on how quantification of social phenomena inherently gives importance (perhaps unduly) to the object of measurement.  Felix Salmon explores this idea to illustrate why quants don’t know everything. He quotes sociologist Donald T. Campbell who observed that quantification may actually change the nature of the quantified:

“The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.”

– Donald T. Campbell (December 1976), Assessing the Impact of Planned Social Change.

This all reminds me of similar logic drawn from the physical sciences. Some interpretations of the double-slit experiment explain the nature of light (i.e. wave or particle?) as dependent on the observer, tool of measurement, or location of observation. This is just food for thought. I will leave it to the physicists to hash out veracity.

My Science Is Harder Than Your Science…Bla, Bla, Bla.

There are several false assumptions that proliferate discourse in the media and among scientists about neuroscience, and science in general, that I believe are largely driven by artificial distinctions drawn between the “hard” and “soft” sciences. I recently came across an article on the fusion of architecture and neuroscience, that acts as just one example of a broader obsession and delusion about anything prefixed with “neuro”. The premise is that this relatively new field  is exciting because it provides an objective “window into the mind” that can better inform technologies and hard sciences than softer sciences like psychology, economics or sociology.

This particular article, examines how knowledge of the mind can improve architectural design. It asks questions like, can neuroarchitecture foster scientific discovery or improve development of social skills among autistic children by clever manipulation of aesthetics and physical design? While I agree that neuroscience can inform many fields including architecture, I object to the explicit tone that is too common in discourse on neuro[fill in the blank]. That is that neuroscience is a blessing because it is the first science of the mind objective enough to be fused with other hard sciences.

Here is  a sample from the article which quotes, Eduardo Macagno, professor of biological sciences at the University of California, San Diego:

“We are now really beginning to understand better how to measure the responses to the built environment without relying on psychology, social science, observational behavior. [Those studies] don’t have the quantitative and objective experimental approach that we believe neuroscience brings to the interface with architecture.”

This is a fundamental misunderstanding of social science that is driven by many things, but language is probably what throws people off the most. Macagno is confusing the tools of research with the method. Sciences that use new and exciting tools cloaked in complex technical language are often considered more objective, despite the fact that they use the same (or less rigorous) research methods as sciences with tools that are more easily understood in plain english.

One tool used in neuroscience is the fMRI which measures changes in blood flow to different areas of the brain. By measuring relative increases in blood flow to certain regions of the brain scientists can develop insights into brain function. While this is a powerful tool, accurate interpretation of results requires advanced training in technical language, physiology, methodology and statistics. Cloaked in complex language, people outside the field often fail to recognized that fMRI studies are usually correlational, relative increases in blood flow are only associated with increased neural activity, and blood flow lags behind neural events in the brain by about 2-6 seconds, which makes it difficult to pinpoint the connection between a stimulus or behavior and it’s associated brain region. This being said many similar methodological limitations are faced by the softer psychological sciences and even the harder sciences like physics.

It is frustrating to see scientists speaking in such absolutes about the quality of research going on in one field versus another. It points to a lack of homework on methodology, and snap judgements based on familiarity of language. Generally, the hard/soft distinction in science is not about rigor of methodology, it is more a distinction between inaccessible and colloquial language used to explain tools of the trade. Of course variability of the object of study might have something to do with it. But that is for another post.

[Featured photo courtesy of Royal Anthropological Institute’s Education Outreach Programme]