Interpreting Data and The Rosenhan Experiment

The lens in which we interpret data and experiments is crucial and can often become a hindrance, a blind spot on our lens if you will. For example, if we are evaluating a patients neurological condition and we have a doctrine in mind, say phrenology, we are more likely than not to find  something that we are looking for (a false positive). If we have instead, a doctrine of localization when we are examining the patient we will more likely than not find evidence for localization. The same holds true for a theory of neuroplasticity and each separate type of plasticity.What can be alarming is when we are presented with identical data, yet come up with two entirely different theories of explanation. This kind of subjective theory is dangerous in science and needs to be considered when we are conducting experiments and interpreting data and as we are reading over results of experiments and studies alike. This is exactly what double-blind studies attempt to guard for by taking away any preconceived notion of what a patient is (or is not) receiving. Double blind studies essentially strip away the subjective thoughts and opinions and leave us with objective data. I do not know if there is a science that isn’t vulnerable to this type of mistake. Certainly, the softer fields of philosophy and psychology are the most vulnerable with chemistry and physics being the least. Although, I would say that theoretical physics is very susceptible to this subjective error. Not only because theoretical physics necessarily involves making difficult predictions but additionally because the field is just so incredibly complicated.

Speaking of theEinstein famously disliked the implications of quantum mechanics that revolutionized physics in the first three decades of the 20th century. During international physics conferences Einstein would routinely pester Niels Bohr and the rest of the quantum mechanics founders, conjuring up thought experiments that seemed to contradict quantum theory.  After much dismay and some clever thinking Bohr would come up with an answer, usually by the end of the conference. The point I would like to make here is that even two of the smartest people who ever lived disagreed fundamentally about the same sets of rules and equations. Science, and especially in complex fields like theoretical physics, how you interpret data can be a very tricky business indeed. In fact, there were many new and often radical theories being proposed to explain the experimental results of quantum mechanics. This lead to something called The Copenhagen Interpretation which lead by Bohr, attempted to lay down a clear concise interpretation of the very odd world we were just beginning to discover. So, more objectivity leads to less room for interpretation but at some point complexity makes simple interpretation all but impossible.

Another great example of how we can misread and misinterpretation is The Rosenhan Experiment. This was a famous experiment conducted by a Stanford psychologist and professor who wanted to investigate the validity of psychiatric diagnoses. He and his colleagues (12 in total) attempted to get admitted into various mental hospitals across the U.S. The pseudo-pataients, as they were dubbed, confessed that they were experiencing tried auditory hallucinations. All twelve of them were admitted and diagnosed with psychiatric disorders. Following admission the pseudo-patients acted normally and after some time (I am not sure exactly how long) they told the staff that their auditory hallucinations had resided. All were forced to admit to having mental disorders and had to agree to take anti-psychotic drugs. The average time spent in the mental hospitals by the pseudo-patients was 19 days and upon release all but one patient was diagnosed with schizophrenia in remission. The second part of the experiment was not planned by Rosenhan but occurred when an offended psychiatric hospital administration challenged Rosenhan to send pseudo-patients to their facilities claiming they could spot the fakes. After a few weeks the psychiatric hospital released their results. Out of 193 new patients the hospital had seen 41 were labelled as potential pseudo-patients, in effect, patients who were lying and faking their symptoms. In reality however, Rosenhan had sent exactly zero pseudo-patients to the hospital. How could this hospital think that more than a fifth of patients they saw were faking it?

In science and medicine mistaking a false positive as a true positive is at best embarrassing and at worst deadly. Interestingly, evolution has programmed organisms to detect false positives and to react to them if they were true positives. This makes sense if we consider how animals interact in the wild. Consider a rabbit getting scared by a rustling in the bushes. Even if the rustling is nothing but the wind it is always optimal for the rabbit to respond as if the rustling was a lion. In other words it is much more favorable, evolutionarily, to be mistaken about a false positive than it is to be eaten.

Further Reading on The Rosenhan Experiment:

Further Reading on The Copenhagen Interpretation:

Further Reading on Data Analysis:


About ben james

"The stars keep me up at night" -some song I heard one time Human. Studying the intersection of Neuroscience and Computation. Wanna be (astro) physicist.
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s