The New Phrenology?

The nineteenth-century was as much a century of science as the twentieth or twenty-first.  Major breakthroughs occurred in all scientific fields, the most well known being Darwin’s theory of evolution.  But it was also a great age of pseudo-science, including dotty ideas about health and nutrition, Social Darwinism, magnetism, and so forth.  One of the more popular pseudo-sciences was phrenology, which purported to be able to determine personality type and intelligence by feeling the pattern of bumps on a person’s skull.  Phrenology maintained its appeal even into the twentieth century by couching its assertions in quantitative and statistical terms; there was even an “automatic electric phrenometer” that purportedly measured the variations in the cranial bones.

Do we now have a new pseudo-science in the shape of the use of brain imaging techniques to explore the correlation between brain states and emotion, personality, and cognition?  Certainly there has been a raft of studies claiming to demonstrate that brain states as measured by fMRI’s correlate with certain emotional states, prejudices, habits, and tendencies to believe in God, and so forth.  Numerous articles in the popular press and many best-selling books have not only reported these findings but have reached seemingly profound, breakthrough conclusions on human nature–some even reaching such extreme conclusions as to deny such age-old and valued qualities as free will and reason.  If these studies are correct, it would seem that human beings are little more than robots operating on software written into our genes.

How such robots could be sufficiently autonomous, as gifted with self-consciousness, to reach such conclusions about themselves has never been addressed.  Perhaps the researchers and writers are superior to the rest of us in this way.  But, perhaps, not to worry.  An article by Edward Vul, et al., “Puzzlingly High Correlatios in fMRI Studies of Emotion, Personality, and Social Cognition” (originally titled “Voodoo Correlation in Social Neuroscience”), applies the principles of statistical analysis to the claims presented in a sample of peer-reviewed papers and found that they did not stand up to such scrutiny.  Interested readers can read the article for themselves, but most of us probably will not be able to follow the technical details.  But the authors’ conclusions can be readily understood.

Vul et al. began their metastudy because they were struck by the statisticaly unlikely “high correlations between measures of individual differences relating to personality, emotion, and social cognition and measures of brain activity using fMRI.”  In sum, after subjecting the studies to careful analysis of the methods employed, the authors found that the studies suffered from the common error of what logicians call card stacking (what the authors call “the nonindepence error”).  The researchers selected statistical “noise that exhibits the effect being searched for, and any measures obtained from such a nonindependent analysis are biased and untrustworthy.”  In other words, the researchers massaged the data until it yielded the desired results.  To illustrate their point, the authors conducted a study which showed a correlation between temperature readings on a particular thermometer on Adal Island, Alaska, to the rise and fall of the stock market during the same period.  Yet of course, the temperature at any one place on the globe has nothing to do with the stock market, anymore than bumps on the head have anything to do with intelligence or a tendency to criminal behavior (part of the problem, of course, is that criminal behavior is open to definition–to commit a crime is to break a law, but if there is no law against, say, cheating on your wife, then cheating on your wife is not criminal behavior; note for example that homosexuality has gone from being a crime to being a psychological disorder to being just another natural form of sexual behavior, now found everywhere in the animal kingdom).

It is perhaps possible that someday, somehow, science will devise a means of truly reading the mind via highly sophisticated mechanical means, but at present it is far too early to reach any global conclusions on the basis of research using the current technologies.  Until that far-off time arrives, we will have to turn to the traditional sources of knowledge of ourselves, to our novelists, historians, sociologists, and anthropologists, even to our psychologists (or at least those who are not enamored of fMRI’s).

Post a comment or leave a trackback: Trackback URL.

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

%d bloggers like this: