Share this article on:


Phillips, Lisa

Original Article

A study that used functional magnetic resonance imaging (fMRI) to examine voters' reactions to political candidates in the heated 2004 presidential race is raising questions about possible uses of brain scan technology – and its limits in assessing cognition.

Investigators used fMRI to scan the brains of 10 Republican and 10 Democratic voters who looked at political advertisements and images of the presidential candidates. While viewing their own candidate, the subjects showed activity in the ventromedial prefrontal cortex. Looking at the opposing candidate activated the anterior cingulate cortex and the dorsolateral prefrontal cortex.

In a January 18th editorial in the New York Times, Joshua Freedman, MD, Assistant Professor of Psychiatry at the University of California-Los Angeles (UCLA), described the activity in the ventromedial prefrontal cortex as indicating “strong instinctive feelings of emotional connection.” He wrote that the anterior cingulate cortex activation suggested “cognitive and emotional conflict” and the dorsolateral prefrontal cortex is an area “that acts to suppress or shape emotional reactions.”

Dr. Freedman interpreted the scans as indicating that voters were “mentally battling their attraction to the other side.” The editorial ultimately argued that political allegiances may not be “driven by a deep commitment to issues” and instead have to do with longtime allegiances, similar to loyalty to a sports team.

Back to Top | Article Outline


The fMRI study was conducted by Marco Iacoboni, MD, PhD, Associate Professor at the UCLA Neuropsychiatric Institute and Director of the Transcranial Magnetic Stimulation Lab of the Ahmanson Lovelace Brain Mapping Center. The sponsor was FKF Research, a company founded by Dr. Freedman; his brother, Tom Freedman, a strategist in the 1996 Clinton campaign; and William Knapp, a consultant for both Clinton presidential campaigns and the Gore campaign in 2000.

According to Dr. Freeman, the company aims to research potential applications for fMRI for political and media consulting. Dr. Freedman said the company partnered with Dr. Iacoboni, who had already begun to examine fMRI as a tool for studying political thinking.

“While none of us foresee fMRI being used by political candidates at any time in the near future, major corporations are beginning to experiment with it to understand the effect of their marketing,” Dr. Freedman said. “We believe it is essential that this work be done in an open and ethical manner.”

The study has yet to be published in a peer-reviewed journal, though Dr. Iacoboni said a manuscript has been submitted for publication. He said the paper considers two more alternative interpretations of the voter brain scans, rather than focusing only on the hypothesis proposed in Dr. Freedman's editorial.

“The editorial was really a write-up that takes a position,” Dr. Iacoboni said. “That's the intrinsic nature of an editorial, which should not be judged with the same criteria used for peer-reviewed scientific articles.”

Back to Top | Article Outline


Dr. Freedman's editorial and other reports of the study in the mainstream press have provoked strong criticism from members of the neurology community, many of who see fMRI as a limited and inconclusive tool for assessing cognition. Kenneth Heilman, MD, the James E. Rooks, Jr., Distinguished Professor of Neurology and Health Psychology at the University of Florida at Gainesville, called the study “physiological phrenology with a whole bunch of psychobabble.”

“There has always been a desire to understand the mind,” he said. “The phrenologist did this by feeling bumps, and now we have this new instrument which shows changes in blood flow, but there are many problems with a poorly controlled functional imaging study because when you see selective activation of a part of the brain, it is difficult to know what this activation means.”



For example, Dr. Heilman said that although Dr. Freedman wrote in his editorial that activation in the anterior cingulate cortex indicated conflict, “with functional imaging the cingulate gyrus lights up with almost any task.”

Dr. Heilman added that there are still questions about the meaning of increased or decreased blood flow in certain areas of the brain and whether activation on an fMRI means increased activity of excitatory or inhibitory neurons.

He said any study that uses fMRI to assess emotions needs to have strict control procedures. “If you are attempting to show that there is some type of emotion being mediated by a specific brain region, you should attempt to make certain there are not other cognitive processes that might account for this activation,” he said. “Hence, you need control procedures and without these procedures all you have is physiological phrenology.”

Back to Top | Article Outline


Bruce H. Price, MD, Chief of Neurology at McLean Hospital in Boston, MA, said he also takes issue with the methodology of the study. The small sample size – 10 Democrats and 10 Republicans – may not take into account “immense individual variability,” he said. He criticized Dr. Freedman's interpretation of the fMRI, calling it “a kind of biological reductionism of a remarkably complex set of behaviors.”

“At best, this is sloppy science which was highly exaggerated and offered premature claims,” he said. “At worst, it strikes me as George Orwell's 1984 circa 2005.”

Dr. Price criticized the “entrepreneurial and self-promoting interests” behind the research, asking: “Why shouldn't McDonald's look at people's brains as they light up, seeking ways to better promote their market share, which in a larger social context contributes to obesity and diabetes? Whose interests are being served and why? Is this really good science or just good marketing?”

And, some experts asked, should the New York Times editorial be used as a forum to promote scientific findings – without having first undergone peer-review and publication?

Back to Top | Article Outline


Yet Christopher M. Filley, MD, Professor of Neurology and Psychiatry at the University of Colorado School of Medicine in Denver, sees positive potential in the use of fMRI to peer into emotional and rational responses to politics.

“fMRI can display the parts of the brain that are involved with these kinds of processes, at least in a preliminary way, and this kind of localization may turn out to be important,” he said. “It's a promising way to get at least some idea of what goes on when we think about political issues and candidates.”

Like other critics, though, Dr. Filley took issue with the interpretation of the scan in the editorial on the voter study. He pointed out that such interpretations are particularly challenging to confirm because of the great complexity of thought and emotion that fMRI purports to demonstrate. For example, whereas postulating internal conflict when viewing the opposing candidate may be one interpretation, “an alternative explanation would be that there is no conflict, and the subjects are simply reviewing quite rationally why it is that they oppose in that candidate,” he said.

Dr. Filley said Dr. Freedman's analysis that voters are battling their attraction to the opposing candidate is “one a psychiatrist is more likely to make because of psychiatry's traditional interest in the conflicts that are thought to exist within a person's mind.”

Back to Top | Article Outline


Dr. Freedman defended his interpretation of the voter brain scans, explaining that his conclusion was also based on the widely-accepted observation from cognitive therapy that people often have “automatic” negative thoughts that induce negative emotions. He also cited a study done at Arizona State University that found that voters tend to join the party their parents joined and only later adopt that party's issues and positions, feeling a loyalty similar to the type felt for sports teams.

“There's a process of people moving to embrace the views that they think they're supposed to have, as opposed to having a view and then pursuing the candidate that reflects that view,” he said.

Dr. Freedman added that other interpretations of the voter brain scans are possible. “I took the data available now and gave my opinion, which is a reasonable interpretation but not an interpretation of certainty,” he said. “I don't think this should be considered the final word.”

He said he shared Dr. Price's concern about the potential misuse of fMRI, but added that potential advantages should not be overlooked, either. “This technology and any technology has the potential for abuse and for benefit,” he said. “But just as marketers may be interested in using fMRI to sell products, researchers could use the technology to determine which ads are manipulative and which are informative … so you can tell which ads are telling people to do things against their better judgment.”

Dr. Iacoboni, said that the experiment is a first step into using fMRI in a new field: politics. “The study was really an experiment within an experiment,” he said. “Can we actually use brain imaging to get some meaningful data out of these questions? The answer is a positive one.”

Beyond that, he said, subsequent research will likely become much more conclusive. “These are experiments with basically no background,” Dr. Iacoboni said. “In science, you do an experiment, you put forth your findings, and you have some interpretation. Some may be correct, some may be wrong, but other scientists are going to pick up your studies, replicate, and elaborate on them. Then you're going to build a body of literature that allows you to interpret your data in a much more reliable way.”

Back to Top | Article Outline


  • ✓ An editorial in the New York Times describing an experiment that used neuroimaging to evaluate voter preferences has provoked debate among neurologists about whether fMRI is a valid tool for assessing cognition.
© 2005 American Academy of Neurology