Illustration: Chris Sharp

Briefing a research group from the U.S. military recently, Boston College psychology professor Lisa Feldman Barrett held up two photographs: one of Vice President Dick Cheney with his lip curled, and the other of 2004 Democratic presidential hopeful Howard Dean, his face livid. Barrett asked the 50 researchers, mostly civilians, what they thought the two men were feeling, and the consensus was that Cheney seemed contemptuous and Dean looked angry.

On both counts, the assumptions were dicey. Barrett point­ed out that in some cases, a lip curl has less to do with state of mind than with a person’s facial muscle tone. Dean, for his part, was probably feeling energized, rather than angry, as became evident when Barrett displayed the en­tire image—the Democratic can­didate with an audience of cheering supporters.

The researchers were reflecting standard wisdom: that we all recognize an emotion such as anger or sadness when we see it in someone’s face or gestures. Barrett believes otherwise—that we often misread how others feel, because human feelings are far more basic than the roster of emotions referenced by psychologists and poets.

That the conventional view could be faulty is of interest to the U.S. Army Research Institute for the Behavioral and Social Sciences, the group briefed by Barrett at the National Academy of Sciences, in Washington. “If you’re a soldier in Iraq, and you open a door,” says Barrett, you want to be able to tell whether the person you face is angry or afraid—and to know how far to trust what you think you know.

Barrett’s views are breeding “a good deal of agitation” among researchers, according to Gerald Clore, a University of Virginia psychologist who admires her work. In the conventional scientific view, emotions are distinct, with firm boundaries around them, and they are hardwired into the brain or body: A scowl will signal anger, which is an emotion branded in human nature, innately recognizable to other humans. However, it remains a “dirty little secret” of emotions research that little of this bears out, Barrett says. Clinical experiments that have looked to link particular emotions to physical manifestations, including changes in the peripheral nervous system, have yielded weak or inconclusive results. Likewise, scientists using imaging technology have searched without success for parts of the brain that might house emotions, Barrett says.

Barrett doesn’t doubt that people have deep feelings, which they name. What she questions is “the idea that there are bounded and distinct kinds of emotion,” and that these feelings “constitute the building blocks of emotional life,” as she wrote in a paper published last year in Perspectives on Psychological Science.

Barrett is advancing her critique with nine research projects under way in her McGuinn Hall laboratory. In one study, subjects listen to music that sounds unpleasant—not dark, or sad, just adverse. In response, their feelings about the music are simply negative, not articulated as a specific emotion. That is, unless (through laboratory techniques) the subjects have been primed with the concept of an emotion like fear and induced into such a mood. In that case, the music evokes fear.

This research illustrates Barrett’s theory that emotions are a human invention. Hard­wired in the brain, she suspects, is a primitive dimension—what she calls “affect,” a spectrum of feelings from pleasant to un­pleasant, from positive to negative. Basically, human beings contrive distinct emotions from this primordial field as it is stirred by specific events in their lives.

Virginia’s Clore says one implication of Barrett’s research is that there are an “infinite number of emotions, limited only by the kinds of situations [stimuli] that can be good or bad in some way,” a possibility broached in 1890 by pioneering psychologist-philosopher William James (whose name graces a prestigious annual lecture sponsored by the Association for Psychological Science and delivered last October by Barrett). Clore likens Barrett’s studies to those in the 1960s by the noted psychologist Walter Mischel, who discomfited many colleagues with his finding that personality is not constant but varies by situation. According to Clore, Barrett’s work could have a similarly “upsetting but ultimately cleansing effect” on the study of emotions.

Read more by William Bole