Many researchers have theorized that unpredictability--and its close relative, lack of control--are important ingredients of anxiety, and research with nonhuman animals has supported that notion. But there has been little experimentation in humans to bolster the idea, until recently.

New research finds that being able to predict when aversive, anxiety-provoking events will occur may make the anxiety more tolerable, particularly for women and people who are vulnerable to anxiety.

The study findings, coupled with further research on why predictability and control have such an important influence on anxiety, should guide clinicians in tailoring treatments to clients' specific needs, argue authors Carl Lejuez, PhD, and Michael Zvolensky of Brown University; Georg Eifert, PhD, of West Virginia University; and Jerry Richards, PhD, of the State University of New York­Buffalo.

"Although we're in the early stages of this path, I'm increasingly excited," says Zvolensky. "Each study brings us closer to understanding the principles enough to incorporate them in interventions down the road."

The research, published in the December issue of APA's Journal of Experimental Psychology: Applied (Vol. 6, No. 4), is the most recent in a series of studies that examines the role that predictability and control of aversive events play in panic and other anxiety disorders.

The trials

In the first phase of their experiment, Lejuez and his colleagues conducted 24 one-minute trials. They explained to participants that on some trials, air enriched with 20 percent carbon dioxide (CO2) would be administered to them through a breathing mask. The researchers used CO2 as an aversive stimulus--as they have done in previous experiments that examined the role of control in anxiety--because it produces many of the same psychological and physiological responses people experience during panic attacks, such as breathlessness, a feeling of tightness in the chest and sweaty palms.

For some experimental trials, participants were told, a tone would warn them if CO2 was coming, but for other trials, there would be no warning tone. The researchers administered CO2-enriched air to participants during six of the 24 trials. Before each trial began, a computer screen displayed either the letter T (indicating that if CO2-enriched air were to be administered on that trial, they would first hear a warning tone) or the letter N (indicating that there would be no tone given as a cue).

A second phase of the experiment was similar to the first, except that participants were allowed to choose, for each trial, whether CO2 administration would be predictable or unpredictable, by pressing the letter T or the letter N. In both phases of the experiment, participants rated how unpleasant each CO2 trial was and how anxious they had felt during the trial. Between phases, participants indicated their preference for either predictable or unpredictable CO2 administrations. Throughout the experiment, the researchers measured participants' skin conductance, a physiological measure of general anxiety, and breathing patterns. None of the participants had been diagnosed with panic disorder, but they varied in their baseline sensitivity to anxiety.

Predictable over unpredictable

As Lejuez and his colleagues expected, the physiological and self-report measures indicated that participants regarded the CO2 administrations as unpleasant and anxiety-provoking. Overall, participants expressed a preference that the CO2 administration be predictable, and when they were given the choice, they more often chose predictable over unpredictable CO2 administrations.

Predictability was especially important for participants whose baseline vulnerability to anxiety was high, the results showed. In addition, women were more likely than men to prefer and choose predictable over unpredictable CO2 administrations, even when baseline anxiety sensitivity was taken into account. In contrast, men who were less sensitive to anxiety were about equally likely to choose unpredictable or predictable CO2 administrations.

The new research is an important advance, says Northwestern University's Susan Mineka, PhD, a clinical psychologist who studies anxiety disorders.

"Although some researchers have picked up on the animal literature indicating that predictability and control are important elements of anxiety, there is still precious little research on this phenomenon in humans, and no experimental studies have shown the effect with such a naturalistic unconditioned stimulus [as CO2 administration]," she remarks.

Further, Mineka observes, "This is also the first study to examine how anxiety sensitivity--a known vulnerability for panic attacks--and gender moderate this preference."