Probing the puzzling workings of 'depressive realism'

Psychologists have thought for decades that depressed people tend to distort the facts and view their lives more negatively than do nondepressed people. Yet, psychological studies have consistently revealed a peculiar exception to that pattern: Depressed people, studies indicated, judge their control of events more accurately than do nondepressed people in a phenomenon that came to be known as "depressive realism."

Now two new studies published in February's Journal of Experimental Psychology: General (Vol. 134, No. 1) are starting to clear up the mystery. By refining an aspect of an experimental task that involved control over a light bulb, researchers uncovered a new twist--that nondepressed people may overestimate their control because they take more aspects of a situation into account in judging their control. The findings may help clinicians to refine therapies for depression.

"This was always a rather puzzling finding that did not sit well with contemporary understanding of depression," in which depression comes from distorted, inaccurate thoughts, says psychologist Chris Brewin, PhD, who studies cognitive theories of depression at University College London. He says researchers wondered how the same depressed person could be wrong about some things and right about others.

To find out, Rachel Msetfi, PhD, conducted a study as part of her doctoral research at the University of Hertfordshire in England with three psychologists: Hertfordshire colleague Diana Kornbrot, PhD, Robin Murphy, PhD, of University College London, and Jane Simpson, PhD, of the University of Lancaster. By introducing new conditions into the experimental paradigm commonly used to study depressive realism, the researchers found that apparent depressive realism may actually come from depressed people not using all the available evidence to judge the facts, relative to nondepressed people.

"This is a very well-conducted piece of research that undermines the evidence that the depressed may in some cases make sounder judgments than the nondepressed," says Brewin. But he notes that the data, while promising, will need further investigation and elaboration as psychologists revise their understanding of depression.

For better or for worse

Depressive realism has been viewed as the opposite of optimistic bias, itself a distortion of reality. In a seminal study by psychologists Lyn Abramson, PhD, Lauren Alloy, PhD, and others in 1979 in the Journal of Experimental Psychology: General (Vol. 108, No. 4, pages 441-485) nondepressed people were more likely than depressed people to think they controlled a light bulb when it came on at least three out of four times--even though they didn't have any objective control. Thus, it appeared that the depressed were more realistic about their degree of control--i.e., they were more likely to realize they had none.

To investigate why, Msetfi's team conducted two experiments in which they manipulated a new variable, the intertrial interval. Varying the length of the pause between task trials gave them a relatively simple way to manipulate an aspect of task context. The first experiment divided 128 people evenly among experimental groups of depressed or nondepressed women or men, matched as closely as possible on factors such as age, education, IQ and working memory. The researchers gauged participants' depressive symptoms through scores on the Beck Depression Inventory.

The participants performed a contingency-judgment task in which they pressed a button and saw a light bulb appear (or not) on a computer screen. After 40 trials, they judged how much control they'd had over the light bulb's appearance--much like judging how much flipping a switch turns on a light. Participants rated their control from zero (no control) to 100 (total control).

Msetfi and her team made the time between trials either short (three seconds) or long (15 seconds). The intertrial interval can, of course, be viewed as a time when nothing happens. But the authors proposed that something still might be happening during these intervals because participants are still exposed to the context, which could influence their judgment.

The long and short of it

After experimental sets that turned on the light 75 percent of the time (one of the conditions under which depressive realism appears), more nondepressed than depressed participants thought they controlled the light even though they didn't, replicating past research. However, the intertrial intervals mattered a lot. When the light bulb went on in the same proportion either way, nondepressed people thought they had significantly more control when they had long intervals. Depressed people thought they had the same amount of control no matter how long between trials.

The researchers ran a second experiment with 96 participants--this time leaving the light bulb always present on the screen. First, because a real light bulb would always be there, it made the experiment a bit more realistic. Second, by asking participants to imagine themselves to be scientists who were testing old, possibly unreliable equipment, the experimenters dampened participants' expectation of control.

As in the first experiment, when the bulb went on three-quarters of the time, nondepressed participants judged their control as significantly higher for the longer waits. Again, depressed participants showed no bias either way.

Msetfi and her colleagues suggest that intertrial length does not affect depressed people's sense of control perhaps because long waits make it harder for people prone to rumination to pay attention or because they do not process information about the context of the task adequately. Because depression causes problems with attention and concentration, depressed people may, consciously or not, be unable to take the context into account when judging control.

Further lab research could, Msetfi speculates, reveal whether depressed people have a fundamental problem in processing context. She wonders whether scientists could train these individuals to attend to relevant information.

Reconciling findings

The results help to fit depressive realism, once an apparent anomaly, into the cognitive-distortion model of depression, Baker says. If depressed people do indeed ignore relevant information, this inattention to reality fits clinical results and theories showing that depression influences cognitive activity and the ability to maintain attention--even in nonclinical populations like the one in this study.

Experts welcome the chance to clear up past confusion. "Msetfi and her colleagues have shown that depressive realism is potentially a consequence of differences in simple information processing and not other more complex processes such as the protection of self-esteem," says Andy Baker, PhD, a cognitive psychologist at McGill University who studies how people judge how events go together.

Thus, Msetfi says it could be useful to train patients to interpret situations in the wider context of all possible information that could be relevant to judgment.

Baker is more circumspect about applying the new findings. First, he notes that depressive realism has appeared only in conditions of high density (the bulb comes on a lot) and zero contingency (no matter what the person does).

"Thus there is no real generality to this phenomenon," he says.

Second, he notes that although participants in these studies "are undeniably sad and many of them are alienated, their level of functioning is quite high--they are generally not clinically depressed." Baker believes that studying this group may shed light on the mechanisms of clinical depression, but that calling them "depressed" obscures the fact that this research may or may not be relevant to the clinical population.

Lyn Abramson cautions, "Although the results of Msetfi et al are quite interesting, they do not explain away the phenomenon of depressive realism in [our] original experiment because [the original] depressed participants were doing what the experimenter asked them to do--figure out how much control they had during the experimental trials. Moreover, the Msetfi et al results don't explain why other factors such as whether an outcome is good or bad predict when depressive realism is observed."

Abramson hopes that this study inspires a second-generation wave of research on depressive realism.

"The implications of the depressive realism effect for explaining how cognitive therapy works remain to be explored," she says. "In the spirit of this study, it will be important to further explore the conditions under which depressed people are more accurate than nondepressed people and vice versa." Future work will be required to fully answer this complex question. In fact, in those rare cases when researchers have looked at prediction of meaningful life events rather than contingency detection tasks, some researchers have shown that depressed people are actually more optimistic than the nondepressed. In the original language of Alloy and Abramson, whether "the sadder really are wiser" depends on how one defines wisdom.