Depressive realism is the proposition that people with depression actually have a more accurate perception of reality, specifically that they are less affected by positive illusions of illusory superiority, the illusion of control and optimism bias. The concept refers to people with borderline or moderate depression, suggesting that while non-depressed people see things in an overly positive light and severely depressed people see things in overly negative light, the mildly discontented grey area in between in fact reflects the most accurate perception of reality.
Studies by psychologists Alloy and Abramson (1979) and Dobson and Franche (1989) suggested that depressed people appear to have a more realistic perception of their importance, reputation, locus of control, and abilities than those who are not depressed.
People without depression may be more likely to have inflated self-images and look at the world through “rose-colored glasses”, thanks to cognitive dissonance elimination and a variety of other defense mechanisms.
This does not necessarily imply that a specific happy person is delusional nor deny that some depressed individuals may be unrealistically negative (as in studies by Pacini, Muir and Epstein, 1998).
Some recent studies argue the contrary to the hypothesis, suggesting that mentally healthy people actually have fewer positive illusions and illusions in general than depressed ones. For example, studies by Msetfi et al. (2005, 2007) found that when replicating Alloy and Abramson’s findings the overestimation of control in nondepressed people only showed up when the interval was long enough, implying that this is because they take more aspects of a situation into account than their depressed counterparts, and other studies such as Joiner et al. (2006) or Moore et al. (2007) found that all forms of illusion, positive or not, were associated with higher depressive symptoms. It might also be that the pessimistic bias of depressives results in “depressive realism” when, for example, measuring estimation of control when there is none, as proposed by Allan et al. (2007). Various other recent studies such as Fu et al.(2003), Carsona et al.(2009) and Boyd-Wilson et al. (2000) reject the idea of depressive realism by showing no link between positive illusions and mental health, well-being or life satisfaction maintaining that accurate perception of reality is compatible with happiness.
A longitudinal study (Colvin et al. 1995) found that self-enhancement biases were associated with poor social skills and psychological maladjustment. In a separate experiment where videotaped conversations between men and women were rated by independent observers, self-enhancing individuals were more likely to show socially problematic behaviors such as hostility or irritability. A 2007 study (Sedikides et al.) found that self-enhancement biases were associated with psychological benefits (such as subjective well-being) but also inter- and intra-personal costs (such as anti-social behavior).
When studying the link between self-esteem and positive illusions, Compton (1992) identified a group which possessed high self-esteem without positive illusions, and that these individuals weren’t depressed, neurotic, psychotic, maladjusted nor personality disordered, thus concluding that positive illusions aren’t necessary for high self-esteem. Compared to the group with positive illusions and high self-esteem, the nonillusional group with high self-esteem was higher on self-criticism and personality integration and lower on psychoticism.
A meta-analysis of 118 studies including 7013 subjects by Moore et al. (2007) found that slightly more studies supported the depressive realism hypothesis (Cohen’s d = -.24, SD = .72). Studies that were more generalizable were more likely to produce depressive realism effects, but both depressed and nondepressed participants were found to be strongly positively biased, which does not go in line with the hypothesis. Studies that used self-reports instead of clinical interviews found the effects more strongly, and studies using student samples found mild depressive realism effects whereas clinical samples found results contrary to depressive realism. There also was a significant moderation to the effect by the method which was used to measure depressive realism, as studies of attentional bias supported depressive realism the most, followed by judgement of contingency studies, whereas recall of feedback and evaluation of performance studies produced results mildly counter to depressive realism. Study quality was also a significant factor in moderating the effect, as studies lower in methodological quality supported depressive realism more.
Since there is evidence that positive illusions may be more common in normally mentally healthy individuals than in depressed individuals, Taylor and Brown (1988) argue that they are adaptive.
However, Pacini, Muir and Epstein (1998) have shown that the depressive realism effect may be because depressed people overcompensate for a tendency toward maladaptive intuitive processing by exercising excessive rational control in trivial situations, and note that the difference with non-depressed people disappears in more consequential circumstances.
Knee and Zuckerman (1998) have challenged the definition of mental health used by Taylor and Brown and argue that lack of illusions is associated with a non-defensive personality oriented towards growth and learning and with low ego involvement in outcomes. They present evidence that self-determined individuals are less prone to these illusions.
Colvin and Block (1995) have criticised the logic and empirical evidence used by Taylor and Brown to link mental health and positive illusions together. They argue that there are fair number of studies that fails to replicate the findings, and that a closer examination of their evidence doesn’t support their thesis.
Dykman et al. (1989) argue that, although depressive people make more accurate judgments about having no control in situations where in fact they have no control, they also believe they have no control when in fact they do; and so their perceptions are not more accurate overall.
Dunning and Story (1991) have shown that when applied to real world settings, depressed individuals are less accurate in their predictions about the future and more overconfident than their nondepressed counterparts.
The French philosopher Voltaire’s classic 1759 novella Candide: Or, Optimism deals with this subject and can be considered an early exploration of this psychological phenomenon. The story is an attack on Leibniz’s optimistic theory that ours is the greatest of all possible worlds, a philosophy that is espoused by the character of Professor Pangloss even though the events around him are presented as unambiguously awful. Much of the humour in the story comes from Pangloss’s rationalizations of these miserable and cataclysmic events as he will not admit that even the worst forms of individual human suffering are not all for the best. His position is counterpointed later in the book by the character of Martin, a more depressive character whose pessimistic philosophy may not be any better for getting along with life, but his viewpoint is certainly the least deluded as to the reality of the world around him. Candide’s own conclusion on the subject can be summed up in his utterance that “Optimism is the madness of insisting that all is well when we are miserable.”