Farmer et al (Reference Farmer, Redman and Harris2002) examined whether self-reports of neuroticism and extraversion represent enduring vulnerabilities to depression. While their findings make a valuable contribution, their conclusions regarding neuroticism warrant additional consideration. They found variables reflecting past and current depression were significant predictors of neuroticism and interpreted this as inconsistent with their hypothesis that neuroticism would ‘exhibit trait-like qualities’ and would not be ‘substantially influenced by alteration in mood-state’ (p. 118). Farmer et al's findings did not directly address this hypothesis because they did not use a longitudinal design needed to observe fluctuations in mood-state. None the less, several longitudinal studies indicate that neuroticism is affective-state dependent. For example, Hirschfeld et al (Reference Hirschfeld, Klerman and Clayton1983) found that patients in remission from depression reported lower levels of neuroticism than they originally reported when depressed. Findings of this nature have been used to argue that neuroticism and related traits are only contaminants of depression. More recently, investigators (e.g. Reference Santor, Bagby and JoffeSantor et al, 1997) studying samples of patients have noted that absolute changes in depression-related traits are associated with changes in mood (i.e. affective-state dependent), but that there is also a consistency in the rank of patients with regard to their scores on these measures (i.e. relative stability). Findings such as these have been interpreted as indicating that depression-related personality traits have both state-like and trait-like properties.

Given the large association between neuroticism and depression, Farmer et al suggested that neuroticism may be largely ‘a proxy measure for present or past depression’ (p. 121) and questioned whether neuroticism reflects a vulnerability for depression. Neuroticism refers to a tendency to experience negative affect, so this high degree of overlap is not surprising. More importantly, longitudinal studies (Reference Hirschfeld, Klerman and LavoriHirschfeld et al, 1989; Reference Krueger, Caspi and MoffittKrueger et al, 1996) have found that high premorbid neuroticism is positively associated with the development of depression.

In summary, Farmer et al's conclusion that neuroticism does not measure a vulnerability to depression and primarily reflects symptoms of depression is not warranted. Self-reports of neuroticism prospectively predict depression. Longitudinal studies support Farmer et al's conclusion that neuroticism is strongly associated with a person's current affective state. However, such studies also suggest that neuroticism is likely to have trait-like properties in addition to the state-like properties noted by Farmer et al.