SCHEMATIC PROCESSING AND THE PREDICTION OF DEPRESSION FOLLOWING CHILDBIRTH

On the basis of performance in an incidental free recall task, a sample of 156 non-depressed pregnant women, all scoring below 10 on the Beck Depression Inventory (BDI), was divided into two groups. The first (S+) was hypothesized as susceptible to depression as a result of showing a bias favouring better recall of self-esteem-threatening (SET) words than emotionally positive words. The second group (S-) showed the reverse recall bias and was considered not to be susceptible to depression. Level of depression in the two groups was reassessed 3 months after delivery and the impact of recent life events on self-esteem was also assessed using a structured interview procedure. The S+ and S- groups were subdivided on the basis of high (HISE) and low (LISE) scores for the impact of life events on self-esteem. The four resulting groups S+/HISE, S+/LISE, S-/HISE and S-/LISE were compared in terms of BDI score by means of an ANOVA (Susceptibility × Impact on self-esteem × Occasion). A significant second-order interaction showed that level of depression increased in the susceptible group experiencing negative impact of life events on self-esteems (S+/LISE). Depression level declined in all other groups. The findings are considered to be consistent with Beck's cognitive theory of depression. © 1991.