The Role of Rumination in the Relationship Between Postnatal Depressive Symptoms and Maternal Attunement

Abstract

The aim of this thesis was to increase understanding of how rumination, defined as the behaviours and thoughts that focus an individual’s attention on their depressive symptoms and on the implications of these symptoms (Nolen-Hoeksema, 1991), affects the relationship between postnatal depressive symptoms and maternal sensitivity. Study 1 examined the impact of self-reported maternal rumination on perceived maternal attunement and mood, and the role of perceived social support and infant temperament in this relationship in a community sample of mothers (N = 203). Rumination mediated the relationship between maternal depressive symptoms and maternal responsiveness when infant negative affect was low but not high. Contrary to predictions, rumination did not mediate the relationship between social support and maternal attunement. Study 2 incorporated a second assessment point approximately six months later with the same sample. Prospective analyses were undertaken to examine the directional relationship between rumination and maternal attunement in the context of depressive symptoms. Unexpectedly, analyses revealed that maternal bonding prospectively predicted rumination at six months, after controlling for rumination at baseline. The converse relationship was not significant. This was contrary to the thesis hypothesis that increased rumination would predict impaired attunement at six months. Partially consistent with the thesis hypotheses, the relationship between rumination and maternal attunement was moderated by depressive symptoms at baseline; such that mothers who were low in depressive symptoms and had lower bonding at baseline reported higher levels of rumination at 6 months. Study 3 also explored the directional relationships between maternal mood, rumination and maternal attunement at a state level in a daily diary study with a community sample of mothers (N = 94) with infants aged between 3 and 14 months. Consistent with the findings of study 2, state maternal bonding at Time 1 predicted both state maternal rumination and state maternal mood at Time 2, and state rumination at Time 2 mediated the relationship between state bonding at Time 1 and state mood at Time 2. In the final study, the causal relationships between an experimentally induced state of ruminative thinking and observed maternal behaviours in a mother-infant interaction task were examined in a sample of dysphoric and non-dysphoric mothers (N = 79) and their infants. The analyses examined change in mother-infant interaction quality from baseline to post rumination induction, and subsequent change following an infant stressor task. Findings revealed a significant reduction in maternal sensitivity and mother-infant dyadic synchrony in the rumination group, but not the control group. For maternal sensitivity, the effect of rumination was exacerbated following the stressor task. Contrary to predictions, this relationship was not moderated by dysphoric symptoms. The findings of this thesis indicate that ruminative thinking directly impairs observed maternal behaviours, but that perceived poorer maternal attunement also increases self-reported ruminative thinking. The significance of these findings for theoretical explanations of rumination in a postnatal context are considered, and the clinical implications for parenting programmes and interventions for both mothers in the community as well as those considered at risk are discussed.