Psychological studies in bipolar affective disorder and analogue conditions suggest that mania may be the product of an abnormal defence against depression. In this study, currently manic bipolar individuals, currently depressed bipolar individuals, and normal controls were assessed using explicit and implicit measures of attributional style, an emotional Stroop test with euphoria-related and depression-related words and a recall measure of the selfschema. Manic individuals showed a normal self-serving bias on a version of the explicit attributional style questionnaire, attributing positive events more than negative events to self, in contrast to bipolar-depressed individuals who attributed negative events more than positive events to self. However, on an implicit test of attributional style, both manic and bipolar-depressed individuals attributed negative events more than positive events to self. Both bipolar-manic and bipolar-depressed individuals demonstrated slowed colour naming for depression-related but not euphoriarelated words on an emotional Stroop test. Manic individuals, like normal controls, endorsed primarily positive words as true to self on a self referent questionnaire, but like bipolar-depressed individuals, recalled primarily negative words in a surprise recall test afterwards. Findings from the implicit tests therefore indicate a common form of psychological organisation in manic and depressed individuals, whereas the contrasts between the scores on the implicit and explicit measures are in accord with the hypothesis of a manic-defence. Future avenues for research and implications for treatment are discussed.

Objective: The present studies aimed to examine temporal patterns of symptom change over the course of cognitive behavioural therapy (CBT) in individuals with treatment-resistant depression (TRD) and evaluate their relationship to outcomes. They further sought to investigate whether case-conceptualisation competence, client hope and processing were associated with therapeutic change. Method: Participants were 156 individuals with TRD receiving 12-18 sessions of CBT as part of a randomised controlled trial. Depressive severity was assessed at each session. Audio-recordings of therapy sessions proximal to sudden gains and control sessions for 25 sudden gainers and 25 non-sudden gainers were rated for client hope and emotional processing, and therapist competence in case-conceptualisation. Results: The overall shape of change was best described as cubic, with frequent discontinuities in symptom trajectories. Sudden gains were associated with reduced depressive severity and a greater likelihood of remission at 12 months. Sudden gainers demonstrated greater hope than non-sudden gainers and emotional processing increased prior to a gain. Therapists showed greater competence in case-conceptualisation with sudden gainers. Conclusion: The present study describes a non-linear shape of change in TRD and extends the phenomenon of sudden gains to this population. It suggests that the combination of hope for change with supported efforts to process and make meaning of experiences may predispose individuals favourably towards experiencing sudden gains in therapy.