Little evidence for a reduced late positive potential to unpleasant stimuli in major depressive disorder

Background: Individuals with major depressive disorder (MDD) present with deficits in emotional reactivity. Conflicting models have been proposed to explain this effect. We sought to confirm reports of blunted reactivity to negatively-valanced emotional stimuli, in line with the emotional context insensitivity hypothesis of depression, in a preregistered study. Methods: Forty-one depressed participants and 41 age- and gender-matched healthy controls were presented a series of unpleasant and neutrally-valanced pictures in a passive view paradigm while acquiring electroencephalography (EEG). The late positive potential (LPP), an EEG correlate of emotional reactivity, was compared between groups using mixed-effects repeated-measures models and exploratory cluster-based permutation tests. A sensitivity analysis was performed to assess the robustness of LPP findings by reanalysing the LPPs using 24 EEG pipelines from studies identified in the literature. Results: We found no difference in LPP amplitudes between MDD and healthy individuals using the preregistered analysis pipeline. The sensitivity analysis revealed that the magnitude and direction of LPP effect sizes were affected by the analysis pipeline. Exploratory permutation analysis revealed an electrode cluster that showed a significant reduction in the LPP for MDD participants while viewing unpleasant pictures. Conclusions: These results do not provide evidence in support of the emotional context insensitivity hypothesis, except for the exploratory data-driven approach. Methodological differences, in particular in the analysis pipeline, contribute to the heterogeneity of LPP modulation in depression. A standardised approach to quantify EEG correlates of emotional reactivity is needed to evaluate alternative models of emotional reactivity in depression.