Self-referential processing in depressed adolescents: A high-density event-related potential study

Abstract

Despite the alarming increase in the prevalence of depression during adolescence, particularly among female adolescents, the pathophysiology of depression in adolescents remains largely unknown. Event-related potentials (ERPs) provide an ideal approach to investigate cognitive-affective processes associated with depression in adolescents, especially in the context of negative self-referential processing biases. In this study, healthy (n = 30) and depressed (n = 22) female adolescents completed a self-referential encoding task while ERP data were recorded. To examine cognitive-affective processes associated with self-referential processing, P1, P2, and late positive potential (LPP) responses to negative and positive words were investigated, and intracortical sources of scalp effects were probed using low-resolution electromagnetic tomography (LORETA). Additionally, we tested whether key cognitive processes (e.g., maladaptive self-view, self-criticism) previously implicated in depression related to ERP components. Relative to healthy female subjects, depressed females endorsed more negative and fewer positive words, and free recalled and recognized fewer positive words. With respect to ERPs, compared with healthy female adolescents, depressed adolescents exhibited greater P1 amplitudes following negative words, which was associated with a more maladaptive self-view and self-criticism. In both early and late LPP responses, depressed females showed greater activity following negative versus positive words, whereas healthy females demonstrated the opposite pattern. For both P1 and LPP, LORETA revealed reduced inferior frontal gyrus activity in response to negative words in depressed versus healthy female adolescents. Collectively, these findings suggest that the P1 and LPP reflect biased self-referential processing in female adolescents with depression. Potential treatment implications are discussed. (PsycINFO Database Record

(c) 2015 APA, all rights reserved).

Figures

Figure 1. Self-Referential Encoding Task: Endorsed, Reaction Time, Recall, and Recognition

Note. Self-referential encoding task data for HC (n = 28) and MDD (n = 22) pertaining to (A) words endorsed, (B) Reaction Time (ms) (HC=17; MDD=21); (C) Recall, and (D) Recognition; Data reflect between-group differences for *p < .05 and ***p < .001.

Figure 2. P1, P2, and Early LPP Activity for Healthy (n = 28) and Depressed (n = 22) Female Adolescents in Response to Positive and Negative Words

Note. P1 (100–200 ms), P2 (200–300 ms), and early LPP (400–600 ms) averaged across electrode sites Pz, POz, P1, P2, PO3, PO4 for: (A) Healthy (n = 28) and (B) Depressed (n = 22) Adolescents.

Figure 3. Associations among P1 Activity, Negative Self-View, and Self-Criticism Following Negative Words

Note. Correlations for Negative Word Condition among (A) P1 Mean Area and Negative View of the Self (Lower Scores = More Maladaptive Views of the Self; r = −.54, p < .001); (B) P1 Mean Area and Self-Criticism (r = .35, p = .014).

Figure 4. Late LPP Activity (600–1200 ms) for Healthy (n = 28) and Depressed (n = 22) Adolescents in Response to Positive and Negative Words

Note. Late LPP (600–1200 ms post-stimulus) activity averaged across electrode sites Fz, FCz, Cz for (A) Healthy (n=28) and (B) MDD (n=22) adolescents.

Figure 5. LORETA Contrasts for Healthy (n = 28) versus Depressed (n = 22) Adolescents Following Negative Words

Note. Results of independent t-tests contrasting current density for healthy versus depressed adolescents following negative words in the Self-Referential Encoding Task (Blue: HC > MDD; Red: MDD > HC). Statistical maps are thresholded at p < .01 (minimum cluster size: 5 voxels) displayed on the MNI template: (A) Medial Temporal Gyrus (100 – 200 ms poststimulus), (B) Inferior Frontal Gyrus (100 – 200 ms poststimulus), (C) Precentral Gyrus (100 – 200 ms poststimulus), (D) Inferior Frontal Gyrus and Middle Temporal Gyrus (400 – 600 ms poststimulus).