Results of ANCOVA models examining group differences in model parameters, when accounting for main effects and interactions with age and sex
| Predictor* | Test, p value | B (95% CI), p value | Post hoc contrasts* |
|---|---|---|---|
| Emotion conflict | |||
| Age | F1, 471 = 5.02, p = 0.026 | 0 (−0.01 to 0.02), p = 0.569 | NS |
| Sex | F1, 471 = 2.41, p = 0.121 | −0.23 (−0.39 to −0.08), p = 0.002 | NS |
| Group | F1, 471 = 24.83, p < 0.001 | D/A: −0.17 (−0.36 to 0.03), p = 0.091 SUD: −0.53 (−0.72 to −0.34), p < 0.001 | HC–D/A: 0.22, t471 = 2.33, p = 0.020 HC–SUD: 0.64, t471 = 6.75, p < 0.001 D/A–SUD: 0.42, t471 = 5.52, p < 0.001 |
| Age × group | F2, 471 = 1.38, p = 0.253 | Age × D/A: 0 (−0.02 to 0.02), p = 0.916 Age × SUD: 0.01 (−0.01 to 0.03), p = 0.153 | NS |
| Sex × group | F2, 471 = 5.55, p = 0.004 | Sex × D/A: 0.14 (−0.05 to 0.33), p = 0.144 Sex × SUD: 0.32 (0.13 to 0.51), p = 0.001 | Male HC–D/A: 0.02, t471 = 0.14, p = 0.887 HC–SUD: 0.21, t471 = 1.40, p = 0.163 D/A–SUD: 0.19, t471 = 1.36, p = 0.174 Female HC–D/A: 0.31, t471 = 2.74, p = 0.007 HC–SUD: 0.85, t471 = 7.03, p < 0.001 D/A–SUD: 0.54, t471 = 5.77, p < 0.001 |
| Decision uncertainty | |||
| Age | F1, 471 = 32.56, p < 0.001 | 0.01 (−0.01 to 0.03), p = 0.27 | NS |
| Sex | F1, 471 = 3.28, p = 0.071 | −0.1 (−0.3 to 0.09), p = 0.301 | NS |
| Group | F1, 471 = 3.69, p = 0.026 | D/A: −0.02 (−0.27 to 0.23), p = 0.871 SUD: 0.26 (0.01 to 0.51), p = 0.042 | HC–D/A: 0.02, t471 = 0.15, p = 0.884 HC–SUD: −0.24, t471 = −1.94, p = 0.053 D/A–SUD: −0.26, t471 = −2.57, p = 0.011 |
| Age × group | F2, 471 = 3.71, p = 0.025 | Age × D/A: 0.02 (−0.01 to 0.04), p = 0.155 Age × SUD: 0.03 (0.01 to 0.06), p = 0.007 | HC–D/A: −0.02, t471 = −1.42, p = 0.155 HC–SUD: −0.04, t471 = −2.72, p = 0.007 D/A–SUD: −0.02, t471 = −1.66, p = 0.097 |
| Sex × group | F2, 471 = 0.17, p = 0.847 | Sex × D/A: −0.01 (−0.26 to 0.24), p = 0.941 Sex × SUD: 0.05 (−0.2 to 0.3), p = 0.688 | NS |
ANCOVA = analysis of covariance; CI = confidence interval; D/A = depression/anxiety disorder group; HC = healthy control group; NS = nonsignificant; SUD = substance use disorder group.
↵* For interpretability, continuous predictors were centred, sum coding was used for sex (female = −1; male = 1), and treatment coding was used for group (with healthy controls as the reference group, coded as 0).