Table 2. Main results – differences within conditions.

Outcome	K(total)	K(condition)	Trainingcondition	ES	95% CI	p
difference between positive and negativeinterpretation bias after training1	75	35	Benign	1.33	1.11; 1.55	<.001
		28	Negative	−0.05	−0.28; 0.17	.66
		6	Neutral	0.49	0.06; 0.93	.03
		6	No training	0.28	−0.16; 0.86	.72
change in positive interpretation biasfrom pre- to post training2	34	20	Benign	0.43	0.17; 0.69	<.01
		5	Negative	−0.22	−0.75; 0.32	.43
		6	Neutral	0.12	−0.33; 0.57	.59
		3	No training	0.32	−0.30; 0.93	.31
change in negative mood from pre- to posttraining3	90	47	Benign	0.25	0.14; 0.36	<.001
		25	Negative	−0.20	−0.35; −0.05	<.01
		14	Neutral	0.22	0.02; 0.41	.03
		4	No training	−0.03	−0.38; 0.33	.88
difference in negative mood from pre- to postemotional challenge4	35	18	Benign	−0.79	−1.04; −0.53	<.001
		10	Negative	−0.80	−1.10; −0.49	<.001
		6	Neutral	−1.03	−1.39; −0.67	<.001
		1	No training	−0.77	−1.49; −0.05	.04

Note. ES = effect size, CI = confidence interval, k = nr of independent samples; ¹positive values reflect higher positive than negative bias; ²positive values reflect increase in positive bias; ³positive values reflect decrease in negative mood; ⁴negative values reflect increase in negative mood.