Table 2.

Results of the three-way ANOVA analysis evaluating the effects of serial reaction time task (SRTT) variants (limited vs. unlimited response time), studied group (schizophrenia, bipolar disorder, and healthy controls), and hand (left vs. right) on the indices of implicit motor learning and normalized medians of reaction times (mRT%) in blocks 2–5.

	Main effect for the SRTT variant		Main effect for the diagnosis		Main effect for the hand
	F (1, 181)	Effect size	F (2, 181)	Effect size	F (1, 181)	Effect size
Indices of implicit motor learning
Difference between mRT2% and mRT4%	1.231	0.004	15.108***	0.09	6.941**	0.015
Rebound	1.457	0.004	37.996***	0.182	1.509	0.004
Normalized medians of reaction times in blocks 2–5
mRT2%	8.940*	0.027	2.594	0.016	3.215	0.008
mRT3%	1.376	0.004	7.097*	0.04	0.340	< 0.001
mRT4%	1.448	0.005	12.208***	0.073	0.587	0.001
mRT5%	0.083	0.000276	2.414	0.016	0.112	< 0.001

Effect size (generalized eta squared) lower than 0.01 was counted as negligible, 0.01–0.06 as small, 0.06–0.14 as medium, and for >0.14 as large.

^*p ≤ 0.05,

^**p ≤ 0.01,

^***p ≤ 0.001.