Table 1.

Numbers of participants included in the analyses, broken down by age, gender and experimental tasks. We aimed for a sample balanced across ages, and within each age group, a balanced gender distribution. Note that for some analyses the number of included participants was reduced because of missing outcome variables (see §§2c, 2d and 2e).

age	total	male	female	figure 1		figure 2	figure 3		figure 4
				rule following		belief formation	prosociality		factor S
				bad example	good example		selfish example	fair example
10	16	10	6	9	7	16	15	8	15
11	21	7	14	10	11	21	15	12	19
12	18	10	8	10	8	18	13	12	17
13	9	4	5	6	3	9	5	5	9
14	13	10	3	7	6	13	8	8	11
15	13	9	4	8	5	13	7	11	12
16	13	7	6	7	6	13	5	10	12
17	12	7	5	7	5	12	8	7	12
18	9	5	4	4	5	9	5	8	9
19	12	7	5	3	9	12	11	7	12
20	10	5	5	6	4	10	4	7	10
sum	146	81	65	77	69	146	96	95	138