Estment was larger (0.5 times for Study 2 and 0.4 times for Study 1). Fourth, we asked the participants to register C0 decisions, which we failed to collect in Study 1. Fifth, there was no showup fee for Study 2. The second and third changes were intended to make it easier for participants to understand the game structure. Registered responses were randomly grouped and game payoffs computed. The results were sent to the participants via postal mail. By the same mail, participants were asked to send back their bank account information, so that payoffs could be transferred to them (1 point = 20). All the procedures were explained before participants logged in to the response webpage. In preparation for the experiment with twin participants, we conducted a preliminary experiment with undergraduates (n = 37; Hiraishi, purchase PBTZ 169 unpublished data). The results were generally consistent with Fischbacher et al. (2001) study; we observed the two major strategies of conditional cooperation (n = 17) and free riding (n = 16). The experimental procedures were approved by the ethics committee at the Faculty of Letters, Keio OPC 8212 site University.TABLE 4 | Mean contributions in Study 2. Study 2 C0 C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 UC2 LC2 (C0 6) MC2 (C7 13) HC2 (C14 20) M 1.00 1.67 2.19 2.53 3.01 3.63 4.15 4.52 5.06 5.57 6.22 6.84 7.26 7.71 7.89 8.43 8.80 8.89 9.50 9.74 9.98 7.03 2.60 6.17 9.03 SD 3.37 3.64 3.70 3.55 3.75 4.04 4.23 4.48 4.79 5.08 5.21 5.59 5.84 6.18 6.49 6.98 7.37 7.76 8.13 8.58 9.14 6.21 3.33 4.98 7.UC, unconditional; LC, lowest conditional; MC, medium C; HC, highest C scores in Study 2.same applied to SD as well (b = 0.982, p < 0.001; R2 = 0.963, p < 0.001). While the majority of participants (n = 123, 43.6 ) adopted a conditional cooperation strategy, 70 participants (24.8 ) adopted a free rider strategy, contributing zero points through C0 20 decisions. Because we had 21 conditional decision scores from each participant (C0 20), we computed three conditional decision scores. They represented low contribution in Study 2 (LC2 scores; average of C0 6 decisions), medium contribution in Study 2 (MC2) scores (average of C7 13), and high contribution in Study 2 (HC2) scores (average of C14 20). As there were no significant differences between MLC (C6 10) scores and MHC scores (C11 15) in Study 1, we decided to merge the MLC and MHC categories to obtain three, rather than four, overall scores.Comparison of Repeaters, Non-Repeaters, and First-ComersSeventy-three participants were repeaters from Study 1. We compared the repeaters' decisions in Study 1 with those of nonrepeaters (those who participated only in Study 1). Repeaters had significantly lower LC and MLC scores (LC score, Wilcoxon test, W = 9582.5, p < 0.05; MLC score, W = 9536.5, p < 0.05). There were no significant differences in UC, MHC, and HC scores for repeaters and non-repeaters. Next, we compared repeaters (n = 73) and newcomers (n = 209) on their decisions in StudyResults Simple StatisticsWe found that as the contribution by others increased, both the mean contribution decisions and the variances of the conditional decisions increased (Table 4). Regression of mean contribution decisions on others' contribution showed significant positive relationship (b = 0.996, p < 0.001; R2 = 0.992, p < 0.001). TheFrontiers in Psychology | www.frontiersin.orgApril 2015 | Volume 6 | ArticleHiraishi et al.Heritability of cooperative behavior2. Repeaters contributed signific.Estment was larger (0.5 times for Study 2 and 0.4 times for Study 1). Fourth, we asked the participants to register C0 decisions, which we failed to collect in Study 1. Fifth, there was no showup fee for Study 2. The second and third changes were intended to make it easier for participants to understand the game structure. Registered responses were randomly grouped and game payoffs computed. The results were sent to the participants via postal mail. By the same mail, participants were asked to send back their bank account information, so that payoffs could be transferred to them (1 point = 20). All the procedures were explained before participants logged in to the response webpage. In preparation for the experiment with twin participants, we conducted a preliminary experiment with undergraduates (n = 37; Hiraishi, unpublished data). The results were generally consistent with Fischbacher et al. (2001) study; we observed the two major strategies of conditional cooperation (n = 17) and free riding (n = 16). The experimental procedures were approved by the ethics committee at the Faculty of Letters, Keio University.TABLE 4 | Mean contributions in Study 2. Study 2 C0 C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 UC2 LC2 (C0 6) MC2 (C7 13) HC2 (C14 20) M 1.00 1.67 2.19 2.53 3.01 3.63 4.15 4.52 5.06 5.57 6.22 6.84 7.26 7.71 7.89 8.43 8.80 8.89 9.50 9.74 9.98 7.03 2.60 6.17 9.03 SD 3.37 3.64 3.70 3.55 3.75 4.04 4.23 4.48 4.79 5.08 5.21 5.59 5.84 6.18 6.49 6.98 7.37 7.76 8.13 8.58 9.14 6.21 3.33 4.98 7.UC, unconditional; LC, lowest conditional; MC, medium C; HC, highest C scores in Study 2.same applied to SD as well (b = 0.982, p < 0.001; R2 = 0.963, p < 0.001). While the majority of participants (n = 123, 43.6 ) adopted a conditional cooperation strategy, 70 participants (24.8 ) adopted a free rider strategy, contributing zero points through C0 20 decisions. Because we had 21 conditional decision scores from each participant (C0 20), we computed three conditional decision scores. They represented low contribution in Study 2 (LC2 scores; average of C0 6 decisions), medium contribution in Study 2 (MC2) scores (average of C7 13), and high contribution in Study 2 (HC2) scores (average of C14 20). As there were no significant differences between MLC (C6 10) scores and MHC scores (C11 15) in Study 1, we decided to merge the MLC and MHC categories to obtain three, rather than four, overall scores.Comparison of Repeaters, Non-Repeaters, and First-ComersSeventy-three participants were repeaters from Study 1. We compared the repeaters' decisions in Study 1 with those of nonrepeaters (those who participated only in Study 1). Repeaters had significantly lower LC and MLC scores (LC score, Wilcoxon test, W = 9582.5, p < 0.05; MLC score, W = 9536.5, p < 0.05). There were no significant differences in UC, MHC, and HC scores for repeaters and non-repeaters. Next, we compared repeaters (n = 73) and newcomers (n = 209) on their decisions in StudyResults Simple StatisticsWe found that as the contribution by others increased, both the mean contribution decisions and the variances of the conditional decisions increased (Table 4). Regression of mean contribution decisions on others' contribution showed significant positive relationship (b = 0.996, p < 0.001; R2 = 0.992, p < 0.001). TheFrontiers in Psychology | www.frontiersin.orgApril 2015 | Volume 6 | ArticleHiraishi et al.Heritability of cooperative behavior2. Repeaters contributed signific.