FIGURE 3.

Relationship between the genotype groups and the D scores obtained from the mIAT. The one-way ANOVA yielded a statistically significant difference between genotype group mean D scores [F(2,111) = 3.432, p = 0.036]. Post hoc analyses revealed that D scores were significantly lower in those homozygous for the S allele (0.68 ± 0.042, mean ± SE, p = 0.024) when compared to those homozygous for the L allele (0.91 ± 0.071, mean ± SE, p = 0.024). There was no statistically significant difference, however, between the heterozygous (L/S)(0.80 ± 0.058, mean ± SE) and the homozygous (S/S and L/L) groups of alleles (p = 0.069 and p = 0.318, respectively). The subsequent ANCOVA showed that the difference between genotype group mean D scores remained statistically significant also when controlling by age and gender [F(2,109) = 3.242, p = 0.043, ηp² = 0.056]. Post hoc analyses further underlined the statistically significant difference between the lower mean D scores among those homozygous for the S allele and the higher D scores among those homozygous for the L allele (p = 0.024), whereas the difference in means between those with heterozygous alleles (L/S) and those with homozygous alleles (S/S and L/L) remained non-significant (p = 0.102 and p = 0.276, respectively). Likewise, the linear regression analysis exhibited a statistically significant linear association between genotype group (number of risk allele “S”) and mIAT D scores [F(3,110) = 2.514, p = 0.012, R² = 0.064], and controlling for age and gender. Accordingly, there is a decrease of 0.117 points in the participants’ D scores as a function of a one increase in S alleles being carried.