Although there is plenty of

Although there is plenty of evidence in the literature that supports Cloninger\'s biosocial personality theory and makes clearly defined assumptions on the underlying neurotransmitter systems, there are also many negative findings, especially concerning genetic association studies (e.g. Munafò et al., 2009). Moreover, the psychometric properties of the TCI have been criticized (e.g. Farmer & Goldberg, 2008). It becomes more and more apparent that there is an ultimate need for replication studies that can be used for meta-analyses (Hirschhorn et al., 2002, Munafo and Flint, 2004). This in mind the aim of the present study was to investigate the theory driven hypothesis that genetic variation on the DBH gene is associated with individual differences in RD in two independent samples of healthy participants. Cloninger postulates low noradrenaline levels in subjects scoring high on RD (Cloninger, 1987b), the functional SNP rs1611115 (C-970T) on the DBH gene is related to alterations in DBH enzymatic activity (Punchaichira et al., 2016, Zabetian et al., 2001) and to drug addiction (Kosten et al., 2013, Xie et al., 2013). Given that rs1611115 (C-970T) and RD are both associated to addiction we hypothesized an effect of the candidate SNP on RD. Results show a significant association between rs1611115 (C-970T) and RD. The effect is stronger in the population based Sample2 in comparison to the data derived from Sample1 which is dominated by younger subjects. However, irrespective of these marked age differences between both samples the effect could be replicated. Results from both samples show that carriers of the TT ARRY-380 sale have significantly lower RD scores than carriers of at least one C allele (genotypes CC and CT). Zabetian et al. (2001) reported that rs1611115 could explain up to 50% of the variance in DBH activity. Two independent studies demonstrated an allelic load effect of rs1611115 on DBH activity, with highest enzymatic activity in carriers of the CC genotype, intermediate activity in heterozygous CT carriers and lowest activity in carriers homozygous for the T allele (Punchaichira et al., 2016, Zabetian et al., 2001). This finding suggests that CC carriers associated with high RD scores have high noradrenaline levels, because DBH metabolizes dopamine to noradrenaline. Therefore, the present study only supports Cloninger\'s general hypothesis that RD is related to noradrenaline but the directionality of the RD – noradrenaline association that high RD (C allele carriers) is related to low noradrenaline levels was not corroborated. A possible explanation is that DBH activity is likely to be dependent on the availability of its substrate, i.e. dopamine. Unfortunately, neither the existing functionality studies nor our study controlled for the availability of dopamine. The present study is a genetic association study using the classic candidate gene approach. This method has been repeatedly criticized in the past to produce highly unreliably findings (e.g. Flint & Munafò, 2013). The alternative would be the conduction of a genome wide association study (GWAS). GWAS is a completely a-theoretical and therefore explorative method, correlating millions of polymorphisms simultaneously with the phenotype under investigation. GWAS although being a great technical achievement also bears its disadvantages. The immense number of investigated polymorphisms demands extremely large samples, potential findings have again to be replicated in independent studies, and statistical findings have to be controlled for an alpha error inflation. Ironically most GWAS findings could not be replicated and sometimes the functionality of polymorphisms of replicated findings is unknown or does not make sense (for an overview see Reuter, Felten, & Montag, 2016). Therefore, the candidate gene approach and GWAS both have their legitimation. Luckily the present candidate gene approach study provides replication data and the sample size is rather large for studies in the field of genetic personality research. This will be even of advantage in the future: The larger the sample the more representative it is and this is a positive effect for future meta-analyses. Meta-analyses relying on data from numerous but underpowered samples could provide unreliable findings.