A large response range can be observed in both behavioral and neuroendocrine responses to environmental challenges. This variation can arise from central mechanisms such as those involved in the shaping of general response tendencies (temperaments) or involves only one or the other output system (behavioral vs. endocrine response). The participation of genetic factors in this variability is demonstrated by family and twin studies in humans, the comparison of inbred strains and selection experiments in animals. Those inbred strains diverging for specific traits of stress reactivity are invaluable tools for the study of the molecular bases of this genetic variability. Until recently, it was only possible to study biological differences between contrasting strains, such as neurotransmitter pathways in the brain or hormone receptor properties, in order to suggest structural differences in candidate genes. The increase of the power of molecular biology tools allows the systematic screening of significant genes for the search of molecular variants. More recently, it was possible to search for genes without any preliminary functional hypothesis (mRNA differential expression, nucleic acid arrays, QTL search). The approach known as quantitative trait loci (QTL) analysis is based on the association between polymorphic anonymous markers and the phenotypical value of the trait under study in a segregating population (such as F2 or backcross). It allows the location of chromosomal regions involved in trait variability and ultimately the identification of the mutated gene(s). Therefore, in a first step, those studies skip the 'black box' of intermediate mechanisms, but the knowledge of the gene(s) responsible for trait variability will point out to the pathway responsible for the phenotypical differences. Since variations in stress-related responses may be related to numerous pathological conditions such as behavioral and mood disorders, drug abuse, cardiovascular diseases or obesity, and production traits in farm animals, these studies can be expected to bring significant knowledge for new therapeutic approaches in humans and improved efficiency of selection in farm animals.