The temperament T score is just an output of our 50-question NPA personality test. The T score could be used as an adjunct to launch some hypotheses. But the main issue with temperament is the necessity to devise a quantitative test to measure it, probably based on some aspect of neuroconduction or physiological response. Then, one would need to do genetic studies of temperament in the light of the various NPA personality types. The research community is still light years away from doing anything meaningful in this area, since both the notions of (1) the discrete (NPA) character traits, and (2) temperament as a variable genetic personality trait have yet to be stumbled upon.
Our initial guess is that temperament will turn out to be a complex polygenic entity, being dependent on 1) male-female gender, 2) the NPA genes themselves, 3) the genes determining A− trait, and 4) an assortment of separate genes that determine the general "activity" or "reactivity" of an individual.