COMPLEX TRAIT TREATMENT INTERACTION ANALYSIS - A POWERFUL APPROACH FOR ANALYZING INDIVIDUAL-DIFFERENCES IN EXPERIMENTAL-DESIGNS

Complex Trait-Treatment-Interaction (CTTI) analysis is introduced as a new and highly efficient statistical approach for analysing individua l differences in experimental designs. CTTI analysis enables the resea rcher to investigate the combined effects of several personality trait s and several treatments on a dependent variable. Thus, the hypothesis can be tested that a specific relationship between some aspect of hum an behavior and some trait and/or treatment variable(s) is moderated b y several other trait and/or treatment variables simultaneously. Unlik e traditional approaches such as zone analysis, CTTI analysis treats t rait variables as metric variables. Thus, the statistical power and, t hereby, the sensitivity of the design to detect complex relationships is enhanced, requiring relatively small sample sizes. CTTI analysis co nsists of three main steps: (1) exploration of trait interactions with in experimental groups by plotting regression surfaces; (2) designing a proper linear model with specified higher-order interactions; (3) te sting the model using a standard general-linear-model algorithm. To de monstrate this, CTTI analysts was applied to data from a study on indi vidual differences in responsiveness to alcohol and antidopaminergic t reatment, in which the combined influence of two trait variables (anxi ety and impulsiveness) and two treatment variables (ethanol and alpha- methyl-para-tyrosine) on CFF performance was investigated in eighty he althy male subjects. The results showed that, under specific pharmacol ogical conditions, anxiety and impulsiveness as well as their mutual m oderating effects are essential for the drug response observed. CTTI a nalysis proved to be a Very powerful and highly sensitive statistical procedure for detecting complex higher-order interactions in this exam ple of experimental personality research.