Understanding the causes and consequences of individual differences in behavior, neurobiology, and physiological regulation is advancing in several biological disciplines, from evolutionary ecology to biomedicine. Social interaction is one intensively studied factor that can modify both behavior and physiology. Pre-existing differences in physiology, whether innate or acquired, may determine the outcome of agonistic interactions and hence social rank. Obviously, it is not straightforward to disentangle causes and consequences in this interaction. Here, I will review work on the HR-LR (high- and low-responsive) rainbow trout model, initially selected for contrasting post-stress cortisol levels. I will dwell on the observation that LR fish usually show a proactive behavioral pattern and also win contests for social dominance against size-matched HR. This outcome is however context-dependent, with HR fish becoming dominant in a new environment after food restriction and transport. More recent examples of subtle environmental changes with high impact on aggression include unpredictable omission of expected rewards (a.k.a. 'frustration'), which has a remarkable stimulatory effect on usually inferior (i.e. smaller bodied) competitors.