Current ability EI measures rely on the assumption that there is a "correct" way to use emotions intelligently and that this way is applicable to all individuals. Furthermore, ability EI tests imply that knowledge about emotions (e.g., how to categorize them, how to use them in different situations, how to regulate them) is a fundamental characteristic for identifying emotionally intelligent individuals. In this presentation I will examine whether the use of a correct answer for scoring the test is a feasible criterion and to what extent explicit emotion knowledge may predict emotionally intelligence performance. As an alternative, the investigation of emotion processes underlying EI will be proposed and new avenues for measuring the fluid component of EI that rely on procedural, rather than declarative knowledge will also be discussed.