The main objective of this research study is to determine the relationship between entropy and behaviors that optimize learning decision-making among student teams during an experimental physics science class. An initial condition (IC) is applied by using a very short video with apparently unrelated content, emerging as a discovery the question that must be answered by each team and its experiment. The behaviors displayed are transformed into time series that are analyzed with a chaos theory software. The results show that the emotions time series, in chaotic dynamics, show that entropy has the highest thermal correspondence, positively influencing team collaborative learning while favoring each student’s training. It is concluded that using ICs within a context, at the beginning of a class, is suitable for inducing a positive emotional disposition among student teams that are committed to an experimental physics course.