This study sought to replicate a model developed to explain the problem-solving process through insight with human participants. This process happens when an organism solves a problem by producing a behavior chain with at least two operant behaviors that share similar consequences during acquisition but in different contexts. We recruited six participants aged between 20 and 30 years old with a low level of previous experience in first-person video games. They also did not have significant experience with the Portal 2 game and had adequate structural conditions to participate in the experiment remotely. The procedure has five phases: familiarization, pretest, practicing of behaviors not necessary for problem solving, practicing of behaviors necessary for problem solving, and posttest. As a result, three participants solved the problem only in the posttest, one solved it in the pretest phase, and two did not solve it in any of their presentations. However, all participants issued the necessary behaviors to solve the problem. For future experiments, it is encouraged that changes in the problem-solving situation occur that seek to increase the probability of solving the problem once the participant emits the necessary behaviors.