Using Brain Computer Interaction to Evaluate Problem Solving Abilities

The ability to solve problems is increasingly important in today’s world, not only for good school performance but also to be successful in today’s world, being one of the most desired skills for the XXI century. However, the existence of tasks with an inadequate cognitive load may discourage the individuals involved in it. Thus, we believe that the effective monitoring of this capacity must be well monitored. To this end, we started an experiment made up of 2 different samples to assess the ability to solve logical problems through the testing of Raven’s Progressive Matrices. The research project developed and presented in this paper sought to assess differences in the ability to solve logical problems considering brain activity when solving them. Therefore, EEG was used to infer the cognitive workload of individuals. Our main interest was to identify specific ERP waveforms, namely the feedback-related negativity (FRN) component about the correctness of the students answers to each question. The analysis presented in this work shows that it is possible to find the FRN potential associated to a greater negativity meaning a greater astonishment for an unconsciousness of the wrong answer. Therefore, this aspect is related with the performance of the participant based on their knowledge of the abstract principle underlying the task. Despite having only 2 samples with few students, these data indicate that our findings demonstrate that cognitive load can be predicted using these features, even using a low number of channels.