Paper: Problem Solving: Rats vs. Astronomy Students Volume: 524, Advancing Astronomy for All (ASP 2018) Page: 185 Authors: Montgomery, M. M. Abstract: Given a choice at a maze intersection, rats tend to take the path that has a guaranteed small morsel of food at maze's end rather than risking an uncertain route that has an uncertain outcome. To test whether general education Astronomy students have similar metamemories in decision making, students are given two sequential visual python numerical coding assignments several weeks prior to follow-up testing. Of the 131 participants, 31 elected to earn zero points on the assignments, and two of the 31 did not take the follow-up exam. Four questions in the exam tested student memory of vpython commands, lines of code, and code outputs and comprehension of the inverse square law concept conveyed in the vpython assignment. Students could opt out for a small reward of similar proportion to that in the rat studies, that is 1/4 of the maximum possible points for that test question. Only 37.9% of the 29 students made decisions like the rats, opting out of testing due to a lack of metamemory to earn the guaranteed small reward. Of the 111 students that completed the assignment, only 8.1% did not remember and opted out of all four test questions for the small reward. More than 60% answered these memory and comprehension questions incorrectly in the follow up testing. These results show that some gain in metamemory affects student meta-reasoning and thus decision making. We find no significant correlation between undergraduate major and decision making. As 100% of the freshman students opted in to both the numerical coding assignments and follow up testing, we find that freshman lack the strength in meta-reasonings that contributed to the decision making of their higher ranked class peers. The most significant reason students reported for not answering a follow-up test question is not having prior exposure to numerical modeling. Because 70% felt they learned the inverse square law after having completed the assignments that had an average score of 78%, we find that nearly all general education astronomy students have the ability to learn via the mode of computational modeling.