Our work has focused on detecting Mild Cognitive Impairment (MCI) by developing Serious Games (SG) on mobile devices, distinct from games marketed as 'brain training' which claim to maintain mental acuity. One game, WarCAT, captures players' moves during the game to infer processes of strategy recognition, learning, and memory. The purpose of our game is to use the generated game-play data combined with machine learning (ML) to help detect MCI. MCI is difficult to detect for several reasons. Firstly, it is a mild impairment and as such difficult to detect in its early stages, Secondly, it is a subtle impairment for which the brain attempts compensation; as a consequence, it is considered rare in light of normal cognitive decline and the brain's ability to mask its manifestation. The problem of early MCI detection is further compounded as people have various cognitive acumen which again can lead to false positives which would exacerbate the rare diagnosis still further. To evaluate the conjecture, ML methods are used to generate synthetic data to plausibly emulate a large population of players. Reinforcement Learning (RL) is used to train bots as RL most closely emulates the way humans learn. Considerable trial and error (training) is required, therefore RL bots were developed that process millions of gameplay training patterns and achieve results comparable to the best human performance. This baseline allows us to create bots to emulate individuals at various stages of learning, or conversely, various levels of cognitive decline. The paper demonstrates the ML work to both generate data and subsequently classify different levels of play. This development stage is necessary as part of the larger objective to create SGs that detect MCI.