Abstract: The purpose of this study was to identify, develop and analyze a new variable associated with informal and incidental learning (Marsick and Watkins, 1990, 2017), the roughness of learning. The variable, based on Robert May’s logistic model (May, 1975), measures the complexity of the learning challenge and assesses how this complexity changes over repeated play. May’s model shows quantitatively the loss of predictive patterns as the rate of adaptiveness approaches a threshold of chaos (Mitchell, 2009). This study used this framework in a repeated measures, longitudinal study to examine human adaptive capacity as influenced by individual incidental learning and uncertainty to predict survival in a serious game [Minecraft]. To answer the research question: Is there a roughness of learning variable that measures complexity in adaptiveness, we developed the variable by calculating the fractal dimension of the shape of learning extracted from player health data at time of death and rescaled the data to use with May’s formula. Using ANOVA, a significant difference (p <= 0.01) was found between groups based on experience. There was no significant difference in the roughness of learning based on play. Based on our analysis of the overall fractal dimension, we concluded a roughness of learning effect can be estimated as a measure of complexity in the game. The fractal dimension of learning as manifest in the roughness of learning variable adds a new quantified variable to the study and assessment of learning effort in HRD.
