Predicting individual skill learning, a cautionary tale

Abstract

People show vast variability in skill learning. What determines a persons individual learning ability? In this study we explored the possibility to predict participants future learning, based on their behavior during initial skill acquisition. We recruited a large online multi-session sample of participants performing a sequential tapping skill learning task. We trained machine learning models to predict future skill learning from raw data acquired during initial skill acquisition, and from engineered features calculated from the raw data. While the models did not explain learning, strong correlations were observed between initial and final performance. In addition, the results suggest that in correspondence with other empirical fields testing human behavior, canonical experimental tasks developed and selected to detect average effects may constrain insights regarding individual variability, relevant for real-life scenarios. Overall, implementing machine learning tools on large-scale data sets may provide a powerful approach towards revealing what differentiates between high and low innate learning abilities, paving the way for learning optimization techniques which may generalize beyond motor skill learning to broad learning abilities.