Learning Rates Are Not All the Same: The Interpretation of Computational Model Parameters Depends on the Context

Abstract Reinforcement Learning (RL) has revolutionized the cognitive and brain sciences, explaining behavior from simple conditioning to problem solving, across the life span, and anchored in brain function. However, discrepancies in results are increasingly apparent between studies, particularly in the developmental literature. To better understand these, we investigated to which extent parameters generalize between tasks and models, and capture specific and uniquely interpretable (neuro)cognitive processes. 291 participants aged 8-30 years completed three learning tasks in a single session, and were fitted using state-of-the-art RL models. RL decision noise/exploration parameters generalized well between tasks, decreasing between ages 8-17. Learning rates for negative feedback did not generalize, and learning rates for positive feedback showed intermediate generalizability, dependent on task similarity. These findings can explain discrepancies in the existing literature. Future research therefore needs to carefully consider task characteristics when relating findings across studies, and develop strategies to computationally model how context impacts behavior. Highlights Efforts in computational cognitive modeling often assume that Reinforcement Learning (RL) modeling parameters will generalize between studies and models, but this is not well established. We empirically investigate whether RL parameters generalize between three tasks and models, using a large developmental dataset and a within-participant design. We find that RL decision noise/exploration parameters generalize fairly well, but RL learning rates do not. Our data support previous conclusions that decision noise/exploration decreases during development (ages 8-17), but suggests that claims about learning rate development cannot be generalized.