A Sequential Response Model With Growth Parameters for Process Data in Complex Problem Solving

Interactive problem-solving assessments allow examinees to learn from trial-and-error, potentially changing their response patterns when revisiting problem states due to learning or fatigue effects. This study introduces the Sequential Response Model With Growth parameters, which captures how examinees’ cognitive processes evolve through accumulated experience during task interaction via growth parameters. An empirical application to Complex problem-solving assessment reveals significant positive effects in intermediate states (mastering task mechanisms) and negative effects at the initial state (accumulated frustration). Simulation results confirmed robust parameter recovery when growth effects were present, while maintaining comparable performance when such effects were absent. These findings suggest that incorporating operational history enhances our understanding of dynamic problem-solving processes and provides insights for assessment design and educational practice.