The use of standardized problem-solving tasks in large-scale studies can provide more rigorous and powerful results, while incorporating the use of experimental models in the same study can yield important insights into factors that affect problem-solving performance.