Well, I agree. What is the use? Well, inaccurately simple models can sometimes help make complex phenomena manageable. If we roughly assume that there are real

Well, I agree. What is the use? Well, inaccurately simple models can sometimes help make complex phenomena manageable. If we roughly assume that there are real variations in people’s motivation (eg variations in curiosity, status-seeking, fear, joy) we can see whether certain variations predict certain kinds of success. If so, we can think about how to support these educationally. If we find that curiosity-based motivation in math is highly predictive of certain kinds of success, we might want to see how to cultivate that motivation through pedagogy.