Willpower is best used with care

A DECADE ago when I was an undergraduate psychologist, a departmental librarian called Anne was doing something any psychologist would say was impossible. Every year, with near-perfect accuracy, she would predict which third-year undergraduates would be awarded first-class degrees.
Anne didn't know how their essays were rated, what A-level grades they had under their belts, or how they scored on IQ tests. (All information many would say was essential to forecasting final results.)
All she knew was how often she had seen students in the department library: reading course notes, photocopying journals, borrowing books. And the handful of students who Anne saw a lot - conspicuously more often than the other students in the same year - were going to get a first.
Anne was working on the principle that in academic achievement it is self-discipline, not talent, that counts. Ten years on, a study published recently in Psychological Science has come to exactly the same conclusion.
Psychologists Angela Duckworth and Martin Seligman descended on the eighth grade of a large public school in the northeast of the US. As the autumn leaves fell, each of the 160-odd children took an IQ test, then they (and their parents and teachers) answered questionnaires that probed self-control. Are you good at resisting temptation, they were asked. Can you work effectively towards long-term goals? Or do pleasure and fun sometimes keep you from getting work done?
The children were also given a real-life test of their ability to delay gratification. Each was handed a dollar bill in an envelope. They could choose either to keep it or hand it back and get $2 a week later. Their decision was carefully recorded.
The researchers returned in spring. They took note of each child's grades and then looked back to see both how clever, and how self-controlled, that student had been in autumn. What, they wanted to know, was the most important factor in school