Measuring individual differences in statistical learning: Current pitfalls and possible solutions