Towards improved model structures for analyzing priming: potential pitfalls of using bulk turnover time