Next, we consider the relationship between participants’ assessments of default risk and their
lending decisions to better understand what happens during the evaluation and weighting stage of
information processing. All else equal, we would expect a positive correlation between assessments
of relative creditworthiness and lending decisions. As expected, these assessments are highly
correlated with participants’ lending decisions across conditions (r ¼ 0.64, p , 0.01) and in each
individual condition (p , 0.05 in each condition). Interestingly, this correlation is (at least
nominally) highest when renewal periods are not capitalized and is lowest when renewal periods are
capitalized and disaggregated (see Figure 2).