A growing empirical literature on how to measure uncertainty has emerged following the 2007-2008 financial crisis. We investigate the impact of uncertainty shocks according to their nature, applying local projection methods in a nonlinear framework. More precisely, we investigate the impact of uncertainty shocks according to their nature (nonfinancial and financial) in different regimes of general uncertainty: high, moderate and low. We get a positive and significant effect of the nonfinancial uncertainty shocks when the general level of uncertainty is high or moderate.