Sensitivity (proportionality) of willingness to pay to (small) risk changes is often used as a criterion to test for valid measures of economic preferences. In a contingent valuation (CV) study conducted in Austria, 1,005 respondents were asked their willingness to pay (WTP) for preventing an increase in the risk of being killed in an avalanche of 1/42,500 and 3/42,500 respectively. WTP for the higher variation in risk is significantly greater than WTP for the lower risk change. We find evidence that those respondents who have had personal experience of avalanches in recent years combine the information about future risk increase - as provided in the survey - with the observed number of fatal avalanche accidents in the past. Proportionality of WTP holds if such prior experience is taken into account and if attitudinal factors in scope tests are controlled for.