The economic value placed on a reduction in the risk of mortality relies more and more frequently on contingent assessment techniques. These consist in questioning a sample of individuals on their "willingness to pay" (WTP) in order to reduce this risk, on the basis of hypothetical scenarios. These WTP nevertheless depend on many factors, especially the nature of the underlying risk and the scenario proposed to reduce it. This article deals with reducing the risk of mortality associated with exposure to atmospheric pollution and tests the effect of a change in the hypothetical context of assessment through three scenarios: a new drug, a house move and new regulations. To analyse the "willingness to pay" stated in the different scenarios, we define a unified, theoretical then econometric framework of analysis, taking into account the preferences of the participants at present, and also those of other household members. There are two main results. The estimated implicit updating rates, specific to each hypothetical scenario, were seen to differ significantly. They were about 7% for the "move" scenario and 24% and 26% for the "drugs" and "regulations" scenarios respectively. Results showed that the average "values set for avoiding death" differed greatly between the "move" scenario (801,000) on the one hand, and the "drugs" (299,000) and "regulations" (252,000) scenarios on the other hand.