According to definition, data bases used and variables considered, it is estimated that between 5 % and 30 % of cancers are due to environmental exposures. In this article, we inventory perceived and actual risks relative to interactions between environmental exposures and cancer. We emphasize methodological challenges of environmental epidemiology studies, central to scientific inquiry and discussion, given the numerous doubts associated with this topic. We then examine how risks’ perception motivates people to appropriate this research domain, traditionally reserved to experts. Finally, we demonstrate that risk perception depends on its quality (known vs. unknown, preferred vs. suffered, high vs. low), its visibility and its proximity to populations.