a statistical trendp < 0.01
Provision of experimentally manipulated evidence about prevalence lowered regression weight for the associations between ‘attitude’ and ‘willingness’ in condition A by 45% relative to the control group, which did not receive any prevalence information; a similar effect (lowered regression weight by 40%) appeared as a statistical trend in condition B ( Figure 2B ; condition A: β stand. = 0.31, p < 0.01; condition B: β stand. = 0.34, p < 0.01; condition C: β stand. = 0.57, p < 0.01).