This study examined the relation between professional affiliation and the framing of expert congressional testimony about nicotine's addictiveness. Experts were chosen from three different types of sponsoring organizations: the tobacco industry, government, and independent research organizations, both pro- and anti-tobacco. The study sought to identify common technical biases and policy concerns that could define an overall “expert” attitude, as well as differences where the experts’ framing of nicotine addiction would reveal attempts to favor their own institutions. Semantic network analysis was (...) applied to each group's discourse, thereby clustering associated words that represented major themes in each type of expert group. Clusters revealed a common preoccupation with narrowly defined, lab-based evidence, but more locally, each group framed the issues to support its sponsor's strategy. (shrink)