In one of the most influential works in the public opinion literature, Philip Converse proposed a “black-and-white” model that divided respondents into two groups: opinion holders and unstable opinion changers. We extend the model by allowing for a group that makes rational opinion changes over time. This enables us to (1) explore hypotheses about the adequacy of Converse’s original model, (2) estimate the percentage of the population that belongs to each of the groups, and (3) examine the evidence for Converse’s basic claim that unstable changers truly exhibit nonattitudes. Contrary to Converse’s suggestion that the unstable group is essentially giving random responses, the results imply that the response behavior of this group may be best interpreted in terms of Zaller’s notion of ambivalence. The results also undermine the measurement-error model, which maintains that unstable responses are mainly attributable to deficient survey instruments, not individual opinion change. We use data collected at four time points over nearly two years, which track Swiss citizens’ support for pollution reduction.