An experiment on minority influence was designed with six experimental conditions. Three conditions imply an important threat and the others a little onci In each case an identification with the Church or the minority is induced in the subjects or there is not identification at all. Results show that: a) in conditions without explicit identification, minority reaches the conversion effect if the cost is high, b) in the cases with explicit identification with the minority direct and indirect influence declines, c) when there are an identification with the Church there is a curious effect: higher is the cost more important is the minority effect. Two complementary interpretations of results are given. © Taylor & Francis Group, LLC.