For example, Martin (2003) gave participants a message supporting a particular viewpoint, and attitudes measured. Then they heard an endorsement of the view from either a minority or a majority. Finally they heard a conflicting view; attitudes measured again.