In 1969, the psychologist Robert Zajonc published an article about a curious study. He'd posted a silly-sounding word—either kardirga, saricik, biwonjni, nansoma, or iktitaf—on the front page of some student newspapers in Michigan every day for several weeks. Then he sent questionnaires to the papers' readers, asking them to guess whether each word referred to “something ‘good'” or “something ‘bad.’” Their answers were consistent, if a little strange: Nonsense words that showed up in print many times were judged to be more positive than those that appeared just once or twice. The fact of their repetition, said Zajonc, gave the words an aura of warmth and trustworthiness. He called this the mere exposure effect.