Zajonc expands the general idea of the mere exposure effect into non-conscious preference. He rejects the necessity of mediation by demonstrating that the effect of repeated exposure on liking occurs even in the absence of conscious processing of the attitude toward object.

Moreland and Zajonc (1977, 1979) conclude that the stimulus recognition is a sufficient but not necessary condition for the mere exposure effect to occur after reporting that repeated mere exposure to the Japanese ideographs increases attitudes and subject recognitions. Kunst-Wilson and Zajonc (1980) demonstrated experimental founding which is similar with Moreland and Zajonc's statistical result. Performing a listening task, they propose that the more one is exposed to the melodies - without the participants' remembering their previous exposure to the music - creates the greater liking. Suggesting in his long article, "Feeling and Thinking: Preferences Needs No Inferences" in the traditional journal, American Psychologists, Zajonc(1980) finally hypothesizes that emotion and cognition, although in virtually continuous interaction, are in fact separate systems and the affective system often commands temporal primacy.

Although "when the paper was published, no neuroanatomical or neurophysiological evidence existed to support Zajonc's hypothesis" (Bargh and Apsley, 2001, p.5), many subsequent researchers have proven Zajon's radical idea on emotion and cognition. Ledoux (1996) suggests the primitive pathway of fear without thought as neurological processes, illustrating how we can experience emotion before cognition. Lazarus (1991) concedes that some emotional responses - especially simple likes, dislikes and fears - involve no conscious thinking while complex feeling such as grief, happiness and love arises from our interpretations and expectations. Those evidences all support Zajonc's belief that some of our emotional reactions involve no deliberate thinking, and that cognition is not always necessary for emotion.

In addition, Byrne, Clore and Smeaton (1986) suggest a Two-Step Process, indicating that we avoid others who are different, then among those who are left, we seek out those people who are the most similar to us. They show that the more similar two individual are, the better are the chances that their relationship will last. Mita, Dermer and Knight (1977) find that the mere exposure effect can also influence our self-evaluation. Because our face is not perfectly symmetrical, the face we see in the mirror is not the same as the ones our friends see. Most of us prefer the familiar mirror image, while our friends like the reverse.