A 1947 study reported a study of rumor formation and dissemination, examining how a story changes as it is passed from one person to another (Allport & Postman, 1947). The experiment used the following procedure:
- Out of a college class, a group of six or seven volunteers are selected. “They are asked to leave the room … They are only told that they must listen carefully to what they will hear when they return to the room and repeat what they heard `exactly as possible.’
- When the subjects have left the room, a slide depicting some detailed situation is thrown on the screen and some member of the audience is assigned the task of describing it (while looking at it) to the first subject. He is requested to include about twenty details in his description.
- After the initial description of the picture, a member of the group of subjects is called back into the room and is placed in a position where he cannot see it … The first subject listens to the `eyewitness’ account given him by the selected member of the audience or by the experimenter.
- A second subject is called into the room, taking his position beside the first subject. Both are unable to see the screen. The first subject then repeats as accurately as he can what he has heard about the scene (still visible to the audience on the screen) … A third subject then takes his position next to the second subject and listens to his report.
- This procedure continues in the same manner until the last subject has repeated the story he has heard, and taken his seat (usually amidst laughter) to compare his final version with the original on the screen” (Allport & Postman, 1947, 65-67).
The description of the pictured scene often changed dramatically as it passed from one person to the next. One of the pictured scenes showed a white man holding a razor, apparently confronting a black man on a subway car. In over half of the experiments using this picture, one or more of the white subjects in this experiment (college students during 1944) reported that the black man was holding the razor (Allport & Postman, 1947, p. 111). These mistaken reports corresponded to the prevailing stereotype of white males regarding black men.
Rumors and Stereotypes
Rumor formation and rumor transmission thrive especially well in response to stereotypes. Stereotypes motivate people to leap to premature conclusions about other people they do not know well (Bodenhausen, 1988). The late Gordon Allport, for many years a distinguished member of the Harvard faculty, spoke of stereotypes as labels. He moreover emphasized: “Labels act like shrieking sirens, deafening us to all finer discriminations that we might otherwise perceive” (Allport, 1954). In response to their prevailing stereotypes, the speculative exchanges of anxious parents and concerned others frequently converge into commonly shared theories.
They can rapidly reach a consensus via processes of “Leveling” and “Sharpening.” In their dialogues, they “sharpen” – or emphasize – impressions consistent with their pre-existing stereotypes. Simultaneously, they “level” – or de-emphasize – any information inconsistent with their pre-existing stereotypes (Gilovich, 1987). In response to the effects of sharpening and leveling, their evolving consensus can convince them that they have discovered important facts – “We agree, therefore we must be right!”
Therefore, via rumor formation and rumor transmission driven by the effects of stereotyping, two or more people can verify for each other that some imaginary event actually transpired. Consequently, what originated as a worrisome possibility (sexual abuse) acquires the unwarranted status of an absolute fact.
Parents who worry that their child may have been sexually abused experience elevated anxiety levels while contending with considerable ambiguity — they do not know exactly what happened (Campbell, 1992). As long as the child denies any abuse, a worried parent struggles with a gnawing, unrelenting sense of anxious uncertainty. Because it is logically impossible to conclusively prove any negative statement, anxious parents can never know for sure that nothing traumatic happened to their child. A parent in these circumstances wonders if the child’s denials are borne out of a perpetrator’s intimidation, or if the child is too embarrassed or traumatized to tell the truth.
Data related to these kinds of circumstances indicate that ambiguous situations increase the probability of people seeking out and relying on the judgments of others (Allport & Postman, 1947; Rosnow, 1991). When anxious parents and concerned others discuss the likelihood of a child they love having been sexually abused, the anxiety of the parents no longer confines itself within themselves. Instead, a situation rapidly develops where two – or more – people share the same anxiety of struggling with ambiguous circumstances affording them limited information. This is the kind of situation that readily facilitates rumor formation and rumor transmission. The intense needs of people to obtain information under these circumstances motivates them to engage in speculative exchanges with each other (Allport & Postman; Rosnow, 1991).