These methods should not be dismissed as unimportant in the study of national characteristics: Yet in principle if one wants to know whether the people of one nation behave differently from those of another, it would seem only reasonable to examine the relevant behavior directly, and to do so under conditions of controlled observations in order to reduce the effects of personal bias and to make measurement more precise.
An important step in this direction was reported in 1954 by an international team of psychologists who worked together as the Organization for Comparative Social Research. This team studied reactions to threat and rejection among school children in seven European nations, using hypotheses advanced by Stanley Schachter of Columbia University. The inquiry was not specifically designed to study national characteristics but chiefly to see if certain concepts regarding threat and rejection would hold up when tested in different countries. In the course of the study certain differences between countries did turn up, but the investigators felt they were not necessarily genuine. Conceivably they were due to defects in the experiment or to inadequacies in the theory behind it. Although its focus was on theory validation, this study is a landmark in cross-national research. Unfortunately the Organization for Comparative Social Research halted its research program when the study was completed.
My own investigation was begun in 1957. My objective was to see if experimental techniques could be applied to the study of national characteristics, and in particular to see if one could measure conformity in two European countries: Norway and France. Conformity was chosen for several reasons. First, a national culture can be said to exist only if men adhere, or conform, to common standards of behavior; this is the psychological mechanism underlying all cultural behavior. Second, conformity has become a burning issue in much of current social criticism; critics have argued that people have become too sensitive to the opinions of others, and that this represents an unhealthy development in modern society. Finally, good experimental methods have been developed for measuring conformity.
The chief tool of investigation was a modified form of the group-pressure experiment used by Solomon E. Asch and other social psychologists [see "Opinions and Social Pressure," by Solomon E. Asch, SCIENTIFIC AMERICAN, November, 1955]. In Asch's original experiment a group of half a dozen subjects was shown a line of a certain length and asked to say which of three other lines matched it. All but one of the subjects had been secretly instructed beforehand to select one of the "wrong" lines on each trial or in a certain percentage of the trials. The naive subject was so placed that he heard the answers of most of the group before he had to announce his own decision. Asch found that under this form of social pressure a large fraction of subjects went along with the group rather than accept the unmistakable evidence of their own eyes.
Our experiment is conducted with acoustic tones rather than with lines drawn on cards. Five of the subjects are confederates of the experimenter and conspire to put social pressure on the sixth subject. The subjects listen to two tones and are asked to say which is the longer. The five confederates answer first and their decisions are heard by the subject, who answers last. The confederates have been instructed to announce wrong answers on 16 of the 30 trials that constitute one experiment.
We elected to use tones rather than lines because they are better suited to an experimental method using "synthetic groups." Two psychologists working at Yale University, Robert Blake and Jack W. Brehm, had discovered that grouppressure experiments can be conducted without requiring the actual presence of confederates. It is sufficient if the subject thinks they are present and hears their voices through headphones. With tape recordings it is easy to create synthetic groups. Tapes do not have to be paid by the hour and they are always available.