The Psychology of "Attitudes": 
Part II

Muzafer Sherif and Hadley Cantril
Princeton University


Toward the end of Part I, which appeared in the November number of the REVIEW, we discussed the formation of frames of reference in individual situations. We proceed now to the consideration of attitudes in social situations, ranging from laboratory experiments involving social factors to complicated social conditions of everyday life. In doing so, we should stress again our conviction that in spite of the enormous variation in the content of attitudes, psychological principles of attitude formation are essentially the same, irrespective of what the attitude is concerned with.

The first stage of attitude formation —in the most complicated social situation as well as in a restricted laboratory experiment—is a perceptual stage. The individual must somehow come into contact with the stimulus situation before any attitude is established. In strict laboratory situations this stimulus situation is generally neutral and the frame of reference established lacks the vital affective and motivational properties so characteristic of social attitudes: in the laboratory experiment the stimuli may be a series of magnitudes, such as lines or weights or perceptual structures; in social life the stimuli may be a person, group, an expressed attitude or prejudice, a threat, or a value-judgment (norm) of any kind. Some of these stimuli in social life may be relatively neutral, may have only a mild affective property, e.g., the color of a dress, the shape of a house, the characterization of a city, while some may involve intense, sometimes even traumatic, experiences, e.g., a girl whom one loves at first sight, a scolding received by a child for cheating, an announcement over the radio that the Japanese have bombed Pearl Harbor.

In dealing with attitudes in social situations, then, we are dealing with stimulus situations rife with affective and motivational properties, with situations that frequently involve problems of the individuals status. It might be mentioned in passing that attitudes to situations involving affective properties seem to be more readily established (learned), just as Pavlov, Tolman and others showed that conditioning was more effective under motivational stress. Whether the perceptual experience involves direct contact with an object, person, or group or whether the perceptual experience involves the verbal transmission of a social norm by some short-cut value-judgment, makes ho difference whatever as far as the basic psychological characteristics of attitude

(2) formation are concerned. Both the soldier who has had direct experience with the enemy and his sister back home who has been subjected to value-judgments concerning the enemy can and often do develop various negative attitudes. It is therefore systematically useless to try to categorize attitudes according to their source or origin. This is, of course, not a denial of the specific properties of particular attitudes formed in specific situations.

Irrespective of the particular way in which an individual acquires an attitude in social life, the literature of social psychology is rife with data which support the formulations reached from our survey of general psychology: that perception and judgment are selective and occur within a referential framework, that frames or points of reference are inevitably established if an individual repeatedly faces the same stimulus situation, that these frames and points of reference are by no means always confined to consciously accepted instructions or imposed norms but can become established without the individuals realization of it, and that once established these frames and points of reference serve as anchorages for perception and judgment.

Experimental formation of a frame in group situations. This series of experiments was undertaken with the methodological conviction that the principles formulated and verified in general psychology are valid principles that operate in individuals participating in group action or collective behavior. Results produced by group interaction may be, and in fact are, different and may have all the earmarks of emergent products. Nevertheless, the psychological principles involved in the group situation are the same in operation. The emergence of new qualities is not a unique property of group action alone—perceptions of forms, relationships and melodies emerge as unique qualities not found in individual parts. This emergence of new qualities on different levels—individual as well as collective—has to be stressed constantly because of the sharp dichotomies drawn by certain schools of thought. For example, the dichotomy of experimental psychology vs. cultural psychology made by the romantic German school represented by Spranger (117) portrays experimental (individual) psychology as atomistic, static, and cultural psychology as dynamic and meaningful with total qualities. Writers of the Durkheim school of spiritualistic sociology, for example, Blondel (13), Halbwachs (59), and Durkheim (46) himself, argue that new qualities emerge only in collective situations, whereas findings in psychology substantiate the view of dialectical materialism that emergence takes place in all levels—physical, biological, psychological and socio-economic. In a recent study full of implications for further research in social psychology, Tresselt and Volkmann experimentally demonstrated the production of uniform opinion by non-social stimulation (125). These investigators express a general psychological fact when they state that "the mechanism for judging social stimuli is the same as the mechanism for judging non-social stimuli" (p. 242).

With this methodological concern in mind, we can turn to the results of Sherif’s group experiments on autokinesis. In the group situations each subject reported his judgments aloud, and naturally was heard by the other subjects in the group. He in turn heard their expressed judgments. No primacy effect was introduced into the situation, a special point was made in the wording of the instructions to the effect that subjects could report their judgments in any order in any presentation. A subject who gave his judgment first in a

(3) particular presentation, might be the last to report in the following presentation. The building up of a frame in group situations is a temporal affair established in the course of the experimental period-the influence of expressed judgments is not restricted to judgments given at particular presentations.

Some groups faced the autokinetic situation without any previous individual experience. An equal number of groups faced it after the members had established their individual frames alone in individual experiments. The members of the groups who first took part directly in the group situation were therefore naive and neutral in relation to this unstructured, ambiguous situation.

The results are clear. When individuals as members of a group face the same unstable, unstructured situation without any previously established personal relationship among them, a scale (frame) and a standard (reference point) within that frame are established in the course of the experimental period.[1] When a member of the group faces the same situation subsequently alone, after a frame is established in the group situation, he experiences the situation in terms of the frame established in the group situation as a consequence of group interactions.

This result was verified by Asch and Fright in experiments performed in 1937. We shall give these results here in Asch’s own preferred terminology.[2]

"The outstanding feature of the investigation of Sherif is the tendency of the subjects in the given experimental situation to reach mutual agreement. S. E. Asch and B. Fright undertook in a series of experiments to examine more closely the specific reasons for this tendency.

"Before proceeding with their experimental variations, they first repeated the main experiment of Sherif with four pairs of subjects. The arrangement of the experiment was as far as possible identical with that of Sherif. The main findings of Sherif were confirmed: a new norm (level of response) developed under the given conditions. However, the convergence proceeded along different paths in the different pairs of subjects. For example, some members exercised more influence on the direction than others. This finding is entirely consistent with the results of Sherif." (Variations of these experiments by Asch and Wright will be reported in their appropriate context later.)

More recently, in a series of studies dealing with some social factors in judgment, Schonbar investigated modifications of judgments in situations of medium and high structure respectively (110). From her results she concludes "our findings confirm and extend the conclusions arising from Sherif’s work on the autokinetic phenomenon" (p. 129). That the convergence she found in cases of high structure is almost identical to the convergence found in cases of medium structure is puzzling, in view of the differentiated results obtained with varied degrees of structure and other factors by Asch, Lewis and Hertzman, Coffin, Chapman and Volkmann, Luchins, and Asch and Wright.

In Sherif’s experiments with autokinesis, when individuals first establish their frames and standards in individual sessions, and are then brought into group situations, their judgments tend

(4) to converge. But the convergence is not so close as when they first work in the group situation. When an individual comes into a group situation, with his own established frames and standards, there is a tendency to stick to them to some extent: individuals bring into the situation their own established frames and are no more naive in relation to the situation (113, p. 31, 41; 114, p. 104).

In one variation of these experiments, Asch and Wright gave contradictory instructions in the individual sessions, by prescribing varying ranges of movement. "Under these conditions of contradictory individual norms, no convergence developed in any of the pairs. Subsequent questioning revealed that fully one-half of the subjects did not realize that they were in contradiction with their partners. They interpreted the addition of the integer 3 (or the dropping of the integer 3) as introduced for the sake of identifying each of the subjects in the dark room. The remaining subjects interpreted the existing difference in terms of such factors as differences of position with regard to the light, differences in eyesight, etc." This precisely means that even following individual sessions, with prescribed contradictory instructions and contradictory standards and norms, half of the subjects converged. The addition or dropping of the integer 3 (which the subjects thought was introduced as an experimental device) accounts for statistical but not for psychological nonconvergence as explained by the subjects who "did not realize that they were in contradiction with their partners."

In a more decisive variation aimed at the complete destruction of convergence, Asch and Wright performed the following experiment. In Asch's words, "The aim of the following experiment was to alter in a decisive way the cognitive character of the situation. The subjects were informed in the individual session concerning the subjective nature of the autokinetic effect (each subject was paired in the present variation with a planted subject whose estimates differed considerably and consistently from the experimental subjects). There were ten subjects in this experiment. Under these conditions, five subjects showed no convergence effect whatsoever; the other five did depart significantly from their previous judgments and in the direction of the planted subject. To explain the two forms of reaction, the subjects were questioned at the conclusion of the investigation. According to the present results, it seems that the subjects who showed the convergence effect continued to think of the autokinetic phenomenon as objective and forgot the preceding instructions. Nevertheless, the change found in the first half of the group conclusively demonstrates how effective the alteration of the cognitive character of the situation can be" (see ftn. 11, italics ours). In spite of the fact that the subjects were told beforehand that the light was not moving at all and in spite of their individually established norms before coming into the group situation, it is surprising that there was any degree of convergence at all. We do not know if the fifty per cent convergence attributed by Asch to forgetting is in line with current forgetting curves. The lapse of time between the individual and group sessions cannot be longer than a few days.

Nevertheless the above variations which introduced into the situation various degrees of familiarity or objective knowledge of the situation account for the variations in the results obtained. They are in harmony with Sherif’s results which showed that once individual frames are established, convergence or agreement in subsequent

(5) group situations is decidedly affected by them. Likewise, the findings in these variations are in harmony with the findings of Chapman and Volkmann that previous familiarity with a situation prevents shifts in the direction of experimentally introduced standards or norms. Likewise, they are in harmony with the finding of Luchin's that social influence varies, among other factors, with variations in structural clarity. And all are in harmony with the formulation reached in the previous section that the influence of internal and social factors increases directly with the increase of the unstructuredness or ambiguity of the situation. and decreases with the degree of the structuredness or unambiguity of frames and standards already established by an individual.

Experimental inculcation of a frame. As we stressed above, frames are formed in the individual in daily life corresponding to objects and norms around him which are compelling in their objective structure. Quite frequently attitudes are formed as a direct consequence of short-cut verdicts or value-judgments of grown-ups, teachers and others around us. These value-judgments may or may not be imbedded in the truth of actual relationships. We are concerned here with the psychological process involved in their acquisition. The formation of such a frame in a social situation has been experimentally demonstrated. The experiment was carried out by Sherif in the spring of 1936 at Columbia University again utilizing the autokinetic technique (115).

The specific problem in the experiment was to see whether naive subjects could be experimentally inculcated with varying ranges of prescribed frames and standards. Preliminary experimentation had shown that a partner in this situation (in this case a partner with considerable prestige) could influence the judgments of the other subject-raise and lower them in his direction. When the naive subject was told exactly what had happened she became very disturbed, an indication that an individual can get emotionally upset by being so fooled (115, p. 92).

In the main experiment there were seven groups of two members each. In each group, one member coöperated with the experimenter by deliberately distributing his judgments around the range (scale) and a standard point within that range prescribed beforehand by the experimenter. The second member (designated as the naive subject) was totally unfamiliar with the situation. In each case the naive subject was not acquainted either with the experimenter or the coöperating subject. To be sure that the conformity was to the prescribed range and standard, a different range and a different standard were prescribed for each group. In order to avoid the factor of primacy, the coöperating subject was instructed to let the naive subject express his judgment first at least half the time. The first session was the group session with both the coöperating and naive subjects participating. In the second session the naive subject was alone.

The results indicate considerable convergence to the prescribed range (scale) and standard in the judgments of the naive subjects. The varying convergence in different subjects clearly indicates personality differences. The convergence of the judgments in the alone session, which took place the day following the group session, was, in general, even greater than that obtained in the group sessions. Introspections reveal that the subjects became conscious of the scale and standard formed during the course of the experiments: Five subjects out of seven reported in their written introspections at the end of the experiments that they were not influenced

(6) by the judgments of their partner. This clearly indicates that a social influence which continues even when the individual is alone need not be consciously experienced. Schonbar has recently obtained results which lead to the same conclusion (110).

It is a significant fact to find convergence greater, in the majority of cases, when the individual is acting alone. It is not a rare occurrence in everyday life to react negatively or hesitatingly on some topic raised by some person while in his presence, but to respond positively when he is no longer in the situation. To yield easily is not a pleasant ego experience.

This experiment embodies in itself the rudiments of the psychology of attitude formation: a scale (or frame) is formed under the influence of the verdicts of another individual in relation to a definite, experimentally controlled stimulus situation. The scale (frame) is carried from one day to another, has to some extent an enduring quality which provides an individual with a state of readiness by means of which he reacts in a characteristic way to the same stimulus situation. Here we have the main earmarks of an attitude. We must admit at once that this formation lacks the affective quality of a real attitude. But we do not believe this at all invalidates our conceptualization. The investigations of Beebe-Center (9) and associates Cohen (41) and Hunt and Volkmann (67) all indicate that the mechanisms of affective judgments are not essentially different from those of other judgments. Besides, as we shall see there is further evidence to support the general scheme so far presented.

Social factors in more concrete and everyday life situations.—The conceptual scheme offered here towards the psychology of attitudes may appear to be an artifact based only on the results of laboratory experiments which have no counterparts in real life situations. So we shall turn now to some examples of more everyday life situations which seem to us to confirm the reality of our formulations. We cite here only a few of the many studiesranging from relatively simple demonstrations to complex reactions of daily life-that confirm the conclusions reached in the more strictly laboratory settings.[3]

In his series of experiments (one of which has already been referred to), Coffin investigated the psychology of suggestion as it was related both to attitudinal structure and to stimulus situations (40). In his first experiment Coffin studied the relationship between an individuals attitude and the type of propaganda to which he was particularly susceptible. He found (in the winter of 1939 and 1940) significant correlations between pro-Allied attitudes on the one hand and the acceptance of specially prepared pro-Allied propaganda on the other hand. Conversely those with pro-German attitudes accepted pro-German propaganda to a significant degree. In a second experiment Coffin used the Rorschach inkblots as stimuli with little structuration. After his subjects had rank-ordered a list of ten occupations according to their social standing, they were divided into

(7) two groups; each group was told the characteristic response given to the Rorschach inkblots by professional men, by business men, by skilled laborers and by those on WPA, but the various characteristics were attributed to different occupations in the two experimental groups. The results clearly showed that when these ambiguous stimuli were used, the subjects were highly influenced by the suggestions given which served as anchorages. They actively structured the imaginative situation according to the characteristic reaction to the blots of occupational groups they believed had high social standing. In a third experiment, testing the relationship between suggestibility and the difficulty of problem solution, Coffin found that when mathematical problems were arranged in order of difficulty with marginal (and usually false) hints (anchorages) beside each problem as to what procedure might be used in its solution, respondents more highly trained in mathematics followed the marginal suggestions less: those who knew the most mathematics accepted less than half as many of the suggestions as those who knew least mathematics. Those persons, therefore, least able to make successful checks for themselves were most susceptible to an imposed instruction.

We have cited Coffins experiments here at some length, since they provide unequivocal evidence of two important conclusions: first, that if a suggested stimulus is related to and consistent with an established frame of reference, it is likely to be accepted; and, second, if a stimulus situation is ambiguous or relatively meaningless because of its difficulty, then a frame of reference verbally imposed by the experimenter is readily accepted as the basis for judgment.

A study by Ruth and Eugene Hartley concerned with poorly defined situations is particularly significant since it shows how individuals establish individually characteristic ranges and reference points by means of which they judge a relatively unstructured social stimulus (59a). The Hartleys had students rate a series of pictures of completely unknown men. Among other ratings, the subjects were asked to estimate each man pictured on his general ability and his likeliness to succeed. They found a constancy of ranges and central tendencies in the judgments and conclude that "In making primary evaluative social responses to people, when little information is available, individuals tend to manifest characteristic ranges and central tendencies with reference to which judgments are made "

The research of McGarvey "was undertaken in order to determine whether relations similar to those which have been found to hold with the psychophysical materials would appear also in judgments of verbal material along value-dimensions" (93, p. 26). In her first experiment, McGarvey had subjects rate the social prestige attached to a number of occupations. After they had rated all occupations on a graphic scale, in further experimental sessions the top or the bottom category of the scale was set by an ‘anchoring value’ —a given occupation was used to define one extreme or the other. In the second experiment the items judged were short descriptions of various types of social behavior to be evaluated in terms of their desirability. Again anchoring stimuli were introduced at either end of the scale. On the basis of these experiments McGarvey concluded that "the effect of the anchoring value was that of bringing about an extension of the absolute scale upward or downward in the direction of the anchoring value. The extension of the absolute scale involves not only a displacement of the scale with reference to the range of

(8) values represented by the stimulus-series, but also a widening of the categories of response—a finding in complete agreement with the results obtained with psychophysical material" (93, p. 78). In other words, judgment of a given stimulus is found to be determined by the frame of reference within which that judgment occurred and the introduction of new anchorages changed the dimensions of reference frames. This precisely parallels on a social level the results obtained by Rogers cited in Part I.

The close relationship between memory and an individuals frame of reference was exhaustively shown by Bartlett who concludes that "remembering is schematically determined" (7, p. 312), "an imaginative reconstruction or construction, built out of the relation of our attitude toward a whole active mass of organized past reactions or experience, and to a little outstanding detail which commonly appears in image or in language form" (7, p. 213). "The circumstances that arouse memory orientation, whether they occur in the laboratory or in everyday life, always set up an attitude that is primarily toward a particular schematic organization" (7, p. 312). What a person remembers, as well as what a person observed of a given stimulus situation, was, he found, clearly influenced by the particular social origin of the individuals attitude. He showed how these attitudes—and consequently recall and observation—varied according to the cultural background of the individual, how they differed among members of various groups within a given social system, and how they differed within the same individual when be was or was not in the actual social presence of other members of his group. Various subsequent studies have shown that material consistent with a persons attitude is much more likely to be remembered than material not consistent. Seeleman, in stating the problem of her experimental study of memory, says that "these differences in standards may be expressed as differences in subjective norms or frames of reference" (111, p. 7). She analyzed the influence of the attitude toward the Negro on the remembrance of pictures of whites and Negroes to which were attached, in one experiment, favorable or unfavorable phrases supposedly describing the individual shown. She found that persons with extremely unfavorable attitudes toward the Negro recognized fewer individual differences between the Negroes shown in the pictures, recognized correctly fewer Negro than white pictures, assigned to Negro pictures more unfavorable phrases and remembered these unfavorable phrases attached to Negro pictures more accurately than persons favorable to Negroes. The evidence from these experiments is, then, that the attitude toward the Negro affects perception as well as memory. Zillig found experimentally that women tend to remember more items favorable to women, while men tend to remember more items favorable to men (142). Watson and Hartman demonstrated that a persons attitude toward atheism clearly determined what he would remember of material concerning atheism and theism (133). Edwards showed that the political attitudes of his subjects significantly determined their recognition of items contained in a speech they had previously heard about the New Deal (47). In another study Edwards demonstrated the extent to which a persons attitude stimulates him to rationalize his answers to factual statements with which he disagrees (48). Levine and Murphy have shown experimentally that both the learning and the forgetting of passages that are favorable or unfavorable to the Soviet Union are significantly af-

(9) -fected by a persons attitude toward communism (88a).

The psychology of testimony presents many vivid illustrations of the influence of attitude on observation, judgment, and memory. Sterns early experiments on Aussage showed the suggestive effect of leading questions and demonstrated that the effect of such suggestions often became stabilized to influence later or allied judgments. He also demonstrated how persons tended to describe occurrences from the point of view of what was to them the customary or usual way for such occurrences to take place. The attitude a witness has toward a dispute and a witness own attitudes toward situations were apt to have considerable influence on the testimony of the witness, even though he might be a highly-educated, cautious individual (120, 121).

In an unpublished study by Lazarsfeld and associates of the reaction of individuals of various nationality and racial backgrounds to the same motion picture, the effect of attitudes in determining what will or will not be observed and remembered and what emotional reactions will be aroused in different people by the same stimulus situation is clearly demonstrated. The film was a short British war picture, Naples is a Battleground, shown to the experimental groups in the early summer of 1944. Native white Americans tended to see in the film an example of American armed might and remembered particularly the showing of General Mark Clark; Negroes in the audience paid special attention to the Negro troops pictured; while most marked of all were the reactions of first generation Italians who -were impressed, saddened, or horrified by the ruins and extent of devastation. The experimenters observed that "There were certain aspects of this film that stood out for some but not for others because of their differing mental sets as they watched it" (87).

In a study of rumors current during 1942, Knapp found that "rumors become harmonized with the cultural traditions of the group in which they circulate" (79, p. 30). Rumors are given particular twists as they penetrate groups, where characteristic norms are found, such as a high degree of antisemitism, anti-Negro and anti-British sentiment. He noted, for example, that in the Italian section of Boston rumors of enemy submarines outside Boston Harbor were similar to rumors in other areas of the city except that people in the Italian section believed the submarines were Italian.

Numerous experiments on the level of aspiration have shown how the same performance can be regarded either as a success or a failure depending on the frame of reference in which the performance occurs, as, for example, the study of Gould and Lewis where an individual defined success differently when comparing himself to the supposed performance of college professors than when comparing himself with the supposed performance of WPA workers (57). Similar results were obtained by Hertzman and Festinger (62). Frank has demonstrated that when a level of aspiration is set within a frame of reference clearly involving the ego it is likely to differ from non-ego involving levels of aspiration by being consistently higher than the level of past performance (52, 53). Gould has pointed out the importance of the acceptance of various cultural norms as determinants of level of aspiration and has indicated that the meaning of differences between level of expectation and level of performance for any individual under any given conditions can only be understood with reference to his genetic development (56). In the work of Chapman and Volkmann previously mentioned the

(10) setting of a level of aspiration was studied under two sets of conditions (39). In one experiment involving familiarity with the authors of various literary passages, each of the different experimental groups was given in advance of their direct acquaintance with the task involved the score supposedly made by another group. The score attributed to each group was the same but the groups varied in their prestige: authors and literary critics, students similar to the subjects, and WPA workers. Results here show a clear-cut tendency for those comparing themselves to a superior group to lower their aspiration level, those comparing themselves to an inferior group to raise their aspiration level. In their second experiment, involving a test of mental ability, all subjects were given a test in two sessions before any attempt was made to change their aspiration level by comparing their performance to inferior or superior groups. Under these conditions, no significant change in the level of aspiration was produced; "the subjects own previous scores provided the most effective anchoring" (p. 235). In the conditions of the first experiment, the stimulus situation is unstructured and an imposed norm is accepted as a frame of reference; in the conditions of the second experiment, the stimulus situation has become structured through experience and imposed norms are ineffectual.

Asch, Block and Hertzman in 1938 showed that judgments concerning the characteristics of different professional groups tend to be considerably modified when an evaluation from an authoritative source is introduced (3). A close relationship was found between judgments on different characteristics of various professions indicating that most of the judgments seemed to derive from underlying attitudes. They conclude that "the judgments of a single situation are related to each other by a person in accordance with an underlying attitude of acceptance or rejection" (p. 245), that "a standard having an authoritative source tends to alter an individuals judgment in its direction" (p. 249), that relationship in judgment becomes most clearly established "for situations which are not well defined objectively" (p. 251) and that "the observer, in the absence of objective criteria, and in the face of the necessity of reaching some conclusions, proceeds to arrange a scale of preference in terms of some generally favorable or unfavorable impression" (p. 229).

Kays study of the relationship between personal frames of reference and social judgments indicates that an individuals evaluation of various occupations is considerably more affected by accepted social norms relating to those occupations than by an individuals own preference for an occupation or his experience with it (75). In other words, the social norms concerning the value and characteristics of common occupations are rather uncritically taken over by individuals as personal frames of reference by means of which specific judgments are made. Analyzing her data further according to the source of information concerning the various occupations rated, Kay found that for a third of the occupations cultural sources were mentioned, most frequently while the personal sources uncovered in the interviews were interpreted or reacted to on the basis of personal frames of reference and in the light of existing social norms rather than on the basis of the objective quality of the experiences themselves (76, p. 363). This interpretation confirms Davis study on the attitudes of children in Soviet Russia with its finding that Russian children rated laboring people high, lawyers and bankers low (42). In their examination of the basis of prestige judg-

(11) -ments of various occupational groups, Osgood and Stagner demonstrated that the "decisions about characteristics of occupational stereotypes tend to conform closely to a framework, which is based on the relative prestige of occupations" (106, p. 287). They conclude that "the mere presentation of a set of occupational stereotypes for a series of judgments caused our subjects spontaneously to establish a prestige framework which then determined in a highly reliable manner judgment on the specific traits listed" (p. 289).

Asch found that an individuals judgment of relatively ill-defined and unclear situations could be changed when the imputed judgment of congenial groups was introduced as a reference (4). On the other hand, subjects tended to reject the judgment of antagonistic groups. Here again we have evidence that an individual tends to accept an imposed norm as his own frame of reference for judging a situation when that situation is itself unclear or when he has no preëxisting, sure, or ego-involved frame of reference of his own. Cantril's analysis of radio listeners who followed a particular commentator indicated that the chief function of a news commentator is to provide frames of reference by means of which listeners can judge the plethora of events going on around them (27). The less people know about objective conditions, the more they depend on a commentator to tell them what these events mean and to help them select items to read in their newspapers.

One of the most penetrating studies of the development of prevalent attitudes in our social system is that of Horowitz on the genesis of attitudes toward the Negro (66). He demonstrated unequivocally that this attitude is imposed bodily and uncritically without any basis in experience or knowledge. He concludes "that attitudes toward Negroes are now chiefly determined not by contact with Negroes, but by contact with the prevalent attitude toward Negroes" (p. 35). Horowitz developed a series of ingenious tests to measure attitudes toward the Negro objectively. These tests were administered to children and adolescents in a variety of social groups—children in the rural and urban South, all white groups in New York City, mixed groups in New York City, and a group of New York communist children. His tests showed that children in New York City were just as prejudiced as the children in the South, that children in mixed schools were as prejudiced as those in all white schools, that contact with popular Negro children had no effect on attitudes, and that the only group of children tested who had no prejudice against the Negro were children of communist parents-people devoid of racial prejudices which they would pass on to their children. Furthermore, Horowitz found that Negro boys in mixed schools tended to accept some of the racial attitudes of the white majority of their group.[4] His findings confirm the earlier conclusions of Lasker's well-known study that race attitudes in children are due chiefly to the absorption of adult attitudes (86, p. 371), and that contacts between children of different races are almost in variably influenced by the adult-made environment (p. 371).

In a significant study concerned with the development of stereotypes toward the Negro, Blake and Dennis (12) had students from grades 4 through 11 in a southern school compare whites and Negroes on 60 characteristics. Their results showed that there was less agreement among the younger children who

(12) had a relatively undifferentiated attitude unfavorable to the Negro. They conclude that "the young white child acquires first of all a generally unfavorable attitude toward the Negro, which makes him unwilling to attribute to the Negro any good traits. With increased age and experience, the child gradually learns to apply the adult stereotypes, a few of which are complimentary" (p. 531). The extent to which the accepted attitude toward the Negro affects judgment and goes against all evidence has recently been illustrated by a public opinion survey on a nationwide sample of the adult white population. A majority of those with opinions believe that Negroes are not as intelligent as white people and cannot learn as well if they are given the same education. And a third of our white population believes that Negro blood somehow differs from white blood, a third of the people believes it is the same, and a third say they don't know (104). Murphy and Likert concluded from their study of attitudes toward minority groups that "the individual usually acquires his prejudices against a minority group not primarily from contact with this minority group but chiefly from contact with the prevailing attitude toward this minority group" (98, p. 136).

The bodily acceptance of attitudes is further confirmed by studies which have shown the ease with which people characterize various nationality groups, irrespective of their lack of knowledge about or experience with these groups. Katz and Braly found that the preferential ranking of students for various nationality and racial groups closely followed a weighted ranking of the judgment of a comparable group of students on the stereotype of these same groups (73). Public opinion surveys have shown that less than 10 per cent of the population feels unable to select from  a list of adjectives those that best describe various nationality groups and that people in our culture have generally similar stereotypes concerning the characteristics of major nationality and racial populations (18). The extent to which stereotypes influence judgment has been clearly demonstrated experimentally. Sherif, for example, showed that college students both in the United States and in Turkey were significantly affected in their rating of a literary passage by the name of the author attributed to that passage (113). The passages used in the United States were all by the same author, those used in Turkey by another. The judgments of these passages, then, were made largely in terms of established values. These results are similar to earlier demonstrations of everyday stereotypes shown by Zillig (142) among others, and were confirmed later on still another type of material (24).

It is obvious that the whole psychology of fads and fashions is to be explained largely in terms of accepted norms and values that provide standards of judgment for style, correctness, and certain criteria for status, beauty, significance, and the like (126). In a brilliant note concerning scientific progress at the end of his History of Sensation and Perception, Boring shows that even scientists, including psychologists, tend to conform to the Zeitgeist, are retarded by habits of thought currently fashionable or by laboratory atmosphere (16). It should be pointed out in this connection, however, that atmospheres of psychological laboratories, as well as more common fads and fashions, sooner or later change with new factors imposed by objective conditions or by the accumulation of evidence. The dresses women have for everyday wear in the western world can no longer be styled without reference to the demands of the machine age in

(13) which they live, the Titchenerian atmosphere at Cornell could not remain unaffected by the facts of Würzburg. As facts pile up in psychology or any other science, laboratory atmospheres and schools of thought become increasingly tenuous.

Newcomb measured the change of attitudes of students in a small college community where non-conservative attitudes were considered to be more proper and to carry more prestige than conservative attitudes (103). He found that this community frame of reference (p. 151) significantly influenced the attitudes of students in the liberal direc- tion during their four years, irrespective of the courses studied in college. Newcomb describes this shift of attitude as general rather than as a shift of a series of specific attitudes toward specific issues. When events or new proposals were reacted to, the more advanced students more consistently reacted to them in a liberal way.

In his study of the psychology of status, Hyman has investigated the rôle of attitude in a complicated social setting (69). After learning from an intensive interview of each of his subjects something of the meaning, genesis, criteria of and satisfaction with status, Hyman constructed scales to measure subjective status along several dimensions: general status, economic, intellectual, cultural, social, and physical attractiveness. Subjects were also asked to indicate their subjective status with reference to different groups: the total adult population in the United States, friends and acquaintances, and their occupational group. He found. that "within each status dimension an individuals judgment of his status shifts when reference groups are changed" (p. 49). Among other results reported, the following are particularly significant for our purposes here: individuals strive for status with respect to those accomplishments or characteristics which they most highly value; when an individuals status is approximately similar to the status of the group he is using as a basis for comparison, then he shows no particular concern for his own status, no great drive to achieve a higher status; persons who regard the difference between their own status and a reference group as being determined by a social system they disapprove of also show little dissatisfaction with their own status. Hymans study clearly indicates that an understanding of the psychology of status is possible only if we have a precise knowledge of the frame of reference or anchorage by means of which an individual judges his status.

The way in which an individuals own present income provides a reference point for judging financial needs and aspiration has been demonstrated by Centers and Cantril (36). In a nationwide survey concerned with the relationship of present income to satisfaction and wants, they found that among those people dissatisfied with their present income, the larger this income is, the more additional money is wanted. An analysis by Fried and Cantril of judgments made by persons in different income groups of the income tax people in various income brackets should pay shows that those in the low income groups have such inadequate standards for judging the incomes of people in high income brackets that they tend to find such judgments difficult or impossible to make. Furthermore, when judgments are made by persons in the lower income groups of the tax which those in the highest brackets should pay, the figure given is significantly less than the tax upper income groups think should be paid by the rich, illustrating again the way in which judgment is anchored in individual frames of reference (55).

Cantrils study on identification with social and economic class has shown

(14) that in 1941 almost nine-tenths of the American people identify themselves with the middle class (32). Even 70 per cent of those who feel they are members of the lower income group still call themselves members of the middle social group. There is a strict correspondence between the two class identifications—social and economic—among only 54 per cent of a representative sample of the population. There is, then, a definite tendency for individuals to regard their social class within a frame of reference provided by the norms of their social system and frequently unsupported by the income necessary to solidify their own positions objectively in the social level they accept as their own. The greatest disparity between income and social identification is found among the low income group in 1941. The same study also indicated that those higher in the social class identification tended to base the identification less on strictly economic criteria. Their subjective social status was more affected by norms commonly accepted as high in the social scale such as family background, education, professional accomplishments, and the like. And the evidence indicates that there is considerably more ego-involvement in maintaining a relatively high social identification than a high economic identification (131).

Kornhauser’s investigations of the class attitudes of various socio-economic groups have shown that in our American social system logically opposed interests do not by any means lead to uniformly opposed class attitudes (83, 84). Kornhauser did find, however, that those persons within each economic group who were most dissatisfied in terms of their present status and opportunity did accept less frequently than others the traditional status quo as measured by attitudes concerning labor, government control, etc. And although the attitudes of income and of occupational groups were not diametrically opposed, still significant differences appeared between the attitudes of these groups on a number of social and political issues. In other words, the objective conditions imposing themselves did not fit currently accepted economic and social norms. Particularly significant for our purposes are Kornhauser’s reports of the instances where no significant differences are found between the attitudes of different economic groups. These attitudes—accepted rather uniformly—were those which reflected the traditional American middle class ideology concerning individual opportunity either for the person himself or for his children.

These exploratory investigations of the psychology of class in contemporary America show that although an individuals attitude concerning the political-economic organization of the social system is affected by his circumstances and the norms of his economic group, the fact that the characteristic middle class American ideology is relatively uncritically superimposed on and accepted by individuals in all economic groups, results in the inconsistency of attitude so frequently pointed out, in the difference between attitudes people logically should have and the attitudes they actually do have.

That the attitudes toward labels describing different political systems are by and large uncritically accepted without any real knowledge of the principles or implications of those political systems has been established in a number of different studies. Stagner demonstrated, for example, that people will have a distinctly unfavorable attitude toward fascism but at the same time will accept certain fascist doctrines (118). Menefee comes to the same conclusion in his studies which reveal that if the statement of a political or economic

(15) principle is labelled as fascist it will call forth a much more negative reaction than it will without the label (95). Edwards concludes from his investigations that "some college students have a far greater degree of sympathy for certain fascist principles than might be expected from their otherwise antagonistic reaction to the fascist label" (49, p. 580). Significantly enough, Edwards found that those persons who labelled themselves politically as independents were considerably more critical of statements of fascist principles than those who called themselves Democrats or Republicans or, in other words, those who accepted uncritically the political labels of their own social system. Katz and Cantril found that the attitude toward fascism or communism bore little relationship to the knowledge of fascism or communism among college students in 1939 (74). Their study also shows that, although there was an overwhelming rejection of the terms fascism and communism, the majority of students felt at the time (before World War II had brought the implications of fascism into clearer focus) that fascism was a good thing for Germany and Italy but would be a bad thing for Russia and certainly for the United States, whereas communism was a good thing for Russia but would be a bad thing for Germany and Italy and especially for the United States. All of these studies show, then, that broad symbols tend to be accepted without objective reference.

Analyses of political attitudes have indicated clearly that for the vast majority of modern Americans the way in which they vote is determined chiefly by the political attitude they have accepted rather than by any analysis of different party platforms or candidates. Hartmann has shown that individuals frequently accept the stated principles of political parties they strongly disapprove of so long as they are unaware that these principles have the endorsement of those parties (60). Cantril and Harding discovered that over half of the voting population of this country were completely unable to tell what the differences were between the major political parties even though the overwhelming majority of these same people voted consistently for one or the other of the major political parties (31). Their analysis of the Congressional election of 1942 shows that the overwhelming proportion of the major issues of the day which might logically be thought to have some relationship to the way a person voted and which were discussed during the election campaign as partisan issues actually had little or no importance in determining vote. They also showed how people rationalize their accepted attitude by claiming that they usually vote for the candidate rather than for the party that candidate represents. At least two-thirds of the people who vote a straight party ticket claim that they vote for the man rather than the party. Even the minority of voters who classify themselves as independents appear, on closer scrutiny, to be considerably less independent than they claim. At least half of the independents vote a straight party ticket and well over half of the independents vote according to their fathers political affiliation.

Lazarsfeld, Berelson, and Gaudet have shown that a fairly reliable index of political predisposition can be constructed for an individual from a knowledge of three social factors: religious affiliation, economic status, and place of residence (88). They found that cross pressures of these three social factors tended to delay a persons decision as to how he would vote. Relatively few voters were sufficiently affected by political propaganda to change their political predisposition. Persons tend to expose themselves mainly to the politi-

(16) -cal propaganda of their own party but those whose predisposition is toward one of the major parties and who expose themselves more than others of their persuasion to opposition propaganda, do tend to vote more than others for the opposition party. The less structured a persons political attitude is—either because of lack of interest or social pressures—the greater is the variability of voting behavior, the greater the influence of propaganda from various media or by personal contact.

Breslaw concluded from his detailed interviews concerned with the development of political attitudes that political attitudes can more appropriately be described as an orientation or bias than a point of view logically arrived at (17). "Attitudes emerge from the particular social life which happens to surround the individual" (p. 65). An attitude "is an end product with no necessary relationship to the particular components of that stimulation. An attitude is something that becomes implanted-as fear of the dark becomes implanted-in many different ways" (p. 66).

The consistency with which an individuals attitude determines his reactions to situations to which he relates the attitude has been shown in a number of different studies. Vetter found that a person who was radical or conservative tended to be radical or conservative in his reaction to a wide variety of social, political, and ethical situations (127). Katz and F. H. Allport noted the "consistency with which different attitudes seem to fit together in their respective patterns" (72, p. 48). Cantril demonstrated that an attitude has a directive influence on numerous specific reactions to which it is related, that an attitude tends to be enduring and constant even though the situations which evoke it may fluctuate, and that an attitude can be quite independent of the particular experiences which may have established it (19). Stagner demonstrated that persons who are intensely chauvinistic tend to react consistently and unfavorably to a wide variety of issues such as tariff reduction, labor unions, or government ownership (119). Analyses of data obtained from public opinion surveys show that in the summer of 1940 various attitudes of the American people toward the war in Europe tended to cluster into three broad, underlying patterns (29).

In a study of the reaction of the American people to World War II, we find that while the religious frame of Catholics previous to Pearl Harbor affected their attitude toward U. S. intervention in the European war, it did not affect their attitude toward U. S. intervention in the Pacific war-in the latter case there was no conflict between religious and nationalistic attitudes influencing the judgments of a stimulus situation (34). The conflict of attitudes in German-American and Italian-American citizens resulting from U. S. intervention in World War II and the effect of this conflict in voting behavior is revealed by the findings of Bean, Mosteller and Williams that these two groups significantly shifted their vote away from Roosevelt in the 1940 election (8).

Analysis of data obtained from public opinion polls shows that major events are judged in terms of frames of reference which enable people to relate these events to their own self-interest. Hence public opinion on specific issues is highly sensitive to events (33). Polls further show that once the vast majority of individuals become aware that a social, economic, or political problem exists, they develop attitudes toward that problem-the proportion of people who remain neutral or who have no opinion about an issue they are aware of is very small indeed.


Studies on the prediction of social events have shown the enormous extent to which attitudes direct the way in which the future is projected (25, 94). A persons attitude toward socialism, for example, largely determines his picture of future economic trends. But the more highly structured and clear-cut the forces determining the resolution of an issue appear to be, the less is an individuals prediction determined by his own attitude. Predictions are especially influenced by frames or points of reference when the issue judged is ambiguous or unstructured because of the variety and apparent inconsistency of direction of the variables concerned.

Although the bulk of our attitudes do seem to be derived from the norms that surround us, the fact remains that—all other things being equal—individuals do acquire attitudes based on knowledge and reasoned analysis. Murphy and Likert found, for example, that, next to parental influence, a students reading habits and scholarship tended more than other factors to affect his attitudes (98). Cantril has shown the tremendous difference between enlightened and unenlightened Americans in their pre-Pearl Harbor attitudes toward U. S. intervention in the war-with enlightenment being based on general knowledge and a feeling of a clear idea of what the war is all about (34). Differences of opinion according to the degree of enlightenment were found within all income groups and were considerably greater than differences of opinion between income groups themselves. Evidence from public opinion polls shows that well-informed people accept less readily than uninformed persons many of the common stereotypes of the day, the facts they know serving as reference points for discrimination (139). Informed people are better able to see the implications of events and proposals to their own self-interest-they show considerably more concern, for example, in international affairs as contrasted to relatively uninformed persons whose predominant concern is with strictly domestic problems (35). The uninformed group generally has a comparatively higher no opinion reaction to most public questions. Differences of opinion according to the amount of information a person has do not appear, however, with respect to issues where wish-fulfillment is clearly involved. For example, a study of the relative importance of education and economic status in determining opinion showed that education was more important than income only in situations where greater knowledge gives an insight into the effect of certain events or proposals that do not deal with an individuals financial return in any clear-cut way (140). Divens analysis of aesthetic appreciation indicates that-all other things being equal-if a well-informed and an uninformed person hold the same attitude with equal intensity the essential difference between them is that the informed man will be better able to rationalize his point of view, supporting it with what seems to him sound evidence (45).

Cantril’s study of the nation-wide panic resulting from Orson Welles broadcast War of the Worlds showed that when individuals are faced with a critical and apparently dangerous situation they readily accept any interpretation offered them as a basis for judgment if they have no appropriate and sure standards by means of which to evaluate the situation (28). Those frightened by the broadcast were highly suggestible, believing what they heard to be true and being unable themselves to make any external or internal checks of the reports. Furthermore, for most of those who became panicked, the story was credible since it fit into preëxisting attitudes such as the belief that God

(18) would someday destroy our planet, an attack by a foreign power, or fanciful notions concerning the possibilities of science. People who lacked appropriate standards to interpret the broadcast properly were found particularly among those who had neither the opportunity nor the ability to acquire information or training that would have protected them with relevant points of reference. This broadcast occurred in the fall of 1938 when vast numbers of the population were unusually bewildered by the prolonged economic insecurity they had experienced and by the precarious and delicate state of world affairs following the Munich settlement. In commenting on the panic, Heywood Broun tersely and aptly summarized this effect of general, social and political unrest by saying that "jitters have come to roost." Current norms were somehow proving inadequate to account for objective conditions, the whole course of recent history created a relatively ambiguous and unstructured situation conducive to high suggestibility.

And just as critical conditions provide fertile soil for panics, so too do they provide the optimum conditions for rumor or for individual re-orientation by means of slogans or simple appeals. Knapp concludes, for example, from his study of rumors that "in proportion as the cognitive world is ambiguous or ill defined and the motivation intense, rumors will find life" (79, p. 31). F. H. Allport and Lepkin in their study of rumors point out that "the more outer facts, or true reports of facts, the individual has within his grasp, and the more he is stimulated to weigh this evidence objectively, the more nearly the picture he forms in his mind will conform to the true reality, and the less altered it will be through the effect of emotion and impulse" (1, p. 14). And Sherif points out from his analysis of slogans "that slogans are short-cut expressions arising in confused and critical situations . . . the more correctly and the more objectively a set of slogans expresses the underlying forces in a critical situation, the more vital and lasting they will prove to be" (116, p. 461). In a more recent study of slogans, Bellak reaches the same conclusion with special emphasis on the rôle of motivational factors (10). Since slogans can only be expected to take hold in critical times when they give meaning to or point a way out of confused situations, it is not surprising that the judgments of slogans in the laboratory are unaffected by conflicting standards as in the experiment of Asch, Block and Hertzman (3) or Blocks later study where only restricted shifts were found in the ratings of slogans when authoritative standards were imposed under laboratory conditions (89). The leaders of any mass movement or revolution show sound psychology when, during a critical situation where old norms have lost their hold, they try first of all to get control of the mass media of communication so they can issue new instructions, spread slogans, and otherwise try to restructure peoples thinking.

Cantril’s analysis of the rise of various social movements shows how persons dissatisfied with their status or the fulfillment of their needs tend to accept new frames of reference provided by a leader or nuclear group which seem to them to explain their situation more appropriately and to offer an apparently more effective course of action than did adherence to the more commonly accepted norms of the social system (30). The suggestibility of a person to new norms was found to be proportional to the inadequacy of existing standards for the interpretation of particular situations, to his desire for a more adequate interpretation or to the ease with which

(19) the norms of a new movement could be related to established frames.

All of the representative studies cited above indicate that only by means of some frame or anchorage can and does the individual judge and react to social stimuli. Individuals in social life cannot long remain normally adjusted if they are in a state of indecision. Sooner or later they must and do make some appropriate judgment or reaction to a stimulus, place it in some way meaningful to them. The psychological process they use to make such judgments and give meaning to their social environment is to refer the stimuli around them to some frame of reference or anchorage they acquire and which is a readiness for reaction. These frames or anchorages which make everyday social judgments possible are, like frames or anchorages discovered in laboratory experiments, inevitably developed with repeated exposure to the same stimuli (objects, persons, groups, values, or norms) and individuals are by no means always aware that frames or anchorages have become established.


The established conformities in the experience and behavior of individual members of a group may be expressed in an empirical way and on the psychological level as the effects of established attitudes. Because of this, attitudes become a major meeting ground for psychologists and sociologists. In spite of the vast literature on attitudes contributed by both psychologists and sociologists, there is as yet no recognized psychology of attitudes with basic concepts applicable to all cases of attitudes. In large part, this may be due to the fact that the term attitude, though useful as an empirical and composite term, is not yet delineated by underlying psychological concepts. An attitude does not designate any specific psychological principle in operation.

Psychologically, an attitude implies an established state of readiness. But the cases of psychological readiness are not exhausted by all cases of attitudes. An attitude has certain characteristic features or criteria which differentiate it from other states of readiness: an attitude is an established readiness which has a subject-object relationship of highly variable content, which is learned (formed), has affective properties with various degrees of motivational components, may refer to whatever stimuli are encompassed in the subject-object relationship, and which determines that an individual will react to a stimulus in a selective way.

Thus characterized, the psychology of value is included in the psychology of attitudes. And since there are states of readiness possessing the above criteria and established in strictly personal situations, all attitudes are not shared by other members of the group (individual attitudes). Hence the meaning of a statement above that the established conformities of members of a group may be thought of as the effects of established attitudes should be modified accordingly. Since there are as many different kinds of attitudes as there are stimulus situations or norms to which they are related, a rigid classification of attitudes becomes meaningless.

The first stage in the actual formation of an attitude is a perceptual stage and the psychology of attitudes is related to the psychology of perception. And since attitudes are inferred from the selective nature of response (verbal or non-verbal), the psychology of judgment also lies at the basis of attitude formation. Perceptions and judgments take place in referential frameworks. The properties of perceptions and judgments are determined or altered by the

(20) properties of the scales or frames to which they are related. An individual inevitably forms such scales or frames if he faces a stimulus situation repeatedly. He may or may not be consciously aware of this fact. But once such scales or frames are formed, he reacts to subsequent stimulation by related stimuli in a characteristic or selective way. In other words, these established scales or frames serve as anchorages to structure or modify subsequent experience or response.

In cases where the stimulus field is well-structured, the frames or scales formed are determined in a compelling way by the objective properties of the stimulus situation and correspond to them. The frames formed by the products of technology are of this kind, determining the mentality of individuals to an important degree. In cases where the stimulus situation is not well-structured, the psychological result is not chaos but again the formation of a frame. In these cases, internal factors (e.g., temperament, motivation) and social factors (e.g., suggestion, prestige and other social influences) contribute in an important degree to the structuration of the frame. The rôle played by internal and social factors decreases with the stability, unambiguity, or structuredness of the stimulus situation, and increases with its instability or ambiguity. But whether the stimulus field is structured or not, the result of repeated stimulation is the formation of a frame.

Studies on perception and judgment furnish ample experimental evidence of the formation of scales and frames of reference and the same basic psychological principles are operative in cases of frame formation where social factors play an important rôle. The effects of anchorages which come in to structure or modify experience are demonstrated with social factors of varying degrees of concreteness as well as with nonsocial factors. Individuals in social life are sooner or later bound to make some decisions with respect to a stimulus situation that confronts them. They do this by referring the stimulus to some frame or standard. Whether or not the individual is aware of it, these frames or standards invariably develop in social life with repeated exposure to the same stimuli. Many of these frames have all the characteristics of attitudes and, once formed, attitudes act as anchorages to related situations and thus have a selective property as inferred from verbal or non-verbal responses.

The more ambiguous and unstructured the social stimulus situation is, the more will established frames or anchorages exert an influence in determining reaction. If the stimulus situation is ambiguous and there are no established frames to which it can be related, new norms will be readily accepted and will effectively structure peoples thinking. In concrete social life, critical periods of unrest, such as prolonged depressions or revolutions, are examples par excellence of ambiguous, unstable situations. During these times, old norms and values are inadequate and new anchorages, such as slogans, provide an apparent basis for judgment or action. This new formulation may or may not account for the critical situation objectively. But the study of the adequacy and validity of the formulation is one that goes beyond psychology alone.

The content and affective properties of attitudinal anchorages will vary according to the motivational and other factors involved in the specific case of the subject-object relationship. In most concrete social situations of everyday life the contents of attitudes have affective, motivational, or status-involving factors of varying degrees of intensity. Whether the perceptual stage in the

(21) formation of a social attitude is based on direct contact with a person or group, or whether it is based on the transmission of a social norm by short-cut verbal value-judgment makes no difference psychologically. Hence the basic psychological properties of attitudes—quite irrespective of what the attitude is concerned with or whether it is studied by the experimental psychologist or the sociologist-are essentially the same.


1. ALLPORT, F. H., & LEPKIN, M. Wartime rumors of waste and special privilege. J. abnorm. & soc. Psychol., 1945, 40, 3-36.

2. ALLPORT, G. W. Attitudes. Chap. 17 in A handbook of social psychology (Ed. by C. Murchison). Worcester: Clark Univ. Press, 1935.

3. ASCH, S. E., BLOCK, H., & HERTZMAN, M. Studies in the principles of judgments and attitudes: I. Two basic principles of judgment. J. Psychol., 1938, 5, 219251.

4. ASCH, S. E. Studies in the principles of judgments and attitudes: II. Determination of judgments by group and by ego standards. J. soc. Psychol., 1940, 12, 433-465.

5. BAIN, R. An attitude on attitude research. Amer. J. Sociol., 1928, 33, 940957.

6. BALDWIN, J. M. Dictionary of philosophy and psychology. (3 vols.) New York: Macmillan, 1901-1905.

7. BARTLETT, F. C. Remembering: a study in experimental and social psychology. Cambridge: Univ. Press, 1932.

8. BEAN, L. H., MOSTELLER, F., & WILLIAMS, F. Nationalities and 1944. Publ. Opin. Quart., 1944, 8, 368-375.

9. BEERE-CENTER, J. G. Pleasantness and unpleasantness. New York: Van Nostrand, 1932.

10. BELLAK, L. The nature of slogans. J. abnorm. & soc. Psychol., 1942, 37, 496510.

11. BENARY, W. The influence of form on brightness contrast. In A source book of Gestalt psychology (Ed. by W. D. Ellis). New York: Harcourt, Brace, 1938, 104-108.

12. BLAKE, R., & DENNIS, W. The development of stereotypes concerning the Negro. J. abnorm. & soc. Psychol., 1943, 38, 525-531. .

13. BLONDEL, C. Introduction d la psychologie collective. Paris: Colin, 1928.

14. BOGARDUS, E. S. Fundamentals of social psychology. (2nd ed.) New York: Century, 1931.

15. BORING, E. G. Attribute and sensation. Amer. J. Psychol., 1924, 35, 301-304.

16. ——. Sensation and perception in the history of experimental psychology. New York: Appleton-Century, 1942.

17. BRESLAW, B. J. The development of a socio-economic attitude. Arch. Psychol., 1938, No. 226.

18. BRUNER, J. S. Mandate from the people. New York: Duell, Sloan & Pearce, 1944.

19. CANTRIL, H. General and specific attitudes. Psycho!. Monogr., 1932, No. 192.

20. ——. Attitudes in the making. Understanding the child., 1934, 4, 13-15.

21. ——. The social psychology of everyday life. Psycho!. Bull., 1934, 31, 297-330.

22. ——. A psychological reason for the lag of non-material culture traits. Soc. Forces, 1935, 13, 376-379.

23. ——. The effect of modern technology and organization upon social behavior. Soc. Forces, 1937, 15, 493-495.

24. ——. Experimental studies of prestige suggestion. Psycho!. Bull., 1937, 34, 528.

25. ——. The prediction of social events. J. abnorm. & soc. psycho!., 1938, 33, 364389.

26. ——, & KATZ, D. Objectivity in the social sciences. In Industrial conflict: a psychological interpretation (Edit. by G. W. Hartmann & T. Newcomb). New York: Cordon, 1939, 9-18.

27. ——. The rôle of the radio commentator. Publ. Opin. Quart., 1939, 3, 654662.

28. ——. The invasion from Mars. Princeton: Univ. Press, 1940.

29. ——. America faces the war: a study in public opinion. Publ. Opin. Quart., 1940, 4, 387-407.

30. ——. The psychology of social movements. New York: Wiley, 1941.

31. ——, & HARDING, J. The 1942 elections: a case study in political psychology. Publ. Opin. Quart., 1943, 7, 222-241.

32. - ——. Identification with social and economic class. J. abnorm. & soc. Psychol., 1943, 38, 74-80.

33. - ——. The use of trends. In Gauging public opinion (Edit. by H. Cantril). Princeton: Univ. Press, 1944, ch. 16.

34. - ——. The use of breakdowns. In Gauging public opinion (Edit. by H. Cantril). Princeton: Univ. Press, 1944, chap. 13.


35. - ——. The issues—as seen by the American people. Publ. Opin. Quart., 1944, 8, 331-347.

36. CENTERS, R., & CANTRIL, H. Income satisfaction and income aspiration. J. abnorm. & soc. Psychol. (To be published.)

37. CHANDLER, A. R. Beauty and human nature. New York: Appleton-Century, 1934.

38. CHAPMAN, D. W. Relative effects of determinate and indeterminate Aufgaben. Amer. J. Psychol., 1932, 44, 163-174.

39. CHAPMAN, D. W., & VOLKMANN, J. A social determinant of the level of aspiration. J. abnorm. & soc. Psychol., 1939, 34, 225-238.

40. COFFIN, T. E. Some conditions of suggestion and suggestibility: a study of some attitudinal and situational factors influencing the process of suggestion. Psychol. Monogr., 1941, No. 241.

41. COHEN, N. E. The relativity of absolute judgments. Amer. J. Psychol., 1937, 49, 93-100.

42. DAVIS, J. Testing the social attitudes of children in the government schools in Russia. Amer. J. Social., 1927, 32, 947-952.

43. DEARBORN, G. V. Blots of ink in experimental psychology. PSYCHOt. REV., 1897, 4, 390-391.

44. - —— A study of imaginations. Amer. J. Psychol., 1898, 9, 183-190.

45. DIVEN, K. Aesthetic appreciation test. In Explorations in personality (H. A. Murray et al). New York: Oxford, 1938, 447-453.

46. DURKHEIM, E. Les formes élémentaires de la vie religieuse. Paris, 1912.

47. EDWARDS, A. L. Political frames of reference as a factor influencing recognition. J. abnorm. & soc. Psychol., 1941, 36, 34-50.

48. ——. Rationalization in recognition as a result of a political frame of. reference. J. abnorm. & soc. Psychol., 1941, 36,


49. ——. Unlabeled fascist attitudes. J. abnorm. & soc. Psychol., 1941, 36, 575582.

50. FERNBERGER, S. W. Instructions and psychophysical limen. Amer. J. Psychol., 1931, 43, 361-376.

51. FOLSOM, J. K. Social psychology. New York: Harper, 1931.

52. FRANK, J. D. Some psychological determinants of the level of aspiration.

Amer. J. Psychol., 1935, 47, 285-293.

53. ——. The influence of the level of performance in one task on the level of aspiration in another. J. exp. Psychol., 1935, 18, 159-171.

54. FRANKLIN, J. H. The free Negro in North Carolina, 1790-1860. Chapel Hill: N. Carolina Press, 1943.

55. FRIED, E., & CANTRIL, H. The meaning of questions. In Gauging public opinion (Edit. by H. Cantril). Princeton: Univ. Press, 1944, ch. 1.

56. GOULD, R. An experimental analysis of level of aspiration. Genet. Psychol. Monogr., 1939, 21, 1-116.

57. ——, & LEWIS, H. B. An experimental investigation of changes in the meaning of level of aspiration. J. exp. Psychol., 1940, 27, 422-438.

58. HAGGARD, E., & ROSE, G. J. Some effects of mental set and active participation in the conditioning of the autokinetic phenomenon. J. exp. Psychol., 1944, 34, 45-59.

59. HALBWACHS, M. Les cadres sociaux de la mémoire. Paris, 1925.

59a. HARTLEY, R., & HARTLEY, E. Analysis of the judgment of pictures of people in a poorly defined situation. (To be published.)

60. HARTMANN, G. W. The contradiction between the feeling-tone of political party names and public response to their platforms. J. soc. Psychol., 1936, 7, 336-357.

61. HENRI, V. Recherches sur la localisation des sensations tactiles. Année psychol., 1895, 2, 168-192.

62. HERTZMAN, M., & FESTINGER, L. Shifts in explicit goals in a level of aspiration experiment. J. exp Psychol., 1940, 27, 439-452.

63. HOLLINGWORTH, H. L. The inaccuracy of movement. Arch. Psychol., 1909, No. 13.

64. ——. The central tendency of judgment. J. Phil., 1910, 7, 461-469.

65. HOPPE, F. Erfolg and Misserfolg. Psychol. Forsch., 1930, 14, 1-62.

66. HOROWITZ, E. L. The development of attitude toward the Negro. Arch. Psychol., 1936, No. 194.

67. HUNT, W. A., & VOLKMANN, J. The anchoring of an affective scale. Amer. J. Psychol., 1937, 49, 88-92.

68. HUXLEY, J. Science and its relation to social needs. In Scientific progress (Sir Halley Stewart Lecture, 1935). London: Allen & Unwin, 1935, ch. 6.

69. HYMAN, H. H. The psychology of status. Arch. Psychol., 1942, No. 269.


70. JAMES, W. Will to believe and other essays in popular philosophy. New York: Longmans, Green, 1896.

71. JUNG, C. G. Psychological types. New York: Harcourt, Brace, 1924.

72. KATZ, D., & ALLPORT, F. H. Students attitudes. Syracuse: Craftsman, 1931.

731 ——, & BRALY, K. W. Racial prejudice and racial stereotypes. J. abnorm. & soc. Psychol., 1935, 30, 175-193.

74. ——, & CANTRIL, H. An analysis of attitudes toward fascism and communism. J. abnorm. & soc. Psychol., 1940, 35, 356-366.

75. KAY, L. W. The relation of personal frames of reference to social judgments. Arch. Psychol., 1943, No. 283.

76. ——. Social norms as determinants in the interpretation of personal experiences. J. soc. Psychol., 1944, 19, 359367.

77. KLOPFER, B., & KELLY, D. M. The Rorschach technique. Yonkers: World Book, 1942.

78. KLJNEBERC, O. Social psychology. New York: Holt, 1940.

79. KNAPP, R. A psychology of rumor. Publ. Opin. Quart., 1944, 8, 22-37.

80. KOFFKA, K. Perception: an introduction to the Gestalt-Theorie. Psychol. Bull., 1922, 19, 531-585.

81. KÖHLER, W. Gestalt psychology. New York: Liveright, 1929.

82. ——. Psychological remarks on some questions of anthropology. In Amer. J. Psychol., Golden Jubliee vol. (Edit. by K. M. Dallenbach), 1937, 50, 271-288.

83. KORNHAUSER, A. W. Attitudes of economic groups. Publ. Opin. Quart., 1938, 2, 260-268.

84. ——. Analysis of class structure of contemporary American society-psychological bases of class divisions. In Industrial conflict: a psychological interpretation (Edit. by G. W. Hartmann & T. Newcomb). New York: Cordon, 1939, 199-264.

85. KüLPE, O. Versuche über Abstraktion. Ber. 1. Kongr. Psychol., 1904, 58-68.

86. LASKER, B. Race attitudes in children. New York: Holt, 1929.

87. LazARsFELD, P. F. Report on "Naples is a battleground." (Unpublished.) June 1944.

88. -, BERELSON, B., & GAUDET, H. The peoples choice: how the voter makes up his mind in a presidential campaign. New York: Duell, Sloan, & Pearce, 1944.

88a. LEVINE, J. M., & MURPHY, G. The learning and forgetting of controversial material. J. abnorm. & soc. psychol., 1943, 38, 507-517.

89. LEWIS, H. BLOCK. Studies in the principles of judgments and attitudes: II. The influence of political attitude on the organization and stability of judgments. J. soc. Psychol., 1940, 11, 121146.

90. LONG, L. A study of the effect of preceding stimuli upon the judgment of auditory intensities. Arch. Psychol., 1937, No. 209.

91. LUCHINS, A. S. On agreement with anothers judgment. J. abnorm. & soc. Psychol., 1944, 39, 97-111.

92. ——. Social influences on perception of complex drawings. J. soc. Psychol., 1945, 21, 257-273.

93. McGARVEY, H. R. Anchoring effects in the absolute judgment of verbal materials. Arch. Psychol., 1943, No. 281.

94. MCGREGOR, D. The major determinants of the prediction of social events. J. abnorm. & soc. Psychol., 1938, 33, 179204.

95. MENEFEE, S. C. The effect of stereotyped words on political judgments. Amer. social. Rev., 1936, 1, 614-621.

96. MURPHY, G., & MURPHY, L. B. Experimental social psychology. New York: Harpers, 1931.

97. ——, & NEWCOMB, T. M. Experimental social psychology. New York: Harpers, 1937.

98. MURPHY, G., & LIKERT, R. Public opinion and the individual. New York: Harpers, 1938.

99. MURRAY, H. A., et al. Explorations in personality. New York: Oxford, 1938.

100. NELSON, E. Attitudes: I. Their nature and development. J. gen. Psychol., 1939, 21, 367-399.

101. ——. Attitudes: II. Social attitudes. J. gen. Psychol., 1939, 21, 401-416.

102. ——. Attitudes: III. Their measurement. J. gen. Psychol., 1939, 21, 417-436.

103. NEWCOMB, T. M. Personality and social change. New York: Dryden, 1943.

104. National Opinion Research Center. Survey, May 1944.

105. ORTH, J. Gefühl and Bewusstseinslage. Abhand. Geb. Padagog. Psychol. u. Physiol., 1903.

106. OSGOOD, C. E., & STAGNER, R. Analysis of a prestige frame of reference by a gradient technique. J. appl. Psychol., 1941, 25, 275-290.

107. POSTMAN, L., & MILLER, G. A. Anchoring of temporal judgments. Amer. J. Psychol., 1945, 58, 43-53.

108. PRATT, C. C. Time-errors in the method of single stimuli. J. exp. Psychol., 1933, 16, 798-814.

109. ROGERS, S. The anchoring of absolute judgments. Arch. Psychol., 1941, No. 261.

110. SCHONRAR, R. A. Some social factors in judgment: the formation of social norms in the judgment of structured stimuli and esthetic preference. Ph.D. thesis, 1945, Columbia University Library.

111. SEELEMAN, V. The influence of attitude upon the remembering of pictorial material. Arch. Psychol., 1940, No. 258.

112. SHELDON, W. H. The varieties of temperament. New York: Harpers, 1942.

113. SHERIF, M. A study in some social factors in perception. Arch. Psychol., 1935, No. 187.

114. ——. The psychology of social norms. New York: Harpers, 1936.

115. ——. An experimental approach to the study of attitudes. Sociometry, 1937, 1, 90-98.

116. ——. The psychology of slogans. J. abnorm. & soc. Psychol., 1937, 32, 450461.

117. SPRANGER, E. Lebensformen. Halle: Niemeyer, 1927.

118. STAGNER, R. Fascist attitudes: an exploratory study. J. soc. Psychol., 1936, 7, 309-319.

119. ——. A correlational analysis of nationalistic opinions. J. soc. Psychol., 1940, 12, 197-212.

120. STERN, W. Beiträge zur Psychologie der Aussage. Leipzig: Barth, 1903-1906.

121. ——. The psychology of testimony. J. abnorm. & soc. Psychol., 1939, 34, 3-20.

122. STRAUSS, A. The concept of attitude in social psychology. J. Psychol., 1945, 19,


123. THOMAS, W. I., & ZNANIECKI, F. The Polish peasant in Europe and America. Boston: Badger, 1918, vol. 1.

124. TITCHENER, E. B. Systematic psychology: prolegomena. New York: Macmillan, 1929.

125. TRESSELT, M. E., & VOLKMANN, J. The production of uniform opinion by nonsocial stimulation. J. abnorm. & soc. Psychol., 1942, _37, 234-243.

126. VERLEN, T. Theory of the leisure class. New York: Macmillan, 1899.

127. VETTER, G. B. Measurement of social and political attitudes. J. abnorm. & soc. Psychol., 1930, 25, 149-189.

128. VOLKMANN, J. The anchoring of absolute scales. Psychol. Bull., 1936, 33, 742-743.

129. VOTH, A. Individual differences in the autokinetic phenomenon. J. exp. Psychol., 1941, 29, 306-322.

130. ——. Psychologists report. In 33rd biennial report (Topeka State Hospital), 1942, 16-20.

131. WALLACE, J. M., JR., WILLIAMS, F. W., & CANTRIL, H. Identification of occupational groups with economic and social class. J. abnorm. & soc. Psychol., 1944, 39, 482-485.

132. WARREN, H. C. (Ed.). Dictionary of psychology. Boston: Houghton, Mifflin, 1934.

133. WATSON, W. S., & HARTMANN, G. W. Rigidity of a basic attitudinal frame. J. abnorm. & soc. Psychol., 1939, 34, 314-336.

134. WATT, H. J. Experimentelle Beiträge zu einer Theorie des Denkens. Leipzig, 1904.

135. WEDELL, C. H. The nature of the absolute judgment of pitch. J. exp. Psychol., 1934, 17, 485-503.

136. WERTHEIMER, M. Untersuchungen zur Lehre von der Gestalt. Psychol. Forsch., 1923, 301-350.

137. ——. Drei Abhandlungen zur Gestalttheorie. Erlangen: Phil. Akad., 1925.

138. WEVER, E. G., & ZENER, E. Method of absolute judgment in psychOphysics. PSYCHOL. REV., 1928, 35, 466-493.

139. WILLIAMS, F. W. Information as a determinant of opinion. In Gauging public opinion (Edit. by H. Cantril). Princeton: Univ. Press, 1944, ch. 15.

140. ——, & MOSTELLER, F. Education and economic status as determinants of opinion. In Gauging public opinion (Edit. by H. Cantril). Princeton: Univ. Press, 1944, ch. 14.

141. WOODWORTH, R. S. Situation-and-goal set. In Amer. J. Psychol., Golden Jubilee vol. (Edit. by K. M. Dallenbach), 1937, 50, 130-140.

142. ZILLIG, M. Einstellung und Aussage. Z. Psychol., 1928, 106, 58-106.


  1. Here we shall not digress to indicate some interesting results which revealed the differential, relative contributions of individual members. These points are important for the psychology of personality and will be considered in a forthcoming volume.
  2. This is a summary statement of investigations the full account of which will be reported in a forthcoming publication by Dr. S. Asch. The results reported here were personally communicated.
  3. We might incidentally call attention here to the fact that most experiments on humans in the psychological laboratory are social situations at least to the extent that they involve verbal material in the instructions given to subjects. The variety of possible meanings of stimulus situations persistently gives trouble since meanings so readily intrude themselves (50, 109, 141). It is well known that a subjects set or attitude can profoundly affect his reactions in psychological experiments. It will also be remembered that Titchener, in trying to create the proper set, advocated the use of trained observers. From our present vantage point these results obtained only from observers trained to rule out objective references seem highly artificial.
  4. This is reminiscent of the fact that in the days of slavery in this country there was a sizeable number of freed Negro slaves who, when they acquired land of their own, retained Negro slaves to work it (54).

Valid HTML 4.01 Strict Valid CSS2