Hovland, Carl I.
Hovland, Carl I.
Carl I. Hovland (1912–1961), American pioneer in communications research, began his career as an experimental psychologist working on classical problems of conditioning and human learning. By the age of 30, when he turned to the newly developing field of research on attitude change, he had already become one of the most eminent psychologists of his generation.
The most important of Hovland’s early research studies were focused on the generalization of conditioned responses. During the 1930s, he also made significant discoveries concerning factors that influence reminiscence effects in human memory functioning, the efficiency of alternative methods of rote-learning, and the modes of resolution of motor conflicts. From 1942 until his untimely death from cancer in 1961, Hovland devoted the major part of his time to careful investigations of the effects of social communication, using research designs and analytic methods derived from the more highly developed fields of experimental psychology. It is primarily for his contributions in this area that he is regarded as one of the foremost social scientists of the twentieth century. Wilbur Schramm (1963, p. 5), in reviewing communications research in the United States, refers to the work that came out of Hovland’s research program at Yale University between 1950 and 1961 as “the largest single contribution … [to this field] any man has made.” The Distinguished Scientific Contribution Award was presented to Hovland by the American Psychological Association, in 1957, “for his original and provocative contributions to the scientific study of persuasive communications and the modification of beliefs and attitudes.” The citation states further:
…Combining a sensitive use of controlled experimentation with penetrating logical analysis, he has done much to isolate the major factors at work when an individual is confronted with the complex informational input of a persuasive argument. By judicious use of psychological theory, he has been able to relate this area of social psychology to basic investigations of the higher mental processes. His work has been of central importance in advancing attitude research from the early stage of merely demonstrating that changes can be produced to the point of making predictions about when and where they will occur. His work has provided a convincing demonstration of the values of a sustained and integrated program of research. (American Psychological Association 1958, p. 158)
Formative years. A native of Chicago, Hovland attended nearby Northwestern University, where he devoted himself to acquiring as thorough a background as possible in mathematics, physics, and biology, as well as in experimental psychology. After obtaining his m.a. degree in 1933, he completed his graduate studies in psychology at Yale University. He remained affiliated with Yale throughout his entire academic career, starting as an instructor in 1936 (immediately after receiving his ph.d. degree), attaining the rank of professor of psychology in 1945 and the chair of Sterling professor two years later.
As a graduate student and junior faculty member during the prewar years at Yale, Hovland participated in the stimulating intellectual environment of the Yale Institute of Human Relations, which helped to shape his interests and approach to the study of human behavior. Of particular importance in Hovland’s training was the influence of the great American psychologist Clark L. Hull. Using a rigorous empirical approach in conjunction with analytic theory construction, Hull was highly successful during the late 1930s in organizing and stimulating talented young psychologists at Yale to carry out research on significant problems of motivation and learning. After serving as Hull’s research assistant for several years, Hovland became a coinvestigator in the series of studies on human learning, which led to his being a coauthor of the well-known book by Hull and his collaborators, Mathematico-Deductive Theory of Rote Learning (1940). Although not sharing Hull’s predilection for far-reaching theoretical formulations, Hovland acquired an extraordinary degree of methodological sophistication, both from Hull himself and from other specialists whom Hull had recruited to participate in his research program. Of equal importance was the optimistic vision he acquired that led him to extend the analytic approach of experimental psychology to other research areas in the human sciences—particularly those suffering from a dearth of dependable generalizations in the midst of an abundance of vague theoretical speculations.
After he had received his ph.d. in 1936, Hov land’s outlook and approach to social science research continued to be fostered by collaboration with the staff of Yale’s Institute of Human Relations, which was at the height of its influence during the late 1930s and early 1940s. Outstanding social scientists from all over the world were brought together and given ample time and resources to pursue the inquiries of their choice. Hopes ran high that this would make for rapid cross-fertilization among traditionally isolated fields and lead to vital new breakthroughs, comparable to those emerging from interdisciplinary developments in the physical and biological sciences. Among the outstanding personalities with whom Hovland came in contact were Dusser de Barenne, Mark May, Walter Miles, Edward Sapir, and Robert Yerkes.
Although the senior members of the institute rarely achieved their high aspirations for interdisciplinary advances, the intellectual ferment created among the research assistants and junior staff members of Hovland’s generation did produce unexpected gains. Coming from different social science disciplines, these well-trained young men began to influence each other as they examined the implications of generalizations that purported to account for complex aspects of human behavior. Among Hovland’s contemporaries at the institute were John Dollard, Leonard Doob, Clellan S. Ford, Neal Miller, O. Hobart Mowrer, George P. Murdock, Robert R. Sears, and John W. M. Whiting. With these men he formed bonds of personal friendship and often participated with them in lively seminars. One well-known product was the collaborative Yale volume Frustration and Aggression (Dollard et al. 1939). Several members of this group, together with Donald Marquis, Ernest R. Hilgard, and Kenneth W. Spence, who were also at Yale during the early 1940s, played an important role in the development of learning theory.
Following the lead of Hull, the Yale group attempted to formulate unambiguous behavioral laws concerning the conditions under which habits are strengthened and weakened. These laws were then used as a basis for explaining complex social phenomena, such as the displacement of hostility from the family to outsiders, observed by specialists in such diverse fields as anthropology, psychoanalysis, and social psychology. Hovland contributed to the work of this group not so much by suggesting comprehensive theoretical insights as by focusing on rigorous analysis of empirical evidence. His originality took the form of discovering new functional relationships by working closely with the available findings, noting inconsistencies and reversals that others might be inclined to overlook, and then proceeding to unravel the puzzles by ingeniously testing a series of alternative explanations with a new set of data. These qualities also characterized his later work on communication effects.
Research on mass communication. In 1942 Hovland took a leave of absence from Yale in order to serve as a research expert on morale problems for the United States government. He became chief psychologist and director of experimental studies in the research branch of the information and education division of the War Department. In this role, he worked closely with two eminent sociologists, Samuel Stouffer, who was then the research director of the research branch, and Leonard C. Cottrell, senior social analyst in the same organization. For four years Hovland participated in the planning of a series of large-scale investigations on social psychological factors in military morale; the empirical findings from these studies were subsequently incorporated into the American Soldier volumes by Stouffer and his collaborators.
Hovland’s main role in the military research organization, however, was to conduct psychological experiments on the effectiveness of training and information programs, including the series of “Why We Fight” films that were intended to influence the motivation of men in the American armed forces. In his own experimental section of the research branch, Hovland assembled a group of six psychology graduate students, who worked with him on these studies for several years: John Finan, Irving L. Janis, Arthur A. Lumsdaine, Nathan Maccoby, Fred D. Sheffield, and M. Brewster Smith. Although partly oriented toward meeting the practical needs of the military services, the studies conducted by Hovland and his group embodied a research approach that led to major advances on many basic problems in social psychology.
Following the pattern of his earlier work at Yale, Hovland set up investigations designed to test hypotheses concerning the conditions under which mass communications are effective and to explore fully the implications of all the relevant data. But instead of confining the research to restricted laboratory settings, the mainstay of experimental social psychology up to that time, Hovland took advantage of the unique opportunities afforded by his military research mission. He and his group investigated the effects of different types of communication on “live” issues by conducting experimental studies with equated groups of soldiers at U.S. Army training centers. One of the most widely cited of these pioneering communications experiments involved testing the effects of a one-sided versus a two-sided presentation of a controversial issue. The results contradicted some of the well-publicized contentions of Nazi propaganda strate gists who claimed that to be successful a communication should never mention the opposing side of an argument. Among men initially hostile to the point of view fostered by a communication (and particularly those familiar with cogent opposing arguments), it was found to be more effective to include mention of the opposing arguments than to give a strictly one-sided presentation.
Many of the investigations by Hovland and his group provide systematic data bearing on the sources of audience resistance to persuasive efforts and call attention to factors that help to overcome such resistance. These wartime studies formed the basis for a book entitled Experiments on Mass Communication by Hovland, Lumsdaine, and Shef field (1949), which was published as part of the same series as the American Soldier volumes, jointly sponsored by the U.S. War Department and the Social Science Research Council.
The Yale communication studies. After the war Hovland returned to Yale University as chairman of the department of psychology and was awarded a Sterling professorship. Having recruited for his department several members of his wartime research team, Hovland continued to devote his energies to systematic research on communication effects. With the support of the Rockefeller Foundation, he organized and directed the Yale studies in attitude and communication, which enabled a large number of junior faculty members and gradu ate students to participate in collaborative research on a variety of communication problems of their own choice.
The main purpose of the research project was to explore systematically the factors that influence the effectiveness of social communications. Hovland himself continued to take a leading role as an active research worker, and his own experiments set a high standard as models of analytic precision. Among his best-known studies are those elucidating the influence of the communicator’s prestige and the ways that prestige effects disappear with the passage of time. Following a lead obtained from the wartime research reported in Experiments on Mass Communication, Hovland and his collabo rators showed that when a persuasive message is presented by an untrustworthy source it tends to be discounted by the audience, so that immediately after exposure there is little or no attitude change; but then, after several weeks, the source is no longer associated with the issue in the minds of the audience and positive attitude changes appear (Hovland & Weiss 1951). This delayed or “sleeper” effect was shown to vanish, as predicted, when after several weeks the unacceptable communicator was “reinstated” by reminding the audience about who had presented the earlier persuasive material (Kelman & Hovland 1953).
For more than fifteen years Hovland system atically investigated factors that determine the effectiveness of persuasive communications, in cluding studies of different sequential arrangements of arguments, the retention of arguments and conclusions, and judgmental processes that enter into attitude change. While pursuing his own research, Hovland continually encouraged his as sociates on the Yale project to select other variables in line with their own research interests, such as the influence of group affiliation, role playing, emotional appeals, and personality predispositions. The major research findings of the first five years of the communications research project, together with theoretical analyses of the problems under investigation, were summarized in a volume entitled Communication and Persuasion by Hovland, Janis, and Kelley (1953). This volume was followed during the next eight years by a series of four multiauthored monographs on more specific topics: The Order of Presentation in Persuasion (Hovland et al. 1957); Personality and Persuasibility (Janis, Hovland et al. 1959); Attitude Organization andChange (1960b); and Social Judgment (Sherif & Hovland 1961). The series of works by Hovland and his co-workers, according to Nathan Maccoby (1963), furnishes the empirical core of “the new scientific rhetoric,” the body of psychological knowledge accumulated from objective description and analysis of the processes of persuasion.
Research on thought processes . In the last decade of his life Hovland’s research on verbal concepts and judgment led him into an intensive analysis of symbolic processes. Once again he played a pioneering role in developing a new field of research—computer simulation of human thought processes. His first major contribution in this field was a “communication analysis” of concept learning (1952) which showed how a newly developed mathematical theory could be applied to computer simulation of the ways in which people form new concepts. His general method of analyzing concept learning and his notational system were soon adopted by many other research workers who were conducting experiments on human learning and cognitive processes.
Several years after Hovland’s pioneering paper, there were some breakthroughs in the programming of digital computers, which Hovland immediately applied in constructing a computer simulation model of the steps a person goes through as he thinks out the solution of problems requiring the attainment of a new concept. Supported by generous research grants from the Ford Foundation and the Bell Telephone Laboratories, Hovland and his collaborators began devising a series of experiments in order to obtain some of the missing in formation needed for an adequate theory to account for human acquisition of complex concepts through experience. One of the main findings, reported in a paper by Hunt and Hovland (1960), was that most human learners readily make use of information about conjunctive concepts (for example, all members of the given class share two characteristics, A and B) but tend to ignore information pointing to disjunctive concepts (for example, all members of the given class possess either characteristic A or B). Accordingly, Hovland developed a computer model of concept formation in which a hierarchy of responses was programmed in such a way that the conjunctive concepts would be the first type tried out and disjunctive concepts would be scanned only after other approaches consistently failed. In a highly influential paper entitled “Computer Simu lation of Thinking” (1960a), Hovland pointed out the potential advantages of making use of new developments in mathematics and computer technology for advancing the human sciences. Many research workers are now implementing the mixed research strategy he recommended, combining experimental studies of human thinking with the development of computer programs that simulate human psychological processes.
Other contributions. Hovland’s influence on the methodology of social science research was consistently directed toward integrating seemingly divergent lines of research. One of his best-known papers deals with the problems of reconciling conflicting results derived from experimental and survey studies of attitude change (1959). He pointed out that one gets the impression from survey research that very few people are affected by mass communications, whereas experiments on opinion change show that from one-third to one-half of the audience is influenced by a single exposure to a persuasive message. This apparent divergence can be accounted for by a number of well-known factors that are often overlooked, such as the use of captive audiences and remote or unfamiliar issues in experimental studies, in contrast to the audience’s self-selective exposure and high ego-involvement in the issues typically studied by survey research. Hovland’s recommendation was that the two research approaches should be used conjointly, “combining their virtues so that we may develop a social psychology of communication with the conceptual breadth provided by correlational study of process and with the rigorous but more delimited methodology of the experiment” (1959, p. 17).
At a memorial session of the New England Psychological Association, held a year after Hovland’s premature death, his former students and associates recalled his “uncanny ability to integrate and focus knowledge” and “to discern the central aspects of a problem” while at the same time fulfilling his leadership role in a “gentle and supportive” way. As Herbert Kelman put it, he was “the world’s most non-authoritarian leader.” Indeed, Hovland welcomed diverse theoretical viewpoints and encouraged those working with him to try out new research strategies. His incisive comments stimulated his co-workers and students to make their studies as rigorous as possible and to pursue fully the substantive inferences that could be drawn from the data. This type of direction, combined with the atmosphere of freedom of inquiry which he consistently fostered, nurtured the talents of the many younger psychologists whose names appear as coauthors of his books and articles, most of whom have subsequently become leading figures in American social psychology.
Not the least of Hovland’s contributions was the public service he rendered in the role of a “states man of the social sciences.” As one of the few psychologists of his generation elected to the National Academy of Sciences, Hovland was invited to be a committee member or consultant to the Rockefeller Foundation, the Ford Foundation, the Russell Sage Foundation, the Bell Telephone Laboratories, the Social Science Research Council, the National Research Council, the Research Development Board, the Fund for Adult Education, and a number of other private national research organizations, as well as to some of the social research agencies of the U.S. government. Hovland fulfilled his consultant role by consistently working to improve the standards and quality of research in psychology and related fields.
Perhaps the most comprehensive statement of the scope of Hovland’s substantive contributions to social science research is contained in the citation of the Warren medal, awarded by the Society of Experimental Psychologists in the last year of his life: “For his systematic analyses …[in] four areas of research—verbal learning, conditioning, concept formation and attitude change.”
Irving L. Janis
[Directly related are the entriesAttitudes, article onAttitude Change; Persuasion; Social Psychology; Suggestion. Other relevant material may be found inCommunication, Mass; and in the hioq-raphies ofHull; Sapir; Stouffer; Yerkes]
WORKS BY HOVLAND
1937a The Generalization of Conditioned Responses: 1. The Sensory Generalization of Conditioned Responses With Varying Frequencies of Tone. Journal of General Psychology 17:125–148.
1937b The Generalization of Conditioned Responses: 2. The Sensory Generalization of Conditioned Responses With Varying Intensities of Tone. Journal of Genetic Psychology51:279–291.
1937c The Generalization of Conditioned Responses: 3. Extinction, Spontaneous Recovery, and Disinhibition of Conditioned and of Generalized Responses. Journal of Experimental Psychology21:47–62.
1937c The Generalization of Conditioned Responses: 4. The Effects of Varying Amounts of Reinforcement Upon the Degree of Generalization of Conditioned Responses. Journal of Experimental Psychology21:261–276.
1938 Hovland, Carl I.; and Sears, Robert R. Experiments on Motor Conflict: 1. Types of Conflict and Their Modes of Resolution. Journal of Experimental Psychology23:477–493.
1940 Hovland, Carl I.; and Sears, Robert R. Minor Studies of Aggression: 6. Correlation of Lynchings With Economic Indices. Journal of Psychology 9:301–310.
1940 Mathematico-Deductive Theory of Rote Learning. New Haven: Yale Univ. Press. → By Carl I. Hovland, C. L. Hull, R. T. Ross, M. Hall, D. T. Perkins, and F. B. Fitch.
1940 Sears, Robert R.; Hovland, Carl I.; and Miller, Neal E. Minor Studies of Aggression: 1. Measurement of Aggressive Behavior. Journal of Psychology 9:275–295.
1949 Hovland, Cabl I.; Lumsdaine, Arthur A.; and Sheffield, Frederick D. Experiments on Mass Communication. Studies in Social Psychology in World War II, Vol. 3. Princeton Univ. Press; Oxford Univ. Press.
1951 Human Learning and Retention. Pages 613-689 in S. S. Stevens (editor), Handbook of Experimental Psychology. New York: Wiley.
(1951) 1954 Hovland, Carl I; and Weiss, Walter The Influence of Source Credibility on Communication Effectiveness. Pages 337-347 in Society for the Psychological Study of Social Issues, Public Opinion and Propaganda. New York: Dryden. → First published in Volume 15 of Public Opinion Quarterly.
1952 A “Communication Analysis” of Concept Learning. Psychological Review 59:461–472.
1952 Hovland, Carl I.; and Mandell, Wallace An Experimental Comparison of Conclusion-drawing by the Communicator and by the Audience. Journal of Abnormal and Social Psychology 47:581–588.
1953 Hovland, Carl I.; Janis, Irving L.; and Kelley, Harold H. Communication and Persuasion: Psycho logical Studies of Opinion Change. New Haven: Yale Univ. Press.
1953 Hovland, Carl I.; and Weiss, Walter Transmis sion of Information Concerning Concepts Through Positive and Negative Instances. Journal of Experimental Psychology 45:175–182.
1953 Kelman, Herbert C; and Hovland, Carl I. “Re instatement” of the Communicator in Delayed Measurement of Opinion Change. Journal of Abnormal and Social Psychology 48:327–335.
1954 Effects of the Mass Media of Communication. Volume 2, pages 1062-1103 in Gardner Lindzey (editor), Handbook of Social Psychology. Cambridge, Mass.: Addison- Wesley.
1955 Kelley, H. H.; Hovland, Carl I.; Schwartz, M.; and Abelson, R. P. The Influence of Judges’ Attitudes in Three Methods of Attitude Scaling. Journal of Social Psychology 42:147–158.
1956 Kurtz, Kenneth H.; and Hovland, Carl I. Concept Learning With Differing Sequences of Instances. Jour nal of Experimental Psychology 51:239–243.
1957 Hovland, Cabl I. et al. The Order of Presentation in Persuasion. New Haven: Yale Univ. Press. → Co authors are W. Mandell, E. H. Campbell, T. Brock, A. S. Luchins, A. R. Cohen, W. J. McGuire, I. L. Janis, R. L. Feierabend, and N. H. Anderson.
1959 Reconciling Conflicting Results Derived From Experimental and Survey Studies of Attitude Change. American Psychologist 14:8–17.
1959 Janis, Ibving L.; Hovland, Carl I. et al. Personal ity and Persuasibility. New Haven: Yale Univ. Press. → Coauthors are P. B. Field, H. Linton, E. Graham, A. R. Cohen, D. Rife, R. P. Abelson, G. S. Lesser, and B. T. King.
1959 Mobbisett, Lloyd N.; and Hovland, Cabl I. A Comparison of Three Varieties of Training in Human Problem Solving. Journal of Experimental Psychology 58:52–55.
1960a Computer Simulation of Thinking. American Psychologist 15:687–693.
1960 Hovland, Cabl I.; and Hunt, Eabl B. Computer Simulation of Concept Attainment. Behavioral Science 5:265–267.
1960b Attitude Organization and Change. Yale Studies in Attitude and Communication, Vol. 3. New Haven:Yale Univ. Press. → By Carl I. Hovland, M. J. Rosen berg, W. J. McGuire, J. W. Brehm, and R. P. Abelson.
1960 Hunt, Earl B.; and Hovland, Carl I. Order of Consideration of Different Types of Concepts. Journal of Experimental Psychology 59:220–225.
1961 Hunt, Earl B.; and Hovland, Carl I. Programming a Model of Human Concept Formation. Pages 145-155 in Western Joint Computer Conference, Los Angeles, 1961, Proceedings. Los Angeles: The Conference.
1961 Sherif, Muzafer; and Hovland, Carl I. Social Judgment: Assimilation and Contrast Effects in Communication and Attitude Change. Yale Studies in Atti tude and Communication, Vol. 4. New Haven: Yale Univ. Press.
American Psychological Association 1958 Distin guished Scientific Contribution Awards, 1957. American Psychologist 13:155–168.
Cohen, Arthur R. 1964 Attitude Change and Social Influence. New York: Basic Books.
Dollard, John et al. 1939 Frustration and Aggression. New Haven: Yale Univ. Press. → A paperback edition was published in 1961. Coauthors are L. W. Doob, N. E. Miller, O. H. Mowrer, and R. R. Sears.
Janis, Irving L.; and Smith, M. Brewster 1965 Effects of Education and Persuasion on National and International Images. Pages 188-235 in H. C. Kelman (editor), International Behavior. New York: Holt.
Maccoby, Nathan 1963 The New “Scientific” Rhetoric. Pages 41-53 in Wilbur Schramm (editor), The Science of Human Communication. New York: Basic Books.
Schramm, Wilbur 1963 Communication Research in the United States. Pages 1-16 in Wilbur Schramm (editor), The Science of Human Communication. New York: Basic Books.
Carl I. Hovland
Carl I. Hovland
The American psychologist Carl I. Hovland (1912-1961) was one of the pioneers in research on the effects of social communication on attitudes, beliefs, and concepts.
Carl I. Hovland was born in Chicago, III. He attended Northwestern University and completed his graduate studies at Yale University, receiving his doctorate in 1936. He then joined the faculty at Yale, where he remained throughout his entire career.
During the late 1930s and early 1940s Hovland made major contributions to several areas of human experimental psychology, such as the efficiency of different methods of rote learning. From his close association with Clark L. Hull and other psychologists working at the Yale Institute of Human Relations, Hovland developed a comprehensive view of the behavioral sciences that led him to extend the analytic experimental approach of research on human learning to underdeveloped areas of research in the human sciences.
Hovland's first opportunity to work intensively in the underdeveloped area of social psychology arose during World War II, when he took a leave of absence from Yale for over 3 years to serve as a senior psychologist in the War Department. His main role was to conduct experiments on the effectiveness of training and information programs that were intended to influence the motivation of men in the American armed forces. He assembled a group of six psychology graduate students who worked with him on these studies for several years. One of the most widely cited of the pioneering experiments on opinion change by Hovland and his group involved testing the effects of a one-sided versus a two-sided presentation of a controversial issue. The results contradicted contentions of totalitarian propagandists, who claimed that a communication that presents only one side of the issue will generally be more successful than one that mentions the opposing side of the argument. These wartime studies were reported in Experiments on Mass Communication (1949), written jointly by Hovland, A. A. Lumsdaine, and F. D. Sheffield.
After the war Hovland returned to Yale University, where he recruited several members of his wartime research team, with whom he continued to study the factors that influence the effectiveness of social communications. Among Hovland's best-known studies are those elucidating the influence of the communicator's prestige and the ways that prestige effects disappear with the passage of time. For example, Hovland and his collaborators showed that when a persuasive message is presented by an untrustworthy source, it tends to be discounted by the audience, so that immediately after exposure there is little or no attitude change; but then, after several weeks, the source is no longer associated with the issue in the minds of the audience and positive attitude changes appear. This delayed, or "sleeper, " effect was shown to vanish, as predicted, if the unacceptable communicator was "reinstated" several weeks later by reminding the audience about who had presented the earlier persuasive material.
For 15 years Hovland and his group systematically investigated different ways of presenting arguments, personality factors, and judgmental processes that enter into attitude change. While pursuing his own research, Hovland continually encouraged his associates on the Yale project to select other problems in line with their own research interests. The work of Hovland's program was described in Communication and Persuasion (1953) by Hovland, Irving L. Janis, and Harold Kelly.
In the last decade of his life Hovland's research on verbal concepts and judgment led him into an intensive analysis of concept formation. Once again he played a pioneering role in developing a new field of research— computer simulation of human thought processes.
A summary of the research developments and theoretical ideas that have grown out of Hovland's pioneering projects is presented in a comprehensive chapter by Irving L. Janis and M. Brewster Smith in Herbert C. Kelman, ed., International Behavior (1965). Hovland's work is also discussed in Arthur R. Cohen, Attitude Change and Social Influence (1964). □