Social Influence and the Power of the Pyramid
Markovsky, Barry, Thye, Shane R., Skeptic (Altadena, CA)
MORE THAN 90% OF ADULT AMERICANS report holding at least one paranormal belief, and the rates for some specific beliefs have risen steeply in recent years. (1-5) Skeptics often bemoan such numbers, knowing as they do that the overwhelming majority of paranormal claims either are perfectly consistent with "normal" explanations unknown to the claimant, or else they unravel completely in those rare cases that are investigated systematically. Skeptics may be over estimating the effect logic and evidence have on popular notions, and underestimating the effect that people have on each other's beliefs.
Our interests encompass both social influence in general, and paranormal beliefs specifically. In this study we combine those interests to fill a gap in the research on how people influence one another's beliefs in the paranormal. While the study that we conducted focuses on "pyramid power," this particular pseudo-phenomenon is incidental to the more general issues involved.
Studies of the nature and sources of paranormal beliefs have taken two general directions. The micro approach leans toward the psychological, emphasizing sensory, perceptual, cognitive, and experiential processes. (6-12) In contrast, the macro approach identifies cultural, demographic, religious, and other variables that, in the aggregate, are associated with paranormal beliefs, typically as expressed in surveys. (13-17) Surprisingly, there is practically no research on paranormal beliefs that fits between the micro and macro. This meso level would include the interpersonal spread of paranormal beliefs, a potentially fertile ground when one considers the kinds of settings it may include: families, friendship circles, classrooms, work groups, voluntary associations, and so on.
We focused our investigation on four key questions:
(1) Are paranormal beliefs acquired from another person merely through exposure the other's belief? If so, this would suggest that paranormal beliefs spread through much larger social groups, whether or not anyone actively promotes those beliefs.
(2) Are paranormal beliefs acquired from other people as readily as non-paranormal beliefs? We suspected that people may be reluctant to accept paranormal claims, compared to nonparanormal ones, thus hindering their spread through larger groups.
(3) Are claims more influential if they come from higher status people? Higher status people generally are more persuasive, even in situations where their status is irrelevant. Will the same be true for paranormal claims, even without any overt attempt to persuade?
(4) Do people influence others beyond the effect of the information they communicate? In other words, would a paranormal claim that emanates from a flesh-and-blood person have more weight than, say, one heard second-hand from an anonymous source?
A crucial point in the dissemination of paranormal beliefs is the moment when individuals are confronted with an alleged paranormal claim and are motivated to judge its validity. We contend that the outcome of this judgment depends not only on individual qualities such as critical thinking ability, but also on social factors.
Social influence refers to changes in feelings, beliefs, values or behaviors that result from the actions of others. Examples of influenced behaviors may include: following orders at work (obedience to authority), dressing like one's peers (conformity to a group), and convincing a friend to change her vote in an upcoming election (persuasion). You also may be influenced to develop a positive attitude toward the Boston Red Sox, to form the belief--beyond all rationality--that they will win the World Series, and to feel the letdown of yet another disappointing season when they fail. Almost any person or group, real or imagined, is a potential source of social influence: parents, sports teams, dictators, corporations, cartoon characters, nations, cult leaders, to list but a few.
Social impact theory (18-22) offers a general, unified approach to understanding and investigating these and other forms of social influence. According to the theory, social impact is affected by four factors: number, strength, immediacy, and impediments (see Figure 1). The number factor refers to the number of people constituting the source of impact, such as whether peer pressure on an adolescent is being exerted by a single friend or by an entire group of friends. Strength indicates the source's power, status or importance. That is, we are more likely to be hurt by insensitive comments from someone we care about a great deal than by the same comments spoken by a mere acquaintance. Immediacy refers to the source's closeness in space or time. For example, we should be more likely to obey a boss's request when he or she delivers it personally than when it arrives in the form of an e-mail. Finally, social impact also is assumed to be negatively affected by any sort of barrier or impediment to the effects of th e three positive factors. (23) This could occur when there is a barrier to effective communication, or a preconceived notion that the claim being made is less than fully legitimate. In the experiment reported below, we tested for social impact on paranormal beliefs by manipulating all four of the above factors.
We decided to capitalize on a paranormal claim that is easy to work with in an experimental setting, marginally familiar and plausible to college students, and unlikely to be the target of strong beliefs pro or con. "Pyramid power" fit the bill. Pyramids have long been the subject of paranormal claims based on ostensive powers to rejuvenate objects stored within them. (24-29) The typical claim is that the pyramid focuses a mysterious "energy" causing the preservation of organic materials, or the sharpening of razor blades, or the curing of illness.
Speculation and anecdote far outweigh evidence in the "pyramidology" literature. So before investigating social impact, we thought it worthwhile to see whether research subjects would be able to detect any pyramid power effects, if they in fact existed at all. Using a double-blind experiment, we had subjects judge the freshness of three bananas, one of which had been stored in a pyramid, another in a box, and a third with no container. The bananas were presented to each subject in a random sequence for rating, and neither the subjects nor the experimenter knew the correct sequence of the storage condidons. The results showed that there was no detectable difference between the pyramid-banana and the box-banana, but that they both looked fresher than the banana stored out in the open. This accounts for the results of one type of poorly-designed pyramid experiment: Suppose that one only compared fruit stored in a pyramid to fruit stored out in the open. The fruit in the pyramid would look fresher, and so one cou ld mistakenly attribute this to pyramid power. We showed that, in order to conclude that the pyramid does indeed possess a special power, we would also have to conclude that the box has it, tool The simper explanation, of course, is that any container that inhibits air circulation will slow the ripening process.
Our main experiment, reported below, did not test for pyramid power per se. Rather we wanted to determine whether social influences could heighten subjects' belief in the pyramid's effect when, in fact, experimental controls assured us that no such effect exists.
Subjects were undergraduate students at the University of Iowa who signed up in classes to participate in research projects. The pyramid--a bit larger than a shoebox--was built to within a 1% tolerance of published specifications. (30) The pyramid was placed at the horizontal end of a T-shaped table along with a box of the same material and internal volume (Figure 2). (31) The experimenter sat at the bottom of the T, with the subject on one side of the vertical section. When present, a "confederate" posed as another subject and sat opposite the real subject.
The experimenter read instructions that included a brief review of "evidence" for pyramid power, such as the well-preserved bodies of Egyptian pharaohs entombed in pyramids. Subjects also were told that they would be asked to rate the freshness of bananas which, one week before, had been placed under the pyramid and the box. (32)
The pairs of bananas used in all conditions were virtually identical, selected on the basis of similarity ratings provided by panels of judges. Prior to the start of every session, the experimenter swapped the bananas between the pyramid and box. This ensured that results could not be biased by any natural differences that might have emerged over the several days each set of bananas was used.
Subjects were assigned randomly to conditions of the experiment. Most of the conditions involved a confederate posing as another subject. Confederates made no attempts to influence or interact with subjects, and no incentives were provided for the subject either to agree or to disagree with the confederate's ratings. With the exception of one condition described below, confederates closely matched subjects in age, race, sex, and education.
Subjects provided oral ratings for both bananas on a series of five 15-point rating scales. The ordering of the rating scales and of the bananas (box first v. pyramid first) was shuffled across sessions. The experimenter always queried the confederate before the subject for three critical rating scales: how fresh, hard, and preserved the bananas appeared. The confederate's responses were predetermined so that he/she always rated the pyramid-banana as considerably fresher than the box-banana by four or five rating points. Researchers long have known that public and private ratings do not necessarily coincide. (33,34) Therefore, we also asked subjects to express their beliefs regarding the pyramid's effect without their having to feel accountable to a confederate or experimenter: After the session with the experimenter, they completed a slightly different set of ratings alone.
We devised an experiment that allowed us to test all four of social impact theory's basic factors in a "pyramid power" context. There were five conditions in all:
Condition 1: Baseline-Confederate. The confederate's ratings indicated that he/she believed the pyramid-banana looked considerably fresher than the one stored in the box. Responses in this condition were contrasted with those from the other four conditions.
Condition 2 No Confederate. The subject judged the bananas without hearing anyone else's ratings.
Condition 3: Non-paranormal Context. A plastic box was substituted for the pyramid, and paranormal connotations were removed from the instructions.
Condition 4: High-status Confederate. Here the confederate was a male professor, 39 years old. While in the presence of the subject, his responses to several questions from the experimenter made it obvious that his occupational status, age, and education exceeded the subject's. To allay suspicions, he added that he was curious about the research and signed up to participate when a project assistant recruited students from a course he taught.
Condition 5: Distant Confederate. Instead of the confederate sitting across the table from the subject, the experimenter read what he described as the ratings of a randomly selected subject who rated the same set of bananas in an earlier session.
In applying the social impact theory to our experimental setting, we can derive four hypotheses to answer the four questions we posed earlier:
Hypothesis 1 (Number Effect): The first question asked whether a person can acquire a paranormal belief from another merely through exposure to that other's belief. Because this is contrasted with a no-confederate condition, it involves the number factor from social impact theory. (36) Subjects in the presence of a credulous confederate were predicted to believe that the pyramid kept the banana fresher than the box, despite the fact that the two pieces of fruit were virtually identical. We tested the hypothesis by seeing whether the pyramid-bananas were rated fresher than the box-bananas more often in the Baseline-Confederate condition (#1) than in the No-confederate condition (#2).
Hypothesis 2 (Impediment Effect): If the paranormal connotations of pyramid power are an impediment from the standpoint of social impact theory, then we may hypothesize that social impact is greater in the Non-paranormal Context (#3) than in the Baseline-Confederate condition (#1).
Hypothesis 3 (Strength Effect): According to the theory, a higher-status confederate should have greater impact than an equal-status confederate due to the strength factor. Therefore, we predicted that social impact would be greater in the High-status Confederate condition (#4) than in the equal status Baseline-Confederate condition (#1).
Hypothesis 4 (Immediacy Effect): The theory predicted that the Distant Confederate condition (#5) would have less impact than the Baseline-Confederate condition (#1).
A total of 199 subjects participated in the experiment. (37) All conditions had near equal numbers of males and females, and gender had no effects in our data analyses. Subjects' responses were coded +1 if they rated the pyramid-banana fresher than the box-banana, 0 if they rated both the same, and -1 if they rated the box-banana fresher. (35) With each subject providing three critical ratings, the range of possible scores was +3 to -3. For the written ratings completed in private after the oral ratings, the possible range was +7 to -7. The average oral and private ratings in each condition are shown in Figure 3. (38)
Hypotheses 1 and 2--the number and impediment predictions--were supported unequivocally in both oral and private ratings. Relative to the No Confederate condition, the Baseline-Confederate condition showed a clear effect on the subjects' ratings: Subjects judged the pyramid-banana to be fresher significantly more often when the confederate expressed this judgment, and they continued to do so in their private ratings. Additionally, although social impact did occur in all of the confederate conditions, the paranormal context inhibited this effect relative to the non-paranormal Context.
Hypothesis 3, the strength prediction, received mixed support. The effect of the Highstatus Confederate was significantly greater than that of the Baseline-Confederate in the oral ratings, but not so in the private ratings.
Hypothesis 4 was the immediacy prediction, and it did not receive support either in the oral ratings or in the private ratings. This was because the impact of the Distant Confederate on both oral and private ratings was unexpectedly strong relative to that of the Baseline-Confederate.
Our findings show that people may acquire a paranormal belief merely by observing another who appears to hold that belief. Bear in mind the mildness of the basic experimental manipulation: a lone confederate possessing no special status or authority over the subjects, making absolutely no attempt to persuade, coerce or otherwise influence the subject. For social influence to occur at all was not a foregone conclusion, especially in view of the predicted suppression of influence due to the impediment effect of the paranormal context.
Why the high status confederate did not have significantly more impact than the equal status confederate on private ratings is a bit of a mystery. Given that the effect was significant for the oral ratings, it would appear that the status difference led subjects to conform. outwardly to the professor/confederate, but not to internalize the paranormal belief beyond a level induced by an equal-status confederate. The status manipulation may not have been quite strong enough, the measure may have been too "noisy" to detect the effect, or perhaps status is not a valid strength factor. (39) Given that the results were in the predicted direction--albeit not statistically significant--the first two possibilities appear more likely than the third.
The immediacy hypothesis received the weakest support in that the Distant Confederate had more impact than expected in both oral and private ratings--a level only a little lower than when the confederate sat just across the table. Social impact theory is quite explicit in its emphasis on physical proximity as an exemplar of the immediacy factor. If our finding holds up in replications, it would have to be taken as a falsification of the theory. More important, however, is the implication for the spread of paranormal beliefs: The results suggest that people may be as likely to acquire paranormal beliefs second-hand from anonymous others as they do from someone present in the situation. Perhaps this contagiousness helps to overcome the impediment of paranormality, and to account for the broad diffusion of paranormal beliefs in our culture.
Public opinion surveys indicate that the overwhelming majority of Americans believe some paranormal claims. At the same time, most specific beliefs are minority viewpoints. It appears that most people internalize only a small proportion of the paranormal claims to which they are exposed. This could mean that beliefs are inhibited by the marginal status of most claims, but still we internalize if they appear to have social support. Social support thus acts as a proxy for valid evidence.
As social researchers, we find such results to be intriguing because they suggest that existing theories may be applicable to paranormal belief formation and maintenance. As educators, however we emphasize that social influence is a two-way street. The same theories that suggest conditions for the spread of paranormal beliefs also suggest methods for spreading rational approaches to evaluating paranormal claims.
[FIGURE 3 OMITTED]
(1.) The term "paranormal" has a variety of meanings. We use it to refer to claims that violate physical laws or precedents, that is, "scientific anomalies." See Truzzi, M. 1971. Definition and Dimensions of the Occult: Towards a Sociological Perspective." Journal of Popular Culture 3: 635-46.
(2.) Gallup, G.H. 1979. The Gallup Poll: Public Opinion, 1978. New York: Random House.
(3.) Gallup, G.H. 1997. The Gallup Poll: Public Opinion, 1996. Wilmington, DE: Scholarly Resources, Inc.
(4.) Gallup, G.H., and F. Newport. 1990. "Belief in Psychic and Paranormal Phenomena Widespread Among Americans." The Gallup Poll Monthly 299 (August):35-44.
(5.) McAneny, L. 1995. "It Was a Very Bad Year: Belief in Hell and the Devil On the Rise." Gallup Poll Monthly 304 (January):14-47.
(6.) Alcock, J. 1980. Parapsychology: Science or Magic? A Psychological Perspective. Toronto: Pergamon.
(7.) Blackmore, S. 1991. "Near-Death Experiences: In or Out of the Body? Skeptical inquirer 16:3445.
(8.) Blackmore, S. 1993. Dying to Live: Near-Death Experiences. Buffalo, NY: Prometheus.
(9.) Gilovich, T. 1991. How We Know What Isn't So: The Fallibility of Human Reason in Everyday Life. New York: Free Press.
(10.) Nisbett, R.E., and L Ross. 1980. Human Inference: Strategies and Shortcomings of Social Judgment. Englewood Cliffs, NJ: Prentice Hall.
(11.) Reed, G.F. 1988. The Psychology of Anomalous Experience: A Cognitive Approach. Buffalo, N.Y.: Prometheus.
(12.) Zusne, L., and W. Jones, 1982. Anomalistic Psychology: A Study of Extraordinary Phenomena of Behavior and Experience. Hillsdale, NJ: Lawrence Erlbaum Associates.
(13.) Gallup and Newport 1990, op cit.
(14.) Greeley, A. 1975. Sociology of the Paranormal: A Reconnaissance. London: Sage.
(15.) Harrold, F.B., and R.A. Eve (eds.). 1987. Cult Archaeology and Creationism. Iowa City, IA: University of Iowa Press.
(16.) MacDonald, W.L. 1994. "The Popularity of Paranormal Experiences in the United States." Journal of American Culture 17(3):35-42.
(17.) McClenon, J. 1994. Wondrous Events. Philadelphia: University of Pennsylvania Press.
(18.) Latane, B. 1981. "The Psychology of Social Impact." American Psychologist 36:343-356.
(19.) Latane, B, J.H. Liu, A. Nowak and M. Bonevento. 1995. "Distance Matters: Physical Space and Social Impact." Personality and Social Psychology Bulletin 21(8): 795-805.
(20.) Latane, B. and T. L'Herrou. 1996. "Spatial Clustering in the Conformity Game: Dynamic Social Impact in Electronic Groups." Journal of Personality and Social Psychology 70(6):1218-30.
(21.) Latane, B. and S. Wolf. 1981. "The Social Impact of Majorities and Minorities." Psychological Review 88:438-53.
(22.) Nowak, A., J. Szamrej and B. Latane. 1990. "From Private Attitude to Public Opinion: A Dynamic Theory of Social Impact." Psychological Review 97:362-376.
(23.) Social impact theory also argues that the strength, immediacy and number of the targets of influence also have negative effects on the amount of impact Our research does not address this facet of the theory.
(24.) Cazeau, C. 1986. Exploring the Unknown: Great Mysteries Reexamined. New York: Plenum.
(25.) Daniels, P. and A. Horan (eds.). 1987. Mystic Places. Alexandria, VA: Time-Life.
(26.) Gardner, M. 1957. Fads and Fallacies in the Name of Science. New York: Dover.
(27.) Ostrander, S. and L Schroeder. 1970. Psychic Discoveries Behind the Iron Curtain. Englewood Cliffs, NJ: Prentice-Hall.
(28.) Stine, G.H. 1992; Mind Machines You Can Build. Largo, FL: Top of the Mountain Publishing.
(29.) Toth, M. 1974. Pyramid Power. New York: Destiny Books.
(30.) Watson, L 1973. Supernature. Garden City, NY: Anchor Press.
(31.) Promoters of pyramidology agree that only the pyramid's shape matters, not the material from which it is constructed. We used mat board (1/8" solid cardboard), and acknowledge Rose Garfinkle's precision craftsmanship in constructing our pyramid and box.
(32.) Prior to the start of a session, subjects also completed a "Background Information Form" and a general "Paranormal Beliefs Questionnaire." These were used as statistical controls in our analyses. in effect, implementing these controls allowed the true effects of the social impact manipulations to emerge more clearly. For details see Markovsky, B. and S. Thye. "Social influences on Paranormal Beliefs." Sociological Perspectives 2001, 44(1):21-44.
(33.) Sherif, Muzafer. 1935. "A Study of Some Social Factors in Perception." Archives of Psychology 27(187).
(34.) Asch, Solomon. 1956. "Studies of independence and Submission to Group Pressure: I. On Minority of One Against a Unanimous Majority." Psychological Monographs 70(9).
(35.) Results were the same when we tested the actual magnitudes of the differences.
(36.) Strictly speaking, social impact theory predicts two effects from increasing the number of confederates: (1) Overall levels of social impact will increase with each additional confederate, but (2) increasing the number of confederates has a diminishing marginal effect. Because we seek to establish the minimal necessary conditions for social impact in a non-persuasive context, our experiment tests only for the impact of increasing the number of confederates from zero to one.
(37.) Fourteen subjects were excluded after post-experiment interviews revealed strong suspicions regarding experimental manipulations. This 7% suspicion rate is quite typical for experiments involving mild deception of subjects.
(38.) Means for the oral ratings in the order presented in the graph are .87, 1.37, 1.56, 2.08, 2.13. For the private ratings the means were .30, 1.17, 1.78, 3.18, 2.63. When we state that a hypothesis is supported, it means that the observed effect was statistically significant.
(39.) Unfortunately, social impact theory does not provide sufficiently precise definitions for its key factors so that one could determine whether or not particular interpretations--such as the status differences we attempted to create--actually qualify as appropriate tests.
Dr. Barry Markovsky is Professor and Chair, Department of Sociology, University of South Carolina. In addition to his interest in how social factors influence paranormal beliefs, he has published research on social power, status processes, perceptions of justice, social influence, group solidarity, social networks, and methods for theory construction. Current projects include the applicability of complexity theory to issues in the emergence and evolution of social structures.
Dr. Shane R. Thye is Assistant Professor of Sociology at the University of South Carolina, and coeditor of the annual series Advances in Group Processes. His current research is aimed at understanding how status. processes lead to the development of power in small groups. He also is conducting research funded by the National Science Foundation on the emergence of commitment and micro social order in networks.…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Social Influence and the Power of the Pyramid. Contributors: Markovsky, Barry - Author, Thye, Shane R. - Author. Magazine title: Skeptic (Altadena, CA). Volume: 9. Issue: 3 Publication date: Fall 2002. Page number: 36+. © 2009 Skeptics Society & Skeptic Magazine. COPYRIGHT 2002 Gale Group.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.