Keith E Rice's Integrated SocioPsychology Blog & Pages

Aligning, integrating and applying the behavioural sciences

Conformity & Obedience #2

More research into conformity
 Timothy Williams & Shunya Sogon (1984) looked at Japanese students belonging to a sports club and found that normative influence was much greater when participants cared about the opinions of other group members. The higher level of conformity found by Williams & Sogon may reflect the collectivistic nature of Japanese society and, therefore, may not be generalisable beyond similar societies. However, it shows clearly the influence of fitting in with those you with whom you have a belonging connection – that connection fulfilling the needs of the PURPLE vMEME.

Supporting this, Paul McGhee &, Richard Teevan (1967) found that students high in the need for affiliation were more likely to conform.

Dominic Abrams et al (1990) found an in-group influence, proposing that their 1st-year Psychology students would show more conformity if the other group members were perceived to be in the same in-group (Psychology students from a nearby university) than if they were from an out-group (Ancient History students from the same university). Accordingly, there was conformity on 58% of trials when in the presence of an in-group but only 8% with an out-group.

Morton Bogdonoff et al (1962) found arousal levels were high in all participants once they were faced with the opposing judgements of the confederates – an indication of their stress. These high levels of arousal were maintained in participants who held firm to their own judgements but dropped in those who conformed.

Deutsch & Gerrard explored the role of normative influence by extending Asch’s research. There were 3 main conditions in their study:-

  1. As with Asch, the 3 participants were together – face-to-face – and the genuine participant had to announce their judgement publicly
  2. The genuine participant gave their judgements anonymously in an isolated cubicle by pressing a button, believing there were 3 other participants
  3. The 3 participants were in a isolated group, believing they were in competition with other groups and the experimenter promising a reward to the groups making the most accurate decisions

As predicted, Deutsch & Gerrard found conformity greatest in the third condition – though it is impossible to separate out normative from informational influence in that condition. They found conformity at its least in the second. In another variation, designed to minimise normative influence, Deutsch & Gerrard had participants write their judgements on a piece of paper and then throw it away. Conformity was only 5%.

Vernon Allen & John Levine (1971) found that the ‘quality’ of a ‘supporter’ could affect levels of conformity. They they had 3 conditions:-

  1. Participants had no support which produced a remarkable 97% conformity
  2. Participants had a ‘valid’ supporter with normal vision which reduced conformity to 36%
  3. Participants had an ‘invalid’ supporter with very poor vision who wore very thick glasses, leading to 64% conformity

The results indicate the participants in condition 2 did not trust the validity of the supporter’s judgement so they took little informational influence from the supporter’s judgement.

One of the more interesting variations on Asch was carried out by Richard Crutchfield (1955). He had groups of 5 participants sitting side by side in individual booths with a panel of lights and switches in front of them. One set of lights, when illuminated, indicated the supposed responses of the other 4 participants. One set of switches provided the participants with the means of giving their own responses. They were presented with slides containing multiple choice questions – described by Crutchfield as: “The slides call for various kinds of judgements – lengths of lines, areas of figures, logical completion of number series, vocabulary items, estimates of the opinions of others, expression of his own attitudes on issues, expression of his own preferences for line drawings….” The participants always had to give their response last, having already seen the supposed responses of the other 4. In fact, the lights were manipulated by the experimenter to create a bogus majority on critical (but not all) items. Over 600 participants were tested, including students, women and army officers attending a 3-day assessment programme. The results included:-

  • 46% conforming to the incorrect majority response when asked to compare the size of a circle and a star – the circle being much larger
  • 37% of the army officers agree with a bogus majority statement that they would not make a good leader – having all said privately beforehand that they believed they would make good leaders!
  • 58% of the student participants agreed with the bogus majority statement: Free speech being a privilege rather than a right, it is proper for a society to suspend free speech when it feels itself threatened”
  • 30% conformed to the bogus majority on Asch-type perceptual tasks

Although they were isolated from others when making their choices, a number of participants still complained they felt discomfort and embarrassment and had their self-confidence severely shaken. However, 17% of the participants indicated they knew what was going on and were aware of the deception. Crutchfield did not comment on how this might have affected the results. However, he did find that less intellectually-effective participants showed more conformity, perhaps being more susceptible to informational influence.

An interesting look into the biology of conformity comes from the work of Gregory Berns et al (2005). The researchers took an fMRI scan of the working brain of 32 volunteers during a conformity experiment on mental rotation tasks. They found that conformity responses showed up as activity in regions of the brain that were entirely devoted to perception rather than in areas involved in conscious decision-making. This suggests that exposure to the majority viewpoint had influenced their perceptions. However, those participants who went against the majority showed activity in both those brain areas involved in decision-making and in emotion, suggesting there is an emotional cost to going against the group consensus.

Conformity factors
The variables that can affect conformity include:-

  • Zeitgeist
    As demonstrated by the research of Stephen Perrin & Christopher Spencer (1980) and Nigel Nicholson, Steven Cole & Thomas Rocklin (1985), conformity can also be affected by the spirit of the times.
    Peter Smith & Michael Harris Bond (1993, p124) were in no doubt that “Levels of conformity in general had steadily declined since Asch’s studies in the early 1950s”.

    Graphic copyright © 2000 Psychology Press Ltd

  • Culture
    Smith & Bond’s 1993 meta-analysis was of 20 studies using the Asch ‘lines’ methodology. Across these studies they found a mean 31.2% of conformity errors. Samples of their results are in the graphic above…
    The Indian teachers in Fiji were the highest in conformity and the Belgian students the lowest, apart from Perrin & Spencer’s 1980 study in England. The meta-analysis included Williams & Sogon’s (1984) study of Japanese students.In a further meta-analysis of another 133 Asch-type studies, Rod Bond & Peter Smith (1996) found that individualistic countries, such as the United States, where emphasis was placed on individual fulfilment, were less conformist generally (25.3% of trials) than collectivistic cultures like China, Asia and much of Africa (37.1%) where group needs were usually prioritised over individual needs. Heejung Kim & Hazel Rose Markus (1999) elaborated on this by showing that a failure to conform to majority influence is perceived positively as ‘uniqueness’ in individualistic cultures but is regarded as deviant in collectivistic ones.
    Howard Bloom’s (1997) take on this would be that collectivistic cultures act as ‘conformity enforcers’ facilitating the ‘cool’ side of the Spiral while individualistic cultures are ‘diversity generators, facilitating the ‘warm’ vMEMES.
  • Personal characteristics
    Smith & Bond found a significant difference in conformity between students (28%) and non-students (37%). However, the researchers were unable to say whether the students learn to be more independent and critical in their thinking or whether their higher level of intelligence makes them more confident in their judgement.
    Jerry Burger & Harris Cooper (1979) found that an individual’s desire for self-control affected how likely they were to conform. The researchers asked participants to rate a set of cartoons in terms of their funniness in the presence of a confederate who was expressing his own opinion. Participants who previously had measured high in terms of their desire for personal control – indicating RED was strong in their vMEME stack – were less influenced by the confederate.
    Correspondingly, those with a high need for social approval and having low self-esteem were more likely to conform. The work of Kaoru Kurosawa (1993) confirms this as he found those low in self-esteem showed more majority influence than those high in it in an Asch replication with 4 confederates.
  • Gender
    From a meta-analysis Alice Eagly & Linda Carli (1981) found that some studies show females to be more conformist? Eagly (1978) attributes this to their social roles making females more concerned with interpersonal goals.
    However, Frank Sistrunk & John McDavid (1971) argue that many conformity studies are biased because they use male task judgements – creating ambiguity for the female participants. They state men will conform more on female task judgements – again because unfamiliarity causes ambiguity. The work of Christina Maslach, Richard Santee & Cheryl Wade (1987) supports this. Using an equal number of feminine-relevant and masculine-relevant stories, they found no gender difference in conformity via majority influence. Kurosawa also found no gender differences in conformity to majority influence.

This is a concept developed by Irving Janis (1972) which explains conformity in small, often tightly-knit groups who perceive no need to take in views from outside the group.

Janis’ research was initiated primarily by his teenage daughter challenging him, as a psychologist, to explain how a group of experts could have made decisions that led to the ‘Bay of Pigs’ disastrous, American-supported 1961 attempt to invade Cuba by reactionary forces opposed to the revolutionary Marxist-Leninism of Fidel Castro. (The invasion failed and the Americans became obliged to pay a ransom of $53M in food and medicine for 1,113 captured prisoners.) Janis found 4 factors in particular that tend to increase conformity and lead to bad decision-making by such tightly-knit groups:-

  1. Acceptance by the group is critical to the group. People in groups do not like to be ostracised. They want to be liked and, therefore, tend to do things to be accepted as one of the group. Clearly normative influence.
  2. When the group feels under pressure to reach a decision, they reduce this sense of stress by reaching a decision quickly and with little argument.
  3. Groups often work in isolation with the result that there are no external challenges to the way they are thinking. Because of this, inherently they tend to believe that they are right and, therefore, do not need external influences because they are either wrong or would support them anyway. A kind of False Consensus Effect.
  4. There may institutional conformity. The way to progress up a promotional ladder in institutions often includes conforming to expectations. This embodies – sometimes erroneously! – the principle that a good soldier makes a good commander.

Following the many gross and horrific atrocities in World War II – especially the Holocaust – many psychologist and sociologists were concerned with understanding the nature and processes of obedience. They were especially concerned with explaining obedience to do harm to others. From their research, there emerged 2 approaches to explaining obedience: situational explanations and dispositional explanations.

Situational explanations
From his work with obedience experiments, Stanley Milgram was very much of the view that obedience results from situational factors. In 1974 he wrote: “Obedience is as basic an element in the structure of social life as one can point to. Some system of authority is a requirement of communal living, and it is only the man living in isolation who is not forced to respond, through defiance or submission, to the commands of others.” For Milgram obedience permeates society. From an early age children are taught to obey their parents as part of their primary socialisation. Then, at school and going into work, as part of their secondary socialisation, older children and young adults are expected to obey experts and authority figures.

Milgram’s Agentic Shift Theory states that, when somebody accepts the command of another person, they surrender their own autonomy and enter an agentic state: “being another person’s agent”. This frees them from being responsible for the consequences of their actions – that responsibility rests with the person issuing the orders. That person should, of course, be a legitimate authority – ie: they have the right (legal and/or moral) right to issue orders and that right is recognised by the person receiving the orders. For example, a soldier is expected to obey orders from his commanding officer; if, by obeying those orders, something goes wrong, then the commanding officer is held responsible.

Legitimate authority is often symbolised by a uniform – eg: a police officer – or is contextual – eg: students are required to obey teachers in the school but would not expect to have to obey those same people in a totally different context such as hanging out with their friends in town on a Saturday afternoon.

According to Milgram, there are 2 ways to mitigate the moral strain people may experience if they are ordered to do something contrary to their own moral standards:-

  1. Use graduated commitment. This is starting with small requests/demands, then increase the size/severity of the demands incrementally. It’s easier to overcome resistance to a small demand than a big one. Then gradually, without people realising it, the demands become steadily but slowly more extreme.
  2. Use terms which minimise the brutal consequences of the obedience such as ‘final solution’, ‘ethnic cleansing’ or ‘collateral damage’.  As Milgram puts it: “Euphemisms come to dominate language as a means of guarding the person against the full moral implications of his acts.”

Resistance can also be overcome by using ‘Just World’ thinking – or as Melvin Lerner (1980) puts it: people deserve what they get in a just world. According to Leon Festinger (1957), cognitive dissonance is a clash between 2 contradictory attitudes held by the same person. In the Milgram experiments, the cognitive dissonance – the desire to obey vs reluctance to harm another – led to huge moral strain. This could be reduced by seeing the ‘learner’s’ suffering as deserved. Some of Milgram’s ‘teachers’ justified going to the maximum voltage by blaming the ‘learner’. As Milgram (1974, p10) wrote: “Once having acted against the victim, these subjects found it necessary to view him as an unworthy individual whose punishment was made inevitable by his own deficiencies of intellect and character.” Clearly the BLUE vMEME underpins ‘Just World thinking.

An interesting piece of research supporting the Just World Hypothesis comes from Amerigo Farina, Charles Holland & Klaus Ring (1966). In what was otherwise a replication of Milgram, the researchers had the ‘teacher’ spend time talking with the confederate ‘learner’ before the experiment began. (On the pretext that getting to know someone helps communication.) With some teachers, the learner talked positively about himself. With others, he talked negatively – eg: unhappy childhood, history of illness, parents divorced. Farina, Holland & Ring found that teachers gave more intense shocks when the learner had talked negatively about himself.
In post-experiment questionnaires, teachers demonstrated that they were less interested in getting to know the learner when he was talking negatively about himself.

Dispostional explanations
These attribute the cause of behaviour as coming from within the individual – either their biology or their personality.

Florence Miale & Michael Selzer (1975) proposed the ‘psychopathic personality’, arguing that Milgram’s experiments made giving shocks in a Psychology laboratory a socially acceptable expression of violent and psychologically disordered impulses. However, Milgram had been at pains to point out that most of his participants gave lower voltage shocks when allowed.

A dispositional explanation with rather more credibility is the authoritarian personality, first proposed by Erich Fromm in 1941. From their research into prejudice – see the resulting model on the Comparison Map –  Theodore Adorno et al (1950) established this ‘personality type’ from use of the Fascism Scale (F-Scale)

Adorno et al administered the test to a sample of 2,000 Americans. Respondents were asked their strength of agreement or disagreement with statements so that researchers could determine their attitudes towards religious and ethnic minorities, their views on politics and economics, and their moral values. About 1/10 of the participants, equally balanced between male and female, were interviewed in more depth under the supervision of Adorno’s co-author, Else Frenkel-Brunswik . This sub-sample consisted of one group which had expressed the most prejudiced and authoritarian views and one group which had expressed the least such views. The males were equally balanced between high and low scorers in authoritarianism but the females were divided between 25 high and 15 low. The groups, matched on age, political and religious affiliation, and national or regional background, were compared to see which other factors seemed to give rise to the authoritarian personality.

From the results of these tests, Adorno et al established the concept of the authoritarian personality with the following characteristics:-

  • personality type submissive to authority figures
  • prone to unquestioning obedience  – and hostile to those who don’t obey
  • treats those beneath them with contempt  – Adorno et al identified disguised sadism as an element in the authoritarian personality.
  • stereotypic thinking
  • intolerant of ambiguity – Adorno et al postulated that the authoritarian personality projects their own unacceptable sexual and aggressive impulses onto minority groups and, thus, finds them threatening.
  • rigid beliefs in conventional values

Frenkel-Brunswik (1942) had found that authoritarian-type personalities were likely to have been strictly treated as a child and punished for disobedience. In Psychoanalytic Theory, which Adorno et al used for explanatory purposes, childhood maltreatment by the father leads to repression and displacement. Adorno et al’s theory was that the maltreated child would repress the hostility towards his father. Consequently, on the face of it the child seems to idolise his parents and in later life acts in a submissive way towards authority figures. However, there is still much hostility towards the parents lying below the surface which is displaced onto non-threatening minority groups as prejudice & discrimination.

The idea that the authoritarian personality developed from children’s upbringing has been criticised as an explanation of Germany’s embrace of the Nazis. Extreme anti-Semitism developed during the Nazi era (1932-45), a relatively short period and one deemed far too short for a large-scale change in German parenting styles. It should be noted here that Adorno had a background in Psychoanalitic Theory which may have influenced him placing undue emphasis on unresolved childhood conflicts with parents having an unconscious influence on behaviour.

Adorno et al’s study was severely criticised by Roger Brown (1965) who thought the structure of the test invited acquiescence responses – ie: the respondents agree with one item because they’ve agreed with other items.  Robert Rosenthal (1966) pointed out that the interviewers knew both the hypothesis and the interviewees’ test scores. Therefore, there was more than a possibility of researcher bias influencing the process and, therefore, the results.


Go back…>