asch experiment group size

Solomon Asch Conformity Line Experiment Study

Saul McLeod, PhD

Editor-in-Chief for Simply Psychology

BSc (Hons) Psychology, MRes, PhD, University of Manchester

Saul McLeod, PhD., is a qualified psychology teacher with over 18 years of experience in further and higher education. He has been published in peer-reviewed journals, including the Journal of Clinical Psychology.

Learn about our Editorial Process

Olivia Guy-Evans, MSc

Associate Editor for Simply Psychology

BSc (Hons) Psychology, MSc Psychology of Education

Olivia Guy-Evans is a writer and associate editor for Simply Psychology. She has previously worked in healthcare and educational sectors.

On This Page:

Solomon Asch experimented with investigating the extent to which social pressure from a majority group could affect a person to conform .

He believed the main problem with Sherif’s (1935) conformity experiment was that there was no correct answer to the ambiguous autokinetic experiment.  How could we be sure that a person conformed when there was no correct answer?

Asch (1951) devised what is now regarded as a classic experiment in social psychology, whereby there was an obvious answer to a line judgment task.

If the participant gave an incorrect answer, it would be clear that this was due to group pressure.

Experimental Procedure

Asch used a lab experiment to study conformity, whereby 50 male students from Swarthmore College in the USA participated in a ‘vision test.’

Using a line judgment task, Asch put a naive participant in a room with seven confederates/stooges. The confederates had agreed in advance what their responses would be when presented with the line task.

The real participant did not know this and was led to believe that the other seven confederates/stooges were also real participants like themselves.

Each person in the room had to state aloud which comparison line (A, B or C) was most like the target line. The answer was always obvious.  The real participant sat at the end of the row and gave his or her answer last.

At the start, all participants (including the confederates) gave the correct answers. However, after a few rounds, the confederates started to provide unanimously incorrect answers.

There were 18 trials in total, and the confederates gave the wrong answer on 12 trials (called the critical trials).  Asch was interested to see if the real participant would conform to the majority view.

Asch’s experiment also had a control condition where there were no confederates, only a “real participant.”

Asch measured the number of times each participant conformed to the majority view. On average, about one third (32%) of the participants who were placed in this situation went along and conformed with the clearly incorrect majority on the critical trials.

Over the 12 critical trials, about 75% of participants conformed at least once, and 25% of participants never conformed.

In the control group , with no pressure to conform to confederates, less than 1% of participants gave the wrong answer.

Why did the participants conform so readily?  When they were interviewed after the experiment, most of them said that they did not really believe their conforming answers, but had gone along with the group for fear of being ridiculed or thought “peculiar.

A few of them said that they did believe the group’s answers were correct.

Apparently, people conform for two main reasons: because they want to fit in with the group ( normative influence ) and because they believe the group is better informed than they are ( informational influence ).

Critical Evaluation

One limitation of the study is that is used a biased sample. All the participants were male students who all belonged to the same age group. This means that the study lacks population validity and that the results cannot be generalized to females or older groups of people.

Another problem is that the experiment used an artificial task to measure conformity – judging line lengths. How often are we faced with making a judgment like the one Asch used, where the answer is plain to see?

This means that the study has low ecological validity and the results cannot be generalized to other real-life situations of conformity. Asch replied that he wanted to investigate a situation where the participants could be in no doubt what the correct answer was. In so doing he could explore the true limits of social influence.

Some critics thought the high levels of conformity found by Asch were a reflection of American, 1950’s culture and told us more about the historical and cultural climate of the USA in the 1950s than then they did about the phenomena of conformity.

In the 1950s America was very conservative, involved in an anti-communist witch-hunt (which became known as McCarthyism) against anyone who was thought to hold sympathetic left-wing views.

Perrin and Spencer

Conformity to American values was expected. Support for this comes from studies in the 1970s and 1980s that show lower conformity rates (e.g., Perrin & Spencer, 1980).

Perrin and Spencer (1980) suggested that the Asch effect was a “child of its time.” They carried out an exact replication of the original Asch experiment using engineering, mathematics, and chemistry students as subjects. They found that in only one out of 396 trials did an observer join the erroneous majority.

Perrin and Spencer argue that a cultural change has taken place in the value placed on conformity and obedience and in the position of students.

In America in the 1950s, students were unobtrusive members of society, whereas now, they occupy a free questioning role.

However, one problem in comparing this study with Asch is that very different types of participants are used. Perrin and Spencer used science and engineering students who might be expected to be more independent by training when it came to making perceptual judgments.

Finally, there are ethical issues : participants were not protected from psychological stress which may occur if they disagreed with the majority.

Evidence that participants in Asch-type situations are highly emotional was obtained by Back et al. (1963) who found that participants in the Asch situation had greatly increased levels of autonomic arousal.

This finding also suggests that they were in a conflict situation, finding it hard to decide whether to report what they saw or to conform to the opinion of others.

Asch also deceived the student volunteers claiming they were taking part in a “vision” test; the real purpose was to see how the “naive” participant would react to the behavior of the confederates. However, deception was necessary to produce valid results.

The clip below is not from the original experiment in 1951, but an acted version for television from the 1970s.

Factors Affecting Conformity

In further trials, Asch (1952, 1956) changed the procedure (i.e., independent variables) to investigate which situational factors influenced the level of conformity (dependent variable).

His results and conclusions are given below:

Asch (1956) found that group size influenced whether subjects conformed. The bigger the majority group (no of confederates), the more people conformed, but only up to a certain point.

With one other person (i.e., confederate) in the group conformity was 3%, with two others it increased to 13%, and with three or more it was 32% (or 1/3).

Optimum conformity effects (32%) were found with a majority of 3. Increasing the size of the majority beyond three did not increase the levels of conformity found. Brown and Byrne (1997) suggest that people might suspect collusion if the majority rises beyond three or four.

According to Hogg & Vaughan (1995), the most robust finding is that conformity reaches its full extent with 3-5 person majority, with additional members having little effect.

Lack of Group Unanimity / Presence of an Ally

The study also found that when any one individual differed from the majority, the power of conformity significantly decreased.

This showed that even a small dissent can reduce the power of a larger group, providing an important insight into how individuals can resist social pressure.

As conformity drops off with five members or more, it may be that it’s the unanimity of the group (the confederates all agree with each other) which is more important than the size of the group.

In another variation of the original experiment, Asch broke up the unanimity (total agreement) of the group by introducing a dissenting confederate.

Asch (1956) found that even the presence of just one confederate that goes against the majority choice can reduce conformity by as much as 80%.

For example, in the original experiment, 32% of participants conformed on the critical trials, whereas when one confederate gave the correct answer on all the critical trials conformity dropped to 5%.

This was supported in a study by Allen and Levine (1968). In their version of the experiment, they introduced a dissenting (disagreeing) confederate wearing thick-rimmed glasses – thus suggesting he was slightly visually impaired.

Even with this seemingly incompetent dissenter, conformity dropped from 97% to 64%. Clearly, the presence of an ally decreases conformity.

The absence of group unanimity lowers overall conformity as participants feel less need for social approval of the group (re: normative conformity).

Difficulty of Task

When the (comparison) lines (e.g., A, B, C) were made more similar in length it was harder to judge the correct answer and conformity increased.

When we are uncertain, it seems we look to others for confirmation. The more difficult the task, the greater the conformity.

Answer in Private

When participants were allowed to answer in private (so the rest of the group does not know their response), conformity decreased.

This is because there are fewer group pressures and normative influence is not as powerful, as there is no fear of rejection from the group.

Frequently Asked Questions

How has the asch conformity line experiment influenced our understanding of conformity.

The Asch conformity line experiment has shown that people are susceptible to conforming to group norms even when those norms are clearly incorrect. This experiment has significantly impacted our understanding of social influence and conformity, highlighting the powerful influence of group pressure on individual behavior.

It has helped researchers to understand the importance of social norms and group dynamics in shaping our beliefs and behaviors and has had a significant impact on the study of social psychology.

What are some real-world examples of conformity?

Examples of conformity in everyday life include following fashion trends, conforming to workplace norms, and adopting the beliefs and values of a particular social group. Other examples include conforming to peer pressure, following cultural traditions and customs, and conforming to societal expectations regarding gender roles and behavior.

Conformity can have both positive and negative effects on individuals and society, depending on the behavior’s context and consequences.

What are some of the negative effects of conformity?

Conformity can have negative effects on individuals and society. It can limit creativity and independent thinking, promote harmful social norms and practices, and prevent personal growth and self-expression.

Conforming to a group can also lead to “groupthink,” where the group prioritizes conformity over critical thinking and decision-making, which can result in poor choices.

Moreover, conformity can spread false information and harmful behavior within a group, as individuals may be afraid to challenge the group’s beliefs or actions.

How does conformity differ from obedience?

Conformity involves adjusting one’s behavior or beliefs to align with the norms of a group, even if those beliefs or behaviors are not consistent with one’s personal views. Obedience , on the other hand, involves following the orders or commands of an authority figure, often without question or critical thinking.

While conformity and obedience involve social influence, obedience is usually a response to an explicit request or demand from an authority figure, whereas conformity is a response to implicit social pressure from a group.

What is the Asch effect?

The Asch Effect is a term coined from the Asch Conformity Experiments conducted by Solomon Asch. It refers to the influence of a group majority on an individual’s judgment or behavior, such that the individual may conform to perceived group norms even when those norms are obviously incorrect or counter to the individual’s initial judgment.

This effect underscores the power of social pressure and the strong human tendency towards conformity in group settings.

What is Solomon Asch’s contribution to psychology?

Solomon Asch significantly contributed to psychology through his studies on social pressure and conformity.

His famous conformity experiments in the 1950s demonstrated how individuals often conform to the majority view, even when clearly incorrect.

His work has been fundamental to understanding social influence and group dynamics’ power in shaping individual behaviors and perceptions.

Allen, V. L., & Levine, J. M. (1968). Social support, dissent and conformity. Sociometry , 138-149.

Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgment. In H. Guetzkow (ed.) Groups, leadership and men . Pittsburgh, PA: Carnegie Press.

Asch, S. E. (1952). Group forces in the modification and distortion of judgments.

Asch, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological monographs: General and applied, 70(9) , 1-70.

Back, K. W., Bogdonoff, M. D., Shaw, D. M., & Klein, R. F. (1963). An interpretation of experimental conformity through physiological measures. Behavioral Science, 8(1) , 34.

Bond, R., & Smith, P. B. (1996). Culture and conformity : A meta-analysis of studies using Asch’s (1952b, 1956) line judgment task.  Psychological bulletin ,  119 (1), 111.

Longman, W., Vaughan, G., & Hogg, M. (1995). Introduction to social psychology .

Perrin, S., & Spencer, C. (1980). The Asch effect: a child of its time? Bulletin of the British Psychological Society, 32, 405-406.

Sherif, M., & Sherif, C. W. (1953). Groups in harmony and tension . New York: Harper & Row.

Asch Conformity Experiments: Line Study

Categories Social Psychology

Will people conform to the group’s opinions, even if they disagree? That was the question behind one of the most famous experiments in psychology history. The Asch conformity experiments were a series of studies by social psychologist Solomon Asch during the 1950s. In the studies, Asch sought to learn more about how social pressure could lead to conformity .

In the studies, people were asked to choose a line that matched the length of another line. When the others in the group chose the incorrect line, participants would often conform to the rest of the group, even though they were clearly wrong.

The experiments are classic studies in social psychology, offering important insights into when and why people conform to group norms and pressures.

Key Takeaways: The Asch conformity experiments demonstrated that people often conform to group opinions, even when they know the group is wrong. Factors such as the desire for social acceptance and uncertainty in decision-making contribute to conformity. Group size, status, and whether responses are public or private influence the likelihood of conforming. Despite criticism, Asch’s experiments significantly impacted the understanding of social behavior and group influence in psychology.

Table of Contents

The Asch Experiments

In the main version of the experiment, Asch told the participants that they were taking part in a vision test. Each participant was then placed in a group of people who were actually confederates in the study. In other words, they were actors who were involved in the experiment.

The group was shown a line on a card and then another card with several lines of varying lengths. They were asked to pick the line that matched the first line.

It was a simple task. When asked on their own, almost all participants were able to easily perform the task correctly. When they were in the group, and the confederates gave the wrong answers, the participants were often go along with the group.

Results of the Asch Conformity Experiments

The results of the Asch conformity experiments were startling. They revealed that a staggering 75% of the participants conformed to the group at least once. Even more surprising, about 25% never conformed, while 5% conformed every single time.

For the control group, where people faced no social pressure, incorrect responses were given less than 1% of the time.

Explanations for the Results

What explains the high rates of conformity in Asch’s experiments? There are several important psychological factors at work. The reasons people went along with the group even when they knew the others were wrong come down to several reasons:

Normative Social Influence

People have a desire for social acceptance. They want to fit in with the group and prefer not to stand out. By agreeing with the rest of the group, they increase the likelihood of being liked and accepted by others.

The fear of embarrassment can also play a role. Being the only one to voice a different answer comes with the risk of appearing foolish or being ridiculed. Even if people knew they were right, fear of social disapproval caused them to conform.

Informational Social Influence

When making decisions under uncertainty, people often look to other people as a source of information. If other people say one thing is correct, people often assume that others know something they don’t, which is why they conform.

Self-doubt in these situations can also play a role. Once others started choosing the wrong answer, the participants may have started to question their response and wondered if they had overlooked something.

Other Factors That Can Influence Conformity

There are also a number of other factors that can affect the likelihood that people with conform like they did Asch conformity experiments.

These include:

• Group size : Conformity usually increases with group size, at least up to a certain point. When 3 to 5 people are present, there is a lot of pressure to conform. When the number of people exceeds that, conformity typically starts to decline.
• Status : People are more likely to conform if the others in the group are seen as having a higher status, more authority, or greater expertise.
• Privacy of responses : People are more inclined to conform if their responses are public. When responses are private, conformity rates drop.
• Uncertainty and difficulty : If the task is ambiguous or difficult, people are less likely to trust their own judgment. They will often look to others for information and assurance, which increases conformity.
• Group unity : Conformity is higher in very cohesive groups. The stronger the bonds between group members, the more likely people are to conform.

In a 2023 replication of Asch’s conformity experiment, researchers found an error rate of 33%, similar to the one in Asch’s original study. They found that offering monetary incentives helped reduce errors but didn’t eliminate the effects of social influence. The study also found that social influence impacted political opinions, leading to a conformity rate of 38% (Franzen & Mader, 2023).

The study also examined how Big Five personality factors might be linked to conformity. While openness was associated with susceptibility to group pressure, other personality traits were not significantly connected.

One 2018 experiment found that the social delivery of information caused 33% of participants to change their political opinions (Mallinson & Hatemi, 2018).

Critiques of the Asch Conformity Experiments

While influential, the Asch experiments were not without criticism. Some of the main criticisms hinge on the following:

• The impact of demand characteristics : Some critics suggest that some participants may have suspected the study’s real intentions and behaved to meet the experimenter’s expectations.
• Lack of relevance in the real world : Critics also suggest that the experimental setup needed to be more contrived and accurately reflect real-world situations where conformity might occur.
• Cultural factors : The time and place of the experiments (the United States and during the 1950s) may also have contributed to the high conformity rates. During that time, conformity to American norms and values was highly valued. Such characteristics may not be universal to other places and periods.
• Simplified approach: While Asch’s experiments demonstrate one aspect of conformity (normative social influence), they don’t address the many other factors that can contribute to this behavior in real-world settings.

Impact and Contributions of the Asch Conformity Experiments

Asch’s conformity experiments had a major impact on the field of psychology. They helped inspire further research on conformity, compliance, and obedience.

The studies demonstrated that conformity is not just about fear of punishment ; it often comes from a deep psychological need for acceptance and group harmony.

These findings have influenced a wide range of fields, from understanding peer pressure and decision-making in groups to exploring the dynamics of social behavior in various cultural and political contexts. Asch’s experiments remain a cornerstone in social psychology , shaping how we think about the relationships between individual judgment and group influence.

EXPERT PICKS: WHAT TO READ NEXT Classic Psychological Experiments : Explore other well-known psychology experiments. The Robbers Cave Experiment : What would happen if you pitted two groups of kids against one another? Learn more about what researchers discovered. Kohlberg’s Stages of Moral Development : This theory explores the stages and factors that play a role in the development of moral behavior. What Is the Ingroup Bias? People tend to favor those who are in their own social groups, which can affect how they respond to people who are different from them.

Asch, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority . Psychological Monographs: General and Applied , 70(9), 1–70. https://doi.org/10.1037/h0093718

Franzen, A., & Mader, S. (2023). The power of social influence: A replication and extension of the Asch experiment . PloS one , 18 (11), e0294325. https://doi.org/10.1371/journal.pone.0294325

Levine J. M. (1999). Solomon Asch’s legacy for group research . Personality and Social Psychology Review : An Official Journal of the Society for Personality and Social Psychology, Inc , 3 (4), 358–364. https://doi.org/10.1207/s15327957pspr0304_5

Mallinson, D. J., & Hatemi, P. K. (2018). The effects of information and social conformity on opinion change . PloS One , 13 (5), e0196600. https://doi.org/10.1371/journal.pone.0196600

• Bipolar Disorder
• Therapy Center
• When To See a Therapist
• Types of Therapy
• Best Online Therapy
• Best Couples Therapy
• Managing Stress
• Sleep and Dreaming
• Understanding Emotions
• Self-Improvement
• Healthy Relationships
• Student Resources
• Personality Types
• Sweepstakes
• Guided Meditations
• Verywell Mind Insights
• 2024 Verywell Mind 25
• Mental Health in the Classroom
• Editorial Process
• Meet Our Review Board
• Crisis Support

The Asch Conformity Experiments

What These Experiments Say About Group Behavior

What Is Conformity?

Factors that influence conformity.

The Asch conformity experiments were a series of psychological experiments conducted by Solomon Asch in the 1950s. The experiments revealed the degree to which a person's own opinions are influenced by those of a group . Asch found that people were willing to ignore reality and give an incorrect answer in order to conform to the rest of the group.

At a Glance

The Asch conformity experiments are among the most famous in psychology's history and have inspired a wealth of additional research on conformity and group behavior. This research has provided important insight into how, why, and when people conform and the effects of social pressure on behavior.

Do you think of yourself as a conformist or a non-conformist? Most people believe that they are non-conformist enough to stand up to a group when they know they are right, but conformist enough to blend in with the rest of their peers.

Research suggests that people are often much more prone to conform than they believe they might be.

Imagine yourself in this situation: You've signed up to participate in a psychology experiment in which you are asked to complete a vision test.

Seated in a room with the other participants, you are shown a line segment and then asked to choose the matching line from a group of three segments of different lengths.

The experimenter asks each participant individually to select the matching line segment. On some occasions, everyone in the group chooses the correct line, but occasionally, the other participants unanimously declare that a different line is actually the correct match.

So what do you do when the experimenter asks you which line is the right match? Do you go with your initial response, or do you choose to conform to the rest of the group?

Conformity in Psychology

In psychological terms, conformity refers to an individual's tendency to follow the unspoken rules or behaviors of the social group to which they belong. Researchers have long been been curious about the degree to which people follow or rebel against social norms.

Asch was interested in looking at how pressure from a group could lead people to conform, even when they knew that the rest of the group was wrong. The purpose of the Asch conformity experiment was to demonstrate the power of conformity in groups.

Methodology of Asch's Experiments

Asch's experiments involved having people who were in on the experiment pretend to be regular participants alongside those who were actual, unaware subjects of the study. Those that were in on the experiment would behave in certain ways to see if their actions had an influence on the actual experimental participants.

In each experiment, a naive student participant was placed in a room with several other confederates who were in on the experiment. The subjects were told that they were taking part in a "vision test." All told, a total of 50 students were part of Asch’s experimental condition.

The confederates were all told what their responses would be when the line task was presented. The naive participant, however, had no inkling that the other students were not real participants. After the line task was presented, each student verbally announced which line (either 1, 2, or 3) matched the target line.

Critical Trials

There were 18 different trials in the experimental condition , and the confederates gave incorrect responses in 12 of them, which Asch referred to as the "critical trials." The purpose of these critical trials was to see if the participants would change their answer in order to conform to how the others in the group responded.

During the first part of the procedure, the confederates answered the questions correctly. However, they eventually began providing incorrect answers based on how they had been instructed by the experimenters.

Control Condition

The study also included 37 participants in a control condition . In order to ensure that the average person could accurately gauge the length of the lines, the control group was asked to individually write down the correct match. According to these results, participants were very accurate in their line judgments, choosing the correct answer 99% of the time.

Results of the Asch Conformity Experiments

Nearly 75% of the participants in the conformity experiments went along with the rest of the group at least one time.

After combining the trials, the results indicated that participants conformed to the incorrect group answer approximately one-third of the time.

The experiments also looked at the effect that the number of people present in the group had on conformity. When just one confederate was present, there was virtually no impact on participants' answers. The presence of two confederates had only a tiny effect. The level of conformity seen with three or more confederates was far more significant.

Asch also found that having one of the confederates give the correct answer while the rest of the confederates gave the incorrect answer dramatically lowered conformity. In this situation, just 5% to 10% of the participants conformed to the rest of the group (depending on how often the ally answered correctly). Later studies have also supported this finding, suggesting that having social support is an important tool in combating conformity.

At the conclusion of the Asch experiments, participants were asked why they had gone along with the rest of the group. In most cases, the students stated that while they knew the rest of the group was wrong, they did not want to risk facing ridicule. A few of the participants suggested that they actually believed the other members of the group were correct in their answers.

These results suggest that conformity can be influenced both by a need to fit in and a belief that other people are smarter or better informed.

Given the level of conformity seen in Asch's experiments, conformity can be even stronger in real-life situations where stimuli are more ambiguous or more difficult to judge.

Asch went on to conduct further experiments in order to determine which factors influenced how and when people conform. He found that:

• Conformity tends to increase when more people are present . However, there is little change once the group size goes beyond four or five people.
• Conformity also increases when the task becomes more difficult . In the face of uncertainty, people turn to others for information about how to respond.
• Conformity increases when other members of the group are of a higher social status . When people view the others in the group as more powerful, influential, or knowledgeable than themselves, they are more likely to go along with the group.
• Conformity tends to decrease, however, when people are able to respond privately . Research has also shown that conformity decreases if people have support from at least one other individual in a group.

Criticisms of the Asch Conformity Experiments

One of the major criticisms of Asch's conformity experiments centers on the reasons why participants choose to conform. According to some critics, individuals may have actually been motivated to avoid conflict, rather than an actual desire to conform to the rest of the group.

Another criticism is that the results of the experiment in the lab may not generalize to real-world situations.

Many social psychology experts believe that while real-world situations may not be as clear-cut as they are in the lab, the actual social pressure to conform is probably much greater, which can dramatically increase conformist behaviors.

Asch SE. Studies of independence and conformity: I. A minority of one against a unanimous majority . Psychological Monographs: General and Applied . 1956;70(9):1-70. doi:10.1037/h0093718

Morgan TJH, Laland KN, Harris PL. The development of adaptive conformity in young children: effects of uncertainty and consensus . Dev Sci. 2015;18(4):511-524. doi:10.1111/desc.12231

Asch SE. Effects of group pressure upon the modification and distortion of judgments . In: Guetzkow H, ed.  Groups, Leadership and Men; Research in Human Relations. Carnegie Press. 1951:177–190.

Britt MA. Psych Experiments: From Pavlov's Dogs to Rorschach's Inkblots . Adams Media. 

Myers DG. Exploring Psychology (9th ed.). Worth Publishers.

By Kendra Cherry, MSEd Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

• Abnormal Psychology
• Assessment (IB)
• Biological Psychology
• Cognitive Psychology
• Criminology
• Developmental Psychology
• Extended Essay
• General Interest
• Health Psychology
• Human Relationships
• IB Psychology
• IB Psychology HL Extensions
• Internal Assessment (IB)
• Love and Marriage
• Post-Traumatic Stress Disorder
• Prejudice and Discrimination
• Qualitative Research Methods
• Research Methodology
• Revision and Exam Preparation
• Social and Cultural Psychology
• Studies and Theories
• Teaching Ideas

Key Study: Conformity – Asch (1955)

Travis Dixon October 4, 2016 Uncategorized

• Click to share on Facebook (Opens in new window)
• Click to share on Twitter (Opens in new window)
• Click to share on LinkedIn (Opens in new window)
• Click to share on Pinterest (Opens in new window)
• Click to email a link to a friend (Opens in new window)

Background Information

Humans are social animals, formign groups and strong bonds naturally. As such, it’s not hard to see the many ways that belonging to a group is important. Conformity is one effect that can happen as a result of this need to belong. Conformity is when behaviour is modified in order to fit in with a larger group.

Methodology and Results

For the following experiments Asch used the same experimental paradigm using the line length cards (which has come to be known as the Asch Paradigm). It involves matching one line with one from a group of three.

Typical Experimental Condition

7 to 9 male college students get together in a room, but only one of them is a naïve subject – the rest are confederates. i.e. the subject thinks everyone is just there like him, but in actual fact they all know what is going on. They sit in a row at a large desk and face the experimenter who holds up cards like those in figure 1. The subject is near the end of the line, so by the time they have given their response they have heard the responses of most of the group. The first two “trials” the confederates give the right answer and it all appears a bit boring. On the third trial, however, they deliberately give the wrong answer. In total there are 18 trials, but only in 12 of these do the group deliberately give the wrong answer (when this happens it’s called a “critical trial”).

In Asch’s first experiment reported in his 1955 paper, there were 123 subjects from three different colleges. The results were as follows:

• When alone: almost 100% accurate (<1% error rate)
• When in the group, the subjects conformed 36.8% of the time

In the results above, there was a range of individual variability:

• About 25% of subjects never agreed with the group
• Some subjects always agreed with the group

The subjects were interviewed after each experiment so the researchers could find out more about why they went along with the group. However, these were not investigated extensively. In his report he makes some generalizations, saying confidence in one’s own opinion was one factor that would have influenced the rate of conformity. There were some who adopted the idea, “I am wrong, they are right” and so changed their answers accordingly. Other subjects changed their answers because they did want to spoil the researchers’ results.

Interestingly, all subjects who conformed underestimated the frequency with which they conformed to the group.

Asch wanted to explore two dynamics that he thought might influence conformity: group size and unanimity.

Experiment 2: Group Size

In this experiment this size of the “opposition” (i.e, number of confederates) ranged from 1 to 15. In this modification, when there was only one confederate the subjects did not change their answers but instead gave conflicting (and correct) responses. However, as the group grew the results changed:

• One confederate: almost 0% conformity
• Two confederates: 13.6% conformity
• Three confederates: 31.8% conformity

Interestingly, this positive correlation between group size and conformity rate only goes up until a group of four (i.e. three confederates, one subject). After more confederates are introduced the rate does not change and is around 30%. However, after 8 or more confederates the rate of conformity begins to drop.

Experiment 3: Unanimity

Unanimity is when all people are in agreement, so in this context it refers to the extent to which all confederates give the wrong answer. In this design, Asch introduced either another naïve subject or a confederate who was told to not give the wrong answer on the critical trials. The results showed that conformity rates dropped when this happened (the conformity rate was ¼ of that in the regular design). During post-experiment interviews it was revealed that subjects felt a sense of closeness and warmth with their “partner.”

Experiment 4: Dissent

Based on the results with a partner, Asch posed another interesting question: “Was the partner’s effect a consequence of his dissent, or was it related to his accuracy?” Dissent is going against what is commonly believed, so in this context it means breaking away from the group and giving different answers. To test the above question, Asch made more modifications. This time a confederate was told to be either a “compromising” or “extremist” dissenter.

• Compromising: going against the group but the difference in answers is close
• Extremist: going against the group and given a very different answer

Both conditions reduced conformity. Moreover, the extremist dissenter “produced a remarkable freeing of the subjects.” When there was an extreme dissenter conformity dropped to 9%.

Experiment 5: Gaining and Losing a Partner

At this point I think it’s important just to mention that all these modifications within the experiment were done on different subjects.

In this next design, Asch tested what would happen if a subject gains a fellow dissenter but then this partner decides to join the incorrect group. A confederate was told to answer correctly on the first 6 critical trials. During these trials, the subjects also had no problems going against the rest of the group and giving the correct answer. However, after 6 trials the dissenting confederate changed and began giving incorrect answers along with the rest of the group – and as a result so did the subject. Having lost the partner to the majority the subject’s conformity rate rose to a similar level as those in Experiment #1 (i.e. about 1/3).

Experiment 5b: Gaining and Losing a Partner from the Group

As a result of the questioning from the previous experiments (27 subjects in total), Asch decided to see what would happen if the partner left the room completely. It was announced at a certain point after giving the correct answer on a few trials the partner said he had an appointment with the dean and had to leave. While there were still errors made (i.e. the participant conformed), the rates were lower than when the partner joined the majority.

Experiment 6: Gradual Dissolution

Asch had all confederates begin by giving correct answers and gradually changing to incorrect ones. By the sixth trial all the confederates were now giving wrong answers. The results showed that as long as there was just one other person against the group the subject could stay independent. However, as soon as he lost all partners the conformity rates increased.

7: Degree of Inaccuracy

Asch hypothesized that if the group’s answers were wrong enough the conformity would disappear. However, even when the different in line lengths were 7 inches (18 cm) there were still participants who conformed.

The above studies provide some interesting insight into factors that influence conformity. Asch suggested the following factors might influence conformity and since this paper was published in 1955 these have been studied:

• Social and cultural factors (Bond and Smith)
• Personality and characte

Travis Dixon is an IB Psychology teacher, author, workshop leader, examiner and IA moderator.

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

The power of social influence: A replication and extension of the Asch experiment

Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing – original draft

* E-mail: [email protected]

Affiliation Institute of Sociology, University of Bern, Bern, Switzerland

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing

• Axel Franzen, 
• Sebastian Mader

• Published: November 29, 2023
• https://doi.org/10.1371/journal.pone.0294325
• Reader Comments

In this paper, we pursue four goals: First, we replicate the original Asch experiment with five confederates and one naïve subject in each group (N = 210). Second, in a randomized trial we incentivize the decisions in the line experiment and demonstrate that monetary incentives lower the error rate, but that social influence is still at work. Third, we confront subjects with different political statements and show that the power of social influence can be generalized to matters of political opinion. Finally, we investigate whether intelligence, self-esteem, the need for social approval, and the Big Five are related to the susceptibility to provide conforming answers. We find an error rate of 33% for the standard length-of-line experiment which replicates the original findings by Asch (1951, 1955, 1956). Furthermore, in the incentivized condition the error rate decreases to 25%. For political opinions we find a conformity rate of 38%. However, besides openness, none of the investigated personality traits are convincingly related to the susceptibility of group pressure.

Citation: Franzen A, Mader S (2023) The power of social influence: A replication and extension of the Asch experiment. PLoS ONE 18(11): e0294325. https://doi.org/10.1371/journal.pone.0294325

Editor: Mike Farjam, Lund University: Lunds Universitet, SWEDEN

Received: March 17, 2023; Accepted: October 30, 2023; Published: November 29, 2023

Copyright: © 2023 Franzen, Mader. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data used in this study is publicly available in the repository of the University of Bern at https://boris.unibe.ch/id/eprint/169645 .

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

A core assumption in sociology is that what humans think and do does not only depend on their own attitudes and disposition, but also to a large extent on what others think and do. The power of social influence on individuals’ behavior was demonstrated already in the 1950s in a series of experiments by Solomon Asch [ 1 – 3 ]. Asch invited individuals into the lab and assigned them the task of judging the length of a line. He also placed 6 confederates into the lab who were assigned to give wrong answers publicly, so that the naïve subject could hear them before he provided his own answer. The results were very surprising: on average 35% of the real subjects followed the opinions of the confederates even if their answer was obviously wrong. The work of Asch has attracted a great amount of attention in the social sciences. Hence, a multitude of replications, extensions, and variations of the original studies have been conducted. However, many of these replications were done with student samples in the US, and fewer studies consist of samples from other countries. Furthermore, many replications were undertaken in the 40 years following the original experiment of Asch, but there are fewer replications thereafter. This raises two important questions: First, are the findings of Asch universal or do they predominantly apply to American students? And second, are the findings still valid today or has the influence of others diminished over time, for instance through increased education and democratization?

Moreover, many experiments in psychology are not incentivized by monetary rewards. This is also true for Asch’s original experiments and for most replications of it. However, in real life outside the lab, decisions are usually associated with consequences, either pleasant in the form of rewards, or unpleasant in the form of some kind of punishment. To make the study of decision-making more realistic, experiments in economics usually use monetary incentives [ 4 ]. To provide a conforming but wrong judgment in the original Asch experiment has no consequences, giving rise to the interesting question of whether the finding of Asch still holds when correct answers are rewarded. So far, the effect of incentives in the Asch decision situation has only been investigated rarely [ 5 – 7 ], with inconclusive evidence. Baron et al. [ 5 ] report that use of monetary incentives actually increased conformity when the task was difficult. A decreased conformity rate was only found in situations with easy tasks. Bhanot & Williamson [ 6 ] conducted two online experiments and found that incentivizing correct answers increases the number of conforming answers. Fujita and Mori [ 7 ] compared group reward and individual rewards in the Asch experiment and found that conformity vanished in the individual reward condition. Thus, the existing evidence on the role of incentives is inconclusive, calling for further investigations of the effect of incentives.

Of course, misjudging the length of lines when others do is not important in itself; the Asch experiment created so much attention because it elicits the suspicion that social influence is also present in other and more important social realms, for instance when it comes to political opinions. Early research by Crutchfield [ 8 ] suggests that the original findings on line judgment also transfer over to political opinions. We are only aware of one further study by Mallinson and Hatemi [ 9 ] that investigates the effect of social influence on opinion formation. However, the authors used a group discussion in the treatment condition, and hence diverged somewhat from the original Asch design. Furthermore, investigations of the effect of social conformity on political opinions are always idiosyncratic making further replications on the transferability from lines to a variety of political opinions important and interesting.

Moreover, behaving in a conforming way and misjudging tasks raises a number of interesting questions. About one third of Asch’s subjects was susceptible to social pressure on average. The rest solved the task correctly irrespective of the confederates’ opinion most of the time. How do those who are not influenced by the group differ from the ones that behave in a conformative manner? Crutchfield [ 8 ] investigated a number of personality traits such as competence, self-assertiveness, or leadership ability on the susceptibility to the pressure to conform to the groups’ judgment. However, many of the measurement instruments used by him or by others [ 10 , 11 ] investigating similar questions are suboptimal, and furthermore produced inconclusive results. Hence, it is worthwhile to further investigate the characteristics of those who conform to social pressure and of those who resist it. We are particularly interested in the Big Five, intelligence, self-esteem, and the need for social approval.

The remainder of the article proceeds in four sections. First, in section two, we present an elaborate literature review of the original Asch experiment, and its various replications. Section three describes how we conducted the replication of the Asch experiment and its variant by using political opinions. Furthermore, we describe how we implemented the incentives, and how we measure the various traits that are presumably related to behavior in the Asch experiment. Section four presents the results and section five concludes and discusses ideas for further research.

2. Literature review

In Asch’s [ 2 ] original experiment 6 to 8 confederates gathered in an experimental room and were instructed to give false answers in matching a line with the length of three reference lines. An additional uninstructed subject was invited into the experimental room and asked to provide his judgment after the next to last of the confederates. Asch [ 2 ] reports a mean error rate of 36.8% of the 123 real subjects in the critical trials in which the group provided the wrong answer. This result was replicated remarkably consistently. Bond and Smith [ 12 ] conducted a meta study including 44 strict replications, and report an average error rate of 25%. As with the study by Asch [ 2 ], the vast majority of these replications were conducted with male university students in the US. However, more recent studies from Japan [ 13 , 14 ], and Bosnia and Herzegovina [ 15 ] also confirm Asch’s findings. Takano and Sogon [ 14 ] found an error rate of 25% in male Japanese university students (n = 40) in groups with 6 to 9 confederates. Mori and Arai [ 13 ] used the fMORI technique in which participants wear polarized sunglasses allowing the perception of different lines from the same presentation. The method allows to abandon the use of confederates in the Asch judgment task. They replicated the conformity rate for Japanese female subjects (N = 16) but found no conformity for male subjects (N = 10). Usto et al. [ 15 ] found an error rate of 35% in 95 university students of both sexes from Bosnia and Herzegovina with five confederates per group. Other studies also show that subjects are influenced by groups, when the confederates provided their judgments anonymously or with respect to different judgment tasks such as judging the size of circles, completing rows of numbers, or judging the length of acoustic signals [ 8 , 12 , 16 – 21 ]. More recent studies conducted the Asch experiment also with children [ 22 – 25 ] suggesting that the conformity effect can also be found in preschool children. However, some studies also found age effects, such that younger children conformed to the groups majority judgment, but the effect decreases for adolescents [ 26 , 27 ]. To summarize, given the results of the literature, we expect to find a substantial conformity rate in the replication of the original Asch line experiment (H 1 ).

2.1 Monetary incentives

An important extension of the original Asch experiment is the introduction of incentives. In everyday life, decisions are usually associated with consequences. However, in the Asch experiment, as in many other experiments in psychology, decisions or behavior in the lab usually have no consequences, besides of standing out in the laboratory group. This raises questions of the external validity of non-incentivized experiments. Theoretically, it can be expected that correct judgments are less important if they are not incentivized. This could imply that the findings of the Asch experiment are partly methodological artifacts. So far there is only limited and inconclusive empirical evidence with respect to monetary incentives in the Asch experiment. Early studies analysed the role of the perceived societal or scientific importance of the task [ 20 ]. Later research incentivized correct answers in various conformity experiments. Andersson et al. [ 28 ] report that individual incentives decreased the effect of conformity on the prediction of stock prices. However, Bazazi et al. [ 29 ] report the opposite. They found that individualized incentives increase conformity in comparison to collective payoffs in an estimation task. In the study of Baron et al. [ 5 ] 90 participants solved two eyewitness identifications tasks (a line-up task and a task of describing male figures) in the presence of two unanimously incorrectly-answering confederates. Additionally, task importance (low versus high) and task difficulty (low versus high) were experimentally manipulated resulting in a 2 x 2 between subject design. Subjects in the high task importance condition received $20 if ranked in the top 12% of participants with regard to correct answers. Subjects in the low task importance condition received no monetary incentive for correct answers. The results of Baron et al. [ 5 ] show a conformity rate that closely replicates Asch’s [ 1 – 3 ] finding in the condition without monetary incentives. In the condition including a monetary incentive for correct answers, conformity rates drop by about half to an error rate of 15%. However, this result only emerges in the condition with low task difficulty. For the high task difficulty condition, the opposite effect of monetary incentives was observed. Thus, monetary incentives increased conformity when the task was difficult and decreased conformity in situations with easy tasks. However, one drawback of the study of Baron et al. [ 5 ] is a rather low sample size, which might explain the differential effects by experimental condition.

Fujita and Mori [ 7 ] analysed the effect of individual vs collective payoff in the Asch experiment. They found that the conformity effect disappears in the individually incentivized condition. However, also this study suffered from low sample sizes since there were only 10 subjects in the individualized minority incentive condition. Furthermore, Fujita and Mori [ 7 ] used the fMORI method and report that some subjects might have noticed the trick.

Bhanot and Williamson [ 6 ] conducted online experiments (using Amazon Mechanical Turk) in which 391 participants answered 60 multiple-choice trivia-knowledge questions while the most popular answer was displayed at each question. Correct answers were incentivized randomly with $0, $1, $2 or $3 each in a within-subject design, i.e. randomized over trials, not over subjects. Bhanot and Williamson [ 6 ] found that monetary incentives increase the proportion of answers that align with the majority. Hence, the studies using incentives yield inconclusive and contradicting results: Particularly, Baron et al. [ 5 ] found both an accuracy-increasing and accuracy-decreasing effect of monetary incentives depending on task difficulty. Bhanot and Williamson [ 6 ] found an increased conformity rate, and Fujita and Mori [ 7 ] found that the conformity bias disappears in the individually incentivized condition. Overall, we follow the economic notion that monetary incentives matter and expect that rewards for nonconformity decrease group pressure (H 2 ).

2.2 Political opinions

Another critical question is, whether matters of fact can be generalized to matters of attitude and opinion. Crutchfield [ 8 ] investigated experimentally the influence of social pressure on political opinions in an Asch-like situation. He found that agreement with the statement “Free speech being a privilege rather than a right, it is proper for a society to suspend free speech whenever it feels itself threatened” was almost 40 percentage points higher in the social pressure condition (58%, n = 50) than in the individual judgment condition (19%, n = 40). Furthermore, he observed a difference of 36%-points if the confederates answer “subversive activities” to the question "Which one of the following do you feel is the most important problem facing our country today? Economic recession, educational facilities, subversive activities, mental health or crime and corruption” as compared to an individual judgment condition (48% vs 12%). However, the results are based on a rather small number of cases and decisions were anonymous, unlike the original design of Asch.

To the best of our knowledge, there is only one further study that experimentally investigates the influence of social pressure on opinions regarding political issues in an Asch-like situation. In the study of Mallinson and Hatemi [ 9 ] participants (n = 58) were asked to give their opinion on a specific local political issue before and after a 30–45 minutes face-to-face group discussion (treatment condition). In the control condition subjects received written information that contradicts their initial opinion. They found that in the control condition only 8% changed their initial opinion when provided with further information, while in the treatment condition 38% of subjects changed their opinion. Yet, in this recent study the sample size is also rather small. To sum up, given the results of these two studies, we expect that groups exert influence also on political opinions (H 3 ).

2.3 Individual differences

Crutchfield [ 8 ] was also the first who investigated the relationship between personality traits and the susceptibility to the pressure of conformity. He found that low conformity rates were related to high levels of intellectual competence, ego strength, leadership ability, self-control, superiority feelings, adventurousness, self-assertiveness, self-respect, tolerance of ambiguity, and freedom from compulsion regarding rules. High levels of conformity were observed for subjects with authoritarian, anxious, distrustful, and conventional mindsets. However, no substantial correlation was found for neuroticism. Obviously, Crutchfield’s [ 8 ] study is limited by a rather low number of subjects (N = 50). Moreover, the measurement instruments used may be debatable from a contemporary point of view. We are aware of one more recent study with a sufficiently high number of study subjects and more rigid measurement instruments to test the influence of personality traits on conformity in Asch-like situations: Kosloff et al. [ 19 ] analysed the association of the Big Five personality traits (agreeableness, conscientiousness, extraversion, neuroticism, and openness) with conformity in public ratings of the humorousness of unfunny cartoons in 102 female college students. Kosloff et al. [ 19 ] found that subjects with low neuroticism, high agreeableness, and high conscientiousness scores show high levels of conformity. Extraversion, and openness were not associated with conformity ratings. Beyond that, we are not aware of any more studies that investigate the influence of the Big Five personality traits in the original Asch situation. However, there is evidence that openness is linked to nonconformity. Eck and Gebauer [ 30 ] argue that “open people engage in independent thought and, thus, rely little on the conformity heuristic”.

Crutchfield [ 8 ] studied the effect of intellectual competence on conformity. He found that higher competence was associated with lower levels of conformity. However, intelligence was measured by the subjective ratings of the experimental staff. Iscoe, Williams, and Harvey [ 10 ] exposed high school students (7 to 15 years) to group pressure in an acoustic task (counting metronome ticks), and approximated intelligence by subjects’ school records. They found no correlation of school records with conformity. Uchida et al. [ 31 ] studied 12 to 14 year-old high school students and assessed scholastic achievements by their school performance. They report that high achievers conformed less to the majority than low achievers. Hence, results of the effect of intelligence on conformity are inconclusive so far and the existing studies use indirect measures (school grades) but do not measure intelligence directly.

The effect of self-esteem (or self-assertiveness, self-consciousness) on conformity was only investigated in a few studies so far. Kurosawa [ 11 ] found no effect on conformity when the decision of the minority subject was preceded by two confederates. In groups of four, confederates’ self-esteem had a negative effect on conformity. Similarly, Tainaka et al. [ 32 ] found in a sample of Japanese female students that those with low self-esteem conformed more often in a co-witness task.

In addition, the need for social approval may explain individual differences in conformity behavior. The urge to please others by adhering to social norms is expected to be positively related to conformity, simply because conformity is socially approved in many situations and because of a general tendency among humans toward acquiescence. Once more, Crutchfield [ 8 ] provided the first hints of a positive relationship between the need for social approval and conforming behavior in an anonymous Asch situation. However, the measurement instrument he used is debatable. Strickland and Crowne [ 33 ] confirmed Crutchfield’s [ 8 ] finding in a sample of 64 female students exposed to an Asch-like acoustic judgment task using the Crowne-Marlowe (CM) social desirability scale [ 34 , 35 ] to gauge the need for social approval. Again, we are not aware of any other more recent study on this aspect. Hence, we investigate the association of the need for social approval using the CM social desirability scale as well as a more recent and supposedly more appropriate instrument to capture the need for social approval [ 36 ]. Summarizing, we expect to find a positive association between social approval and conformity (H 4 ), and negative associations for intelligence (H 5 ) and self-esteem (H 6 ). With respect to the Big Five we follow Eck and Gebauer [ 30 ] and expect a negative relation between openness and conformity (H 7 ).

Finally, Crutchfield [ 8 ] also analysed the influence of gender on conformity in a sample of 40 female and 19 male college students (study two). He found that young women show higher conformity rates than young men. Yet, in a third study he found that female college alumnae (N = 50) show lower conformity rates than in study one. Hence, Crutchfield’s [ 8 ] findings for the gender effect are inconclusive. However, Bond and Smith [ 12 ] report in their meta-analysis higher conformity rates for females. The study by Griskevicius et al. [ 18 ] shows that gender-differences in conformity depend on the activation of behavioral motives. Men who were primed to attract a mate revealed more independent judgments than women primed to attract a mate, supposedly because of differing mating preferences in men and women. Therefore, we wonder, whether we can replicate the finding that females are more conformative than males in the Asch experiment.

3. Design and method

3.1 procedure and materials.

The experiment consisted of three parts. Part 1 was designed to replicate the original Asch experiment. For this purpose, we recruited 210 subjects on the campus of the University of Bern. Informed consent was obtained verbally before participants entered the experimental room. We randomized subjects into two groups. In group one subjects had to judge the length of lines, as in the original Asch experiment. For this purpose, we placed 5 confederates in addition to a naïve subject in a room. The confederates were asked to behave as naïve subjects and entered the room one after the other. The front row of the seats in the experimental room were numbered such that subjects sat next to each other. The naïve subject was always assigned to seat number 5, leaving the last seat to another confederate. First, we presented some instructions to the subjects: “Welcome to our study on decision-making behavior and opinions. This study consists of two parts: In the first part in this room, we ask you to solve a total of 10 short tasks. In the second part in the room next door, we ask you to complete a short questionnaire on the laptop. In total, this study takes about 40 minutes. As compensation for your participation, you will receive 20 Swiss francs in cash after completing the study.” We then presented a reference line to subjects next to three other lines that were numbered 1 through 3 on projected slides. Subjects were asked to judge the length of the reference line by naming the number of the line that corresponds to the reference line in length. We presented 10 such line tasks (see Fig A1 in the S1 Appendix ). In the first two trials as well as in trials number 4 and 8, confederates pointed out the correct lines. Four trials were easy tasks, since the difference between the reference line and two of the other lines was large. The other six trials were more difficult, since the differences were small. Subjects were asked to call out the number of the correct line always starting with subject 1 through 6.

After the line task in part 2 of the experiment subjects were confronted with 5 general questions on different political issues. The statements were selected because we believe they describe fundamental attitudes towards different political or social groups in a democracy. The five statements read (1) “Do you think that the Swiss Federal Government should be given more power?”, (2) “Do you think trade unions should be given more power in Switzerland?”, (3) “Do you think that the employers’ association in Switzerland should be given more power?”, (4) “Do you think that citizens should be given more liberties in Switzerland?”, and (5) “Do you think that companies in Switzerland should be given more freedom?”. Subjects were asked to answer all 5 questions with either yes or no. The confederates in this group were instructed to answer “yes” to the first question and “no” to the rest. We chose this sequence of “yes” and “no” to prevent that subjects discover the existence of confederates. Finally, part 3 of the experiment consisted of an online questionnaire which subjects were asked to complete. To conceal that some participants were confederates all 6 participants were accompanied to separate rooms where the online-questionnaire was installed on a laptop. The questionnaire was designed to measure a number of different personality traits. Particularly, we measured the Big Five using a 10-item scale (two items for each of the 5 traits) as suggested by Rammstedt et al. [ 37 ] (see Table A1 in the S1 Appendix for item wording); a 10-item scale measuring self-esteem as suggested by Rosenberg [ 38 ] (see Table A2 in the S1 Appendix ); a short version of the Hagen Matrices Test [ 39 ] to measure intelligence and the 10-item version of the Martin Larson Approval Motivation Scale (MLAM) [ 36 ].

In group 2 the experimental design and procedure was the same as in group 1 besides the fact that correct answers in the length of lines judgment task were incentivized. In addition to the 20 Swiss francs show-up fee, subjects received one Swiss franc for every correct answer in the line judgment task, and hence, could earn up to 30 Swiss francs in total. Since there are no correct answers to political opinions these were not incentivized. However, we randomized the confederates’ answers to political opinion questions independently of whether a subject was in the incentivized or non-incentivized group. In one version confederates answered “yes” to the first question and “no” to the four other questions. In the other version the sequence of the confederates’ response was “no” to the first question and “yes” in response to the other four. The experiment was conducted by three different research teams consisting of 7 student assistants each. In every group 5 students acted as confederates and 2 as research assistants, recruiting subjects, welcoming and instructing them in the laboratory room, and reading out loud the projected instructions.

A power analysis suggested that we need about 100 subjects per experimental condition to find statistically significant (α = 0.05) differences of 5 percentage points for a power of 0.8. Hence, we stopped recruiting subjects after reaching 210 participants. The experiment was conducted between March 16, 2021 and April 30, 2021. The authors had no access to any information that links individual identifiers to the data. Subjects were debriefed after the end of the study by email.

Overall, 210 subjects participated in the experiment (female = 61%, mean age = 22.6). 102 subjects were randomly assigned to the non-incentivized group and 108 into the group with incentives. Moreover, 113 subjects were assigned to the sequences of “yes” and four “no” of the political opinion task and 97 to the reversed sequence, suggesting that the randomization procedure worked well. The questionnaire also contained an attention check. The question reads “In the following we show you five answer categories. Please do not tick any of the answers”. Four subjects failed to comply and ticked an answer, suggesting that they did not pay proper attention to the question wording. These subjects were excluded from the analysis. Furthermore, we asked subjects at the end of the questionnaire what they think the experiment was about. Three subjects recognized that the experiment was the line task experiment of Asch or expressed the suspicion that some of the other group members were confederates. We also excluded these three subjects from the analysis. Moreover, one subject answered the question about their gender with “other” and was also excluded from the analysis. Hence, these exclusions result in 202 valid cases. However, the results presented do not depend on these eight excluded observations.

Fig 1 presents the results of the ten line length tasks for the non-incentivized (grey bars) and for the incentivized conditions (blue bars). As can be clearly seen, almost none of the naïve subjects gave an incorrect answer when the group provided the correct answer which was the case in decision situations 1, 2, 4 and 8. However, when the group provides the false answer a substantial number of naïve subjects provided this incorrect answer as well (decision situations 3, 5, 6, 7, 9, and 10). The proportion of incorrect answers in the non-incentivized condition is relatively small in decision 3 (10%), but relatively high in decisions number 6 and 7 (44% and 47%). The average of incorrect answers is 33% in the non-incentivized group, which is a perfect replication of Asch’s (1955) original 36.8% result (two sample two-sided T-test, t(16) = 0.59, p = 0.57).

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

Note: Percent of correct answers by experimental group and trial including 95% confidence intervals. The numbers on top of the bars denote the trial numbers. “correct” stands for uncritical trials, “false” for critical trials. “easy” denotes easy trials with big differences between the lines, and “hard” denotes more difficult trials with smaller differences between the lines. The numbers between the bars denote the difference in proportions between the groups in percentage points. One-sided T-tests: * = p < 0.05. N without incentive (no) = 99, n with incentive (yes) = 103.

https://doi.org/10.1371/journal.pone.0294325.g001

When correct answers are incentivized, the proportion of incorrect answers decreases by on average 8%-points. The difference between the groups is statistically significant in 2 out of 6 critical trials (p < 0.05 for one-sided T-tests). The difference also becomes evident when we consider the number of incorrect answers in the 6 critical trials. When decisions were not incentivized subjects gave on average 1.97 incorrect answers. In the incentivized condition the average number dropped to 1.47, leading to a statistically significant difference of 0.5 incorrect answers (t(208) = 2.24, p = 0.03 for two-sided T-test).

Next, Fig 2 presents the results concerning the five political questions. When the group said “yes” to the question of whether the Swiss Federal Council (the government in Switzerland) should have more power, 27% of the naïve subjects did so as well. When the group said “no” only 3% of the subjects said “yes” resulting in a difference of 23.4%-points. When the group said that trade unions should have more power 72% of the subjects answered “yes” as compared to only 29% when the group said “no” resulting in a difference of 43%-points. Similarly, the question of whether the employers’ association should have more power is agreed to by 44% and 6% respectively, depending on the group agreeing or disagreeing. Moreover, 81% of the subjects agreed that citizens in Switzerland should be given more liberties when the group does so, and 33% agreed to this question when the group says “no”. Finally, 46% said that companies should be given more freedom when the group agreed but only 8% did so when the group denied this question. The average difference in the proportion of yes-answers is 38%-points and all 5 differences are statistically highly significant. This result corresponds astonishingly close to the result in the length of line experiment and suggests that the influence of group pressure can be generalized to the utterance of political opinions.

Note: Percent of ‘yes’ answers to five general questions on political opinions in which all confederates answered uniformly ‘yes’ or ‘no’, by experimental group including 95% confidence intervals. The numbers on top of the bars stand for the difference in proportions between the respective groups in percentage points. Two-sided T-tests: *** = p < 0.001. n (sequence yes, no, no, no, no) = 109, n (sequence no, yes, yes, yes, yes) = 93.

https://doi.org/10.1371/journal.pone.0294325.g002

One interesting question is whether the susceptibility to group pressure is linked to certain personality traits. To investigate this question, we count the number of wrong answers in the six critical trials of the length of line task. This variable is our dependent variable and runs from 0 when a subject always gave correct answers to 6 for subjects who gave only wrong answers. First, we wondered whether conformity is linked to the Big Five personality traits. We measured the Big Five using a short 10-item version as suggested by Rammstedt et al. [ 37 ] which measures each trait (openness, extraversion, agreeableness, conscientiousness, and neuroticism) with two questions (see Table A1 in the S1 Appendix ).

Second, we incorporate a 10-item measure of self-esteem, as suggested by Rosenberg [ 38 ], into the analysis (see Table A2 in the S1 Appendix ). Each item of the scale has four answer categories ranging from 1 = “disagree strongly”, 2 = “disagree”, 3 = “agree” to 4 = “agree strongly”. Subjects that score high on self-esteem are expected to have stronger confidence in their own perception and should be less influenced by the group’s opinion. Third, we measured individuals’ intelligence using a short version of the Hagen Matrices Test (HMT) [ 39 ]. The HMT consists of six 9-field matrices that show graphical symbols that follow a logical order. The last field is missing and the task of the subjects is to pick the correct symbol, out of eight, that fits and completes the pattern of the matrix. Hence, the HMT ranges from 0 if no answer is correct to 6 for subjects who provided six correct answers. The hypothesis is that subjects who score high on the HMT are less susceptible to the pressure of the group and also provide more correct answers in the line task.

Finally, conformity might be linked to the need for social approval. We measured the need for social approval with a 10-item version of the Martin Larson Approval Motivation Scale (MLAM) [ 36 ] (see Table A3 in the S1 Appendix ). Individuals that score highly on the MLAM display high need for social approval by others. Hence, we expect that subjects with higher values on the MLAM should also conform more often to the opinions of others in order to receive social approval. A summary of the descriptive information of the considered variables is depicted in Table A4 in the S1 Appendix . To investigate whether any of the measured personality traits are linked to the answering behavior in the line task we conducted multiple OLS regression analysis. The results of this analysis are depicted in the coefficient plots in Fig 3 (see also Table A5 in the S1 Appendix ).

Note: N = 202. Unstandardized coefficients of multiple linear OLS regressions including robust 95% confidence intervals. Poisson and negative binomial models do not alter the results in any substantial way. Variables marked with an ‘*’ indicate statistically significant differences in the coefficients between models (2) and (3).

https://doi.org/10.1371/journal.pone.0294325.g003

First, model 1 presents the effects on the number of conforming answers for the whole sample. In the incentivized condition subjects gave on average 0.43 fewer conforming answers as compared to the unincentivized condition. This effect mirrors the bivariate result already presented in Fig 1 and is statistically significant for the 5% level. In tendency, females show more conforming answers, but this effect is statistically only significant for the 10% level. Besides “openness” none of the personality traits contained in the Big Five show any statistically significant effects. This is also true for the other effects of intelligence, self-esteem, and the measure for social approval seeking. Models 2 and 3 show the results for men and women separately. The separate results suggest that women react somewhat more strongly to incentives than do men. However, a test for differences in coefficients suggests that the effects do not differ (χ 2 (1) = 1.11, p = 0.29). Intelligence seems to have greater importance for men, leading to 0.27 fewer conforming answers for every correct answer of the HMT. However, this effect does not differ statistically from the effect for females (χ 2 (1) = 1.35, p = 0.25). No difference in effects can be observed for self-esteem. However, in the female sample the need for social approval is positively linked to the number of conforming answers, which is not the case in the male sample (χ 2 (1) = 4.23, p = 0.04); but the effect of social approval in the female sample is relatively small.

We conducted a number of robustness checks with the presented analyses. Since our dependent variable is a count variable (number of conforming answers) the models can also be estimated using Poisson regressions or negative-binomial models. However, none of our presented results change in any substantial way using these alternatives. Furthermore, we excluded 24 more subjects who when asked at the end of the experiment about the goal of the study said that the experiment was about group pressure or conformity, although they did not explicitly mention Asch or the suspicion that other participants were confederates. But these additional exclusions also did not change the results substantially (see Table A6 in the S1 Appendix ). Finally, we also incorporated the 10-item version of the Marlowe-Crowne Scale [ 34 , 35 ] suggested by Clancy [ 40 , 41 ]. However, inclusion of the scale did not show any statistically significant effects or did change any of the other estimates.

5. Conclusion and discussion

In this study we first replicated the original experiment of Asch [ 1 – 3 ] with 5 confederates and ten line tasks. We find an average error rate of 33% which replicates the original findings of Asch very closely and which is in line with other replications that were conducted predominately with American students [ 12 ]. Together with recent studies from Japan [ 14 ], and Bosnia and Herzegowina [ 15 ], our study provides further evidence that the influence of groups on individuals’ judgments is a universal phenomenon, and is still valid today. Furthermore, we incentivized the decisions and find a drop of the error rate by 8%-points to 25%. Hence, monetary incentives do not eliminate the effect of group pressure. This finding sheds doubt on former results which predominately show the opposite effect, namely that incentives increase compliance [ 5 , 6 ].

Moreover, our study suggests that group pressure is not only influential in the simple line task but also when it comes to political opinions. We randomized the groups’ response to five different political statements and find an average conformity rate of 38%. Hence, these results suggest that the original finding of Asch can also be generalized to matters of opinion. This result is in line with former evidence by Crutchfield [ 8 ], and Mallinson and Hatemi [ 9 ]. However, both of these studies had only small sample sizes of 50 and 58 subjects respectively, which called for further replication studies. Finally, we measured the Big Five, intelligence, self-esteem and social approval. With the exception of openness, our study finds no support that these personality traits are statistically significantly related to the susceptibility of group pressure.

Of course, our study has some limitations, which suggest a number of further research questions. First, we used a relatively large sample of 202 subjects providing more statistical power than former replications and extensions of the Asch experiment; however, our subjects were also students, and hence, it would be important to have further replications with non-student samples. This would allow further investigations of the susceptibility to group pressure with respect to age, different occupational groups, different social backgrounds, and different levels of social experience.

Second, the subjects we investigate are strangers. That means the single naïve subjects did not know the confederates. An interesting question for further research would be, whether group pressure is stronger among non-strangers or whether dissent becomes more acceptable among a group of friends.

Third, we demonstrate that monetary incentives reduce the error rate. However, our incentives were one Swiss franc for every correct answer, and hence small. Thus, the interesting question remains whether larger incentives reduce the error rate further, or can even lead to the elimination of it.

Fourth, the political statements we choose are relatively moderate and general. This leaves the question open as to whether subjects would also conform to more extreme or socially less acceptable statements. Furthermore, our subjects might have rarely thought about the statements we provided, leaving the question of what would happen with respect to statements about which subjects had stronger opinions or which are more related to their identity.

With the exception of openness all personality traits considered (e.g. intelligence, self-esteem, need for social approval) are not related to conformity. This raises a number of very interesting research questions. One possibility is that the traits were not measured good enough, and that measurement errors impede the identification of these individual differences. This concern applies particularly to the measurement of the Big Five where we relied on the short 10-item version suggested by Rammstedt et al. [ 37 ]. Hence, the puzzling result that openness leads to less conformity must be replicated before it can count as a reliable finding. However, the finding is in line with the assumption of Eck and Gebauer [ 30 ]. Another possibility is that other personality traits are more important when it comes to conformity behavior. Hence, there is much room for further interesting research concerning conformity behavior in situations of group pressure.

Supporting information

S1 appendix..

https://doi.org/10.1371/journal.pone.0294325.s001

Acknowledgments

We like to thank our student assistants for helping us with the data collection. Their names are: Yvonne Aregger, Elias Balmer, Ambar Conca, Davide Della Porta, Shania Flück, Julian Gerber, Anna Graf, Ina Gutjahr, Kim Gvozdic, Anna Häberli, Chiara Heiss, Paula Kühne, Jenny Mosimann, Remo Parisi, Elena Raich, Virginia Reinhard, Fiona Schläppi, Maria Tournas, Angela Ventrici, Marco Zbinden, Sarah Zwyssig.

• 1. Asch, Solomon E. (1951): Effects of group pressure upon the modification and distortion of judgments. In: Guetzkow, H. (Ed.) Groups, leadership and men; research in human relations, p. 177–190. Carnegie Press.
• View Article
• Google Scholar
• PubMed/NCBI
• 35. Crowne Douglas P., and Marlowe David (1964): The Approval Motive. Studies in Evaluative Dependence. New York: John Wiley & Sons.
• 37. Rammstedt, Beatrice, Christoph J. Kemper, Mira Céline Klein, Constanze Beierlein, and Anastassyia Kovaleva (2014): Big Five Inventory 10 (BFI-10).
• 38. Rosenberg Morris (1965): Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

The Asch Line Study (+3 Conformity Experiments)

Sheep. Complicit. Doormat. There are a lot of negative connotations associated with conformity, especially in the United States. Individualist societies push back on "going along with what everyone is doing." And yet, we conform more than we think. According to studies like the Asch Line Study, humans have a natural tendency to conform.

The Asch Line Study is one of the most well-known experiments in modern psychology, but it's not without its faults. Keep reading to learn about how the Asch Line Study worked, its criticisms, and similar experiments!

How Did the Asch Line Study Work?

In his famous “Line Experiment”, Asch showed his subjects a picture of a vertical line followed by three lines of different lengths, one of which was obviously the same length as the first one. He then asked subjects to identify which line was the same length as the first line.

Solomon Asch used 123 male college students as his subjects, and told them that his experiment was simply a ‘vision test’. For his control group, Asch just had his subjects go through his 18 questions on their own.

However, for his experimental group, he had his subjects answer each of the same 18 questions in a group of around a dozen people, where the first 11 people intentionally said obviously incorrect answers one after another, with the final respondee being the actual subject of the experiment.

Who was Solomon Asch?

Solomon E. Asch was a pioneer in social psychology. He was born in Poland in 1907 and moved to the United States in 1920. Asch received his Ph.D. from Columbia University in 1932 and went on to perform some famous psychological experiments about conformity in the 1950s.

One of these studies is known as the “Asch Line Experiment”, where he found evidence supporting the idea that humans will conform to and accept the ideas of others around them, even if those ideas are obviously false. This study is one of the most influential studies in social psychology .

Findings of Asch's Conformity Study

​ Asch found that his subjects indeed were more likely to give a false response after the other members of their group (the actors) gave false responses. As shown in this ‘Table 1’ from his experiment, during 18 trials, the ‘Majority Error’ column shows no error when the group response was the correct response, such as in Trial #1.

However, when the entire group intentionally gave a false answer (these situations are designated with an * under the “Group Response” column), the ‘Majority Error’ did exist and was slanted toward the opinion of the group.

For example, if the group answered with a line that was too long, such as in Trial #3, the ‘Majority Error’ column shows that the subjects generally estimated the line to be longer than it really was (denoted with a ‘+’), and vice-versa for when the group answered with a line that was too short, such as in Trial #4.

As for his control group, Asch found that people generally said the correct answer when they did not have a group of actors saying answers before them.

Interviewing the Participants

​ After the experiment, Asch revealed the true experiment to his subjects and interviewed them. Some subjects had become very agitated during the experiment, wondering why they kept disagreeing with the group. When the group pressed one particular subject on why he thought that he was correct and the entire group was wrong, he replied defiantly, exclaiming: “You're probably right, but you may be wrong!”

Other subjects admitted during the interview that they changed their answers after hearing others in their group reply differently. One was recorded saying, “If I’d been the first I probably would have responded differently.” Another subject admitted, “...at times I had the feeling: 'to heck with it, I'll go along with the rest.' "

Conclusions from the Asch Line Study

​ Asch found that his subjects often changed their answers when they heard the rest of the group unanimously giving a different response.

After the interviews, Asch concluded in his study that his subjects conformed to the opinions of the group for three different reasons:

Distortion of perception due to the stress of group pressure: This group of subjects always agreed with the group and said during the interview that they wholeheartedly believed that their obviously incorrect answers were correct. Asch concluded that the stress of group pressure had distorted their perception.

Distortion of judgment: This was the most common outcome, where subjects assumed that their individual answers were incorrect after seeing the rest of the group answer differently, so they changed their answer to align with the group.

Distortion of action: These subjects never doubted that they were correct and the group was wrong, but out of fear of being perceived as different, they suppressed their opinions and intentionally lied when it was their turn to give an answer.

Asch Line Study vs. Milgram Experiment

Both the Asch Line Study and the Milgram Experiment look at conformity, obedience, and the negative effects of going along with the majority opinion. Those negative effects are slightly awkward, like in the Asch Line Study, or dangerous, like in the Milgram Experiment. Both experiments were conducted in the Post-WWII world as a response to the conformity that was required for Nazi Germany to gain power. The premise of Asch's study was not nearly as dramatic. Milgram's was. 

To test conformity, Milgram and his researchers instructed participants to press a button. Participants believed that the buttons would shock another "participant" in a chair, who was really an actor. (No one was shocked.) The study continued as long as participants continued to shock the participant at increasingly dangerous levels. The participants knew that they could cause serious harm to the person in the chair. Yet, many obeyed.

Further Experiments and Variations

Solomon Asch didn't just conduct one experiment and move on. He replicated his experiment with new factors, including:

• Changing the size of the actor group
• Switching to a non-unanimous actor group
• Having a unanimous actor group, except for one actor who sticks to the correct response no matter what the group or subject says
• Instructing the one actor who gives the correct response come in late
• Having one actor decide to change their answer from the group’s answer to the subject’s answer

There are also many reproductions and replications of this study online. Not all of them come to the same conclusions! Read through the following texts to get a sense of how other psychologists approached this subject:

• Mori K, Arai M. No need to fake it: reproduction of the Asch experiment without confederates. Int J Psychol. 2010 Oct 1;45(5):390-7. doi: 10.1080/00207591003774485. PMID: 22044061.

Why Is The Asch Line Study Ethnocentric? And Other Criticisms

​ One big issue with the Asch line study is that the subjects were all white male college students between the ages of 17 and 25, with a mean age of 20. Since the experiment only shows results for this small and specific group of people, it alone cannot be applied to other groups such as women or older men.

Experimenter Bias in the Asch Line Study

Only choosing subjects from one demographic is a form of Experimenter Bias . Of course, researchers can use one demographic if they are specifically studying that demographic. But Asch was not just looking at young, white men. If he had expanded his research to include more participants, he may have produced different answers.

We assume Asch did not go about his study with the intention of being biased. That's the tricky thing about biases. They sneak up on us! Even the way that we share information about psychology research is the result of bias. Reporter bias is the tendency to highlight certain studies due to their results. The Asch Line Study produced fascinating results. Therefore, psychology professors, reporters, and students find it fascinating and continue to share this concept. They don't always share the full story, though.

Did you know that 95% of the participants actually defied the majority at least once during the experiment? Most textbooks don't report that. Nor did they report that the interviewees knew that they were right all along! Leaving out this key information is not Asch's fault. But it should give you, a psychology student, some pause. One thing that we should take away from this study is that we have a natural tendency to conform. This tendency also takes place when we draw conclusions from famous studies! Be critical as you learn about these famous studies and look to the source if possible.

Further Criticisms of the Asch Line Study

Does the Asch Line Study stand the test of time? Not exactly. If we look at what was happening in 1950s society, we can see why Asch got his results. Young white men in the early 1950s may have responded differently to this experiment than young white men would today. In the United States, which is where this experiment was performed, the mid-1950s was a historic turning point in terms of rejecting conformity. Youth were pushing toward a more free-thinking society. This experiment was performed right around the time that the movement was just starting to blossom, so the subjects had not grown up in the middle of this new anti-conformist movement. Had Asch performed this experiment a decade later with youth who more highly valued free-thinking, he may have come across very different results.

Another thing to note is that, at least in the United States, education has evolved with this movement of encouraging free-thinking. Teachers today tell students to question everything, and many schools reject ideas of conformity. This could once again mean that, if done again today, Asch would have found very different results with this experiment.

Another problem with this experiment is that, since subjects were not told it was a psychological experiment until after it was over, subjects may have gone through emotional and psychological pain during what they thought was just a simple ‘vision test’.

Finally, it’s good to remember that the ‘Asch Line Experiment’ is just that: an experiment where people looked at lines. This can be hard to apply to other situations because humans in group settings are rarely faced with questions that have one such obvious and clear answer, as was the case in this experiment.

Related posts:

• Solomon Asch (Psychologist Biography)
• 40+ Famous Psychologists (Images + Biographies)
• Stanley Milgram (Psychologist Biography)
• Experimenter Bias (Definition + Examples)
• The Monster Study (Summary, Results, and Ethical Issues)

Reference this article:

About The Author

PracticalPie.com is a participant in the Amazon Associates Program. As an Amazon Associate we earn from qualifying purchases.

Follow Us On:

Youtube Facebook Instagram X/Twitter

Psychology Resources

Developmental

Personality

Relationships

Psychologists

Serial Killers

Psychology Tests

Personality Quiz

Memory Test

Depression test

Type A/B Personality Test

© PracticalPsychology. All rights reserved

Privacy Policy | Terms of Use

The power of social influence: A replication and extension of the Asch experiment

Axel franzen.

Institute of Sociology, University of Bern, Bern, Switzerland

Sebastian Mader

Associated data.

The data used in this study is publicly available in the repository of the University of Bern at https://boris.unibe.ch/id/eprint/169645 .

In this paper, we pursue four goals: First, we replicate the original Asch experiment with five confederates and one naïve subject in each group (N = 210). Second, in a randomized trial we incentivize the decisions in the line experiment and demonstrate that monetary incentives lower the error rate, but that social influence is still at work. Third, we confront subjects with different political statements and show that the power of social influence can be generalized to matters of political opinion. Finally, we investigate whether intelligence, self-esteem, the need for social approval, and the Big Five are related to the susceptibility to provide conforming answers. We find an error rate of 33% for the standard length-of-line experiment which replicates the original findings by Asch (1951, 1955, 1956). Furthermore, in the incentivized condition the error rate decreases to 25%. For political opinions we find a conformity rate of 38%. However, besides openness, none of the investigated personality traits are convincingly related to the susceptibility of group pressure.

1. Introduction

A core assumption in sociology is that what humans think and do does not only depend on their own attitudes and disposition, but also to a large extent on what others think and do. The power of social influence on individuals’ behavior was demonstrated already in the 1950s in a series of experiments by Solomon Asch [ 1 – 3 ]. Asch invited individuals into the lab and assigned them the task of judging the length of a line. He also placed 6 confederates into the lab who were assigned to give wrong answers publicly, so that the naïve subject could hear them before he provided his own answer. The results were very surprising: on average 35% of the real subjects followed the opinions of the confederates even if their answer was obviously wrong. The work of Asch has attracted a great amount of attention in the social sciences. Hence, a multitude of replications, extensions, and variations of the original studies have been conducted. However, many of these replications were done with student samples in the US, and fewer studies consist of samples from other countries. Furthermore, many replications were undertaken in the 40 years following the original experiment of Asch, but there are fewer replications thereafter. This raises two important questions: First, are the findings of Asch universal or do they predominantly apply to American students? And second, are the findings still valid today or has the influence of others diminished over time, for instance through increased education and democratization?

Moreover, many experiments in psychology are not incentivized by monetary rewards. This is also true for Asch’s original experiments and for most replications of it. However, in real life outside the lab, decisions are usually associated with consequences, either pleasant in the form of rewards, or unpleasant in the form of some kind of punishment. To make the study of decision-making more realistic, experiments in economics usually use monetary incentives [ 4 ]. To provide a conforming but wrong judgment in the original Asch experiment has no consequences, giving rise to the interesting question of whether the finding of Asch still holds when correct answers are rewarded. So far, the effect of incentives in the Asch decision situation has only been investigated rarely [ 5 – 7 ], with inconclusive evidence. Baron et al. [ 5 ] report that use of monetary incentives actually increased conformity when the task was difficult. A decreased conformity rate was only found in situations with easy tasks. Bhanot & Williamson [ 6 ] conducted two online experiments and found that incentivizing correct answers increases the number of conforming answers. Fujita and Mori [ 7 ] compared group reward and individual rewards in the Asch experiment and found that conformity vanished in the individual reward condition. Thus, the existing evidence on the role of incentives is inconclusive, calling for further investigations of the effect of incentives.

Of course, misjudging the length of lines when others do is not important in itself; the Asch experiment created so much attention because it elicits the suspicion that social influence is also present in other and more important social realms, for instance when it comes to political opinions. Early research by Crutchfield [ 8 ] suggests that the original findings on line judgment also transfer over to political opinions. We are only aware of one further study by Mallinson and Hatemi [ 9 ] that investigates the effect of social influence on opinion formation. However, the authors used a group discussion in the treatment condition, and hence diverged somewhat from the original Asch design. Furthermore, investigations of the effect of social conformity on political opinions are always idiosyncratic making further replications on the transferability from lines to a variety of political opinions important and interesting.

Moreover, behaving in a conforming way and misjudging tasks raises a number of interesting questions. About one third of Asch’s subjects was susceptible to social pressure on average. The rest solved the task correctly irrespective of the confederates’ opinion most of the time. How do those who are not influenced by the group differ from the ones that behave in a conformative manner? Crutchfield [ 8 ] investigated a number of personality traits such as competence, self-assertiveness, or leadership ability on the susceptibility to the pressure to conform to the groups’ judgment. However, many of the measurement instruments used by him or by others [ 10 , 11 ] investigating similar questions are suboptimal, and furthermore produced inconclusive results. Hence, it is worthwhile to further investigate the characteristics of those who conform to social pressure and of those who resist it. We are particularly interested in the Big Five, intelligence, self-esteem, and the need for social approval.

The remainder of the article proceeds in four sections. First, in section two, we present an elaborate literature review of the original Asch experiment, and its various replications. Section three describes how we conducted the replication of the Asch experiment and its variant by using political opinions. Furthermore, we describe how we implemented the incentives, and how we measure the various traits that are presumably related to behavior in the Asch experiment. Section four presents the results and section five concludes and discusses ideas for further research.

2. Literature review

In Asch’s [ 2 ] original experiment 6 to 8 confederates gathered in an experimental room and were instructed to give false answers in matching a line with the length of three reference lines. An additional uninstructed subject was invited into the experimental room and asked to provide his judgment after the next to last of the confederates. Asch [ 2 ] reports a mean error rate of 36.8% of the 123 real subjects in the critical trials in which the group provided the wrong answer. This result was replicated remarkably consistently. Bond and Smith [ 12 ] conducted a meta study including 44 strict replications, and report an average error rate of 25%. As with the study by Asch [ 2 ], the vast majority of these replications were conducted with male university students in the US. However, more recent studies from Japan [ 13 , 14 ], and Bosnia and Herzegovina [ 15 ] also confirm Asch’s findings. Takano and Sogon [ 14 ] found an error rate of 25% in male Japanese university students (n = 40) in groups with 6 to 9 confederates. Mori and Arai [ 13 ] used the fMORI technique in which participants wear polarized sunglasses allowing the perception of different lines from the same presentation. The method allows to abandon the use of confederates in the Asch judgment task. They replicated the conformity rate for Japanese female subjects (N = 16) but found no conformity for male subjects (N = 10). Usto et al. [ 15 ] found an error rate of 35% in 95 university students of both sexes from Bosnia and Herzegovina with five confederates per group. Other studies also show that subjects are influenced by groups, when the confederates provided their judgments anonymously or with respect to different judgment tasks such as judging the size of circles, completing rows of numbers, or judging the length of acoustic signals [ 8 , 12 , 16 – 21 ]. More recent studies conducted the Asch experiment also with children [ 22 – 25 ] suggesting that the conformity effect can also be found in preschool children. However, some studies also found age effects, such that younger children conformed to the groups majority judgment, but the effect decreases for adolescents [ 26 , 27 ]. To summarize, given the results of the literature, we expect to find a substantial conformity rate in the replication of the original Asch line experiment (H 1 ).

2.1 Monetary incentives

An important extension of the original Asch experiment is the introduction of incentives. In everyday life, decisions are usually associated with consequences. However, in the Asch experiment, as in many other experiments in psychology, decisions or behavior in the lab usually have no consequences, besides of standing out in the laboratory group. This raises questions of the external validity of non-incentivized experiments. Theoretically, it can be expected that correct judgments are less important if they are not incentivized. This could imply that the findings of the Asch experiment are partly methodological artifacts. So far there is only limited and inconclusive empirical evidence with respect to monetary incentives in the Asch experiment. Early studies analysed the role of the perceived societal or scientific importance of the task [ 20 ]. Later research incentivized correct answers in various conformity experiments. Andersson et al. [ 28 ] report that individual incentives decreased the effect of conformity on the prediction of stock prices. However, Bazazi et al. [ 29 ] report the opposite. They found that individualized incentives increase conformity in comparison to collective payoffs in an estimation task. In the study of Baron et al. [ 5 ] 90 participants solved two eyewitness identifications tasks (a line-up task and a task of describing male figures) in the presence of two unanimously incorrectly-answering confederates. Additionally, task importance (low versus high) and task difficulty (low versus high) were experimentally manipulated resulting in a 2 x 2 between subject design. Subjects in the high task importance condition received $20 if ranked in the top 12% of participants with regard to correct answers. Subjects in the low task importance condition received no monetary incentive for correct answers. The results of Baron et al. [ 5 ] show a conformity rate that closely replicates Asch’s [ 1 – 3 ] finding in the condition without monetary incentives. In the condition including a monetary incentive for correct answers, conformity rates drop by about half to an error rate of 15%. However, this result only emerges in the condition with low task difficulty. For the high task difficulty condition, the opposite effect of monetary incentives was observed. Thus, monetary incentives increased conformity when the task was difficult and decreased conformity in situations with easy tasks. However, one drawback of the study of Baron et al. [ 5 ] is a rather low sample size, which might explain the differential effects by experimental condition.

Fujita and Mori [ 7 ] analysed the effect of individual vs collective payoff in the Asch experiment. They found that the conformity effect disappears in the individually incentivized condition. However, also this study suffered from low sample sizes since there were only 10 subjects in the individualized minority incentive condition. Furthermore, Fujita and Mori [ 7 ] used the fMORI method and report that some subjects might have noticed the trick.

Bhanot and Williamson [ 6 ] conducted online experiments (using Amazon Mechanical Turk) in which 391 participants answered 60 multiple-choice trivia-knowledge questions while the most popular answer was displayed at each question. Correct answers were incentivized randomly with $0, $1, $2 or $3 each in a within-subject design, i.e. randomized over trials, not over subjects. Bhanot and Williamson [ 6 ] found that monetary incentives increase the proportion of answers that align with the majority. Hence, the studies using incentives yield inconclusive and contradicting results: Particularly, Baron et al. [ 5 ] found both an accuracy-increasing and accuracy-decreasing effect of monetary incentives depending on task difficulty. Bhanot and Williamson [ 6 ] found an increased conformity rate, and Fujita and Mori [ 7 ] found that the conformity bias disappears in the individually incentivized condition. Overall, we follow the economic notion that monetary incentives matter and expect that rewards for nonconformity decrease group pressure (H 2 ).

2.2 Political opinions

Another critical question is, whether matters of fact can be generalized to matters of attitude and opinion. Crutchfield [ 8 ] investigated experimentally the influence of social pressure on political opinions in an Asch-like situation. He found that agreement with the statement “Free speech being a privilege rather than a right, it is proper for a society to suspend free speech whenever it feels itself threatened” was almost 40 percentage points higher in the social pressure condition (58%, n = 50) than in the individual judgment condition (19%, n = 40). Furthermore, he observed a difference of 36%-points if the confederates answer “subversive activities” to the question "Which one of the following do you feel is the most important problem facing our country today? Economic recession, educational facilities, subversive activities, mental health or crime and corruption” as compared to an individual judgment condition (48% vs 12%). However, the results are based on a rather small number of cases and decisions were anonymous, unlike the original design of Asch.

To the best of our knowledge, there is only one further study that experimentally investigates the influence of social pressure on opinions regarding political issues in an Asch-like situation. In the study of Mallinson and Hatemi [ 9 ] participants (n = 58) were asked to give their opinion on a specific local political issue before and after a 30–45 minutes face-to-face group discussion (treatment condition). In the control condition subjects received written information that contradicts their initial opinion. They found that in the control condition only 8% changed their initial opinion when provided with further information, while in the treatment condition 38% of subjects changed their opinion. Yet, in this recent study the sample size is also rather small. To sum up, given the results of these two studies, we expect that groups exert influence also on political opinions (H 3 ).

2.3 Individual differences

Crutchfield [ 8 ] was also the first who investigated the relationship between personality traits and the susceptibility to the pressure of conformity. He found that low conformity rates were related to high levels of intellectual competence, ego strength, leadership ability, self-control, superiority feelings, adventurousness, self-assertiveness, self-respect, tolerance of ambiguity, and freedom from compulsion regarding rules. High levels of conformity were observed for subjects with authoritarian, anxious, distrustful, and conventional mindsets. However, no substantial correlation was found for neuroticism. Obviously, Crutchfield’s [ 8 ] study is limited by a rather low number of subjects (N = 50). Moreover, the measurement instruments used may be debatable from a contemporary point of view. We are aware of one more recent study with a sufficiently high number of study subjects and more rigid measurement instruments to test the influence of personality traits on conformity in Asch-like situations: Kosloff et al. [ 19 ] analysed the association of the Big Five personality traits (agreeableness, conscientiousness, extraversion, neuroticism, and openness) with conformity in public ratings of the humorousness of unfunny cartoons in 102 female college students. Kosloff et al. [ 19 ] found that subjects with low neuroticism, high agreeableness, and high conscientiousness scores show high levels of conformity. Extraversion, and openness were not associated with conformity ratings. Beyond that, we are not aware of any more studies that investigate the influence of the Big Five personality traits in the original Asch situation. However, there is evidence that openness is linked to nonconformity. Eck and Gebauer [ 30 ] argue that “open people engage in independent thought and, thus, rely little on the conformity heuristic”.

Crutchfield [ 8 ] studied the effect of intellectual competence on conformity. He found that higher competence was associated with lower levels of conformity. However, intelligence was measured by the subjective ratings of the experimental staff. Iscoe, Williams, and Harvey [ 10 ] exposed high school students (7 to 15 years) to group pressure in an acoustic task (counting metronome ticks), and approximated intelligence by subjects’ school records. They found no correlation of school records with conformity. Uchida et al. [ 31 ] studied 12 to 14 year-old high school students and assessed scholastic achievements by their school performance. They report that high achievers conformed less to the majority than low achievers. Hence, results of the effect of intelligence on conformity are inconclusive so far and the existing studies use indirect measures (school grades) but do not measure intelligence directly.

The effect of self-esteem (or self-assertiveness, self-consciousness) on conformity was only investigated in a few studies so far. Kurosawa [ 11 ] found no effect on conformity when the decision of the minority subject was preceded by two confederates. In groups of four, confederates’ self-esteem had a negative effect on conformity. Similarly, Tainaka et al. [ 32 ] found in a sample of Japanese female students that those with low self-esteem conformed more often in a co-witness task.

In addition, the need for social approval may explain individual differences in conformity behavior. The urge to please others by adhering to social norms is expected to be positively related to conformity, simply because conformity is socially approved in many situations and because of a general tendency among humans toward acquiescence. Once more, Crutchfield [ 8 ] provided the first hints of a positive relationship between the need for social approval and conforming behavior in an anonymous Asch situation. However, the measurement instrument he used is debatable. Strickland and Crowne [ 33 ] confirmed Crutchfield’s [ 8 ] finding in a sample of 64 female students exposed to an Asch-like acoustic judgment task using the Crowne-Marlowe (CM) social desirability scale [ 34 , 35 ] to gauge the need for social approval. Again, we are not aware of any other more recent study on this aspect. Hence, we investigate the association of the need for social approval using the CM social desirability scale as well as a more recent and supposedly more appropriate instrument to capture the need for social approval [ 36 ]. Summarizing, we expect to find a positive association between social approval and conformity (H 4 ), and negative associations for intelligence (H 5 ) and self-esteem (H 6 ). With respect to the Big Five we follow Eck and Gebauer [ 30 ] and expect a negative relation between openness and conformity (H 7 ).

Finally, Crutchfield [ 8 ] also analysed the influence of gender on conformity in a sample of 40 female and 19 male college students (study two). He found that young women show higher conformity rates than young men. Yet, in a third study he found that female college alumnae (N = 50) show lower conformity rates than in study one. Hence, Crutchfield’s [ 8 ] findings for the gender effect are inconclusive. However, Bond and Smith [ 12 ] report in their meta-analysis higher conformity rates for females. The study by Griskevicius et al. [ 18 ] shows that gender-differences in conformity depend on the activation of behavioral motives. Men who were primed to attract a mate revealed more independent judgments than women primed to attract a mate, supposedly because of differing mating preferences in men and women. Therefore, we wonder, whether we can replicate the finding that females are more conformative than males in the Asch experiment.

3. Design and method

3.1 procedure and materials.

The experiment consisted of three parts. Part 1 was designed to replicate the original Asch experiment. For this purpose, we recruited 210 subjects on the campus of the University of Bern. Informed consent was obtained verbally before participants entered the experimental room. We randomized subjects into two groups. In group one subjects had to judge the length of lines, as in the original Asch experiment. For this purpose, we placed 5 confederates in addition to a naïve subject in a room. The confederates were asked to behave as naïve subjects and entered the room one after the other. The front row of the seats in the experimental room were numbered such that subjects sat next to each other. The naïve subject was always assigned to seat number 5, leaving the last seat to another confederate. First, we presented some instructions to the subjects: “Welcome to our study on decision-making behavior and opinions. This study consists of two parts: In the first part in this room, we ask you to solve a total of 10 short tasks. In the second part in the room next door, we ask you to complete a short questionnaire on the laptop. In total, this study takes about 40 minutes. As compensation for your participation, you will receive 20 Swiss francs in cash after completing the study.” We then presented a reference line to subjects next to three other lines that were numbered 1 through 3 on projected slides. Subjects were asked to judge the length of the reference line by naming the number of the line that corresponds to the reference line in length. We presented 10 such line tasks (see Fig A1 in the S1 Appendix ). In the first two trials as well as in trials number 4 and 8, confederates pointed out the correct lines. Four trials were easy tasks, since the difference between the reference line and two of the other lines was large. The other six trials were more difficult, since the differences were small. Subjects were asked to call out the number of the correct line always starting with subject 1 through 6.

After the line task in part 2 of the experiment subjects were confronted with 5 general questions on different political issues. The statements were selected because we believe they describe fundamental attitudes towards different political or social groups in a democracy. The five statements read (1) “Do you think that the Swiss Federal Government should be given more power?”, (2) “Do you think trade unions should be given more power in Switzerland?”, (3) “Do you think that the employers’ association in Switzerland should be given more power?”, (4) “Do you think that citizens should be given more liberties in Switzerland?”, and (5) “Do you think that companies in Switzerland should be given more freedom?”. Subjects were asked to answer all 5 questions with either yes or no. The confederates in this group were instructed to answer “yes” to the first question and “no” to the rest. We chose this sequence of “yes” and “no” to prevent that subjects discover the existence of confederates. Finally, part 3 of the experiment consisted of an online questionnaire which subjects were asked to complete. To conceal that some participants were confederates all 6 participants were accompanied to separate rooms where the online-questionnaire was installed on a laptop. The questionnaire was designed to measure a number of different personality traits. Particularly, we measured the Big Five using a 10-item scale (two items for each of the 5 traits) as suggested by Rammstedt et al. [ 37 ] (see Table A1 in the S1 Appendix for item wording); a 10-item scale measuring self-esteem as suggested by Rosenberg [ 38 ] (see Table A2 in the S1 Appendix ); a short version of the Hagen Matrices Test [ 39 ] to measure intelligence and the 10-item version of the Martin Larson Approval Motivation Scale (MLAM) [ 36 ].

In group 2 the experimental design and procedure was the same as in group 1 besides the fact that correct answers in the length of lines judgment task were incentivized. In addition to the 20 Swiss francs show-up fee, subjects received one Swiss franc for every correct answer in the line judgment task, and hence, could earn up to 30 Swiss francs in total. Since there are no correct answers to political opinions these were not incentivized. However, we randomized the confederates’ answers to political opinion questions independently of whether a subject was in the incentivized or non-incentivized group. In one version confederates answered “yes” to the first question and “no” to the four other questions. In the other version the sequence of the confederates’ response was “no” to the first question and “yes” in response to the other four. The experiment was conducted by three different research teams consisting of 7 student assistants each. In every group 5 students acted as confederates and 2 as research assistants, recruiting subjects, welcoming and instructing them in the laboratory room, and reading out loud the projected instructions.

A power analysis suggested that we need about 100 subjects per experimental condition to find statistically significant (α = 0.05) differences of 5 percentage points for a power of 0.8. Hence, we stopped recruiting subjects after reaching 210 participants. The experiment was conducted between March 16, 2021 and April 30, 2021. The authors had no access to any information that links individual identifiers to the data. Subjects were debriefed after the end of the study by email.

Overall, 210 subjects participated in the experiment (female = 61%, mean age = 22.6). 102 subjects were randomly assigned to the non-incentivized group and 108 into the group with incentives. Moreover, 113 subjects were assigned to the sequences of “yes” and four “no” of the political opinion task and 97 to the reversed sequence, suggesting that the randomization procedure worked well. The questionnaire also contained an attention check. The question reads “In the following we show you five answer categories. Please do not tick any of the answers”. Four subjects failed to comply and ticked an answer, suggesting that they did not pay proper attention to the question wording. These subjects were excluded from the analysis. Furthermore, we asked subjects at the end of the questionnaire what they think the experiment was about. Three subjects recognized that the experiment was the line task experiment of Asch or expressed the suspicion that some of the other group members were confederates. We also excluded these three subjects from the analysis. Moreover, one subject answered the question about their gender with “other” and was also excluded from the analysis. Hence, these exclusions result in 202 valid cases. However, the results presented do not depend on these eight excluded observations.

Fig 1 presents the results of the ten line length tasks for the non-incentivized (grey bars) and for the incentivized conditions (blue bars). As can be clearly seen, almost none of the naïve subjects gave an incorrect answer when the group provided the correct answer which was the case in decision situations 1, 2, 4 and 8. However, when the group provides the false answer a substantial number of naïve subjects provided this incorrect answer as well (decision situations 3, 5, 6, 7, 9, and 10). The proportion of incorrect answers in the non-incentivized condition is relatively small in decision 3 (10%), but relatively high in decisions number 6 and 7 (44% and 47%). The average of incorrect answers is 33% in the non-incentivized group, which is a perfect replication of Asch’s (1955) original 36.8% result (two sample two-sided T-test, t(16) = 0.59, p = 0.57).

Note: Percent of correct answers by experimental group and trial including 95% confidence intervals. The numbers on top of the bars denote the trial numbers. “correct” stands for uncritical trials, “false” for critical trials. “easy” denotes easy trials with big differences between the lines, and “hard” denotes more difficult trials with smaller differences between the lines. The numbers between the bars denote the difference in proportions between the groups in percentage points. One-sided T-tests: * = p < 0.05. N without incentive (no) = 99, n with incentive (yes) = 103.

When correct answers are incentivized, the proportion of incorrect answers decreases by on average 8%-points. The difference between the groups is statistically significant in 2 out of 6 critical trials (p < 0.05 for one-sided T-tests). The difference also becomes evident when we consider the number of incorrect answers in the 6 critical trials. When decisions were not incentivized subjects gave on average 1.97 incorrect answers. In the incentivized condition the average number dropped to 1.47, leading to a statistically significant difference of 0.5 incorrect answers (t(208) = 2.24, p = 0.03 for two-sided T-test).

Next, Fig 2 presents the results concerning the five political questions. When the group said “yes” to the question of whether the Swiss Federal Council (the government in Switzerland) should have more power, 27% of the naïve subjects did so as well. When the group said “no” only 3% of the subjects said “yes” resulting in a difference of 23.4%-points. When the group said that trade unions should have more power 72% of the subjects answered “yes” as compared to only 29% when the group said “no” resulting in a difference of 43%-points. Similarly, the question of whether the employers’ association should have more power is agreed to by 44% and 6% respectively, depending on the group agreeing or disagreeing. Moreover, 81% of the subjects agreed that citizens in Switzerland should be given more liberties when the group does so, and 33% agreed to this question when the group says “no”. Finally, 46% said that companies should be given more freedom when the group agreed but only 8% did so when the group denied this question. The average difference in the proportion of yes-answers is 38%-points and all 5 differences are statistically highly significant. This result corresponds astonishingly close to the result in the length of line experiment and suggests that the influence of group pressure can be generalized to the utterance of political opinions.

Note: Percent of ‘yes’ answers to five general questions on political opinions in which all confederates answered uniformly ‘yes’ or ‘no’, by experimental group including 95% confidence intervals. The numbers on top of the bars stand for the difference in proportions between the respective groups in percentage points. Two-sided T-tests: *** = p < 0.001. n (sequence yes, no, no, no, no) = 109, n (sequence no, yes, yes, yes, yes) = 93.

One interesting question is whether the susceptibility to group pressure is linked to certain personality traits. To investigate this question, we count the number of wrong answers in the six critical trials of the length of line task. This variable is our dependent variable and runs from 0 when a subject always gave correct answers to 6 for subjects who gave only wrong answers. First, we wondered whether conformity is linked to the Big Five personality traits. We measured the Big Five using a short 10-item version as suggested by Rammstedt et al. [ 37 ] which measures each trait (openness, extraversion, agreeableness, conscientiousness, and neuroticism) with two questions (see Table A1 in the S1 Appendix ).

Second, we incorporate a 10-item measure of self-esteem, as suggested by Rosenberg [ 38 ], into the analysis (see Table A2 in the S1 Appendix ). Each item of the scale has four answer categories ranging from 1 = “disagree strongly”, 2 = “disagree”, 3 = “agree” to 4 = “agree strongly”. Subjects that score high on self-esteem are expected to have stronger confidence in their own perception and should be less influenced by the group’s opinion. Third, we measured individuals’ intelligence using a short version of the Hagen Matrices Test (HMT) [ 39 ]. The HMT consists of six 9-field matrices that show graphical symbols that follow a logical order. The last field is missing and the task of the subjects is to pick the correct symbol, out of eight, that fits and completes the pattern of the matrix. Hence, the HMT ranges from 0 if no answer is correct to 6 for subjects who provided six correct answers. The hypothesis is that subjects who score high on the HMT are less susceptible to the pressure of the group and also provide more correct answers in the line task.

Finally, conformity might be linked to the need for social approval. We measured the need for social approval with a 10-item version of the Martin Larson Approval Motivation Scale (MLAM) [ 36 ] (see Table A3 in the S1 Appendix ). Individuals that score highly on the MLAM display high need for social approval by others. Hence, we expect that subjects with higher values on the MLAM should also conform more often to the opinions of others in order to receive social approval. A summary of the descriptive information of the considered variables is depicted in Table A4 in the S1 Appendix . To investigate whether any of the measured personality traits are linked to the answering behavior in the line task we conducted multiple OLS regression analysis. The results of this analysis are depicted in the coefficient plots in Fig 3 (see also Table A5 in the S1 Appendix ).

Note: N = 202. Unstandardized coefficients of multiple linear OLS regressions including robust 95% confidence intervals. Poisson and negative binomial models do not alter the results in any substantial way. Variables marked with an ‘*’ indicate statistically significant differences in the coefficients between models (2) and (3).

First, model 1 presents the effects on the number of conforming answers for the whole sample. In the incentivized condition subjects gave on average 0.43 fewer conforming answers as compared to the unincentivized condition. This effect mirrors the bivariate result already presented in Fig 1 and is statistically significant for the 5% level. In tendency, females show more conforming answers, but this effect is statistically only significant for the 10% level. Besides “openness” none of the personality traits contained in the Big Five show any statistically significant effects. This is also true for the other effects of intelligence, self-esteem, and the measure for social approval seeking. Models 2 and 3 show the results for men and women separately. The separate results suggest that women react somewhat more strongly to incentives than do men. However, a test for differences in coefficients suggests that the effects do not differ (χ 2 (1) = 1.11, p = 0.29). Intelligence seems to have greater importance for men, leading to 0.27 fewer conforming answers for every correct answer of the HMT. However, this effect does not differ statistically from the effect for females (χ 2 (1) = 1.35, p = 0.25). No difference in effects can be observed for self-esteem. However, in the female sample the need for social approval is positively linked to the number of conforming answers, which is not the case in the male sample (χ 2 (1) = 4.23, p = 0.04); but the effect of social approval in the female sample is relatively small.

We conducted a number of robustness checks with the presented analyses. Since our dependent variable is a count variable (number of conforming answers) the models can also be estimated using Poisson regressions or negative-binomial models. However, none of our presented results change in any substantial way using these alternatives. Furthermore, we excluded 24 more subjects who when asked at the end of the experiment about the goal of the study said that the experiment was about group pressure or conformity, although they did not explicitly mention Asch or the suspicion that other participants were confederates. But these additional exclusions also did not change the results substantially (see Table A6 in the S1 Appendix ). Finally, we also incorporated the 10-item version of the Marlowe-Crowne Scale [ 34 , 35 ] suggested by Clancy [ 40 , 41 ]. However, inclusion of the scale did not show any statistically significant effects or did change any of the other estimates.

5. Conclusion and discussion

In this study we first replicated the original experiment of Asch [ 1 – 3 ] with 5 confederates and ten line tasks. We find an average error rate of 33% which replicates the original findings of Asch very closely and which is in line with other replications that were conducted predominately with American students [ 12 ]. Together with recent studies from Japan [ 14 ], and Bosnia and Herzegowina [ 15 ], our study provides further evidence that the influence of groups on individuals’ judgments is a universal phenomenon, and is still valid today. Furthermore, we incentivized the decisions and find a drop of the error rate by 8%-points to 25%. Hence, monetary incentives do not eliminate the effect of group pressure. This finding sheds doubt on former results which predominately show the opposite effect, namely that incentives increase compliance [ 5 , 6 ].

Moreover, our study suggests that group pressure is not only influential in the simple line task but also when it comes to political opinions. We randomized the groups’ response to five different political statements and find an average conformity rate of 38%. Hence, these results suggest that the original finding of Asch can also be generalized to matters of opinion. This result is in line with former evidence by Crutchfield [ 8 ], and Mallinson and Hatemi [ 9 ]. However, both of these studies had only small sample sizes of 50 and 58 subjects respectively, which called for further replication studies. Finally, we measured the Big Five, intelligence, self-esteem and social approval. With the exception of openness, our study finds no support that these personality traits are statistically significantly related to the susceptibility of group pressure.

Of course, our study has some limitations, which suggest a number of further research questions. First, we used a relatively large sample of 202 subjects providing more statistical power than former replications and extensions of the Asch experiment; however, our subjects were also students, and hence, it would be important to have further replications with non-student samples. This would allow further investigations of the susceptibility to group pressure with respect to age, different occupational groups, different social backgrounds, and different levels of social experience.

Second, the subjects we investigate are strangers. That means the single naïve subjects did not know the confederates. An interesting question for further research would be, whether group pressure is stronger among non-strangers or whether dissent becomes more acceptable among a group of friends.

Third, we demonstrate that monetary incentives reduce the error rate. However, our incentives were one Swiss franc for every correct answer, and hence small. Thus, the interesting question remains whether larger incentives reduce the error rate further, or can even lead to the elimination of it.

Fourth, the political statements we choose are relatively moderate and general. This leaves the question open as to whether subjects would also conform to more extreme or socially less acceptable statements. Furthermore, our subjects might have rarely thought about the statements we provided, leaving the question of what would happen with respect to statements about which subjects had stronger opinions or which are more related to their identity.

With the exception of openness all personality traits considered (e.g. intelligence, self-esteem, need for social approval) are not related to conformity. This raises a number of very interesting research questions. One possibility is that the traits were not measured good enough, and that measurement errors impede the identification of these individual differences. This concern applies particularly to the measurement of the Big Five where we relied on the short 10-item version suggested by Rammstedt et al. [ 37 ]. Hence, the puzzling result that openness leads to less conformity must be replicated before it can count as a reliable finding. However, the finding is in line with the assumption of Eck and Gebauer [ 30 ]. Another possibility is that other personality traits are more important when it comes to conformity behavior. Hence, there is much room for further interesting research concerning conformity behavior in situations of group pressure.

Supporting information

S1 appendix, acknowledgments.

We like to thank our student assistants for helping us with the data collection. Their names are: Yvonne Aregger, Elias Balmer, Ambar Conca, Davide Della Porta, Shania Flück, Julian Gerber, Anna Graf, Ina Gutjahr, Kim Gvozdic, Anna Häberli, Chiara Heiss, Paula Kühne, Jenny Mosimann, Remo Parisi, Elena Raich, Virginia Reinhard, Fiona Schläppi, Maria Tournas, Angela Ventrici, Marco Zbinden, Sarah Zwyssig.

Funding Statement

The author(s) received no specific funding for this work.

Data Availability

Learning Materials

• Business Studies
• Combined Science
• Computer Science
• Engineering
• English Literature
• Environmental Science
• Human Geography
• Macroeconomics
• Microeconomics
• Asch Conformity Experiments

Conformity occurs when we change our behaviour or opinions to match those of a group. We all conform in some way to the explicit rules of the groups we belong to (e.g. by adhering to dress codes) or to implicit ones (e.g. respecting queues). 

Millions of flashcards designed to help you ace your studies

• Cell Biology

What does the word confederate mean?

What task did Asch (1951) use to investigate conformity?

Review generated flashcards

to start learning or create your own AI flashcards

Start learning or create your own AI flashcards

Vaia Editorial Team

Team Asch Conformity Experiments Teachers

• 9 minutes reading time
• Checked by Vaia Editorial Team
• Approaches in Psychology
• Basic Psychology
• Biological Bases of Behavior
• Biopsychology
• Careers in Psychology
• Clinical Psychology
• Cognition and Development
• Cognitive Psychology
• Data Handling and Analysis
• Developmental Psychology
• Eating Behaviour
• Emotion and Motivation
• Famous Psychologists
• Forensic Psychology
• Health Psychology
• Individual Differences Psychology
• Issues and Debates in Psychology
• Personality in Psychology
• Psychological Treatment
• Relationships
• Research Methods in Psychology
• Schizophrenia
• Scientific Foundations of Psychology
• Scientific Investigation
• Sensation and Perception
• Social Context of Behaviour
• Activation Synthesis Theory
• Addiction Treatment
• Adornos Theory
• Altered States of Consciousness
• An Introduction To Mental Health
• Anger Management Programmes
• Antidepressant Medications
• Autonomic Nervous System in Psychology
• Bickman Obedience Study
• Biological Explanation of Depression
• Body Language
• Brain During Sleep
• Brain Structure
• Brain and Neuropsychology
• Bystander Effect
• Caspi et al 2003
• Characteristics of Addiction
• Characteristics of Mental Health
• Clinical Depression
• Cognitive Neuroscience
• Collective Behaviour
• Community Sentencing
• Concepts of the Self
• Crime Punishment
• Crime and Culture
• Criminal Psychology
• Depression Treatment
• Depression Vs Sadness
• Dispositional Factors
• Drug Abuse Vs Addiction
• Erikson's Stages Of Development
• Ethical Issues in Social Influence Research
• Eye Contact
• Fight-or-Flight Response
• Flow States
• Freud Wolfman Study
• Freud's Theory of Dreaming
• Functions of Sleep
• Hebbs Theory
• How Crime Is Measured
• Human Language and Animal Communication
• Humanistic Theory of Self
• Identity and Free Will
• Improving Sleep
• Language and Thought
• Language, Thought And Communication
• Lateralisation
• Localisation of Brain Function
• Majority and Minority Influence
• Management Of Insomnia
• Milgrams Agency Theory
• Nervous System
• Neuroimaging Techniques
• Neurological Damage On Behaviour
• Neuropsychology
• Non Verbal Communication
• Non-Verbal Behaviour
• Penfield's Study of The Interpretive Cortex
• Personal Space
• Personality Scales
• Personality Types
• Piaget vs Vygotsky
• Prosocial Behaviour
• Psychological Problems
• Rehabilitation
• Restorative Justice
• Self Management Psychology
• Self Report
• Siffre Cave Study Psychology
• Sleep Hygiene
• Sleep and Dreaming
• Sleep and Zeitgebers
• Social Effects
• Stages of Sleep
• Studying the brain
• Symptoms Of Schizophrenia
• The James Lange Theory
• Theories of Depression
• Theories of Schizophrenia
• Theories of addiction
• Trait Theory of Personality
• Tulving's Gold Memory Study
• Types of crime
• Understanding Crime
• Social Psychology

Jump to a key chapter

However, how can we investigate conformity and the factors that affect it in a lab? In the 1950s, Solomon Asch tackled this problem and created a replicable laboratory procedure to observe conformity .

Asch's conformity experiment

To study the effects of group pressure in a laboratory environment Asch (1951, 1956) investigated whether participants' judgements on a simple visual perception task will be affected by group pressure. The sample consisted of 123 male, American students.

Deception was used to elicit conformity. Participants took part in a task in a group and didn't know they were the only subject of the experiment. Participants thought they were a member of a larger panel making the judgements. They met the other group members who were taking part in the task, not knowing they were confederates .

Confederates are actors that pretend to be participants in an experiment.

After the stimuli were presented to the group in some trials, all other group members unanimously gave a wrong answer to the task. Participants had to decide whether to provide the right answer and stand out from other group members or whether to conform to the group that is making a mistake and avoid the anxiety of being the odd one out.

This experimental paradigm was the first one to investigate conformity concerning an unambiguous task . Previously conformity was investigated using ambiguous tasks that required some interpretation; participants were often unsure about the correct answer and so relied on the answers of others. This type of conformity relates to our need to be right. Using an unambiguous task allows us to understand whether conformity occurs in situations when we know the group is incorrect. It can be argued that this type of conformity relates to our need to be liked by the group or simply the need to fit in.

Asch's conformity experiment: hypothesis

Asch hypothesised that in the critical trials, when the other group members (confederates) unanimously give a wrong answer to the task, participants will conform to the group and provide the same answer as others, even though they know it's incorrect.

Asch's conformity experiment: procedure

Asch used an unambiguous visual perception task to measure conformity. The experimental stimuli consisted of a standard line and three comparison lines. Participants had to make judgements about which comparison line matches the length of the standard line.

In the control condition , participants made judgements about the lines alone, without any group influence to establish the task's difficulty.

In the experimental condition , participants had to make judgements in groups ranging from 7 to 9.

After experimental stimuli were presented, each group member had to voice their answer publicly.

Participants were the second last person to state their judgement, which means they heard the answers of almost the entire group before stating their own.

The task was repeated 18 times (18 trials). In 12 of the trials, confederates unanimously gave a wrong answer to the task (either chose a longer or shorter line than the original line). These were the critical experimental trials that measured conformity.

Asch's conformity experiment: results

In the absence of the group, when participants made judgements alone they were correct over 99% of the time, suggesting that the task was obvious.

Most participants (75%) conformed to the group at least once in the experimental condition. Half of the participants conformed in at least 50% of the 12 critical trials. On average, 37% of participants conformed in each of the twelve critical trials.

When asked about why they conformed, some participants admitted that they started doubting the accuracy of their perception, and some conformed to avoid standing out from the rest of the group. The majority's motivation to conform was to avoid social rejection. Therefore, it can be concluded that they conformed due to the normative social influence .

Summary of Asch's conformity experiment

In a nutshell, Asch's conformity experiment investigated the degree of conformity concerning an obvious task (matching the length of a line to comparison lines).

Each participant was tested in a group of confederates. In the 12 critical trials, confederates unanimously gave a wrong answer, putting pressure on the participant to conform to the rest of the group. Under the influence of group pressure, participants experienced distress related to the fear of rejection. 37% of participants conformed on an average critical trial, with 75% conforming at least once.

Asch's conformity experiment: evaluation

Asch's study was one of the first experimental investigations of conformity and has helped us understand how we can often conform. However, it's not without its limitations.

Validity of findings

Asch's conformity study was a laboratory experiment, which allowed to control of potential confounding variables and therefore has high internal validity and minimal issues with extraneous variables. However, it has been criticised for low ecological validity. The task used is artificial and quite different from how we experience conformity in our daily lives.

Generalisability

Asch's sample consisted of only male, American students, limiting the findings' generalisability to the wider population and might not reflect conformity across cultures. The US is an individualist culture ; it can be argued that the degree of conformity would be higher in collectivist cultures that have a greater emphasis on the group. It's also possible that conformity differs in women compared to men, as women in many cultures can be more oriented towards maintaining social relationships.

Failed replications

Later, Perrin and Spencer (1980) replicated Asch's experiment on a sample of UK Engineering students. Across 396 trials, only one participant conformed. Some have argued that this failed replication suggests that Asch's findings were limited to his time (The United States population in the 1950s) and didn't necessarily apply to other contexts. It is also possible that a greater degree of expertise and confidence of Engineering students prevented conformity on this task.

Asch's conformity experiment: ethical issues

The first ethical issue to consider concerning Asch's experiment is the use of deception . Participants were deceived about the character of the study (they thought they were taking part in a visual test experiment) and about the other group members (they thought the confederates were real participants). While deceiving participants is unethical, it can be argued it was necessary for conducting this environment. If participants knew the study investigated conformity and they were the only subject, they wouldn't conform.

Another issue is the lack of protection from harm . The experience of being the only one that perceives the lines differently and gives different answers than the group was distressing to most participants. After the experiment, participants reported experiencing a fear of rejection or anxiety related to the pressure to conform. During the experiment, participants were not protected from distress. However, participants were debriefed after the experiment and took part in an interview about their experience, which could reduce their distress.

Variations of Asch's conformity experiment

Asch conducted multiple variations of his original experiment to investigate which factors affect the degree of conformity to clearly incorrect majority influence and take situational factors into account.

To investigate how the size of the group impacts conformity rates, Asch (1956) tested participants in groups ranging from 2 to 15. When only one confederate was present, conformity dropped to 3%. When two confederates were present, conformity rapidly increased to 13.6%. When three confederates were present conformity reached 33% and mostly levelled off as the number of confederates increased further.

These findings suggest that a lower group size reduces conformity. However, as many as 3 other people can be a source of pressure to conform.

In one of Asch's experiment variations, participants wrote down their answers privately without disclosing them publicly to the other group members. Anonymity decreased conformity. Only 12.5% of participants conformed in this variation of the study. However, it is important to note that even when participants could give anonymous responses, some were still influenced by the group.

Task difficulty

To examine how the difficulty of the task impacts our tendency to rely on the judgements of others, Asch conducted a variation of his study but with smaller differences between comparison lines. As the comparison lines were much closer together in length, matching the standard line became harder. Asch reported that when the task's difficulty increases, conformity also increases. This effect can be attributed to the informational social influence .

Informational social influence occurs when we are uncertain what behaviour is right in a particular situation so we refer to what others do for guidance.

Asch Conformity Experiments - Key takeaways

To study the effects of group pressure in a laboratory environment Asch (1951, 1956) investigated whether participants' judgements on a simple visual perception task will be affected by group pressure.

Asch's sample consisted of 123 male American students. Each participant was tested in a group of confederates.

Asch hypothesised that when the confederates will unanimously give a wrong answer to the task in the critical trials, participants will conform to the group, even though they know the group is incorrect.

Asch used an unambiguous visual perception task to measure conformity. The experimental stimuli consisted of a standard line and three comparison lines.

Under the influence of group pressure, participants experienced distress related to the fear of rejection. 37% of participants conformed on an average critical trial, with 75% conforming at least once.

Asch's study (1951) was criticised for low ecological validity and limited sample use. Failed replications also question how transferable Asch's findings are across cultures and time.

The study raises ethical issues that the use of deception and lack of protection from psychological harm.

Variations of Asch's experiment found that group size, anonymity and task difficulty affect conformity, alongside unanimity.

Flashcards in Asch Conformity Experiments 2

Someone you can confide in

Unambiguous visual perception task  

Learn with 2 Asch Conformity Experiments flashcards in the free Vaia app

We have 14,000 flashcards about Dynamic Landscapes.

Already have an account? Log in

Frequently Asked Questions about Asch Conformity Experiments

How many conformity experiments did Asch conduct?

Asch conducted many variations of his conformity experiments. The main variations tested for the effects of group size, unanimity, anonymity and task difficulty.

What type of experiment was the Asch conformity study?

Asch's conformity study was a laboratory experiment.

What was the procedure in the Asch conformity experiment?

After experimental stimuli were presented each group member had to voice their answer publicly. Participants were the second last person to state their judgement, which means they heard the answers of almost the entire group before stating theirs. On 12 of the 18 trials, confederates unanimously gave a wrong answer to the task (either chose a longer or shorter line than the original line). These were the critical trials that measured conformity.

What was the surprising result of Asch's experiment on conformity?

Most participants (75%) conformed to the group at least once. Half of the participants conformed on at least 6 of the twelve critical trials. On average, 37% of participants conformed in each of the twelve critical trials.

What year was Asch conformity experiment?

Asch conducted his conformity experiment in 1951.

Discover learning materials with the free Vaia app

Vaia is a globally recognized educational technology company, offering a holistic learning platform designed for students of all ages and educational levels. Our platform provides learning support for a wide range of subjects, including STEM, Social Sciences, and Languages and also helps students to successfully master various tests and exams worldwide, such as GCSE, A Level, SAT, ACT, Abitur, and more. We offer an extensive library of learning materials, including interactive flashcards, comprehensive textbook solutions, and detailed explanations. The cutting-edge technology and tools we provide help students create their own learning materials. StudySmarter’s content is not only expert-verified but also regularly updated to ensure accuracy and relevance.

Team Psychology Teachers

Study anywhere. Anytime.Across all devices.

Create a free account to save this explanation..

Save explanations to your personalised space and access them anytime, anywhere!

By signing up, you agree to the Terms and Conditions and the Privacy Policy of Vaia.

Sign up to highlight and take notes. It’s 100% free.

Join over 22 million students in learning with our Vaia App

The first learning app that truly has everything you need to ace your exams in one place

• Flashcards & Quizzes
• AI Study Assistant
• Study Planner
• Smart Note-Taking

Privacy Overview

The Asch Conformity Experiments

What Solomon Asch Demonstrated About Social Pressure

• Recommended Reading
• Key Concepts
• Major Sociologists
• News & Issues
• Research, Samples, and Statistics
• Archaeology

The Asch Conformity Experiments, conducted by psychologist Solomon Asch in the 1950s, demonstrated the power of conformity in groups and showed that even simple objective facts cannot withstand the distorting pressure of group influence.

The Experiment

In the experiments, groups of male university students were asked to participate in a perception test. In reality, all but one of the participants were "confederates" (collaborators with the experimenter who only pretended to be participants). The study was about how the remaining student would react to the behavior of the other "participants."

The participants of the experiment (the subject as well as the confederates) were seated in a classroom and were presented with a card with a simple vertical black line drawn on it. Then, they were given a second card with three lines of varying length labeled "A," "B," and "C." One line on the second card was the same length as that on the first, and the other two lines were obviously longer and shorter.

Participants were asked to state out loud in front of each other which line, A, B, or C, matched the length of the line on the first card. In each experimental case, the confederates answered first, and the real participant was seated so that he would answer last. In some cases, the confederates answered correctly, while in others, the answered incorrectly.

Asch's goal was to see if the real participant would be pressured to answer incorrectly in the instances when the Confederates did so, or whether their belief in their own perception and correctness would outweigh the social pressure provided by the responses of the other group members.

Asch found that one-third of real participants gave the same wrong answers as the Confederates at least half the time. Forty percent gave some wrong answers, and only one-fourth gave correct answers in defiance of the pressure to conform to the wrong answers provided by the group.

In interviews he conducted following the trials, Asch found that those that answered incorrectly, in conformance with the group, believed that the answers given by the Confederates were correct, some thought that they were suffering a lapse in perception for originally thinking an answer that differed from the group, while others admitted that they knew that they had the correct answer, but conformed to the incorrect answer because they didn't want to break from the majority.

The Asch experiments have been repeated many times over the years with students and non-students, old and young, and in groups of different sizes and different settings. The results are consistently the same with one-third to one-half of the participants making a judgment contrary to fact, yet in conformity with the group, demonstrating the strong power of social influences.

Connection to Sociology

The results of Asch's experiment resonate with what we know to be true about the nature of social forces and norms in our lives. The behavior and expectations of others shape how we think and act on a daily basis because what we observe among others teaches us what is normal , and expected of us. The results of the study also raise interesting questions and concerns about how knowledge is constructed and disseminated, and how we can address social problems that stem from conformity, among others.

Updated  by Nicki Lisa Cole, Ph.D.

• An Overview of the Book Democracy in America
• The Social Transformation of American Medicine
• McDonaldization: Definition and Overview of the Concept
• Understanding Durkheim's Division of Labor
• Émile Durkheim: "Suicide: A Study in Sociology"
• Malcolm Gladwell's "The Tipping Point"
• The Presentation of Self in Everyday Life
• Definition of the Sociological Imagination and Overview of the Book
• Stigma: Notes on the Management of Spoiled Identity
• The Main Points of "The Communist Manifesto"
• Overview of The History of Sexuality
• A Book Overview: "The Protestant Ethic and The Spirit Of Capitalism"
• Savage Inequalities: Children in America’s Schools
• 15 Major Sociological Studies and Publications
• Learn About Various Sanctions in Forcing Compliance With Social Norms
• Understanding Resocialization in Sociology

Reference Library

Collections

• See what's new
• All Resources
• Student Resources
• Assessment Resources
• Teaching Resources
• CPD Courses
• Livestreams

Study notes, videos, interactive activities and more!

Psychology news, insights and enrichment

Currated collections of free resources

Browse resources by topic

• All Psychology Resources

Resource Selections

Currated lists of resources

Group size

Asch identified group size as a variable that influences conformity. Asch found that as he increased the size of the majority, conformity levels increased. With two confederates, conformity occurred on 12.8% of trials, rising to 32% for trials with three confederates. However, after that group size did not make a significant difference to the rate of conformity.

• Share on Facebook
• Share on Twitter
• Share by Email

Conformity - Asch (1951)

Study Notes

Free Revision Crosswords for Social Psychology

9th March 2017

Conformity - Variations of Asch (1951)

Social psychology "red herring" activity.

Quizzes & Activities

Social Psychology "Connection Wall" Activity

Social influence: mcq revision test 1 for aqa a level psychology.

Topic Videos

Model Answer for Question 4 Paper 1: AS Psychology, June 2016 (AQA)

Exam Support

Our subjects

• › Criminology
• › Economics
• › Geography
• › Health & Social Care
• › Psychology
• › Sociology
• › Teaching & learning resources
• › Student revision workshops
• › Online student courses
• › CPD for teachers
• › Livestreams
• › Teaching jobs

Boston House, 214 High Street, Boston Spa, West Yorkshire, LS23 6AD Tel: 01937 848885

• › Contact us
• › Terms of use
• › Privacy & cookies

© 2002-2024 Tutor2u Limited. Company Reg no: 04489574. VAT reg no 816865400.