The five "others" in this demonstration are scripted, not real people. On some trials all five will give the same wrong answer on purpose. Unlike Asch's participants, you know this in advance — the demonstration is about feeling the pull of a unanimous room anyway. Your answers stay in this tab and are discarded when you leave.
The interactive station needs JavaScript, which is off — nothing is hidden from you because of that. The full explanation below covers everything the station demonstrates.
The five others below are scripted. On some trials they will all give the same wrong answer on purpose — you know this in advance, which is exactly what Asch’s participants did not.
The experience
Six line judgments, spoken in company. The station above seats you at the end of a
row of five others who answer before you do — and, unlike the people who walked
into Solomon Asch's classroom in the early 1950s, you are told the truth in
advance: the five are scripted, and on some trials every one of them will give the
same wrong answer on purpose. Nothing about you is scored, and your answers stay
in this tab. What the warning cannot switch off is the interesting part — the beat
of hesitation before contradicting a confident, unanimous room. But be clear about
what a disclosed, solo demonstration cannot do: it cannot recreate real social
pressure from real people — the eyes of a physically present group, or the cost of
dissenting out loud in front of strangers. Treat the station as a sketch of the
arrangement; the entry is about the real thing.
What is happening
Volunteers matched the length of a line to one of three
comparison lines — a task so clear that, judging alone, control participants were
accurate on over 99% of estimates, while the scripted majority's wrong answers
were off by three quarters of an inch to an inch and three quarters. Seated in a
group whose other members had secretly agreed to give the same wrong answer on 12
of 18 trials, the 123 participants in the core experiments moved to the majority's
wrong answer on 36.8% of those critical judgments.Historical / descriptive The group was not
arguably right; it was plainly wrong. That is what made the situation so strange
to be inside.
Participants were recruited for what appeared to
be a simple experiment in visual discrimination; the prearrangement with the
majority was hidden. Asch ended every session with an individual interview in
which the purpose of the experiment, the participant's role and the majority's
role were fully explained — he considered it neither advisable nor justified to
let anyone leave without that explanation.Historical / descriptive
What this shows
Unanimity mattered more than numbers. A single fellow
dissenter cut pro-majority errors from 32% to about 10% (5.5% when the partner was
instructed to answer accurately), and Asch reported that a unanimous majority of
three was far more effective than a majority of eight containing one dissenter —
while majorities larger than three added almost nothing to the pressure.Historical / descriptive The
pressure, in other words, is structural: it takes very little to build it, and one
honest voice to break it.
The average hides the range: about one quarter of
participants (24%) never once went with the majority, roughly as many (27%) went
with it on eight or more of the twelve critical trials, and the rest fell in
between — Asch himself counted the wide individual differences among the study's
significant findings.Historical / descriptive His interviews suggested that yielding was not one
thing either: a few came to see the majority as right, most decided their own
perception must be at fault, and some knew exactly what they saw but could not
bear to stand out.
What this does not show
Most of what happened in Asch's laboratory was not
conformity: across the core experiments about two thirds of all critical
judgments were correct despite the pressure, and Asch read his results as much as
a study of independence as of yielding. A content analysis of 99 social psychology
textbook accounts published between 1953 and 1984 found that authors increasingly
accentuated the role of conformity and underestimated that of independence.Supported, with limits
The experiments therefore do not show that people are sheep, and they do not show
that groups can make people believe anything: the task was public agreement about
a clear fact, under the eyes of strangers, and what most yielders surrendered was
their report, not their eyesight.
Research and limitations
The size of the effect is not a constant
of human nature: a meta-analysis of 133 Asch-type studies from 17 countries found
that conformity in US studies had declined since the 1950s, and that collectivist
countries tended to show higher levels of conformity than individualist ones —
with the cultural measures mattering more than classic situational moderators such
as majority size.Supported, with limitsThe basic effect still
replicates: a 2024 systematic review of conformity research reports recent
Asch-type studies finding rates close to the originals, including a 2023
replication with a conformity rate of 33%.Supported, with limits
Remember also who the original numbers describe: male American college students,
judging lines, in the early 1950s. Asch treated the effect as a joint product of
the stimulus, the group structure and the person — not a fixed quantity — and his
own variations moved it across nearly its whole range. And a browser page has a
limit worth repeating: five drawings that answer in sequence are not seven people
in a room. If the station gives you even a flicker of the pull, the honest
inference is not "so that is what it was like" but "and the real thing was
stronger".
Research drawer — claims, sources and limits
Every highlighted claim above is recorded here with its evidence grade and sources. Grades are per claim; this site never computes an overall truth score.
↩ Volunteers matched the length of a line to one of three comparison lines — a task so clear that, judging alone, control participants were accurate on over 99% of estimates, while the scripted majority's wrong answers were off by three quarters of an inch to an inch and three quarters. Seated in a group whose other members had secretly agreed to give the same wrong answer on 12 of 18 trials, the 123 participants in the core experiments moved to the majority's wrong answer on 36.8% of those critical 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, Whole No. 416), 1-70. https://doi.org/10.1037/h0093718 · accessed 2026-08-24 Support note: Read directly from the scanned monograph: control estimates 'showed an accuracy of over 99 per cent' (p. 3); the majority's estimates 'departed from the true value by amounts ranging from 3/4 to 1 3/4 inches' (p. 3, Table 1); 12 critical trials among 18 comparisons (pp. 6-7); Table 3 (p. 10) gives N=123 across three experimental groups, mean per cent errors 36.8, against 0.7 per cent in the control group (N=37).
Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgments. In H. Guetzkow (Ed.), Groups, Leadership and Men: Research in Human Relations (pp. 177-190). Pittsburgh: Carnegie Press. https://gwern.net/doc/psychology/1952-asch.pdf · accessed 2026-08-24 Support note: The program's first report confirms the structure: a group of eight with one naive critical subject, unanimous wrong answers on 12 of 18 trials (Table 1), mean 3.84 errors against 0.08 in controls (Table 2), and one-third of critical-group estimates displaced toward the majority; its text describes the majority's errors as ranging between 1/2 and 1 3/4 inches.
↩ Participants were recruited for what appeared to be a simple experiment in visual discrimination; the prearrangement with the majority was hidden. Asch ended every session with an individual interview in which the purpose of the experiment, the participant's role and the majority's role were fully explained — he considered it neither advisable nor justified to let anyone leave without that explanation.
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, Whole No. 416), 1-70. https://doi.org/10.1037/h0093718 · accessed 2026-08-24 Support note: P. 3: subjects gathered 'to take part in what appeared to be a simple experiment in visual discrimination' and the critical subject 'was not aware of this prearrangement'; p. 9: an individual interview 'followed immediately upon the conclusion of the experiment', 'the purport of the experiment was explained in full', and 'We did not consider it advisable or justified to allow subjects to leave without receiving an explanation of the procedure and the reasons for the investigation.'
Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgments. In H. Guetzkow (Ed.), Groups, Leadership and Men: Research in Human Relations (pp. 177-190). Pittsburgh: Carnegie Press. https://gwern.net/doc/psychology/1952-asch.pdf · accessed 2026-08-24 Support note: P. 225 of the reprint: 'Toward the conclusion of the interview each subject was informed fully of the purpose of the experiment, of his role and of that of the majority', with the surrounding passage stating the experimenter's responsibility to clarify doubts and explain why the subject was placed in the situation.
↩ Unanimity mattered more than numbers. A single fellow dissenter cut pro-majority errors from 32% to about 10% (5.5% when the partner was instructed to answer accurately), and Asch reported that a unanimous majority of three was far more effective than a majority of eight containing one dissenter — while majorities larger than three added almost nothing to the pressure.
Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgments. In H. Guetzkow (Ed.), Groups, Leadership and Men: Research in Human Relations (pp. 177-190). Pittsburgh: Carnegie Press. https://gwern.net/doc/psychology/1952-asch.pdf · accessed 2026-08-24 Support note: Reports pro-majority errors dropping to 10.4 per cent with a second naive subject and 5.5 per cent with an instructed accurate partner, against 32 per cent under the unanimous baseline; states 'a unanimous majority of three is, under the given conditions, far more effective than a majority of eight containing one dissenter'; and Table 3 with the surrounding text: 'The effect appeared in full force with a majority of three. Larger majorities of four, eight, and sixteen did not produce effects greater than a majority of three.'
Asch, S. E. (1955). Opinions and social pressure. Scientific American, 193(5), 31-35. doi:10.1038/scientificamerican1155-31 https://www.panarchy.org/asch/social.pressure.1955.html · accessed 2026-08-24 Support note: Corroborates both series for the larger sample: with a supporting partner present, subjects 'answered incorrectly only one fourth as often as under the pressure of a unanimous majority'; against two opponents errors were 13.6 per cent and against three 31.8 per cent, while a single opponent swayed subjects little.
↩ The average hides the range: about one quarter of participants (24%) never once went with the majority, roughly as many (27%) went with it on eight or more of the twelve critical trials, and the rest fell in between — Asch himself counted the wide individual differences among the study's significant findings.
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, Whole No. 416), 1-70. https://doi.org/10.1037/h0093718 · accessed 2026-08-24 Support note: P. 11: 'One-fourth of the experimental groups (24 per cent) gave errorless estimates, while an approximately equal number (27 per cent) gave majority-determined estimates from eight to twelve times. Between these extremes is to be found one-half of the critical group'; and 'That the majority elicited widely different reactions is one significant aspect of the present findings.'
Asch, S. E. (1955). Opinions and social pressure. Scientific American, 193(5), 31-35. doi:10.1038/scientificamerican1155-31 https://www.panarchy.org/asch/social.pressure.1955.html · accessed 2026-08-24 Support note: States that about one quarter of the subjects were completely independent and never agreed with the erroneous judgments of the majority, while at the other extreme some individuals went with the majority nearly all the time.
↩ Most of what happened in Asch's laboratory was not conformity: across the core experiments about two thirds of all critical judgments were correct despite the pressure, and Asch read his results as much as a study of independence as of yielding. A content analysis of 99 social psychology textbook accounts published between 1953 and 1984 found that authors increasingly accentuated the role of conformity and underestimated that of independence.
Friend, R., Rafferty, Y., & Bramel, D. (1990). A puzzling misinterpretation of the Asch 'conformity' study. European Journal of Social Psychology, 20(1), 29-44. https://doi.org/10.1002/ejsp.2420200104 · accessed 2026-08-24 Support note: Official abstract (read via Crossref metadata): Asch 'equally, if not more, intended and interpreted it as demonstrating the powers of independence'; the evidence is '99 accounts in social psychology textbooks published between 1953 ... and 1984'; the authors found accounts 'have increasingly accentuated the role of conformity and underestimated that of independence'.
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, Whole No. 416), 1-70. https://doi.org/10.1037/h0093718 · accessed 2026-08-24 Support note: Table 3 (p. 10): mean per cent errors 36.8 for all experimental groups, i.e. about 63 per cent of critical estimates correct; p. 10 states 'the preponderance of estimates was, in each of the experimental groups, correct or independent of the majority'. The 1951 report gives 68 per cent correct for its 50 subjects, so 'about two thirds' covers both reports.
↩ The size of the effect is not a constant of human nature: a meta-analysis of 133 Asch-type studies from 17 countries found that conformity in US studies had declined since the 1950s, and that collectivist countries tended to show higher levels of conformity than individualist ones — with the cultural measures mattering more than classic situational moderators such as majority size.
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), 111-137. https://doi.org/10.1037/0033-2909.119.1.111 · accessed 2026-08-24 Support note: Official abstract: 'The literature search produced 133 studies drawn from 17 countries. An analysis of US studies found that conformity has declined since the 1950s. ... Collectivist countries tended to show higher levels of conformity than individualist countries.' Abstract-level access only; effect sizes and moderator tables were not read.
Capuano, C., & Chekroun, P. (2024). A Systematic Review of Research on Conformity. International Review of Social Psychology, 37(1), 13. https://doi.org/10.5334/irsp.874 · accessed 2026-08-24 Support note: Open-access systematic review summarizing Bond and Smith (1996): 'conformity appears to be higher in collectivist cultures than in so-called individualist cultures. More importantly, the impact of these cultural values was greater than other situational variables considered key moderators, such as majority size' — the source for the 'mattering more than majority size' clause.
↩ The basic effect still replicates: a 2024 systematic review of conformity research reports recent Asch-type studies finding rates close to the originals, including a 2023 replication with a conformity rate of 33%.
Capuano, C., & Chekroun, P. (2024). A Systematic Review of Research on Conformity. International Review of Social Psychology, 37(1), 13. https://doi.org/10.5334/irsp.874 · accessed 2026-08-24 Support note: States that 'recent investigations have indeed reported conformity rates closely resembling those observed by Asch in the 1950s, exemplified by the replication conducted by Franzen and Mader (2023), which observed a conformity rate of 33%, mirroring Asch's rates'. The Franzen and Mader paper itself was not opened; the claim is attributed to the review.