Normative and informative social influence


The Asch conformity experiments were conducted in the 1950s by Solomon Asch, a psychologist at Swarthmore College. The experiments were designed to examine the extent to which an individual would conform to the majority opinion of a group, even if the individual's own perception differed from the group's. In the experiments, a group of confederates (individuals who were in on the experiment and had been instructed to give the wrong answer) were placed in a room with the participant and asked to identify the length of a line. The confederates consistently gave the wrong answer, and the participant was asked to give their own answer. The results of the experiments showed that the majority of participants conformed to the group's answer at least once, even if it conflicted with their own perception. The study found that conformity was more likely to occur when the task was ambiguous and the group was unanimous in their response. The findings of the Asch conformity experiments support Deutsch and Gerard's theory of normative and informative influence, which suggests that conformity is influenced by both the desire to fit in with the group and the desire for accurate information. However, the latter study has been criticized for its low ecological validity (the extent to which the results can be generalized to real-world situations) and for its lack of external validity (the extent to which the results can be generalized to other populations).

To test your understanding of the material, try the English quiz below. Scroll down for the Norwegian version of the text.

Normative and informative social influence

Man is born free; and everywhere he is in chains. – Jean Jacques Rousseau, The Social Contract

Solomon Asch's line study (1952, 56), a experimental setup called the Asch paradigm, and Muzafer Sherif's study of the autokinetic effect (1936) are among the most influential studies in social influence in general, and the experiments each formed their own school. It seems necessary to discuss some of the findings in the hundreds of replications in the traditions. The two studies can be interpreted as competing and oxymoronic: Asch's original intention was to refute Sherif's hypothesis. Here, we will argue that the two experiments manipulate each their own component in Deutsch's and Gerrard's theory (1955) of normative and informative influence, and estimate that recent fMRI studies (1995) may quell the dissonance between the two pioneering experiments.

Informative social influence

The need for accurate knowledge is considered one of the fundamental driving forces in humans. In very unclear situations, people turn to each other to orient themselves. According to Berkeley Professor Sara Cheng, the effect of informative influence is especially high in crises, in ambiguous situations, and when it comes from people we consider experts. The question is how strong and deep-seated this influence can be. The left-radical Sherif had served time in a Turkish prison, and knew how the police, among other things, removes all points of reference in the cell during hard interrogations, creating the most ambiguous and unclear situation possible, often generating false confessions. Sherif's hypothesis was that we only see what the norm wants us to see. Based on the principle of evolution, he believed that norms arise precisely to guide us in unclear situations, and that this has been so essential for the survival of the species that socially created norms can change the very perception of reality: an Orwellian dystopia where two plus two equals five.

Sherif's experiment is based on an optical illusion (the autokinetic effect). In an absolutely dark room, a fixed light point can appear to move because the eye loses all reference points, but does not stop moving itself. The participants in the experiment were exposed to a flash of light and then asked to guess how far the light source had moved. When there was only one participant in the room, his estimates gradually converged on his own average; when there were two or more participants, the estimates converged on a common average. The participants therefore used their own or others' estimates as a reference frame when guessing. To control for normative influence (group pressure), the participant was called back and tested alone a few weeks later. All participants then gave approximately the same estimate as the first time. This suggests that the norm had been internalized, and supports Sherif's hypothesis. As the situation in the experiment is not just ambiguous and unclear, but actually an illusion, one can question the external validity of the study. Asch and other critics claimed that Sherif's study did not measure anything of interest to the participants and that the experiment had low ecological validity (the same criticism that was later directed at Asch's study). That the independent variable in the experiment is informative social influence is not as easy to refute.

Normative influence

Asch wanted to falsify Sherif's hypothesis that social influence can change the perception of reality itself. In Sherif's experiment, the task was impossible and the participants did not invest anything of themselves in the result, but what would happen if Asch himself created an experiment with a task that was impossible to fail? Asch's design started with a sleep-inducing simple task given to a control group in which he assumed that conformity would not yield results. The conditions were then varied. The young Asch got a flying start in his career when it turned out that the degree of conformity (change in behavior as a result of real or perceived influence from other people) in the control group was shocking:

”That we have found the tendency to conformity in our society so strong that reasonably intelligent and well-meaning young people are willing to call White Black is a matter of concern. It raises questions about our ways of education and about the values that guide our conduct.”

— S. Asch's evaluation of the line experiment in a Scientific American article from 1955.

The line task consisted of assessing which of three lines was the same length as a fourth. Before the experiment, almost all participants correctly answered the task, but under pressure from Asch's research assistants, who consistently gave incorrect answers, only a quarter correctly answered all the tasks. A quarter conformed in half of the answers. On average, participants gave incorrect answers in about 33% of the presentations. Some participants explained afterwards that they answered incorrectly because they did not want to stand out. An equal number explained that they began to doubt themselves and their own judgment, and a somewhat smaller group explained that they actually saw that the wrong answer was the right one.

From an evolutionary perspective, it is not difficult to imagine that adapting to the group norm has been essential for survival, and that humans therefore have an innate ability, in some cases surpassing complex rationalizations, to create and follow unspoken norms. An example of this is that researchers in the Netherlands wanted to find the optimal marking of a parking space. After trying a number of variations, it turned out that the most effective parking space was completely unmarked. There were also experiments for road markings—this task should not speculate whether it is only conformity that still makes road signs exist in Holland. The strength of normative influence was also made visible when Asch reversed his original design: one helper consistently answered all the line tasks incorrectly in a group of 16 participants. The situation eventually became so absurd that even the experiment leader couldn't help but laugh at the poor assistant.

In later versions of the original study, it was found that the conformity rate does not significantly increase with the introduction of more than three to five assistants, and it seems that the most important thing is that they are perceived as separate sources of information. Three distinct sources have a stronger effect than ten people representing the same perspective. If there is someone else in the group who gives deviant answers (even if they are wrong), the conformity rate drops significantly. Interestingly, it drops even if the deviant answer comes from an assistant with such thick eyeglasses that the research participant has reason to doubt that they can see the lines presented at all. In tasks where there is no specific answer (such as music taste), conformity increases linearly with the number of assistants. The conformity rate also varies greatly across cultures—from about 14% in Belgium to 51% in India. It also appears that the conformity level has generally decreased in the past 50 years.

It can be argued that the Asch paradigm measures normative social influence: participants are not in doubt about what the correct answer is, but yield to the norm of fitting in with the group; but what about those who reported that they saw that the wrong answer was the right one? Deutsch and Gerrard combine Asch's and Sherif's theories to explain their findings in their studies. The background for their experiment was the desire to remove normative influence in the Asch paradigm. Deutsch and Gerrard's hypothesis was that conformity would drop to zero if participants were allowed to give their answers completely anonymously, but despite their efforts, they never got the conformity rate below 23%. They therefore concluded in their Dual Process Theory that it was normative and informative influence that together created conformity.

—Seeing is believing what the group tells you to believe

In 1995 (one year before Asch died), Gregory S. Berns et al published the first conformity experiment using fMRI in Biological Psychiatry. 32 volunteers came and were presented with four accomplices. While they were in a brain scanner, they were asked to rotate three-dimensional objects to find those that were the same, and before they gave their answer, they heard the accomplices' wrong answers. Berns et al had two hypotheses about what would happen: If social conformity was a result of conscious processes, they would see activity in the prefrontal cortex, while they would observe activity in the posterior regions of the brain if conformity stemmed from changes in perception.

When the participants followed the group's wrong answers, the researchers measured activity in brain areas that control spatial processes, and in participants who did not conform, they measured high activity in the amygdala. A very superficial interpretation of this finding is that it appears that both Sherif and Asch were partly right, and that conformity is as complex a phenomenon as anything else the human organism contains.

Normativ og informativ sosialinnflytelse

Solomon Asch’ linjestudie (1952, 56), et eksperimentoppsett som kalles Asch-paradigmet, og Muzafer Sherifs studie av autokinetisk effekt (1936) er blant de mest innflytelsesrike studiene innen sosial påvirkning generelt, og forsøkene dannet hver for seg skole. Det virker nødvendig å drøfte noen av funnene i de mange hundre replikasjonene i tradisjonene. De to studiene kan tolkes som konkurrerende og oksymorone: Asch’ opprinnelige intensjon var å avkrefte Sherifs hypotese. Her vil vi argumentere for at de to eksperimentene snarere manipulerer hver sin komponent i Deutsch’ og Gerrards teori (1955) om normativ og informativ påvirkning, og anslå at nyere fMRI-studier (1995) kanskje stilner dissonansen mellom de to pionerforsøkene.
Man is born free; and everywhere he is in chains. – Jean Jacques Rousseau, The Social Contract

Informativ sosialinnflytelse

Behovet for nøyaktig kunnskap regnes som en av de grunnleggende drivkreftene i mennesket. I svært uklare situasjoner henvender mennesker seg til hverandre for å orientere seg. Effekten av informativ innflytelse er ifølge Berkeley-professor Sara Cheng spesielt høy i kriser, i tvetydige situasjoner og når den kommer fra personer vi anser som eksperter. Spørsmålet er hvor sterk og dyptgripende denne innflytelsen kan bli. Venstreradikaleren Sherif hadde sonet i tyrkisk fengsel, og visste hvordan politiet under harde forhør blant annet fjerner alle holdepunkter i cella for å skape en mest mulig tvetydig og uklar situasjon, noe som ofte genererer falske tilståelser. Sherifs hypotese var at vi kun ser det normen vil at vi skal se. Fundert i evolusjonsprinsippet mente han at normer oppstår nettopp for å veilede oss i uklare situasjoner, og at dette har vært så essensielt for artens overlevelse at sosialt skapte normer kan endre selve persepsjonen av virkeligheten: en orwellsk dystopi hvor to og to er fem.

Sherifs forsøk bygger på en optisk illusjon (autokinetisk effekt). I et absolutt mørkt rom vil et fiksert lyspunkt kunne se ut til å bevege seg fordi øyet mister alle referansepunkter, men ikke selv slutter å bevege seg. Deltakerne i forsøket ble eksponert for et lysglimt og deretter bedt om å gjette hvor langt lyskilden hadde forflyttet seg. Når det bare var én deltaker i rommet konvergerte dennes estimater gradvis mot hans eget gjennomsnitt; når det var to eller flere deltakere konvergerte estimatene mot et felles gjennomsnitt. Deltakerne brukte altså henholdsvis egne eller andres estimater som referanseramme når de gjettet. For å kontrollere for normativ påvirkning (gruppepress), ble deltakeren noen uker senere innkalt og testet alene. Alle deltakerne ga da omtrent samme estimat som første gang. Det antyder at normen var blitt internalisert, og gir støtte for Sherifs hypotese. Ettersom situasjonen i forsøket ikke bare er tvetydig og uklar, men faktisk er en illusjon, kan man stille spørsmål ved den eksterne validiteten i studien. Asch og andre kritikere hevdet at Sherifs studie ikke målte noe av interesse for deltakerne og at eksperimentet hadde lav økologisk validitet (den samme kritikken som senere ble rettet mot Asch’ studie). At den uavhengige variabelen i eksperimentet er informativ sosialpåvirkning, er ikke like enkelt å bestride.

Normativ innflytelse

Asch ønsket å falsifisere Sherifs hypotese om at sosial påvirkning kan endre selve persepsjonen av virkeligheten. I Sherifs eksperiment var oppgaven umulig og deltakerne investerte ikke noe av seg selv i resultatet, men hva ville skje dersom Asch selv lagde et forsøk med en oppgave man umulig kunne bomme på? Asch’ design startet med en søvndyssende enkel oppgave gitt til en slags kontrollgruppe hvori han antok at konformitet ikke ville gi utslag. Deretter skulle betingelsene varieres. Den unge Asch fikk en flyvende start på karrieren da det viste seg at graden av konformitet (endring av atferd som følge av reell eller innbilt påvirkning fra andre mennesker) i kontrollgruppen var rystende:

”That we have found the tendency to conformity in our society so strong that reasonably intelligent and well-meaning young people are willing to call White Black is a matter of concern. It raises questions about our ways of education and about the values that guide our conduct.”
– S. Asch’ vurdering av linjeforsøket i en Scientific American-artikkel fra 1955.

Linjeoppgaven bestod i å bedømme hvilken av tre linjer som var like lang som en fjerde. Før forsøket svarte nesten alle deltakerne korrekt på oppgaven, men utsatt for presset fra Asch’ forskningsassistenter, som konsekvent ga gale svar, svarte bare en fjerdedel korrekt på alle oppgavene. En annen fjerdedel konformerte i halvparten av svarene. I gjennomsnitt avga deltakerne galt svar i omtrent 33 prosent av presentasjonene. En del av deltakerne forklarte i etterkant at de svarte feil fordi de ikke ville skille seg ut. En like stor del forklarte at de begynte å tvile på seg selv og på sin egen vurdering, og en noe mindre gruppe forklarte at de faktisk at det gale svaret var det riktige.

Ut fra evolusjonsperspektivet er det ikke vanskelig å tenke seg at det å tilpasse seg gruppenormen har vært livsnødvendig, og at mennesker derfor har en medfødt evne, som i noen tilfeller overgår komplekse rasjonaliseringer, til å skape og følge uuttalte normer. Et eksempel på dette er at forskere i Nederland ønsket å finne den optimale oppmerkingen av en parkeringsplass. Etter å ha forsøkt en rekke varianter viste det seg at den mest effektive parkeringsplassen var helt markeringsfri. Det ble også gjort eksperimenter for veimarkering – denne oppgaven skal imidlertid ikke spekulere i om det kun er konformitet som gjør at det fortsatt finnes veiskilt i Holland. Styrken i normativ innflytelse ble også synliggjort da Asch speilvendte det opprinnelige designet sitt: Én medhjelper svarte konsekvent feil på alle linjeoppgavene i en gruppe med 16 deltakere. Situasjonen ble til slutt så absurd at ikke en gang forsøkslederen klarte å la være og le av den stakkars assistenten.

I andre utgaver av originalstudien har man funnet at konformitetsgraden ikke øker vesentlig med innføringen av mer enn tre til fem assistenter, og det ser ut til at det viktigste er at de oppfattes som separate informasjonskilder. Tre distinkte kilder påvirker sterkere enn ti personer som representerer det samme perspektivet. Dersom det er en annen i gruppa som gir avvikende svar (selv om det er galt), synker konformitetsprosenten signifikant. Den synker interessant nok selv om det avvikende svaret kommer fra en assistent med så tykke brilleglass at forsøksdeltakeren har grunn til å betvile om denne i det hele tatt skimter de foreviste linjer. I oppgaver hvor det ikke finnes noe konkret svar (for eksempel musikksmak) øker konformiteten lineært med antallet assistenter. Konformitetsgraden varierer også mye på tvers av kulturer – fra omtrent 14 prosent i Belgia til 51 prosent i India. Det ser for øvrig ut til at konformitetsnivået generelt har sunket de siste 50 årene.

Det kan argumenteres for at Asch-paradigmet måler normativ sosialpåvirkning: Deltakerne er ikke i tvil om hva som er det riktige svaret, men gir etter for normen om å tilpasse seg gruppa; men hva da med dem som rapporterte at de så at det gale svaret var det riktige? Deutsch og Gerrard forener Asch’ og Sherifs teorier for å forklare funnene i sine studier. Bakgrunnen for deres eksperiment var ønsket om å fjerne normativ innflytelse i Asch-paradigmet. Deutsch’ og Gerrards hypotese var at konformiteten ville synke til null dersom deltakerne fikk avgi sine svar totalt anonymt, men sine anstrengelser til tross fikk de aldri konformitetsprosenten under 23 prosent. De konkluderte derfor i sin Dual Process Theory med at det var normativ og informativ innflytelse som til sammen skapte konformitet.

– Seeing is believing what the group tells you to believe

I 1995 (ett år før Asch døde) publiserte Gregory S. Berns et al i Biological Psychiatry det første konformitetseksperimentet som benyttet fMRI. 32 frivillige møtte opp og ble presentert for fire medsammensvorne assistenter. Mens de lå i en hjerneskanner fikk de i oppgave å dreie tredimensjonale objekter for å finne de som var like, og før de avga sitt svar fikk de høre medhjelpernes gale svar. Berns et al hadde to hypoteser om hva som ville skje: Dersom sosial konformitet var et resultat av bevisste prosesser ville man se aktivitet i forhjernen, mens man ville observere aktivitet i bakre regioner av hjernen dersom konformiteten stammet fra endringer i persepsjonen.

Når deltakerne fulgte gruppens gale svar målte forskerne aktivitet i hjerneområder som styrer spatielle prosesser, og hos deltakerne som ikke konformerte målte man stor aktivitet i amygdala. En svært overfladisk fortolkning av dette funnet er at det ser ut til at både Sherif og Asch hadde litt rett, og at konformitet er et like komplekst fenomen som noe annet den menneskelige organisme rommer. 


Berns, Gregory, S. (1995). Neurobiological Correlates of Social Conformity and Independence During Mental Rotation.

Bond, R., & Smith, P. (1996). Culture and conformity: A meta-analysis of studies using Asch’s (1952b, 1956) line judgment task. Psychological Bulletin, 119, 111-137.

Podcast ov lecture (psych160) at UC Berkely (


  1. What was the main goal of Asch's study?
    A) To test the influence of the group on an individual's perception of reality
    B) To prove that social influence can change an individual's behavior
    C) To demonstrate the importance of normative influence in shaping behavior
    D) To falsify Sherif's hypothesis that social influence can alter our perception of reality

  2. What was the task given to the control group in Asch's study?
    A) To identify which of three lines was the same length as a fourth line
    B) To guess how far a light source had moved in a dark room
    C) To rate the attractiveness of a series of faces
    D) To taste test different brands of soda and determine which was the most popular

  3. How did the rate of conformity in the control group change when confederates were introduced?
    A) It remained the same
    B) It increased significantly
    C) It decreased slightly
    D) It fluctuated randomly

  4. What factors contributed to the influence of the group in Asch's study?
    A) The task was ambiguous and the group was unanimous in their response
    B) The confederates were perceived as experts in the field
    C) The participants were in a crisis or ambiguous situation
    D) All of the above

  5. What was the main focus of Deutsch and Gerard's theory of normative and informative influence?
    A) The influence of the group on an individual's perception of reality
    B) The influence of the group on an individual's behavior
    C) The influence of norms on an individual's perception of reality
    D) The influence of norms on an individual's behavior

  6. What was the hypothesis of Deutsch and Gerard's experiment?
    A) That normative influence would be eliminated if participants were allowed to give their answers anonymously
    B) That normative influence would be eliminated if participants were given a clear and unambiguous task
    C) That normative influence would be eliminated if the group consisted of more than five individuals
    D) That normative influence would be eliminated if the group was not unanimous in their response

  7. What was the result of Deutsch and Gerard's experiment?
    A) Normative influence was eliminated and conformity rates dropped to zero
    B) Normative influence was reduced but still present, with conformity rates dropping to 23%
    C) Normative influence remained unchanged, with conformity rates remaining at 50%
    D) Normative influence increased, with conformity rates rising to 75%

  8. How does the fMRI study conducted by Berns et al. in 1995 relate to the theories of normative and informative influence?
    A) It supports the idea that normative influence is a result of conscious processes
    B) It supports the idea that normative influence is a result of changes in perception
    C) It supports the idea that both normative and informative influence contribute to conformity
    D) It supports the idea that conformity is driven by evolutionary forces

  9. What was the main finding of the fMRI study conducted by Berns et al. in 1995?
    A) That normative influence is driven by conscious processes in the frontal lobes
    B) That normative influence is driven by changes in perception in the posterior regions of the brain
    C) That both normative and informative influence contribute to conformity
    D) That conformity is driven by evolutionary forces in the amygdala

  10. What is the main criticism of Asch's study?
    A) Its low ecological validity
    B) Its lack of external validity
    C) Its use of confederates
    D) All of the above

 Correct answers

  1. D
  2. A
  3. B
  4. A
  5. D
  6. A
  7. B
  8. C
  9. C
  10. D



What is Blockchain?

Bitcoin and Blockchain – A New Backbone for Human Trust

MBT – Gyldendal

MBT Mentaliseringsbasert terapi av Sigmund Karterud, Espen Folmo og Mickey Kongerslev, Gyldendal

Mentalizing and MBT

A movie about Mentalization-based treatment (MBT) featuring Anthony Bateman and Espen Folmo


LOOK UP — The science of cultural evolution