Pastoral Psychology

, Volume 66, Issue 1, pp 65–77 | Cite as

Muslims Love Jesus, Too? Corrective Information Alters Prejudices Against Islam

  • Steffen Moritz
  • Anja S. Göritz
  • Simone Kühn
  • Brooke C. Schneider
  • Eva Krieger
  • Jana Röhlinger
  • Sarah Zimmerer


The present study examined whether prejudices towards Islam can be altered through corrective information. A total of 1715 German participants were first asked to appraise their opinions towards Judaism, Christianity and Islam (the ratings pertained to progressiveness, tolerance, peacefulness). Subsequently, questions regarding knowledge about religious topics were posed to participants. Questions were selected to elicit common prejudices pertaining to Islam. The correct answers were then displayed along with detailed explanations. Finally, participants were asked to rate their current opinion towards the three religions once again. Opinions towards Islam were largely negative at baseline but improved significantly after presentation of the correct answers. The present study suggests that prejudices against Islam are partially fueled by knowledge gaps.


Religion Islam Christianity Judaism Prejudices Opinion 

The majority of Muslims have nothing to do with terrorist attacks. One should not forget that. Therefore, we have to think about new labels. I think it is terrible if the Taliban, Boko Haram or Isis are called “radical Islamic.” These people are criminals and they should be named criminals or terrorists. The Serbs who in the Bosnian War raped and murdered Muslims were also not called radical Christians by us.

—Rupert Neudeck (interview, Die Welt, August 9, 2014)

The number of people of the Muslim faith in Western countries is continuously on the rise (Savage 2004). To illustrate, before the 1960s when the German government fostered immigration of Turkish workers, the population of Muslims in Germany was negligible. In 2009, the share of the Muslim population in Germany was estimated at 3.8 million to 4.3 million people; that is, 4.6–5.2 % of the entire German population (Haug et al. 2009). To date, approximately 15–20 million Muslims are living in the European Union, and this population is expected to double by 2025 (Open Society Institute 2010).

Although the majority of the non-Muslim population in the Western hemisphere has limited knowledge about the rituals and religious beliefs of Muslim citizens (Ajzen et al. 2011; Dunn 2005; Hussain 2000; Panagopoulos 2006), Muslims often face resentment due to the implicit association of Islam with violence and terrorism. Although prejudices against Islam were already evident before 2001 (Hussain 2000), attitudes have further deteriorated in the wake of 9/11 and recent attacks by so-called Islamist groups (Fetzer and Soper 2003; Jackson 2007; Panagopoulos 2006). According to a survey of 1110 Americans, Muslims received the lowest favorability ratings among religious groups (Arab American Institute 2014). Similar results emerged in European countries (Pollack 2010) and Canada (Angus Reid Global 2013). Biases against Arab-Muslims are reportedly even higher when tested with implicit rather than explicit measures (Park et al. 2007). Empirical studies as well as polls (ABC News/Washington Post Poll 2008; Jackson 2007; Panagopoulos 2006; Pew Research Center 2003; Pollack 2010) indicate that in Western countries Islam is often perceived as linked to discrimination against women, fanaticism, and a propensity toward violence; there is also concern about loyalty to and compatibility with Western values (Aly 2007; Panagopoulos 2006). To illustrate, according to the aforementioned poll by the Arab American Institute (2014), respondents were divided as to whether Arab Americans and American Muslims, if appointed to a government post, could do the job without their ethnicity or religion influencing their work. Only a minority of respondents thought that Arab Americans who held an important position of influence in the government would be able to carry out their job unbiased. Republican voters, in particular, endorsed concerns about the risk of religious or ethnic influence. The media has been ascribed a major role in echoing and augmenting such preoccupations (Aly 2007; Hervik 2012; Hussain 2000; Jackson 2007; Schneider et al. 2013). According to a poll of 9200 people in Germany, both those with and without a migration background complained that Muslims are portrayed in overly negative ways in the media (Schneider et al. 2013).

Recent civil wars and armed conflicts committed by terrorists in the name of Islam are equated by many non-Muslims with the faith itself and considered an orthodox interpretation of Islam that is a radical but still valid reflection of the faith of Islam (Cesari 2010; Fink 2014; Hervik 2012; Jackson 2007). According to survey data, approximately 40 % of Americans thought the attacks on the World Trade Center and the Pentagon on September 11, 2001, represent the true teachings of Islam (Panagopoulos 2006). Muslims are sometimes regarded by non-Muslims as fanatics who want to defeat liberalism (Aly 2007). In contrast, the Christian religion is predominantly judged as favorable relative to Islam (Angus Reid Global 2013; Pew Research Center 2014), which has also been shown using implicit measures, such as the Implicit Association Test (Rowatt et al. 2005). This schism is problematic, particularly as it is often based on false convictions.

Tensions among religious groups are increasing and are not confined to the Middle East. The present study set out to explore whether filling knowledge gaps (for example, by highlighting shared values/heritage of the three monotheistic religions) may help to alter hostile attitudes and foster understanding towards Islam. To date, the evidence is mixed. A study (Ajzen et al. 2011) examined the relationship between knowledge and pro-Muslim behavior. Although one of their cross-sectional experiments did indeed find a positive correlation between the two, a subsequent study revealed that the association largely depended on underlying opinions.

Authors have begun to look into predictors of Islam-critical opinions (e.g., González et al. 2008; Lee et al. 2009). It appears that older people (ABC News/Washington Post Poll 2008; Angus Reid Global 2013; Arab American Institute 2014; Pew Research Center 2003, 2014) hold more negative views about Islam. People on the right-wing and conservative side of the political spectrum also show more negative views (ABC News/Washington Post Poll 2008; Arab American Institute 2014; Pew Research Center 2003; Rowatt et al. 2005). Higher education has sometimes been linked with holding more favorable views towards Islam (Angus Reid Global 2013; Pew Research Center 2003, 2014). For example, this association was demonstrated among French but not German participants in a 2003 study (Fetzer and Soper 2003). However, no clear picture emerged regarding gender; female gender was associated with more favorable views on Islam among Germans but not French participants (Fetzer and Soper 2003). A newspaper poll identified more critical views among males (ABC News/Washington Post Poll 2008).

The present study pursued three main goals. First, we sought to gather the opinions of a large sample of the German population towards the three monotheistic religions (Judaism, Christianity, and Islam). We predicted that opinions would be least favorable towards Islam, followed by Judaism and Christianity. Second, we expected that we could change these opinions after the presentation of information inconsistent with an image of Islam as intolerant, along with detailed explanations. Third, in case the second goal was attained, we were interested in exploring whether this led to a hindsight bias, that is, the phenomenon that inconsistent information overrides or blends with a prior judgment or estimate (the “knew it all along” effect). Thus, we tested whether participants recollected more positive opinions about Islam than they had previously expressed.

Our approach was inspired by metacognitive training for psychosis (MCT), a therapy program that aims to correct cognitive biases in schizophrenia (a syndrome characterized by severe delusions and other symptoms of reality distortion). MCT aims to “plant the seed of doubt” by luring participants with schizophrenia into high-confident false judgments and then providing corrective information (for an alternative design, see Kunst and Thomsen 2015). A number of studies suggest that MCT ameliorates delusional beliefs (Moritz et al. 2014). Whether such a technique also works for biased—though not pathological per se—views towards religions was tested for the first time in the present study.



Participants from the general population were recruited via WiSo Panel, a German online service providing researchers with the opportunity to advertise scientific studies (for the reliability of data of this and related services, see Göritz 2007, 2014; Judge et al. 2006; Piccolo and Colquitt 2006). A total of 12,049 individuals from diverse backgrounds were invited to participate in the web-based study, which was presented in the German language. Of these, 1789 completed the questionnaire. Blind to results, we discarded data of 74 participants who had either entered the same value throughout the scales (n = 37), had made nonsensical/inadequate entries in one of the open text fields raising questions about the validity of the entries (n = 8) or responded too fast (as measured by timestamps reflecting the duration a participant viewed a particular webpage, n = 29). The final sample consisted of 1715 participants. As an incentive for participation, participants were offered a free downloadable manual containing mindfulness and relaxation exercises at the end of the questionnaire (the relaxation exercises were deemed to be of interest even for a nonclinical population). The research was completed in accordance with the Declaration of Helsinki.

Most participants in the final sample were in their mid-40s (M = 46.8, SD = 14.5) and were more likely to be female (59.2 %) than male. With respect to education, 35.9 % had a formal school education below the 13th grade (13 years of formal school education, so-called Abitur). Approximately half of the sample (48 %) considered themselves to be religious or very religious. With regard to religious affiliation, one third reported having no affiliation with any religion, 61.8 % with Christianity, 9.4 % with Buddhism, 1.3 % with Islam, 1.3 % with Judaism, and 1 % with Hinduism (participants were allowed to make multiple endorsements). A total of 2.3 % said they had very good knowledge of world religions, 19.8 % considered their knowledge good, 66.6 % had some knowledge, and 11.3 % reported no knowledge at all (self-report). Similarly, 2.4 % of the sample endorsed having very good subjective knowledge about the religious background of the Arab-Israeli conflict in the Middle East, whereas 12.1 % endorsed good knowledge, 62.6 % endorsed some knowledge, and 22.9 % claimed to have no knowledge at all.


Questions relating to faith

Participants were asked whether they were religious (yes, very much, a little, not at all) and if they felt close to any of the following religions (sorted in alphabetical order; multiple endorsements were possible): none, Buddhism, Christianity, Hinduism, Islam, Judaism, other/more specific. Participants were also asked about their self-rated knowledge of world religions and their perception of their knowledge about the religious background of the current Arab-Israeli conflict in the Middle East (both ratings: very good, good, some, no knowledge at all).

Pre-test opinions indices

Participants were asked to rate their opinion regarding progressiveness [progressive (=1) vs. regressive (=7)], tolerance [tolerant (=1) vs. intolerant (=7)], and peacefulness [peaceful (=1) vs. militant (=7)] separately for the three monotheistic religions (Judaism, Christianity, Islam) on a 7-point Likert scale, in which 4 demarcated the midpoint (= cannot make a judgment). In total, nine questions were posed (three religions x three opinions). These questions will be referred to as pre-ratings.

Knowledge questions and answers

We posed the 11 questions depicted in Table 1. Participants endorsed one out of four response options and then rated their degree of confidence on a 4-point Likert scale (100 % sure, rather sure, rather unsure, guessing). Following each presentation of the 11 items, the correct answers were shown accompanied by extensive explanations quoting religious (New and Old Testaments, Quran) as well as historical sources. These questions were derived from a literature search and discussions with experts and were aimed to elicit common incorrect prejudices toward Islam.
Table 1

Questions and Response Options with Endorsement in Percentages


Response option

Endorsement (%)

1. Are God (Christianity), Yawheh (Judaism), and Allah (Islam) different gods?

Yes, as different as Jupiter (highest God in ancient Roman religion) and Zeus (highest God in ancient Greek religion), analogous but mythic figures of very different narratives of creation.


Yes, in the holy books God and Allah even fight each other [Islam-critical]


No, all three are one and the same god [correct]


Similar to previous response, but the Hindu God Krishna can also be counted the same as the others


2. Which holy scripture named an entire chapter after Jesus’ mother Mary?

Torah/Old Testament


New Testament


Quran [correct]


None of the above


3. Which religious book requires the veiling of a woman’s hair as God’s command?

New Testament


Old Testament/Torah


Quran [Islam-critical]


Cannot be clearly answered [correct]


4. From which holy scripture has the following verse been taken: “O Children of Israel, remember My favor which I have bestowed upon you and that I preferred you over the worlds”

Quran [correct]


Torah/Old Testament


New Testament


None of the above scriptures


5. Is female genital mutilation a cruel ritual of Islam?

Yes [Islam-critical]


Required by all religions that also stipulate the circumcision of men (i.e., Judaism, Islam)


Is a tribal ritual without clear religious foundation [correct]


This ritual is only performed by the Muslim subgroup of the Sunnites [Islam-critical]


6. Stoning is a punishment of the Sharia (religious law of Islam); the religious commandments of Jews and Christians do not specify this punishment.

Yes, correct [Islam-critical]


No, the Old Testament also recognizes stoning as punishment [correct]


No, only the New Testament stipulates stoning as punishment


None of the three holy books (Old and New Testament, Quran) stipulate stoning for any crime


7. The Bible is more progressive and compatible with democratic beliefs than the Quran, which can be described as anti-scientific and conservative.

Yes [Islam-critical]




Both the Bible and the Quran can be described as anti-scientific and regressive. The only difference is that the biblical view of the world is less observable because of moral decay in Christian societies.


Cannot be clearly answered [correct]


8. Which of the following religions acknowledge Christ as a prophet?

Only Christianity


Christianity and Judaism


Christianity and Islam [correct]


Only Judaism


9. Which epoch is often referred to as the “golden age” of Jewish culture?

The time under Islamic rule (by the “Moors”) in Spain (began ca. 711 CE) [correct]


The Babylonian exile (began 597 BCE)


During the Renaissance (15th–16th centuries), Jewish scriptures regained great importance


The 1920s in Berlin


10. The Torah and the Bible are openly racist and clearly endorse the enslaving of Africans



Only the New Testament is racist, not the Torah


Bible texts have been distorted for this purpose [correct]


The Quran can be called racist, but not the Torah and Bible [Islam-critical]


11. Which holy book advocates the killing of unbelievers or people of different faiths

Quran [Islam-critical]




Quran, Torah, and Bible [correct]


None of the above


Numbers may not add up to 100 % due to rounding

Recollection phase

Participants were asked to recollect their initial ratings on all three religions in terms of progressiveness, tolerance, and peacefulness (see above) using the same scale as before—irrespective of their current opinion.

Post-test opinion indices

Participants were asked to rate their current opinion towards the three monotheistic religions (post-rating), again using the same items and scale as in the pre-rating phase.


We first informed participants that the study dealt with knowledge about religions. No clue was given about the specific rationale of the study (i.e., probing clichés towards Islam; researching the changeability of religious opinions). After this introduction, questions relating to faith were posed (see above) and participants were asked about their opinions of the three main monotheistic world religions. Subsequently, the correct answers to the 11 knowledge questions were presented. Participants were then asked to recollect their initial ratings on all three religions (hindsight bias). Afterwards, they were asked to rate their current opinion towards the three monotheistic religions. On the next page, participants were given links to webpages for further information. Finally, participants could leave comments. On the last page, participants were debriefed about the purpose of the study and could download the relaxation exercises (this incentive was announced at the beginning of the study).


Endorsement of knowledge questions

Table 1 shows the percentage of endorsement for the 11 knowledge questions. The table also highlights response options considered instances of Islam-critical incorrect response options (Islam-critical responses are responses that capture typical—but incorrect—prejudices towards Islam). Correct answers ranged from 69.8 % (Question 1, that the Christian God, Yahweh and Allah are the same) to only 7.9 % (Question 4, that the Quran emphasizes the special rank of the Israelites as the chosen people). Table 2 summarizes the means and standard deviations for the correct and incorrect answers. Eight of the response options were labeled as Islam-critical (i.e., that female genital mutilation is a ritual of Islam); on average, participants endorsed Islam-critical options in 20 % of the cases.
Table 2

Correct and Incorrect Responses on the Religion Quiz, Separated for Response Confidence





 Frequency of correct responses

4.38 (40 %)a


 Frequency of incorrect Islam-critical responses (maximum: 8)

1.58 (20 %)b


100 % confidence

 Frequency of high-confident correct responses

1.09 (25 %)b


 Frequency of high-confident incorrect responses

1.11 (17 %)b


 Frequency of all high-confident responses (correct and incorrect)

2.19 (20 %)a


 Frequency of high-confident incorrect Islam-critical responses

0.43 (27 %)b


100 % confidence or very sure responses

 Frequency of confident correct responses

2.80 (64 %)b


 Frequency of confident incorrect responses

3.34 (50 %)b


 Frequency of all confident responses (correct and incorrect)

6.13 (56 %)a


 Frequency of confident incorrect Islam-critical responses

1.03 (65 %)b


a ratio determined for all 11 items

b ratio determined for the subset of items, answered either correctly or incorrectly by individuals

Opinion towards the three monotheistic religions and change over time

We conducted a 2 X 3 X 3 three-way ANOVA (all within-subject factors) with Time (pre and post test), Opinion (Progressiveness, Tolerance, and Peacefulness) and Faith (Judaism, Christianity, and Islam) as factors. Valence of the Opinion (1 = positive, 7 = negative) served as the dependent variable. To facilitate comprehension, only the results of Faith and Time, as well as their shared interactions, are reported in detail below. The three levels of the factor Opinion revealed quite similar results, and we therefore only report results for the main effect of Opinion and its interactions.

The main effect of Faith reached significance at a high effect size, F(2,1714) = 983.21, p < .001, η2 partial = .37. As can be seen in Fig. 1, opinions towards Islam were overall more negative than those towards Judaism or Christianity; post-hoc paired t-tests were significant (p < .001) for all comparisons (both pre and post combined). No main effect of Time emerged, F(1,1714) = 1.13, p = .29, η2 partial < .001, which, however, was qualified by a significant interaction of Faith X Time, F(2,3428) = 192.39, p < .001, η2 partial = .10, at a medium-to-large effect size. The interaction emerges because opinions towards Islam (marginal means pre: M = 5.22, post: M = 5.13) improved, while opinions towards Christianity (marginal means, pre: M = 3.73, post: M = 3.93) and Judaism (marginal means, pre: M = 4.20, post: M = 4.32) somewhat deteriorated (see Fig. 1); paired t-tests for pre and post scores were significant for all comparisons (at least p = .005) except for peacefulness of Islam, for which a trend emerged towards a more peaceful rating at post test, p = .09. The three-way interaction, although significant, can be considered negligible as the effect size was very small, F(4,6856) = 15.09, p < .001, η2 partial < .001. The test-retest reliability for all nine pairs (pre-post scores of Faith X Opinion) was satisfactory (all rs > .68, mean: r = .71).
Fig. 1

Pre and post judgments. Except for peacefulness, participants largely revised negative opinions towards Islam after being confronted with the correct answers from the quiz. For the other two religions, opinions somewhat deteriorated on several parameters. Higher scores designate more negative opinions

Hindsight bias

We performed a 2 X 3 X 3 three-way ANOVA (all within-subject factors) with Recollection (pre rating, recollected pre rating), Opinion (Progressiveness, Tolerance, Peacefulness) and Faith (Judaism, Christianity, and Islam) as factors. The analyses on hindsight bias are partially redundant with the prior analysis of opinion change as we only substituted the post ratings with the recollected pre ratings. To facilitate comprehension, only data on Faith and Time, as well as its interactions, will be reported on in detail. As before, a negligible main effect of Time emerged, F(1,1714) = 4.67, p = .03, η2 partial < .001, which was qualified by an interaction of Faith X Time, F(2,3428) = 44.49, p < .001, η2 partial = .03, at a small-to-medium effect size (improvement for Islam, marginal means, pre: M = 5.39, post: M = 5.26; Christianity, pre: M = 3.73, post: M = 3.79; Judaism, pre: M = 4.20, post: M = 4.20). The interaction derives from the fact that initial opinions towards Islam were recollected as more favorable than they actually had been (see Fig. 2). In contrast, initial opinions towards the other two religions were recollected in a manner that was not systematically biased; paired t-tests for pre vs. recollected scores were significant for all comparisons (p < .001) except for opinions regarding the intolerance of Judaism and Christianity as well as the peacefulness of Islam. Again, the three-way interaction was significant but very small, F(4,6856) = 12.32, p < .001, η2 partial = .007. Test-retest reliability for all nine pairs (pre-recollected scores of Faith X Opinion) was satisfactory (all rs > .73; mean: r = .80).
Fig. 2

Participants showed a strong hindsight bias. Recollections of the initial opinions were largely biased towards current opinions (see Fig. 1). For Islam, participants recollected their initial opinions as much more positive than it had actually been for Progressiveness and Tolerance. Higher scores designate more negative opinions

Predictors of negative opinions towards Islam

We explored which variables predicted negative opinions towards Islam (mean score for the baseline judgments on progressiveness, tolerance, and peacefulness). We entered all variables collected in the study (religious faith, subjective knowledge on religions, and the religious background of the Arab-Israeli conflict in the Middle East, as well as initial opinions on the three religions (sum score), (high-confident) correct/incorrect responses to knowledge questions, and number of Islam-critical incorrect answers). In a stepwise regression, the following variables showed an independent contribution, R = .31, p < .001: number of incorrect Islam-critical responses (standardized beta: .206), older age (standardized beta: .113), high-confident incorrect responses (beta: .073) and male gender (beta: .064). Approximately 10 % of the variance was predicted by the model.

Predictors of positive opinion change towards Islam

We split the sample into a subgroup of individuals who showed a positive opinion change towards Islam (n = 706) versus those with no change or whose opinion became even more negative (n = 1009; of these, 331 made more negative ratings than before [19.3 %]). Next, we analyzed which variables were associated with positive change. Due to the high number of entered variables and the large sample size, which inflates the chance of finding significant results even with small effects, we adjusted the level of significance to p < .001. Only three variables discriminated the two groups at a small effect size: people whose opinions towards Islam improved had more positive initial opinions towards Judaism, t(1713) = 6.27, p < .001, d = .30, and Christianity, t(1713) = 5.56, p < .001, d = .27, and people with stronger reservations against Islam versus Christianity and Judaism (difference score), t(1713) = 7.16, d = .35.

Regression to the mean

We explored whether regression to the mean could partially account for the observed positive changes in opinions towards Islam. Indeed, baseline opinion scores towards Islam were positively correlated with change scores on opinion (r = .20, p < .001). To follow up on this result, we split the sample into a subgroup with slightly negative baseline opinions (>4 to 5), negative baseline opinions (>5 to 6), and very negative baseline opinions (>6). For all three groups, significant changes were noted (p < .001), resulting in an almost medium effect size for the subgroup with slightly negative opinions (ηpartial 2 = .052) and high effect sizes for the subgroups with negative (ηpartial 2 = .170) and very negative opinions (ηpartial 2 = .177).


The main aims of the present study were to examine the opinions of a large sample of the German population on the three monotheistic religions (Judaism, Christianity, and Islam) and to explore whether these opinions changed following presentation of information that was inconsistent with an image of Islam as intolerant.

As expected from other studies (Arab American Institute 2014) and in line with our first hypothesis, participants’ opinions about Islam were mostly negative, followed by Judaism and Christianity. Islam still received the most negative ratings at post test with respect to progressiveness, tolerance, and peacefulness. However, as predicted in our second hypothesis, ratings were significantly shifted in a positive direction at a very large effect size, particularly for the subgroup that initially had more negative opinions towards Islam relative to the other two religions. Importantly, the change in opinion could not be fully explained by regression to the mean. Moreover, since opinions of the other two religions became more negative, the effect cannot be reconciled by a mere exposure effect (i.e., repeated exposure to an initially neutral and novel stimulus increases positive evaluations of that stimulus) (Zajonc 1968).

Some of the prejudices against Islam may reflect stylistic and linguistic difficulties in translating the Quran into other languages (Ali et al. 2012). For example, the word jihad, which is often translated as “holy war,” more literally means “to struggle” and “to exert” (Ali and Rehman 2005). Depending on the translation, the word can thus have a militant versus a more peaceful notion.

Interestingly, in line with previous experimental research, participants were not good at remembering their exact initial judgment. As predicted (third hypothesis), a hindsight bias was evident for most of the judgments. Participants demonstrated a bias to recollect ratings in the direction of their current (i.e., revised) belief. Unlike most other studies on hindsight bias (Bernstein et al. 2011; Pohl 2004), we analyzed changes in opinion and not in knowledge. Thus, the hindsight bias found in our study likely does not reflect the urge of participants to appear/feel smarter than they are.

The study corroborates earlier evidence showing that many non-Muslims have little knowledge about Islam (Ajzen et al. 2011; Dunn 2005; Hussain 2000; Panagopoulos 2006). To illustrate, only 35.4 % of the participants knew that the Quran acknowledges Jesus as a prophet, and only 11.4 % knew that a chapter in the Quaran is devoted to Mary. In contrast, most participants (69.8 %) correctly answered the question of whether Allah, Yahweh, and God (Christianity) are the same. Perhaps most concerning, 22.5 % associated Islam with female genital mutilation. The following variables emerged as the strongest predictors of negative opinions about Islam: number of incorrect Islam-critical responses, older age, high-confident incorrect responses overall, and male gender. These results are partially in line with prior studies (see introduction).

We would like to note some limitations of the study. This was a single-arm proof-of-concept study exploring whether corrective information could change opinions towards Islam. Since no control group was employed, we can only speculate which of the design aspects fostered change (e.g., items encouraging high-confident responses, passage of time, confrontation with religious topics per se, delayed feedback). Therefore, follow-up studies should focus on single aspects to learn more about the driving mechanisms. Furthermore, our study recruited very few Muslim participants. How the present quiz would work in a sample that includes more individuals of the Muslim faith should be tested, as well as whether the approach could also be applied to correct clichés about Christianity in a Muslim population.

Moreover, future studies are needed to elucidate whether the observed change in opinion is stable over time or how quickly (or whether) participants revert back to their original opinions. As another extension of our work, it would be interesting for future studies to consider not only explicit opinions but also implicit attitudes by using one or more implicit measures (for an example from the domain of preferences, see Gebauer et al. 2012). We would expect a similar direction of effects, but given that explicit and implicit measures tap into somewhat different aspects, interesting insights might be gained. In the framework of metacognitive training for psychosis (MCT), one could also test whether a set-up with delayed feedback, as used in the present study, is superior or inferior to a set-up with immediate feedback. We also suspect that more questions and more corrective information will lead to stronger change. We expect a nonlinear relationship, however, as too much information may reduce salience or even lead to resentment. A paradigm such as ours could also be used for campaigns aimed at raising understanding of other religions, in particular by demonstrating the common roots and heritage of all three monotheistic religions. Such an endeavor may be overdue in view of the many conflicts that are fought for religious reasons or claims. Followers of Christianity and Islam make up nearly half of the world’s population. Therefore, the relations between these two major religions are important for a peaceful world and its future (Zdanowski 2008).

One should be aware, however, that improving the image of one religion might be to the detriment of other religions. For example, in our study, participants gave more progressive/positive ratings on Islam after the knowledge quiz, but ratings on the other two religions turned more negative (for simlar effects see Jorgensen et al. 2013). Future studies may try to counter such influences.

We think that this study has important implications. It has long been known that clichés tend to be resistant to change (Pitz 1969), reminiscent of delusional beliefs. Our metacognitive approach, adopted from psychosis research, fared well. Wrong answers for seemingly simple questions were often given with high confidence: 50 % chose wrong options with 100 % sure or very sure ratings. Although we think that the “aha effect” following the correct responses elicited the significant decrease in negative opinions towards Islam, more rigorous research designs will need to examine this in further detail.


  1. ABC News/Washington Post Poll (2008). Broad skepticism of Islam marks post-9/11 sentiment. Washington, DC: ABC News/Washington Post.Google Scholar
  2. Ajzen, I., Nicholas, J., Sheikh, S., & Gilbert Cote, N. (2011). Knowledge and the prediction of behavior: The role of information accuracy in the theory of planned behavior. Basic and Applied Social Psychology, 33, 101–117.CrossRefGoogle Scholar
  3. Ali, S. S., & Rehman, J. (2005). The concept of Jihad in Islamic international law. Journal of Conflict and Security Law, 10, 321–343.CrossRefGoogle Scholar
  4. Ali, A., Brakhw, M. A., Bin Nordin, M. Z. F., & Ismail, S. F. S. (2012). Some linguistic difficulties in translating the holy Quran from Arabic into English. International Journal of Social Science and Humanity, 2, 588–590.Google Scholar
  5. Aly, A. (2007). Australian Muslim responses to the discourse on terrorism in the Australian popular media. Australian Journal of Social Issues, 42, 27–40.CrossRefGoogle Scholar
  6. Angus Reid Global (2013). Canadians view non-Christian religions with uncertainty, dislike. Vancouver: Angus Reid Global.Google Scholar
  7. Arab American Institute (2014). American attitudes toward arabs and muslims. Washington, DC: Arab American Institute.Google Scholar
  8. Bernstein, D. M., Erdfelder, E., Meltzoff, A. N., Peria, W., & Loftus, G. R. (2011). Hindsight bias from 3 to 95 years of age. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37, 378–391.PubMedPubMedCentralGoogle Scholar
  9. Cesari, J. (2010). Islam is a religion, not a terror ideology, Accessed 22 Nov 2014.
  10. Dunn, K. M. (2005). Australian public knowledge of Islam. Indonesian Journal for Islamic Studies, 12, 1–32.Google Scholar
  11. Fetzer, J. S., & Soper, J. C. (2003). The roots of public attitudes toward state accommodation of European Muslims’ religious practices before and after September 11. Journal for the Scientific Study of Religion, 42, 247–258.CrossRefGoogle Scholar
  12. Fink, S. (2014). Fear under construction: Islamophobia within American Christian zionism. Islamophobia Studies Journal, 2, 26–43.CrossRefGoogle Scholar
  13. Gebauer, J. E., Göritz, A. S., Hofmann, W., & Sedikides, C. (2012). Self-love or other-love? Explicit other-preference but implicit self-preference. PLoS ONE, 7(7), e41789.CrossRefPubMedPubMedCentralGoogle Scholar
  14. González, K. V., Verkuyten, M., Weesie, J., & Poppe, E. (2008). Prejudice towards Muslims in the Netherlands: Testing integrated threat theory. British Journal of Social Psychology, 47, 667–685.CrossRefGoogle Scholar
  15. Göritz, A. S. (2007). Using online panels in psychological research. In A. N. Joinson, K. Y. A. McKenna, T. Postmes, & U.-D. Reips (Eds.), The Oxford handbook of internet psychology (pp. 473–485). Oxford: Oxford University Press.Google Scholar
  16. Göritz, A. S. (2014). Determinants of the starting rate and the completion rate in online panel studies. In M. Callegaro, R. Baker, J. Bethlehem, A. S. Göritz, J. A. Krosnick, & P. J. Lavrakas (Eds.), Online panel research: A data quality perspective (pp. 154–170). Chichester: Wiley.CrossRefGoogle Scholar
  17. Haug, S., Müssig, S., & Stichs, A. (2009). Muslim life in Germany: A study conducted on behalf of the German conference on Islam (Vol. 1). Berlin: Bundesamt für Migration und Flüchtlinge.Google Scholar
  18. Hervik, P. (2012). The Danish Muhammad cartoon conflict. Malmö: Malmö University.Google Scholar
  19. Hussain, M. (2000). Islam, media and minorities in Denmark. Current Sociology, 48, 95–116.CrossRefGoogle Scholar
  20. Jackson, R. (2007). Constructing enemies: ‘Islamic Terrorism’ in political and academic discourse. Government and Opposition, 42, 394–426.CrossRefGoogle Scholar
  21. Jorgensen, O., Backstrom, M., & Bjorklund, F. (2013). Bidirectional correction in social judgments: How a cue to the risk of bias causes more favorable ratings of some groups but less favorable of others. Journal of Social Psychology, 153, 131–148.CrossRefPubMedGoogle Scholar
  22. Judge, T. A., Ilies, R., & Scott, B. A. (2006). Work-family conflict and emotions: Effects at work and at home. Personnel Psychology, 59, 779–814.CrossRefGoogle Scholar
  23. Kunst, J. R., & Thomsen, L. (2015). Prodigal sons: Dual Abrahamic categorization mediates the detrimental effects of religious fundamentalism on Christian-Muslim relations. International Journal for the Psychology of Religion, 25, 293–306.CrossRefGoogle Scholar
  24. Lee, S. A., Gibbons, J. A., Thompson, J. M., & Timani, H. S. (2009). The Islamophobia Scale: Instrument development and initial validation. International Journal for the Psychology of Religion, 19, 92–105.CrossRefGoogle Scholar
  25. Moritz, S., Andreou, C., Schneider, B. C., Wittekind, C. E., Menon, M., Balzan, R. P., & Woodward, T. S. (2014). Sowing the seeds of doubt: Anarrative review on metacognitive training in schizophrenia. Clinical Psychology Review, 34, 358–366.CrossRefPubMedGoogle Scholar
  26. Open Society Institute (2010). Muslims in Europe. A report on 11 EU cities. New York: Open Society Institute.Google Scholar
  27. Panagopoulos, C. (2006). The polls-trends: Arab and Muslim Americans and Islam in the aftermath of 9/11. Public Opinion Quarterly, 70, 608–624.CrossRefGoogle Scholar
  28. Park, J., Felix, K., & Lee, G. (2007). Implicit attitudes toward Arab-Muslims and the moderating effects of social information. Basic and Applied Social Psychology, 29, 35–45.CrossRefGoogle Scholar
  29. Pew Research Center (2003). Religion and politics: Contention and consensus. Washington, DC: Pew Research Center.Google Scholar
  30. Pew Research Center (2014). How Americans feel about religious groups—Jews, Catholics & Evangelicals rated warmly, atheists and Muslims more coldly. Washington, DC: Pew Research Center.Google Scholar
  31. Piccolo, R. F., & Colquitt, J. A. (2006). Transformational leadership and job behaviors: The mediating role of core job characteristics. Academy of Management Journal, 49, 327–340.CrossRefGoogle Scholar
  32. Pitz, G. (1969). An inertia effect (resistance to change) in the revision of opinion. Canadian Journal of Psychology, 23, 24–33.CrossRefGoogle Scholar
  33. Pohl, R. F. (2004). Hindsight bias. In R. F. Pohl (Ed.), Cognitive illusions: A handbook on fallacies and biases in thinking, judgement and memory (pp. 363–378). Hove: Psychology Press.Google Scholar
  34. Pollack, S. (2010). Wahrnehmung und Akzeptanz religiöser Vielfalt [Perception and acceptance of religious diversity]. Münster: Westfälische Wilhelms-Universität Münster.Google Scholar
  35. Rowatt, W. C., Franklin, L. M., & Cotton, M. (2005). Patterns and personality correlates of implicit and explicit attitudes toward Christians and Muslims. Journal for the Scientific Study of Religion, 44, 29–43.CrossRefGoogle Scholar
  36. Savage, T. M. (2004). Europe and Islam: Crescent waxing, cultures clashing. Washington Quarterly, 27, 25–50.CrossRefGoogle Scholar
  37. Schneider, J., Fincke, G., & Will, A.-K. (2013). Muslime in der Mehrheitsgesellschaft: Medienbild und Alltagserfahrungen in Deutschland [Muslims in the majority society: Reflection in the media and everyday experiences in Germany]. Berlin: Sachverständigenrat deutscher Stiftungen für Integration und Migration (SVR) GmbH.Google Scholar
  38. Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9, 1–27.CrossRefGoogle Scholar
  39. Zdanowski, J. (2008). Christianity and Islam. Focus on Oriental Studies, 17, 12–15.Google Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Steffen Moritz
    • 1
  • Anja S. Göritz
    • 2
  • Simone Kühn
    • 3
  • Brooke C. Schneider
    • 1
  • Eva Krieger
    • 1
  • Jana Röhlinger
    • 1
  • Sarah Zimmerer
    • 1
  1. 1.Department of Psychiatry and PsychotherapyUniversity Medical Center Hamburg-EppendorfHamburgGermany
  2. 2.Freiburg UniversityOccupational and Consumer PsychologyFreiburgGermany
  3. 3.Max Planck Institute for Human DevelopmentCenter for Lifespan PsychologyBerlinGermany

Personalised recommendations