[Proposal] Conspiracy Beliefs and Dishonest Behavior


In my recent paper (Alper, 2021), I showed that there are higher levels of conspiracy beliefs in ‚Äúless inclusive‚ÄĚ (weaker rule of law, and lack of equality and press freedom) countries; and the association between inclusiveness and conspiracy beliefs is stronger among the highly educated. This is reasonable, because there is in fact higher likelihood of actual conspiracies in these countries. In other words, conspiracy believers expect others to be dishonest, and sometimes others are in fact dishonest. The accuracy of their expectation is likely to depend on the context, as there is more dishonesty in more corrupt countries (G√§chter & Schulz, 2016; Muthukrishna et al., 2017).

To this date, there has been no research on how accurate conspiracy believers’ expectation of dishonesty is, and whether this accuracy depends on the context. Furthermore, although past research showed that believers of conspiracy theories are more willing to conspire (Douglas & Sutton, 2011) and engage in crime (Jolley et al., 2019), their actual level of dishonesty has not been studied before.


I, Karen Douglas, and Valerio Capraro aim to look into the following:

  • Are conspiracy believers more likely to engage in dishonest behavior?

Past research has showed that conspiracy believers themselves have higher intentions to engage in dishonest behavior (Douglas & Sutton, 2011; Jolley et al., 2019). However, these studies looked into intention, and the current proposed study will be the first one to look into the actual dishonest behavior.

  • Are they more likely to perceive others as dishonest?

Conspiracy believers would be more likely to expect other people to act dishonestly. After all, strong suspicion over other people’s integrity is central to conspiracy beliefs. However, the current study will be the first to investigate conspiracy believers’ perception of other people by using a behavioral paradigm. Thus, for the first time, the situation will not be hypothetical.

  • How accurate are their perception? Do they overestimate others‚Äô dishonesty?

These are also questions that are yet to receive an evidence-based answer. We will, for the first time, investigate this by comparing their predictions and the actual dishonest behavior by others.

  • Is the accuracy of their perception dependent on the context? Are conspiracy believers more right to be suspicious in places with higher likelihood of actual conspiracies (i.e., countries with lack of rule of law)?

Past research shows that people from countries with higher corruption are more likely to act dishonestly (Gächter & Schulz, 2016; Muthukrishna et al., 2017). Thus, we expect that increased suspicion over others would be relatively more warranted in countries with lower rule of law. In other words, conspiracy believers would be relatively more right in their suspicions in these countries.

Based on these questions, we aim to test four hypothesis:

H1: Conspiracy mindset (a generic, generalized conspirational belief) and specific conspiracy beliefs (specific, well-known conspiracy theories like the one on 9/11) will be positively related to dishonest behavior.

H2: Conspiracy mindset and specific conspiracy beliefs would be positively related to expecting others to behave dishonestly.

H3: Conspiracy mindset and specific conspiracy beliefs would be positively related to overestimating others’ dishonesty.

H4: The association proposed in H3 would be weaker in countries with lower rule of law.


Cheating paradigm. A cheating paradigm (Capraro et al., 2019) will be used to measure dishonesty. Participants will be told that they were randomly assigned to one of two groups and they are matched with another participant who is assigned to the opposite group. Then the participants will be provided with a choice: If they honestly report the group they were assigned to, both the participant and the other person will receive £0.25 as a bonus. However, if they report being assigned to the other group, they would get £0.40 while the other person will receive only £0.10. A dummy variable (0 = honest, 1 = dishonest) will be computed.

Expectation regarding others’ dishonesty. Participants will be told that all participants of this study are from their country, and they will be asked to guess what percentage of other participants would act dishonestly (on the same dishonesty paradigm they completed) on a scale ranging from 0% to 100%.

Accuracy of expectations regarding others’ dishonesty . This will be automatically calculated by subtracting the actual dishonesty of others (the mean dishonesty of all participants except for the target participant making the prediction) from the target participant’s prediction.

Conspiracy mindset. The 5-item conspiracy mindset questionnaire (Bruder et al., 2013) will be completed.

Conspiracy beliefs. The 15-item generic conspiracist beliefs scale (Brotherton et al., 2013) will be completed.

Rule of law. Country-level Rule of Law scores will be retrieved from the World Justice Project (2020).

The demographic form including age, sex, and education status information will also be completed.

The analyses including exploratory variables (perceived competence, income, honesty-humility) will be reported separately from the confirmatory analyses (see the preregistration).


The sample will include participants who are nationals of the United Kingdom, Canada, South Africa, or India. All participants will have to have at least 90% approval rate from their previous participation on Prolific.

Assuming a small correlation of .10, power of .80, and an alpha of .05, the minimum number of participants was calculated to be 616. The aim is to collect 650 participants from each country. The total sample size is planned to be 2600.

The study is expected to last for 10 minutes on average. Participants will be compensated with £1.25 (equivalent of £7.50/hour) which amounts to a total of £4,333.33, including the 33% service fee. All participants will have the opportunity to earn a bonus up to £0.50 (see the Design) which amounts to a total of £1,726, including the 33% service fee. Therefore, in total, I am applying for £6,062.33.

The total amount of unclaimed bonus will be equally divided among all participants.


The study is preregistered at https://osf.io/785rk/?view_only=79ebe18bbe74439e82d96a6fcc5f5f6e


All data and materials will be made available at https://osf.io/t7pjr/?view_only=580938685cf2499faaf83c32341340f2

The pre-print version of the paper will be freely available on https://psyarxiv.com/.