Effect of religious priming in prosocial and destructive behaviour

• DOI: http://doi.org/10.1111/1468-0106.12293
• Author: Elizabeth Brown,Huan Xie,Jipeng Zhang
• Published date: 01 February 2020
• Date: 01 February 2020

ORIGINAL MANUSCRIPT

Effect of religious priming in prosocial

and destructive behaviour

Jipeng Zhang

1

| Elizabeth Brown

2

| Huan Xie

2

1

Research Institute of Economics and

Management, Southwestern University of Finance

and Economics, Chengdu, China

2

Department of Economics, Concordia University,

Quebec, Canada

Correspondence

Jipeng Zhang, Research Institute of

Economics and Management, Southwestern

University of Finance and Economics,

55 Guanghuacun Street, Chengdu, Sichuan

Province, China 610074.

Email: jpzhang@swufe.edu.cn

Funding information

FQRSC, Grant/Award Number: 2010-NP-133118

Abstract

In this paper, we study the behavioural impact of religious

priming by showing participants religious words in a

scrambled sentence task before a dictator game and a joy-

of-destruction game. We also elicited data on individual

religiosity and religious affiliation using a questionnaire.

Priming religious words significantly increased prosocial

behaviour in the dictator game, and the effect was espe-

cially striking among those reporting no religion, atheists

and agnostics. The religious prime has no significant

effect in mitigating destructive behaviour or own expecta-

tions of the other’s destruction choice, but both destructive

behaviour and expectations correlate positively with the

multi-dimensional religiosity measure.

KEYWORDS

dictator game, joy-of-destruction game, prosocial

behaviour, religious priming

1|INTRODUCTION

Religion has been linked to a range of social and economic decisions as far back as Smith (1776). In

modern research, economists have investigated the relationship between religion and income

(Crabtree, 2010), GDP growth (Barro & McCleary, 2003), labour supply (Granger & Price, 2007)

and other variables. McCleary and Barro (2006) discuss religion’s two-way interaction with the polit-

ical economy, both as a dependent variable and an independent variable. The present paper contrib-

utes to the economics research on religion and prosocial behaviour by soliciting donations to a real

charity, adding the dimension of antisocial behaviour, and using a scrambled sentence task to prime

participants with religious words. Moreover, we collected information on individual religiosity and

religious affiliation in a questionnaire to evaluate the differentiated priming effects among different

religious groups.

Received: 2 August 2017 Revised: 28 July 2018 Accepted: 12 January 2019

DOI: 10.1111/1468-0106.12293

Pac Econ Rev. 2020;25:47–68. wileyonlinelibrary.com/journal/paer © 2019 John Wiley & Sons Australia, Ltd 47

In the previous literature, researchers found that religions tend to espouse prosocial values such

as charity, forgiveness, honesty and tolerance. Theories of religion generally posit that religion fosters

prosocial behaviour (Norenzayan & Shariff, 2008).

1

Many empirical researches have studied the

effects of religious priming on prosocial behaviour (see also the recent meta-analyses by Shariff, Wil-

lard, Andersen, and Norenzayan (2015) and by van Elk et al. (2015)).

2

Different from most previous

studies, subjects in our study do not play a standard dictator game, but an Eckel and Grossman

(1996) dictator game, in which they choose an amount to be donated to a charity instead of an anony-

mous stranger. One research question in our study is to investigate whether religious priming would

affect the decision to donate for a good cause differently compared to the decision to allocate money

to a stranger.

Religions tend to discourage antisocial behaviours such as deception and spite. There is an eco-

nomics literature on negative behaviours (e.g. negative reciprocity, defection in the prisoner dilemma,

free-riding in public goods games and spiteful behaviour in ultimatum games). However, the study of

direct antisocial behaviour that does not arise from retaliation against other negative behaviour (per-

ceived or otherwise) is limited. Abbink and Sadrieh (2009) introduced the joy-of-destruction game in

which two players each receive an endowment and simultaneously decide on how much of the other

player’s endowment to destroy. Again in the joy-of-destruction game, Abbink and Herrmann (2011)

find that approximately one-quarter of participants will destroy half their anonymous partner’s

endowment when the source of the destruction is hidden. To our knowledge, there has not been any

experimental research on the subject of proactive antisocial behaviour and religiosity or religious

priming.

Our study also adjoins the experimental economics literature that has studied the relationship

between self-reported measures of religiosity and prosocial behaviour in various economic games.

Most of these studies have found no significant relationship between religiosity and prosocial behav-

iour.

3

One novel part of our study is that we incorporate the self-reported religiosity measures into

the experimental analysis on the priming effects in prosocial and destructive behaviour, so that we

can shed light on the possible differentiated impact of religious priming on different religious groups

categorized by self-reported religiosity measures.

In summary, we present an experimental study in which half of our participants were primed with

religious words before they made a dictator game and a joy-of-destruction game decision. We

observed that priming with religious words increased prosocial behaviour in the dictator game but

1

An interesting recent experimental study conducted in a Muslim country by Haruvy, Ioannou, and Golshirazi (2018) shows

that observant workers who are in the midst of fasting during Ramadan are far more generous to recipients than workers who

have had their evening meal, but it is reversed when decision-makers are outside Ramadan.

2

Randolph-Seng and Nielsen (2007) found that participants cheated significantly less in an honesty task when they had been

primed with religious words. Pichon, Boccato, and Saroglou (2007) found that priming participants with religious words led to

them taking more charity pamphlets when leaving an experimental session. Malhotra (2010) shows that religious participants

were more likely to keep bidding on Sundays, and Malhotra concludes that Sunday served as a prime that activated religious

norms in religious subjects. Benjamin, Choi, and Fisher (2016) found no significant effect of religious priming on dictator

game giving. Lambarraa and Riener (2012) found that Moroccan subjects gave more in a dictator game when the experiment

was administered in Arabic versus French; the Arabic language served as a prime of Arabic and Muslim identity. Shariff and

Norenzayan (2007) found that priming people with religious words made them more generous in the dictator game.

3

Eckel and Grossman (1996) found no significant difference between religious and non-religious people in the amount of giv-

ing to secular charities. Similar results are reported by Anderson and Mellor (2009) and Ahmed and Salas (2011) for the public

goods game. Using a multi-dimensional measure of religiosity, Tan (2006) finds no influence of religiosity on pro-sociality

measured by dictator and ultimatum games because of independent religiosity dimensions producing counter-working effects

that cancel each other out. In a trust game that reveals a responder’s religiosity score, Tan and Vogel (2008) find that both a

proposer’s trust and a responder’s trustworthiness increase with the responder’s religiosity.

48 ZHANG ET AL.