Summary and New Study Proposal: God Priming and Prosocial Behavior

This article supports the claim that the presence of God concepts and supernatural agents creates an environment for increases in prosocial behavior by providing two studies involving these two variables. Prosocial behavior is defined as the intent to benefit others and society as a whole. This relationship between imagined omnipresent beings or supernatural agents and prosocial behavior has been theorized for many years. A piece of evidence theorists often use for this claim is the state of our societies. We would never have been able to develop such large-scale, functional societies without cooperation between unrelated strangers. The factor behind this cooperation to reach a superordinate goal is thought to be the imagined presence of supernatural agents and their impact on humans.

Previous studies about religiosity and prosocial behavior have yielded mixed results. Some studies have supported the relationship, while others have failed to show that there is a relationship. The experimenters were uncertain of the past research, due to its focus on a correlational relationship between the two variables. The relationship could be due to a third variable (guilt proneness, social desirability bias, etc.). Due to this problem, the experimenters decided to move in a different direction, in which they used priming to create the presence of God concepts, possibly indicating a one-way relationship that could have causal conclusions.

For the first study, the experimenters gathered a sample of 50 college age students from the University of British Columbia in Canada. Within this sample, 26 participants identified being religiously affiliated, while 24 of the subjects were classified as either atheist (n=19) or theists without an organized religion (n=5). These distinctions were made through measures that assessed religious beliefs by requesting identification of religion and a belief-in-God question, which was then quantified. This measurement of religious affiliation and belief was measured after the experiment.

These participants were then randomly assigned to either a religious-prime condition or no-prime condition. This assignment showed the manipulation of the IV, religious priming. This religious priming was introduced in the form of a pretest exercise. The participants in the religious-prime group were asked to unscramble 10 five-word sentences, 5 of these containing religious target words including divine, God, spirit, sacred and prophet.

The procedure involved one of the participants being put into a private room behind closed doors. They were then taken through an anonymous version of the dictator game, a design commonly used in psychology and economics. Following this design, the participants were given 10 one-dollar coins and told they were the giver in an economic decision-making task. The participant was told they could take and keep as many coins as they’d like, and those that were left would be given to the receiver (a confederate), to keep. After this decision was made, the subject answered demographic information and tests to measure religiousness that were mentioned previously.

For the results of this study, the researchers used the amount of money left by the subject as the dependent variable. The researchers also applied descriptive words to the amount such as fairly (leaving exactly $5) and selfishly (leaving nothing). The previous research conducted while using the dictator game showed most givers to act selfishly. Following this trend, subjects in the control condition left an average of $1.84 for the other subject, compared to the average of the religious prime condition of $4.22. The prime-control amount difference for theists and atheists was $1.88 and $2.95, respectively. These values proved to be statistically significant.

The experimenters concluded that religious priming of God concepts did implicitly improve prosocial behavior. This was measured as an increase of amount left by the religious-prime condition. This increase was nonspecific, it affected both religious and nonreligious individuals within the religious-prime condition. The problem the researchers addressed about this study was the sample. Since they used students, the researchers were concerned this study could not be generalized to a heterogeneous and more representative population. This concern would later be addressed in Study 2.

In the second study, the researchers replicated the first study, but made several changes. First, they recruited subjects from the community of Vancouver, Canada. This was meant to provide a more representative sample. Second, the experimenters replaced the no-prime control with a neutral-prime condition. Third, they added a secular-prime condition. This was meant to find whether secular moral beliefs would influence subjects similarly to the religious-prime condition. Finally, the researchers added a dimension in which the subjects reported any awareness that the priming had activated religious thoughts.

The researchers recruited 78 participants from newspaper ads and posters around Vancouver. 3 participants were dropped due to suspicion regarding the study’s hypothesis, so 75 were left to be randomly assigned to the three conditions. The participants ranged in age from 17 to 82 which is diverse compared to the first study, which involved only college age students. Of these 75 participants, 28 reported religious affiliations. 22 were classified as atheists by their own identification and scoring below the midpoint on the belief-in-God test in the first study. The other 25 were classified as theists.

The same procedure in the first study was conducted, but participants were now told they had been randomly selected as givers, so the action they make will happen to affect the next subject. The neutral-prime condition involved unscrambling 10 five-word sentences, but these sentences did not contain any target words that would evoke a significant response. The secular-prime group were given sentences with the words civic, jury, court, police and contract. After these priming tasks, and the participant had made a decision as a giver, they were asked the demographic questions and religious questions described previously. The same IV was manipulated, this time with a change to a condition, and an addition of another level in the experiment.

The findings of this study were relatively similar to the results of the first study. For the dependent variable, the experimenters used the same variable measured in the first study, amount left by the “giver.” The religious-prime condition left an average of $4.56, which was exactly $2 more than the neutral-prime condition. There was a greater variance in the amount of money given, which the researchers attributed to a more representative sample. They also found that there was no statistical difference between theists and atheists in any of the conditions, which contradicts the findings of the first study. Lastly, the experimenters found that the secular-prime condition had a very similar effect on participants to the religious priming, with an average amount left of $4.44 ($1.88 more than the control condition).

The authors also included a suspicion probe at the end of the experiment, which was to test whether the subjects were somewhat aware of the purpose of the priming before the test was run. Most of the subjects were unaware of the priming, which showed that the findings could not be due to subjects’ awareness of the hypothesis or nature of the study (demand characteristics).

The authors concluded that the implicit priming of God concepts and religious terms increased prosocial behavior in both studies. They found it interesting that even if the behavior was anonymous, eliminating any possible repercussions, the experimental groups were still likely to leave more money. Even though the authors were able to support their claim about God concept priming and prosocial behavior with these two studies, they still had several ambiguous findings. Specifically, they focused on the measures recorded for atheists in both studies. In the first study, amount left by atheists was statistically significant and larger than theists, but this discovery proved inconsistent when observed in the second study. In the second study, theists had an average amount left that was larger than atheists, even if it wasn’t significant. The authors ultimately concluded that this was due to a stricter measurement for atheism in the second study. This potentially excluded participants from the first study that should have been labeled atheists.

This article showed two scenarios where religiosity could cause an impact on prosocial behavior. There had been previous correlational studies run with these two variables, but the researchers were able to find a significant, causal relationship. Research had never been conducted by priming participants with God concepts, meaning this research can be used as a template for conducting many more studies in the future. People should care about the results because of the possible real life applications of the information. The results come from a large, heterogeneous sample size, one whose data has supported this phenomenon, meaning there can be further generalization to larger populations. Using this research, theorists can support a genuine argument that this phenomenon truly exists in societies around the world.

For a study that could expand or add to the findings of these researchers, I would create a study where the participants in each group acted as a receiver as well. My research question would still test the effect of God concepts on prosocial behavior, but it will also test whether prosocial behavior will be affected if a variable of antisocial cues and behavior is introduced. I believe that if antisocial cues are introduced, it could affect the prosocial behavior previously recorded in the two earlier studies. This factor would have negative effects on prosocial behavior, because antisocial behavior works in direct contrast with prosocial. Antisocial is defined as covert or overt actions that disrupt the goals of a group. God concepts may have a considerable effect on us and our interactions with others, but the behavior of others is equally as important.

The expected sample would be 100 participants from the community around Indianapolis, Indiana. This allows for a heterogenous and representative sample that will have a high variability to observe a wider variety of responses. The procedure for this proposed study would be similar to the procedure in the first two studies. The participants would be randomly assigned to either a religious-prime condition or neutral-prime condition, acting as the IV. They would then be primed before assuming the role as the giver. These priming conditions would be the same as in the previous studies, using 10 five-word scrambled sentences. Once they have finished their role as the giver, they will then wait 15 minutes in the same room before an experimenter enters the room and asks if they will act as a receiver for one trial, due to several subjects being ill. If they agree, they will be shown 10 more scrambled sentences. These sentences will have similar target words to the original priming words if they are in the religious-prime group, but no significant target words if they are in the control. The participant will then receive either $1 or $5 from a confederate. After receiving the money, they will be asked how much money they would give if they had the opportunity to be the giver again. After answering, they will answer demographic questions about religion and the belief-in-God question used in the first two studies. The $1 and $5 conditions act as IV’s for the second portion of the experiment. For the first portion of the experiment, the DV remains amount left for receiver. For the second portion of study, the DV is estimated amount left if the participant assumed the role of giver again.

For this study, I’d expect to see some differences from the first two studies. The first part of the study would remain relatively the same, since this is a replication of the original studies. For the receiver part of the study, I’d expect participants from both groups who received $1 to leave less money the second time around. For the $5 condition, I think I would see little change in the religious-prime group, but a slight increase in money left for the control group. Religious and nonreligious people will be affected similarly, so even people with religious affiliations from the religious-prime condition will want to give less money.

These results show that when antisocial behavior of a perceived member of society is observed, people are less likely to demonstrate prosocial or helping behavior. It will support my claim that antisocial behavior can negate the effect of supernatural agents on prosocial behavior, or make the effect smaller.

The implications of this study would be considerably important. It could explain sources of aggression and hate within society. Even with cues of supernatural agents or omnipresent beings, a productive member of society could still show hostility towards uncooperative and unproductive members. This could explain hostility and clashes between political parties. If one party labels certain members of society as being uncooperative or unhelpful, they will advocate for the allocation of resources to be limited to productive members who demonstrate prosocial behavior, while another party could advocate equal allocation. In conclusion, God concepts in society could facilitate cooperation and development, but if another member of society is conflicting with the community, others will be less likely to demonstrate prosocial behavior and this could lead to social issues.