Religiosity has been present in societies throughout history and several theories hold that religion serves to foster trust and a sense of community within the religious ingroup. In many societies today, it is not uncommon to lack religious beliefs and religion is no longer a natural part of everyday life. The studies included in this thesis investigated how religious groups perceive each other or how generous they are to each other, both in the more secular Sweden and in the more religious USA.Paper 1 examined Swedes' perceptions of atheists and religious people. Specifically, how often they associated atheists or religious people with extreme immoral behaviour by making a conjunction error. Previous studies using the same methodological paradigm have found that more people associate atheists, rather than religious people, with immoral behaviour. We found no significant association between target (atheist or religious person) and conjunction errors, indicating that Swedes do not associate immoral behaviour with atheists to a greater degree than they associate immoral behaviour with religious people. We compared the results to those presented in a previous study and found that the Swedish participants in our study made significantly fewer conjunction errors when the target was an atheist than a sample from the USA. They also made significantly more conjunction errors when the target was a religious person than a Finnish sample and the American sample. The results suggests that anti-atheist bias is lower in Sweden compared to the USA, but anti-religious bias is higher than in both Finland and the USA. However, it is also possible that the type of sample used affected the resultswe recruited participants from social media while the other two samples were solely or mainly student samples. The study shows that the clear anti-atheist bias found in similar studies is not universal.Paper 2 investigated Christians' and atheists' perceptions of Christian, Muslim, and atheist job applicants in four studies, two with Swedish samples (studies 1 and 3) and two with samples from the USA (studies 2 and 4). Participants rated the perceived competence and likeability of a target applicant (Christian, Muslim, or atheist) and a control applicant (with no information about religious affiliation). In the last two studies, participants also specified if they would have hired the target or control applicant. Participants generally rated the control applicant as being more competent (USA) and more likeable (Sweden iv and USA) than the target applicant. Both Christian and atheist participants rated targets with the same religious affiliation higher in likeability than targets from one or both religious outgroups in two of the studies. The only significant difference in competence ratings between the targets were in study 3, where Christians rated Muslims as less competent than Christians. However, Christians' likeability ratings in study 3 did not differ between targets. More atheists in study 3 hired the control applicant than the Chris...