I’ve just had some new friends over, which had me making a ton of social inferences - what kind of food would they like? Should I serve drinks or not? What kind of conversation topics would be appropriate, while still being interesting enough? Does the house need to be in top shape, or is it OK if they see a few of my daughter’s toys lying around? As we go about our lives and interact with others, we are constantly making these kinds of inferences about others’ thoughts, opinions and beliefs.
How do we come up with answers to these questions? One way would be to use stereotypes - talk with young people about Tik Tok; serve people food according to their home country. Another would be to use self knowledge, taking myself as a guide - ask myself what would I want to eat, drink or talk about, and offer my guests the same. Finally, I could use target specific knowledge - I might know about my friends’ preferences from previous conversations with them, or with people who know them better than me. But what determines which one of these strategies would I use?
A recent paper we published in the British Journal of Social Psychology asks how the way we perceive our relationships to others determines which information sources we rely on when making social inferences about them. In a series of studies conducted at the Princeton Social Neuroscience Lab, led by Diana Tamir, we explored how the extent to which people feel familiar with others, similar to them, or like them influences the way they make inferences. Study design and data collection were led by Tony Phan, with additional work by Mark Thornton and Sara Verosky.
Disentangling relationship types
These studies are by no means the first to examine the connection between these interpersonal dimensions - familiarity, similarity and liking - and the way we make social inferences. For a more comprehensive review see the full paper, but one key theory worth mentioning here is Daniel Ames’ work on similarity, stereotyping and projection (assuming that others think like us, which we refer to as using self knowledge). In a series of papers, Ames and colleagues (2004, 2004, 2012) show that when feel more similar to others, we tend to rely less on stereotypes and more on self-knowledge when making social inferences about them. They also demonstrated that this effect has more to do with similarity than with liking. However, other research has shown that familiarity also tends to increase the use of self-knowledge, and that familiarity, similarity and liking are all associated with increased use of target-specific knowledge. Part of the reason findings might be so similar between these three interpersonal dimensions is that they all tend to go together. We usually like people more when we feel more similar to them and are more familiar with them. These links are causally multidirectional: for example, liking someone can lead us to learn more about them, while the mere exposure effect suggests that familiarizing ourselves with someone will make us like them more. This means that studies that only look at one interpersonal dimension risk finding the effects of another.
The current set of studies addressed that issues in several ways. In four studies, participants were asked to infer the opinions and beliefs of others, and to rate the extent to which they felt familiar with them, similar to them or liked them. In two studies, these were celebrities such as Oprah Winfrey. In another study, these were fictional people, whose opinions were constructed according to known correlations between items on personality questionnaires. In the last study, these were actual acquaintances of the participants. To assess the use of self-knowledge, we also asked participants to rate their own opinions and beliefs. To assess the use of stereotypes, we looked at mean results for targets of specific genders, races and professions.
Findings
When examined separately, our findings followed previous research - familiarity was consistently associated with more use of self-knowledge and target-specific knowledge, while similarity and liking were associated with more use of self-knowledge. However, the more interesting results were revealed when looking at all three interpersonal dimensions together. We found that familiarity was the only dimension associated with increased use of target-specific knowledge when controlling for the other dimensions. This suggests that liking people or feeling that they are similar to us, without familiarizing ourselves with them, will not help us gain specific understanding of their opinions or beliefs. Following previous studies, similarity was uniquely associated with more use of self-knowledge. Finally, liking was uniquely associated with increased reliance on stereotypes.
Beyond the theoretical implications, what does this all mean? As familiarity, similarity and liking tend to go together in most cases, we might expect an increase in all three dimensions to be associated with increased reliance on target-specific and self-knowledge, as well as reduced reliance on stereotype knowledge. In these cases, fine distinctions between the three might not matter that much. However, there are real life occasions in which the distinction might be important. One is that of media figures and celebrities. People might really like a famous actor or musician, or even a niche social media personality or podcaster, without being actually familiar with them. The other is that of “model minorities” - in which people ascribe “positive” stereotypes to social outgroups (A classic example is that of benevolent sexism, in which men perceive women as pure and moral and deserving of protection; Bareket & Fiske, 2023). People might like members of the model minority, without feeling similar to them. Our findings suggest that in both cases, people are likely to rely more on stereotypes when trying to understand famous figures or members of model minorities. This might be part of what causes cases of harassment and other illicit behavior towards these populations.
Ames, D. R. (2004). Strategies for social inference: a similarity contingency model of projection and stereotyping in attribute prevalence estimates. Journal of personality and social psychology, 87(5), 573.
Ames, D. R. (2004). Inside the mind reader’s tool kit: projection and stereotyping in mental state inference. Journal of personality and social psychology, 87(3), 340.
Ames, D. R., Weber, E. U., & Zou, X. (2012). Mind-reading in strategic interaction: The impact of perceived similarity on projection and stereotyping. Organizational Behavior and Human Decision Processes, 117(1), 96-110.
Bareket, O., & Fiske, S. T. (2023). A systematic review of the ambivalent sexism literature: Hostile sexism protects men’s power; benevolent sexism guards traditional gender roles.Psychological Bulletin, 149(11-12), 637–698. https://doi-org.ezproxy.haifa.ac.il/10.1037/bul0000400