Trusting the experts: The domain-specificity of prestige-biased social learning

Prestige-biased social learning (henceforth “prestige-bias”) occurs when individuals predominantly choose to learn from a prestigious member of their group, i.e. someone who has gained attention, respect and admiration for their success in some domain. Prestige-bias is proposed as an adaptive social-learning strategy as it provides a short-cut to identifying successful group members, without having to assess each person’s success individually. Previous work has documented prestige-bias and verified that it is used adaptively. However, the domain-specificity and generality of prestige-bias has not yet been explicitly addressed experimentally. By domain-specific prestige-bias we mean that individuals choose to learn from a prestigious model only within the domain of expertise in which the model acquired their prestige. By domain-general prestige-bias we mean that individuals choose to learn from prestigious models in general, regardless of the domain in which their prestige was earned. To distinguish between domain specific and domain general prestige we ran an online experiment (n = 397) in which participants could copy each other to score points on a general-knowledge quiz with varying topics (domains). Prestige in our task was an emergent property of participants’ copying behaviour. We found participants overwhelmingly preferred domain-specific (same topic) prestige cues to domain-general (across topic) prestige cues. However, when only domain-general or cross-domain (different topic) cues were available, participants overwhelmingly favoured domain-general cues. Finally, when given the choice between cross-domain prestige cues and randomly generated Player IDs, participants favoured cross-domain prestige cues. These results suggest participants were sensitive to the source of prestige, and that they preferred domain-specific cues even though these cues were based on fewer samples (being calculated from one topic) than the domain-general cues (being calculated from all topics). We suggest that the extent to which people employ a domain-specific or domain-general prestige-bias may depend on their experience and understanding of the relationships between domains.

Example screenshots for each stage of the experiment, and each Condition.
Please note, Round 1 is identical for each Condition.
Example of a question for the "Language" topic in Round 1. Players have two possible answers to choose from, or they can select "Ask Someone Else." The number '8' here is the countdown timer, telling players they have 8 seconds left to answer. Each player gets 15 seconds to decide on their first response.
Example of Round 1, after choosing "Ask Someone Else." On this question, 3 out of the other 9 participants answered for themselves. Those three player's scores are displayed to each participant who chose to "ask someone else", as below: Example of for Round 2, Condition A after selecting "Ask Someone Else" Example of Round 2, Condition A after choosing "Times Chosen on a Different Topic" Example of for Round 2, Condition B after selecting "Ask Someone Else" Example of Round 2, Condition B after choosing "Times Chosen Altogether" Example of for Round 2, Condition C after selecting "Ask Someone Else" Example of Round 2, Condition C after choosing "Times Chosen on This Topic" Example of for Round 2, Condition D after selecting "Ask Someone Else" Example of Round 2, Condition D after choosing "Their Player ID"

Model specifications
Full code and analysis scripts are available at https://github.com/lottybrand/Prestige_2_Analysis All models were written according to the Statistical Rethinking course (McElreath 2016) using the Rethinking() package in R.

Exploratory analyses:
All code and data for exploratory analyses are available at: https://github.com/lottybrand/Prestige_2_Analysis In exploratory analyses we found that more copying throughout the quiz does lead to a higher score on the quiz overall (mean estimate: 0.24, 89%CI: [0.20, 0.28]) and that higher asocial quiz score on the quiz overall does lead to a higher prestige score overall (mean estimate: 0.22, 89%CI: [0.14, 0.30]).
The below plot shows the relationship between asocial quiz score and prestige score, comparable to previous results (Brand et al. 2020).
The below plot shows the individual differences in prestige score, comparable to previous results (Brand et al. 2020). This shows that, the vast majority of participants were never copied, or copied under 20 times, but that a handful of participants became extremely prestigious in their group, being copied over 40 times, in each condition except Condition D.