The emergence and adaptive use of prestige in an online social learning task

Prestige-biased social learning occurs when individuals preferentially learn from others who are highly respected, admired, copied, or attended to in their group. This form of social learning is argued to reflect novel forms of social hierarchy in human societies, and, by providing an efficient short-cut to acquiring adaptive information, underpin the cumulative cultural evolution that has contributed to our species’ ecological success. Despite these potentially important consequences, little empirical work to date has tested the basic predictions of prestige-biased social learning. Here we provide evidence supporting the key predictions that prestige-biased social learning is used when it constitutes an indirect cue of success, and when success-biased social learning is unavailable. We ran an online experiment (n = 269) in which participants could copy each other in real-time to score points on a general-knowledge quiz. Our implementation of ‘prestige’ was the number of times someone had previously been copied by others. Importantly, prestige was an emergent property of participants’ behaviour during the experiment; no deception or manipulation of prestige was employed at any time. We found that, as predicted, participants used prestige-biased social learning when the prestige cue was an indirect cue of success, and when direct success information was unavailable. This highlights how people flexibly and adaptively employ social learning strategies based on the reliability of the information that such strategies provide.

Round 3: 150 ppts, split into three conditions. Conditions A: can copy based on hobbies or number of times copied (in the previous round). Condition B: can copy based on hobbies or number of times copied. Condition C: can copy based on score or number of times copied: Predictions (preregistered at https://osf.io/g3mhj) 1. When participants choose to copy based on score, they will choose the highest scoring participant available to copy from 2. When participants choose to copy based on copy-frequency, they will choose to copy the most-copied participants 3. Participants choose to view the copy-frequency information more in Condition B than the other two conditions because i) in Condition B copiers can access success information, unlike Condition A where copiers only have access to irrelevant information, and ii) in Condition B copy frequency is the only relevant cue available, unlike condition C where direct pay-off information is also available 4. Copying rate will be higher in conditions B & C compared to A because copying can be based on success in Conditions B & C.
5. Participants perform better in Conditions B & C compared to Condition A because copying can be based on success information in Conditions B & C. 7

Analysis:
All data were analysed with multi-level models using Bayesian HMC estimation in the Rethinking package in R (McElreath 2016) and using a model comparison approach. Condition A was modelled as the baseline, thus parameter estimates for Conditions B and C are in comparison to Condition A. All parameter estimates are displayed with the default 89% confidence intervals .

Results:
Was prestige info used more in condition B compared to A & C?
Parameter estimates for likelihood of choosing Prestige information

Condition
Proportion of times they chose prestige/score/hobby information

Likelihood of choosing prestige
Did people copy more in Condition B?
Parameter estimates for copy frequency Proportion of times they chose to copy Did people score higher in condition B & C compared to A?
Parameter estimates for score Participants' scores in each condition 9 Did participants "copy the most copied"?
Did they copy the least copied? Parameter estimates for choosing the most copied Did they copy the highest scorer?
Parameter estimates for choosing the most copied (only available in Condition C) Did they copy the least copied? 10

Pilot 1 Conclusions
• People choose to copy based on prestige less when pay-off information is freely available • No difference between copying based on prestige or copying based on an "uninformative" trait like favourite hobby • People copied very rarely (reported wanting to see how they did on their own, and copying is "bad") • When they did use prestige information, they didn't "copy the most copied"… but they didn't copy at random (avoided the least copied). Equally they didn't choose to copy the highest scorer, but they avoided the lowest scorer.
• Participants weren't using the social information to their advantage (didn't gain a higher score when score information was freely available) • Probably not incentivised enough to use copying to score highly on this task….

Overall Conclusions
• People will choose to copy based on prestige information when success information is costly or unavailable • This is more evident on the abstract task, people get distracted by other characteristics in the knowledge-based task (they may think it's useful even though it isn't) • People don't reliably copy the most copied, but they don't appear to be copying at random either… they avoid copying the least copied.
• People also don't choose to copy the highest scorer available, although they do avoid choosing the lowest scorer available.
• Results should be interpreted with caution: participants reported not wanting to 'copy' in the Qualtrics game, associating it with cheating, and also due to intrinsic enjoyment of quiz, wanting to assess their own individual knowledge. Perhaps incentive for achieving top score was not high enough in the Qualtrics payment scheme to overcome this.

Practice Round
Question: 'Which is the largest of The Canary Islands?',