Table 4. Social learning and optimal choices from experiment 2.

Results are from logistic regressions that model social learners choosing the social learner optimum. Independent variables include an index for the 20 learning blocks in each session, a dummy specifying if left was the optimal urn for social learners, a dummy indicating concordance between demonstrators and social learners, a dummy indicating if a majority of demonstrators chose the demonstrator optimum, and the interaction between these last two dummies. Robust standard errors clustered on social learner account for any correlations due to multiple observations per social learner [48, 49], and we show these robust clustered standard errors in parentheses.

Parameter	Prior	No prior
Intercept	0.745 (0.468)	0.297 (0.453)
Trial block	0.006 (0.012)	0.003 (0.012)
Left optimal (soc)	0.114 (0.307)	-0.413 (0.228)
Concordant	-1.007 (0.762)	-1.403 (0.780)
Majority demo opt	-1.192* (0.517)	-0.276 (0.521)
Concordant × Majority demo opt	1.623 (0.956)	1.702 (0.919)

*** (0.001)

** (0.01)

* (0.05)