Figure 8.
Average sensitivity to social influence 〈S〉 against the number of shared estimates τ, in the Sorted (black), Unsorted (blue) and Aggregated (red) treatments, when the average social information M is higher than the personal estimate Xp (D = M − Xp > 0; squares) and when it is lower (D < 0; triangles). Subjects follow the social information more when M is higher than Xp, than when it is lower. Filled symbols represent the data, while empty symbols (with solid lines for squares and dashed lines for triangles) are model simulations. Error bars were computed using a bootstrap procedure described in the Statistical methods, and roughly represent 1 s.e.