Extended Data Fig. 4: Supplementary Study 3b, Minimal Pay Versus Points (Gamification) Condition in the US.

Effects of the minimal monetary incentive (1 cent per rating 10 images; in green) and a non-monetary gamification treatment (1 extra point per rating 10 images; in blue) in the US (N = 537 on Prolific). Panel A shows the central tendency and distribution of effort by incentive condition. The black line within each box represents a median; the red dot shows a mean. Upper and lower bounds show the third and the first quartile, respectively. The whiskers represent the 1.5 times the interquartile range, with black points showing observations outside of this range. The width of each violin corresponds to the frequency of observations at any given number of images rated on the y-axis. The difference in the number of images rated in the two conditions is statistically significant, Welch’s t(446.98) = 4.67, P < 0.001, Mean_difference = 10.61, d = 0.40, 95% CI 6.15 to 15.08. Panel B shows the money advantage—that is, how much more effective the minimal monetary incentive is compared to the gamification condition. Panel C shows the central tendency and distribution of cost-effectiveness (effort per dollar spent) of each incentive. Graph elements are analogous to those in Panel A, with the width of each violin corresponding to the frequency of observations at any given level of cost-effectiveness (effort per dollar spent) rated on the y-axis. The minimal monetary incentive is more cost-effective than the gamification treatment, Welch’s t(471.93) = 4.37, P < 0.001, Mean_difference = 7.15, d = 0.38, 95% CI 3.93 to 10.36. In Panel B, the error bar is a bootstrapped 95% CI for the mean relative difference in the number of images rated in the minimal-monetary-incentive versus social-norm condition.