Fig. 5

Number of measurements taken at each location during initial strategy selection (magic number heuristic, right column) and average location interval discrepancy (equal spacing heuristic, left column) in expert geoscientist (top row) and novice undergraduate participants (bottom row). Average location interval discrepancy was computed by ordering interval sizes smallest to largest and taking the mean difference of intervals—a score of zero indicates perfectly uniform location intervals, and a threshold of one (dotted vertical line in plots B and D) was used to identify participants who selected non-uniform intervals. Novice undergraduates were randomly assigned to complete either bounded or unbounded versions of the task—participants were limited to 10 measurements per location and selecting from pre-labeled locations, or the number of measurements and location selection was unconstrained. Bounded versus unbounded conditions were included to test the extent to which constraints on measurement and location may have impacted sampling strategy