Highly dangerous road hazards are not immune from the low prevalence effect

Cognitive Research: Principles and Implications

Table 1 Results of permutation tests for each individual experiment and quartile

Experiment	LPE (Threshold)	p	LPE Q1	p	LPE Q2	p	LPE Q3	p	LPE Q4	p
Full Feedback	0.30	< 0.001	0.32	< 0.001*	0.2	< 0.001*	0.21	< 0.001*	0.12	< 0.001*
Response Correction	0.28	< 0.001	0.37	< 0.001*	0.26	< 0.001*	0.18	< 0.001*	0.06	0.02
No Feedback	0.13	< 0.001	0.09	0.13	< 0.001	0.99	0.04	0.25	0.06	0.02
Partial Feedback	0.23	< 0.001	0.25	< 0.001*	0.09	0.05	0.16	< 0.001*	0.04	0.05
Lower Prevalence	0.32	< 0.001	0.20	0.011	0.16	0.05	0.36	< 0.001*	0.12	0.09

All LPE measures were calculated such that the more positive the value, the higher the miss rate was under low prevalence compared to high prevalence. The leftmost column shows the overall LPE calculated as the shift in threshold, and the remaining columns show the difference in miss rates between low and high prevalence conditions separately for each hazard quartile (Q1: least hazardous quartile; Q4: most hazardous quartile)
*Significant based on Bonferroni-corrected critical alpha was 0.0025