Table 3.

Results of a repeated-measures ANOVA comparing encoded mapping (Mapping) × hotspot lightness (HSLightness) × color scale (Colors) × legend text position (LegText) for balanced cue images and the same analysis but without hotspot lightness for hotspot-more images (Experiment 3). Degrees of freedom for all tests were (1, 29). **p < .01, ***p < .001. Results of the full analysis (including LegText) are in Supplementary Table S3.

	Balanced cue images			Reliable cue images
Source	F	p	${\eta }_p^2$	F	p	${\eta }_p^2$
Mapping	9.34	0.005**	0.244	14.45	0.001**	0.333
HSLightness	2.06	0.162	0.066
Colors	2.20	0.149	0.070	0.85	0.364	0.029
Mapping * HSLightness	37.50	***	0.564
Mapping * Colors	0.29	0.594	0.010	0.59	0.448	0.020
HSLightness * Colors	0.00	0.951	0.000
Mapping * HSLightness * Colors	0.16	0.691	0.006