Table 4.

Test statistics on performance related feedback signal parameters. A paired t-test was used on the pre and post-test parameters within each group and a mixed-design analysis of variance was used to test between group differences.

	Within group		Between group
	Experimental	Control
Signal decline
Slope of a linear fitted line based on trials mean (lowest point)	t(11) = 2.099, p =.060	t(7) = .708, p = .502	F(1,18) = .960, p = .340
Mean number of scans until apple picked over trials	t(9) = 3.388, p =.008 **	t(10) = -1.147, p =.278	F(1,19) = 5.219, p =.034 *
Slope based on the decrease (until below ladder) divided by the number of scans	t(11) = 2.861, p =.015 *	t(10) = -1.106, p =.295	F(1,21) = 7.484, p =.012 *
Slope of a linear fitted line based on trials mean (first below ladder).	t(10) = 2.824, p =.018 *	t(6) = -1.582, p =.165	F(1,16) = 8.771, p =.009 **
Signal elevation
Mean number of scans until apple picked over trials	t(10) = 1.687, p = .122	t(6) = -.455, p = .665	F(1,16) = 2.100, p =.167
Slope of a linear fitted line based on trials mean (first above ladder).	t(9) = -2.925, p =.017 *	t(5) = .072, p =.945	F(1,14) = .890, p =.361
Undershoot
Average value of undershoots	t(12) = -.295, p = .773	t(9) = 1.019, p = .335	F(1,21) = 1.321, p =.263
Overshoot
Average value of overshoots	t(10) = .106, p = .918	t(6) = 2.249, p = .066	F(1,16) = 3.043, p =.100

^**Significant at the 0.01 level (2-tailed)

^*Significant at the 0.05 level (2-tailed)