Skip to main content
. 2020 Jul 6;9:e53850. doi: 10.7554/eLife.53850

Figure 8. Larger D values were associated with more gaze shifts from D to HV and more accurate decisions.

(a) A multivariate regression analysis showed that larger D values were associated with attentional capture effects as indexed by more fixations at D (right, purple bar). In addition, larger HV-LV difference was associated with fewer fixations at LV (left, blue bar) and larger total HV+LV values were associated with fewer fixations in general (middle). (b) As D value increased so did gaze shifts between D and HV (right, green bar), while gaze shifts between D and LV decreased (right, blue bar). These effects could not merely be due to more fixations at HV or LV per se because the effects of fixations at HV, LV and D on the gaze shifts were partialled out before testing the relationships between D value and gaze shifts. (c) The effect was directionally specific; larger D values were associated with more D-to-HV shifts and fewer D-to-LV shifts but not the opposite (right, green and blue bars; HV-to-D or LV-to-D shifts). (d) In turn, more D-to-HV shifts and fewer D-to-LV shifts predicted greater decision accuracy. FixHV fixation at HV; FixLV fixation at LV; FixD fixation at D; ShiftHV-LV gaze shift between HV and LV; ShiftHV-D gaze shift between HV and D; ShiftLV-D gaze shift between LV and D. # p<0.1, *p<0.05, **p<0.01, ***p<0.001. Error bars indicate standard error.

Figure 8.

Figure 8—figure supplement 1. The same analysis performed in Figure 8 was performed by including one additional participant and the results remained similar.

Figure 8—figure supplement 1.

A posterior power analysis indicated that 22 participants were required to meet a power of 80% and alpha of 0.05. Initially an inclusion criterion of eye movement data validity >85% resulted in 21 participants, which is one participant less than the required number. Then, the inclusion criterion was relaxed to 70% to include one additional participant (i.e. n = 22). The results remained similar to those reported in Figure 8. FixHV fixation at HV; FixLV fixation at LV; FixD fixation at D; ShiftHV-LV gaze shift between HV and LV; ShiftHV-D gaze shift between HV and D; ShiftLV-D gaze shift between LV and D. *p<0.05, **p<0.01, ***p<0.001. Error bars indicate standard error.