Supplemental Materials to Visual Processing is Biased in Peripersonal Foot Space

For each of our four experiments, we provide summary data and analyses that incorporate the factor of cue validity.

EXPERIMENT 1

Mean reaction times were calculated to targets on either side of the monitor under each of the three foot-placement conditions for both valid and invalid trials and are displayed in Figure S1.

Figure S1. Results from Experiment 1. Error bars represent ± 1 SEM.

A 3x2x2 within-subjects ANOVA with factors of foot placement (right, left, or no foot on display), cue validity (valid vs. invalid), and target side (target near vs. far from foot) yielded significant main effects of validity (F(1,42)=87.375, p<.001, ηp2=.675) and target side (F(1,42)=8.334, p=.006, ηp2=.166), and significant interactions between foot placement and validity (F(2,84)=3.615, p=.031, ηp2=.079) and foot placement and target side (F(2,84)=3.136, p=.049, ηp2=.069). Although the overall difference between reaction times for valid versus invalid trials was greater in the foot conditions than the control condition, critically, the three-way interaction between foot placement, validity, and target side was not statistically significant (F(2,84)=1.398, p=.253), a result that does not support the idea that cueing magnitude differs in near versus far foot space. No other main effects or interactions were statistically significant (all p-values > .15). Bonferroni-adjusted pairwise comparisons indicated that participants were significantly faster to detect targets appearing near the right foot than far from the right foot in the valid cue conditions (t(42)=3.165, p=.018). All other pairwise comparisons were not significant (p-values > .1).

EXPERIMENT 2

Mean reaction times were calculated to targets on either side of the monitor under each of the three foot-placement conditions for both valid and invalid trials and are displayed in Figure S2.

Figure S2. Results from Experiment 2. Error bars represent ± 1 SEM.

A 3x2x2 within-subjects ANOVA with factors of foot placement (right, left, or both feet on display), cue validity (valid vs. invalid), and target side (target near vs. far from foot) yielded significant main effects of validity (F(1,42)=224.324, p<.001, ηp2=.842) and target side (F(1,42)=14.011, p=.001, ηp2=.250), and a significant interaction between validity and target side (F(1,42)=6.916, p=.012, ηp2=.141). Although the overall difference between trials classified as target far versus target near was greater for invalid than valid trials (potentially due to the overall increase in RTs for invalid trials), critically, the three-way interaction between foot placement, validity, and target side was not statistically significant (F(2,84)=1.249, p=.292). These results are not consistent with an interpretation that cueing magnitude is dependent upon whether targets appear in peripersonal foot space. No other main effects or interactions were statistically significant (all p-values > .15). Bonferroni-adjusted pairwise comparisons indicated that in the right foot condition, participants were significantly faster to detect targets appearing near their foot than far from their foot in both valid (t(42)=2.972, p=.03) and invalid (t(42)=3.226, p=.012) trials (all other p-values > .05). Participants were also marginally faster to detect targets appearing near their right foot than their left foot on invalid trials when both feet were on the display (t(42)=2.972, p=.054). All other pairwise comparisons were not significant (p-values > .2).

EXPERIMENT 3

Mean reaction times were calculated to targets on either side of the monitor under each of the three block placement conditions for both valid and invalid trials and are displayed in Figure S3.

Figure S3. Results from Experiment 3. Error bars represent ± 1 SEM.

A 3x2x2 within-subjects ANOVA with factors of block placement (right, left, or no block on display), cue validity (valid vs. invalid), and target side (target near vs. far from block) yielded significant main effects of block placement (F(2,84)=5.219, p=.007, ηp2=.111), validity (F(1,42)=170.882, p<.001, ηp2=.803) and target side (F(1,42)=11.199, p=.002, ηp2=.211), and a significant interaction between validity and target side (F(1,42)=4.209, p=.046, ηp2=.091). Although the difference between reaction times for near versus far targets was somewhat greater for invalid trials than for valid trials, critically, the three-way interaction between block placement, validity, and target side was not statistically significant (F(2,84)=.073, p=.930), a result that does not support the idea that cueing magnitude differs in near versus far block space compared to the control condition. No other interactions were statistically significant (all p-values > .50). Bonferroni-adjusted pairwise comparisons indicated that participants were not significantly faster to detect targets appearing near a block than to detect targets appearing far from a block (all p-values > .1).

EXPERIMENT 4

Mean reaction times were calculated to targets on either side of the monitor under each of the three occluded foot placement conditions for both valid and invalid trials and are displayed in Figure S4.

Figure S4. Results from Experiment 4. Error bars represent ± 1 SEM.

A 3x2x2 within-subjects ANOVA with factors of foot placement (right, left, or no foot on display), cue validity (valid vs. invalid), and target side (target near vs. far from foot) yielded significant main effects of foot placement (F(2,84)=6.027, p=.004, ηp2=.125) validity (F(1,42)=160.033, p<.001, ηp2=.792) and target side (F(1,42)=8.914, p=.005, ηp2=.175), and a significant interaction between validity and target side (F(1,42)=17.139, p<.001, ηp2=.290). Although the overall difference between trials classified as target far versus target near was greater for invalid than valid trials, critically, the three-way interaction between foot placement, validity, and target side was not statistically significant (F(2,84)=.970, p=.383). As with previous experiments, these results are not consistent with an interpretation that cueing magnitude is dependent upon whether targets appear in peripersonal foot space. No other main effects or interactions were statistically significant (all p-values > .15). Bonferroni-adjusted pairwise comparisons indicated that participants were marginally faster to detect targets appearing near their right foot than far from their right foot in the invalid condition (t(42)=2.529, p=.09). All other pairwise comparisons failed to reach significance (p-values > .2).