Figure 1
A graphical representation of the body parts included in Experiment 1 (A) note that this figure is only for graphical reasons, as it is not standard in its proportions). A heatmap representing the linguistic distances across body parts words in English (B). The positive relationship between semantic and real body parts distances across the six languages tested (C).
Figure 2
The timeline of events in Experiment 2 and Experiment 3. In Experiment 2, participants were shown with the names several pairs of body parts (e.g., hand – nose) and were asked to indicate which one was closer to the eyes or the feet (depending on the experimental condition). Experiment 3 was identical to Experiment 2, but in this case participants were shown images of body parts.
Table 1
Results of the LMM on RTs and of the GLMM on accuracy for Experiment 2.
| FIXED EFFECT | REACTION TIMES | ACCURACY | ||||
|---|---|---|---|---|---|---|
| F-VALUE | NumDF, DenDF | p-VALUE | χ2-VALUE | DF | p-VALUE | |
| ΔBDist | 213.5 | 1,4092 | <.001 | 102.93 | 1 | <.001 |
| ΔLDist | 240.6 | 1,6216 | <.001 | 5.11 | 1 | .02 |
| Reference | 28.1 | 1,11955 | <.001 | 25.72 | 1 | <.001 |
| ΔBDist : Reference | 111.5 | 1,11288 | <.001 | .72 | 1 | .39 |
| ΔLDist : Reference | 74.2 | 1,5527 | <.001 | 2.67 | 1 | .10 |
Figure 3
Plots of the significant effects observed in the statistical model comprising real body and linguistic distances in Experiment 2. For RTs, the higher the distances, the faster participants’ response, with this effect being moderated by the reference point. Specifically, the effect of linguistic distance is stronger in the eyes condition then in the feet condition (A), while for body distance the opposite pattern was observed (B). For accuracy, the higher the distance, the higher the participants’ accuracy for both semantic (C) and body distance (D) predictors.
Figure 4
Examples of images of body parts used in Experiment 3: nose (A), shoulder (B), hand (C) and foot (D). Images were obtained using DAZ3D (Daz Productions, Inc; https://www.daz3d.com/).
Table 2
Results of the LMM on RTs and of the GLMM on accuracy for Experiment 2.
| FIXED EFFECT | REACTION TIMES | ACCURACY | ||||
|---|---|---|---|---|---|---|
| F-VALUE | NumDF, DenDF | p-VALUE | χ2-VALUE | DF | p-VALUE | |
| ΔBDist | 266.12 | 1,1604 | <.001 | 94.53 | 1 | <.001 |
| ΔLDist | 13.36 | 1,587 | <.001 | 2.74 | 1 | .09 |
| Reference | .01 | 1,44 | .92 | 1.56 | 1 | .21 |
| ΔBDist : Reference | 26.69 | 1,1948 | <.001 | .47 | 1 | .49 |
| ΔLDist : Reference | 31.21 | 1,478 | <.001 | .27 | 1 | .60 |
Figure 5
Plots of the significant effects observed in the statistical model comprising real body and linguistic distances in Experiment 3. For RTs, the higher the distances, the faster participants’ response, with this effect being moderated by the reference point. Specifically, the effect of linguistic distance is stronger in the eyes condition then in the feet condition, with the latter being not significant (A), while for body distance the opposite pattern was found (B). For accuracy, no effect was found for the semantic predictor (C), while for the body distance predictor, the higher the distance, the higher participants’ accuracy (D).