Table 2.
Results of Bayesian linear mixed effect models.
| Variables | Type of sentences | Kinds of concepts | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Abstract | Concrete | EMSS | PS | PSTQ | Animal | Food | Tool | ACs vs. CCs | 95% CI | PP | |
| Uncertainty expressions | 0.24 | 1.81 | 0.76 | 0.69 | 0 | 0.3 | 0.6 | [0.2, 1.1] | ACs > CCs, 99 | ||
| Number of questions | 0.38 | 0.55 | 0.50 | 0.57 | 0.45 | 0.41 | 0.0 | [− 0.04, 0.04] | ACs > CCs, 50 | ||
| Number of target word repetition | 2.5 | 5.1 | 2.5 | 6.2 | 5.4 | 1 | − 0.8 | [− 1.9, 0.1] | ACs > CCs, 3.6 | ||
| Turn-taking | 39.1 | 58.2 | 51.8 | 58.4 | 44.8 | 41.5 | 1.5 | [− 1.4, 4.3] | ACs > CCs, 83.6 | ||
| 1st or 2nd point of view | 40.1 | 35.3 | 32.8 | ||||||||
| 1st & 2nd point of view | 10. 2 | 5.8 | 7.5 | ||||||||
| 1st point of view | 9 | 18.7 | 7.4 | ||||||||
| 3rd point of view | 0.72 | 0.13 | 0.48 | 0.6 | 0.25 | 0.13 | 0.11 | [− 0.2, 0.5] | ACs > CCs, 75.6 | ||
| General statements | 8.5 | 10.6 | 6.9 | 3.3 | 1.7 | 2.3 | 6.2 | [4.7, 7.9] | ACs > CCs, 100 | ||
| Number of evoked contexts | 5.5 | 5 | 5 | 4 | 4.5 | 4.4 | 0.87 | [− 0.22, 1.95] | ACs > CCs, 93 | ||
| Vision | 0.5 | 0.5 | 1.1 | 7.8 | 5.9 | 7. 1 | 6.2 | [5, 7.4] | CCs > ACs, 100 | ||
| Touch | 0 | 0 | 0.1 | 0.8 | 0.1 | 0. 72 | 0.5 | [0.2, 1.0] | CCs > ACs, 100 | ||
| Hearing | 0.1 | 0 | 0 | 0.5 | 0 | 0.6 | 0.3 | [0.08, 0.7] | CCs > ACs, 99.9 | ||
| Taste | 0 | 9.9 | 9.9 | [8.7, 11.2] | CCs > ACs, 100 | ||||||
| Smell | 0 | 0.12 | 0.1 | [4.5e−5, 3.04e−3] | CCs > ACs, 99.6 | ||||||
| Materials | 0 | 0 | 0 | 0.7 | 6.3 | 3.5 | 3.47 | [2.5, 4.5] | CCs > ACs, 100 | ||
| Space | 1 | 0.2 | 0.6 | 7.6 | 0.9 | 7 | 4.6 | [3.4, 5.9] | CCs > ACs, 100 | ||
| Time | 2.1 | 3.4 | 4 | 2.9 | 8.9 | 5.9 | − 2.6 | [− 3.8, − 1.5] | ACs > CCs, 0 | ||
| Events | 1.0 | 0.3 | 0.6 | 2.6 | 1.8 | 3.5 | − 2 | [− 2.9, − 1.2] | ACs > CCs, 0 | ||
| Concrete actions | 2.3 | 3.3 | 8.2 | 27.8 | 27.6 | 36.6 | 26 | [23.4, 28.9] | CCs > ACs, 100 | ||
| Abstract actions | 6.3 | 11.1 | 6.6 | 0.5 | 1 | 1.2 | 7.14 | [5.6, 8.8] | ACs > CCs, 100 | ||
| Interoception | 0.2 | 0 | 0.1 | 0.1 | 1.9 | 0.4 | − 0.6 | [− 1, − 0.3] | ACs > CCs, 0 | ||
| Emotions | 11. 7 | 3.6 | 3.8 | 4 | 1.5 | 2.3 | 3.8 | [2.6, 5] | ACs > CCs, 100 | ||
| Metacognition | 0.63 | 0.04 | 0.6 | [0.2, 1.1] | ACs > CCs, 100 | ||||||
| Beliefs | 13.6 | 14.8 | 10.8 | 4.4 | 3.1 | 5.9 | 8.6 | [6.9, 10.6] | ACs > CCs, 100 | ||
| Introspection | 1.4 | 2 | 1.1 | 1 | 0.3 | 07 | 0.8 | [0.3, 1.4] | ACs > CCs, 99.9 | ||
| Associations | 1.4 | 1.4 | 2.2 | 0.7 | 2.1 | 0.8 | 0.5 | [0, 1.1] | ACs > CCs, 95.8 | ||
| Subordinates | 1.7 | 2.6 | − 0.8 | [− 1.6, 0.1] | ACs > CCs, 100 | ||||||
| No Perceptual evaluations | 9.1 | 7.3 | 1.8 | [0.3, 3.4] | ACs > CCs, 99 | ||||||
| Why-Questions | 4.9 | 1.9 | 9.7 | 3.5 | 1 | 11 | 0.2 | [− 0.9, 1.4] | ACs > CCs, 68 | ||
| Who-Questions | 12.2 | 1.32 | 0.88 | 0.99 | 0.55 | 0.88 | 4.7 | [3.5, 5.8] | ACs > CCs, 100 | ||
| What-Questions | 14.13 | 6.89 | 7.2 | [5.4, 9] | ACs > CCs, 100 | ||||||
| Where-Questions | 0.8 | 0.2 | 0 | 20.4 | 3 | 4.6 | 8.9 | [7.6, 10.2] | CCs > ACs, 100 | ||
| When-Questions | 0.4 | 0.2 | 0 | 1 | 1.4 | 0.5 | 0.7 | [0.3, 1.2] | CCs > ACs, 100 | ||
| How-Questions | 8.3 | 11.2 | 7.5 | 5 | 17.7 | 2.8 | − 0.5 | [− 2, 1] | CCs > ACS, 25.9 | ||
We showed the main probability of conditional effects according to the type of sentence (abstract, concrete) or kind of concepts (i.e., emotional EMSS, philosophical PS, quantitative PSTQ, Animals, Food, Tool) in each variable. We report the percentage of all variables for each kind of sentence, with the exception of the variables “number of questions” and “numbers of evoked contexts” where the count frequency is reported. This discrepancy is due to the fact that for the variables “number of questions” and “number of evoked contexts” we did not have a total numerical reference value, while for the others we had it, consequently we reported the percentages. For each variable, we report the comparative analysis between Abstract and Concrete sentences (ACs vs. CCs) and the 95% Credible Intervals (CI) and the probability of posterior observations major to zero (PP > 0) reported in percentage terms according to our hypothesis. The PPs values highlighting meaningful differences are in bold.