University of Groningen Experimental investigations into the semantics of distributive marking Bosnić, Ana

(1)

University of Groningen

Experimental investigations into the semantics of distributive marking

Bosnić, Ana

DOI:

10.33612/diss.171644158

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from

it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date:

2021

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Bosnić, A. (2021). Experimental investigations into the semantics of distributive marking: Data from

Serbian, Korean and Dutch. University of Groningen. https://doi.org/10.33612/diss.171644158

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

A

Appendices

Chapter 2 Appendix A

Results separated by number of objects, for all three main response types:

Distributive answers (left for number three; right for number two):

Collective answers (left for number three; right for number two):

Cumulative (1-to-1) answers (left for number three; right for number two):

Appendix B

The best model fit for distributive answers for both languages:

Model: Answer_Dist ~ Type * Age + Number + (1 + Type | Subjects) + (1 | Items)

Estimate SE Z value P value (Intercept)Null-three-Dutch-8 -1.154 0.665 -1.735 0.082 . Quantifier 7.989 1.112 7.182 <0.0001 *** Serbian-7 -1.070 0.978 -1.094 0.273 Serbian-8 -2.833 1.040 -2.724 0.006 ** Serbian-9 -3.816 1.085 -3.518 0.0004 *** Dutch-9 -1.226 0.928 -1.321 0.186 Number two 0.678 0.173 3.919 <0.0001 *** Quantifier:Serbian-7 -5.509 1.362 -4.044 <0.0001 *** Quantifier:Serbian-8 -2.940 1.370 -2.147 0.031 * Quantifier:Serbian-9 -0.528 1.403 -0.376 0.707 Quantifier:Dutch-9 0.317 1.289 0.246 0.806

Appendix C

The best model fit for collective answers for both languages:

Model: Answer_Coll ~ Type + Age + Number + (1 + Type + Number |Subjects) + (1|Items)

Estimate SE Z value P value (Intercept) Null-three-Dutch-8 -4.829 1.002 -4.817 <0.0001 *** Quantifier -7.634 1.443 -5.292 <0.0001 *** Serbian-7 0.751 1.149 0.654 0.513 Serbian-8 2.714 1.114 2.438 0.015 * Serbian-9 3.623 1.079 3.358 0.0008 *** Dutch-9 2.316 1.010 2.292 0.022 * Number two 2.164 0.453 4.776 <0.0001 ***

Appendix D

The best model fit for cumulative (1-to-1) answers for both languages:

Model: Answer_Cumulative ~ Type * Age + Number + (1 + Type + Number |Subjects) + (1|Items)

Estimate SE Z value P value (Intercept) Null-three-Dutch-8 -1.851 0.662 -2.798 0.005 * Quantifier -6.112 1.479 -4.133 <0.0001 *** Serbian-7 1.894 0.946 2.003 0.045 * Serbian-8 3.190 0.960 3.323 0.0009 *** Serbian-9 0.375 0.965 0.388 0.698 Dutch-9 0.674 0.908 0.743 0.458 Number two -3.121 0.411 -7.601 <0.0001 *** Quantifier:Serbian-7 4.562 1.606 2.851 0.004 ** Quantifier:Serbian-8 3.570 1.606 2.222 0.026 * Quantifier:Serbian-9 3.387 1.549 2.186 0.029 * Quantifier:Dutch-9 0.379 1.564 0.242 0.808

(3)

A

Appendices

Chapter 2 Appendix A

Results separated by number of objects, for all three main response types:

Distributive answers (left for number three; right for number two):

Collective answers (left for number three; right for number two):

Cumulative (1-to-1) answers (left for number three; right for number two):

Appendix B

The best model fit for distributive answers for both languages:

Model: Answer_Dist ~ Type * Age + Number + (1 + Type | Subjects) + (1 | Items)

Estimate SE Z value P value (Intercept)Null-three-Dutch-8 -1.154 0.665 -1.735 0.082 . Quantifier 7.989 1.112 7.182 <0.0001 *** Serbian-7 -1.070 0.978 -1.094 0.273 Serbian-8 -2.833 1.040 -2.724 0.006 ** Serbian-9 -3.816 1.085 -3.518 0.0004 *** Dutch-9 -1.226 0.928 -1.321 0.186 Number two 0.678 0.173 3.919 <0.0001 *** Quantifier:Serbian-7 -5.509 1.362 -4.044 <0.0001 *** Quantifier:Serbian-8 -2.940 1.370 -2.147 0.031 * Quantifier:Serbian-9 -0.528 1.403 -0.376 0.707 Quantifier:Dutch-9 0.317 1.289 0.246 0.806

Appendix C

The best model fit for collective answers for both languages:

Model: Answer_Coll ~ Type + Age + Number + (1 + Type + Number |Subjects) + (1|Items)

Estimate SE Z value P value (Intercept) Null-three-Dutch-8 -4.829 1.002 -4.817 <0.0001 *** Quantifier -7.634 1.443 -5.292 <0.0001 *** Serbian-7 0.751 1.149 0.654 0.513 Serbian-8 2.714 1.114 2.438 0.015 * Serbian-9 3.623 1.079 3.358 0.0008 *** Dutch-9 2.316 1.010 2.292 0.022 * Number two 2.164 0.453 4.776 <0.0001 ***

Appendix D

The best model fit for cumulative (1-to-1) answers for both languages:

Model: Answer_Cumulative ~ Type * Age + Number + (1 + Type + Number |Subjects) + (1|Items)

Estimate SE Z value P value (Intercept) Null-three-Dutch-8 -1.851 0.662 -2.798 0.005 * Quantifier -6.112 1.479 -4.133 <0.0001 *** Serbian-7 1.894 0.946 2.003 0.045 * Serbian-8 3.190 0.960 3.323 0.0009 *** Serbian-9 0.375 0.965 0.388 0.698 Dutch-9 0.674 0.908 0.743 0.458 Number two -3.121 0.411 -7.601 <0.0001 *** Quantifier:Serbian-7 4.562 1.606 2.851 0.004 ** Quantifier:Serbian-8 3.570 1.606 2.222 0.026 * Quantifier:Serbian-9 3.387 1.549 2.186 0.029 * Quantifier:Dutch-9 0.379 1.564 0.242 0.808

(4)

A

Appendix E

1) The best model fit for distributive answers in Serbian - 7-year-olds:

Model_Serbian7: Answer_Dist ~ Type + Number + (1 + Type |Subjects) + (1|Items)

Estimate SE Z value P value (Intercept) svaki-three 0.132 0.990 0.133 0.894 Type null -3.471 1.070 -3.243 0.001 ** Type po -1.778 0.753 -2.362 0.018 * Number two 0.528 0.312 1.693 0.090 .

2) The best model fit for distributive answers in Serbian - 8-year-olds:

Model_Serbian8: Answer_Dist ~ Type + Number + (1 + Type|Subjects) + (1|Item)

Estimate SE Z value P value (Intercept) svaki-three 1.278 1.539 0.831 0.406 Type null -5.478 1.551 -3.532 0.0004 *** Type po -4.304 1.539 -2.796 0.005 ** Number two 1.016 0.353 2.877 0.004 **

3) The best model fit for distributive answers in Serbian - 9-year-olds:

Model_Serbian9: Answer_Dist ~ Type + Number + (1+Type|Subjects)

Estimate SE Z value P value (Intercept) svaki-three 2.090 0.872 2.395 0.017 * Type null -7.082 1.181 -5.995 <0.0001 *** Type po -1.467 0.758 -1.935 0.053 * Number two 0.496 0.319 1.555 0.120

Chapter 3 Appendix A

Experiment 1

Serbian experiment:

Formula: Answer ~ Picture Type + (1 + Picture Type | Subjects)

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type A) 9.033 3.170 2.850 0.001 ** Picture Type B – non-exhausted spaces -1.701 3.427 -0.496 0.620 Picture Type C – non-exhausted monkeys -16.601 3.466 -4.790 0.001 ***

Korean experiment:

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type A) 11.263 5.925 1.901 0.057. Picture Type B – non-exhausted spaces -0.934 6.741 -0.139 0.890 Picture Type C – non-exhausted monkeys -18.917 6.683 -2.830 0.001 **

Experiment 2

Serbian experiment:

Formula: Answer ~ Picture Type + (1 + Picture Type | Subjects) + (1 | Items)

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type B1) 7.779 1.988 3.914 0.001 *** Picture Type A1 – all exhausted 0.311 3.011 0.103 0.918 Picture Type C1 – non-exhausted monkeys

outside cages -11.406 2.573 -4.432 0.001 ***

Korean experiment:

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type B1) 3.220 1.123 2.868 0.004 ** Picture Type A1 – all exhausted 4.312 2.981 1.446 0.148 Picture Type C1 – non-exhausted monkeys

(5)

A

Appendix E

1) The best model fit for distributive answers in Serbian - 7-year-olds:

Model_Serbian7: Answer_Dist ~ Type + Number + (1 + Type |Subjects) + (1|Items)

Estimate SE Z value P value (Intercept) svaki-three 0.132 0.990 0.133 0.894 Type null -3.471 1.070 -3.243 0.001 ** Type po -1.778 0.753 -2.362 0.018 * Number two 0.528 0.312 1.693 0.090 .

2) The best model fit for distributive answers in Serbian - 8-year-olds:

Model_Serbian8: Answer_Dist ~ Type + Number + (1 + Type|Subjects) + (1|Item)

Estimate SE Z value P value (Intercept) svaki-three 1.278 1.539 0.831 0.406 Type null -5.478 1.551 -3.532 0.0004 *** Type po -4.304 1.539 -2.796 0.005 ** Number two 1.016 0.353 2.877 0.004 **

3) The best model fit for distributive answers in Serbian - 9-year-olds:

Model_Serbian9: Answer_Dist ~ Type + Number + (1+Type|Subjects)

Estimate SE Z value P value (Intercept) svaki-three 2.090 0.872 2.395 0.017 * Type null -7.082 1.181 -5.995 <0.0001 *** Type po -1.467 0.758 -1.935 0.053 * Number two 0.496 0.319 1.555 0.120

Chapter 3 Appendix A

Experiment 1

Serbian experiment:

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type A) 9.033 3.170 2.850 0.001 ** Picture Type B – non-exhausted spaces -1.701 3.427 -0.496 0.620 Picture Type C – non-exhausted monkeys -16.601 3.466 -4.790 0.001 ***

Korean experiment:

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type A) 11.263 5.925 1.901 0.057. Picture Type B – non-exhausted spaces -0.934 6.741 -0.139 0.890 Picture Type C – non-exhausted monkeys -18.917 6.683 -2.830 0.001 **

Experiment 2

Serbian experiment:

Formula: Answer ~ Picture Type + (1 + Picture Type | Subjects) + (1 | Items)

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type B1) 7.779 1.988 3.914 0.001 *** Picture Type A1 – all exhausted 0.311 3.011 0.103 0.918 Picture Type C1 – non-exhausted monkeys

outside cages -11.406 2.573 -4.432 0.001 ***

Korean experiment:

Estimate Std. Error z-value Pr (>|z|) Intercept (Picture Type B1) 3.220 1.123 2.868 0.004 ** Picture Type A1 – all exhausted 4.312 2.981 1.446 0.148 Picture Type C1 – non-exhausted monkeys

(6)

A

Appendix B

Examples of two experimental conditions:

Picture C Picture C1

Complete list of items in Serbian (with English translations):

(1) Leti po jedan papagaj. – Po one parrot is flying.

(2) Peva po jedan kanarinac. – Po one canary is singing.

(3) Leti po jedna sova. – Po one owl is flying.

(4) Leži po jedan tigar. – Po one tiger is lying.

(5) Maše po jedan lav. – Po one lion is waving.

(6) Leži po jedan panter. – Po one panther is lying.

(7) Sedi po jedan slon. – Po one elephant is sitting.

(8) Pleše po jedan nilski konj. – Po one hippo is dancing.

(9) Peva po jedan nosorog. – Po one rhino is singing.

(10) Skače po jedan majmun. – Po one monkey is jumping.

(11) Sedi po jedna gorila. – Po one gorilla is sitting.

(12) Pleše po jedna šimpanza. – Po one chimpanzee is dancing.

(13) Maše po jedan medved. – Po one bear is waving.

(14) Skače po jedan konj. – Po one horse is jumping.

(15) Plače po jedna zebra. – Po one zebra is crying.

(16) Plače po jedna žirafa. – Po one giraffe is crying.

Complete list of items in Korean (with English translations):

(1) 앵무새 한 마리씩이 날고 있다. – One parrot ssik is flying.

(2) 카나리아 새 한 마리씩이 노래하고 있다. – One canary ssik is singing. (3) 부엉이 한 마리씩이 날고 있다. – One owl ssik is flying.

(4) 호랑이 한 마리씩이 엎드려 있다. – One tiger ssik lying. (5) 사자 한 마리씩이 손을 흔들고 있다. – One lion ssik is waving. (6) 표범 한 마리씩이 엎드려 있다. – One panther ssik is lying. (7) 코끼리 한 마리씩이 앉아 있다. – One elephant ssik is sitting. (8) 하마 한 마리씩이 춤을 추고 있다. – One hippo ssik is dancing. (9) 코뿔소 한 마리씩이 노래를 하고 있다. – One rhino ssik is singing. (10) 원숭이 한 마리씩이 뛰고 있다. – One monkey ssik is jumping. (11) 고릴라 한 마리씩이 앉아 있다. – One gorilla ssik is sitting.

(12) 침팬지 한 마리씩이 춤을 추고 있다. – One chimpanzee ssik is dancing. (13) 곰 한 마리씩이 손을 흔들고 있다. – One bear ssik is waving.

(14) 말 한 마리씩이 뛰고 있다. – One horse ssik is jumping. (15) 얼룩말 한 마리씩이 울고 있다. – One zebra ssik is crying. (16) 기린 한 마리씩이 울고 있다. – One giraffe ssik is crying.

All experimental material available on:

https://osf.io/m75zk/?view_only=611d9a7e071c4c30a2e4dc572804ab2a

Chapter 4 Appendix A

A1: The best model output – whole population:

Answer ~ Picture * Group + (1 + Picture |Subjects)

Estimate Std. Error Z value Pr(>|z|)

(Intercept) Four - A 1.413 0.581 2.432 0.015 * Four - C -4.915 0.594 -8.279 0.000 *** Four - B -4.232 0.555 -7.632 0.000 *** One - A 6.855 0.796 8.613 0.000 *** One - C -5.671 0.847 -6.693 0.000 *** One - B -5.772 0.844 -6.841 0.000 ***

A2: Tukey contrasts

Results are given on the logit (not the response) scale. Confidence level

used: 0.95; P value adjustment: tukey method for comparing a family of

6 estimates

Contrast Estimate Std. Error Z value P value

Four-A : Four-C 4.9151709 0.5937260 8.279 <.0001 Four-A : Four-B 4.2324117 0.5545881 7.632 <.0001 Four-A : One-A -6.8551681 0.7959551 -8.613 <.0001 Four-A : One-C 3.7314045 0.5542638 6.732 <.0001 Four-A : One-B 3.1496332 0.5252151 5.997 <.0001 Four-C : Four-B -0.6827592 0.5024893 -1.359 0.7517 Four-C : One-A -11.7703391 1.0108700 -11.644 <.0001 Four-C : One-C -1.1837664 0.2891111 -4.095 0.0006 Four-C : One-B -1.7655377 0.4697285 -3.759 0.0024 Four-B : One-A -11.0875799 0.9921515 -11.175 <.0001 Four-B : One-C -0.5010072 0.4539321 -1.104 0.8801 Four-B : One-B -1.0827785 0.2779503 -3.896 0.0014 One-A : One-C 10.5865727 0.9886908 10.708 <.0001 One-A : One-B 10.0048013 0.9766699 10.244 <.0001 One-C : One-B -0.5817713 0.4173874 -1.394 0.7308

A3: The best model output – YES sayers:

Answer ~ Picture + Group + (1 + Picture |Subjects)

(Intercept) Four - A 5.678 1.423 3.991 0.000 ***

Four - C -7.250 1.479 -4.901 0.000 ***

Four - B -6.326 1.426 -4.436 0.000 ***

(7)

A

Appendix B

Examples of two experimental conditions:

Picture C Picture C1

Complete list of items in Serbian (with English translations):

(1) Leti po jedan papagaj. – Po one parrot is flying.

(2) Peva po jedan kanarinac. – Po one canary is singing.

(3) Leti po jedna sova. – Po one owl is flying.

(4) Leži po jedan tigar. – Po one tiger is lying.

(5) Maše po jedan lav. – Po one lion is waving.

(6) Leži po jedan panter. – Po one panther is lying.

(7) Sedi po jedan slon. – Po one elephant is sitting.

(8) Pleše po jedan nilski konj. – Po one hippo is dancing.

(9) Peva po jedan nosorog. – Po one rhino is singing.

(10) Skače po jedan majmun. – Po one monkey is jumping.

(11) Sedi po jedna gorila. – Po one gorilla is sitting.

(12) Pleše po jedna šimpanza. – Po one chimpanzee is dancing.

(13) Maše po jedan medved. – Po one bear is waving.

(14) Skače po jedan konj. – Po one horse is jumping.

(15) Plače po jedna zebra. – Po one zebra is crying.

(16) Plače po jedna žirafa. – Po one giraffe is crying.

Complete list of items in Korean (with English translations):

(1) 앵무새 한 마리씩이 날고 있다. – One parrot ssik is flying.

(2) 카나리아 새 한 마리씩이 노래하고 있다. – One canary ssik is singing. (3) 부엉이 한 마리씩이 날고 있다. – One owl ssik is flying.

(4) 호랑이 한 마리씩이 엎드려 있다. – One tiger ssik lying. (5) 사자 한 마리씩이 손을 흔들고 있다. – One lion ssik is waving. (6) 표범 한 마리씩이 엎드려 있다. – One panther ssik is lying. (7) 코끼리 한 마리씩이 앉아 있다. – One elephant ssik is sitting. (8) 하마 한 마리씩이 춤을 추고 있다. – One hippo ssik is dancing. (9) 코뿔소 한 마리씩이 노래를 하고 있다. – One rhino ssik is singing. (10) 원숭이 한 마리씩이 뛰고 있다. – One monkey ssik is jumping. (11) 고릴라 한 마리씩이 앉아 있다. – One gorilla ssik is sitting.

(12) 침팬지 한 마리씩이 춤을 추고 있다. – One chimpanzee ssik is dancing. (13) 곰 한 마리씩이 손을 흔들고 있다. – One bear ssik is waving.

(14) 말 한 마리씩이 뛰고 있다. – One horse ssik is jumping. (15) 얼룩말 한 마리씩이 울고 있다. – One zebra ssik is crying. (16) 기린 한 마리씩이 울고 있다. – One giraffe ssik is crying.

All experimental material available on:

https://osf.io/m75zk/?view_only=611d9a7e071c4c30a2e4dc572804ab2a

Chapter 4 Appendix A

A1: The best model output – whole population:

Answer ~ Picture * Group + (1 + Picture |Subjects)

(Intercept) Four - A 1.413 0.581 2.432 0.015 * Four - C -4.915 0.594 -8.279 0.000 *** Four - B -4.232 0.555 -7.632 0.000 *** One - A 6.855 0.796 8.613 0.000 *** One - C -5.671 0.847 -6.693 0.000 *** One - B -5.772 0.844 -6.841 0.000 ***

A2: Tukey contrasts

Results are given on the logit (not the response) scale. Confidence level

used: 0.95; P value adjustment: tukey method for comparing a family of

6 estimates

Contrast Estimate Std. Error Z value P value

Four-A : Four-C 4.9151709 0.5937260 8.279 <.0001 Four-A : Four-B 4.2324117 0.5545881 7.632 <.0001 Four-A : One-A -6.8551681 0.7959551 -8.613 <.0001 Four-A : One-C 3.7314045 0.5542638 6.732 <.0001 Four-A : One-B 3.1496332 0.5252151 5.997 <.0001 Four-C : Four-B -0.6827592 0.5024893 -1.359 0.7517 Four-C : One-A -11.7703391 1.0108700 -11.644 <.0001 Four-C : One-C -1.1837664 0.2891111 -4.095 0.0006 Four-C : One-B -1.7655377 0.4697285 -3.759 0.0024 Four-B : One-A -11.0875799 0.9921515 -11.175 <.0001 Four-B : One-C -0.5010072 0.4539321 -1.104 0.8801 Four-B : One-B -1.0827785 0.2779503 -3.896 0.0014 One-A : One-C 10.5865727 0.9886908 10.708 <.0001 One-A : One-B 10.0048013 0.9766699 10.244 <.0001 One-C : One-B -0.5817713 0.4173874 -1.394 0.7308

A3: The best model output – YES sayers:

Answer ~ Picture + Group + (1 + Picture |Subjects)

(Intercept) Four - A 5.678 1.423 3.991 0.000 ***

Four - C -7.250 1.479 -4.901 0.000 ***

Four - B -6.326 1.426 -4.436 0.000 ***

(8)

A

A4: The best model output – NO sayers:

Answer ~ Picture * Group + (1 + Picture |Subjects)

(Intercept) Four - A -2.561 0.372 -6.879 0.000 *** Four - C -5.272 2.658 -1.983 0.047 * Four - B -4.623 2.646 -1.747 0.081 . One - A 5.979 0.658 9.086 0.000 *** One - C -5.011 1.064 -4.707 0.000 *** One - B -5.159 1.150 -4.487 0.000 ***

Appendix B

The best model output - Definiteness follow up with and without po

Answer ~ Picture * Group * Sentence Type + (1 + Picture + Group +

Sentence Type |Subjects)

(Intercept) Four – A with po -1.387 0.520 -2.667 0.008 **

Four – B with po -3.733 0.653 -5.712 0.000 *** One – A with po 8.343 0.906 9.208 0.000 *** Four – A without po 0.197 0.440 0.448 0.654 One – B with po -5.383 0.983 -5.475 0.000 *** Four – B without po 1.487 0.499 2.981 0.003 ** One – A without po -2.923 0.894 -3.269 0.001 ** One – B without po 2.070 1.000 2.070 0.038 *

Appendix C

Homogeneity effects with po

Yes answers – whole population:

Formula: Answer_yes ~ Group + (1 + Group | Subjects)

(Intercept) Negative po -0.8830 0.4271 2.068 0.038672 *

Negative definite 1.8897 0.5667 -3.335 0.000854 ***

Negative svaki 3.2807 0.5366 6.114 0.000 ***

Gap answers – whole population:

Formula: Answer_gap ~ Group + (1 | Subjects)

(Intercept) Negative po -1.2937 0.3565 -3.629 0.000284 ***

Negative definite 0.1581 0.2498 0.633 0.526801

Negative svaki -3.5359 0.4988 -7.089 0.000 ***

Yes answers – population: UQ pattern:

Formula: Answer_yes ~ Group + (1 | Subjects)

(Intercept) Negative po 1.0361 0.2482 4.175 0.000 ***

Negative definite -2.8139 0.3988 -7.057 0.000 ***

Negative svaki 0.8565 0.4077 2.101 0.0356 *

Gap answers – population: UQ pattern:

Formula: Answer_gap ~ Group + (1 | Subjects)

Negative definite 1.4722 0.4367 3.371 0.000748 ***

Negative svaki -2.7797 1.0685 -2.602 0.009280 **

Yes answers – population: Homogeneity pattern:

Formula: Answer_yes ~ Group + (1 | Subjects)

Negative definite -0.6496 0.5213 -1.246 0.213

Negative svaki 5.3840 0.6383 8.435 0.000 ***

Gap answers – population: Homogeneity pattern:

Formula: Answer_gap ~ Group + (1 | Subjects)

(Intercept) Negative po -0.2691 0.5063 -0.532 0.5950

Negative definite -0.7388 0.3436 -2.150 0.0316 *

(9)

A

A4: The best model output – NO sayers:

Answer ~ Picture * Group + (1 + Picture |Subjects)

(Intercept) Four - A -2.561 0.372 -6.879 0.000 *** Four - C -5.272 2.658 -1.983 0.047 * Four - B -4.623 2.646 -1.747 0.081 . One - A 5.979 0.658 9.086 0.000 *** One - C -5.011 1.064 -4.707 0.000 *** One - B -5.159 1.150 -4.487 0.000 ***

Appendix B

The best model output - Definiteness follow up with and without po

Answer ~ Picture * Group * Sentence Type + (1 + Picture + Group +

Sentence Type |Subjects)

(Intercept) Four – A with po -1.387 0.520 -2.667 0.008 **

Four – B with po -3.733 0.653 -5.712 0.000 *** One – A with po 8.343 0.906 9.208 0.000 *** Four – A without po 0.197 0.440 0.448 0.654 One – B with po -5.383 0.983 -5.475 0.000 *** Four – B without po 1.487 0.499 2.981 0.003 ** One – A without po -2.923 0.894 -3.269 0.001 ** One – B without po 2.070 1.000 2.070 0.038 *

Appendix C

Homogeneity effects with po

Yes answers – whole population:

Formula: Answer_yes ~ Group + (1 + Group | Subjects)

(Intercept) Negative po -0.8830 0.4271 2.068 0.038672 *

Negative definite 1.8897 0.5667 -3.335 0.000854 ***

Negative svaki 3.2807 0.5366 6.114 0.000 ***

Gap answers – whole population:

Formula: Answer_gap ~ Group + (1 | Subjects)

Negative definite 0.1581 0.2498 0.633 0.526801

Negative svaki -3.5359 0.4988 -7.089 0.000 ***

Yes answers – population: UQ pattern:

Formula: Answer_yes ~ Group + (1 | Subjects)

(Intercept) Negative po 1.0361 0.2482 4.175 0.000 ***

Negative definite -2.8139 0.3988 -7.057 0.000 ***

Negative svaki 0.8565 0.4077 2.101 0.0356 *

Gap answers – population: UQ pattern:

Formula: Answer_gap ~ Group + (1 | Subjects)

Negative definite 1.4722 0.4367 3.371 0.000748 ***

Negative svaki -2.7797 1.0685 -2.602 0.009280 **

Yes answers – population: Homogeneity pattern:

Formula: Answer_yes ~ Group + (1 | Subjects)

Negative definite -0.6496 0.5213 -1.246 0.213

Negative svaki 5.3840 0.6383 8.435 0.000 ***

Gap answers – population: Homogeneity pattern:

Formula: Answer_gap ~ Group + (1 | Subjects)

(Intercept) Negative po -0.2691 0.5063 -0.532 0.5950

Negative definite -0.7388 0.3436 -2.150 0.0316 *