Table 1
Example items for the grammatical comprehension task.
| STRUCTURE | EXAMPLE | QUESTION (RESPONSE OPTIONS) |
|---|---|---|
| Complex NP | (2) Linda complained that the fact that cycling in the main square is forbidden annoys tourists. | What did Linda complain about? (That tourists are annoyed. / That cycling is forbidden in the main square.) What is forbidden? (Complaining about the cycling restrictions. / Cycling on the main square.) What annoys the tourists? (That one is not allowed to cycle in the main square. / That Linda complained about cycling restrictions.) |
| X-Is-Difficult-to-Answer | (3) James will be easy to persuade Walter to help. | Who might be helped? (James / Walter) What will be easy? (Persuading Walter to help James. / For James to help Walter.) Who will find it easy to do something? (Someone not mentioned in the sentence / James) |
| Reduced Relatives | (4) A child staring at a dog chasing a postman was afraid. | Who was afraid? (A child / A postman) Who chased someone? (A human / An animal) (control) |
| Ditransitives | (5) Mr Peters showed her baby the pictures. (6) Mr Peters showed her the baby pictures. | Who saw something? (A woman / A baby) Who showed someone something? (Mr Peters / Someone not mentioned in the sentence) (control) |
Figure 1
Experimental design.
Table 2
By-participant mean value, standard deviation (SD), median value, interquartile range (IQR, with indication of the first and third quartiles), range (minimum and maximum), split-half reliability (Spearman-Brown corrected reliability estimate) and Cronbach’s coefficient alpha for all measures.
| MEASURE | MEAN | SD | MEDIAN | IQR | RANGE | SPLIT-HALF RELIABILITY | CRONBACH’S ALPHA |
|---|---|---|---|---|---|---|---|
| Complex NPs | .33 | .33 | .25 | 0–.56 | 0–1 | .85 | .70 |
| X-Is-Difficult-to-Answer | .22 | .28 | .12 | 0–.31 | 0–.88 | .84 | .59 |
| Red. Relatives | .85 | .22 | 1 | .75–1 | 0–1 | .78 | .61 |
| Ditransitives | .86 | .15 | .88 | .75–1 | .25–1 | .42 | .23 |
| Control | .97 | .05 | 1 | .94–1 | .75–1 | - | - |
| Compr.aon | .57 | .18 | .56 | .45–.69 | .22–.97 | .79 | .88 |
| Compr.RT | 8256 | 3588 | 6946 | 5718–9927 | 3863–22494 | .70 | - |
| Calculation | .91 | .09 | .94 | .89–.96 | .42–1 | .88 | .87 |
| WM | 55.13 | 15.29 | 58 | 49–65 | 4–75 | .87 | .89 |
| Print.exp | 1.30 | 0.76 | 1.24 | 0.78–1.81 | -0.11–2.87 | .90 | .94 |
| Implicit | .58 | .15 | .57 | .48–.67 | .26–.95 | .75 | .77 |
| Explicit | .60 | .21 | .56 | .40–.79 | .23–.98 | .90 | .90 |
| LgAnalytic | .57 | .23 | .59 | .47–.75 | .03–.94 | .88 | .91 |
| Attention | 4.39 | 0.8 | 4.53 | 4.23–4.93 | 0.5–5.07 | .76 | .98 |
[i] Note: Compr.aon = mean proportion correct (all-or-nothing score) for each item on the comprehension task; Compr.RT = comprehension reaction time (for correct and incorrect responses); Calculation = mean proportion of correct responses in the calculation part of the Ospan; WM = working memory (Operation span task); Print.exp = Print exposure (Author Recognition Task, d’), Implicit = implicit memory; Explicit = explicit memory, LgAnalytic = language analytic ability (Sentence Pairs test); Attention = Sustained Attention (d’).
Table 3
Correlation matrix for all measures.
| COMPR.AON | COMPR.RT | CALCULATION | WM | PRINT.EXP | IMPLICIT | EXPLICIT | LGANALYTIC | ATTENTION | |
|---|---|---|---|---|---|---|---|---|---|
| Compr.aon | –.07 | .26* | .22* | .58*** | .26* | .21 | .68*** | .11 | |
| Compr.RT | –.07 | .17 | .01 | –.20 | –.05 | .02 | .13 | .01 | |
| Calculation | .26* | .17 | .22 | .22 | .14 | .27* | .43*** | .14 | |
| WM | .22* | .01 | .22 | .05 | .16 | .16 | .27* | .29* | |
| Print.exp | .58*** | –.20 | .22 | .05 | .14 | .19 | .45*** | .07 | |
| Implicit | .26* | –.05 | .14 | .16 | .14 | .53*** | .19 | .20 | |
| Explicit | .21 | .02 | .27* | .16 | .19 | .53*** | .30** | .36 | |
| LgAnalytic | .68*** | .13 | .43*** | .27* | .45*** | .19 | .30** | .21 | |
| Attention | .11 | .01 | .14 | .29* | .07 | .20 | .36** | .21 |
[i] Note: Correlations of .22 and above are significant at the .05 level (without correction for multiple comparisons).
* p < .05.
** p <.01.
*** p < .001 level.
Table 4
Correlations between structures on the comprehension task (all-or-nothing scores).
| COMPLEX NPS | X-IS-DIFFICULT-TO-ANSWER | RED. RELATIVES | DITRANSITIVES | |
|---|---|---|---|---|
| Complex NPs | .64 | .38 | .07 | |
| X-Is-Difficult-to-Answer | .64 | .30 | .18 | |
| Red. Relatives | .38 | .30 | .37 | |
| Ditransitives | .07 | .18 | .37 |
Table 5
Fixed effects for the model predicting comprehension scores.
| ESTIMATE | STANDARD ERROR | Z VALUE | p VALUE | |
|---|---|---|---|---|
| Intercept | 0.6482 | .4271 | 1.518 | .1291 |
| LgAnalytic | 0.8735 | .1528 | 5.715 | 1.10 × 10-8 |
| Print.exp | 0.5469 | .1384 | 3.953 | 7.73 × 10-5 |
| Implicit | 0.2854 | .1434 | 1.990 | .0466 |
| Calculation | –0.1018 | .1326 | –0.768 | .4425 |
| Explicit | –0.1126 | .1508 | –0.746 | .4555 |
| WM | 0.0940 | .1304 | 0.721 | .4710 |
| Attention | –0.0851 | .1294 | –0.658 | .5105 |
Figure 2
Generalisation as a proportional analogy problem.
Note: CAPITALS and pictures represent semantic structure while lower case letters represent phonological forms.