Have a personal or library account? Click to login
Employing Natural Control for Confounding Factors in the Hunt for the Bilingual Advantage in Attention: Evidence from School Children in Gibraltar Cover

Employing Natural Control for Confounding Factors in the Hunt for the Bilingual Advantage in Attention: Evidence from School Children in Gibraltar

Open Access
|Mar 2020

Figures & Tables

joc-3-1-94-g1.png
Figure 1

Schematic illustration of the bilingual picture vocabulary test (BilVoc).

Table 1

Questions asked in the BilQ to assess language usage. For questions with sub-questions, participants were required to answer again separately for each language.

#QuestionsSub-questionPossible responses
1eI speak:English…1. Never
1sSpanish…2. A few times a month
3. Every day
4. Always
2eIf someone speaks English and Spanish, what language do you speak to them in?English…1. Never
2sSpanish…2. Sometimes
3. Most of the time
4. Always
3e                                            I started learning English:1. At school
3s                                            I started learning Spanish:2. Years before school
3. I’ve always known it
4                                            I talk to friends in:1. English
5                                            I talk to family in:2. Spanish
6                                            I watch television in:3. Both
joc-3-1-94-g2.png
Figure 2

Schematic illustration of the ANT2 conflict and orienting trials (ANT2:C&O) in top panel, and of the ANT2 alerting trials (ANT2:A) in bottom panel. In the ANT2:C&O, response time was measured following the second tone (up to 5000ms).

joc-3-1-94-g3.png
Figure 3

Top panel: BilVoc performance – Response latencies. Left panel: average latencies in English and Spanish plotted against each other. Right panel: density histogram of corresponding BilVoc bilingualism scores. Middle panel: BilVoc performance – error percentages. Left panel: average error scores in English and Spanish plotted against each other. Right panel: density histogram of corresponding BilVoc bilingualism scores. Bottom panel: BilQ results. Left panel: average scores on the BilQ (0–1) in English and Spanish plotted against each other. Right panel: density histogram of corresponding BilQ bilingualism scores. In all plots, points represent individual participants and have been slightly jittered to avoid overplotting.

Table 2

Correlation matrix for disparity between English and Spanish language measures (BilVoc response times; BilVoc errors; and BilQ) and demographic factors (individualist culture; SES; and immigrant status).

Bilingualism score for BilVoc RTsBilingualism score for BilVoc ErrorsBilingualism score for BilQIndividualist cultureSES
Bilingualism scorer–.206
for BilVoc Errorsp.024
BF010.70
BF101.42
Bilingualism scorer–.249.377
for BilQp.011<.001
BF010.330.01
BF103.07285.09
Individualistr.065–.284–.124
culturep.502.003.159
BF016.660.113.43
BF100.159.440.29
SESr–.133–.126–.016–.018
p.165.191.857.810
BF013.243.609.0410.36
BF100.310.280.110.10
Immigrantr.085–.017–.170–.131.008
statusp.382.865.052.081.912
BF015.728.231.412.3510.71
BF100.180.120.710.430.093

[i] Note: Bayes Factor (BF) values which suggest at least ‘moderate’ evidence (BF ≥ 3) are shown in bold.

joc-3-1-94-g4.png
Figure 4

Performance on the ANT2 conflict and orienting trials (ANT2:C&O) task. Response latencies (in ms; left panel) varied by Frequency and Location; Response time differences corresponding to orientation and conflict resolution (in ms; right panel). Error bars indicate standard error of the means.

joc-3-1-94-g5.png
Figure 5

Performance on the ANT2 conflict and orienting trials (ANT2:C&O) task. Error rates (in percent; left panel) varied by Frequency and Location; Error rate differences corresponding to orientation and conflict resolution (in percent; right panel). Error bars indicate standard error of the means.

Table 3

Correlation matrix for measures of bilingualism (BilVoc response times; BilVoc errors; and BilQ) and of attentional control (global measures in the ANT2:C&O and ANT2:A, as well as the conflict, orienting and alerting components from these tasks).

Global MeasuresComponent Measures
ANT2:C&OANT1:AConflictOrientingAlerting
RTErrorRTErrorRTErrorRTErrorRTError
Bilingualismr.33.20.06.24.10.06.03.21.03.12
score forp.002.061.675.090.341.617.809.059.822.400
BilVoc RTsBF010.071.325.361.444.736.537.171.285.704.14
BF1014.080.760.190.690.210.150.140.790.180.24
Bilingualismr–.03–.02.04.04–.05–.07–.03–.02–.10–.07
score forp.781.835.789.762.651.520.799.887.492.609
BilVoc ErrorsBF017.107.225.645.586.676.027.157.314.645.14
BF100.140.140.180.180.150.170.140.140.220.20
Bilingualismr–.12–.06.09–.07.12.08.20–.07–.16.02
score forp.233.567.481.588.244.471.049.486.179.867
BilQ scoresBF013.856.605.215.763.986.001.156.112.756.56
BF100.260.150.190.170.250.170.870.160.360.15

[i] Note: Bayes Factor (BF) values which suggest at least ‘moderate’ evidence (BF ≥ 3) are shown in bold.

DOI: https://doi.org/10.5334/joc.94 | Journal eISSN: 2514-4820
Language: English
Submitted on: Dec 3, 2019
Accepted on: Feb 25, 2020
Published on: Mar 20, 2020
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2020 Chris M. Moreno-Stokoe, Markus F. Damian, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.