How to study interference between languages during gender selection: An experiment with Portuguese-German bilinguals

Abstract

Research in bilingualism faces a significant challenge: grammatical gender. This aspect poses difficulties in studying its retrieval during real-time language processing and stands as one of the most formidable features to acquire in a second language. In this presentation, I explain the essential methodologies for investigating gender retrieval in bilingual language production and I discuss the main theories on the representation of gender within the bilingual mind. I conclude by presenting a recent study where we examined gender processing in bilingual speakers of European Portuguese and German, acquired at varying stages of life.

Full text

How to study interference between languages during
gender selection:
An experiment with Portuguese-German bilinguals
Ana Rita Sá-Leite
Funding:
FCT and FEDER through COMPETE2020 (PT2020 Partnership
Agreement POCI-01-0145-FEDER-007653); FCT: UIDB/00305/2020
Ayuda FPU16/06983 financiada por MCIN/AEI /10.13039/501100011033 y
FSE “El FSE invierte en tu futuro”
14th Heritage Language Research Institute

Contents
I
Grammatical
gender: concept
II
Grammatical
gender: mental
representation
III
Methodology:
naming tasks
IV
Methodology:
translation tasks
Main considerations and
resources
VI
Our study
V
14th Heritage Language Research Institute

Gender
system
Grammatical gender
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Gender
system
Grammatical gender
classification of nouns
Arbitrary/abstract
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Gender
system
Grammatical gender
European Portuguese:
M
M
mesa
[table]
relógio
[watch]
classification of nouns
F
Arbitrary/abstract
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Gender
system
Grammatical gender
European Portuguese:
M
M
Fmesa
[table]
relógio
[watch]
Arbitrary/abstract
German:
N
M
F
Zimmer
[room]
Tisch
[table]
Uhr
[watch]
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
classification of nouns
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Representation of gender within the architecture of lexical access
European
Portuguese TABLE
mesa
/mesa/
conceptual
stratum
lexical-syntactic
stratum
word-form
stratum
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
M
F
Levelt et al. (1999)
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Representation of gender within the architecture of lexical access
TABLE
mesa
/mesa/
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
M
FM F
Threshold for selection
=
difference
Accumulation of activation
Levelt et al. (1999)
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Gender systems are integrated within the bilingual mind, so that:
M
F
German (L1)
European Portuguese (L2)
TABLE
mesa tisch
/mesa/
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
N
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Producing L2 European Portuguese
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Gender systems are integrated within the bilingual mind, so that:
M
F
German (L1)
European Portuguese (L2)
TABLE
mesa tisch
/mesa/
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
gender incongruency →competition for selection
N
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Producing L2 European Portuguese
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Representation of gender within the architecture of lexical access
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
M F
Threshold for selection
Accumulation of activation
Levelt et al. (1999)
M
F
TABLE
mesa tisch
/mesa/
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
N
Producing L2 European Portuguese
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Representation of gender within the architecture of lexical access
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
M F
Threshold for selection
Accumulation of activation
Levelt et al. (1999)
M
F
TABLE
mesa tisch
/mesa/
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
N
Producing L2 European Portuguese
previous threshold for selection
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

N
M
F
DOOR
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
porta
/porta/
tür
Gender systems are integrated within the bilingual mind, so that:
gender congruency →facilitation for selection
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Our study
V. Conclusions
14th Heritage Language Research Institute
German (L1)
European Portuguese (L2)
Producing L2 European Portuguese

Representation of gender within the architecture of lexical access
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
M F
Threshold for selection
Accumulation of activation
Levelt et al. (1999)
N
M
F
DOOR
conceptual
stratum
lexico-syntactic
stratum
word-form
stratum
porta
/porta/
tür
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

Gender systems are integrated within the bilingual mind, so that:
gender congruency →facilitation for selection
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
gender incongruency →competition for selection
Translations of the same gender
Translations of different gender
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

+
porta
500 ms
Procedure:
max.: 3000 ms
500 ms
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
German (L1) - European Portuguese (L2)
Participants:
Gender congruent and gender incongruent translations
Materials:
Measures:
Response times:
•Gender congruent condition
•Gender incongruent condition
+
mesa
500 ms max.: 3000 ms
500 ms
Gender congruency Gender incongruency
Tür Tisch
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

+
porta
500 ms
Procedure:
max.: 3000 ms
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Tür
German (L1) - European Portuguese (L2)
Participants:
Gender congruent and gender incongruent translations
Materials:
Measures:
Response times:
•Gender congruent condition
•Gender incongruent condition
+
mesa
500 ms max.: 3000 ms
Tisch
Gender congruency Gender incongruency
500 ms 500 ms
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Results:
The cross-linguistic
gender congruency
effect (L2)
Response Times
gender CONGRUENCY
<
Response Times
gender INcongruency
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants)

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size
- Control of the materials in both languages

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size
- Control of the materials in both languages
•Frequency and length

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size
- Control of the materials in both languages
•Frequency and length
•Form overlap between translations
•Whether one noun can have more than one translation of different gender:

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size
- Control of the materials in both languages
•Frequency and length
•Form overlap between translations
•Whether one noun can have more than one translation of different gender:
Spanish: Portuguese:
“Jarra” (fem) “Vaso” (masc) [vase]
“Caneca” (fem) [mug]
“Jarra” (fem) [jug/vase]

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size
- Control of the materials in both languages
•Frequency and length
•Form overlap between translations
•Whether one noun can have more than one translation of different gender:
•Neighboorhood size, familiarity, etc.
Spanish: Portuguese:
“Jarra” (fem) “Vaso” (masc) [vase]
“Caneca” (fem) [mug]
“Jarra” (fem) [jug/vase]

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
-Sample size (>50 participants) →gender = effects of small size
- Control of the materials in both languages
Word databases:
For orthographic overlap: NIM database (Guasch et al., 2013)
SUBTLEX databases: Catalonian, Dutch, German, Greek, Polish, Portuguese, Spanish…
•Frequency and length
•Form overlap between translations
•Whether one noun can have more than one translation of different gender:
•Neighboorhood size, familiarity, etc.
Spanish: Portuguese:
“Jarra” (fem) “Vaso” (masc) [vase]
“Caneca” (fem) [mug]
“Jarra” (fem) [jug/vase]

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
- Conduct it online? (free, open-source experiment builder software)

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
- Conduct it online? (free, open-source experiment builder software)
* PCIbex:
* OpenSesame (OSWeb):
*jsPsych:
Zehr, J., & Schwarz, F. (2018). PennController for Internet Based Experiments
(IBEX). https://doi.org/10.17605/OSF.IO/MD832
Mathôt, S., Schreij, D., & Theeuwes, J. (2012). OpenSesame: An open-source,
graphical experiment builder for the social sciences. Behavior Research Methods,
44(2), 314–324. https://doi.org/10.3758/s13428-011-0168-7
de Leeuw, J. R. (2015). jsPsych: A JavaScript library for creating behavioral
experiments in a Web browser. Behavior Research Methods, 47(1), 1–12.
https://doi.org/10.3758/s13428-014-0458-y

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
- Fill a gap:
A. Test new languages

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
- Fill a gap:
A. Test new languages
B. Use new measuring techniques (EEGs)

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
14th Heritage Language Research Institute
Considerations to conduct this type of task:
IV. Methodology: translation tasks
V. Main considerations and resources
VI. Our study
- Fill a gap:
A. Test new languages
B. Use new measuring techniques (EEGs)
C. Go beyond that…
* What happens when the gender is not shared between languages (neuter)?
* Heritage speakers?

I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

GAPS
•What is the role of the age of acquisition? Most studies
used late learners.
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
AIMS

GAPS
•What is the role of the age of acquisition? Most studies
used late learners.
•What is the role of proficiency? Never included in the
analyses.
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
AIMS

GAPS
•What is the role of the age of acquisition? Most studies
used late learners.
•What is the role of proficiency? Never included in the
analyses.
•Is the cross-linguistic gender congruency effect
generalizable to the L1?
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
AIMS

Participants:
74 (
M
age = 38.12 years,
SD
= 9.73)
European Portuguese (L1) –German (L2)
METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Participants:
74 (
M
age = 38.12 years,
SD
= 9.73)
Early bilinguals: were born or lived in Germany
European Portuguese (L1) –German (L2)
METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Participants:
74 (
M
age = 38.12 years,
SD
= 9.73)
Early bilinguals: were born or lived in Germany
Late bilinguals: acquired German in classroom setting
European Portuguese (L1) –German (L2)
METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Participants:
74 (
M
age = 38.12 years,
SD
= 9.73)
Early bilinguals: were born or lived in Germany
Late bilinguals: acquired German in classroom setting
Dialang Vocabulary Size Placement Test (VSPT, v. 1) for Portuguese
and German →Language balance
European Portuguese (L1) –German (L2)
METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Participants:
74 (
M
age = 38.12 years,
SD
= 9.73)
Early bilinguals: were born or lived in Germany
Late bilinguals: acquired German in classroom setting
Dialang Vocabulary Size Placement Test (VSPT, v. 1) for Portuguese
and German →Language balance
European Portuguese (L1) –German (L2)
METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
The effect of
proficiency

180 nouns →
90 nouns Portuguese →German;
90 nouns German →Portuguese –counterbalanced
60 nouns per German gender value (F/M/N)
METHOD
Materials:
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Forward and backward translation task (online)
Portuguese (L1) to German (L2)
German (L2) to Portuguese (L1)
METHOD
Procedure:
+
Tisch
mesa
+
mesa
Tisch
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

METHOD
Hypotheses:
(a) The higher the imbalance between languages, the greater the effect of gender
congruency.
(b) The effect can be obtained in both the L1 (European Portuguese) and the L2
(German), but it may depend on opaqueness
(a) The effect does not depend on the Age of Acquisition.
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
GENDER TRANSPARENCY:
Relationship between the NOUN FORM and GENDER VALUE
+ -

METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
GENDER TRANSPARENCY:
Relationship between the NOUN FORM and GENDER VALUE
+ -
Portuguese: German:
Masculine nouns: “-o”
Feminine nouns: “-a”

METHOD
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
GENDER TRANSPARENCY:
Relationship between the NOUN FORM and GENDER VALUE
+ -
Portuguese: German:
Masculine nouns: “-o”
Feminine nouns: “-a”
Lots of endings and
exceptions (more intricate)
Salamoura & Williams, 2007
Bordag & Pechmann, 2008

METHOD
Design:
Gender Congruency x Target Language (L1/L2) x Language balance score x Age of acquisition
Hypotheses:
(a) The higher the imbalance between languages, the greater the effect of gender
congruency.
(b) The effect can be obtained in both the L1 (European Portuguese) and the L2
(German), but it may depend on opaqueness (only for European Portuguese?)
(a) The effect does not depend on the Age of Acquisition.
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
x Target Gender x Contact with German Language x Dominance

The higher the imbalance, the greater the effect of interference (gender incongruency) when
producing Portuguese (L1), regardless of age of acquisiton.
Gender congruency
Gender incongruency
German Portuguese
Response times
RESULTS
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Hypotheses:
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
(a) The higher the imbalance between languages, the greater the effect of gender
congruency.
(b) The effect can be obtained in both the L1 (European Portuguese) and the L2
(German), but it may depend on opaqueness (only for European Portuguese?)
(a) The effect does not depend on the Age of Acquisition.

Hypotheses:
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
(a) The higher the imbalance between languages, the greater the effect of gender
congruency.
(b) The effect can be obtained in both the L1 (European Portuguese) and the L2
(German), but it may depend on opaqueness →Only for L1 (European
Portuguese)
(a) The effect does not depend on the Age of Acquisition.

Hypotheses:
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute
(a) The higher the imbalance between languages, the greater the effect of gender
congruency.
(b) The effect can be obtained in both the L1 (European Portuguese) and the L2
(German), but it may depend on opaqueness →Only for L1 (European
Portuguese)
(a) The effect does not depend on the Age of Acquisition.

Future studies: replication plus…
L1 Transparent
L2 Transparent
L1 Transparent
L2 Opaque
L1 Opaque
L2 Transparent
L1 Opaque
L2 Opaque
+ transparency
+ chance of observing competition between gender values
I. Grammatical gender: concept
II. Grammatical gender: mental representation
III. Methodology: naming tasks
I. Methodology: translation tasks
IV. Main considerations and resources
V. Our study
14th Heritage Language Research Institute

Thank
you!
anar.saleite@gmail.com
14th Heritage Language Research Institute
The Article!
My ResearchGate