L2 Rhythm Acquisition: The question of learning direction

Lieke van Maastricht, Emiel Krahmer, Marc Swerts

Research output: Contribution to conferencePosterOther research output

36 Downloads (Pure)

Abstract

Dutch and Spanish are distinguishable from each other based solely on rhythmic cues (Ramus, Dupoux & Mehler, 2003) due to typological contrasts between both languages: Dutch has relatively complex syllable structure and uses extensive final (FL) and accentual lengthening (AL), while Spanish has less complex syllable structure and uses less FL and AL (Prieto et al., 2012; Cambier- Langeveld, 1999). This interplay of factors complicates speech rhythm acquisition by second language (L2) learners (Carter, 2005; Li & Post, 2014; White & Mattys, 2007). To determine whether learning direction affects the aforementioned correlates of speech rhythm, our study compares speech of 10 Dutch learners of Spanish with either low, i.e., A2, or high, i.e., B2, proficiency (Council of Europe, 2001) (DLS- and DLS+) with data by 10 less and more proficient Spanish learners of Dutch (SLD- and SLD+), as well as 10 native (L1) speakers of both languages as control groups. Eckman’s (1977, 2008) Markedness Differential Hypothesis (or MDH, used recently in Rasier & Hiligsmann, 2009; Ordin & Polyanskaya, 2015) predicts that Dutch rhythm is more marked, and thus more difficult to acquire than Spanish rhythm (cf. Payne et al., 2012). When comparing FL and AL data collected in an oral reading task similar to the one used in Prieto et al. (2012), DLS are thus expected to advance more towards their target L1 control group than SLD. Following Li & Post (2014), the raw data were converted to lengthening measures as a percentage of the baseline condition (unstressed and non-final syllables) to control for speech-rate differences. A Generalized Linear Mixed Model analysis reveals that for AL the SLD- differ significantly from the L1 Dutch while the proficient SLD+ do not, suggesting they approach target values. However, both DLS groups do not differ from the L1 Spanish, indicating that both are already on target, outperforming the SLD. Regarding FL, both L2 groups do not differ from the L1 Dutch, while the DLS+ do differ significantly from the L1 Spanish, implying that the SLD outperform the DLS in this aspect. Further analyses using rhythm metrics and including speakers up to C1 level will present a more definitive answer as to whether learning direction affects rhythm acquisition.
Original languageEnglish
Publication statusPublished - 2016
EventNew Sounds - Aarhus University, Aarhus, Denmark
Duration: 10 Jun 201612 Jun 2016

Conference

ConferenceNew Sounds
CountryDenmark
CityAarhus
Period10/06/1612/06/16

Fingerprint

learning
Group
language
Council of Europe
model analysis
Values

Cite this

van Maastricht, L., Krahmer, E., & Swerts, M. (2016). L2 Rhythm Acquisition: The question of learning direction. Poster session presented at New Sounds, Aarhus, Denmark.
van Maastricht, Lieke ; Krahmer, Emiel ; Swerts, Marc. / L2 Rhythm Acquisition : The question of learning direction. Poster session presented at New Sounds, Aarhus, Denmark.
@conference{af7faea9749e44c2b8e9ebbec96c37e7,
title = "L2 Rhythm Acquisition: The question of learning direction",
abstract = "Dutch and Spanish are distinguishable from each other based solely on rhythmic cues (Ramus, Dupoux & Mehler, 2003) due to typological contrasts between both languages: Dutch has relatively complex syllable structure and uses extensive final (FL) and accentual lengthening (AL), while Spanish has less complex syllable structure and uses less FL and AL (Prieto et al., 2012; Cambier- Langeveld, 1999). This interplay of factors complicates speech rhythm acquisition by second language (L2) learners (Carter, 2005; Li & Post, 2014; White & Mattys, 2007). To determine whether learning direction affects the aforementioned correlates of speech rhythm, our study compares speech of 10 Dutch learners of Spanish with either low, i.e., A2, or high, i.e., B2, proficiency (Council of Europe, 2001) (DLS- and DLS+) with data by 10 less and more proficient Spanish learners of Dutch (SLD- and SLD+), as well as 10 native (L1) speakers of both languages as control groups. Eckman’s (1977, 2008) Markedness Differential Hypothesis (or MDH, used recently in Rasier & Hiligsmann, 2009; Ordin & Polyanskaya, 2015) predicts that Dutch rhythm is more marked, and thus more difficult to acquire than Spanish rhythm (cf. Payne et al., 2012). When comparing FL and AL data collected in an oral reading task similar to the one used in Prieto et al. (2012), DLS are thus expected to advance more towards their target L1 control group than SLD. Following Li & Post (2014), the raw data were converted to lengthening measures as a percentage of the baseline condition (unstressed and non-final syllables) to control for speech-rate differences. A Generalized Linear Mixed Model analysis reveals that for AL the SLD- differ significantly from the L1 Dutch while the proficient SLD+ do not, suggesting they approach target values. However, both DLS groups do not differ from the L1 Spanish, indicating that both are already on target, outperforming the SLD. Regarding FL, both L2 groups do not differ from the L1 Dutch, while the DLS+ do differ significantly from the L1 Spanish, implying that the SLD outperform the DLS in this aspect. Further analyses using rhythm metrics and including speakers up to C1 level will present a more definitive answer as to whether learning direction affects rhythm acquisition.",
author = "{van Maastricht}, Lieke and Emiel Krahmer and Marc Swerts",
year = "2016",
language = "English",
note = "New Sounds ; Conference date: 10-06-2016 Through 12-06-2016",

}

van Maastricht, L, Krahmer, E & Swerts, M 2016, 'L2 Rhythm Acquisition: The question of learning direction' New Sounds, Aarhus, Denmark, 10/06/16 - 12/06/16, .

L2 Rhythm Acquisition : The question of learning direction. / van Maastricht, Lieke; Krahmer, Emiel; Swerts, Marc.

2016. Poster session presented at New Sounds, Aarhus, Denmark.

Research output: Contribution to conferencePosterOther research output

TY - CONF

T1 - L2 Rhythm Acquisition

T2 - The question of learning direction

AU - van Maastricht, Lieke

AU - Krahmer, Emiel

AU - Swerts, Marc

PY - 2016

Y1 - 2016

N2 - Dutch and Spanish are distinguishable from each other based solely on rhythmic cues (Ramus, Dupoux & Mehler, 2003) due to typological contrasts between both languages: Dutch has relatively complex syllable structure and uses extensive final (FL) and accentual lengthening (AL), while Spanish has less complex syllable structure and uses less FL and AL (Prieto et al., 2012; Cambier- Langeveld, 1999). This interplay of factors complicates speech rhythm acquisition by second language (L2) learners (Carter, 2005; Li & Post, 2014; White & Mattys, 2007). To determine whether learning direction affects the aforementioned correlates of speech rhythm, our study compares speech of 10 Dutch learners of Spanish with either low, i.e., A2, or high, i.e., B2, proficiency (Council of Europe, 2001) (DLS- and DLS+) with data by 10 less and more proficient Spanish learners of Dutch (SLD- and SLD+), as well as 10 native (L1) speakers of both languages as control groups. Eckman’s (1977, 2008) Markedness Differential Hypothesis (or MDH, used recently in Rasier & Hiligsmann, 2009; Ordin & Polyanskaya, 2015) predicts that Dutch rhythm is more marked, and thus more difficult to acquire than Spanish rhythm (cf. Payne et al., 2012). When comparing FL and AL data collected in an oral reading task similar to the one used in Prieto et al. (2012), DLS are thus expected to advance more towards their target L1 control group than SLD. Following Li & Post (2014), the raw data were converted to lengthening measures as a percentage of the baseline condition (unstressed and non-final syllables) to control for speech-rate differences. A Generalized Linear Mixed Model analysis reveals that for AL the SLD- differ significantly from the L1 Dutch while the proficient SLD+ do not, suggesting they approach target values. However, both DLS groups do not differ from the L1 Spanish, indicating that both are already on target, outperforming the SLD. Regarding FL, both L2 groups do not differ from the L1 Dutch, while the DLS+ do differ significantly from the L1 Spanish, implying that the SLD outperform the DLS in this aspect. Further analyses using rhythm metrics and including speakers up to C1 level will present a more definitive answer as to whether learning direction affects rhythm acquisition.

AB - Dutch and Spanish are distinguishable from each other based solely on rhythmic cues (Ramus, Dupoux & Mehler, 2003) due to typological contrasts between both languages: Dutch has relatively complex syllable structure and uses extensive final (FL) and accentual lengthening (AL), while Spanish has less complex syllable structure and uses less FL and AL (Prieto et al., 2012; Cambier- Langeveld, 1999). This interplay of factors complicates speech rhythm acquisition by second language (L2) learners (Carter, 2005; Li & Post, 2014; White & Mattys, 2007). To determine whether learning direction affects the aforementioned correlates of speech rhythm, our study compares speech of 10 Dutch learners of Spanish with either low, i.e., A2, or high, i.e., B2, proficiency (Council of Europe, 2001) (DLS- and DLS+) with data by 10 less and more proficient Spanish learners of Dutch (SLD- and SLD+), as well as 10 native (L1) speakers of both languages as control groups. Eckman’s (1977, 2008) Markedness Differential Hypothesis (or MDH, used recently in Rasier & Hiligsmann, 2009; Ordin & Polyanskaya, 2015) predicts that Dutch rhythm is more marked, and thus more difficult to acquire than Spanish rhythm (cf. Payne et al., 2012). When comparing FL and AL data collected in an oral reading task similar to the one used in Prieto et al. (2012), DLS are thus expected to advance more towards their target L1 control group than SLD. Following Li & Post (2014), the raw data were converted to lengthening measures as a percentage of the baseline condition (unstressed and non-final syllables) to control for speech-rate differences. A Generalized Linear Mixed Model analysis reveals that for AL the SLD- differ significantly from the L1 Dutch while the proficient SLD+ do not, suggesting they approach target values. However, both DLS groups do not differ from the L1 Spanish, indicating that both are already on target, outperforming the SLD. Regarding FL, both L2 groups do not differ from the L1 Dutch, while the DLS+ do differ significantly from the L1 Spanish, implying that the SLD outperform the DLS in this aspect. Further analyses using rhythm metrics and including speakers up to C1 level will present a more definitive answer as to whether learning direction affects rhythm acquisition.

M3 - Poster

ER -

van Maastricht L, Krahmer E, Swerts M. L2 Rhythm Acquisition: The question of learning direction. 2016. Poster session presented at New Sounds, Aarhus, Denmark.