“Beyond the literal meaning” processing implied emotion in second language discourse: an ERP study

ABSTRACT

Emotion is processed incrementally during sentence comprehension in a first language (L1) due to unification operations. Since processing multiple types of information is cognitively more demanding in a second language (L2), we investigated implied emotion processing in L1 and L2 using event-related brain potentials. We presented native Spanish speakers with sentences whose context rendered neutral words either negative-congruent, neutral-congruent or neutral-incongruent in Spanish (L1) and English (L2). Results showed differences in early emotion processing across languages: a larger N100 for neutral than negative sentences in L1 and a more positive P300 for negative than neutral sentences in L2. Effects in the N400 and LPP time windows were not language dependent. An N400 semantic incongruity effect was present in both languages. A gradual LPP effect was larger for negative than incongruent sentences in both languages, with neutral sentences in between. Results indicate early time-window differences in implied emotion processing across languages.

KEYWORDS:

• Implied emotion
• Event-related potentials (ERP)
• Second language processing
• N400
• LPP

Acknowledgements

The study was supported by the Spanish Government (FEDER/Ministerio de Ciencia, Innovación y Universidades, PID2020-115175RB-I00). Andrea González-García was supported by a grant from the Spanish Government (FPU20/06880) and received the Language Learning Dissertation Grant (2022) that partially financed the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data will be available upon publication at OSF link: https://osf.io/v43pr/?view_only=c739ab99270842f3b82c210e40bc3e69

Notes

1 The minimally required sample size (i.e. 24 participants) was determined with an a priori power analysis using G*Power with an effect size of 0.25 (ŋ²) across a set of 4 electrodes, power of 0.90 and alpha level of 0.05 (Faul et al., 2007). Thirty Spanish-native L2 English speakers took part in the study but 4 were excluded from analyses because the ERP data exceeded the threshold of artifact rejections (over 50%).

2 We thank Guillaume Thierry for pointing out this possibility during the poster presentation at the Society of Neurobiology of Language meeting 2023. Although this potential experimental bias should have equally affected the two language conditions similarly, we are currently conducting a follow-up study in our lab to discard this possibility and to better identify processing differences in L1 and L2.

Additional information

Funding

This work was supported by Language Learning Dissertation Grant; Spanish Government Grant [grant number FPU20/06880]; FEDER/Ministerio de Ciencia, Innovación y Universidades [grant number PID2020-115175RB-I00].