Infant event-related potentials to speech are associated with prelinguistic development

Abstract

Highlights Speech processing and prelinguistic skills studied in a large longitudinal sample. Auditory ERPs predicted prelinguistic development in infancy in LCS models. P1 amplitude at 6 months predicted prelinguistic development between 6 and 12 months. MMR to a frequency change was associated with prelinguistic skills at 6 months. Infants’ neural speech processing can help to predict early language development.

Longitudinal trajectories of electrophysiological mismatch responses in infant speech discrimination differ across speech features

“…Taken together, while our findings are in line with the general notion of MMR amplitude development from positive to negative polarity, they offer new insights on potential multi-stage developments. Our study was the first to systematically investigate and compare MMR amplitude trajectories to different language-related contrasts across the first year of life, while most previous studies were based on single time-point or cross-sectional observations (e.g., Kailaheimo-Lönnqvist et al, 2020 ; Mueller et al, 2012 ; Partanen, Kujala et al, 2013 ). Moreover, the few studies following the MMR development longitudinally did not investigate its featurespecific shape ( Cheng et al, 2013 , Cheng et al, 2015 , Pena et al, 2012 , Pihko et al, 1999 , Weber et al, 2004 ) and differences in study design and analytic approach might have further caused discrepant results.…”

“…However, despite the general development from p-MMR to n-MMR, the MMR’s characteristics are also influenced by experimental features (e.g., interstimulus-interval (ISI), attentional demands of the design), language background, and stimulus features ( Cheng et al, 2013 , Cheng et al, 2015 , Garcia-Sierra et al, 2016 , Leppänen et al, 1999 , Morr et al, 2002 , Sambeth et al, 2009 ). Consequently, depending on design and stimulus features, either p- or n-MMRs have been found for comparable age groups across different studies (e.g., Cheour, Ceponiene et al, 1998 ; Kailaheimo-Lönnqvist et al, 2020 ; Pihko et al, 1999 ; Ragó et al, 2014 ; Schaadt et al, 2015 ). Crucially, design and stimulus features not only influence single-time-point MMR amplitudes, but also differently affect the MMR’s maturational trajectory (e.g., Cheng et al, 2015 ; Garcia-Sierra et al, 2016 ).…”

Neural phoneme discrimination in variable speech in newborns – associations with dyslexia risk and later language skills

“…The signals were then re-referenced to an average of LM, RM, P7, and P8 electrodes. An average of four electrodes close to each other was chosen as the reference in order to improve signal-to-noise ratio in the often poor-quality data in the peripheral electrodes, and to allow for referencing in situations where one of the reference electrodes was considered broken (in which case it and its contralateral pair were excluded from referencing) (see also Thiede et al, 2019, Kailaheimo-Lönnqvist et al, 2020).…”

Challenges and new perspectives of developmental cognitive EEG studies

“…Importantly, the term "Mismatch Response " or MMR is often preferred in developmental studies because of its high intersubject variability. Indeed, this component tends to be positive in younger infants and sometimes up to school age, depending on the developmental status of children ( Bisiacchi et al, 2009 ;Mento and Bisiacchi, 2012 ;François et al, 2017 ;Moser et al, 2020 ;Kailaheimo-Lönnqvist et al, 2020 ;Hoehl and Wahl, 2012 ;Partanen et al, 2013 ;Maurer et al, 2003 ;Putkinen et al, 2014 ; see also Virtala et al, 2002 for a recent discussion on positive vs. negative MMR in a large longitudinal study). Similarly, positive and negative MMRs have been reported in infants between 2 and 6 months of age, suggesting that experiments with small sample sizes might be more likely to result in erroneous results than adult studies ( Trainor et al, 2003 ).…”