Word recognition and phonological representation in very low birth weight preterms

https://doi.org/10.1016/j.earlhumdev.2012.07.019Get rights and content

Abstract

Background

Several studies have shown impaired neurocognitive development in infants born very preterm. Language is one of the areas that may be affected. Early lexical development measurements have revealed possible delays associated with low gestational age, but no studies have analyzed lexical processing using real-time measures in this at-risk population.

Aims

To explore the effects of preterm birth on the robustness of phonological representations and lexical processing speed.

Subjects and methods

Eighteen two-year-old VLBW healthy preterms (≤ 32 weeks of gestation; ≤ 1500 g) and a matched group of at term infants, equivalent in age, gender, SES, linguistic environment and expressive lexicon were compared in a ‘looking-while-listening’ task, using correctly pronounced and mispronounced known words involving a vowel change. Percentage of target fixation, longest look duration, shift rate and orientation latency measures were used to analyze possible between-group differences in phonological representation and familiar word recognition processes.

Results

Based on the percentage of fixation time measure both groups succeeded at word recognition and responded similarly to mispronunciations. However, preterms significantly differed from full‐terms in processing speed measures, showing longer look duration, lower shift rate and slower orientation latencies to target from distracter.

Conclusions

Preterm birth negatively affects lexical processing speed. Birth weight and gestational age are two critical variables in these results. Slower language processing in preterms can compromise the acquisition of more complex lexical and grammatical representations later in development and may underlie poor language outcomes frequently observed in children born very prematurely.

Introduction

It has been shown that babies born very preterm (< 32 weeks of gestation) and with very low birth weight (< 1500 g) run a higher risk of encountering problems in their cognitive and behavioral development [1], [2], [3]. Recent studies suggest that premature interruption of time in the intra-uterine environment has consequences in cerebral growth and in the establishment of connectivity among structures to the extent that neurocognitive development is affected, even when no significant anomalies are observed with conventional ultrasound or MRI [4]. In effect, white matter microstructural abnormalities have been identified in preterm infants at term equivalent age using diffusion tensor imaging with tract-based analysis [5] and diffuse myelination disturbances linked to cognitive sequelae can still be found later in the development [6], [7]. Studies evaluating the cognitive development, behavior, and school performance of preterms and adolescents at varying ages, compared to control groups of full-term babies, have shown, with few exceptions [8], [9], the presence of perceptive and visual-motor alterations, attention deficits and executive function disorders, difficulties with language, learning problems, and, in general, poor academic performance [10], [11], [12], [13]. One of the affected areas is language development. Both receptive and expressive difficulties have been demonstrated in measurements made at 4–6 years of age [14], [15], [16], and even later in development [17], [18], with significant differences in comparison to control groups. Recent work involving preterm participants at 6 and 8 years of age has confirmed that, even in the absence of brain damage, preterm birth has long-term effects on language and literacy development, and differences in acquisition rate and learning strategies are present in this population when compared to fullterm children [19], [20]. Even in studies involving samples of less immature preterm children (30–34 weeks gestation), lower scores in receptive language measures have been found at five years of age [21]. Preterms not only tend to lag behind full-terms in many language acquisition measures, but they also show an increased risk for language impairment [22].

A closer look to the early stages in language development reveals that preterms very often show a slow development in early social-communicative skills, with less responsiveness and fewer initiatives to engage in dyadic and triadic interactions [23] and delays in the production of communicative gestures [24], [25]. They also differ from full-terms in their initial expressive lexicon at 12 months [26], [27] and in lexical and grammatical development at two years of age [28], [29], [30]. Although some studies indicate that the size of the expressive lexicon for preterms and controls may be similar at 2 years [31] and at 2 1/2 years of age [32], they did find differences in the composition of the vocabulary (fewer grammatical words in preterms with vocabularies > 425 words [31]) and they highlight the roles of sex and birth weight/gestational age as critical variables in linguistic development (extremely low birth weight preterm boys are at increased risk of presenting lexical and grammatical delays [32]).

Just as studies on visual perception and recognition memory in the first year of life have shown that processing of information in preterms is slower [33], [34] and that the difficulties in processing [35], [36], [37] and attention [38] are strongly related with subsequent cognitive development, it is possible to posit the idea that the problems in language development might be related to difficulties in the processing of speech. Studies in this area have revealed important differences between preterms and controls in electrophysiological measurements of sound discrimination and recognition of the mother's voice at 40 weeks of gestational age [39]. Differences in the evoked auditory potentials with syllable shift have also been observed throughout the first year of life [40], [41], [42], and they continue at older ages [43], [44], predictive of the difficulties seen in an object-naming task carried out at 6 years of age [45]. Taken together, these results show poorer performance in auditory discrimination and speech perception tasks, which may be related to the language problems observed in this population. It has been questioned whether these difficulties are specific to the processing of linguistic stimuli, characterized by rapid changes in the patterns of acoustic energy, or rather are of a more general nature, equally affecting different areas of cognitive functioning. This issue has been suggested by a study in which nine-month-old preterms resolved visual and audio-visual discrimination tasks and they differed from controls in the measured rate of processing, with shallower slopes of habituation, a greater number of trials in order to achieve the criteria, and longer visual fixation times [46]. Recent data exploring both native-language discrimination and early word-form segmentation abilities in a sample of very preterm infants, tested before 9 months of age, has revealed specific difficulties only in the latter, which places greater cognitive demands in terms of information processing, pattern extraction and memory load [47]. Conclusions from a different study, in which four basic cognitive processes (i.e. memory, representational competence, processing speed and attention) were analyzed at 12 months of age, go one step further and suggest that language deficits in preterms are related to information processing abilities in two specific domains: memory and representational competence [24]. Critically, even though preterms and controls differed in attention and encoding speed measures at age 1 year, these results were not found to be related to language outcomes at 12 or 36 months of age. The lack of predictive value of these two basic cognitive abilities found in that study seems to be in conflict with previous evidence obtained with normally developing children, showing that look duration declines along the first year of life [48] and increasing speed in word recognition [49] were related to language outcomes in the second year of life. Moreover, not only the speed of word recognition was shown to be closely related to lexical–grammatical development measured at 25 months of age [49], but it also predicted cognitive and linguistic abilities at 8 years of age [50]. Thus, the assessment of preterm's word recognition processes at 2 years of age can be particularly relevant because of their subsequent role in further developing and improving oral language skills, both in the comprehension and production domains. Besides word recognition studies, results from research on toddlers' phonological representation of known words have clearly shown that they have adequately detailed lexical representations and react when presented with changes in pronunciation relative to both consonants and vowels [51], [52], [53], [54], [55], [56]. Taken together, these results set the ground for the present research. The evidence of initial difficulties in information processing awakens interest in analyzing how lexical recognition processes are carried out by this at-risk population, once it reaches the second year of life.

The main aim of this study is, thus, to explore the effects of preterm birth on word recognition processes and phonological encoding based on the ability to detect incorrect pronunciations (via a vowel change) in known words. In an attempt to explore the efficiency of lexical processing in preterms and to identify possible differences when compared to a control group of full-terms, a version of the looking-while-listening task has been implemented [57], [58], [59]. In this task, toddlers' eye movements in response to spoken words are monitored and analyzed for accuracy and speed in word recognition. Specifically, measures involving proportion of visual fixation time to target will capture accuracy, while measures relative to look duration, fixation shifts and orientation latencies will offer relevant information about processing speed.

Regarding accuracy in word recognition (i.e. how reliably toddlers look at the target image), previous studies have shown that 2-year-old-toddlers can easily identify the correct image to target words (as seen in proportion fixation above chance level) and can detect unusual pronunciations (as seen in lower percentage fixation to incorrectly pronounced items compared to the standard ones) [53], [54], [60]. In the present research, the two groups under study (very preterm and full‐term participants) were closely matched in age and expressive vocabulary size, therefore we expected no between-group differences in correctly identifying the images to the target words and in noticing the unusual pronunciations. Only if phonological representation of vowels in known words were less detailed in the preterm group than in the control sample, then their proportion of looking time to correct and incorrect pronunciations would be equivalent as mispronunciations would remain undetected. As for processing speed measures in this task (i.e. longest look duration, fixation shifts and orientation latencies from distracter) it was hypothesized that between group differences would be observed and they could be even more evident in the incorrect pronunciation condition, where a fast and straightforward word recognition response might be hindered by the mismatch between the stimulus presented and its stored representation. Slower responses in terms of language processing were, thus, predicted in the preterm group, paralleling previous results obtained in the visual perception domain [33], [35], [36]. In addition, the impact of specific neonatal characteristics such as birth weight, gestational age and clinical risk factors on preterms' accuracy and processing speed measures were also analyzed.

Section snippets

Subjects and methods

The participants were 18 preterms (≤ 32 weeks of gestation; ≤ 1500 g birth weight, without severe congenital, physical, or neurological abnormalities) and 18 full-terms (control group). All participants were born in the Sant Joan de Déu maternity–children's hospital in Barcelona. Preterms had had magnetic resonance imaging (MRI) at 40 weeks of gestational age and cases with observable brain injury were excluded from the final sample. Also excluded were infants born small for gestational age (SGA)

Procedure

Word recognition and phonological representation were evaluated by means of an intermodal visual fixation task, presenting familiar words that were correctly and incorrectly pronounced [54]. Word selection was constrained by several factors: items had to belong to children's vocabularies, correspond to picturable objects, and have a cognate counterpart in the other language (two comparable versions of the same test were built so that participants could be tested in their native language). Four

Fixation percentages

The values for this measure were normally distributed and no outliers were identified (values ± 3 SD greater than the mean of the distribution). The two groups behaved in a similar fashion (Fig. 1), with percentages of 70.9% (CP) and 59.1% (IP) for the preterm group and 71.6% (CP) and 57% (IP) for the control group, all significantly greater than chance (50%) under both conditions [preterms: t(17) = 6.6; p = 0.0001, d = 2.28 and t(17) = 3.7; p = 0.002, d = 1.27, for CP and IP respectively. Controls: t(17) = 

Discussion

In this research two groups of 2-year-old children differing in gestational age and birth weight (a sample of healthy very preterms and a sample of full‐term participants) were compared on word recognition processes and their ability to detect a mispronunciation in known words. Critically, the two groups were matched in terms of age, gender, language background, parental education and expressive vocabulary to control for possible confounding effects of these variables on the measures under

Conflict of interest

The authors of this study have indicated no conflicts of interest.

Acknowledgments

We would like to thank the families of the participants in this research and the preterm baby follow-up team at the Hospital Sant Joan de Déu for having arranged contact with them. We are also grateful to Eva Águila for her help in testing and coding. This research was supported by a grant from the Spanish Ministry of Science and Innovation (projects PSI2008-01253 and PSI2011-25376).

References (62)

  • K. Mikkola et al.

    Auditory event-related potentials and cognitive function of preterm children at five years of age

    Clin Neurophysiol

    (2007)
  • E. Jansson-Verkasalo et al.

    Neurophysiological correlates of deficient phonological representations and object naming in prematurely born children

    Clin Neurophysiol

    (2004)
  • L. Bosch

    Precursors to language in preterm infants: speech perception abilities in the first year of life

  • T.M. Bailey et al.

    Phonological specificity in early words

    Child Dev

    (2002)
  • D. Swingley et al.

    Spoken word recognition and lexical representation in very young children

    Cognition

    (2000)
  • N. Mani et al.

    Phonological specificity of consonants and vowels in early lexical representations

    J Mem Lang

    (2007)
  • D. Swingley et al.

    Continuous processing in word recognition at 24 months

    Cognition

    (1999)
  • M. Ramon-Casas et al.

    Vowel categorization during word recognition in bilingual toddlers

    Cogn Psychol

    (2009)
  • G.P. Aylward

    Neurodevelopmental outcomes of infants born prematurely

    J Dev Behav Pediatr

    (2005)
  • C.S.H. Aardnoudse-Moens et al.

    Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children

    Pediatrics

    (2009)
  • M.L. Krishnan et al.

    Relationship between white matter apparent diffusion coefficients in preterm infants at term-equivalent age and developmental outcome at 2 years

    Pediatrics

    (2007)
  • R.T. Constable et al.

    Prematurely born children demonstrate white matter microstructural differences at 12 years of age, relative to term control subjects: an investigation of group and gender effects

    Pediatrics

    (2008)
  • J.S. Andrews et al.

    Reading performance correlates with white-matter properties in preterm and term children

    Dev Med Child Neurol

    (2010)
  • J. Litt et al.

    Learning disabilities in children with very low birthweight: prevalence, neuropsychological correlates and educational interventions

    J Learn Disabil

    (2005)
  • B. Caravale et al.

    Cognitive development in low risk preterm infants at 3–4 years of life

    Arch Dis Child Fetal

    (2005)
  • A.T. Bhutta et al.

    Cognitive and behavioral outcomes of school-age children who were born preterm. A meta-analysis

    JAMA

    (2002)
  • L.A. Foulder-Hugues et al.

    Motor, cognitive and behavioural disorders in children born very preterm

    Dev Med Child Neurol

    (2003)
  • M. Hack et al.

    Outcomes in young adulthood for very-low-birth-weight infants

    N Engl J Med

    (2002)
  • S.H. Foster-Cohen et al.

    High prevalence/low severity language delay in preschool children born very preterm

    J Dev Behav Pediatr

    (2010)
  • L. Luoma et al.

    Speech and language development of children born at ≤ 32 weeks gestation: a 5-year prospective follow-up study

    Dev Med Child Neurol

    (1998)
  • D. Wolke et al.

    Cognitive status, language attainment and pre-reading skills of 6-year-old very preterm children and their peers: the Bavarian longitudinal study

    Dev Med Child Neurol

    (1999)
  • Cited by (14)

    • Eye tracking based assessment of lexical processing and early lexical development in very preterm children

      2022, Early Human Development
      Citation Excerpt :

      Additionally, at preschool age they score lower than full-term controls in assignments for attention, processing speed and working memory [15]. Studies report slower language processing of preterm children at 18 months [1] and at 24 months [16], while other studies found no differences between preterm and full-term children's processing speed or accuracy at 18 months, when corrected age was applied [17]. The specific connection between early processing of linguistic items and lexical growth during the second year of life is not fully known [9].

    • Do the early development of gestures and receptive and expressive language predict language skills at 5;0 in prematurely born very-low-birth-weight children?

      2016, Journal of Communication Disorders
      Citation Excerpt :

      Individual differences in the subskills (e.g. auditory discrimination and processing, phonological discrimination, phonological memory, word retrieval, verbal memory, grammatical processing) already at an early age would then lead to different types of end results. Preterm children have been reported to have for example weaker phonological working memory and slower lexical processing speed than the full-term controls (e.g. Ramon-Casas, Bosch, Iriondo, & Krauel, 2013; Sansavini et al., 2007). Secondly, it has been shown that there exists a strong link between the expressive lexicon size and the emergence of expressive grammar at the end of the second year in children with different native languages (e.g. Bates & Goodman, 1999; Caselli, Casadio, & Bates, 1999; Stolt, Haataja, Lapinleimu, & Lehtonen, 2009b; Thodrardottir, Weismer, & Evans, 2002).

    • Electrophysiological auditory response to acoustically modified syllables in preterm and full-term infants

      2016, Journal of Neurolinguistics
      Citation Excerpt :

      For preterm infants, the corrected age was considered. The auditory habituation task is particularly suitable for our purpose because it is a measure that is present from the first months of life (González-Frankerberger et al., 2008; Ramon-Casas et al. 2013; Rose et al., 2002) and is an indirect measure of processing speed. We hypothesized that the extension of the syllable formant transitions would facilitate auditory habituation in preterm infants who are at risk for language development problems.

    • Effect of maturation on suprasegmental speech processing in full- and preterm infants: A mismatch negativity study

      2014, Research in Developmental Disabilities
      Citation Excerpt :

      Jansson-Verkasalo et al. (2010) found qualitatively different ERP response pattern in case of 4 year-olds born with very low birth weight. A more recent study of Ramon-Casas, Bosch, Iriondo, and Krauel (2013) showed diminished lexical processing speed at the age of two. Based on all the suggestions of these studies one may assume that a better specification of the nature of early maturational differences would allow us to develop more effective intervention programs to be used from the early months of life on.

    View all citing articles on Scopus
    View full text