Les difficultés d’accès lexical, aussi appelées trouble d’accès lexical, se manifestent par une incapacité fréquente à retrouver le mot voulu pour s’exprimer. Bien que les difficultés d’accès lexical soient couramment rapportées chez les enfants vus en orthophonie, celles-ci bénéficient de peu de recherches en orthophonie. Les recherches auprès des enfants francophones ou ayant recours à des tâches discursives, comme la narration, sont encore plus rares. Cette thèse visait, dans un premier temps, à mettre à l’essai l’analyse de l’accès lexical en discours du Test of Word Finding in Discourse (German, 1991) auprès des enfants francophones d’âge scolaire. Dans un deuxième temps, cette thèse avait pour objectif 1) de mieux comprendre à quelle étape de la production de la parole se situent les difficultés d’accès lexical et 2) de déterminer quelles tâches, parmi celles les plus fréquemment utilisées pour évaluer l’accès lexical, permettent de distinguer des enfants avec et sans difficultés d’accès lexical. Pour répondre à ces objectifs, trois études distinctes ont été développées. Deux études auprès d’enfants francophones participant à des tâches de narration suggéraient que nos participant·e·s produisaient plus de manifestations pouvant être associées à des difficultés d’accès lexical, appelées caractéristiques d’accès lexical (répétitions, révisions, etc.), que celleux de l’échantillon de normalisation du Test of Word Finding in Discourse (German, 1991). Nos résultats suggèrent que la fréquence des caractéristiques d’accès lexical peut être associée au genre de l’enfant, à la langue parlée et au type de tâche (narration vs description d’images). Les résultats de la troisième étude, réalisée auprès d’enfants avec et sans difficultés d’accès lexical, suggèrent qu’en général, les difficultés d’AL impliquent un déficit sémantique, soit à développer des représentations sémantiques précises sur les mots appris. Cela rendrait ensuite ardu l’accès à ces mots. Lorsque les enfants avec difficultés d’accès lexical sont catégorisés en profils de difficultés, un profil de difficultés sémantiques et phonologiques émerge, de même que des profils intermédiaires, sans être spécifiques à des difficultés sémantiques ou phonologiques. Ces résultats concordent avec une autre étude faisant usage des analyses par profils, mais ne concordent pas avec la majorité de la littérature qui propose de catégoriser les enfants avec difficultés d’accès lexical en profils sémantiques et phonologiques. Concernant l’identification de difficultés d’accès lexical, nos résultats suggèrent qu’un questionnaire de dépistage des difficultés d’accès lexical, une tâche de définitions de mots et une tâche de dénomination sont prometteurs. Bien qu’il existe des différences significatives entre les enfants avec et sans difficultés d’accès lexical en discours, nous recommandons d’utiliser le discours pour permettre de décrire les impacts des difficultés d’accès lexical plutôt que de statuer sur leur présence. D’autres études sont nécessaires, surtout impliquant le discours, étant donnés les défis de l’évaluation en discours. / Word-finding difficulties are best described as a frequent inability to find the target words to express oneself. Although word-finding difficulties are commonly reported in children receiving speech and language services, research on these difficulties is sparse. Research with French-speaking children or including narrative tasks is even rarer. French-speaking children and for narration. This thesis aimed to test the use of the Test of Word Finding in Discourse’s analysis of word-finding behaviours with French-speaking school-aged children. This thesis also aimed 1) to better understand at which stage of speech production word-finding difficulties may arise and 2) to determine which task, among the ones frequently used to assess word-finding difficulties, can best distinguish children with and without word-finding difficulties. To this end, three distinct studies were developed. Two studies of French-speaking children participating in a narrative task suggest that our participants produce more word-finding behaviours (repetitions, word reformulations, etc.) than those from the Test of Word Finding in Discourse’s (German, 1991) normalization sample. Our results suggest that the frequency of word-finding behaviours could be associated with a child’s gender or language spoken as well as the type of task (narration vs picture description). The third study, conducted with children with and without word-finding difficulties, suggests that, in general, word-finding difficulties involve a semantic deficit, that is, difficulties developing precise semantic representations when learning new words. This would then make it hard for the child to retrieve these words. When children with word-finding difficulties were categorized into word-finding difficulties profiles, one with both semantic and phonological difficulties emerged, as well as intermediary profiles that were not specific to semantic or phonological difficulties. These results agree with another study of word-finding profiles, but not with the majority of the literature that proposes that children with word-finding difficulties can be categorized into semantic and phonological profiles. Regarding the identification of word-finding difficulties, our results indicate that a screening questionnaire for word-finding difficulties, a word definition task and a naming task are all promising tools. Although there were significant differences between children with and without word-finding difficulties in discourse, we recommend using discourse measures to describe the impacts of word-finding difficulties rather than to identify word-finding difficulties. Further studies are needed, especially ones involving discourse measures, given the challenges in using discourse in speech-language pathology.

Forskning visar att Sveriges unga rapporterar lägst psykiskt välbefinnande av alla EU-länder. Trots att den psykiska ohälsan orsakar höga samhällskostnader och lidande både för de drabbade och dess anhöriga, finns det inget entydigt svar på varför unga svenskar och barn mår så dåligt. Denna studie genomfördes då det finns en brist på forskning gällande främst yngre barn om vilka faktorer som orsakar försämrad psykosocial hälsa. En kvalitativ metod valdes för att fånga skolkuratorers och skolsköterskors subjektiva upplevelser av vilka faktorer som orsakar psykosociala problem hos barn i grundskoleålder. Det insamlade materialet har bearbetats med hjälp av en tematisk analys. Resultaten visar att faktorer som sociala medier, föräldrars engagemang, föräldrars skilsmässa, neuropsykiatriska diagnoser, ångest, skav i kamratskap, ökade krav på prestation och den fysiska skolmiljön hade betydelse för barns psykosociala hälsa. / Research shows that Sweden's young people report the lowest psychological well-being of all EU countries. Despite the fact that mental illness causes high social costs and suffering both for the affected and their relatives, there is no clear answer to why young Swedes and children feel so badly. This study was carried out as there is a lack of research concerning mainly younger children about which factors cause deteriorating psychosocial health. A qualitative method was chosen to capture school counselors' and school nurses' subjective experiences of which factors cause psychosocial problems in children of primary school age. The collected material has been processed using a thematic analysis. The results show that factors such as social media, parental involvement, parental divorce, neuropsychiatric diagnoses, anxiety, friction in friendships, increased performance demands and the physical school environment were important for children's psychosocial health.

Introduction The health of young people in England is an area of concern, nationally and internationally. This has prompted a range of strategies and policies to try to address how health may be improved. However, there has not necessarily been agreement as to how this should best be done. There appeared a case for consideration of alternative or additional approaches to health promotion. Aim This research aimed to construct an assets based model to shape health promotion practice and policy for young people in England. Methods A narrative synthesis was undertaken and highlighted the lack of information regarding which assets might be important for young people’s health in England. This programme of research was developed from those initial findings. Quantitative and qualitative methods were employed to gain a more comprehensive understanding than could be gained by individual methods. This mixed methods research involved secondary data analysis of the Health Behaviours in School Aged Children (HBSC) dataset using regression analysis to identify the assets associated with life satisfaction for English youth. Focus groups and interviews were employed to capture the views of young people regarding assets, health and health promotion. Findings were discussed with practitioners to gather their ideas as to the potential of an assets approach. The different research methods were drawn together by the underpinning theoretical frameworks provided by Assets models and the New Social Studies of Childhood. Results Two themes emerged from the narrative synthesis providing suggestions for health promotion; the ecological approach acknowledged the range of settings that young people inhabit, whilst the holistic approach recognised the interrelationship between risks and assets. Critical discussion consolidated the research findings to propose a list of health promoting assets for young people in England; constructive relationships, safety, positive attributes, independence and opportunity. These findings were brought together into a descriptive model to guide health promotion policy and practice (Figure 1). Constructive relationships appeared as a core asset, providing a foundation from which young people could develop. Having positive attributes was also fundamental to this process, which emphasises the importance of promoting physical and mental health simultaneously. Safety was the third core asset identified through the research strands. There was variation between young people regarding the definition of, and priority assigned to, the additional assets of independence and opportunity. Figure 1: Assets model to shape health promotion with young people Conclusion This research contributes to previous work in the field of assets models by providing new insight into the relatively little researched area of assets approaches to health promotion with young people in England. The complex interrelationships between mental health and assets have been highlighted; providing challenge to frameworks that focus on the quantitative accumulation of assets. The inclusion of young people’s perspectives provided new depth to previous theoretical models and interpretation of quantitative findings. The variation highlighted within this research raises implications for tackling health inequalities. This assets based model provides a framework to shape professional practice and policy thus providing the potential to improve young people’s health and wellbeing in a sustainable and non-stigmatising way.

La présente recherche explore les conséquences d’une perte auditive périphérique sur le traitement de l’information auditive. Des études ont montré que les enfants malentendants ont de la difficulté à effectuer des tâches d’écoute complexes. De plus, des études menées auprès d’adultes malentendants montrent que l’activité corticale associée à l’écoute de stimuli auditifs est différente de celle d’adultes entendants. Cependant, les résultats de ces études ne mettent pas en lumière la nature des difficultés de traitement de l’information auditive des enfants malentendants. Cette recherche examine donc cet aspect en ayant recours à des mesures comportementales et neurophysiologiques. Les données ont été recueillies auprès de 40 enfants âgés de 9 à 12 ans : 12 enfants ayant une surdité neurosensorielle, 12 enfants ayant trouble de traitement auditif et 16 enfants normo-entendants. Les enfants ont reproduit dans l’ordre des séquences de deux, trois et cinq stimuli verbaux ou non verbaux avec un intervalle interstimuli de 425 ms. Les enfants ont également reproduit des séquences de deux stimuli avec un intervalle interstimuli de 20 et 1000 ms. Enfin, les enfants ont été soumis à des mesures neurophysiologiques à partir de potentiels évoqués auditifs de latence longue et de négativité de discordance avec des paires de stimuli verbaux et non verbaux. Les résultats obtenus permettent d’avancer que les participants du groupe d’enfants malentendants ont un trouble spécifique de traitement auditif. En effet, les résultats de la tâche comportementale montrent que les enfants malentendants ont de la difficulté à traiter des séquences de stimuli lorsque ceux-ci sont verbaux et acoustiquement similaires. Quant aux données neurophysiologiques, les résultats ont démontré que l’amplitude de l’onde tardive N2 était réduite chez les enfants malentendants comparativement à celle de l’onde N2 des deux autres groupes d’enfants. Cette onde pourrait être considérée comme étant un marqueur neurophysiologique reflétant l’influence d’une perte auditive sur le traitement auditif central. De plus, l’amplitude de l’onde de négativité de discordance pourrait être aussi un marqueur pour distinguer les enfants malentendants de ceux ayant un trouble de traitement auditif.Mots-clés : organisation séquentielle auditive, potentiels évoqués auditifs de latence longue, négativité de discordance, enfants malentendants d’âge scolaire / The present research explores the effects of peripheral hearing loss on central auditory processing. Previous studies showed that children with hearing loss have significant difficulties in performing complex listening tasks. Moreover, studies of adults with hearing loss revealed that cortical activity related to listening to acoustic stimuli differed from that of adults without hearing loss. However, results of these studies do not clarify the nature of the difficulties in processing auditory information among children with hearing loss. The present research examines this issue using behavioural and neurophysiological measures. Behavioural and neurophysiological data were collected with forty 9- to 12-year-old children: 12 with hearing loss, 12 with central auditory processing disorder (CAPD) and 16 with normal hearing. Children repeated, in order, two, three, and five verbal and nonverbal stimuli with an interstimulus interval of 425 ms. They also repeated sequences of two stimuli with an interstimulus interval of 20 or 1000 ms. Finally, late-latency auditory evoked potentials and mismatch responses were recorded in the participants using pairs of verbal and nonverbal stimuli. Results suggest that children with hearing loss have a specific central auditory processing disorder. Results of the behavioural task showed that children with hearing loss had difficulty processing sequences of stimuli when the stimuli were verbal as well as similar and complex acoustically. As for the neurophysiological data, results indicated that the amplitude of late N2 waveform was smaller in children with hearing loss than in the other two groups of children. The N2 waveform has the potential to be a neurophysiological marker revealing the influence of hearing loss on central auditory processing. Moreover, the amplitude of the mismatch response could be another marker to distinguish the children with hearing loss from those with central auditory processing disorder. Keywords : Auditory sequential organization, late-latency auditory evoked potentials, mismatch responses, school-aged children with hearing loss

La présente recherche explore les conséquences d’une perte auditive périphérique sur le traitement de l’information auditive. Des études ont montré que les enfants malentendants ont de la difficulté à effectuer des tâches d’écoute complexes. De plus, des études menées auprès d’adultes malentendants montrent que l’activité corticale associée à l’écoute de stimuli auditifs est différente de celle d’adultes entendants. Cependant, les résultats de ces études ne mettent pas en lumière la nature des difficultés de traitement de l’information auditive des enfants malentendants. Cette recherche examine donc cet aspect en ayant recours à des mesures comportementales et neurophysiologiques. Les données ont été recueillies auprès de 40 enfants âgés de 9 à 12 ans : 12 enfants ayant une surdité neurosensorielle, 12 enfants ayant trouble de traitement auditif et 16 enfants normo-entendants. Les enfants ont reproduit dans l’ordre des séquences de deux, trois et cinq stimuli verbaux ou non verbaux avec un intervalle interstimuli de 425 ms. Les enfants ont également reproduit des séquences de deux stimuli avec un intervalle interstimuli de 20 et 1000 ms. Enfin, les enfants ont été soumis à des mesures neurophysiologiques à partir de potentiels évoqués auditifs de latence longue et de négativité de discordance avec des paires de stimuli verbaux et non verbaux. Les résultats obtenus permettent d’avancer que les participants du groupe d’enfants malentendants ont un trouble spécifique de traitement auditif. En effet, les résultats de la tâche comportementale montrent que les enfants malentendants ont de la difficulté à traiter des séquences de stimuli lorsque ceux-ci sont verbaux et acoustiquement similaires. Quant aux données neurophysiologiques, les résultats ont démontré que l’amplitude de l’onde tardive N2 était réduite chez les enfants malentendants comparativement à celle de l’onde N2 des deux autres groupes d’enfants. Cette onde pourrait être considérée comme étant un marqueur neurophysiologique reflétant l’influence d’une perte auditive sur le traitement auditif central. De plus, l’amplitude de l’onde de négativité de discordance pourrait être aussi un marqueur pour distinguer les enfants malentendants de ceux ayant un trouble de traitement auditif.Mots-clés : organisation séquentielle auditive, potentiels évoqués auditifs de latence longue, négativité de discordance, enfants malentendants d’âge scolaire / The present research explores the effects of peripheral hearing loss on central auditory processing. Previous studies showed that children with hearing loss have significant difficulties in performing complex listening tasks. Moreover, studies of adults with hearing loss revealed that cortical activity related to listening to acoustic stimuli differed from that of adults without hearing loss. However, results of these studies do not clarify the nature of the difficulties in processing auditory information among children with hearing loss. The present research examines this issue using behavioural and neurophysiological measures. Behavioural and neurophysiological data were collected with forty 9- to 12-year-old children: 12 with hearing loss, 12 with central auditory processing disorder (CAPD) and 16 with normal hearing. Children repeated, in order, two, three, and five verbal and nonverbal stimuli with an interstimulus interval of 425 ms. They also repeated sequences of two stimuli with an interstimulus interval of 20 or 1000 ms. Finally, late-latency auditory evoked potentials and mismatch responses were recorded in the participants using pairs of verbal and nonverbal stimuli. Results suggest that children with hearing loss have a specific central auditory processing disorder. Results of the behavioural task showed that children with hearing loss had difficulty processing sequences of stimuli when the stimuli were verbal as well as similar and complex acoustically. As for the neurophysiological data, results indicated that the amplitude of late N2 waveform was smaller in children with hearing loss than in the other two groups of children. The N2 waveform has the potential to be a neurophysiological marker revealing the influence of hearing loss on central auditory processing. Moreover, the amplitude of the mismatch response could be another marker to distinguish the children with hearing loss from those with central auditory processing disorder. Keywords : Auditory sequential organization, late-latency auditory evoked potentials, mismatch responses, school-aged children with hearing loss

ABSTRACT In standardized tests, culturally and linguistically diverse (CLD) children usually perform at similar level as children with language impairment (LI), largely due to the fact that most standardized assessment tools have been developed on the basis of a monolingual population with mainstream backgrounds. As an alternative to standardized tests, a dynamic assessment approach has been advocated to assess the language abilities of CLD children. Dynamic assessment is considered to reduce test bias, since it measures the child's learning potential (i.e. modifiability) instead of the child's static performance on a certain task at one point in time. The aim of the present study was to explore if dynamic assessment was as successful in a Swedish context as it has been in American contexts. The study had a test-teach-retest design with every session conducted individually. Five typically developing monolingual Swedish children, aged six to seven, were asked to tell two stories on the basis of two comparable picture sequences. In between, a teaching session, that consisted of a mediated learning experience (MLE), was conducted. During the MLE session, a certain aspect of storytelling, chosen from the analysis of the test narrative, was targeted and 5 trained. After the MLE session, the child's modifiability was measured using five-point scales for teaching effort and student responsiveness. The test and retest results were then compared. The DA results indicated that three of the children were typically developing and that two of the children, contrary to fact, had some kind of language learning disability. The material used was found to have limitations in its narrow rating criteria and complicated scripts for the MLE sessions. The conclusions were that (a) the assessment was influenced by the examiner's subjective rating and therefore unable to show the children's true narrative abilities, (b) that the MLE material was not fully suited for younger school aged children and (c) not suitable in the Swedish context. Further studies have to be made with a larger number of children, including CLD children, to be able to tell whether dynamic assessment is a suitable method in Swedish contexts or not. A special DA material for Swedish assessments would then be useful. / Typiskt utvecklade barn med atypisk kulturell bakgrund, flerspråkighet eller låg socioekonomisk status (dvs. barn med språkskillnad) som genomgår språktestning med standardiserade tester förefaller prestera lika på dessa som barn med språkstörning. Detta pga. att de flesta standardiserade tester är skapade för enspråkiga barn med typisk bakgrund. Barnen med språkskillnad riskerar således att bli feldiagnostiserade med språkstörning. Dynamic assessment (DA) har i främst amerikanska studier visat sig vara en fördelaktig bedömningsmetod vid sådana fall, då den har lyckats särskilja barn med språkskillnad från barn med språkstörning. Detta eftersom DA, till skillnad från standardiserade tester, inte undersöker barnets förmåga att klara en viss uppgift vid ett tillfälle, utan istället undersöker barnets förmåga att lära sig att klara av en uppgift (dvs. barnets modifierbarhet). Föreliggande studie avsåg att undersöka hur DA av narrativ förmåga fungerade med svenskspråkiga typiskt utvecklade barn i åldern sex-sju år. Deltagarna genomgick individuell DA med test-teach-retest-design och fick först berätta en saga till en bildsekvens. En aspekt av narrativ förmåga, t.ex. kausalitet, valdes sedan ut och tränades vid en undervisningssession som utgjordes av en så kallad mediated learning experience (MLE). Barnets modifierbarhet under MLE skattades med hjälp av femgradiga skalor för läraransträngning och barnets responsivitet. Därefter följde retest, där varje barn fick berätta en saga till en ny bildsekvens. En jämförelse mellan test och retest skedde. Resultaten visade att tre av deltagarna fick de resultat som förväntats av barn med typisk utveckling. Två deltagare fick resultat som indikerade språkstörning. Materialet som användes bedömdes ha snävt satta bedömningskriterier och svåranvända skript för undervisningssessionerna. Slutsatserna som drogs var (a) att DA-materialet som användes gav en mycket subjektiv bedömning som inte stämde överens med barnets faktiska narrativa förmåga, (b) att MLE-skripten inte var designade på ett helt barnvänligt sätt och (c) inte passade i svensk kontext. Fler och större studier behöver göras för att kunna ge en tydligare bild av hur DA fungerar i svensk kontext. I detta fall vore ett svenskt DA-material önskvärt.

Abstract Children with hearing disorders face particular challenges in developing their vocabulary. They often have difficulty in acquiring spoken language at the same rate as their normal hearing peers, despite progress regarding hearing aid technology.Aim. The main aim of this study was to describe aspects of expressive vocabulary for school-age children with mild to moderately severe hearing impairment (HI-group), and to compare with results for a group of normal hearing children (NH-group) of the same age. One aim was also to test whether there were differences, between children with hearing impairment and normal hearing children, in the ability to repeat non-words and sentences and to compare these results within each group with results on two newly constructed vocabulary tests. An additional aim was also to investigate whether background factors, such as age of diagnosis, age of hearing aid fitting and amount of hearing aid use, were related to results on vocabulary tests.Method: 24 children age 7-11 with mild to moderately severe bilateral hearing impairment, and 82 normal hearing children of similar age, were included in this cross-sectional study. Assessments of picture naming, word definition, repetition of nonwords and repetition of sentences were performed with all participants. Age, hearing, use of hearing aid and parents' level of education were included in the analyzes.Results: The HI-group performed significantly below the results of the NH-group regarding picture naming, word definition, nonword repetition and repetition of sentences. The HI-group also produced significantly more words with deficiencies in phonological structure, than the NH-group, during picture naming. In the HI-group there was a significant positive correlation between amount of hearing aid use per day and nonword repetition. Age of diagnose and age of hearing aid fitting did not show statistically significant associations with outcomes on the language tests in this study.Conclusion: Despite the technical development of hearing aids, the expressive vocabulary in school-aged children with bilateral, mild to moderately severe, hearing impairment, does not reach the same level as in normal-hearing peers, although there is a variation within the group. Amount of hearing aid use seems to be important for the ability to perceive new words.

No description available.

No description available.

Cette thèse examine le développement du langage des enfants sourds qui ont reçu un implant cochléaire (IC) en bas âge. Une première étude rapporte une revue systématique qui avait pour but d’évaluer les connaissances actuelles concernant le développement du vocabulaire et de la grammaire chez les enfants qui ont reçu un IC avant l’âge de trois ans. Vingt-huit études ont été sélectionnées; une analyse descriptive de même qu’une méta-analyse ont été effectuées séparément pour chaque aspect du langage évalué (vocabulaire et grammaire, aspect réceptif et expressif). Au résultat, en dépit de la variabilité observée dans les études, il appert que l’implant cochléaire influence positivement le développement langagier; toutefois, seule une minorité de participants aux études a atteint des niveaux de langage comparables à ceux d’enfants entendants de même âge chronologique. La majorité des enfants continuent de présenter divers degrés de retard de langage, tant au plan réceptif qu’expressif, et ce, après jusqu’à cinq années de port de l’appareil. Les résultats suggèrent aussi, malgré la variabilité observée dans les études, que les bénéfices langagiers sont influencés par le fait de recevoir l’implant à deux ans plutôt qu’à trois ans. À partir des tendances retrouvées dans la littérature, les habiletés de vocabulaire et de grammaire chez 27 enfants qui ont reçu l’implant cochléaire en bas âge (entre 8 et 28 mois) ont été comparées avec celles d’un groupe d’enfants entendants, en utilisant des outils d’évaluation standardisés. Alors que les résultats de groupe montrent que les enfants qui reçoivent un IC autour de l’âge de deux ans atteignent des niveaux de langage dans les limites de la normale, les résultats individuels d’un sous-groupe formé de enfants les plus âgés font état de quatre profils de développement, soit des niveaux de langage dans les limites de la normale pour l’ensemble des composantes, un retard généralisé à l’ensemble des composantes, des habiletés lexicales dans la norme assorti d’un retard morphosyntaxique et enfin un profil atypique montrant des disparités importantes à travers les composantes du langage. Dans trois des quatre profils, la compréhension des phrases était particulièrement faible. Ces résultats suggèrent que le fait de recevoir un implant cochléaire entre l’âge d’un et deux ans ne garantit pas l’atteinte de niveaux de langage dans les limites de la normale malgré une expérience de port de l’appareil d’une durée appréciable. Une étape antérieure du développement linguistique a été examinée de plus près dans la troisième étude. La taille et la composition du vocabulaire expressif de onze enfants ayant reçu un IC à un âge moyen de 15 mois ont été comparées à celles de l’échantillon d’enfants entendants ayant servi à établir les normes en français québécois pour le questionnaire Mots et énoncés des Inventaires MacArthur-Bates du développement de la communication (IMBDC). Les scores d’âge équivalent selon la taille totale du vocabulaire des enfants avec IC étaient supérieurs à l’âge auditif (correspondant à la durée de port de l’appareil) mais inférieurs à l’âge chronologique. La représentation grammaticale en fonction de la taille du vocabulaire des enfants avec IC suit la tendance observée dans la norme. Ces résultats suggèrent que le profil lexical des enfants avec implant est très similaire à celui des enfants entendants lorsque le nombre total de mots acquis est le même. Les résultats de cette thèse suggèrent que l’implant peut, de manière générale, avoir un effet « normalisant » sur le langage ; toutefois, il semble que l’amélioration de l’accès auditif ne suffise pas pour rattraper à coup sûr le niveau de langage des pairs entendants dans l’ensemble des composantes du langage. Alors que les habiletés lexicales se rapprochent du profil typique, les habiletés de compréhension morphosyntaxique sont fortement atteintes chez une majorité d’enfants, suggérant un profil apparenté à un trouble de langage. / This dissertation examines language development in children who received a cochlear implant (CI) at a young age. A systematic review studied the main outcomes reported in the literature concerning vocabulary and grammar development of profoundly deaf children who received a CI before the age of 3 years. A total of 28 studies met inclusion criteria. Studies varied widely in participant characteristics, assessment tools, study designs, and overall methodological quality. A descriptive synthesis and a meta-analysis were completed separately for each language domain (receptive and expressive vocabulary and grammar). Results showed a positive impact of cochlear implantation on language development, but only a minority of children achieved language levels on par with their hearing age-mates. The majority of the children involved in the primary studies continued to exhibit varying levels of delay in receptive and expressive vocabulary and grammar after up to 5 years of cochlear implant use. Results showed that the minority of children who achieved language levels on par with hearing children were likely to have received their implant by the age of two, thus suggesting that age at implantation influences language achievement. A second study examined receptive and expressive vocabulary and grammar achievement of 27 French-speaking children who received a CI between the age of 8 and 28 months. Standardized measures were administered and the language levels attained by children with CIs were compared with those of the normative sample of same-age hearing peers for each measure. As a group, children exhibited language levels within normal limits on all standardized language measures. Examination of individual patterns in a subgroup of children revealed 4 different language profiles: a) normal language levels in all domains, b) general language delay, c) vocabulary within the norm with morphosyntactic delay, and d) an atypical profile (discrepancies across language domains). In three of these profiles, comprehension of sentences was impaired. Findings suggest that receiving a cochlear implant between the age of 1 and 2 years does not ensure that language abilities will be within normal limits after up to 6 years of experience with the implant. An earlier stage of formal language was examined in the third study. Vocabulary size and grammatical composition in 11 children who received their CI at a mean age of 15 months were compared to that of the Quebec French normative sample for the Words and Sentences questionnaire of the MacArthur-Bates Communicative Development Inventories (MBCDI). Results showed that age equivalent scores according to total vocabulary size were superior to hearing age (equivalent to the duration of device use) but inferior to chronological age. Distribution of grammatical categories according to vocabulary size followed the same pattern as in the normative sample. These results suggest that the lexical profile of children with implants was very similar to that of normally-hearing children who had the same number of words. Taken together, results of this dissertation suggest that the cochlear implant can have a “normalizing” effect on language. However, they also suggest that improved access to auditory input does not seem sufficient to allow children to attain language levels within normal limits in all components. Whereas early lexical abilities were comparable to typical development, receptive morphosyntactic abilities remain severely impaired in a majority of children.