Phonemic discrimination and the relationship with other linguistic levels in children with typical phonological development and phonological disorder

Purpose: To compare children with typical language development (TLD) and evolutional phonological disorder (EPD) regarding the phonemic discrimination and the linguistic performance of language levels (morphological, syntactic, semantic, and perceptual and productive vocabulary). Methods: The sample comprised 36 children, aged between 5 years and 7 years and 11 months, with TLD and EPD. Children with EPD were awaiting care in the speech units of two higher education institutions and children with TLD were screened in public schools. For inclusion in the study, the criteria were the following: being authorized by the informed consent and being within the required age group. After the inclusion of children, all subjects underwent the phonemic discrimination test with figures, to the average phrase value test, which assesses the morphosyntactic and semantic/lexical aspects, and the expressive vocabulary test. For statistical evaluation of the influence of DOI: 10.1590/2317-1782/20152014168 Trabalho realizado no Programa de Pós-Graduação em Distúrbios da Comunicação Humana, Universidade Federal de Santa Maria – UFSM – Santa Maria (RS), Brasil. (1) Programa de Pós-Graduação em Distúrbios da Comunicação Humana, Universidade Federal de Santa Maria – UFSM – Santa Maria (RS), Brasil. (2) Curso de Fonoaudiologia e Programa de Pós-Graduação em Distúrbios da Comunicação Humana, Universidade Federal de Santa Maria – UFSM – Santa Maria (RS), Brasil. (3) Curso de Fonoaudiologia, Universidade Federal de Ciências da Saúde Porto Alegre – UFCSPA – Porto Alegre (RS), Brasil. Conflito de interesses: nada a declarar. 237 Discriminação fonêmica e os níveis linguísticos CoDAS 2015;27(3):236-41 INTRODUÇÃO A discriminação fonêmica é um processo de diferenciação de sons acusticamente semelhantes com frequência, duração e/ou intensidade diferentes em que precisamente nessas diferenças se encontra a informação transportada pelo som. Logo, é a capacidade do indivíduo de perceber diferenças acústicas mínimas, presentes no discurso, sendo considerado como um aspecto fundamental para a produção correta dos sons da fala — a representação mental e o armazenamento de estímulos linguísticos são estabelecidos por meio da recepção, análise e organização de informações pelo processamento auditivo. Portanto, a capacidade de discriminar fonemas é fundamental para o início desse processo e da aquisição fonológica. Para a aquisição dos sons da fala, é essencial que essa habilidade de perceber diferenças mínimas entre as características distintivas que acontecem por oposições binárias de valor se estabeleça. Essas são descritas, com valores, para as seguintes combinações: [+soante] e [-soante], [+aprox.] e [-aprox.], [+cont.] e [-cont.], [+voz] e [-voz], e valores monovalentes de traços de lugar: [labial] e [cor.], [cor.] e [dorsal], [labial] e [dorsal], [cor.+ant.] e [cor.-ant.]. Para tanto, são necessárias condições para que esses sons sejam discriminados, como a integridade das estruturas orgânicas envolvidas na detecção, recepção e condução do som, além de processos de interpretação para que haja a percepção do som. As crianças devem aprender a discriminar sons específicos para que sua fala seja adequada ao padrão-alvo adulto de sua língua materna. Para que a criança aprenda os sons da fala, é necessário que haja condições orgânicas para que os fonemas sejam discriminados. Em outras palavras, depende da integridade e maturação neuromuscular. Assim, a integridade e o desenvolvimento dos aspectos sensoriais e motores estão envolvidos no processo de aprendizagem de uma língua padrão. É por meio da associação de aspectos auditivos para o gesto articulatório motor que os fonemas da língua são memorizados e ficam prontos para ser utilizados no discurso. Assim sendo, a aquisição fonológica depende da capacidade perceptiva da criança em ouvir sons inseridos em palavras e ser capaz de analisá-los de acordo com suas características acústicas e articulatórias. A aquisição dos fonemas de uma língua necessita, ainda, do conhecimento, por parte da criança, sobre o sistema de contrastes válido para aquela comunidade linguística. Embora cada fonema não possua significado em si, a sua omissão, inserção, translocação ou substituição na palavra pode gerar mudanças de sentido. Na aquisição fonológica típica, o domínio do sistema fonológico de uma língua-alvo é atingido de forma espontânea, em uma sequência de idade comum para a maioria das crianças (de 4 a 6 anos de idade), e parece obedecer a um sistema universal de hierarquia e restrições, responsável, em última análise, por prever uma ordem de aquisição dos fonemas de uma determinada língua, podendo apresentar variações individuais. Assim, os traços e as co-ocorrências mais simples dos mesmos são aprendidos antes daqueles mais complexos. Por sua vez, a aquisição fonológica desordenada e/ou atrasada é aquela em que a adequação do sistema fonológico não é espontaneamente alcançada e/ou não é conseguida na mesma sequência e no mesmo tempo observados na maioria das crianças. A fase de maior expansão do sistema fonológico ocorre entre 1 ano e 6 meses e 4 anos, quando há um aumento do inventário fonético/fonológico das crianças, possibilitando a produção de palavras polissilábicas e de estruturas silábicas mais complexas. A idade de 4 anos é considerada um marco importante para a conclusão do inventário fonológico, sendo que, nessa idade, a grande maioria das crianças já adquiriu os contrastes do sistema fonêmico adulto e usa a língua para se comunicar efetivamente. Portanto, deficiências na discriminação auditiva durante o período de desenvolvimento fonológico podem comprometer a constituição e organização dos sons da fala. Além disso, a dificuldade de compreender os sons da fala pode estar relacionada com a ocorrência de alterações fonológicas durante o período da infância. Com base nisso, pode-se supor que a desordem na produção de sons está relacionada à dificuldade de discriminação fonêmica. Concomitantemente com a aquisição fonológica, a aquisição do vocabulário se inicia, aproximadamente, no momento que a criança aprende a relacionar corretamente sequências de sons (significantes) a conjuntos de situações (referentes), utilizando as representações mentais (significados) correspondentes como intermediárias. A construção dessas representações mentais é um trabalho que a criança deve realizar para descobrir as regularidades que governam a utilização dos lexemas por parte do adulto. A partir dessas primeiras palavras, o vocabulário utilizado pela criança começa a se ampliar, até que, por volta dos 24 meses, ocorre um fenômeno conhecido como explosão de vocabulário. Na verdade, tal fenômeno se relaciona com aspectos cognitivos da criança, que, por meio da formação de conceitos, dá nome às coisas que a cercam. Nesse sentido, o variables, Spearman’s nonparametric correlation coefficient was used, and for comparison between the groups regarding evaluation performance, Mann-Whitney test was used. Results: Only children with EPD showed influence of phonemic discrimination in other linguistic levels. In addition, a significant difference was observed between the performances of both groups in relation to the phonemic discrimination, vocabulary, and all levels assessed in the questions and description modality. Conclusion: The comparison between both groups showed a statistically significant difference in phonemic discrimination and morphosyntactic and lexical/semantic development, with better performance in the TLD group. 238 Freitas CR, Mezzomo CL, Vidor DCGM CoDAS 2015;27(3):236-41 vocabulário está intimamente ligado com as experiências vividas pelo sujeito, uma vez que a curiosidade da criança e suas experiências diante de novas situações explicam a aquisição e a busca pelas palavras desconhecidas. Espera-se, portanto, uma considerável variação individual nos padrões do crescimento do vocabulário inicial. Esse padrão de aquisição se deve ao fato de o vocabulário ser um sistema aberto, isto é, em constante expansão ao longo de toda a vida do sujeito. Do ponto de vista linguístico, o aprendizado de novas palavras parece estar intimamente relacionado com a aquisição da sintaxe, da morfologia e da fonologia, pois, sem as palavras, os falantes não conseguem concretizar padrões sintáticos, estruturas morfológicas ou até mesmo padrões sonoros de sua língua. A hipótese deste estudo é a de que, uma vez que os subsistemas estão estreitamente relacionados, as alterações no subsistema fonológico podem influenciar alterações nos subsistemas semântico, morfossintático e lexical, sendo de grande relevância clínica e teórica a realização de estudos com tais relações. Desse modo, espera-se que estes achados contribuam para que se tenham subsídios adequados para a realização da avaliação, do diagnóstico e do tratamento das alterações de fala e linguagem, prevenindo ou minimizando possíveis alterações em outras áreas que possam estar envolvidas. A partir do exposto acima, o presente estudo teve como objetivo comparar crianças com desenvolvimento típico de linguagem (DTL) e com desvio fonológico evolutivo (DFE) quanto à discriminação fonêmica e ao desempenho linguístico dos níveis da linguagem (morfológico, sintático, semântico e vocabulários perceptivo e produtivo) de crianças na faixa etária de 5 a 7 anos e 11 meses.


INTRODUCTION
Phonemic discrimination is a process of differentiation of acoustically similar sounds with different frequency, duration, and/or intensity when the information carried by the sound depends on these differences (1) .Therefore, it is the individual's ability to perceive minimum acoustic differences present in the discourse (2,3) , and it is considered as a key aspect for the correct production of speech sounds -the mental representation and storage of linguistic stimuli are established through reception, analysis, and organization of information by auditory processing (1)(2)(3)(4)(5) .Therefore, the ability to discriminate phonemes is critical to the beginning of this process and for phonological acquisition (3) .
For the acquisition of speech sounds, it is essential that the ability to perceive minor differences between the distinctive features that occur by binary oppositions is established.These are described with values for the following combinations: [+sonorant] and [-sonorant], [+ approx.]and [-approx.]] and [coronal-ant.] (5).Therefore, conditions are necessary so that these sounds are discriminated (6) , such as the integrity of organic structures involved in the detection, reception, and conduction of sound, as well as interpretation processes, so that the sound is perceived (7) .
Children should learn to discriminate specific sounds so that their speech is adjusted to the adult target-pattern of their mother tongue (5) .For the child to learn the sounds of speech, there must be organic conditions for the discrimination of phonemes.In other words, it depends on neuromuscular integrity and maturation.Thus, the integrity and the development of sensory and motor aspects are involved in the learning process of a standard language (8) .It is through the association of auditory aspects for the motor articulatory gesture that language phonemes are memorized and become available for use in speech (3) .Thus, phonological acquisition depends on the child's perceptive capacity to hear sounds embedded in words and be able to analyze them according to their acoustic and articulatory characteristics.
The acquisition of the phonemes of a language also needs knowledge by the child of the contrast systems valid for that language community.Although each phoneme has no meaning in itself, its omission, insertion, translocation, or substitution in a word can cause changes in meaning.
In typical phonological acquisition, the domain of the phonological system of a target-language is reached spontaneously, in an age sequence that is common for the majority of children (9) (4-to 6-year-olds).It seems to obey a universal system of hierarchy and restrictions, responsible, ultimately, for predicting an order of acquisition of the phonemes of a particular language, but which may present individual variations (10) .Thus, the simpler characteristics and co-occurrences are learned before those more complex (11) .In turn, the disordered and/or delayed phonological acquisition happens when the adaptation of the phonological system is not achieved spontaneously and/or is not achieved in the same sequence and at the same time observed in most children (10) .
The phase of greater expansion of the phonological system occurs between 1 year and 6 months and 4 years, when there is an increase in children's phonetic/phonological inventory, enabling the production of polysyllabic words and more complex syllable structures.The age of 4 is considered an important milestone for the completion of the phonological inventory, and at this age, most children have acquired the phonemic contrasts of the adult phonemic system, and use language to communicate effectively (10) .
Therefore, deficiencies in auditory discrimination during the phonological development period may compromise the establishment and organization of speech sounds (3,7) .In addition, the difficulty in understanding speech sounds may be related to the occurrence of phonological changes during the period of childhood (12) .On the basis of this, it can be assumed that the disorder in the production of sounds is related to the difficulty in phonemic discrimination (1) .
Concurrently with the phonological acquisition, the acquisition of vocabulary begins approximately at the time the child learns how to relate properly sequences of sounds (signifiers) to sets of situations (referents) using corresponding mental representations (meanings) as intermediates.The construction of these mental representations is a job that the child must do to discover the regularities that dictate the use of lexemes by adults (13) .
From these first words, the vocabulary used by the child begins to enlarge, until, at around 24 months, there is a phenomenon known as vocabulary explosion.In fact, this phenomenon is related to the child's cognitive aspects, which, through the formation of concepts, name things that surround the child (14,15) .In this sense, the vocabulary is closely connected with the experiences of the subject because the child's curiosity and coeficiente de correlação não paramétrico de Spearman, e para a comparação dos grupos em relação aos desempenhos das avaliações, o teste U de Mann-Whitney.
experiences in new situations explain the acquisition of and the search for unknown words.Therefore, a considerable individual variation in the growth patterns of the initial vocabulary is expected (16) .This pattern of acquisition is because the vocabulary is an open system (17) , that is, constantly expanding throughout the subject's life (15) .
From a linguistic point of view, learning new words seems to be closely related to the acquisition of syntax, morphology, and phonology because without words, speakers cannot construct syntactic patterns, morphological structures, or even sound patterns of their language (16) .
The hypothesis of this study was that because the subsystems are closely related, changes in the phonological subsystem can influence changes in semantic, morphosyntactic, and lexical subsystems, and the conduction of studies on such relations is of great clinical and theoretical relevance.
Thus, it is hoped that these findings will contribute with adequate resources for the completion of the evaluation, diagnosis, and treatment of speech and language disorders, preventing or minimizing possible changes in other areas that may be involved.
Considering the above, this study aimed to compare children with typical language development (TLD) and evolutional phonological disorders (EPD) regarding the phonemic discrimination and the linguistic performance of language levels (morphological, syntactic, semantic, and perceptive and productive vocabulary) of children aged 5 to 7 years and 11 months.

METHODS
This was an experimental, descriptive, and prospective study, with quantitative analyses.It is linked to a research project filed by the research ethics committee of a federal educational institution, under protocol no.360.535, and has a Certificate of Presentation for Ethical Assessment (CAAE), no.17803713.9.0000.5346.
The study sample consisted of 37 children with TLD and EPD, 13 females and 24 males, with ages ranging from 5 years and 3 months to 7 years and 11 months at the time of initial evaluation.Children with EPD were awaiting care in the speech departments of Speech-Language Pathology and Audiology services linked to two higher education institutions, and children with TLD were screened in public schools that authorized this research.
The criteria for the inclusion of children with TLD were the following: authorization from legal guardians by signing the free and informed consent form, being within the age range stipulated and having TLD.For children with EPD, the main criterion was the diagnosis of EPD.
In addition, children had to be authorized by parents or legal guardians to participate in the study by signing the informed consent and be aged between 5 and 7 years and 11 months.These age limits were established to ensure the diagnosis because the majority of studies indicate that the age of 5 is when the stabilization of the phonological inventory can be expected.In turn, children as young as 8 who show these characteristics raise suspicions about the possible causes for the symptoms, which can be classified as residual speech errors.
The exclusion criteria of children with TLD and EPD were as follows: subjects who had received or were receiving any kind of speech therapy; whose legal guardians have not authorized their participation through the FICF; whose auditory thresholds suggest alterations; and who present oral motor skills and structures that could compromise speech, as well as any evident neurological, emotional, or cognitive damage.
For the diagnosis of EPD, the following were assessed: speech and hearing screening, composed of the initial interview (anamnesis), and clinical observation, composed of phonological and oral and/or written language assessment, and audiological and orofacial motricity assessment.
After confirming the EPD, the children were submitted to data collection for assessment of phonological aspects: this aspect was assessed by spontaneous naming of figures that make up the Phonological Assessment Instrument: Children (Avaliação Fonológica da Criança -AFC) (18) .Then, the contrastive analysis was performed to establish the child's phonological system, and, finally, the Percentage of Consonants Correct-Revised (19) was calculated.For the classification of degrees of phonological disorder, the proposed rates were used, which classify the deviation in mild -MD (86-100%), mildmoderate -MMD (66-85%), moderately-severe -MSD (51-65%), and severe -SD (<50%).
For the evaluation of the productive vocabulary, the two groups (TLD and EPD) underwent further evaluation of morphosyntactic and semantic elements, through the average phrase value test (20) .For this analysis, the oral production of the first five sentences spoken by children was collected and recorded, under three different conditions of enunciation (describing a picture, telling a story, and answering questions).
According to this proposal, the scoring was as follows: nouns and verbs, because they are considered the first to emerge in language acquisition and give meaning to the phrase, were considered semantic elements, with two points given each time they were used; adverbs, adjectives, prepositions, conjunctions, pronouns, and articles were considered syntax elements, and four points were given for each one because the use of these words would show greater grammatical knowledge and linguistic evolution.
In addition, the total score was counted for each sentence, to obtain the total complexity (construction), and the number of words in the sentence was also counted, to obtain its total extension.
The productive vocabulary was assessed in its expressive form (21) , using the expressive vocabulary test.This test has been validated and standardized for the 18 months to 7-yearold age group, originally with 100 items for oral appointment, and the total number of correct answers was counted (for this, exchanges, omissions, and substitutions were disregarded).
To evaluate the perceptual aspects, we used the Phoneme Discrimination Test with Figures (PDTWF) (5) .This tool evaluates the phonemic discrimination of children aged 4 to 8 years old.The test consists of 30 minimal pairs (60 words) and 4 demonstration items.These minimal pairs were organized in Phonemic discrimination and linguistic levels CoDAS 2015;27(3):236-41 40 presentations; of which, 30 are presentations with two different words and 10 are presentations with two equal words.Presentations with two equal words were included in the test to make the participating children pay more attention.
Of the 30 presentations with two different words, 3 differed for the opposition [+/-sonorant], 1 for the opposition [+/-approx.], 3 for [+/-cont.],5 for [+/-voice], 3 for the opposition [coronal+/-ant.], 2 for [labial] x [coronal], 2 for [dorsal] x [coronal], 4 for [dorsal] x [labial], and 7 for the opposition of syllabic structures, of which 2 differ for V x CV, 2 for CV x CVC, and 3 for CV x CCV.The order of presentations follows the same sequence.The test score is obtained by adding one point for each correct answer and zero point for each incorrect answer, or obtained through repetition, totaling 40 points.
For statistical analysis, the methods used were the nonparametric Spearman correlation coefficient, for analyzing the influence of phonological aspects in relation to perceptual and productive vocabularies in both groups, and the Mann-Whitney test to compare the scores between the two groups.In addition, the Spearman correlation coefficient varies between -1 and 1, and the closer it is to these extremes, the greater the association between variables.For these analyses, the statistical program Statistic 9.1 and the significance level of 5% were used.

RESULTS
Table 1 shows the correlation of the performances of TLD and EPD groups, between the perceptual aspect of phonology (phonemic discrimination) and each linguistic variable (syntax, semantics, total construction, and total extension) in the enunciation patterns of language, as well as vocabulary.For this, the Spearman's correlation coefficient was used, in which p-values lower than 0.05 were considered significant, which are highlighted with an asterisk in the table.
Spearman's correlation coefficient indicated that there are differences regarding the influence of phonemic discrimination in other linguistic levels between both groups studied, as only the group with EPD presented significant correlations in these aspects.
As can be seen, there were statistically significant, positive correlations only in the group of children with EPD.These occurred between semantics and syntax in the description modality; in semantics, between the total construction and total extension category of the story modality, which is to say that the greater the capacity of discrimination, the greater the semantic and syntactic development.
Also in the group with EPD, there was a significant negative correlation between discrimination and vocabulary.However, it was negative, indicating that the higher the discrimination, the lower the vocabulary.
However, the TLD group had negative correlation, which indicates that the findings of the correlation are inversely proportional to the growth of the variable (in this case, phonemic discrimination), but without statistical significance.
The analysis of the results between groups highlights the fact that, in the deviations (EPD group), language levels seem to correlate significantly, unlike what occurs in typical development (TLD group), as there was no alterations in discrimination or in any of the subsystems assessed in this population.
Table 2 shows the comparison of the performance of children with TLD and EPD in relation to the perceptual aspect of phonology and other linguistic levels.For this, Mann-Whitney U-test was used, in which p-values lower than 0.05 were considered significant, which are highlighted with an asterisk in the table.Mann-Whitney U-test indicated that there is a significant difference between the performance of children with and without phonological deviations regarding phonemic discrimination, vocabulary, and all levels assessed in the questions and description modalities.There was also no significant difference in the total length of the story modality.In all cases, higher performance scores can be observed in the group with TLD.

DISCUSSION
This study aimed to analyze the influence of phonological aspects in the linguistic performance and correlate phonological perceptive performances with other language levels in children with EPD and TLD, by comparing the scores between the two groups.
From the data analysis, differences were observed in the performance of other linguistic levels in relation to phonemic discrimination between the groups studied.However, only the group with EPD was presented significant correlations in these aspects.
Related to this result, a study on phonemic discrimination in children aged 4 to 7 years and 11 months found that phonemic discrimination disability can be an aggravating factor in cases with EPD, and errors involving the phonological processes may highlight the difficulty of children with speech disorders in discriminating sonority and articulation point (22) .In addition, the percentage of errors found in the auditory discrimination test may be indicative of the association between this ability, phonemic discrimination, and the alterations in speech (23) .
In addition, it is clear that phonemic discrimination is important for the development of language skills, semantic/lexical morphosyntactic, and aspects of children with EPD.This may indicate that the difficulty in auditory discrimination is associated with linguistic alterations, more specifically in speech, such as the linguistic levels (24) .Therefore, this fact undermines stabilization in the phonological system.Another factor that can explain this result is that the difficulty in auditory discrimination may be a causal factor or just an aggravating factor in EPD, a situation improves with increasing age (5,7,23) .
As can be seen in the results explained earlier, phonemic discrimination showed a statistically significant relationship in the population with EPD.This fact corroborates the literature, which states that the auditory discrimination is a relevant factor in the process of typical language acquisition (5) , because, with the perceptive ability to differentiate sounds in words and, consequently, to analyze them according to their acoustic and articulatory characteristics, the process of phonological acquisition takes place concurrently, and often children with alterations in phonology, EPD, present difficulties with this perceptual ability (3,22,25) .
Regarding the comparison between the phonemic discrimination and the relationship with language, there are a limited number of studies.But when compared with the other language levels studied in this article (morphology, syntax, and lexicon), there are no studies that analyze this relationship.
It was found that the higher the auditory discrimination of phonemes in EPD, the better the development of language levels would be.Similarly, another study suggested that phonemic discrimination may be related to the severity of the deviation (1) , which agrees with the findings of this study.
By observing the results, it can see that the group with TLD does not seem to be influenced by the performance of other linguistic levels.A fact that may explain the lack of correlation with this population is that the tasks of PDTWF are seemingly simple, but require attention, discrimination of stimuli, lexical access, and audiovisual association; another fact is that children with normal phonological development reached the maximum values allowed in the test without noticeable difficulty (3) .
Regarding the result that shows statistically significant differences between the vocabulary of children with TLD and EPD, these findings corroborate the literature, which suggests that performance in the vocabulary of children with phonological disorders is lower than that of children with typical phonological development (22,26) .A fact that can explain this is that the relationship between phonological development and early lexical development is so close, it is not possible to separate these two aspects in the early stages of language acquisition (9) .However, these results disagree with the findings of other studies (5,19) , which showed that children with phonological disorders showed similar vocabulary to that of children with typical phonological development.
Regarding the other linguistic subsystems, results showed statistically significant differences between phonology and all modalities (semantics, syntax, total construction, and total extension) of the description and story categories.This fact agrees with the literature, which indicates that these subsystems work together throughout the development of language skills and may undergo mutual influences (15) .These language skills work together, providing effective communication.Other searches related to certain semantic factors influence the phonetic-phonological accuracy (13,17) .Regarding syntax, the phonological subsystem shows direct influence (13) .
Regarding the morphosyntactic and semantic aspects, it can be observed that there is a statistically significant difference regarding the performance of the two groups, EPD and TLD.This fact agrees with the literature, which indicates that the phonological subsystem directly influences the syntax (13) .Another study reported that the meaning of a sentence depends on its syntactic organization, the proper use of morphemes, the acquisition of their meaning, and that the access to the name of an object depends on phonological skills.Other authors also reported that all people have a mental lexicon, which is accessed when you want to represent, through words, an object, an action, an attribute, or an event (15) .Phonemic discrimination and linguistic levels CoDAS 2015;27(3):236-41 In addition, a phonological acquisition deficit may cause difficulties at various levels of language, such as errors that are unexpected for the age and alterations in lexicon (27) .According to another study (27) , pragmatic, semantic, morphosyntactic, and phonological aspects cannot be separated because they act together in the development of language skills.
According to the literature, there is a strong synchronization between the development of the semantic aspect and the phonological aspect.On the one hand, there are children with a small phonetic/phonological repertoire, who tend to have few words stored in their lexicon; on the other hand, there are children with a broad vocabulary and phonetic/phonological repertoire (9) .
In general, it can be observed that phoneme discrimination, or perceptual aspects, is closely related to both the phonology and other language levels.
Regarding the lexicon, studies suggested that the limited phonological repertoire with restrictions in syllabic classes and positions can interfere with good linguistic understanding of communication, as the child ends up using homonyms, producing the same word to refer to different objects (28) .

CONCLUSION
This study met the proposed objective and, through its findings, it could be observed that phonemic discrimination plays an important role in the development of the other linguistic levels studied.Children with phonological disorders presented alterations in phonemic discrimination as in other linguistic levels, being possible to analyze the interrelationship between perception levels and language production.However, no alterations were observed in the population with typical development, for they had no alterations in phonemic discrimination skills as well as in the language levels analyzed.
*CRF was responsible for the drafting of the project, data collection, and analysis; CLM was responsible for the drafting of the project, data analysis, and literature reviews; DCGMV was responsible the drafting of the project, data analysis, and general review.

Table 1 .
Correlation of the performances of perceptual aspects of phonology in relation to other linguistic levels in the groups with typical language development and evolutional phonological disorder 2015;27(3):236-41