ARTICLES
HUMAN DEVELOPMENT

Language regression in autism spectrum disorder: a systematic review

Bárbara Backes^I; Regina Basso Zanon^II; Cleonice Alves Bosa^III

^IFederal University of Rio Grande do Sul, Porto Alegre, RS, Brazil
^IISociedade Educacional Três de Maio, Três de Maio, RS, Brazil
^IIIFederal University of Rio Grande do Sul, Porto Alegre, RS, Brazil

Mailling address

ABSTRACT

A significant portion of children with Autism Spectrum Disorder (ASD) is affected by the loss of language skills. This study's objective was to systematically review papers addressing this phenomenon characterizing the loss of skills in ASD. Bibliographic references were systematically searched in five databases: PubMed, Web of Science, PsycINFO, Lilacs and Scielo. Two independent judges selected and extracted data. The search resulted in 30 studies. Language regression tended to occur around 24 months of age and was generally accompanied by the loss of other skills. Beginning in 2000, studies in the field became more frequent, with a refined operational definition of the phenomenon. Important gaps were identified and can motivate the development of future studies, such as a lack of prospective studies and the need to discuss the conceptual definition of developmental regression.

Key words: autistic disorder; child development: regression; language; review.

Introduction

Individuals with Autism Spectrum Disorder (ASD) present social communication disorders, in addition to repetitive and stereotyped behavior, though the presentation and symptoms vary (American Psychiatric Association [APA], 2013). Likewise, both the nature of ASD and the age when its first signs appear vary among individuals (Zanon, Backes, & Bosa, 2014), as well as how the disorder progresses (Zwaigenbaum, Bryson, & Garon, 2013).

Zwaigenbaum et al. (2013) hold that ASD may begin early, in which symptoms become apparent in the first year of life, or be regressive, in which symptoms are preceded by a period of an apparently typical, development and then followed by a loss of skills previously acquired. There is no consensus, however, whether development attained before the loss of skills is typical, or delays and behavioral changes could have been observed but were missed. Thus, some individuals with ASD present developmental regression (DR) defined as definitive or significant loss of previously acquired skills (Baird et al., 2008; Meilleur & Fombonne, 2009).

One meta-analysis addressed, among other factors, the prevalence, and age when loss of skills take place in ASD (Barger, Campbell, & McDnough, 2013). The authors reviewed 85 papers that represented 29,035 participants with ASD. The weighted mean indicated a prevalence rate of RD of 32.1% and the mean age when losses began was 21.36 months; the last information was based on 24 studies (Barger et al., 2013). Note that the prevalence of DR may vary according to the operational definition of regression, the way it is assessed, and the sample under study (e.g., clinical, survey). In general, however, the loss of skills is a phenomenon that affects a significant number of children with ASD.

Studies show that DR may concomitantly affect different areas of development, resulting in loss of language skills, social interaction and the ability to play (Backes et al., 2013; Baird et al., 2008; Castillo et al., 2008; Jones & Campbell, 2010; Lord, Schulman, & DiLavore, 2004; Luyster et al., 2005; Meilleur & Fombonne, 2009; Tamanaha, Machado, Loebmann, & Perissinoto, 2014; Thurm, Manwaring, Luckenbaugh, Lord, & Swedo, 2014). Explanation for the concomitant occurrence of loss of language, social interaction skills and the ability to play is based on the socio-pragmatic approach in which the acquisition of language is linked to socio-cognitive and socio-interactive bases; that is, the emergence of the linguist symbols (i.e., words) takes place after establishing social and cognitive skills during social interaction (Tomasello, 1999/2003). Hence, the process of symbolism production permeates the development of language and children's play and occurs in social exchanges (Tomasello, 1999/2003). Hence, language, social interaction and play are correlated developmental areas.

Note that language regression, more specifically the loss of words, and not the loss of other skills, tends to be characteristic among children with ASD, in contrast with other conditions, such as the Landau-Kleffner syndrome (Lord et al., 2004; Pickles et al., 2009; Shinnar et al., 2001; Valente & Valério, 2004; Wilson, Djukic, Shinnar, Dharmani, & Rapin, 2003), and this is the focus of this study. Considering a lack of consensus regarding the conceptual definition of language regression, it has been operationally defined, based on the measures used. These measures include the Autism Diagnostic Interview-Revised (ADI-R), an instrument considered to be the gold standard for the diagnostic assessment of ASD. This instrument operationally defines language regression as the loss of communicative (spontaneous) use of three to five words, except for of "papa" and "mama" (Lord, Rutter, & LeCouteur, 1994). Additionally, ADI-R uses two time-parameters that refer to a period equal to or longer than three months, one of which is the period before the loss (i.e., period in which words previously learned were used communicatively) and the other is the duration of loss (i.e., time during which the child did not use the words lost). For this reason, in this study, we use the term 'loss of words'. Some studies report oral language as the skill most commonly affected by RD; loss of words occurs in approximately 20% of the individuals with ASD (Backes et al., 2013; Lord et al., 2004).

In this sense, we consider 'word' the linguistic symbol that consistently represents a given object or event, even if its form of presentation (or structure) differs from the conventional use but is close to the model provided to the child (e.g., 'eat' instead of 'food', 'woof-woof' instead of 'dog'). Obviously, one is not neglecting the different facets of oral language acquisition, which involves various complexities, such as inferring different meanings of linguistic symbols, grammatical categorization and understanding potential linguistic constructions related to certain words (Clark, 2012). Acquisition of the first words, however, occurs when children began to understand the meaning assigned to them, in general beginning at the end of the first year of life (Clark, 2012).

Note that language regression, and more specifically, the loss of words, has been highlighted as a potential early indicator of ASD for that subgroup whose onset pattern might be termed a "regressive start" (APA, 2013; Lord et al., 2004). This is due to the fact this loss more frequently occurs among children with ASD, as some comparative studies, using controls with Down Syndrome and specific language disorder, report (Pickles et al., 2009; Shinnar et al., 2001; Thurm et al., 2014; Wilson et al., 2003).

Nonetheless, despite the apparent importance of linguistic regression, and more specifically loss of words, both regarding diagnosis and intervention, this phenomenon remains unclear, which results in a lack of specific preventive and therapeutic measures. Additionally, results of studies in the field vary greatly. To clarify the heterogeneity of findings in the field of language regression, and therefore, to understand better the phenomenon itself, some questions need to be considered. For how long has language regression been studied? How has it been defined and investigated? Are there factors that can explain the occurrence of this phenomenon? Is it an isolated factor or is it accompanied by the loss of other skills? How did language develop before its loss? Because it is a still expanding field of research, these (and many other) questions remain unanswered. In an attempt to find the answers to such questions, this study was designed to provide tools for professionals, clinicians, and scholars, as well as to identify the occurrence and characteristics of language regression, more specifically the loss of words in the population with ASD. Note that, even though reviews addressing DR in ASD have been conducted, no systematic literature reviews were found specifically focusing on language regression.

Therefore, this study's objective was to systematically review studies addressing language regression in ASD, characterizing the loss of skills in this population (i.e., definition of the phenomenon, measures to investigate language regression, age when loss takes place, duration of loss, associated factors, co-occurrence with the loss of other abilities), focusing on the loss of words. Additionally, this study presents information regarding the development of language prior to its loss, whenever such information is available.

Method

Material analysis

Papers addressing language regression in ASD were included. A search was conducted in Brazilian and international databases, considering studies published up to February 17^th, 2016. No chronological restriction was established, and papers written in Spanish, English, Portuguese or French from 1985 to 2014 were analyzed.

Procedures

Bibliographic references were searched in five electronic databases: PubMed, Web of Science, PsycINFO, Lilacs, and Scielo. The following descriptors were used in the Brazilian databases: "autismo" OR "transtornos globais do desenvolvimento" AND "regressão" OR "perda" AND "linguagem" while their equivalents were searched in international databases: "autism" OR "pervasive developmental disorder" AND "regression" OR "setback" AND "language". The terms "autism" and "regression" and "language" were cross-referenced, then "autism" and "loss" and "language", and so on until all possibilities were exhausted, which resulted in four combinations per language. We performed only one search per database using the Boolean OR because, depending on the database, it is less precise in comparison to searches using AND.

The initial inclusion of papers was based on their abstracts, while the application of exclusion criteria and data extraction was based on the full texts. Two independent judges (this study's authors) performed both procedures (selection and extraction). One expert with recognized clinical and academic experience in this field was consulted whenever the judges disagreed in regard to the studies to be selected or information collected in order to reach consensus. Inter-judge agreement in regard to the final selection of studies was 72.6%. Empirical studies addressing aspects related to the loss of skills among individuals with ASD were initially included. Based on the full texts, the following were excluded: (1) studies that involved only children with conditions other than ASD (e.g., Rett syndrome, Fragile X syndrome, Landau-Kleffner syndrome); (2) studies that did not address language regression; and (3) studies that did not characterize the loss of language skills. In regard to this last criterion, characterization of loss of language was operationally defined as presenting two or more of the following: age when loss started, duration of loss, association with external factors (e.g., fever, convulsions, stressful events), co-occurrence of loss of other skills (e.g., social interaction skills, ability to play or practice self-care). Characterization of the sample with language regression (e.g., mean age, scores of diagnostic instruments, intelligence quotient) was not considered in criterion 3. Also, note that studies, the full texts of which were not available in the database, were excluded after directly requesting them of the authors and/or periodicals and receiving no response. Figure 1 presents the flowchart concerning the selection of papers.

Information analysis

The selected studies were initially characterized according to objectives, participants, periodical and year of publication. Additionally, information concerning language regression was selected according to data available in each study, that is, definition and instrument used to investigate the phenomenon, mean age, and date of loss, mean duration of loss, association with health or family problems, and co-occurrence with the loss of other abilities. Finally, information concerning language development before loss took place is also presented, when available.

Results

Thirty out of the 315 studies initially selected met the inclusion and composed the corpus of this review. The complete references of the selected studies are highlighted with an asterisk (*) in the list of references. Table 1 presents the characterization of the studies, including authors, objective, number of participants and ages, year and periodical. The studies were published between 1985 and 2014; the number of participants ranged from one to 585; and the studies encompassed various objectives, most of which focused on regression. Additionally, 86.6% (n = 26) of the studies were published in North American or European journals, the remaining in Brazilian (n = 2) and Asian journals (n = 2).

A total of 86.6% (n=26) of the studies presented the operational definition used to investigate language regression and 43.3% (n = 13) used the ADI-R. Age at the beginning of language regression was available in 93.3% (n = 28) of the papers, ranging from 12 to 42 months. Information concerning the duration of loss could be analyzed only in 23.3% (n = 7) of the studies. Table 2 presents characterizations of language regression according to data available in each study, that is, the definition and instrument used to investigate the phenomenon, mean age when loss initiated, and mean duration of loss.

Additionally, 76.6% (n = 23) of the studies investigated the relationship between language regression and health problems, while 73.9% (n = 17) of these analyzed aspects that concerned electroencephalographic alterations or a history of convulsion. On the other hand, only 23.3% (n = 7) of the studies investigated the relationship with family problems or stressful events, and all depended on the parents' reports. Of the 22 studies that found language regression concomitantly with the loss of other abilities, 90.1% (n = 20) reported the loss of social interaction and/or playing abilities. Table 3 presents this information in detail.

Finally, information regarding language development, more specifically, the age when the first language milestones were achieved, was available in 43.3% (n = 13) of the papers. In regard to the age when the child's first words were spoken, (n = 13) it ranged from 11 to 41.2 months, while in 84.6% (n = 11) of these studies, children were aged between 11 and 14 months. Information regarding the age when children spoke their first phrases was available in six studies and involved the period between 18.8 and 59 months of age. Additionally, two studies investigated a number of words used by children before the loss of words, while one study reports an average of 5.2 words and another 6.33.

Discussion and Final Considerations

This systematic literature review focused on language regression in ASD, a phenomenon that is characteristic in a portion of children with the disorder; it appears less frequently in other disorders (APA, 2013; Lord et al., 2004; Pickles et al., 2009, Thurm et al., 2014). Together, information gathered in this review reveals that language regression in ASD has been a topic of growing interest among researchers in recent years. In this sense, note its operational definition has been refined, since criteria currently used involve temporal parameters, both in regard to the use of the ability lost and the loss itself, in order to investigate the stability of these components (i.e., behavior and loss).

Studies began to address this phenomenon at least three decades ago, but most of the reviewed studies were developed after the 2000s. Williams, Brignell, Prior, Bartak and Roberts (2015) assert that research on DR in ASD has helped develop an understanding of the disorder and even its heterogeneity, though authors stress that this field still lacks information regarding the mechanisms implicated in the loss and its prognosis, for instance, while findings are very diverse.

The variability of results may be explained by different factors, such a lack of consensus concerning the conceptual definition of DR (Lampreia, 2013), which results in the use of different measures to investigate the phenomenon and different sampling methods (Barger et al., 2013). In fact, the results of this study show considerable variability in the objectives, sample sizes, and to a less extent, in the instruments used to investigate language regression in ASD. In regard to this last aspect, the ADI-R was the measure most frequently used in the studies addressed here. The ADI-R is, however, a diagnostic interview that requires formal training, in addition to the fact it is privately owned. Therefore, in addition to not being an easily accessible instrument, it was not specifically developed to investigate regression, which probably motivated the development of other measures used in other studies such as the Regression Supplement Form (Goldberg et al., 2003) and Regression Validation Interview-Revised (Thurm et al., 2014). Additionally, as most of the studies addressed in this review are retrospective studies, some authors analyzed domestic videos to better understand the development of children before their linguistic loss, as well as the occurrence of the phenomenon itself (Bernabei et al., 2007; Ekinci et al., 2012). In fact, the use of domestic videos in the DR field, still seldom used, is promising because it also validates the reports of parents concerning such a loss. This validation is very relevant to the extent that, for some time, it was thought that DR was a parental "illusion", meaning that the affected children had never actually acquired the so-called lost skills. Hence, the analysis of domestic videos has attested to and clarified important aspects concerning the occurrence of this phenomenon.

This review also identified the operational definition of language regression used by the studies. Such a definition was used by most of the studies and tended to involve communicative loss of at least five words in a minimum period of three months and was clearly based on the way loss of abilities is investigated by the ADI-R (Lord et al., 1994). This is, therefore, an examination of loss of words, and not the language itself, since the latter involves broader and more complex components. The concept of word implied in such an investigation is not discussed in the studies; that is, understanding of the configuration of this linguistic element (e.g., its form and structure) may not be the same between researchers and parents, for instance. The way a word is used, on the other hand, tends to be addressed in the operational definition of language, since it involves loss of communicative use. In other words, it should not refer to the mere repetition of another person's speech, rather to symbolic elements used spontaneously in social interaction. Additionally, in order to examine the stability of loss, studies considered a period of time in which the phenomenon persisted, generally corresponding to three months. Together, different information shows that the operational definition of language regression involves different aspects that complement each other. Nonetheless, the issue related to a lack of consensus in the conceptual definition of this phenomenon remains. As highlighted by Lampreia (2013), it is imperative to consider the conception of regression in order to deepen scientific knowledge in the field and contribute to understanding ASD itself, considering that children affected by regression follow a developmental path within the spectrum (ASD) that contrasts with that followed by children who do not lose abilities, at the same time they share with each other characteristics that are inherent to the disorder.

In regard to the prevalence of language regression in ASD, data found in this study ranged from 9% to 51%, which may be explained by the different sampling methods used (Barger et al., 2013), while the prevalence of language regression found in a meta-analysis developed by Barger et al., 2013 was 24.9%. In regard to the age when loss took place, even though the interval was from 12 to 42 months, language regression after 30 months was reported only in two studies (Canitano & Zapella, 2006; Pickles et al., 2009). In fact, research shows that DR tends to occur up to 30 months of age in ASD (Barger et al., 2013), which contrasts with loss that occurs in Landau-Kleffner syndrome (LKS), also called acquired epileptic aphasia, in which loss of abilities tends to occur between 4 and 7 years of age (Valente & Valério, 2004). This information, however, is not sufficient to differentiate these two conditions because children tend to keep playing and maintain social interest while suffering from LKS, while the occurrence of repetitive movement is rare and there is also an association with abnormal electroencephalographic findings, such as idiopathic partial epilepsy (Valente & Valério, 2004). Unfortunately, most of the studies failed to provide information regarding the duration of loss. When such information was provided, it presented great variation. This may be explained by the difficulty in precisely establishing the age when abilities are acquired, since both acquisition and loss of abilities tend to be gradual (Malhi & Singhi, 2012).

The studies investigating the relationship between language regression and health problems used different analyses, from parents' reports up to statistical analysis of associations. In any case, most studies addressed aspects related to electroencephalographic changes, such as convulsions and epilepsy. This potential etiological factor was widely studied in the field of DR in ASD, but findings remain controversial (Williams et al., 2015). Additionally, the few studies addressing the relationship between language regression and stressful events were based on parents' reports, generally involving changes related to family routines (e.g., the birth of a sibling, moving house). The fact these children were later diagnosed with ASD, however, suggests that loss of abilities already constituted the developmental trajectory within the spectrum, and did not necessarily arise from stressors or changes in the children's families, situations considered potentially stressful events from an emotional point of view.

Most studies also report that language regression occurred concomitantly with the loss of other skills, especially the ability to socially interact and play. This result is in agreement with socio-interactionism and socio-cognitive notions of language acquisition (Tomasello, 1999/2003). Tomasello considers the acquisition of the first words to occur based on social-interaction experiences with caregivers during so-called dyadic (i.e., face-to-face interaction) and triadic situations (i.e., interaction mediated by objects) and constitutes the basis for the formation of symbolic capacity, expressed through speech and "make-believe" play. Therefore, from a conceptual point of view, it seems logical that a developmental rupture (or cessation) in any of these abilities affects correlates behaviors of this representational chain (symbolic). Nevertheless, this notion still leaves a question unanswered: how do we explain cases in which the loss of language does not occur concomitantly with the loss of other abilities? There are two possible explanations, both of a methodological nature. The first is related to the fact that most measures are based on parental reports. Thus, loss in other areas could be subtle to the point of not being perceived accurately, especially because the greatest concern of parents is focused on language alterations, more specifically, in speech delay (Zanon et al., 2014). The other potential explanation concerns the measure most frequently used to examine the occurrence of regression, ADI-R, since it investigates language regression in more detail, which does not occur in regard to the loss of other abilities, such as those related to social interaction and play.

Finally, information on the linguistic development of children before the loss was available in less than half the studies. In most papers, the production of the first words took place around 12 months of age, suggesting these children had achieved this linguistic milestone within the expected age range (Clark, 2012). In regard to the age when first phrases emerged, among the four studies that report this information, only one found that children with language regression achieved this milestone within the expected age range (Clark, 2012), which may be related to the fact that the loss of words generally occurs around a child's second birthday (Barger et al., 2013), a period when the first phrases emerge (Clark, 2012). In other words, such a loss could compromise acquiring this linguistic milestone. Considering the aspects previously presented, note that this review identified some gaps that can motivate other studies. Thus, future research addressing DR in ASD, more specifically addressing the loss of words, should verify the acquisition of abilities before such loss occurs to improve understanding regarding development prior to regression and the process of loss itself. Such aspects are especially relevant when one considers the role of language in establishing the severity of ASD and its prognosis (APA, 2013). Note that the investigation of these aspects can be facilitated by analyzing home videos filmed before the loss, a procedure that the research group of which this review's authors are a part has adopted. Additionally, it is imperative that researchers expand discussions regarding the conceptual definition of DR and also of abilities (e.g., what is the concept of the word that should be adopted). Prospective research monitoring the development of children with ASD, with and without the loss of abilities, can significantly contribute to a better understanding of how regression progresses and its potential consequences. Therefore, studies addressing the efficiency of speech-language interventions among children with a history of language regression are potentially useful, especially from a therapeutic perspective. Finally, it is important to more deeply investigate the characteristics and socio-communicative performance of children with language regression who recovered their abilities. Together, these data can contribute to better understanding how this phenomenon occurs in ASD.

This study presents some limitations that should be addressed. For instance, only studies that characterized language regression were analyzed, while other studies addressing DR in ASD were excluded. Additionally, theoretical and systematic reviews were not included since this study's objective was to investigate empirical data. Finally, the descriptors may not have covered all related publications due to the possible use of different nomenclatures to designate language regression, for instance. Nevertheless, this study gathered a considerable amount of data in the field of language regression in ASD, indicating that children with a history of word loss may compose a subgroup within the spectrum. These results provide tools to professionals involved with child development in regard to the importance of the phenomenon, both regarding diagnostic investigation and interventions. In regard to the last aspect, we highlight the need for children to be referred to treatment as soon as the loss of abilities is perceived in order to minimize negative consequences of this phenomenon in child socio-communication development.

References

American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed). Arlington, VA: American Psychiatric Publishing. DOI: 10.1176/appi.books.9780890425596        [ Links ]

Backes, B., Zanon, R. B. & Bosa, C. A. (2013). A relação entre regressão da linguagem e desenvolvimento sociocomunicativo de crianças com transtorno do espectro do autismo. CoDAS, 25(3),268-273. DOI: 10.1590/S2317-17822013000300013        [ Links ]

Baird, G. Robinson, R. O., Boyd, S. & Charman, T. (2006). Sleep electroencephalograms in young children with autism with and without regression. Developmental Medicine & Child Neurology, 48,604-608. DOI: 10.1017/S0012162206001265        [ Links ]

Baird, G., Charman, T., Pickles, A., Chandler, S., Loucas, T., Meldrum, D. Carcani-Rathwell, I., Serkana, D., & Simonoff, E. (2008). Regression, developmental trajectory and associated problems in disorders in the autism spectrum: the SNAP study. Journal of Autism and Developmental Disorders, 38,1827-1836. DOI: 10.1007/s10803-008-0571-9        [ Links ]

Barger, B., Campbell, J. M., & McDonough, J. (2013). Prevalence and onset of regression within autism spectrum disorders: a meta-analytic review. Journal of Autism and Developmental Disorders, 43(4),817-828. DOI: 10.1007/s10803-012-1621-x        [ Links ]

Bernabei, P., Cerquiglini, A., Cortesi, F. & D'Ardia, C. (2007). Regression versus no regression in the autistic disorder: developmental trajectories. Journal of Autism and Developmental Disorders, 37,580-588. DOI: 10.1007/s10803-006-0201-3        [ Links ]

Canitano, R. & Zapella, M. (2006). Autistic epileptiform regression. Functional Neurology, 21(2),97-101. Retrieved on 2017 July 31^st, from http://www.functionalneurology.com/common/php/portiere.php?ID=b303fd4a025b44d2aed7eed38fb7787f        [ Links ]

Castillo, H., Patterson, B., Hickey, F., Kinsman, A., Howard, J. M., Mitchell, T., & Molloy, C. A. (2008). Difference in age at regression in children with autism with and without Down syndrome. Journal of Developmental & Behavioral Pediactris, 29(2),89-93. DOI: 10.1097/DBP.0b013e318165c78d        [ Links ]

Chilosi, A. M., Brovedani, P., Ferrari, A. R., Ziegler, A., Guerrini, R., & Deonna, T. (2014). Language regression associated with autistic regression and electroencephalographic (EEG) abnormalities: a prospective study. Journal of Child Neurology, 29(6),855-859. DOI: 10.1177/0883073813482767        [ Links ]

Christopher, J. A., Sears, L. L., Williams, P. G., Oliver, J., & Hersh, J. (2004). Familial, medical and developmental patterns of children with autism and a history of language regression. Journal of Developmental and Physical Disabilities, 16(2),163-170. DOI: 1056-263X/04/0600-0163/0        [ Links ]

Clark, E. V. (2012). First language vocabulary acquisition. Oxford: Blackwell Publishing Ltd. DOI: 10.1002/9781405198431.wbeal0414        [ Links ]

Ekinci, O., Arman, A. R., Melek, I., Bez, Y., & Berkem, M. (2012). The phenomenology of autistic regression: subtypes and associated factors. European Child & Adolescent Psychiatry, 21(1),23-29. DOI: 10.1007/s00787-011-0228-7        [ Links ]

Goldberg, W. A., Osann, K., Filipek, P. A., Laulhere, T., Jarvis, K., Modahl, C., Flodman, P., & Spence, M. A. (2003). Language and other regression: assessment and timing. Journal of Autism and Developmental Disorders, 33(6),607-616. DOI: 10.1023/B:JADD.0000005998.47370.ef        [ Links ]

Hansen, R. L., Ozonoff, S., Krakowiak, P., Angkustsiri, K., Jones, C., Deprey, L. J., Le, D. N., Croen, L. A., & Hertz-Picciotto, I. (2008). Regression in autism: prevalence and associated factors in the CHARGE study. Ambulatory Pediatrics, 8(1),25-31. DOI: 10.1016/j.ambp.2007.08.006        [ Links ]

Jones, L. A. & Campbell, J. M. (2010). Clinical characteristics associated with language regression for children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 40,54-62. DOI: 10.1007/s10803-009-0823-3        [ Links ]

Kobayashi, R., & Murata, T. (1998). Setback phenomenon in autism and long-term prognosis. Acta Psychiatrica Scandinavica, 98,296-303. DOI: 10.1111/j.1600-0447.1998.tb10087.x        [ Links ]

Kumar, S., Karmakar, P. & Mohanan, A. (2014). Language regression in children with autism spectrum disorders. International Journal of Pediatric Otorhinolaryngology, 78,334-338. DOI: 10.1016/j.ijporl.2013.12.00        [ Links ]

Kurita, H. (1985). Infantile autism with speech loss before the age of thirty months. Journal of the American Academy of Child Psychiatry, 24(2),191-196. DOI: 10.1016/S0002-7138(09)60447-7        [ Links ]

Kurita, H., Kita, M. & Miyake, Y. (1992). A Comparative study of development and symptoms among disintegrative psychosis and infantile autism with and without speech loss. Journal of Autism and Developmental Disorders, 22(2),175-188. DOI: 10.1007/BF01058149        [ Links ]

Lampreia, C. (2013). A regressão do desenvolvimento no autismo: pesquisa e questões conceituais. Revista Educação Especial, 26(47),573-386. DOI: 10.5902/1984686X10071        [ Links ]

Lord, C., Rutter, M., & LeCouteur, A. (1994). Autism diagnostic interview-revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24(5),659-685. DOI: 10.1007/BF02172145        [ Links ]

Lord, C., Shulman, C. & DiLavore, P. (2004). Regression and word loss in autistic spectrum disorders. Journal of Child Psychology and Psychiatry, 45(5),936-955. DOI: 10.1111/j.1469-7610.2004.t01-1-00287.x        [ Links ]

Luyster, R., Richler, J., Risi, S., Hsu, W., Dawson, G., Bernier, R., Dunn, M., Hepburn, S., Hyman, S., McMahon, W., Goudie-Nice, J., MInshew, N., Rogers, S., Sigman, M., Spence, M. A., Goldberg, W., Tager-Flusberg, H., Volkmar, F., & Lord, C. (2005). Early regression in social communication in autism spectrum disorders: a CPEA study. Developmental Neuropsychology, 27(3),311-336. DOI: 10.1207/s15326942dn2703_2        [ Links ]

Malhi, P. & Singhi, P. (2012). Regression in children with autism spectrum disorders. Indian Journal of Pediatrics, 79(10),1333-1337. DOI: 10.1007/s12098-012-0683-2        [ Links ]

McVicar, K. A., Ballaban-Gil, K., Rapin, I., Moshé, S. L. & Shinnar, S. (2005). Epileptiform EEG abnormalities in children with language regression. Neurology, 64,129-131. DOI: 10.1212/01.wnl.0000167193.53817.0f        [ Links ]

Meilleur, A. A. S. & Fombonne, E. (2009). Regression of language and non-language skills in pervasive developmental disorders. Journal of Intellectual Disability Research, 53(2),115-124. DOI: 10.1111/j.1365-2788.2008.01134.x        [ Links ]

Nizamie, A., Sengupta, U., Mishra, B. R., Praharaj, S. K., & Nizamie, S. H. (2010). Role of early multimodal interventions in a case with autistic regression. Acta Neurologica Taiwanica, 19,51-56. Retrieved on 2017 July 31^st, from http://www.ant-tnsjournal.com/Mag_Files/19-1/dw201041513470_19-1%20c-3%209827.pdf        [ Links ]

Parr, J. R., Couteur, A. L., Baird, G., Rutter, M., Pickles, A., Fombonne, E., & Bailey, A. J. (2010). Early developmental regression in autism spectrum disorder: evidence from an international multiplex sample. Journal of Autism and Developmental Disorders, 41,332-340. DOI: 10.1007/s10803-010-1055-2        [ Links ]

Pickles, A., Simonoff, E., Conti-Ramsden, G., Falcaro, M., Simkin, Z., Charman, T., Chandler, S., Loucas, T., & Baird, G. (2009). Loss of language in early development of autism and specific language impairment. Journal of Child Psychology and Psychiatry, 50(7),843-852. DOI: 10.1111/j.1469-7610.2008.02032.x        [ Links ]

Shinnar, S., Rapin, I., Arnold, S., Tuchman, R., Shulman, L., Ballaban-Gil, K., Maw, M., Deuel, R. K., & Volkmar, F. R. (2001). Language regression in childhood. Pediatric Neurology, 24(3),185-191. DOI: 10.1016/S0887-8994(00)00266-6        [ Links ]

Shumway, S., Thurm, A. Swedo, S. E., Deprey, L. Barnett, L. A., Amaral, D. G., Rogers, S. J., & Ozonoff, S. (2011). Brief report: symptom onset patterns and functional outcomes in young children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 41,1727-1732. DOI: 10.1007/s10803-011-1203-3        [ Links ]

Tamanaha, A. C., Machado, G. M. G., Loebmann, C., & Perissinoto, J. (2014). Trajetória de aquisição e desenvolvimento de fala de crianças autistas com e sem história de regressão autística. CoDAS, 26(4),265-269. DOI: 10.1590/2317-1782/201420130021        [ Links ]

Thurm, A., Manwaring, S. S., Luckenbaugh, D. A., Lord, C., & Swedo, S. E. (2014). Patterns of early skill attainment and loss in young children with autism. Development and Psychopathology, 26(1),203-214. DOI: 10.1017/S0954579413000874        [ Links ]

Tomasello, M. (2003). Origens culturais da aquisição do conhecimento humano. (C. Berliner, Trad) São Paulo: Martins Fontes.         [ Links ]

Tuchman, R. F. & Rapin, I. (1997). Regression in pervasive developmental disorders: seizures and epileptiform electroencephalogram correlates. Pediatrics, 99(4),560-566. DOI: 10.1542/peds.99.4.560        [ Links ]

Valente, K. D. R. & Valério, R. M. (2004). Transtorno invasivo do desenvolvimento e epilepsia. Journal of Epilepsy and Clinical Neurophysiology, 10(4),41-46. Retrieved on 2017 July 31^st, from http://epilepsia.org.br/epi2002/J10-4p41-46.pdf        [ Links ]

Williams, K., Brignell, A., Prior, M., Bartak, L., & Roberts, J. (2015). Regression in autism spectrum disorders. Journal of Paedietrics and Child Health, 51,61-64. DOI: 10.1111/jpc.12805        [ Links ]

Wilson, S., Djukic, A., Shinnar, S., Dharmani, C., & Rapin, I. (2003). Clinical characteristics of language regression in children. Developmental Medicine & Child Neurology, 45,508-514. DOI: 10.1111/j.1469-8749.2003.tb00950.x        [ Links ]

Zanon, R. B., Backes, B., & Bosa, C. A. (2014). Identificação dos primeiros sintomas do autismo pelos pais. Psicologia: Teoria & Pesquisa, 30(1),25-33. DOI: 10.1590/S0102-37722014000100004        [ Links ]

Zwaigenbaum, L., Bryson, S., & Garon, N. (2013). Early identification of autism spectrum disorders. Behavioral Brain Research, 251,133-146. DOI: 10.1016/j.bbr.2013.04.004        [ Links ]

Mailling address:
Bárbara Backes
Instituto de Psicologia, UFRGS
Ramiro Barcelos Street, 2600
Porto Alegre (RS), Brazil, CEP: 90035-003
Telefone/Fax: +55 51 33085261
E-mail: barbara.edas@gmail.com

Submission: 15.3.2016
Acceptance: 20.6.2017