The assessment of maternal communicative style is an essential component of the assessment process, particularly given the existence of a relationship between the child’s linguistic development and the maternal communicative style. One of the primary objectives of an initial assessment is that of identifying appropriate therapy aims.

In Italy, a number of researchers have recently been focused on the mother’s role as the child’s support-figure within the development of early communicative and social-interactive skills. This maternal capacity for support may best be explored through analysis of the individual interactive style adopted by the mother during conversational exchanges with her child. It is generally accepted that an adult tutorial communication style that supports the child’s action and communication correlates positively with the development of early linguistic skills, whilst an unphased/devaluing style has been found to be less effective in this regard (Longobardi 1992, 1995; Hvastja Stefani 1997).

The assessment + intervention model used in this study is focused primarily on the communicative and linguistic interaction between the parent and child. The primary objective of the interactive language intervention model adopted in children presenting with expressive specific language impairment is to improve the adult’s level of linguistic and communicative input, in this case specifically the mother’s. More specifically, the mother’s responsive behaviours are increased with the aim of developing interactions focused on the same topic with more equally balanced and less complex conversational turns and a reduction in direct questioning. A communicative style that is focused primarily on the adult’s needs, with more directive communicative purposes (ordering, requesting, instructing) or an unphased/ devaluing interactive style that does not consider the child’s focus of attention and comprises many devaluations, correlates negatively with language acquisition, also because it offers a limited amount of relevant linguistic information. In general, one can say that a "child-centred" interactive style, with numerous expansions and extensions of the child’s utterances, exerts positive effects on linguistic development (Girolametto 1988; Girolametto et al.1996).

If, on one hand, the linguistic input facilitates language acquisition and the lack of input cannot be considered as the cause of the delay, we know that the impact of the child’s atypical progression on the adult’s communicative style itself requires consideration. It is therefore necessary to base the programme of intervention on the knowledge obtained through analysis of the assessment results, in particular on the information obtained through comparison of the maternal communicative style and the communicative and linguistic capacities of the child. It becomes essential to recognise the contextual factors that are fundamental for the child’s linguistic development. The parents learn to act in the so-called " Zone of Proximal Development" (or potential development zone, Vygotskij 1931) which is determined by the difference between what the child is able to do alone and what he or she could do with adequate support or a "tutor" (Bonifacio, Hvastja Stefani 1998).

The interactive approach on which our assessment and intervention system is based derives from the theoretical model that interprets the child’s development in context (Bronfenbrenner 1986; Hvastja Stefani 1995, 1998).

The aims of this assessment protocol vary and consequently the underlying targets differ in their characteristics and implications. It is important to distinguish between objectives which may be considered strictly evaluative, and those aimed at promoting and developing language development. The aims of the assessment goals are to provide a quantitative dimension with specific points of reference aimed at facilitating follow-up and comparison over time whilst simultaneously monitoring the effects of the intervention. In this case, the unit of analysis is represented by the interaction of the mother-child dyad. The second group of targets, however, aim to create a climate of trust between the speech therapist, the child and the family, with the aim of working together jointly to improve the child’s performance. The unit of analysis in this case is the system adopted by the speech therapist to effectively support the mother-child dyad. The qualitative analysis permits further accuracy of the intervention targets.

This study describes two subjects with Specific Expressive Language Impairment (SLI-E) who have been followed by the Speech Therapy Department at the "Burlo Garofolo" Children’s Hospital in Trieste, Italy, where an individual intervention programme was implemented for each child.

Goals of the study:

• To give prominence to changes in the maternal communicative style as a consequence of the focused stimulation implemented.

• To analyse the individual differences in language measures in both SLI-E subjects, with the aim of identifying the most apparent point of development whilst simultaneously considering the specific intervention plan adopted for each subject for the duration of the study.

Two Italian monolingual males of 38 and 41 months of age respectively with no neurological or hearing disorders were selected as subjects for the study. Both subjects presented with Expressive Specific Language Impairment (SLI-E ) as evidenced by an expressive vocabulary of less than 50 words and no two-word utterances. The children were active conversationalists (Fey 1986).

The first child’s family educational level was medium whilst the second one was low. Both mothers were homemakers. While both fathers were invited to participate in the intervention programme, only the mothers were included as the focus of this study.

Ten children (5 males and 5 females) with typical development (TD) were chosen with the "PVB" tool (Il Primo Vocabolario del Bambino: Caselli, Casadio 1995). The longitudinal data collected from the 12th to the 24th month of age was done in order to analyse the vocabulary size + composition obtained with this tool (Visini et al. 1998). To avoid discrepancies, the "PVB" tool was used giving the same instructions to the SLI-E and to the TD subjects’ mothers. The criteria concerning scoring of words were the same.

The intervention programme consisted of three sessions per month for six months for a total of 18 sessions. Each session was 45 minutes long. The intervention aimed at increasing the mother’s communicative and tutorial function, reducing as much as possible the unphased/devaluing behaviours whilst considering that the intervention’s effects vary as a function of maternal responsivity (Joder 1998). The intervention programme used in this study according to the interactive model of language intervention adopts the focused stimulation approach. The parents learn to use strategies such as: to follow the child’s initiative to establish a joint focus, to model language referring to the child’s focus (with labels, expansions, descriptions and comments), to recast the child’s verbal and non verbal intentions, to paraphrase, to stimulate the child’s participation in play routines, in role plays or in dramatic play with the aim of promoting turn-taking at the appropriate time.

The speech therapist selects the intervention strategies and she or he proposes them to the mother in order to:

1. facilitate and increase the child’s communicative abilities while interacting with her

2. emphasise the communicative intention expressed by the child and to promote its effectiveness,

and

3. understand the limitations of the intervention within the mother- child dyad considering in the first instance their ability to influence the selection of strategies, and secondly, the result of intervention throughout its implementation.

The choice of the strategies, not more than three, in each step of the intervention, is not based on a standard list, but it is connected with the priorities that arise from a comparison between the maternal communicative style and the linguistic profile of the child. In the first six months the specific goals concern the parents’ communicative behaviour, and are reformulated according to any changes that take place in the adult and in the child over time (Bonifacio, Hvastja Stefani 1998; Visini, Bonifacio 1998). If necessary, in the following months a mixed intervention is arranged. The three sessions per month are maintained, and they are focused on the mother-child dyad; there are also sessions of direct individual therapy on the child with specific goals.

Each child was examined for four separate assessments, and for a period of twelve months, with the aim of obtaining quantitative and qualitative measures of linguistic behaviour. The two assessment sessions were focused on the observation of the mother – child interaction and on the linguistic abilities of the child in two different contexts: " pretend play" with a set of 14 standard object-toys (a family, animals, vehicles and a house), and "reading of a picture book". Each context was audio-video taped for 15 minutes. Each assessment and treatment session was conducted by the same speech therapist.

• The coding manual of the "Maternal Communicative Style", elaborated by Hvastja Stefani (1997), modifying the manual of Longobardi (1992) and of Ninio et al. (1994).

The tool examines five functions: tutorial, didactic, conversational, controlling and unphased/ devaluing one. Each macro category contains a range of possible behaviours that include various verbal and non verbal features of the maternal behaviour. The behaviours are calculated quantitatively considering the frequency per minute. Such a criterion represents a useful measure of recognising the prevalent behaviour in the session, and facilitates a comparison between the various communicative contexts and moments observed beyond the duration of the assessment itself.

Assessment Goals:

• To recognise the prevailing behaviour within the two communicative functions, i.e. the tutorial and the unphased/devaluing.

Intervention Goals:

• To give prominence to the parent’s strengths and potential in order to facilitate her "scaffolding" abilities (Bruner 1977), that is her ability to support her child in the communicative process.

• To increase responsiveness and to decrease the tendency to be directive.

• To accentuate and to promote the child’s communicative effectiveness.

• To sensitise the parent about the identification and the interpretation of the child’s communicative functions.

• The Parent Report "Il Primo Vocabolario del Bambino" ("PVB": Caselli, Casadio 1995) in the version "Parole e Gesti" and "Parole e Frasi" that represent the Italian version of the MacArthur Communicative Development Inventories ("CDI: WG", "CDI: WS"; Fenson et al. 1993).

The "PVB" was administered every three to four months to the SLI-E subjects in order to calculate the quantitative and the qualitative measures of the lexical vocabulary produced.

Other authors ( Visini et al. 1998; Thal et al. 1999) have documented the validity of using this tool to measure vocabulary, even with language-delayed children who do not fall within the test’s chronological age range, but whose language levels are comparable to those measured by the tool. The same tool was used with the TD subjects every two/ three months in order to obtain longitudinal data relative to the increase in vocabulary size in the words produced and to the changes in the lexical composition.

Assessment Goals:

• To record the increase in words produced by the SLI-E subjects in comparison to that of the TD subjects in order to assign linguistic age to the SLI-E subjects.

• To identify the greatest point in lexical development, thus illustrating the outcome of the intervention programme.

• To analyse the vocabulary composition considering the expressive vocabulary recorded in the first recording and the point of greatest vocabulary growth in the SLI-E subjects.

• To compare the expressive vocabulary measures in the SLI-E and the TD subjects with respect to the amount of words produced, independently of age, with the aim of investigating the harmonised distribution of lexical categories.

Intervention Goals:

• To develop parent awareness of lexical development whilst monitoring the changes that occur as a result of the intervention.

• To utilise the context that most stimulates the child’s interest level and communication.

• To increase lexical development by means of selected words chosen according to the criteria of frequency (a datum included in the "PVB" manual) and production (words containing the speech sounds already produced by the child).

Following each session, the full repertoire of both the parent and the child’s verbalisations were transcribed on the computer together with annotations of both verbal and non-verbal behaviour and details of the communicative context using the CHAT coding of the CHILDES system (Child Language Data Exchange System: MacWhinney 1995).

Assessment Goals:

• To record the increase in the different words produced.

• To analyse the composition of the lexical inventory.

• To quantify the number of phonemes produced.

• To calculate the Mean Length of Utterance (MLU in words).

Intervention Goals:

• To sensitise the parent to recast the child’s productions that are very different to the adult’s model, but that are typical and recurring in the child. The context in which the child produces these lexical forms helps the parent to give a meaning to them.

• To use responsive strategies in order to support and to expand the child’s language.

Figure 1 illustrates the data obtained by analysing the maternal communicative style before and after twelve months from the start of the intervention (Bonifacio, Hvastja Stefani 1997; Hvastja Stefani et al. 2000). At the beginning of the intervention, it was observed that the communicative behaviours of the two mothers were very similar. Both mothers used very long utterances, unbalanced turn-taking, direct requests to elicit the child’s words and demonstrated interfering and non contingency behaviours that were not focused on the child’s verbal and non verbal behaviour.

In the " book reading" context, the marked presence of the tutorial behaviour is obvious in the first and in the last recording. Instead, in the "pretend play" context, the unphased/devaluing function is more remarkable, but it reduces notably in the last recording. This demonstrates the effective activation and the consequent change that took place in the mother’s behaviour, even if the change results more considerably in the "pretend play" context.

The tutorial function is more evident with this mother, with similar values to those in the " book reading " context. In the "pretend play" the unphased/devaluing behaviours are used less then in favour of the tutorial behaviour in the last recording.

It is apparent that the "pretend play" context favoured a greater use of tutorial behaviours in both mothers.

Figure 2 shows the comparison in expressive vocabulary gain between the SLI-E and the TD Subjects. During the initial assessment the SLI-E subjects demonstrated a vocabulary size similar to the TD subjects with an average chronological age of 15 months. The greatest point of expansion in the vocabulary of the SLI-E children is reported in the third session. In the last assessment the SLI-E values clearly correspond to the TD subjects of 24 months of age, even if they have a chronological age of 50- 51 months.

Figure 3 represents the comparison in expressive vocabulary composition (number of words produced), between the SLI-E and the TD subjects. This comparison, included in the tool, has been useful for understanding if the lexical inventory of the SLI-E subjects was comparable to the inventory of the TD sample (i.e. with a similar percentage of names, predicates, function words, etc.). The first comparison (number of words produced 21-50) shows the data recorded at the beginning of the intervention. The size of the 201-300 words produced refers to the session with the greatest gain i.e. after eight months of intervention. For this comparison the TD subjects that had the same vocabulary size independent of chronological age were selected from the sample group. In doing so, we obtained a sub-sample of nine subjects with an average age of 22 months and an average number of 244 words produced.

In the first recording, the comparison illustrates the most marked difference between the category of the common nouns, which are not so well represented, and the routines and function words, which are more prevalent. Bearing the latter in mind, during the first months the intervention was focused on the development of nouns. The mother’s role was to teach the child a list of words selected from the "PVB". The target words were selected on the basis of greatest frequency and balanced on semantic category of appearance ( e.g. food, familiar objects, part of the body etc.). The favourable outcome of this strategic application is evident in the recording that illustrates a vocabulary size of 201- 300 words. The distribution results are shown to resemble those of the TD subjects.

The initial recording illustrates less use of common nouns in this SLI-E subject. Instead, the categories of the predicates, function words and adverbs reveal higher distribution values with respect to the TD subjects. This information suggests more developed cognitive skills in the SLI-E subjects compared to younger children, and this observation was regarded as central to the lexical objectives of the intervention. These categories were consolidated and the noun category was developed. In the successive measures, an equal vocabulary size in both SLI-E and TD subjects was revealed.

The data drawn from this study were transcribed into the CHILDES data base. CLAN programmes were adopted for the analysis.

Each subject was significantly unintelligible, particularly during the initial assessments and consequently many verbalisations were necessarily classified as non-target words and eliminated from the linguistic analysis.

Figure 4 shows the increment in types and tokens in the SLI-E subjects during the different assessments. While analysing the data in the "PVB" tool, the greatest point in vocabulary gain was apparent during the third assessment. However, an analysis of both subjects’ spontaneous lexical sample revealed that the greatest increment of types was observed in the final assessment.

Figure 5 demonstrates a comparison of the spontaneous lexical sample (number of types) as assessed using the "PVB" in the initial and fourth assessments. It is interesting to note the presence of similar values in the first assessments. Subsequent assessments, however, revealed significant differences. The results obtained enable us to reflect that it is easier to obtain a spontaneous language sample that accurately reflects a child’s linguistic abilities if his/her vocabulary knowledge is limited. In addition, with an increase in vocabulary knowledge this measure is no longer an accurate reflection of the parent’s observations, as the clinical setting represents only a fraction of the child’s daily routines. In the first two assessments, for example, the children produced 70% of their potential verbal output, whilst in the subsequent assessments, only 35% was produced.

In the initial assessment the greatest difference is the common nouns and the adverbs, which are significantly fewer in the "PVB" data. In the fourth assessment, however, this disparity is reduced.

In the initial assessment, the common nouns are less prevalent in the "PVB" data, whilst the verbs and function words are well represented. In the fourth assessment, the common nouns are very prevalent in the "PVB" data. The function words and the adverbs are marked in the spontaneous observation.

The differences that appeared when comparing the two samples on the initial assessment suggest that both mothers were not able to give meaning to the nonstandard productions of the children and these were interpreted by the speech therapist. This result was considered in the planning of the intervention with steps included for listening and identifying the child’s verbalisation and adopting clarifying strategies.

The linguistic measures described in Table 1 represent a few indices of lexical abilities regarding increments in the Number of Different Words ("NDW") obtained in each assessment and calculated as a function of types and the Total Number of the Words produced ("TNW"). This result reveals a greater semantic ability in the first subject that is particularly apparent in the fourth assessment. In the second subject, growth observed in the same assessment period is not particularly consistent.

In order to calculate the Mean Length of Utterance (MLU in words), only the intelligible utterances produced by each subject have been selected. In the first assessment the initial stage was the single words. Only a year later in the final assessment do word combinations begin to emerge. It is important to point out that the MLU values were always higher in the "pretend play" context. This fact was considered during the intervention when more time was allocated to this context for the development of syntactical abilities, particularly when the children had reached an expressive vocabulary of at least 200 words as recorded by the "PVB" tool.

Table 2 illustrates the data collected in each assessment as a total number of different phones acquired and calculated as function of lexical size according to guidelines adapted from Stoel-Gammon (1985) and Dyson (1988).

Table 2 - Size and growth of phonetic inventory

 

In both subjects, the comparison of the types/tokens reveals a phonetic variability which was not so evident. The growth in speech sounds refers to the third assessment in the first subject; whilst in the second subject there is a more linear growth despite a more consistent token size in the last assessment than the previous one. This result may be considered in the light of the difficulties previously identified in the learning of different words.

One of the goals of assessment is to contribute to the comprehensive diagnostic definition of the disability. Equally important is the need to establish priorities in the intervention programme without viewing the two objectives as separate entities. The results of recording the maternal communicative style and the child’s communicative and linguistic abilities must become a complex unit of elements on which the therapy input must be based. Each element represents a measure to which equal balance should be assigned.

Should measuring the results of an intervention programme be considered an impossible task? It is the author’s opinion that it is possible, if we are able to clearly define what we intend to measure, and if we are able to recognise the factors that influence the measures.

The following aspects should be considered if we are to compare the various results of an intervention programme:

1. Defining the theoretical model of reference and the intervention procedures. The study presented is based on a structured interactive approach for the first six months and a "mixed" approach for the next six, although the interactive approach continues to be maintained even when integrated with sessions of direct therapy input with specific objectives.
2. Defining the child’s individual communicative-linguistic profile at the start of the intervention in order to ascertain if he/she is an active or passive conversationalist (Fey 1986) despite his limited linguistic capacities as this will define the objectives and strategies implemented in the intervention programme.
3. Carefully assessing which measures to consider and the appropriate time-intervals between recordings given that some measures are more sensitive than others or fluctuate at different times. In addition, some values may offer a more useful quantitative dimension when comparing assessment results over time, whilst others are more qualitative and therefore more pertinent when planning the intervention programme.

The following are observations from the study with reference to the costs and benefits:

• The recording of maternal communicative style:

a) This was found to be useful for identifying the parent’s strengths during the intervention. In particular, the increases in frequency values per recorded minute during the tutorial function as opposed to the unphased/devaluing style facilitated the measurement of changes brought about by the mother;

b) the most significant recording for the assessment results was that completed after 12 months;

c) the accurate and detailed analysis of the results proved significantly time consuming.

• The "PVB" Tool:

a) allowed for a longitudinal quantitative and qualitative comparison with data from the TD Subjects;

b) was useful for monitoring the changes even with a brief comparison over a two-month period and for evaluating significant changes in the lexical composition in the first six months of intervention;

c) proved invaluable for organising the chosen intervention by selecting the appropriate lexical targets for each subject;

d) proved useful for measuring what that the parent had learned to record;

e) was a time-effective instrument for recording and analysing the information obtained.

• Measures of the Spontaneous Linguistic Behaviours:

a) The growth in target words produced and MLU are measures in which consistent changes occurred over time after 10-12 months of the intervention programme. These proved to be contextually sensitive. In fact, in this study comparison between the " book reading " and the "pretend play" context in both subjects revealed the most marked verbal output in the latter.

b) These proved to be sensitive measures for understanding the level of difficulty in lexical learning. This difficulty was most apparent in the second subject during the course of the recording and the results obtained with the "PVB" instrument masked this. This fact was also confirmed by the increase in verbal output as measured by the amount of words per minute (Girolametto et al. 1996). In fact, in the first subject the most evident increment was in the third session (13 words) and developed even more in the last (17 words); in the second subject, however, the increment is linear, with 10 words produced in the final session.

c) The use of the phonetic inventory as a measure of acquired phonemes proved to be a relative measure and therefore not a reliable means of assessing the child’s true phonetic abilities because the assessment explored only a limited repertoire of the speech sounds of the language within a restricted range of contexts.

d) The detailed and accurate recording of the linguistic measures obtained necessitated the use of detailed, useful, yet time-efficient systems of coding and transcription to obtain a comparison over time.

 

 

This study was supported by a research grant from I.R.C.C.S. Istituto per l’Infanzia "Burlo Garofolo" of Trieste, Italy. I wish to thank Cristiana Visini for her help with the CHILDES data analysis. I also wish to thank Sandra Grosso for her help with translation and her insightful comments on an earlier version of the manuscript.

Bonifacio S., Hvastja Stefani L. (1998), L’interazione nel contesto dell’intervento: una risorsa da conoscere per orientarla allo sviluppo. In S. Bonifacio, L. Hvastja Stefani (a cura di): L’interazione comunicativa e linguistica nel bambino con ritardo di linguaggio. Contenuti e metodi nella valutazione, nella clinica e nel trattamento logopedico. Del Cerro, Tirrenia (Pisa).

Bonifacio S., Hvastja Stefani L. (1998), Il bambino con ritardo di linguaggio nel rapporto comunicativo con la madre : problematiche e soluzioni possibili nel trattamento logopedico. In S. Frasson, L. Lena, S. Menin (a cura di) : Diagnosi precoce e prevenzione dei disturbi del linguaggio e della comunicazione : procedure e metodi di trattamento nei disordini della comunicazione. Del Cerro, Tirrenia (Pisa).

Bonifacio S. (1998), La valutazione del bambino con ritardo di linguaggio espressivo nel contesto interattivo con la madre. In Bonifacio S., Hvastja Stefani L. (a cura di ) : L’interazione comunicativa e linguistica nel bambino con ritardo di linguaggio. Contenuti e metodi nella valutazione, nella clinica e nel trattamento logopedica. Del Cerro, Tirrenia (Pisa).

Bonifacio S., Cappelli G., Visini C., Lanzetta E., (1999), A longitudinal study of three children with expressive specific language impairment: procedures in clinical follow-up. Poster IASCL VIIIth International Congress for the study of Child Language. San Sebastian Donostia Spain.

Bronfenbrenner U. (1986), Ecologia dello sviluppo umano. Ed. Il Mulino, Bologna.

Bruner J. S. (1977), Early social interaction and language development. In H. R. Schaffer (a cura di) : Studies in mother-infant interaction. Londres : Academic Press.

Caselli M.C., Casadio P. (1995), Il Primo Vocabolario del Bambino. Guida all’uso del questionario MacArthur per la valutazione della comunicazione e del linguaggio nei primi anni di vita. Franco Angeli, Milano.

Fey M. (1986), Language intervention with young children. San Diego (Ca) : College Hill Press.

Girolametto E. (1988), Improving the social-conversational skills of developmnentally delayed children : an intervention study. Journal of Speech and Hearing Disorders, Vol. 53, 156-167.

Girolametto L., Steig Pearce P., Weitzma E. (1996), Interactive focused stimulation for toddlers with expressive vocabulary delays. Journal of Speech and Hearing Disorders, Vol.39, 1274-1283.

Hvastja Stefani L. (1995), Interazione sociale e sviluppo comunicativo-linguistico. In L. Lena (a cura di) : I disturbi del linguaggio nella prima infanzia: per un approccio integrato. Del Cerro, Tirrenia (Pisa).

Hvastja Stefani L. (1997), Lo stile comunicativo materno come indicatore di rischio. nello sviluppo comunicativo-linguistico. In Czerwinsky L. (a cura di) : Obiettivo bambino. Franco Angeli, Milano.

Hvastja Stefani L., Visini C., Bonifacio S. (2000), Maternal-parental communicative style as index of intervention effectiveness in SLI-E children. Poster 4^th European Congress C.P.L.O.L.- Paris 2-4 june 2000.

Joder P. J., Warren S.F. (1998), Maternal responsivity predicts the prelinguistic communication intervention that facilitates generalised intentional communication. Journal of Speech and Hearing Disorders, Vol. 41, 1207-1219.

Longobardi E. (1992), Funzione comunicativa del comportamento materno e sviluppo comunicativo-linguistico del bambino nel secondo anno di vita. Giornale Italiano di Psicologia, 3, 425-448.

Longobardi E. (1995), Funzioni comunicative materne nel secondo anno di vita del bambino : come varia il supporto materno in relazione allo sviluppo linguistico. Rassegna di Psicologia, XII, 67-83.

Ninio A., Goren H. (1993), PICA 100. Hebrew University, Jerusalem, Israel.

MacWhinney B. (1995), The CHILDES Project. Hillsdale, N.J. Lawrence Erlbaum Associates.

Rescorla L., Schwartz E. (1990), Outcome of toddlers with specific expressive language impairment. Applied Psycholinguistics, 11 : 393-407.

Stoel-Gammon C. (1985), Phonetic inventories, 15-24 months: A longitudinal study. Journal of Speech and Hearing Research. Vol. 28 : 505-512.

Thal D. J., O’Hanlon L., Clemmons M., Fralin L. (1999), Validity of a Parent report measure of vocabulary and syntax for preschool children with language impairment. Journal of Speech and Hearing Research. Vol. 42 : 482-496.

Visini C., Bonifacio S. (1998), La trascrizione e la codifica del linguaggio infantile patologico. In Bonifacio S., Hvastja Stefani L. (a cura di ) : L’interazione comunicativa e linguistica nel bambino con ritardo di linguaggio. Contenuti e metodi nella valutazione, nella clinica e nel trattamento logopedico. Del Cerro, Tirrenia (Pisa).

Visini C., Bonifacio S., Zocconi E. (1998), Aspetti lessicali in bambini con ritardo di linguaggio espressivo. Pediatria Medica Chirurgica. 20: 255-260.

Vygotski L. S. (1931), Trad. it.: Storia dello sviluppo delle funzioni psichiche superiori. Firenze : Giunti, 1974.