elicited and spontaneous communicative functions and stability of conversational measures with...

Research Report

Elicited and spontaneous communicativefunctions and stability of conversationalmeasures with children who have pragmaticlanguage impairments

Catherine Adams and Julian Lloyd

University of Manchester, UK

(Received 5 March 2004; accepted 2 December 2004)

Abstract

Background: The preliminary phase of a project aimed at establishing appropriateoutcome measures for intervention with children who have pragmatic languageimpairments (PLI) is reported. Assessment methods for children with PLI areconsidered in the context of developing outcome measures for interventionstudies. Communicative function assessments in elicitation and conversationalcontexts are compared. The stability of measures derived from conversationalprofiling is also considered.Aims: To investigate the utility of an elicited communicative function assessmentin discriminating the pragmatic characteristics of children with PLI and tocompare this method to conversational profiling. An additional aim was toestimate the degree of variation on conversational indices derived frominteractions with children with PLI.Methods & Procedures: Fifteen children with PLI (mean age 9;5 years) and an age-matched control group were assessed on two occasions on a newcommunicative function elicitation task and on a conversation task. A checklistof communicative functions was employed in analysing the elicitation and theconversation tasks. An analysis of conversation was carried out to deriveconversational indices such as verbosity and meshing.Outcomes & Results: The elicitation task failed to discriminate between the PLIand control groups and showed a strong ceiling effect. Significant between-group differences were found for both communicative function in conversationmeasures and on conversational indices. The variation in conversation indices isrelatively small compared with the baseline measures.

International Journal of Language & Communication Disorders

ISSN 1368-2822 print/ISSN 1460-6984 online # 2005 Royal College of Speech & Language Therapistshttp://www.tandf.co.uk/journals

DOI: 10.1080/13682820400027768

Address correspondence to: Catherine Adams, Human Communication and Deafness, School ofPsychological Sciences, University of Manchester, Manchester M13 9PL, UK; e-mail: [email protected]

INT. J. LANG. COMM. DIS., JULY–SEPTEMBER 2005,

VOL. 40, NO. 3, 333–347

Conclusions: A more representative picture of communicative function abilityarises from unstructured tasks rather than from structured elicitation tasks forthe age group of children with PLI. The elicitation procedure outlined in thispaper might be better suited to a younger age group. Differences betweencommunicative function performance on elicitation and conversation situationssuggest a strong effect of context on the pragmatic performance of childrenwith PLI. The conversational indices reported show sufficient stability to beemployed in an outcome measure provided the minor variations indicated arefactored into estimates of change.

Keywords: pragmatics, assessment conversation, child language, intervention.

Introduction

In the population of children with developmental language disorders, a subgrouphas been identified whose difficulties lie principally with the use of language in socialcontexts. Bishop (2000) has termed this subgroup ‘Pragmatic Language Impairment’(PLI) but it has also been discussed under its previous label of ‘semantic–pragmaticlanguage disorder’. Children with PLI are typically verbose, have poor turn-takingskills, a difficulty staying on topic and significant difficulty in developingconversational skills. PLI is thought to be closely related to autism (Boucher1998) and these children share many of the social and cognitive features of autisticchildren, with the exception that their social withdrawal is less extreme.

Work on the characteristics and long-term outcomes of PLI has been reportedover the last decade or more (Leinonen and Letts 1997, Botting and Conti-Ramsden1999, Bishop 2000). Speech and language practitioners have received limited supportfrom the research base in implementing communication interventions for thesechildren in school and clinic settings. There is little existing high-quality evidencethat pragmatic ability can be enhanced by facilitation or by direct intervention andyet a significant proportion of services in educational speech and language therapy isdirected to the amelioration of pragmatic difficulties. Evidence regarding pragmaticintervention exists in the form of quasi-experimental studies and expert opinion fromlanguage-disordered populations (Conant et al. 1984, Bedrosian and Willis 1987,Camarata and Nelson 1992, Richardson and Klecan Aker 2000) and in children withautism (Gray 1998), in addition to some case studies of children with PLI (Letts andReid 1994, Willcox and Mogford-Bevan 1995, Adams 2001).

The intervention evidence base for PLI has suffered from the difficulty ofconstructing pragmatic tasks or assessments that will function as reliable outcomemeasures. These instruments need to be sensitive to the particular nature of theproblems of children with PLI and sensitive to change, while at the same timedemonstrating reliability and stability. Language pragmatics includes a number ofobservable and covert behaviours interpretable in specific contexts and withreference to a set of implicit rules of reference and inference. Controllingexperimental variables in such a complex system presents substantial challenges anddesigning assessment tasks which have content validity is a further challenge.

Existing standardized tests tap aspects of language pragmatics such as inference(the Test of Language Competence, Wiig and Secord 1989; the Assessment ofComprehension and Expression (6–11), Adams et al. 2001) and ambiguity resolution(Understanding Ambiguity; Rinaldi 1996). Less attention has been paid recently to

334 Catherine Adams and Julian Lloyd

the clinical assessment of aspects of pragmatics to do with expressive discourse,such as the speech act or communicative function. The speech act has been definedas ‘the peformative, or action accomplishing, aspects of language use and particularlythe (illocutionary) force associated with an utterance’ (Grundy 2000: 276). The Testof Pragmatic Language (Phelps-Teraski and Phelps-Gunn 1992) assesses the use ofsome speech acts in structured tasks. However, practitioners generally employ non-standardized schemes to supplement any testing such as Rinaldi’s (2001) Social Useof Language Programme. Speech acts observation schedules such as Fey’s (1986)Responsive/Assertive classification scheme and the Social Interactive CodingSystem of Rice et al. (1990) may be employed to profile speech acts.

Tests have limited potential as outcome measures for pragmatic skills due torestrictions on re-testing and lack of local norms. Observation schemes generallypresent analytical frameworks but are based on spontaneous outputs that render re-assessment problematic and which are time-consuming to analyse. An elicitationprocedure married to an observation scheme, however, might hold promise as anoutcome measure as well as constraining transcription and analysis time. In this project,a new elicitation procedure is outlined that prompts the child to use speech acts in astructured context. From now on, this will be referred to as the communicative functionelicitation task. The term ‘communicative function’ is preferred to that of ‘speech act’since it more closely describes the functional role of the utterance in context.

It is known that children with specific language impairments (SLI) demonstratea comparable range of communicative functions to the child with typical languagedevelopment (Gallagher and Craig 1984, Meline and Brackin 1987, Prutting andKirchner 1987, Craig 1995) limited only by their capacity to encode these functionsin grammatically well-formed utterances. Less is known about the ability of childrenwith PLI to produce communicative functions in situated contexts. Bishop andAdams (1989) and Bishop et al. (1994, 2000) showed that children with PLI couldmake use of exchange structure for the purposes of conversation but dealtprincipally with responsiveness and initiations in their analysis. Ziatas et al. (2003)demonstrated a lower rate of explanations and descriptions in populations ofchildren with autism compared with children with SLI or Asperger’s syndrome.Single case studies (e.g. Willcox and Mogford-Bevan 1995) have also shown thatchildren with PLI have mastery over some forms of communicative function. Thereare, however, no systematic group studies of either the communicative functions ofchildren with PLI or the potential of such an assessment to contribute towardpragmatic profiling. The context of communicative function sampling might alsohave an impact on the emerging pattern of behaviour, particularly for children withPLI who are known to have difficulty as the amount of contextual informationincreases. It is not known whether the child with PLI will produce a similar range ofcommunicative acts in conversation and in elicitation tasks.

Conversational profiling provides another option for monitoring change but, byits relatively unstructured nature, poses difficulties of natural variation incommunication performance from one time to another (Bishop et al. 2000). Touse such measures (e.g. Bishop and Adams 1989) to evaluate changes in pragmaticperformance as a result of an intervention, it is necessary to have factored in the‘allowable’ variation in the chosen conversation measures on different occasions. Ifthis is not done, there is a danger of showing significant changes that might havearisen simply as a result of some typical variation in the unstructured assessmentprocedures. Bishop et al. (1994) found reasonable stability in conversations with

Children with pragmatic language impairments 335

children with PLI between differing conditions of stimuli and interlocutor. Usingsimilar methods, it should therefore be possible to estimate the natural perturbationin these measures that need to be factored into outcome measures.

This study considered the potential of two assessment tools developed to assesspragmatic behaviour in children with PLI: a speech act elicitation task and aconversation task. Specific research questions were as follows:

N Are the pragmatic problems of children with PLI manifested on acommunicative function elicitation task?

N Do elicitation and conversation tasks have similar discriminatory potential indifferentiating the pragmatic skills children with PLI from their typicallydeveloping peers?

N What is the degree of variation (expressed as a proportion of conversationalindices) found in repeated measures of conversations in children with PLI?

Methods

Subjects

Fifteen children (10 boys and five girls, mean age 9;5 years, SD52;5, range 7;3–11;6)identified as having PLI by their speech and language therapists were included in thestudy. All subjects with PLI were recruited from three specialist language provisionsin the North West and West Midlands of England and Northern Ireland and wereattached to mainstream primary schools. All had received statements of specialeducational needs. The speech and language therapists in charge of the specialistunits had considerable experience in diagnosis of pragmatic language impairments.All children had marked impairments in pragmatics and this was the primary reasonthey were receiving speech and language therapy. Clinician opinion was preferred inrespect of diagnosis since there is no current accepted valid method of differentialdiagnosis of PLI and other types of language impairment. The instrument of usualchoice — the Children’s Communication Checklist (CCC) (Bishop 1998), which hasbeen used in previous research — is designed to indicate the presence of pragmaticlanguage impairment in individual children and provides a mean of confirming aclinical opinion. Recent research (Norbury et al. 2004), however, indicates that eventhe revised version of this checklist (CCC-2; Bishop 2003) cannot differentiate PLIfrom other language disorders. CCC-2 provides a pragmatic composite score aimedat partialling out PLI from SLI, but Norbury et al. were unable to validate thisdifferentiation. This recent evidence suggests that there is no advantage of usingCCC scores to discriminate children with PLI and SLI. Moreover in this study wewere not attempting to compare SLI and PLI children. No children in the PLI grouphad a diagnosis of frank autism in clinical records.

An age-matched control group (mean age59;4 years, SD51;5, range57;1–11;0)also took part in the study. Control subjects were recruited from the mainstreamprimary schools attached to the language units provision. Subjects were matchedwithin 3 months to PLI children on age and by sex. Control children had no historyof speech and language delay or received any special attention for communication orother learning disability. Using the British Picture Vocabulary Test (BPVS) (Dunnet al. 1997) as a screen all control children functioned at above the 20th centile, withthe mean centile being 53.2 (range 20–99).


Procedure

All children participated in the elicitation task. A subset of ten children from the PLIgroup (six boys and four girls) with a mean age of 9;10 years (SD59;10,range57;11–11;6), and ten children from the control group with a mean age of 9;8years (SD51;3, range57;3–11;0) also took part in the conversation study.Assessment on elicitation and conversation tasks took place in a quiet room withonly the child and tester present. The sessions were videoed and a protocol wasdeveloped to score the recordings for the elicitation task. Conversational coding wascompleted from the video using the procedure described below. To permit anassessment of variation in pragmatic performance, each child was assessed on asecond occasion. A 2-week interval between Times 1 (T1) and 2 (T2) wasconsidered the optimum interval, but due to school access problems, the periodbetween testing ranged from 2 to 16 weeks.

Communicative function elicitation task procedure

A scripted procedure for eliciting pragmatic behaviours was developed based on the‘communicative-demand’ strategy (Creaghead 1984) for eliciting and assessingpragmatic behaviours (Paul 1995). (This was originally known as the ‘Peanut ButterProtocol’.) This approach involves setting up situations that tempt the child toexhibit typical communicative functions using a series of tasks and props. Nineteencommunicative functions were targeted in the procedure. The communicativefunctions elicitors were contained within a prepared written script available only tothe tester. The aim of the script was to provide a repeatable but naturalistic andinteresting task for the age group of children chosen. Tasks and props used includeda map task to elicit clarification requests and responses to clarification requests(Lloyd 2003); toy telephones to elicit greetings, closings, questions and answers; amagic trick to elicit hypothesizing, explaining and predictions; fake biscuits to elicitchoices and comments; and locked boxes to elicit requests for an object. Thecommunicative functions elicited and examples from the script are set out in theappendix.

Each child in both groups took part in the elicitation script task on twooccasions. Three assessors completed the assessments between the groups with thesame child being seen twice by the same assessor. Assessors were trained in advancein the administration of the elicitation script and pre-test checks were made toensure accuracy of delivery.

Each communicative function was coded according to the following:

N Yes: the behaviour was elicited.

N No: the behaviour was not elicited.

N n/o, i.e. not observable: the behaviour could not be observed due tocontextual factors (e.g. the child became fidgety, or one of the prompts didnot work properly).

Conversation task

The subjects were assessed through a conversation task that was based on theAnalysis of Language Impaired Children’s Conversation (ALICC), developed by


Bishop, Adams and colleagues (Adams and Bishop 1989, Bishop and Adams 1989,Bishop et al. 2000, Adams et al. 2002). Conversations were gathered using the semi-structured method devised by Adams and Bishop (1989). This method uses black-and-white photographs as prompts to initiate particular topics:

N T1: A visit by the doctor; a birthday party; a car breaking down.

N T2: Building a bonfire; a family day trip; taking a pram from a rubbish tip.

The aim was to use the photograph to initiate conversation with the child but thento move on to discuss the child’s own similar experiences. The recordings weretranscribed; turns and utterances were identified and numbered using the guidelinespresented by Bishop et al. (2000). A particular focus of attention was adultsolicitations: utterances by the adult that solicit either information or acknowledgementfrom the child. Solicitations for information include wh-questions (‘Where did yougo on holiday?’), yes–no questions (‘Do you like chocolate?’), and clarificationrequests (C: I went with Paul; A: Who’s Paul?). In later analysis, the ability of thechild to respond to adult solicitations is used in constructing conversational indices.

Communicative function analysis of conversation samples

To provide a measure of the types of communicative functions used in conversation,every transcribed conversation was coded using utterance as the unit of analysis. Theacts chosen reflect the major categories used in previous studies of PLI children’sconversations such as Bishop and Adams (1989) and Bishop et al. (2000). Inconversations, the range of communicative functions is likely to be limited by theconstraints of the context. A reduced number of communicative functions weretherefore selected for analysis. These are as follows:

N Questions: correspond to Request for information and Asking questions inthe elicited communicative functions task.

N Answers: incorporates Explanation, Predicting, Hypothesizing/giving rea-sons, Making choices, Answering.

N Statements: incorporates Comment on object, Volunteering to communicate.

N Requests for clarification by children in the conversational data were alsorecorded. These were mostly non-specific requests (e.g. What?), whereasmost requests for clarification by children in the scripted procedure would bepotential requests for elaboration (What colour door?) in response to anambiguous instruction (Go to the church).

N Clarifications: corresponds to item 19 on the communicative function task:Clarifying.

N Denial/correction: these utterances are unsolicited, and would correspond toitem 8 on the communicative function task: Denial/correction.

N Recapitulations are utterances that restate or summarize information whichhas already been stated in the conversation. These are importantconversational devices but were not elicited from the children in thecommunicative function task.

N Conversational mechanics: do not correspond to any of the communicativefunctions and have no topical content. These are utterances that are used toestablish mutual engagement and turn management. These include requestsfor acknowledgements and acknowledgements, utterances used to reinforce


either one’s own or one’s partner’s contribution, affective expressions, andutterances that are used to retain a speaking turn.

Conversational indices

The same data were then subject to further analysis as follows. Using methods fromAdams et al. (2002) and Bishop et al. (2000), the following indices of conversationalbehaviour were employed:

N Discourse participation.

N Conversational dominance.

N Assertiveness.

N Verbosity.

N Responsiveness.

N Meshing.

Discourse participation is the ratio of child utterances to tester utterances. It can beused to evaluate each partner’s contribution to the conversational floor time. A ratioof 1.0 suggests that both partners have contributed equally to the conversation. Aratio less than 1.0 suggests the tester has dominated the conversation. A ratio greaterthan 1.0 suggests the child has dominated the conversation (Lloyd et al. 2001).

Conversational dominance is an index that can be used to estimate the degree towhich the child dominates the conversation by repeatedly requesting information orby providing unsolicited information. The following formula was derived:

Conversationaldominance~child first partszstatements½ �=adultzchild first partszstatements½ �

Assertiveness is an index that indicates the child’s tendency to initiateconversational exchanges. The following formula was derived:

Assertiveness~Child first parts=Total child utterances

An index of verbosity was used to identify a tendency to verbosity in the form ofextended unsolicited talk in a single turn. It was defined as the number of child turnsthat included four or more statements in sequence as a proportion of the total turnsfor that conversation:

Verbosity~number of child turns which contain four or more statements insequence=total number of turns in conversation

Responsiveness5proportion of child responses to tester utterances ending in asoliciting or neutral intonation. In this analysis, nods and shakes of the head weretreated as yes/no answers.

Meshing was defined as the quality of fit between adult solicitations and childresponses. Four categories of meshing have been defined (for a more detaileddescription of these categories, see Bishop et al. 1998):

N Adequate: the response is judged to be a good fit with the first soliciting part(A: Where are you going on Saturday?; C: To the zoo).

N Inadequate: the child produces a less than optimal response because of somelinguistic limitation, comprehension failure, or lack of general knowledge.


When using this code no allowance is made for age (e.g. A: Where’s Tenby?;C: Long way away).

N Pragmatically inappropriate: the response does not fit the social and/orcommunicative context of the soliciting part. This may be because the childhas failed to take into account previously given information, ignores an adultsolicitation, produces a tangential response, an over-literal response that doesnot appreciate the speaker’s intention, or uses ‘don’t know’ uncooperatively(e.g. A: Have you ever been to the doctor?; C: I had an apple a day; A Wheredid your dad take you last Saturday?; C: Can you please not ask that question).

N No response: the child is given an opportunity to respond but does not do so.

An index of problematic responses was derived using the following formula:

Response problems~child (inadequate responseszpragmatically inappropriateresponseszno responses)=total child responses

An index of pragmatically inappropriate responses was calculated as follows:

Pragmatic problems~pragmatically inappropriate responses=total child responses

Results

Communicative function elicitation task

The elicitation procedure did not distinguish between the PLI and control groupson the number of communicative functions elicited. Both groups were performingat or close to ceiling level for most of the items. Similarly, little variation was foundbetween T1 and T2 for either the PLI or the control group. Though the results didsuggest that knowledge gained in the task during T1 could affect the performance ofsome children in T2 (for example, in T2, the child already knows that a pen ishidden in the locked box and therefore does not have to ask about the location ofthe pen), similar results were found for both the PLI and control groups. Figure 1shows a pattern of responses for one item on the elicitation task (Requesting anObject) which is representative of all other items. The majority of responses arecorrect with little variation between the groups and little change over time and apronounced ceiling effect. Children with PLI tended to show a little more variabilityin their responses when the data is displayed graphically but this is not significant forany communicative function items.

Conversation task

In the following analyses, a 262 mixed factorial ANOVA design was employed.Conversational measures described in the last section were the dependent variables.The independent variables were time of testing (two levels: T1 and T2) and group(two levels: PLI and control). Time was manipulated within subjects, while groupwas manipulated between subjects.

Table 1 shows the total and mean number of turns and utterances for the PLItester and controls tester conversations. No significant main effect of time or group,and no significant interaction between time and group, was found for any of thesemeasures of turns and utterances. These findings suggest no significant differences


between the PLI tester and control tester conversations concerning the amount oftalk that took place between the participants. Both groups of children therefore wereoffered balanced opportunities to participate in the discourse.

Communicative functions in conversation

Table 2 shows the mean proportions of the communicative function types for thePLI and control groups at T1 and T2. No significant main effect of time, and nosignificant interaction between time and group, was found for any of thecommunicative function types. A significant main effect of group was found forquestions (F(1, 18)55.62, p50.029), clarification requests (F(1, 18)54.83, p50.041),clarifying (F(1, 18)56.29, p50.022) and recapitulation (F(1, 18)55.31, p50.033).The PLI group used higher proportions of questions, clarification requests,clarifying responses and recapitulations than the control children.

Table 1. Total and mean number of turns and utterances for the PLI tester and control testerconversations at Times 1 and 2

Time 1 Time 2

Total Mean SD Total Mean SD

PLI conversationsTurns 1498 149.80 (54.37) 1452 145.20 (51.24)Adult utterances 1371 137.10 (46.66) 1331 133.10 (36.99)Child utterances 1212 121.20 (40.16) 1172 117.20 (64.73)Discourse participationa – 0.87 (0.29) – 0.80 (0.27)

Control conversationsTurns 1589 158.90 (59.01) 1648 164.80 (81.54)Adult utterances 1406 140.60 (31.08) 1503 150.30 (46.69)Child utterances 1407 140.70 (81.50) 1468 146.80 (93.56)Discourse participationa – 0.92 (0.36) – 0.89 (0.33)

aUn-codeable utterances were excluded from the calculation of discourse participation ratios.

Figure 1. Distribution of elicitation codes for communicative function item 26, ‘Requesting an Object’,from Times 1 and 2.


Conversational indices

Table 3 shows the mean conversational behaviour indices for the PLI and controlgroups at T1 and T2. A significant main effect of group was found for responseproblems (F(1, 18)513.954, p50.002) and pragmatic problems (F(1, 18)56.29,p50.000), with the PLI group displaying more problematic responses than thecontrols in both categories.

Variation in conversational indices over time

Table 4 shows the differences between conversational indices derived for bothgroups at T1 and T2. There is no significant difference between-groups on any ofthe indices in the degree to which T1 and T2 performance differed from each other.

Table 2. Mean proportions of the communicative function types in conversation for the PLIand control groups at Times 1 and 2

Communicative acts

PLI Control

T1 T2 T1 T2

Mean SD Mean SD Mean SD Mean SD

Questionsa 2.08 (1.98) 3.40 (3.47) 0.78 (1.45) 1.18 (1.48)Answers 45.31 (8.67) 46.14 (21.49) 50.72 (19.52) 46.51 (17.76)Statements 18.99 (9.13) 19.10 (16.80) 24.36 (18.51) 27.40 (18.77)Clarification requestsa 0.42 (0.75) 0.60 (1.36) 0 (0) 0 (0)Clarifyinga 4.87 (2.67) 4.05 (6.37) 1.17 (1.37) 1.58 (1.26)Denial/correction 0.33 (0.56) 0.35 (0.77) 0 (0) 0 (0)Recapitulationa 1.24 (1.97) 1.65 (1.04) 0.12 (0.28) 0.64 (0.81)Evaluation 0.07 (0.22) 0.35 (0.77) 0.03 (0.98) 0.24 (0.49)Conversational mechanics 16.27 (10.08) 16.05 (8.59) 14.40 (6.52) 14.97 (5.92)Other 0.97 (0.84) 0.57 (0.66) 0.12 (0.38) 0.52 (1.42)Un-codeable 9.46 (4.22) 7.59 (6.58) 8.00 (3.48) 6.96 (4.40)

aPLI.control, p,0.05.

Table 3. Mean conversational behaviour indices scores for the PLI and control groups atTimes 1 and 2

Communicative indices

PLI Control

T1 T2 T1 T2

Mean SD Mean SD Mean SD Mean SD

Conversational dominance 0.28 0.15 0.25 0.20 0.30 0.24 0.31 0.24Assertiveness 0.03 0.02 0.04 0.03 0.01 0.02 0.01 0.01Verbosity 0.02 0.02 0.01 0.02 0.03 0.02 0.02 0.02Responsiveness 0.95 0.04 0.93 0.05 0.95 0.28 0.93 0.08Response problemsa 0.27 0.13 0.33 0.13 0.08 0.09 0.12 0.16Pragmatic problemsa 0.20 0.11 0.22 0.12 0.01 0.01 0.02 0.02

aPLI.control, p,0.01.


Discussion

Examining pragmatic behaviours through communicative function elicitationprocedures with a ‘yes/no’ approach either was not sufficient to discriminate thepragmatic problems of the PLI group or was not taxing enough to have potential toshow change over time. Although children engaged with the procedure satisfactorilyas an assessment technique, clearly the production of communicative functions isnot problematic by this age for children with PLI. This is not an entirely unexpectedresult. Practitioners have noted that the addition of structure into assessment taskfavours the support of pragmatics for these children (Leinonen et al. 2000). Bishopand Adams (1991) found similar results in their appraisal of referentialcommunication in these children in which a formal barrier task failed to elicit theknown problematic reference behaviours of children with PLI. Similar ceiling effectshave been found on tasks of inference where the task, although challenging, simplyfails to tap into the complex nature of the cognitive and linguistic processes involved(Botting and Adams 2005), whereas when the task is made more abstract, distinctgroup differences emerge (Adams in preparation).

The findings illustrate the difficulty of selecting appropriate assessmentprocedures for clinical pragmatics. Tasks employed in the assessment of pragmaticlanguage performance for children with PLI, including checklists and elicitationtasks, can be useful in identifying aspects of behaviour in individuals who might betargeted in intervention. Tasks that attempt to discriminate children with PLI fromother groups, however, must be employed with reference to the child’s age andlinguistic ability and must take the availability of supportive context into account. Inthis case, the elicitation task is well within the PLI children’s capacity. It may be thata more taxing elicitation procedure might be developed that would pick updifficulties with sophisticated communicative function production in more realisticcontexts. It would also be interesting to investigate the natural history ofcommunicative function development in children with PLI in a longitudinal study.In the meantime, the present assessment procedure might have more clinicalpotential with younger children who are still in the process of developingcommunicative functions.

The conversation analyses used in this study did identify differences in pragmaticbehaviour between groups. In particular, the framework called meshing, whichexamines the goodness-of-fit between adult solicitations and child responses, helpedto distinguish between the PLI and control groups, and gave some indication aboutwhere PLI children’s problems with conversational interaction might lie. Thisreplicates the findings of Bishop and Adams (1989) and Bishop et al. (2000).

Table 4. Differences between the mean conversational behaviour indices scores for the PLIand control groups for Times 1 and 2

PLI Control

Conversational dominance 20.03 0.01Assertiveness 0.01 0Verbosity 20.01 20.01Responsiveness 20.02 20.02Response problems 0.06 0.04Pragmatic problems 0.02 0.01


Although the PLI and control groups both generally displayed fairly highdegrees of responsiveness to adult solicitations, the PLI group was more likely togive pragmatically inappropriate, inadequate and no responses than were thecontrols.

An examination of communicative act types observed in conversation revealeddifferences between the PLI and control groups. Compared with the controls, thePLI group used higher proportions of questions, clarification requests, clarificationsand recapitulations. This pattern of results might be predicted based on the findingsof previous work on PLI children (Adams and Bishop 1989). The contrast betweenthe ceiling effects in the elicitation task and the discriminative power of theconversation task suggests there is a strong effect of context for processingpragmatic language for the PLI group. Capturing the additional load involved inmoving from pragmatics as a formal or abstract level of linguistics and theconversation task that taps into ‘situated’ pragmatics (Duchan 2000) is the key toaccessing communicative functions and conversational behaviours which will formthe basis of intervention.

One of the aims of this study was to consider which assessments have thepotential to function as outcome measures. The measures will be dependent, ofcourse, on the age and nature of the children receiving the intervention. For theage group of children under consideration here, communicative function elicitationtasks are an inadequate means of measuring change since they cannot tap into thechild’s difficulties with the complexities of verbal interaction. Conversationalmeasures — both communicative functions within conversations and conversationalindices — have more discriminatory power and therefore more potential to showchange.

The third research question addressed was concerned with the estimationof variation in pragmatic performance on these two assessment procedures. Thereis obviously no need to consider the elicited communicative function task sincethis lacked variation in the ceiling T1 scores. There was a small variation inpragmatic performance between T1 and T2 on conversational indices. An estimateof the (admittedly small) degree of variation in performance on the conversationcoding is now available to factor in to the outcome measures in the interventionstage of the study. Change in pragmatic performance which claims to be causallyrelated to the presence of the intervention would need to be significantly greaterthan this.

Conversation coding therefore holds promise for outcome measurement, butit is time consuming and requires more work to ensure its clinical applicability.The conversation coding system employed here requires further refinement toproduce a time-limited procedure that can be used in intervention studies (Adamset al. submitted). The resulting data might help us to determine whether theconversational, linguistic and/or social behaviour of children with PLI can change asa result of intervention.

There is a danger in any outcome study that even well-designed and evaluatedmeasures will not tell us what we can see happening before our eyes. To ward offsuch failures with this population, it will be necessary to develop outcome measuresthat are detailed enough to show changes in subtle complex and high-level pragmaticabilities (such as time-limited conversation analysis), but which are sufficiently broadsuch to detect generalization of gains expected to pervade family life and schoolperformance. In addition to the measures of communicative function outlined in


this paper, assessment protocols must extend to other aspects of language behaviour(narrative and inferential comprehension), to social attribution and participation, tothe child’s ability to interact verbally at home and in school, and the impact ofintervention on educational performance.

Acknowledgements

Thanks to the assessors Esther Gilman and Yvonne McGoldrick, and to the NorthWest, West Midlands and Northern Ireland language units and therapists. Thanksalso to Dorothy Bishop for continued support and the use of ALICC. The projectwas supported by an ESRC Realising Our Potential Award to the first author. Anearlier version of the paper was presented at the joint RCLST/CPLOL Conference,Edinburgh, UK, 2003.

References

ADAMS, C., 2001, Clinical diagnostic studies of children with semantic–pragmatic language disorder.International Journal of Language and Communication Disorders, 36, 289–306.

ADAMS, C., Inferential comprehension in children with pragmatic language impairments (in preparation).ADAMS, C. and BISHOP, D. V. M., 1989, Conversational characteristics of children with semantic–

pragmatic language disorder. I: Exchange structure, turn taking, repairs and cohesion. British

Journal of Disorders of Communication, 24, 211–239.ADAMS, C., COOKE, R., CRUTCHLEY, A., HESKETH, A. and REEVES, D., 2001, Assessment of Comprehension and

Expression 6–11 (ACE 6–11) (Windsor: NFER-Nelson).ADAMS, C., GREEN, J., GILCHRIST, A. and COX, A., 2002, Conversational behaviour of children with

Asperger syndrome and conduct disorder. Journal of Child Psychology and Psychiatry, 43, 679–690.ADAMS, C., LLOYD, J., ALDRED, C. and BAXENDALE, J. (Submitted) Exploring the effects of communication

intervention for developmental pragmatic language impairments: a signal-generation study.BEDROSIAN, J. L. and WILLIS, T. L., 1987, Effects of treatment on the topic performance of a school age

child. Language, Speech and Hearing Services in Schools, 18, 158–167.BISHOP, D. V. M., 1998, Development of the Children’s Communication Checklist (CCC): a method of

assessing qualitative aspects of communication impairment in children. Journal of Child Psychology

and Psychiatry, 39, 879–891.BISHOP, D. V. M., 2000, Pragmatic Language Impairment: a correlate of SLI, a distinct subgroup, or part

of the autistic continuum? In D. V. M. Bishop and L. Leonard (eds), Speech and Language

Impairments in Children: Causes, Characteristics, Intervention and Outcome (Hove: Psychology Press),pp. 99–114.

BISHOP, D. V. M., 2003, The Children’s Communication Checklist — 2 (London: Psychological Corporation).BISHOP, D. V. M. and ADAMS, C., 1989, Conversational characteristics of children with semantic–

pragmatic disorder. II: What features lead to a judgement of inappropriacy? British Journal of

Disorders of Communication, 24, 241–263.BISHOP, D. V. M. and ADAMS, C., 1991, What do referential communication tasks measure? Some

evidence from language-impaired children. Applied Psycholinguistics, 12, 199–216.BISHOP, D. V. M., CHAN, J., ADAMS, C., HARTLEY, J. and WEIR, F., 2000, Conversational responsiveness in

specific language impairment: Evidence of disproportionate pragmatic difficulties in a subset ofchildren. Development and Psychopathology, 12, 177–199.

BISHOP, D. V. M., HARTLEY, J. and WEIR, F., 1994, Why and when do some language-impaired childrenseem talkative? A study of initiation in conversations of children with semantic–pragmaticdisorder. Journal of Autism and Developmental Disorder, 24, 177–197.

BOTTING, N. and CONTI-RAMSDEN, G., 1999, Pragmatic language impairment without autism. Autism, 3,371–396.

BOTTING, N. and ADAMS, C., 2005, Semantic and inferencing abilities in children with communicationdisorders. International Journal of Language and Communication Disorders, 40, 49–66.


BOUCHER, J., 1998, SPD as a distinct diagnostic entity: logical considerations and directions for futureresearch. International Journal of Language and Communication Disorders, 33, 71–108.

CAMARATA, S. M. and NELSON, K. E., 1992, Treatment efficiency as a function of target selection in theremediation of child language disorders. Clinical Linguistics and Phonetics, 6, 167–178.

CONANT, S., BUDOFF, M., HECHT, B. and MORSE, R., 1984, Language intervention: a pragmatic approach.Journal of Autism and Developmental Disorders, 14, 301–317.

CRAIG, H., 1995, Pragmatic impairments. In P. Fletcher and B. MacWhinney (eds), The Handbook of ChildLanguage (Cambridge, MA: Blackwell), pp. 623–640.

CREAGHEAD, N., 1984, Strategies for evaluating and targeting pragmatic behaviours in young children.Seminars in Speech and Language, 5, 241–252.

DUCHAN, J., 2000, Preface: Assessing children’s language: present, past and future. Topics in LanguageDisorders, 21, 189–192.

DUNN, L. M., DUNN, L. M. and WHETTON, C. W., 1997, The British Picture Vocabulary Scale, 2nd edn(Windsor: NFER-Nelson).

FEY, M. E., 1986, Language Intervention with Young Children (Boston: College Hill).GALLAGHER, T. and CRAIG, H., 1984, Pragmatic assessment: analysis of highly frequent repeated

utterances. Journal of Speech and Hearing Disorders, 49, 368–377.GRAY C, A., 1998, Social stories and comic strip conversations with students with Asperger Syndrome

and High-Functioning Autism. In E. Schopler and G. Mesibov (eds), Asperger Syndrome or High-Functioning Autism? (New York: Plenum), pp. 167–198.

GRUNDY, P., 2000, Doing Pragmatics (London: Arnold).LEINONEN, E. and LETTS, C., 1997, Why pragmatic impairment? A case study in the comprehension of

inferential meaning. European Journal of Disorders of Communication, 32, 35–52.LEINONEN, E., LETTS, C. and SMITH, R., 2000, Children’s Pragmatic Communication Difficulties (London:

Whurr).LETTS, C. A. and REID, J., 1994, Using conversational data in the treatment of pragmatic disorder in

children. Child Language Teaching and Therapy, 10, 1–22.LLOYD, J., 2003, Spoken communication between deaf and hearing pupils. In C. Gallaway and A. Young

(eds), Deafness and Education in the UK: Research Perspectives (London: Whurr), pp. 53–74.LLOYD, J., LIEVEN, E. and ARNOLD, P., 2001, Oral conversations between hearing-impaired children and

their normally hearing peers and teachers. First Language, 21, 83–107.MELINE, T. and BRACKIN, S., 1987, Language impaired children’s awareness of inadequate messages.

Journal of Speech and Hearing Disorders, 52, 263–270.NORBURY, C. F., NASH, M., BAIRD, G. and BISHOP, D. V. M., 2004, Using a parental checklist to identify

diagnostic groups in children with communication impairment: a validation of the Children’sCommunication Checklist — 2. International Journal of Language and Communication Disorders, 39,345–364.

PAUL, R., 1995, Language Disorders from Infancy Through Adolescence: Assessment and Intervention (St Louis:Mosby).

PHELPS-TERASKI, D. and PHELPS-GUNN, T., 1992, Test of Pragmatic Language (Hove: PsychologicalCorporation).

PRUTTING, C. A. and KIRCHNER, D. M., 1987, A clinical appraisal of the pragmatic aspects of language.Journal of Speech and Hearing Disorders, 52, 105–109.

RICE, M., SELL, M. and HADLEY, P., 1990, The Social Interactive Coding System: An on-line, clinicallyrelevant descriptive tool. Language, Speech and Hearing Services in Schools, 21, 2–14.

RICHARDSON, K. and KLECAN AKER, J., 2000, Teaching pragmatics to language-learning disabled children: atreatment outcome study. Child Language Teaching and Therapy, 16, 1–23.

RINALDI, W., 1996, Understanding Ambiguity: An Assessment of Pragmatic Meaning Comprehension (Windsor:NFER-Nelson).

RINALDI, W., 2001, Social Use of Language Programme (SULP) — Revised (Windsor: NFER-Nelson).WIIG, E. and SECORD, W., 1989, Test of Language Competence — Expanded Edition (Hove: Psychological

Corporation).WILLCOX, A. H. E. and MOGFORD-BEVAN, K., 1995, Conversational disability: assessment and remediation.

In M. Perkins and S. Howard (eds), Case Studies in Clinical Linguistics (London: Whurr),pp. 146–180.

ZIATAS, K., DURKIN, K. and PRATT, C., 2003, Differences in assertive speech acts produced by childrenwith autism, Asperger Syndrome, specific language impairment and normal development.Development and Psychopathology, 15, 73–94.


Appendix: Pragmatic behaviours assessed via the elicitationprocedure

Communicative functions Conversational devices

1. Greeting 12. Answering2. Request for an object 13. Volunteering to communicate3. Request for information 14. Attending to the speaker4. Comment on an object 15. Taking turns5. Demonstration/explanation 16. Maintaining a topic6. Predicting 17. Asking questions7. Hypothesizing/giving a reason 18. Requesting clarification8. Denial/correction 19. Clarifying9. Making choices

10. Giving reasons11. Closing

Extract from an elicitation script.Communicative function: explaining.Context from the elicitation script: the tester takes out a gift box. The tester says, ‘I know what we cando next. My friend has gone on holiday and he’s asked me to give this to his mum. It’s her birthday.Should I take a peek inside to see what he’s bought her?’ They open the box together: it contains anunfamiliar object. The tester says, ‘I’ve never seen one of those before. How does it work?’


elicited and spontaneous communicative functions and stability of conversational measures with...

Documents