PHONOLOGICALLY BASED ASSESSMENT AND INTERVEN- TION IN SPASTIC CEREBRAL PALSY: A CASE ANALYSIS

The articulation errors of one adult subject demonstrating a spastic variety of congenital cerebral palsy were evaluated via a phonological process analysis. This analysis indicated that a stopping process (replacement of fricatives with homorganic stops) was the most detrimental to the subject's, intelligibility. Subsequent to this analysis a phonemic contrasting programme was initiated toward the goal of minimizing the influence of the stopping process. Results of spontaneous speech sample analyses indicated that this approach was successful in increasing the percentage of correctly produced fricative patterns. Success in this case suggests the applicability of a linguistically based intervention approach in structural/functional disturbances of speech articulation.

Assessment and intervention of articulatory disorders has traditionally focused on individual sound errors that detract from a client's intelligibility.This approach has been employed commonly for both functional and organic disorders of speech articulation.'Though this approach has been successful to some degree, it often fails to bring about clinically significant improvement in those individuals demonstrating multiple articulation errors.'  The past decade has yielded an emphasis on the identification of "patterns of errors" in those cases; where multiple sound errors are observed.A recent development related to this pattern approach of assessment is the identification of general phonological processes in the

The South African Journal of Communication Disorders, Vol. 28, 1981
Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012) defective speech pattern.A phonological process is considered to be a type of "mental operation" which simplifies a production by removing certain difficult properties from that production (Oiler 6 ; Stampe 8 ).In his Dissertation on Natural Phonology, David Stampe 8 points out that although a phonological substitution is a mental operation, it is clearly motivated by the physical character of speech -its neurophysiological, morphological, mechanical, temporal, and acoustic properties (p. 6).Given this description, it is surprising that relatively little attention has been focused on the identification and elimination of phonological processes in organically based articulation disorders.Ingram 5 reviews literature which identified error patterns in the speech of hard-ofhearing, cleft palate, and mentally retarded subjects.Paskowitz and Bond 7 have also used the phonological process approach to identify error patterns in the speech of hearing impaired children.Crary and Hunt 3 (in preparation) have demonstrated the clinical effectiveness of a pattern approach in the identification and subsequent elimination of phonological processes in the speech of one child with a repaired palatal cleft.In a recent didactic article, Crary and Fokes 2 have argued that this approach may be clinically efficient in the identification and elimination of defective articulatory patterns in the speech of neurologically impaired adults.
The present study reflects an attempt to: (1) identify phonological processes in the speech of a neurologically impaired adult, (2) quantify the influence of those processes on intelligibility, and (3) reduce the influence of the processes determined to be the most detrimental to perceived intelligibility.

SUBJECT:
The subject was a twenty-five-year-old female with congenital cerebral palsy of the spastic variety.She was enrolled in a public school special class and received one hour of speech therapy per week.Her therapist judged her to be "fairly intelligible" with multiple sound errors evident in her speech pattern.Therapy for the preceding eight months had focused on facilitating correct fricative production at the sound level.Although the subject could produce some fricatives in isolation, there was no reported or observed carryover to the syllable or word level.

ASSESSMENT: PROCESS IDENTIFICATION
In an attempt to identify phonological processes in the speech pattern of our subject two procedures were employed: (1) the Phonological Process Analysis (Weiner 9 ), and (2) analysis of a spontaneous speech sample using procedures outlined by Crary. 1 Table I presents the results of Weiner's analysis procedure.It can be seen that cluster reduction and stopping were the "strongest" (i.e., most frequent) processes, occurring in 100% and 88% of the test items

Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwy kings, Vol. 28, 1981
Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012) respectively.Cluster reduction was also observed in 75% of the applicable non-test items, while stopping was observed in 89% of the applicable non-test items.In addition, the stopping process was observed in 92% of those clusters involving fricatives.Other processes noted include deletion of final consonants and prevocalic voicing.Table II presents the results of the spontaneous speech analysis.To comprehend these data a brief discussion of the analysis procedures is required.The columm marked "RS" indicates the "Relative Strength" of each process.This is determined by dividing the number of actual occurrences of a process by the number of potential occurrences."RA" indicates "Relative Appearance" of those phonological patterns simplified by processes.This is determined by dividing the potential occurrence of each process by the total number of words in the sample."RIU" indicates the "Relative Influence on Unintelligibility".This is calculated by multiplying RS and RA.The higher the RIU value, the greater the detrimental influence on intelligibility.As shown in this table, the stopping process had the greatest influence on our subject's speech (highest RIU value) followed' by cluster reduction, deletion of final consonants, and prevocalic voicing.In considering the rank orderings of processes in each assessment procedure it can be seen that there is excellent agreement between the two procedures.They both suggest that our subject was simplifying her productive patterns by reducing cluster complexity, replacing fricatives with homorganic stops, deleting final consonants, and voicing prevocalic voiceless stops.Because of the high RIU values for stopping in the speech sample analysis and the frequent occurrence of this process in

The South African Journal of Communication Disorders, Vol. 28, 1981
Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012) both test and non-test items in the Weiner procedure, the stopping process was chosen as our intervention target.

THERAPY PROGRAMME
In an attempt to reduce the influence of the stopping process on our subject's speech, a phonemic contrasting programme was developed which focused on the feature contrast ± continuant.Following the general outline for phonemic contrasting prepared by Fokes a six-step programme was developed (two carryover steps were added to the basic programme to facilitate correct production beyond the word level) Briefly, the steps consisted of presenting words individually or in a minimal pair and emphasizing: (1) the perceptual qualities of the target features, (2) the differences in physical production patterns for the contrasted feature, and (3) imitation of the word pairs combined with perceptual identification of the feature contrast.The final training step involves spontaneous production attempts by the client.A copy of this programme is provided in Table V.

RESULTS OF THERAPY
Figure 1 presents the charted results of the therapy programme.The subject attended twenty half-hour sessions.Data presented in this figure indicate that the subject acquired improved continuant production from a pretest performance of 50% to a post-test performance of 80%.Note that step 5 appears to be the "critical step" in the programme for our subject.In this step the subject had to first correctly identify the feature contrast and then imitate a target word.It is apparent from the data on step 5 that our subject was more successful on the imitative task than the perception task (imitation scores noted by ( )).Even over repeated trials and sessions no improvement was noted in the ability to correctly identify the presented feature.This was confusing to us since our subject had performed the same perceptual task at the 80% level or above for three sessions prior to step 5.The final comment to be made on these data regards the abrupt termination of the programme.This was not a planned feature.Rather, it was necessitated by the end of our subject's academic semester.Note that the post-test did reach the preset criteria even though criteria was not demonstrated in step 6.

POST-THERAPY SPEECH ANALYSIS
At the end of the therapy programme a second speech sample was collected and analysed to assess changes in the spontaneous production of fricative patterns.The analysis performed on this sample was identical''to that performed on the pre-therapy sample.The results of this analysis are presented in Table III.Several changes

The South African Journal of Communication Disorders, Vol. 28, 1981
Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012) in the RS and RIU values can be seen.The relative strength of the stopping process decreased from 92% in the pre-therapy sample to 48% in this sample.In -the post-therapy sample decreases were also noted in RS for Deletion of Final Consonants and for Cluster Reduction.
The final table (Table IV) summarizes RIU changes from the pre-therapy sample to the post-therapy sample.Improvement is noted for the stopping process (RIU .34 to .21),deletion of final consonants (RIU .20 to .13), and cluster reduction (RIU .25 to .14).

CONCLUSION
The present case study attempted to demonstrate that patterns of articulatory errors could be identified in the speech of a neurologically impaired adult via a phonological process analysis.A second consideration of this study was to attempt to reduce the influence of observed patterns via a phonemic contrasting approach to therapy.Pretherapy speech samples revealed a strong and influential stopping process indicating a systematic replacement of fricatives with homorganic stops.The results of a therapy programme aimed at reduction of the stopping process indicated an increase in correct fricative production at the word level.Results of a post-therapy speech sample analysis indicated a reduction in the frequency and influence of the stopping process which reflects an increase in correct fricative production in spontaneous speech.
The results of this case study indicate the clinical effectiveness of using a pattern approach in the assessment and intervention of organically based articulatory disorders.Future clinical research in this area should address the issue of improved clinical efficiency and should attempt to define the potential use and limitations of this type of approach with clients presenting structural and/or functional impairments of the speech mechanism.
A CKNOWLEDGEMENT The authors would like to thank Susan Patra-McPherson for her assistance in the data analysis portion of this report.Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012)

5 .
Problem: Quadraplegic Spastic Cerebral Palsy Pretraining Activities la.Select words which emphasize ± continuancy (try to maintain one feature difference).lb.Be sure the client knows what the words mean.*2.Explain client's present pattern (e.g., stopping) in terms of: (a) air stream dynamics (b) oral cavity postures *3.Explain the "normal" pattern for the feature contrast to be taught.(See 2 ± continuancy).Note:* For steps 2 and 3, it is recommended you use 2 balloons -one tied without air being released (for -continuant) and one with air being released (for + continuant).Training Programme (critieria = 8/10 on each step) Note: Make 2 boxes to hold cards.Make a picture (balloons letting out air or tied) for each contrast box.la.Present words one at a time.b.Discuss key feature.c.Client places word in appropriate contrast box.2a.Present 2 words (contrast pair).b.Say one word.c.Client puts the word in the appropriate box.3a.Present one word.b.Present a physical activity depicting the key feature (e.g., blowing vs closed lips).c.Client imitates and puts card in correct box.4a.Word pair is presented.Say one word./ b.Client performs airstream feature and, y' c.Puts it in the correct box.Word pair is presented.Say one word.Client imitates and puts it in the appropriate box.ι 6a.Client becomes the teacher.Word pair is presented.b.Client produces one.c.Therapist picks up correct word and puts it in the box.****If step 6 fails, repeat steps 1-5 with client as teacher.7. Put each word in a 3-word imitative phrase.Word at the end of the phrase.8. Present word pair.Client says it in a sentence.Client: Female Age: 25 The South African Journal of Communication Disorders, Vol. 28, 1981

TABLE I :
Results of Phonological Process Analysis(Weiner 9 )

TABLE II :
Spontaneous Speech Analysis: Pre-Therapy

TABLE V :
Phonemic Contrasting Programme for ± Continuancy