UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phonological development in twins after non-linear phonological and morphosyntactic intervention Feehan, Angela M 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_spring_feehan_angela.pdf [ 7MB ]
Metadata
JSON: 24-1.0071464.json
JSON-LD: 24-1.0071464-ld.json
RDF/XML (Pretty): 24-1.0071464-rdf.xml
RDF/JSON: 24-1.0071464-rdf.json
Turtle: 24-1.0071464-turtle.txt
N-Triples: 24-1.0071464-rdf-ntriples.txt
Original Record: 24-1.0071464-source.json
Full Text
24-1.0071464-fulltext.txt
Citation
24-1.0071464.ris

Full Text

PHONOLOGICAL DEVELOPMENT IN TWINS AFTER NON-LINEAR PHONOLOGICAL AND MORPHOSYNTACTIC INTERVENTION  by Angela M. Feehan B.A., Simon Fraser University, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  The Faculty of Graduate Studies  (Audiology and Speech Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) November 2010 © Angela M. Feehan, 2010  Abstract This study examined the phonological development of a set of 6-year-old male dizygotic twins with protracted phonological and morphosyntactic development. Each twin was given 16 weeks of therapy, 8 weeks of morphosyntactic treatment and 8 weeks of phonology treatment. These 8week treatment blocks were ordered oppositely so that one twin received morphosyntactic treatment first while the other received phonology treatment first. The study investigated phonological changes after morphosyntax treatment (cross-domain generalization), after phonological treatment (direct treatment effect), and after the full 16 weeks (overall effects and treatment order effect). Specific changes within the phonological systems of the twins were interpreted from a constraint-based non-linear phonology point of view. Twin 1, who received morphosyntax treatment first, did not show an increase in phonological matches to adult targets after the morphosyntax treatment block. After the phonology treatment block, an increase in percentage match for treated word structures and phonological features was seen. Twin 2, who received phonology treatment first, showed an increase in percentage match for treated structures and features after phonology treatment. After morphosyntax treatment, a continued increase in phonological matches to adult targets was seen. These results imply that it may be important to target phonology directly with children who have both protracted phonological and protracted morphosyntactic development. Results also indicate that both morphosyntax treatment first and phonology treatment first strategies can facilitate phonological development and that phonology treatment first may allow for greater generalization in the phonological domain during morphosyntax treatment.  ii  Preface This study was part of a collaboration. The planning and assessment for this project was completed by the two co-supervisors of the research, Dr. May Bernhardt and Dr. Paola Colozzo, and two Master of Science students in speech-language pathology, Angela Feehan (author), and Charmaine Sadler. Angela Feehan conducted the treatment for speech development, under the supervision of Dr. May Bernhardt, Dr. Paola Colozzo, and Lisa Avery. The data analysis for this thesis was carried out by Angela Feehan under the supervision of Dr. May Bernhardt. This research was approved by the University of British Columbia Behavioural Research Ethics Board, certificate number H09-00460.  iii  Table of Contents Abstract............................................................................................................................................ ii Preface ............................................................................................................................................ iii Table of Contents ............................................................................................................................iv List of Tables................................................................................................................................ viii List of Figures ..................................................................................................................................x List of Abbreviations.......................................................................................................................xi Acknowledgements ....................................................................................................................... xii Dedication..................................................................................................................................... xiii Chapter 1 - Introduction...................................................................................................................1 Literature Review.........................................................................................................................1 Phonological Theory and Goal Selection for Intervention ......................................................2 Phonological Theories.........................................................................................................3 The Constraints of Optimality Theory (OT).......................................................................9 Non-linearity .....................................................................................................................14 Constraint-Based Non-linear Phonology ..........................................................................15 Typical Phonological Targets for Intervention.................................................................16 Interactive Psycholinguistic and Dynamic Systems Theories..........................................18 Child Capability and Focus...............................................................................................20 Summary of Key Concepts ...............................................................................................22 Intervention for PPD ..............................................................................................................22 Treatment Methods ...........................................................................................................22 Goal Attack Strategies.......................................................................................................26  iv  Morphosyntax-Only Intervention......................................................................................28 Summary of Treatment Approaches .................................................................................29 Phonological Development and Intervention with Twins .....................................................30 Summary of the Literature Review: Key Points....................................................................35 Research Questions and Hypotheses .........................................................................................36 Chapter 2 - Methods.......................................................................................................................40 Participants.................................................................................................................................40 Overview of Study Design.........................................................................................................42 Assessments ...............................................................................................................................44 Assessment 1: Pre-Baseline (four months prior to treatment)...............................................44 Assessment 2: Baseline (prior to treatment block 1) .............................................................45 Assessment 3: Post-Block 1 (between treatment block 1 and block 2) ................................49 Assessment 4: Post-Block 2 (following treatment block 2) ..................................................49 Pre-Treatment Assessment Results ........................................................................................50 Treatment Strategies and Goals .................................................................................................56 Morphosyntax Goals and Intervention Approach ..................................................................56 Goal Selection for Phonology ................................................................................................57 Goal 1: Syllable and Word Structure................................................................................57 Goal 2: Position or Sequence............................................................................................58 Goal 3: Segments/Features or Feature Combinations ......................................................59 Summary of Goals and their Constraint Rankings...........................................................62 Research Questions: Phonological Treatment Effects ...........................................................64 Treatment Plan........................................................................................................................65 Data Analysis .............................................................................................................................67  v  Chapter 3 - Results.........................................................................................................................70 Global Measures ........................................................................................................................70 T1: Global Measures - Consonants, Word Shapes, and Vowels...........................................71 T2: Global Measures - Consonants, Word Shapes, and Vowels...........................................72 Unacquired Elements: Treated, Direct Generalization, and Indirect Generalization ...............74 T1: All Unacquired Elements.................................................................................................76 T2: All Unacquired Elements.................................................................................................77 T1: Treated Elements .............................................................................................................78 T2: Treated Elements .............................................................................................................82 Statistical Analysis..................................................................................................................86 T1: Direct Generalization Elements.......................................................................................86 T2: Direct Generalization Elements.......................................................................................90 T1: Indirect Generalization Elements ....................................................................................93 T2 Indirect Generalization Elements......................................................................................95 Summary of Phonological Changes .......................................................................................96 Generalization to Connected Speech .........................................................................................98 Qualitative Results .....................................................................................................................99 Similarities and Differences.....................................................................................................100 Twin Similarity.....................................................................................................................100 Twin Difference....................................................................................................................100 Chapter 4 - Discussion .................................................................................................................104 Research Questions ..................................................................................................................104 Discussion of Phonological Changes.......................................................................................111 Limitations of the Study ..........................................................................................................114  vi  Directions for Future Research................................................................................................116 Clinical Implications ................................................................................................................118 Conclusions ..............................................................................................................................119 Works Cited..................................................................................................................................120 Appendices ...................................................................................................................................130 Appendix A: Outcomes Survey ...............................................................................................130 Appendix B – Case History Form ...........................................................................................131 Appendix C - Match Percentages and Ratios for Unacquired Elements ................................135 Appendix D – Transcriptions of CAPES.................................................................................136 Appendix E – Summary of Acquired and Unacquired Non-Linear Elements: Pre-Baseline Assessment ...............................................................................................................................145 Appendix F – Summary of Acquired and Unacquired Non-Linear Elements: Baseline  Assessment ...............................................................................................................................147 Appendix G – [CorLab] Match Data by Item .........................................................................149 Appendix H – [+Lateral] Match Data by Item ......................................................................150 Appendix I – Stress Pattern Match Data by Item ...................................................................151 Appendix J – Word Shape Match Data by Item .....................................................................152 Appendix K – Cluster Match Data by Item ............................................................................154 Appendix L – [-anterior] Match Data by Item ........................................................................157 Appendix M – [-grooved] Match Data by Item ......................................................................159 Appendix N – Treatment Word List........................................................................................160 Appendix O – Sample Treatment Plan....................................................................................161  vii  List of Tables Table 1. English Phoneme Feature Matrix ......................................................................................6 Table 2. Rules and Processes for Productions of /sw!m!"/ at Three Assessments (Ax) ..............10 Table 3. Constraint Ranking for /sw!m!"/ at Assessment 1 ..........................................................12 Table 4. Constraint Ranking for /sw!m!"/ at Assessment 2 ..........................................................13 Table 5. Constraint Ranking for /sw!m!"/ at Assessment 3 ..........................................................13 Table 6. Capability-Focus Framework Typology..........................................................................21 Table 7. Assessment and Treatment Schedule ..............................................................................43 Table 8. Baseline Assessment Results for Non-Phonological Tasks ............................................51 Table 9. Pre-Baseline Assessment: Summary of Unacquired Structures and Features ................53 Table 10. Baseline Assessment: Summary of Unacquired Structures and Features.....................53 Table 11. Pre-treatment Global Measures .....................................................................................56 Table 12. Place of Articulation (POA) of English Approximants ................................................61 Table 13. Mismatch Patterns for Approximants at Pre-Baseline and Baseline Assessment ........62 Table 14. Week-by-Week Treatment Plan.....................................................................................66 Table 15. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match Across Assessment Points for T1..........................................................................................71 Table 16. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match  Across Assessment Points for T2..........................................................................................73  Table 17. Match Data for Treated Structural Elements Across Assessment Points for T1..........81 Table 18. Match Data for Treated Segments Across Assessment Points for T1 ..........................81 Table 19. Match Data for Treated Structural Elements Across Assessment Points for T2..........85 Table 20. Match Data for Treated Segments Across Assessment Points for T2 ..........................85 Table 21. Match Data for Direct Generalization Structures Across Assessment Points for T1...89  viii  Table 22. Match Data for Direct Generalization Features/Segments Across Assessment Points for T1 .....................................................................................................................................90 Table 23. Match Data for Direct Generalization Structure Across Assessment Points for T2 ....93 Table 24. Match Data for Direct Generalization Features/Segments Across Assessment Points for T2 .....................................................................................................................................93 Table 25. Match Data for Indirect Generalization Elements Across Assessment Points for T1..94 Table 26. Match Data for Indirect Generalization Elements Across Assessment Points for T2..95 Table 27. Match Increase by Phonological Element on CAPES Across Assessment Points for T1 ...............................................................................................................................................97 Table 28. Match Increase by Phonological Element on CAPES Across Assessment Points for T2 ...............................................................................................................................................98 Table 29. Match Data for Single Words in Narrative Task for T1 and T2 ..................................99 Table 30. Capability-Focus Framework - Outcome Measurement .............................................101 Table 31. Capability-Focus Typology for T1 and T2..................................................................102  ix  List of Figures Figure 1. English Phoneme Feature Geometry................................................................................7 Figure 2. Interactive Model of Speech Processing ........................................................................19 Figure 3. Feature Matrix: Summary of Needs at Baseline Assessment for T1 ............................54 Figure 4. Feature Matrix: Summary of Needs at Baseline Assessment for T2 ............................55 Figure 5. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match  Across Assessment Points for T1..........................................................................................72  Figure 6. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match  Across Assessment Points for T2..........................................................................................74  Figure 7. Percentage Match for Unacquired Elements across Assessment Points for T1............77 Figure 8. Percentage Match for Unacquired Elements across Assessment Points For T2 ...........78 Figure 9. Treated Element Match Data Across Assessment Points for T1...................................82 Figure 10. Treated Element Match Data Across Assessment Points for T2.................................86  x  List of Abbreviations S – primary stress (strong) w – unstressed syllable (weak) s – secondary stress (strong) C – consonant V – vowel WI – word-initial WM – word-medial WF – word-final DZ – dizygotic MZ – monozygotic Ax – assessment Tx – treatment OT – optimality theory PPD – protracted phonological development MLU – mean length of utterance PCC – percentage consonants correct SD – standard deviation  xi  Acknowledgements Firstly, I would like to thank my primary co-supervisor Dr. May Bernhardt for all of the help and support throughout this entire project, but particularly, throughout the writing phases of this thesis. May’s straightforward feedback and inquisitive commenting were ever so helpful in making this thesis better. I feel so thankful to have had the opportunity to work with a supervisor that shares my same passion. I hope we will have more time to discuss phonology and learning in the future. Thank you to the rest of the UBC research team on this project, especially my co-supervisor Dr. Paola Colozzo. Paola’s expertise in morphosyntactic development was paramount in the planning and assessment phases of this project and her positive attitude, dedication, and thorough work made it happen. Thank you to Charmaine for being my ally during the assessment and treatment portions of the study and for making me laugh. The team at the twins’ school also deserves a thank you. Their dedication and hard work during this project was remarkable and made this study possible. It is truly inspiring to see the great things that their school program is doing. Thank you to my parents who have always believed in me. Thank you to my fiancé Daniel for unconditional love and forthright advice. Thank you to my roommates Dana, Oshy, and Bun for their support and encouragement during the writing phases of this thesis, for listening to me complain and babble about boring technical issues, and for hearing me out when I discovered new and interesting things in my data. Finally, a special thank you to the family for participating in this research. It was truly a pleasure to work with such devoted parents and such affectionate children.  xii  Dedication To the two children who participated in this study.  xiii  Chapter 1 - Introduction Protracted phonological development (PPD), as a distinct and persistent delay in speech (phonological) development, has been the source of considerable academic and practitioner discourse over the last few decades. First used by Bernhardt and Stemberger (1998), the term PPD applies to children with greater mismatch to adult phonological targets than would be expected based on chronological age. Mismatch patterns typically persist past the age at which they generally disappear and include deletion of weak syllables and consonants, substitution of stops or glides for other sound classes, and changes in voicing or place of articulation (either further forward or further back in the oral cavity). The benefits of intervention for children with protracted phonological and morphosyntactic development have been well-documented with experimental evidence (Almost & Rosenbaum, 1998; Law, Garrett, & Nye, 2004; Tyler, Lewis, Haskill, & Tolbert, 2002; Tyler, Lewis, Haskill, & Tolbert, 2003). However, there has been minimal research on phonological intervention with twins (Edwards & Bernhardt, 1973; Von Bremen, 1990). The current study addresses this gap. The primary purpose of this thesis is to evaluate the phonological development of two twin boys over the course of a 16-week treatment study, and specifically, to quantify and explain the changes in each twin’s phonological system during that period. The following literature review outlines foundations for the research: (1) theoretical frameworks in phonology and psycholinguistics, (2) intervention studies concerning children with PPD, and (3) twin language development.  Literature Review In an age of evidence-based practice, the methods used in the assessment and remediation of PPD are developed through a process of asking questions, evaluating potential answers, and arriving at increasingly effective methods of practice. In order to understand PPD  1  it is important to consider (1) how the phonological system is defined, organized, and represented, (2) how the phonological system is related to other areas of the linguistic system (e.g., syntax, morphology, and semantics) and other cognitive processes (e.g., memory, problem solving), and (3) how the phonological system grows and changes. The first two issues will be addressed in the section of the literature review entitled ‘Phonological Theory and Goal Selection for Intervention’, which focuses primarily on constraint-based non-linear phonology as a framework to consider in phonological development and disorders. A discussion of the contribution that psycholinguistic theories have made in the area of phonological assessment and intervention provides the context for considering the interconnection between phonology and other areas of language and learning. The section entitled ‘Intervention for PPD’ considers how the phonological system grows and changes by outlining some of the methods that have been implemented in previous intervention studies for PPD. Where it exists, empirical evidence surrounding the strengths and weaknesses of these approaches is discussed. The third and final section of this literature review, ‘Intervention with Twins,’ discusses phonological and morphosyntactic development in twins and summarizes research in phonology involving twin pairs. This section sheds light on questions specific to phonological development in twins: (1) Are phonological development and general language learning different in twin pairs than in singleton children? (2) Are the phonologies of twins likely to be similar, and if so, to what degree?  Phonological Theory and Goal Selection for Intervention Current clinical practice has been shaped by a host of theories from a variety of disciplines including, but not limited to, biology, linguistics, neuroscience, computer science, and psychology. These theories have aimed to account for developmental data, cross-linguistic  2  data, information processing, and learning. Linguistic theory (specifically phonological theories) has been particularly influential in how phonological development is understood. Phonological Theories Over the last several decades, the linguistic theories that have informed phonological treatment have evolved significantly. Traditionally, phonological difficulties were thought to result from an articulatory-motor disability concerning specific consonants and vowels (e.g., Van Riper, 1953). The works of Leija McReynolds (1975), David Ingram (1976), and others precipitated a shift in the field of clinical practice from an articulatory-motor focus to a phonological or linguistic focus. Ingram suggested that the principles of generative phonology (Chomsky & Halle, 1968) be applied to the analysis of articulation difficulties in children. Generative phonology breaks sounds down into smaller units called features. Features describe aspects of how speech sounds are produced: manner class (e.g., continuant or noncontinuant), laryngeal characteristics, (e.g., voiced or voiceless), and place of articulation (e.g., labial, with the lips). For example, according to the feature system of Bernhardt and Stemberger (1998, 2000), the phoneme /s/ is made up of the manner features [-sonorant], [+consonantal] and [+continuant], the laryngeal features [-voiced] and [+spread glottis], and the place features [Coronal], [+anterior], and [+grooved]. Conversely, the phoneme /f/ is made up of all of the same manner and laryngeal features but a different set of place features [Labial] and [+labiodental]. These two phonemes can thus be distinguished by their place features, [Coronal] vs. [Labial]. The original set of features proposed by Chomsky and Halle (1968) were followed by many other proposals. The list of features used for this project are as described by Bernhardt and Stemberger (1998)1. The matrix in Table 1 shows the features that make up each  1  The features that have been included here are only those required to give an overview of  English phonology as it applies to the participants in this study.  3  English phoneme. Some of these features are labeled using common terminology (e.g., consonantal and voiced), making their definitions transparent. Others, however, are labeled using less commonly used terms (i.e., continuant, sonorant, spread glottis, labial, coronal, anterior, grooved, and dorsal). These are defined as follows: [+continuant] or [+cont]: Speech sounds in which a continuous airflow travels out of the oral cavity. [+sonorant] or [+son]: Speech sounds in which the vocal folds continue to vibrate without interruption and without a rise in pressure above the larynx. [+spread glottis]: Speech sounds in which the vocal folds are abducted, creating a low amplitude noise. [Labial]: Speech sounds that involve one or both lips as articulators. [Coronal]: Speech sounds that are made with the tip or blade of the tongue. [+anterior]: Coronal speech sounds in which the tongue articulates at or in front of the alveolar ridge. [+grooved]: Coronal speech sounds that involve the creation of a narrow groove or space in the middle of the tongue. [Dorsal]: Speech sounds that are made with the back of the tongue body. Figure 1 presents a geometric diagram of the features and featural organization within the phonological system of a typical adult English speaker. This feature hierarchy embeds the notion that certain features have dominance over others in the phonological system, adding to the concept of non-linearity in phonology. In addition to representing manner, laryngeal characteristics, and place of articulation, this geometric figure uses hatched lines to represent multiple manners or multiple places of articulation that occur in English. The affricates /!/, /"/ can be described as a sequence of [-continuant] and [+continuant]. Multiple places of articulation (co-articulations) are common in the approximant category. Thus, the phoneme /#/ is  4  both [Coronal] and [Labial] (also called [CorLab]) word-initially, the phoneme /l/ is [Coronal] and [Dorsal] word-finally, /w/ is [Labial] and [Dorsal], and /j/ is [Coronal] and [Dorsal]. These combinations are specifically relevant to the phonological systems of the participants in this study. They are discussed further in the Methods chapter when goal selection is described.  5  Table 1. English Phoneme Feature Matrix p  t  k  b  d  g  f  !  s  "  #  v  $  z  %  &  w  l  j  '  m  n  (  h  )  consonantal  +  +  +  +  +  +  +  +  +  +  +  +  +  +  +  +  -  +  -  +  +  +  +  +  +  continuant  -  -  -  -  -  -  +  +  +  +  -/+  +  +  +  +  -/+  +  +  +  +  -  -  -  +  -  sonorant  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  +  +  +  +  +  +  +  +  +  nasal  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  +  +  +  -  -  lateral  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  +  -  -  -  -  -  -  -  voiced  -  -  -  +  +  -  -  -  -  -  +  +  +  +  +  +  +  +  +  +  +  +  -  -  spread glottis  +  +  +  +  +  +  +  +  Manner  Laryngeal +  +  Place Labial  !  !  !  !  !  Coronal  !  !  !  !  !  !  !  !  !  !  !  !  (!)a !  anterior  +  +  +  +  -  -  +  +  -  -  +  -  -  grooved  -  -  +  +  +  -  +  +  + (!)b  (!)  Dorsal  !  !  !  Note. Adapted from Bernhardt and Stemberger (1998, 2000). a  /'/ has a [Labial] component in syllable-initial position. It also has a [Dorsal] component across positions.  b  /l/ has a [Dorsal] component in syllable-final position and when it occurs with a back vowel.  6  ! ! + !  Figure 1. English Phoneme Feature Geometry  Note. Adapted from Bernhardt and Stemberger (1998)  These features of generative phonology are used to identify linguistically-based patterns in children’s mismatches with adult targets. Generative phonology identifies the rules2 or the processes3 responsible for these mismatches, using them to explain the relationship between the  2  Rules are procedures that change the pronunciation of words in the presence of certain sounds  (Bernhardt & Stemberger, 1998, p. 21). 3  Processes are procedures that change the underlying representation of a word, causing a  child’s production to be different from an adult’s (Bernhardt & Stemberger, 1998, p. 23).  7  underlying representation of a word’s structure and the child’s (surface) production (Baker, Croot, McLeod, & Paul, 2001). The following example taken from the data of the present study illustrates how rules can be written. When the word “swimming” (/sw!m!"/) is misproduced as [fw!m!"] the segmental pattern is that /s/ becomes [f] before /w/, or (1) [s] # [f] / __ [w] This pattern could be stated by listing features, rather than by listing segments as follows: (2) Coronal # Labial / __ Labial The process illustrated above is place assimilation, with the /s/ taking on the place of articulation (i.e., the place feature ‘Labial’) from the phoneme that follows it, /w/, to become [f]. Describing differences between adult targets and child productions using features is particularly useful because a single rule can be written to represent a class of related differences. For example, the feature-based example (2) represents place assimilation for all coronal consonants that precede labial consonants. The system described in examples 1 and 2 can be used to describe patterns in child pronunciations and to select intervention goals with the purpose of eliminating groups of phonological processes. A number of phonological process approaches were introduced in the 1980s and 1990s, including those of Hodson and Paden (1991), Stampe (1979), and Howell and Dean (1991). Using phonological processes to describe phonological changes in children’s speech is advantageous because it describes the patterns in children’s productions. However, processes do not limit which phonological changes can occur and therefore do not necessarily predict outputs that are typical of human language. Furthermore, each process is generally viewed as independent of other processes, and therefore there is no internal coherence. Finally, no satisfactory basis for the existence of processes has been posited (Bernhardt & Stemberger, 1998). This is contrary to the goals of linguistic theory, which aims to develop a system of  8  language learning and storage that is efficient and predictive, and the goals of psycholinguistic theory, which aims to develop a system consistent with what is known about information processing in neural networks (discussed further below in ‘Interactive Psycholinguistic and Dynamic Systems Theories’). Having said this, a system that uses principles and patterns rather than listing processes creates a theory that organizes and represents the phonological system efficiently. The process-based system applied in standard generative phonology has given way to a theory that is better able to address the questions posed at the beginning of this chapter regarding the structure, growth, and interactions of the phonological system. This is optimality theory (OT, first described by Prince & Smolensky, 1993). OT describes phonological patterns as the result of a single set of interacting constraints. The Constraints of Optimality Theory (OT) According to OT, two general types of violable constraints exist and compete with one another: faithfulness (positive) constraints and markedness (negative) constraints. Faithfulness constraints attempt to keep the form that is produced in speech (the output form) the same as (or faithful to) the input form (or the‘underlying’ representation), whereas markedness constraints attempt to keep the output form as simple (or unmarked) as possible. These constraints apply to each aspect of the phonological system and are ranked relative to one another. Constraints limit the types of pronunciation changes that can occur in the productions of speakers because important, or highly ranked, constraints are rarely violated. As mentioned above, a processbased framework is basically unmotivated (arbitrary) (Bernhardt & Stemberger, 1998). In contrast, a constraint-based framework applies principles that are neither arbitrary nor complex. Here, the surface form produced is an attempt by the speaker to be faithful to the input form (‘positive’ constraints) without violating highly ranked negative constraints on the linguistic output. For a child learning the rankings in his or her own language, once faithfulness outranks markedness, then the child has matched the adult target pronunciation. However, many  9  markedness constraints will remain highly ranked in the adult system, ensuring that language forms do not become excessively marked. Thus, the output realized in this system is a balance between being as unmarked as possible and being as faithful to the input as possible. Observable ‘processes’ are a result of the interaction of the constraints, and in fact, many processes or repairs can occur as a result of the same set of constraints. This constraint-based system of OT can be illustrated by using the production of the target ‘swimming’ by one of the twin participants (see Table 2). At the first assessment point ‘swimming’ was produced as [fw!m!"], four months later as [s!m!"], and four months after that, correctly as [sw!m!"]. Using a processes account to describe these productions, the first elicitation would be explained as place assimilation (/sw!m!"/ # [fw!m!"]), the second as consonant cluster reduction through deletion (/sw!m!"/ # [s!m!"]), and the third would be considered process-free (/sw!m!"/ = [sw!m!"]). Table 2. Rules and Processes for Productions of /sw!m!"/ at Three Assessments (Ax) Ax  Production  Rule  Process  1.  /sw!m!"/ # [fw!m!"]  Coronal # Labial / __  Place assimilation  Labial 2.  /sw!m!"/ # [s!m!"]  C # $ / C__  Cluster reduction  3.  /sw!m!"/ = [sw!m!"]  None  None  As illustrated in Table 2, two rules corresponding to two processes are required to explain the productions at the first two assessment points. Also, in order to explain the correct production at assessment 3, the two processes that were previously observed must somehow have disappeared. It is difficult to motivate the disappearance of these processes, especially without an adequate explanation for their initial presence. The same set of three elicitations presented in Table 2 can be explained using OT, based on the reordering of a single set of four constraints. OT posits that there are constraints against  10  producing marked forms such as consonant clusters and non-homorganic consonant clusters (clusters of consonants with differing places of articulation). The child must deal with these markedness constraints while attempting to stay faithful to the input form. The dynamic ordering of constraints represents the process of organization that the child’s language system (in this example the phonological system) goes through. The terminology of Bernhardt and Stemberger (1998) is used here in the constraint tabels: faithfulness is indicated by “Survived” and markedness by “No” or “Not”. 1) Survived(Timing Unit): No deletion of the timing unit associated with each segment in a word. This is a faithfulness constraint. Ranking this constraint highly disallows output forms such as [s!m!"] due to deletion of /w/. 2) Survived(Place) or Survived(Coronal): Place of articulation in the output must match the input. This is a faithfulness constraint. Ranking this constraint highly disallows output forms such as [fw!m!"] due to the lack of faithfulness to place of articulation of the /s/. 3) NotComplex(Onset): No clusters in the onset. This is a markedness constraint. Ranking this constraint highly disallows output forms such as [fw!m!"] and [sw!m!"] due to the presence of consonant clusters. 4) NoSequence(Coronal...Labial): No sequences of [Coronal]-[Labial] in adjacent phones. This is a markedness constraint. Ranking this constraint highly disallows forms such as [sw!m!"] due to the different place features in the adjacent consonants /sw/. Up until the final assessment point, this was the highest-ranked constraint. (See Tables 3-5.) Tables 3 to 5 show how different constraint orderings resulted in different output forms for /sw!m!"/ at the three assessment points. The rankings represent the relative importance of each constraint, with more highly ranked constraints being more important. In each constraint  11  table, a star (*) indicates when a constraint has been violated. According to OT, all output candidates violate constraints; however, the optimal candidate is one that does not violate the most highly ranked constraints. This optimal output form is realized in production. Exclamation points (!) are used to indicate ‘fatal’ violations; fatal violations occur when an output candidate is determined not to be optimal. At the first assessment point, when the form [fw!m!"] was produced, Survived(Timing Unit) and NoSequence(Coronal…Labial) were ranked above Survived(Coronal) and NotComplex(Onset). This high ranking of Survived(Timing Unit) caused a fatal violation for the output form [s!m!"]. Also, the high ranking of NoSequence(Coronal…Labial) caused a fatal violation for the output form [sw!m!"]. Table 3 represents this constraint ranking and shows how it resulted in [fw!m!"] as the optimal form. Table 3. Constraint Ranking for /sw!m!"/ at Assessment 1 /sw!m!"/  Survived(Timing Unit)  NoSeq(Cor...Lab) NotComplex(Onset) Survived(Coronal)  ! fw!m!" s!m!" sw!m!"  * (optimal)  *  *! *!  At the second assessment point the faithfulness to the place of articualtion of /s/ changed relative to the faithfulness to timing units as represented in clusters. The constraints Survived(Coronal), NoSequence(Coronal...Labial), and NotComplex(Onset) conspired against Survived(Timing Unit). This made the markedness of consonant clusters more highly ranked than faithfulness to the structural input, but also fostered accurate production of the /s/. Thus, both [fw!m!"] and [sw!m!"] incurred fatal violations due to their initial consonant clusters (see Table 4).  12  Table 4. Constraint Ranking for /sw!m!"/ at Assessment 2 /sw!m!"/ Survived(Coronal) NotComplex(Onset) Survived(Timing NoSeq(Cor...Lab) Unit) fw!m!"  *!  * *(optimal)  !s!m!" sw!m!"  *!  *  At the third assessment point, Survived(Timing Unit) and Survived(Coronal) were ranked above NoSequence(Coronal...Labial) and NotComplex(Onset). The markedness constraints against complex consonant clusters and non-homorganic clusters were ranked below the faithfulness constraints, resulting in [sw!m!"] as the optimal candidate (see Table 5). Table 5. Constraint Ranking for /sw!m!"/ at Assessment 3 /sw!m!"/  Survived(Timing Unit)  fw!m!" s!m!" !sw!m!" a  Survived(Place)a  NoSeq(Cor...Lab) NotComplex(Onset)  *!  *  *! * (optimal)  *  Using Survived(Place) instead of Survived(Coronal) preserves both /w/ and /s/.  As this example shows, OT accounts for phonological changes by identifying relevant constraints and constraint rankings. It satisfies the goals of linguistic theory in that it efficiently represents changes to production. Similarly, it is able to predict which productions will and will not occur. The constraints of OT and their rankings may be motivated phonetically by speech perception, by the characteristics of the vocal tract, or by cognitive processing limitations (Bernhardt & Stemberger, 1998). The constant change in constraint rankings represents an attempt by the child to produce increasingly adult-like outputs. These changes often require what appear to be temporary regressions. The child’s attempt to arrive at the adult-like  13  constraint rankings may be governed by algorithms, similar to solving a Rubik’s cube. The relationship that is formed by ranking and re-ranking constraints may represent a dynamic balance between ease of perception and ease of articulation, thus addressing how top-down and bottom-up information interacts in language processing. Psycholinguistic theories regarding how production and perception interact in processing will be discussed further below. Overall, the constraints of OT have been important for understanding how children who have PPD may arrive at different non-adult-like productions throughout development. The current thesis utilizes constraint rankings and processes to describe the phonologies of the twins. Non-linearity Prior to the development of OT, linguistic theories regarding phonology began to focus on the non-linear aspect of phonological representation. Concepts of non-linear phonological representation began to gain momentum in the late 1970s with Goldsmith’s dissertation on tone (Goldsmith, 1976, as cited in Bernhardt, 1990). The term non-linear refers to (1) the representation of prosodic and segmental aspects of phonology as hierarchical rather than linear, and (2) the conceptualization that changes to output forms in production are caused and constrained by tier associations (Bernhardt, 1990), as portrayed in example 3 below. Using these assumptions, children’s productions can be understood in relation to a larger structure of surrounding sounds, syllables, and words. As an example, consider the phonological process ‘place assimilation’ presented in example 2 above, whereby the feature [Coronal] changes to [Labial] in the first consonant of a cluster when the second consonant is [Labial]. This pattern can be represented by considering two tiers or levels, the phoneme tier and the feature tier. In example 3 below, the feature [Coronal] becomes delinked from the first consonant (C) and the feature [Labial] spreads to it.  14  (3) C || ||! ! Coronal  C | Labial  The use of non-linear representations such the one in example 3 for children with PPD can help clinicians to understand how unacquired phonological elements are related to the other aspects of the phonological system. Constraint-Based Non-linear Phonology Constraint-based non-linear phonology emerged along with OT. This framework can guide phonological intervention by considering both constraint rankings and non-linear geometry/hierarchy of the phonological systems of children with PPD. Empirical support for this method of addressing PPD is growing (Bernhardt & Stemberger, 1998; Bernhardt, 1990; Von Bremen, 1990; Major & Bernhardt, 1998; Bernhardt & Major, 2005; Shoaf, Iyer, & Bothe, 2009). Constraint-based non-linear phonology analyzes the phonological system in terms of its hierarchically organized features, segments, and word structures and their interactive constraint rankings (Bernhardt & Stemberger, 1998, 2000; Barlow & Gierut, 1999). Goals for intervention are selected by determining which aspects of this hierarchical framework are absent or unorganized. Intervention deliberately targets various levels within the hierarchy. Faithfulness to the adult target forms is promoted, which reduces the impact of high-ranked negative markedness constraints. One important advantage of non-linear constraint-based phonology is that it is able to predict certain phenomena seen in phonological development. For example, the dynamic nature of the constraint rankings is consistent with both variability in child pronunciations and with developmental progression/regression over time. Also, non-linear representation provides an explanation for certain types of generalization. The focus on features,  15  structures, and their relationships shows how treatment can result in generalization to untreated sounds, positions, sequences, and word shapes. The application of non-linear constraint-based phonology requires information from phonetics, feature geometry, word structure, constraints, and repairs/processes. Bernhardt has applied non-linear theories in intervention with children with PPD in a number of studies (e.g., Bernhardt, 1990, 1992, 1994; Major & Bernhardt, 1998; Bernhardt & Major, 2005). One follow-up study of 19 children with PPD who had received non-linear phonological intervention showed better long-term outcomes for phonology, language, and literacy than has generally been reported in the literature based on other approaches (Bernhardt & Major, 2005). Constraint-based non-linear phonology combines the hierarchical and non-linear organization of features with the interaction of constraints (from OT) to form a theory that explains the structure of the phonological system and shows how changes can occur throughout development. This theory creates a framework that can help clinicians understand the speech of children with PDD and can aid in planning phonological intervention. This theory has guided the methods of the present study. Unacquired non-linear elements of the participants’ phonological systems and relevant constraint rankings were identified for the participants and were targeted in therapy. Bernhardt and Stemberger (2000) propose some specific guidelines for selecting targets for intervention. These guidelines are described in the next section, and were followed in the goal selection process for the current study. Typical Phonological Targets for Intervention Determining the established and unestablished aspects of children’s phonological systems involves collecting a speech sample that represents a variety of features and segments in different word positions, and a variety of word shapes, stress patterns, vowels, and consonant  16  combinations. The Computerized Articulation and Phonology Evaluation System (CAPES) wordlists have been developed for this purpose (Masterson & Bernhardt, 2001). Based on a non-linear analysis, four general types of targets can be identified for phonological intervention (Bernhardt & Stemberger, 2000). The four categories consider (1) the element that is targeted (i.e., syllable and word structure or segment and feature), and (2) whether or not that element is new in the child’s system or pre-existing (old). “New” elements are defined as “syllable and word structures or features that are missing (or very marginal) in the child’s system” (p. 52). “Old” elements are elements that have already been acquired by the child but in a different position or in a different combination. Thus the four goal types are: 1. New syllable and word structures, e.g., specific stress pattern: weak-strong (wS) (as in /b!"lun/); or e.g., specific word structure: consonant-consonant-vowel-consonant (CCVC) (as in /bl#k/) 2. New individual features and segments, e.g., [+lateral] /l/ 3. Old features in new combinations e.g., [Labial] & [+continuant] (/f/, /v/) for a system that already contains [Coronal] continuants (/s/) and labial stops (/p/, /b/). 4. Old segments in new word positions or in new sequences for those segments e.g., [Dorsal] in word-initial position, when the feature already is present wordfinally. When selecting goals, the Bernhardt and Stemberger (2000) approach suggests that both structural and segmental targets should be selected, and also both ‘old’ and ‘new’ elements. This ensures that already acquired elements are used to support the acquisition of new elements, thus “using strengths to bolster needs” (Bernhardt & Stemberger, 2000, p. 52). For example, in the  17  treatment of a new word shape, phonemes that are already used consistently by the child would be used in the treatment words. The idea that new and old elements should be combined is consistent with interactive psycholinguistic theories, which propose that all elements of a task place cognitive demands on the child, particularly when complex or unfamiliar information is presented. With only a finite set of resources available for speech processing, presenting several new elements at once may overload the child’s system (Bernhardt, Stemberger, & Charest, 2010). Thus, using acquired segments to target new word shapes allows the child to dedicate resources toward processing the new word shape with no need to articulate new phones. Interactive Psycholinguistic and Dynamic Systems Theories Further to system interactions, psycholinguistic and dynamic systems theories have attempted to explain how information is processed at a cognitive level. Interactive psycholinguistic models postulate that the flow of information in speech processing is multidirectional. Neural representations are encoded through activation patterns that represent various levels of linguistic detail (i.e., morpheme, syllable, rime, phoneme, feature) that have been shown to be psychologically significant. The multidirectional interactive connection between different levels shows how effects on one level can propagate to other levels. One example of an interactive model is the spreading-activation model developed by Dell (1986). This model is able to represent children’s mismatches at specific levels of the linguistic system, showing how representations are related to one another. Figure 2 is a simplified model of how the activation of different levels can affect specific productions. A lack of adequate activation at one of these levels would lead a child to mismatch the adult production. For example, low activation of the feature [+continuant] in word-initial position might lead the words /st!"/, /sl!#d/, and /spul/ to be produced as [t!"], [l!#d], and [pul].  18  Figure 2. Interactive Model of Speech Processing  Note. Based on Dell (1986)  Figure 2 illustrates a simplified set of interactions between morphemes, syllables, onsets, phonemes, and features. However, interactions also involve other levels of the linguistic system, i.e., semantics, syntax, and pragmatics and other areas of cognition such as attention and perception (Bernhardt et al., 2010). Interactive models represent cognitive processing as a system where all aspects of the linguistic system and cognition are interrelated. Dynamic systems theory models also consider the interactions between interrelated variables, and specifically how new behaviours are learned through the interaction of developmental, task, and environmental variables (Rvachew & Bernhardt, 2010). This complex and dynamic system of interactions involves different domains of the child’s linguistic system and non-linguistic systems plus factors external to the child. Periods of equilibrium and diseqilibrium are reached as learning takes place. Like interactive theories, dynamic system theory predicts that change in one area of a child’s system (e.g., through intervention) can lead to changes in other systems. Like optimality theory, dynamic systems theory can account for temporary regressions in development. Given these interconnections, developing stability in one area may necessitate changes in another area that had previously stabilized. In addition, dynamic  19  systems theory considers both external and internal factors in a child’s learning. It follows that clinicians should consider the interaction of variables such as encoding and retrieval, depth of processing, and contextual effects, and adjust treatment accordingly. As interactive and dynamic models become increasingly able to explain how all aspects of development dynamically interact, they lead to new assessment and treatment approaches and to deeper understandings of the nature of PPD in the context of child development. The interactive connections proposed by these theories are important to consider when looking at how intervention in specific areas of language results in development of other linguistic systems, such as the phonological system. This cross-over of learning from one domain to another is of particular interest in the present study. Another important consideration is how different aspects of intervention interact and place demands on the child’s cognitive resources. The current study was designed with both a phonological treatment block and a morphosyntactic treatment block in order to focus the learning on one aspect of the system at a time, but with the perspective that cross-domain learning might ensue. In addition, other factors relative to system interactions were considered during the treatment planning process: e.g., the length and phonological complexity of the items chosen for intervention, the vocabulary items used, and the complexity of the activities chosen. Within-child variables were also considered important for the planning process as described in the next section. Child Capability and Focus Dynamic systems theory highlights the perspective that many factors interact in development, including within-child variables. Although many of these child-internal factors are difficult to measure, it is important to consider what each child brings to his or her own language learning experience and the potential effects of within-child variables on treatment outcomes. In order to address these additional variables, Kwiatkowski and Shriberg (1998)  20  developed the ‘capability-focus’ treatment framework. This framework posits that treatment outcomes can be better understood through considering a child’s capability (phonological production, phonological comprehension, oral mechanism, and cognitive-linguistic and psychosocial abilities) and focus (attention, motivation, and effort). Understanding how a child’s capability and focus interact with learning can help the clinician to adjust treatment to the child’s individual needs. In Kwiatkowski and Shriberg’s model, the typology for the capability-focus framework consists of a four-box matrix: high capability - high focus; high capability - low focus; low capability - high focus; and low capability - low focus (see Table 6). Table 6. Capability-Focus Framework Typology Focus  High Capability Low  High  Low  High  High  Capability  Capability  High Focus  Low Focus  Low  Low  Capability  Capability  High Focus  Low Focus  Adapted from Kwiatkowski and Shriberg (1998)  Like the psycholinguistic perspective, the capability-focus framework is concerned with how treatment targets are taught rather than which targets are taught. Each child is viewed as an active participant in learning and no single intervention approach is necessarily considered to be suitable for all children. The capability-focus framework is also useful in understanding how individual children respond to treatment. It enables the clinician to quantify within-child variables that non-linear phonological, constraint-based, and interactive theories do not address. For the current study, this framework is useful for understanding how the twin participants, who are similar in many ways, may respond differently to treatment based on within-child variables.  21  Summary of Key Concepts The preceding overview of phonological and psycholinguistic theories has described the basic tenets and advantages of constraint-based non-linear theories in phonology and how these theories have guided goal selection for this study. The overview has also addressed phonological development from an interactionist point of view, with the perspective that learning in one language domain may be linked to learning in other domains. Finally, this overview has considered variables that are specific to the child as an active individual in phonological therapy by presenting the capability-focus framework. The capability-focus framework is used to consider how individual differences are related to treatment outcomes in the participants of this study. With these important background concepts established, the approaches to intervention used in this project will now be outlined in the next major section of the literature review.  Intervention for PPD Treatment Methods Phonological treatment studies frequently report positive changes to the phonological systems of participant children (Almost & Rosenbaum, 1998; Law, et al., 2004; Tyler et al., 2002; Tyler et al., 2003). However, the majority of intervention outcome studies for PPD have used small sample sizes and few have used randomized control (Baker & McLeod, 2004; Hodson, 1983; Pascoe, Stackhouse, & Wells, 2005; Tyler, Edwards, & Saxman, 1987). Although the level of evidence in the body of literature concerning the effectiveness of phonological intervention has generally been low, there have been a few studies using randomized control (Almost & Rosenbaum, 1998) and at least one meta-analysis (Law et al., 2004). Almost and Rosenbaum’s (1998) randomized control trial study used two groups of 15 children between 33 and 61 months of age. All children had severe phonological delays based on the Assessment of Phonological Processes-Revised (Hodson, 1986), and performed more  22  than one standard deviation (SD) below the mean on the Reynell Developmental Language Scales-Revised (Reynell, 1977). A modified cycles approach using a traditional hierarchy and minimal pairs was used. These treatment approaches are commonly used in phonological intervention (see below). Almost and Rosenbaum found that four months of speech intervention in the experimental group (called the ‘early treatment’ group) resulted in significantly greater improvements in articulation and in mean length of utterance than four months of spontaneous development in a delayed intervention control group (called the ‘late treatment’ group). Furthermore, these changes were maintained in the experimental group four months posttreatment. The meta-analysis by Law et al. (2004) combined 13 studies in order to evaluate the efficacy of both phonological and morphosyntactic intervention. The results specific to phonology showed that participants who received phonological intervention improved significantly more than participants receiving no treatment. These studies have begun to establish the importance of phonological intervention in children with PPD. In addition to asking whether or not phonological treatment is effective, research studies have also investigated outcomes in terms of efficiency. Many diverse and well-established treatment approaches have originated from the theoretical frameworks that underlie the diagnosis and assessment of PPD. The five approaches described here have all gained some empirical support and informed the design of the current project. They are: the traditional approach, the cycles approach, the minimal pair approach, the involvement of significant others and perceptually focused treatment. The traditional approach (initiated by Van Riper, 1953) begins with auditory discrimination of contrasting speech sounds and moves toward sound production. Unacquired sounds are targeted individually and in one word position at a time. Production is trained first in isolation (generally in imitation) and then in contexts of increasing complexity and increasing spontaneity (i.e., syllables, familiar words, phrases, and conversation). This approach is  23  designed to foster better perceptual and motor skills in order to incorporate new phonological targets into speech. The major disadvantage of using a pure traditional approach is that it does not consider the structure of the child’s phonological system. The traditional approach is based on single phonemes and articulatory-motor facilitation. Treatment proceeds from segment to segment without consideration for overall patterns affecting groups of segments. One retrospective study compared outcomes in 36 children who had received either traditional therapy or phonological therapy, specifically a minimal pairs therapy approach (described below) that targeted processes (Klein, 1996). Consonant error scores were calculated using the Arizona Articulation Proficiency Scale (Fudala, 1970) based on existing articulation test results (i.e., Goldman-Fristoe Test of Articulation, Goldman & Fristoe, 1986). The mean error score for children who received traditional speech therapy was 25.95 (SD 5.20) before therapy and 12.84 (SD 5.40) after therapy. The mean error score for children who received phonological therapy was 23.76 (SD 3.94) before therapy and 1.53 (SD 3.94) after therapy. This difference in outcomes was statistically significant. Although newer approaches to articulation therapy have overtaken the traditional approach, several aspects of the traditional approach are still applied, including the move from imitation to spontaneous speech, the gradual increase in contextual complexity (syllable to word to phrase), and perceptual training. Like the traditional approach, the cycles approach (as introduced by Hodson and Paden, 1991) includes aspects of both perception and production of target sounds, often within a phonological process approach (although a cycles approach has also been incorporated into a non-linear approach to intervention, e.g., Bernhardt, 1990, 1992). Target phonological processes, speech sounds, or word structures are treated in turn and continually revisited or recycled until the child has mastered each. As noted, the treatment includes both perception and production training. ‘Auditory bombardment,’ or repeated auditory exposure to a target phonological form, is used to enhance speech perception; imitation training is used to pair  24  auditory and kinesthetic information; and phonological awareness skills are targeted in language and literacy tasks (Prezas & Hodson, 2010). One advantage of this approach is that it exposes the child to a wide range of sounds and structures. According to Gierut (1998), the cycles approach allows the child to process targeted sounds or structures in parallel throughout the cycle. Tyler et al. (1987) compared a minimal pairs approach to a cycles approach in a treatment study of four children aged 3;1 to 5;1. The study targeted patterns in the speech of participants including gliding, depalatalization, stopping, and fronting. Tyler et al. (1987) suggest that, because the cycles approach gives equal focus to a number of processes or goals, it may be particularly adapted for children who exhibit multiple idiosyncratic patterns. Minimal pair treatment is a process-based approach derived from generative phonology. It focuses on production of words that differ in only one speech sound, feature, or structural component in order to demonstrate to the client that errors can lead to communication breakdowns (Barlow & Gierut, 2002). A word that the child produces as a mismatch is often paired with the target word. Showing that confusion arises when the two words are pronounced identically, emphasizes the importance of their minimally contrastive pronunciations. For example, in order to target cluster reduction, the words /sp!t/ and /p!t/ can be contrasted, thereby demonstrating that the production of the /s/ is essential in order to distinguish the two words from one another. According to Weiner (1981), minimal pairs treatment is particularly useful for speech sounds not used contrastively by a child (the target and its substitution), although Gierut and others (Gierut, 1992, Barlow & Gierut, 2002; Gierut & Neumann, 1992) have proposed that it is possible to pair speech sounds/features that are both outside the child’s inventory. However, as noted by Tyler et al. (1993), the successful implementation of this approach requires that the child possess the perceptual and conceptual abilities to understand the contrasts used.  25  A basic assumption of phonological intervention is that a child requires opportunities to practice. For this reason, significant individuals in the child’s life (parents, teachers, peers, etc.) are often encouraged to be actively involved in treatment activities both within treatment sessions and outside of therapy (e.g., Bernhardt, 1990). One approach, the parents and children together (PACT) approach has been described by Bowen and Cupples (1999). Parents are asked to complete therapy activities with the child five to six times per week and the child participates in treatment sessions with the clinician only every 7 to 10 days. As Fey (1999) points out, there is unfortunately little information available about for whom the PACT approach works best. Perceptually-focused treatment may involve discrimination of minimal pairs as described above, and/or involve training children to distinguish between accurate and inaccurate productions of phonemes in words (see e.g., Rvachew, Nowak, & Cloutier, 2004). Rvachew (1994) studied the articulation of 27 preschoolers who misarticulated the phoneme /!/ in an object naming task. They were given either perceptual training on the word ‘shoe’ or a control treatment. Rvachew found that the children who had received the perceptual training for ‘shoe’ improved more in their articulations of /!/ on the object naming task than the control group. Rvachew suggests that some children with PPD have weak perceptual skills while others do not and that this treatment approach may only be effective when underlying perceptual skills are weak. Choosing the intervention approach or approaches for phonological therapy is an important part of intervention planning. These decisions are closely related to how specific goals are targeted, which is the topic of the next sub-section. Goal Attack Strategies In addition to determining the child’s phonological needs and choosing a treatment approach, clinicians must decide when and for how long to target each goal. Several goal attack strategies have been investigated in past research. These studies have primarily focused on  26  children with both phonological and morphosyntactic delays. Tyler et al. (2002) compared a phonology followed by morphosyntax goal strategy, with a morphosyntax followed by phonology goal strategy. Each child received two 12-week blocks of treatment. The participants included 20 preschoolers (10 in each condition) aged between 3;0 and 5;11 with phonology and morphosyntax scores at least one SD below the mean on standard measures. These tasks included the Bankson-Bernthal Test of Phonology (Bankson & Bernthal, 1990), a conversation sample, and a narrative elicitation. The measures were percentage of correctly articulated consonants (percentage consonants correct, PCC), mean length of utterance (MLU), and change for specific morphemes. PCC pre-treatment scores ranged from 33% to 81%, a wide range from severe to mild impairment. The study found that both phonology and morphosyntax improved regardless of the goal attack strategy, but that morphosyntax improved more in the morphosyntax followed by phonology group. Unfortunately, there was no analysis of whether or not improvement after each goal attack strategy was related to level of phonological impairment. A second study by Tyler et al. (2003) included four different goal attack strategies over 24 weeks of therapy. Participants were 47 children between 3;0 and 5;11 years of age with phonology and morphosyntax skills at least 1 SD below the mean according to standardized tests. The measures used were the Preschool Language Scale-3 (Zimmerman, Steiner, & Pond, 1992), Clinical Evaluation of Language Fundamentals-Preschool (Wiig, Secord, & Semel, 1992), (MLU), and Bankson-Bernthal Test of Phonology (Bankson & Bernthal, 1990). Participants obtained scores within 1.5 SD on the Columbia Mental Maturity Scale (Burgemeister, Blum, & Lorge, 1972), a test of nonverbal cognitive functioning, and were judged to have normal oral-motor functioning. In addition to the phonology-followed-bymorphosyntax, and morphosyntax-followed-by-phonology strategies (in two 12-week blocks), this second study also included a weekly alternating goal intervention, and a simultaneous goal  27  attack intervention. Ten children were assigned to each of the block strategies, 9 to the simultaneous strategy, 11 to the alternating strategy, and 7 to a control group. Phonological performance improved equally using each of these four goal attack strategies, but the weekly alternating approach facilitated the greatest change in morphosyntactic development. Although research has not found clear phonological advantages for any one goal attack strategy, it remains possible that subtle differences exist. This possibility has been considered in the research design of the present study. Morphosyntax-Only Intervention The approaches and strategies of phonological treatment described thus far differ in many ways: focus, implementation, and goal attack strategy. However, all treatment approaches discussed so far are similar in that they target the phonological system directly. Alternatively, it is possible that phonological development may be stimulated through more generalized language intervention. There exists limited and somewhat conflicting evidence as to whether or not it is possible for the effects of intervention to generalize across domains in this way. Fey, Cleave, Ravida, and Long (1994) studied 26 children aged 3;8 to 5;10, all with language delays and phonological delays varying from moderate to profound. The authors found no gains in percentage consonants correct after five months of focused stimulation treatment targeting morphosyntax. The Tyler et al. (2002) study (described above) also addressed whether or not treatment of morphosyntax would generalize to phonology. That study included 20 preschool children whose scores fell at least 1 SD below the mean on standardized measures of morphosyntax and phonology. The authors found that after 12 weeks of therapy targeting only morphosyntax, phonology had improved significantly (as measured by single word elicitations). These conflicting results indicate that cross-domain effects can occur, but it is not clear what other factors are important for these effects. Also, unfortunately, neither of these studies examined cross-domain effects as a function of severity of the delay.  28  Another closely-linked line of research has investigated whether or not providing phonological treatment results in improvement to the morphosyntactic system. For example, Bopp (1995) studied nine preschool children, seven with severely protracted phonological and morphosyntactic development and two reference participants with severely protracted phonological development only. She found that 16-18 weeks of therapy resulted in significant gains in morphosyntax in the seven with combined difficulties. Although the relationship between phonological and morphosyntactic development has been demonstrated in some research findings, still little is known about the nature of this relationship and about how these two systems interact throughout development. Summary of Treatment Approaches The treatment approaches described in the sub-section above (traditional, cycles, minimal pair training, involvement of significant others, and perceptual training) have received some empirical support and are implemented variously in current phonological treatment practice. The present study has combined aspects of each of the five approaches described in its intervention program (see the Methods chapter). The diversity in treatment approaches available to practitioners can be overwhelming, but there is some limited evidence as to which approaches are most effective for which children. Tyler et al. (1987) proposed that a minimal pairs approach may be most appropriate for focusing on a single error pattern in children who use very few idiosyncratic patterns, whereas a modified cycles approach may be more useful in targeting multiple patterns in children who are more unintelligible. Overall, evidence that children require individualized management plans is increasing (Gierut, 1998). Also, using more than one treatment approach with a single child in intervention can be effective. For example, Almost and Rosenbaum (1998) used a modified cycles approach with minimal pair treatment in the early stages and a traditional approach in the later stages. Bernhardt (1990, 1992, 1994) utilized cycles, minimal pair contrasts (in perception  29  and production), metaphonological activities, and active parent involvement in a non-linear approach to intervention. The limited, conflicting, and inconclusive evidence with regard to cross-domain effects of treatment and goal attack strategy efficiency have highlighted the need for continued investigation of this topic, as is described in the present study. Before proceeding to the specific research questions for this study, the final section of the literature review outlines previous research on phonological (and language) development and intervention with twins, because the participants for the current study were a twin pair.  Phonological Development and Intervention with Twins The following discussion of twin speech and language development addresses three primary issues concerning language (and in particular, phonological) development in twins. First, it will consider how group twin studies have contributed to our understanding of the similarities between speech and language abilities in twins. Secondly, it will highlight the unique speech and language learning experience that growing up as a twin creates. Finally, two phonological intervention case studies of twin pairs will be discussed. Much of the research on speech and language in twins has aimed to understand to what extent heritability plays a role in language abilities (Kovas et al., 2005) and has used the ‘twin method.’ This method estimates the importance of genetic factors in the expression of a trait (in this case speech and language abilities). Comparisons are made for traits occurring in pairs of monozygotic (MZ) twins, who develop from one fertilized egg and are genetically identical, versus sets of dizygotic (DZ) twins, who develop from separately fertilized eggs and share 50% of their genetic information (Plomin, DeFries, McClearn, & McGuffin, 2008). Because sets of twins, for the most part, share both pre- and post-natal environmental factors, if MZ twins are more similar for a given trait than DZ twins, then genetics are likely to be important for that trait.  30  In recent research, studies have used correlations, concordance, and calculations of heritability to describe the contribution of genetics to phonological and language difficulties. Correlations (with value varying between -1.0 and +1.0) have been used to describe the direction and the degree of the relationship between speech and language scores of one twin and the scores of the other twin. Concordance is described by Plomin et al. (2008) as an index of risk. Each twin pair is reported as either concordant (same) or discordant (different) for the presence of a trait, with concordance expressed as a percentage. For example, if 10% of a twin population sample is concordant for a trait, then the probability is 10% that one twin will possess that trait if the other does. Heritability estimates the genetic effect size, or how much genetics contribute to a specific trait (Plomin et al., 2008). It can be calculated by doubling the difference between MZ and DZ twin correlations or concordances. Heritability values range from 0 to 1. Hayiou-Thomas, Kovas, Harlaar, Plomin, Bishop, and Dale (2006) studied the phonological and other language skills of a population-based sample of 281 MZ and 275 DZ four-and-a-half year-old twin pairs. The tests used were the McCarthy Scales of Children’s Abilities (McCarthy, 1972), The Renfrew Bus Story Test (Renfrew, 1997b), The Renfrew Action Picture Test (Renfrew, 1997a), the British Ability Scale (Elliott, Smith, & McCulloch, 1996) and The Children’s Test of Nonword Repetition (Gathercole, Willis, Baddeley, & Emslie, 1994). Phonological abilities were specifically tested using The Phonological Awareness Task (Bird, Bishop, & Freeman, 1995) and the Goldman-Fristoe Test of Articulation (Goldman & Fristoe, 1986). Hayiou-Thomas et al. (2006) estimated heritability at 0.56 for phonological skills and 0.34 for language skills (mid-range of levels of heritability). An older study by Matheny and Bruggemann (1973) evaluated the articulation abilities of 64 sets of MZ twins and 37 sets of DZ twins between the ages of 3 and 8 years, all of whom were typically developing. The standard score on the Templin-Darley Screening Test of  31  Articulation (Templin & Darley, 1960) was used to quantify phonological skill level. Correlations between MZ twin pairs ranged from 0.84 to 0.90 and correlations between DZ twin pairs ranged from 0.54 to 0.68. The fact that MZ twin correlations were higher than DZ twin correlations points to a hereditary component in the development of articulation. Based on this research, a contribution of genetics to phonological and language abilities has been established in population-based samples of twin children. The heritability of phonological and language skills has also been investigated in twins who are delayed in their development of phonology and/or language. Viding, Spinath, Price, Bishop, Dale, and Plomin (2004) compared the speech-language abilities of 310 pairs of control twins (some MZ and some DZ) and 393 pairs (some MZ and some DZ) of twins in which at least one twin fell below the 15th percentile on a language composite score. This composite was created based on scores from a battery of speech and language tests including the GoldmanFristoe Test of Articulation (Goldman & Fristoe, 1986), the Phonological Awareness Task (Bird, Bishop, & Freeman, 1995), The Children’s Test of Nonword Repetition (Gathercole et al., 1994), The Renfrew Language Scales - Bus Story Test (Renfrew 1997b) and Action Picture Test (Renfrew, 1997), the McCarthy Scales of Children’s Abilities-Verbal Index (McCarthy, 1972), and the British Ability Scales-Verbal Comprehension Scale (Elliot, Smith, & McCulloch, 1996). The estimates of heritability were .41 for children with composite scores below -1.5 SD, .50 for children with composite scores below -2.0 SD, and .72 for children with composite scores below -3.0 SD. This research suggests that speech-language impairment has a component of heritability and that this heritability may be greater for children with more severe impairments. However, the authors of this study did not explore any additional factors that could be contributing to greater monozygotic twin similarity for those that score farther from the mean, such as additional inter-twin influences on language development or unique environmental influences.  32  This line of research has begun to establish similarity in twins in terms of their speech and language abilities and is important for evaluating similarities in the participants of the current study. In addition to this, research using the twin method has led to important gains in our understanding of how genetic factors may contribute to phonological and language abilities. One possible limitation of the twin method is its assumption that environmental influences on MZ twin language development and DZ twin language development cancel each other out because they remain constant. However, it is possible that being a MZ twin results in somewhat different environmental influences than being a DZ twin. How twins differ and are similar to singletons in development because of their differences in genetics and environment is an ongoing question beyond the scope of the current study. The language learning experience of twin children is, however, clearly unique in many ways. Twins share many environmental factors including prenatal environment and as well as the home environment and constant peer interaction of someone the same age. Twins similarities in speech and language have been investigated in a phenomenon called ‘autonomous language,’ also termed ‘twin language,’ ‘idioglossia’ or ‘cryptophasia’. It refers to “a language that young children may come to speak that is different from language(s) used in their environment and incomprehensible to others” (Bakker, 1987, p. 234). Hay et al. (1987) studied 15 pairs of twins and, according to parent reports, seven out of nine male twin pairs and one out of six female twin pairs exhibited autonomous language. Bakker (1987) did a review of literature up to 1987 on this topic. The studies reviewed estimated that twin autonomous language occurred in 40 to 47% of twin pairs during early childhood and involved both invented grammar (language) and invented pronunciations (phonology). This autonomous language was reported to disappear between 4 and 6 years of age. Whether these aspects of twin language development were measurably different from singleton development or sibling pairs has not been thoroughly investigated, however. Although there is no evidence that autonomous  33  language is directly related to phonological and language delay, Bakker suggests that the simultaneous acquisition of speech and language that twins experience may cause reciprocal reinforcement of non-ambient language forms and lead them to be maintained. Thus, the solidification of non-target forms could delay the acquisition of target phonological and language forms. Mittler (1976) found that the presence of autonomous language in fact did not have a significant influence on scores on the Illinois Test of Psycholinguistic Abilities (McCarthy & Kirk, 1961). However, this has yet to be systematically investigated for phonological skills. Up to this point all of the twin studies discussed have investigated groups of twins. The present study is about phonological development in one set of twins. Two past studies that investigated the phonologies of single sets of twins are described below. The first is a study of phonological development in DZ twins (Edwards & Bernhardt, 1973) and the second is a nonlinear phonological intervention study with MZ twins (Von Bremen, 1990). Edwards and Bernhardt (1973) compared the phonological systems and processes of a set of DZ twins, a girl and a boy, aged 5;2. The phonological processes used by these twins were separated into two types, those affecting the structure of the word (i.e., deletions and insertions) and those affecting the individual consonants (i.e., feature changes such as frication and labialization). Remediation targeted both the structural and the segmental levels. Based on their analyses, Edwards and Bernhardt (1973) concluded that, although the twins had different substitutions for the same targets, they each had underlying difficulties with the same set of segments, and thus were more alike than different. This study predated OT, but provides evidence that the same markedness constraints can underlie output, in this case constraints disallowing fricatives and codas. Intervention outcomes progressed in the same direction, but more slowly for the male twin. This difference in outcomes may have been related to other within-child variables, including cognitive differences.  34  Von Bremen (1990) conducted a non-linear intervention study with a set of male MZ twins aged 5;7. These participants had very similar phonological systems, showing a need for phonological change in nearly identical areas. Like the twins studied by Edwards and Bernhardt (1973), Von Bremen’s participants showed a failure to match adult targets in both structural and segmental areas of the phonological system. Von Bremen’s study compared intervention that targeted structure first with intervention that targeted segments first. The twins were randomly assigned to these two conditions. Results showed that the twin whose treatment began with structural goals acquired not only these goals but also showed an increase in the number of matches to the adult target for segments. On the other hand, the twin whose treatment began with segmental goals did not show an increase in matches to the adult target for segments or structure until structure was targeted directly. In general, the results of this study provide general support for using a non-linear approach to intervention, and a possible treatment order benefit (i.e., structure before segments and features). The twin sets in these two studies both showed very similar phonological systems within the twin pair, even though there were some output differences. This exemplifies the findings of Matheny and Bruggemann (1973) who found that correlations between twin pairs in articulation were quite high, with estimates varying: from 0.84 to 0.90 for MZ twins and from 0.54 to 0.68 for DZ twins.  Summary of the Literature Review: Key Points This literature review has touched on several areas that have informed the methods and research questions of the current study. It has highlighted the underpinnings of constraint-based non-linear phonological theory in order to show how this theory informs goal selection. It has considered how interactive theories of language processing can explain cross-domain treatment effects and how within-child variables (capability and focus) can explain treatment outcomes. It has also reviewed the background on several treatment approaches, all of which have been  35  incorporated into the phonological treatment of this study. This literature review has also presented current research on goal selection and goal attack, two areas that are directly addressed by the research questions presented below. Finally, it has considered similarities in twin speech and language development as well as how capability and focus of twins as individuals may be related to intervention outcomes.  Research Questions and Hypotheses There remain many unanswered questions concerning outcomes of phonological intervention for children who have both phonological and morphosyntactic delays. These include questions concerning the relevance of severity of impairment (including both phonology and other aspects of development) for approaches and outcomes, types of treatment targets, and both general and specific treatment outcomes concerning the phonological system. Two major aspects concern efficiency, i.e.: (a) which exact aspects of treatment approaches may promote the fastest change, and (b) how, when, and to what extent do cross-domains effects of treatment occur, both within the phonological system (structure versus segments) and between other linguistic domains. At least some of the ambivalent results reflect the small n in the various studies, and the heterogeneous phonological systems of the children with protracted morphosyntactic development. One way to enhance the homogeneity of phonological systems in a study is to have twins as participants. Certainly, monozygotic twins are more likely to show greater correlations in their phonology, but dizygotic twins (the participants for the current study) are also likely to have similar phonological systems; DZ twins share 50% of their genetic information, and many environmental factors. The following specific questions were addressed in response to the questions raised by the literature on phonological intervention:  36  Question 1. Effects of phonological intervention Does phonological treatment result in specific changes to either twin’s phonological system and, if so, what type of changes?  Research hypotheses concerning phonological system effects 1. It was expected that, for both participants, change to the phonological system would be seen during the phonology treatment block because of the specificity of treatment to the individual needs of the child. This result would be consistent with the findings of Almost and Rosenbaum (1998), Law et al. (2004), Tyler et al. (2002, 2003), and Von Bremen (1990). 2. It was also expected that more change would be seen in treated elements of the phonological system compared with untreated elements of the phonological system for both structure and features. This would support the findings of Von Bremen (1990) whose participants showed minimal development of unrelated segments after intervention. 3. It was expected that generalization would occur from treated contexts to untreated contexts and from treated structures and features to untreated structures and features as has been reported in past literature (Bankson, Bernthal, & Flipsen, 2009). Also, generalization was expected from the structural domain to the segmental domain as seen in Von Bremen (1990). As will become clearer in Chapter 2, the present study focused more on structure in the intervention (i.e., two structural goals and one segmental goal with a structural component), making generalization from the structural domain to the segmental domain possible to evaluate. Further expansion of this hypothesis is given at the end of the methods section, after goal selection is discussed.  37  Question 2: Phonological effects after morphosyntactic intervention Does morphosyntax treatment result in specific changes to either twin’s phonological system and, if so, what are these changes?  Research hypotheses concerning effects of morphosyntactic intervention 1. Although research findings have conflicted, it was expected that change to the phonological systems of both twins would result from a morphosyntax-only treatment, as seen in Tyler et al. (2002). This hypothesis was based on the connections between different linguistic domains predicted by interactive psycholinguistic models 2. However, in comparison to the phonology-only treatment block, less change was expected following morphosyntactic intervention.  Question 3: Order of interventions Are morphosyntax-first and phonology-first intervention strategies effective in facilitating phonological development?  Research hypothesis concerning order of intervention effects 1. It was expected that both morphosyntax-first and phonology-first intervention strategies would facilitate phonological development as reported in Tyler et al. (2002, 2003).  Question 4: Twin similarities and differences. What are the similarities and differences between the two participants and how do they influence intervention outcomes?  Research hypothesis concerning individual twin differences 1. It was expected that the phonologies of the twins would be similar as seen with the dizygotic twins studied by Edwards and Bernhardt (1973). Twin correlations for articulation were expected to be similar to those found by Matheny and Bruggemann (1973) for DZ twins (i.e., from 0.54 to 0.68).  38  2. It was expected that higher measures of each twin’s capability and focus based on the framework of Kwiatkowski and Shriberg (1998) would correlate with more desirable intervention outcomes.  39  Chapter 2 - Methods This chapter includes five sections. The first gives information about the participants. The second provides an overview of the study’s design. The third describes the assessments that were carried out throughout the course of the study. This third section also includes the results for the non-phonological tasks of the baseline (second) assessment, because they give information that is useful for understanding the participants’ individual profiles and for goal planning. The fourth section provides information about the treatment, including goal selection, treatment approaches, and a detailed analysis of the phonological assessment at baseline (i.e., just prior to intervention). The final section describes how the results were compiled and analyzed.  Participants The two participants in this study are a pair of monolingual English-speaking male dizygotic twins. They were aged 6;7 when the study’s therapy began. The twins were first identified as having both a phonological and a morphosyntactic delay upon referral to their community health unit at the age of 3;11. They received direct and indirect speech and language intervention and support through community programs throughout the remainder of their preschool years. When the twins entered kindergarten at the age of 5;7, they began to receive speech and language therapy at school. Both boys were reported to have made slow progress throughout their preschool and kindergarten years and continued to be delayed in phonology and morphosyntax when this study began. Information about the twins’ birth, development, and family history was obtained through a parent questionnaire. The twins were delivered through a planned cesarean section. Twin 1 (T1) was born first and Twin 2 (T2) was born second. The presence of borderline gestational diabetes was the only remarkable complication related to the pregnancy and delivery  40  of the participant children. Motor milestones in the first year of life were reportedly normal for both children. Language milestones were similar and significantly delayed for both boys: cooing at 2-3 months, first words after age 2, and first word combinations at age 4 years. Both boys have had good overall health, hearing, and eyesight. The twins have two sisters. Their older sister (10 years old) has experienced learning difficulties in mathematics and their younger sister (4 years old) was receiving speech and language therapy at the time of the study. The severity of her delay is not known precisely, but the parents reported that her difficulties were less severe than those of her brothers. The twins’ father has a history of stuttering, whereas the mother reported no language or learning difficulties. Both parents have completed a university Bachelor degree. Although these two boys are similar in many ways, their strengths and interests differ. According to the twins’ parents: [T1] is very observant of spiders and insects and knows a lot about them. He loves animals [and] has great patience, skill, and endurance when collecting spiders. He loves to know how things work and is good at fixing things and using tools. He is imaginative when playing with his siblings. He is sensitive to how others feel. He is brave and skilled on the monkey bars and in the water (loves to go swimming)! He enjoys being helpful and cleaning [and] enjoys hands on activities like colouring, playdough, hammering, and cutting. [He] enjoys learning science facts, about flags, [and] other countries. He likes planting in a garden, being playful, and singing too. [T2] is very gentle and affectionate. He usually always tries his best, perseveres, and produces neat and careful work. He likes to build with Lego. He is very imaginative when playing with his siblings, especially his brother. He likes hands-on crafts like building boats and painting. [He] enjoys playing games like ‘go fish’, tag, [and] board games. [He] enjoys collecting things, from skeleton toys,  41  growing toys, and nutcrackers to dragon toys. He likes to help out around the house too (like cleaning, cooking). In addition to the differences pointed out by the parents, the results of the assessments conducted as part of the current study suggested not only similarities, but also some differences in the phonological and morphosyntactic skills of the two boys (described later in the chapter). The differences described by the parents and in the assessment results were predicted to have a potential impact on rate of change in therapy. Each twin’s capability in phonology and focus in interactions with the clinician was analyzed using the capability-focus framework (Kwiatkowski & Shriberg, 1998). This framework allowed some of the intrinsic characteristics of each twin to be measured and analyzed in relation to intervention outcomes. The procedure used for assessing capability and focus is explained further at the end of this chapter.  Overview of Study Design The current thesis study was designed to address potential effects of both phonological and morphosyntactic intervention on the phonological development of a pair of DZ school-aged twins (as outlined at the end of Chapter 1). The study had a multiple baseline single subject design and took place over 11 months. The timeline included a four-month baseline period (2 months of therapy by the school-based speech-language pathologist and a 2-month no-treatment phase)4, an 8-week treatment block, followed by a 4.5-week break, and another 8-week block followed by a 3.5-week break5 (see Table 7).  4  Both twins participated in two sessions of a language group during the no-treatment period.  5  The length of breaks varied from 3.5 weeks to 4.5 weeks due to scheduling constraints.  42  Table 7. Assessment and Treatment Schedule Time  Assessment  April 2009  Assessment 1  4 months  Baseline period  Sept 2009  Twin 1  Twin 2  2 months of weekly  2 months of weekly  months of no treatment  months of no treatment  Morphosyntax treatment  Phonology treatment  Phonology treatment  Morphosyntax treatment  intervention and 2  intervention and 2  Assessment 2 Treatment block 1  3.5 months  (8 weekly sessions and a 4.5 week break)  January 2010  Assessment 3 Treatment block 2  3.5 months  (8 weekly sessions and  April 2010  Assessment 4  a 3.5 week break)  Although Tyler et al. (2003) showed that a treatment strategy where morphosyntax and phonology were alternated each week may result in better outcomes for morphosyntax, they found no obvious advantage for phonological development in alternating versus blocked treatment. Thus, for the current study, single-domain treatment blocks were used. The 8-week single-domain blocks also (a) allowed follow-up activities to be simply and clearly described to the school personnel and family, (b) took into account the twins’ previous slow progress, and (c) allowed for a concentration on one aspect of language development with minimal interference from other domains. Twins were assigned randomly to a treatment order: T1 received a block of morphosyntax treatment followed by a block of phonology treatment, whereas T2 received a block of phonology treatment followed by a block of morphosyntax treatment. This design allowed for a measurement of change in matches to the adult target after each individual treatment block and thus addressed the research questions of the study. Change in matches after phonology treatment were used in evaluating whether or not phonological  43  intervention led to phonological change (research question 1) and change in matches after morphosyntax treatment were used in evaluating whether or not morphosyntax intervention led to phonological change (research question 2). This design also allowed phonological changes across the whole treatment period (block 1 and 2) to be evaluated, and thus, the effectiveness of each goal attack strategy in facilitating phonological development.  Assessments Four assessments were completed for each participant: (1) the pre-baseline assessment completed four months prior to the onset of therapy; (2) the baseline assessment prior to the onset of treatment block 1; (3) the post-block 1 assessment between treatment block 1 and block 2; and (4) the post-block 2 assessment at the end of treatment block 2. The components of each assessment can be seen below. Results from the pre-baseline assessment for the phonological tasks and from the baseline assessment for phonological and non-phonological tasks are presented below separately. See the Results chapter for reporting of the remaining assessments. The majority of assessments were completed at the University of British Columbia (UBC). All assessment tasks except probes were completed outside of the treatment environment by the same person who conducted the treatment. All sessions were both audioand video-recorded. Digital audio recordings of assessment sessions were made with a Marantz Professional Handheld Solid State Recorder PMD661 and an Audio Technica ProSeries Pro 44 Cardioid Condenser Boundary Microphone. Video recordings were made with a Sony Digital Video Camera Recorder DCR-TRV39 and Sennheiser EW100 wireless microphones.  Assessment 1: Pre-Baseline (four months prior to treatment) The pre-baseline assessment provided basic information about the participants and helped to begin building a relationship with the twins. It also served as an early baseline measure of the production skills of the participants. The following tasks were completed:  44  1. Computerized Articulation and Phonology Evaluation System (CAPES, Masterson & Bernhardt, 2001): This tool was chosen because it allows for evaluation of the multiple levels of the phonological hierarchy. Words were elicited using pictures, sentence completion, and imitation when necessary. A sentence completion elicitation example for /g!m/ is: “he stepped on…”. The children’s productions for the CAPES Profile and Level 2 lists (86 single words) were audio-recorded (see Appendix D). This pre-baseline CAPES assessment served to document the phonological systems of the twins before the four-month baseline period. These results could then be compared to those from the CAPES at Baseline in order to determine which aspects of phonological development were occurring with minimal intervention. This development was an important consideration in the selection of targets for intervention. 2. Language Sampling: A play-based language sample was collected during a 25-minute interaction between the two participants and two student clinicians.  Assessment 2: Baseline (prior to treatment block 1) The baseline assessment was completed immediately prior to the onset of therapy. The objectives were two-fold: (1) to obtain norm-referenced data to compare the twins’ performance with peers and within-child across domains or measures, and (2) to collect detailed descriptive data regarding phonology and morphosyntax for goal-setting and as a basis to measure change at later time points. This was thus the most in-depth of the four assessments. It included measures of phonological production, speech processing, language production, language comprehension, the oral speech mechanism, nonverbal intelligence, and verbal and nonverbal memory as follows: 1. CAPES: Profile list and List 2, as described above. 2. Phonology probes: An 18-word probe list was created in order to elicit additional tokens of phonological elements to be treated. This probe was completed in the last phase of the  45  baseline assessment, after the CAPES data were analysed and goals selected. Productions of these words were elicited using photos and the same type of verbal cues used in the administration of CAPES (sentence completion or imitation). In addition, eight sets of minimal pairs were chosen in order to test each child’s ability to discriminate between target-matching and target-non-matching adult productions, an important basic skill for learning production according to the findings of Rvachew (1994). See Appendix D for the probe and minimal pair word lists. The minimal pair distinctions corresponded with phonological mismatches in the twins’ productions that were to be targeted in therapy. 3. Oral Speech Mechanism Screening Examination (OSMSE, St. Louis & Ruscello, 2000): This screening tool documents the state of oral structures, function, and diadochokinesis. Diadochokinesis is the rapid alternating movement of the speech articulators while articulating /p!t!k!/ and evaluates motor skill, coordination, and neuromotor integration (Duffy, 1995). The OSMSE was used in order to measure whether or not the twins’ oral structures and functions were typical or atypical. 4. Syllable Repetition Task (SRT, Shriberg & Lohmeier, 2008): This test requires the participant to repeat 18 non-word items with Consonant-Vowel (CV) syllable structure presented auditorally from an audio-recording. The SRT was administered because it purports to investigate speech processing abilities, and has been established as an accurate identifier of language impairment (Shriberg, Lohmeier, Campbell, Dollaghan, Green, & Moore, 2009). Also, it uses only early developing consonants, making it appropriate for use with children whose phonological repertoire is limited. The items require the use of both verbal memory and auditory perceptual skills. Two syllable (i.e., /b"d"/), three syllable (i.e., /d"b"m"/), and four syllable (i.e., /b"m"d"n"/) items were administered.  46  5. Clinical Evaluation of Language Fundamentals-4 (CELF-4, Semel, Wiig, & Secord, 2003): Five subtests of the CELF were selected based on the twins’ known or suspected areas of difficulty, in order to provide a norm-referenced language measure.  Word Structure – The participant completes sentences targeting specific morphological structures.  Recalling Sentences – The participant imitates sentences that are presented verbally by the examiner. This task draws on both morphogyntactic knowledge and verbal memory.  Formulated Sentences – The participant is asked to formulate a sentence that includes a target word based on a visual stimulus.  Sentence Structure – The participant is asked to point to the picture that illustrates a sentence read by the examiner.  Number Repetition Forward – The participant repeats number sequences presented verbally by the examiner. This subtest places demands on attention and on auditory/verbal memory. Performance on all three tasks that tap into verbal memory (i.e., Syllable Repetition, Recalling Sentences, and Number Repetition) can be contrasted against performance on a visual memory task (see below). Verbal memory is thought to be important for speech processing and thus oral language learning (Shriberg & Lohmeier, 2008), while visual memory is not. 6. Language Sampling: This sampling included two tasks: a play task and a story production task. These data served as the basis for determining the intervention goals for morphology and syntax.  Play Task: A language sample was collected during a 10-minute play session with each parent, and a 5-minute play session with one experimenter and each child.  Story Production Task: A narrative sample was elicited using the wordless picture book “Frog where are you?” (Mayer, 1969). The participants were first encouraged to look through the pictures of the book and asked to produce a story to a naïve listener. The book was then sent  47  home with the family along with a written script (SALT Reference Database, n.d.) and the mother was asked to tell the story once following this written script. The next day each child was asked to tell the story to a new listener. The Systematic Analysis of Language Transcripts program (SALT, Miller & Iglesias, 2006) was used to analyze the narrative language transcript. 7. Morphosyntax probes: Morphosyntactic probe activities were designed to elicit the production of specific forms in order to confirm and refine the intervention targets in this domain. These forms were elicited in a game format where the clinician modeled the target structure multiple times and the child was given opportunities to complete sentences with a parallel structure. 8. Test of Non-verbal Intelligence (TONI-3, Brown, Sherbenou, & Johnsen, 1998): The TONI-3 is meant to be a language-free measure of cognitive ability. It is recommended for assessing aptitude in individuals whose linguistic skills might affect performance on other intelligence testing tools. The test evaluates problem solving, a primary cognitive ability that is considered to represent the construct termed “non-verbal intelligence.” The directions to the test are pantomimed and the test items are language-free, using only abstract visual figures. The individual must identify the relationships among figures and select the correct response from an array of alternatives. 9. Illinois Test of Psycholinguistic Abilities (ITPA, McCarthy & Kirk, 1961):  Visual and Sequential Memory Task – This task measures visual memory span. It assesses the participant’s ability to reproduce sequences of non-meaningful figures (lacking semantic information) and purports to measure abilities that are independent of verbal skills. 10. Parent Questionnaire: Information about each child’s birth, development, and family history was obtained through a questionnaire that was sent home with the parents and returned  48  by mail. The areas evaluated included pregnancy, child’s health, early development, later development, and family history. See Appendix B for a copy of this questionnaire.  Assessment 3: Post-Block 1 (between treatment block 1 and block 2) The third assessment served as a measure of progress after the first block of treatment. The areas assessed included phonological and morphosyntactic production. As per the baseline assessment, the following tasks were administered: CAPES, phonology probes, language sampling (play task and story production task, “Frog on his own,” Mayer, 1973), and morphosyntax probes.  Assessment 4: Post-Block 2 (following treatment block 2) The fourth assessment served as a measure of progress after the second block of treatment. The areas assessed included phonological and language production, speech processing, nonverbal intelligence, and qualitative measures of improvement. Both speech processing and non-verbal intelligence were re-assessed in order to verify whether scores obtained at the second assessment point were reliable. In particular, after having spent significant time with the participants, the experimenters believed that scores derived at that second assessment may not have been representative of the participants’ non-verbal skills.6 The following tasks were administered: CAPES, phonology probes, language sampling (play task and story production task, “Frog where are you?”, Mayer, 1969), morphosyntax probes, SRT (three and four syllable items), and TONI-3.  6  Although both participants complied with testing procedure, the results were judged not to be  a reliable representation of the participants’ best effort/ability. The participants were fatigued because this assessment followed several others, and thus their levels of attention were questionable. This test was re-administered in at the final assessment (post-block 2).  49  Outcomes Survey: The final assessment procedure was a survey for the classroom teachers and classroom assistants evaluating progress qualitatively. They were asked to comment on changes in speech, language, and communication in each participant over the course of the study giving examples. (See Appendix A for survey wording.)  Pre-Treatment Assessment Results Scores for the non-phonological tasks of the baseline assessment are reported in Table 8. Both twins showed a discrepancy between skills related to verbal tasks and those related to non-verbal tasks. While scores on verbal tasks, for the most part, fell more than two standard deviations below the mean (2nd percentile), scores on non-verbal tasks were within the normal range. This same pattern was observed when the speech/language processing tasks with a verbal memory component (SRT, Recalling Sentences, Number Repetition) were contrasted with performance on the Visual and Sequential Memory Task of the ITPA. In general, these findings indicate that the phonological and morphosyntactic skills of these two participants were weaker than at least some cognitive abilities (i.e., those tested by the TONI-3 and visual memory subtest of the ITPA). Thus, it is not likely that that the participants’ speech and language difficulties were attributable to a more general cognitive delay. A few other relative strengths that were indicated by these results are structure and function of the oral mechanism in both twins (based on the OSMSE) and speech processing at the two-syllable level (based on the SRT). The results of diadochokinesis (OSMSE) and the speech processing task results for multisyllabic nonsense words, indicate that both participants were delayed in their motor coordination of speech-like oral productions.  50  Table 8. Baseline Assessment Results for Non-Phonological Tasks Area Evaluated  Test  T1 Results  T2 Results  Oral Motor Skills  OSMSE  - Structure:  Pass  Pass  - Function:  Pass  Pass  - 2 Syllable:  4th %ile  71st %ile  - 4 Syllable:  <1st %ile  <1st %ile  - Diadochokinesis: Speech Processing  Did not pass  Did not pass  SRT - 3 Syllable:  <1st %ile  <1st %ile  Language production,  CELF-4  - Word Structure:  0.1%ile  0.4th %ile  processing  - Recalling Sentences:  Incomplete  0.1 %ile  - Sentence Structure:  0.1%ile  5th %ile  - Number Repetition-Forward:  0.1 %ile  1st %ile  - Mean Length of Utterance in words:  1.90  3.91  - Number of Different Words:  31  88  comprehension, and  - Formulated Sentences:  Language Sampling  0.1%ile  0.1%ile  a  Visual and Sequential  ITPA  - Visual and Sequential Memory Task:  15th %ile  14th %ile  Non-Verbal Intelligence  TONI-3 b  13th %ile  16th %ile  Memory a  Mean values (and SD) were taken from a criterion-referenced sample of 65 children from  Wisconsin and California within 6-months of the twins’ age who were asked to retell the “Frog Where Are You?” (Mayer, 1969) story after having heard the same script as used in the current study: 7.53 (SD 1.08) for MLU in words and 99.8 (SD 19.0) for number of different words (based on the SALT Retell Database, Miller & Iglesias, 2006). b  Upon re-administration of the TONI-3 at the post-block 2 assessment (7 months later), T1  scored at the 50th percentile and T2 scored at the 30th percentile. This placed both participants in the average range for this measure.  51  As previously mentioned, the results of the CAPES pre-baseline and baseline assessments were used in order to select targets for intervention. For T2, the pre-phonology treatment period was the period between the pre-baseline and baseline assessments. For T1 the pre-phonology treatment period was longer. It included the time between pre-baseline and postblock 1 assessments. A detailed description of the acquired and unacquired elements for both twins can be seen in Appendix E and F. These appendices summarize each twin’s acquired and unacquired structures and acquired and unacquired features. The specific phonological changes seen over this baseline period are described in the section below on goal selection. Table 9 shows the potential intervention targets (i.e., the unacquired phonological elements) as noted at the pre-baseline assessment. At the time of the first assessment several features remained unacquired. A 75% match was used as the criterion for acquisition, as has been done in past research (Olswang & Bain, 1985). For T1, potential feature/segmental targets were [-grooved] interdentals /!/ and /"/, [-cont]-[+cont] affricates /t#/ and /d$/, [CorLab] /%/ and [+lateral] /l/; potential structural targets were trisyllable words, iambic stress, consonant sequences, and word shapes containing specific consonant sequences. For T2 potential feature/segmental targets were [-grooved], [-anterior] (/#/, /t#/, /d$/), [-cont]-[+cont], [CorLab] and [+lateral]; potential structural targets were trisyllable words, iambic stress, consonant sequences, and word shapes containing specific consonant sequences. Table 10 shows all of the potential intervention targets (the unacquired phonological elements) at the next assessment, the baseline assessment. For both twins at the baseline assessment, potential targets remained essentially the same (but see discussion below for details): [-grooved], [-anterior], [-cont]-[+cont], [CorLab], [+lateral], trisyllabic words, iambic stress, consonant sequences, and word shapes containing sequences.  52  Table 9. Pre-Baseline Assessment: Summary of Unacquired Structures and Features Features  T1 Potential Targets  T1 and T2 Potential Targets  1. /!/, /"/  1. /!/, /"/  2. /#/, /$/  2. /#/, /$/, /&/  3. /%/  3. Syllable-initial /%/  4. Word-medial and word-final /l/  4. Word-medial and word-final /l/  Trisyllabic words  Trisyllabic words  Iambic stress pattern: wS  Iambic stress patterns: wSw, wS  related word shapes  related word shapes  [-grooved] [-cont]-[+cont]  [CorLab] [+lateral]  Structure  Most consonant sequences and  [-grooved] [-cont]-[+cont] and [-anterior] [CorLab]  [+lateral]  Most consonant sequences and  Table 10. Baseline Assessment: Summary of Unacquired Structures and Features Features  T1 Potential Targets  T2 Potential Targets  1. /!/, /"/  1. /!/, /"/  2. /#/, /$/, /&/  2. /#/, /$/, /&/  3. Syllable-initial /%/  3. Syllable-initial /%/  4. Word-medial and word-final /l/  4. Word-medial and word-final /l/  Three syllable words  Three syllable words  Iambic stress pattern: wSw, wS  Iambic stress patterns: wSw, wS  related word shapes  related word shapes  [-grooved] [-cont]-[+cont] and [-anterior]  [CorLab]  [+lateral]  Structure  Most consonant sequences and  [-grooved] [-cont]-[+cont] and [-anterior] [CorLab]  [+lateral]  Most consonant sequences and  The changes for matches to structure and features that occurred prior to phonology treatment included both increases and decreases in matches, some of which affected goal selection. For T2, needs at the baseline assessment were identical to pre-baseline needs. For T1,  53  there was some change in needs from the pre-baseline assessment to the baseline assessment. Specifically word-final /!/ appeared to no longer be a need after the baseline. The phoneme /"/ and the stress pattern wSw surfaced as new needs after the baseline. The rest of the needs for T1 remained the same after the baseline period. The system of phonological features that each twin demonstrated on CAPES immediately preceding the study’s first treatment block can be portrayed in a geometrical matrix (see Figure 3). This matrix is based on the feature framework used by Bernhardt and Stemberger (1998).  Figure 3. Feature Matrix: Summary of Needs at Baseline Assessment for T1  Note. Adapted based on Bernhardt and Stemberger (1998)  54  Figure 4. Feature Matrix: Summary of Needs at Baseline Assessment for T2  Note. Adapted based on Bernhardt and Stemberger (1998)  A calculation of global measures (percentage consonant match, percentage word shape match, and percentage vowel match) was also made based on the CAPES word elicitations. Percentages for both the pre-baseline and baseline assessments are presented in Table 11. The probe words elicited at the baseline assessment have not been included in these calculations so that calculations at both time points are based on the same set of words. These global measures show that some natural change was occurring in the phonological systems of the twins during their baseline periods, particularly for T1.  55  Table 11. Pre-treatment Global Measures T1 Consonant Match  Pre-baseline  Baseline  63.6%  70.3%  (147/231)  (163/232)  70.9%  79.1%  (61/86)  (68/86)  76.0%  92.6%  Word Shape Match Vowel Match T2 Consonant Match  (95/125)  (112/121)  Pre-baseline  Baseline  71.1%  63.2%  (160/225)  (144/228)  82.1%  81.3%  (69/84)  (70/86)  79.3%  80.3%  Word Shape Match Vowel Match  (92/116)  (98/116)  Treatment Strategies and Goals The study’s treatment consisted of one 45-50 minute session each week, a set of activities to be completed by the parents five to six times per week (minimum 15 minutes), and a set of activities to be completed by the classroom assistant four times per week (minimum 15 mininutes). The school-based speech-language pathologist also continued to see each twin for one session per week and offered each twin treatment within the same domain (phonology or morphosyntax) as the study.  Morphosyntax Goals and Intervention Approach For the morphosyntactic intervention block, goals were selected for each child individually on the basis of language sample analysis and information gleaned from the language probe tasks. Consideration was also given to the sequence of grammatical morpheme acquisition as outlined by Brown (1973) and to which language forms could potentially have the most impact on the child’s social interactions and further language development. The specific  56  goals targeted were hafta, can’t, and the Subject-Verb-Object structure for T1, and the copula is,  don’t, and coordination using because for T2. A horizontal approach was used to target the morphology and sentence structure goals (i.e., all three goals were targeted during every session). Treatment activities included books, games, and crafts. Special attention was given to creating activities that were interesting and fun, and that elicited communicative use of the target form in order to foster generalization (Fey, 1986). Recasting and focused stimulation were the primary treatment strategies implemented throughout all treatment sessions, as this has been shown to facilitate learning of grammatical targets (Fey, Cleave, Long & Hughes, 1993). Elicited production was also used in the form of cloze tasks (e.g. “Birds can fly but dogs ____”).  Goal Selection for Phonology Three phonology goals were selected for the 8 weeks of phonology intervention using a non-linear phonological analysis of the CAPES (2001) Profile List and List 2 single word elicitations from the pre-baseline and baseline assessments. The similarities in the two boys’ productions allowed the same phonology goals to be chosen for each twin (see Tables 8, 9; Figures 3, 4). Spontaneous change between pre-baseline and baseline, or lack thereof, was taken into consideration as well. As recommended by Bernhardt and Stemberger (2000), four goal types were considered, (1) new structure, (2) new features, (3) acquired structure in new places or in new combinations, and (4) new combinations of independently acquired features. The twins showed greater need for structure development overall. Two structural goals were chosen for intervention. Goal 1: Syllable and Word Structure Iambic stress patterns: wSw and wS and the pattern Sww. Stress patterns were chosen as targets to facilitate production of unstressed syllables in words and, in the case of morphosyntax, unstressed grammatical morphemes (see Sadler, 2010). For T1, 4/5 iambic stress  57  patterns were matched in the pre-baseline assessment. The one mismatch for T1 was a strengthening of the weak syllable in /k!"nu/, producing ["kæ"nu]. However, three of these four matches were elicited though imitation. At the next assessment, the baseline assessment, only 1/5 iambic stress patterns was matched and this one was again elicited through imitation. Mismatches at this point were weak initial syllable deletions (i.e., ["g#n] for /!"g#n/). It is difficult to say whether a decrease in matches after the baseline period was related to the imitation cues used or to a regression in this particular area of T1’s phonology. However, after another three and a half month period (the morphosyntactic treatment period), T1 matched only 1/5 iambic stress patterns, again elicited in imitation. The lack of change suggested that this was a viable structural target. At the pre-baseline assessment, T2 matched 2/5 iambic stress patterns using weak syllable deletion in the mismatches, and at the baseline assessment, 3/5 words. The minimal change also suggested that this would be a viable target for T2. The stress pattern Sww was chosen for intervention in addition to iambic patterns because reduction of three syllable words with Sww stress was observed in connected speech during the baseline phonology probe sessions the word ‘Canada’ (/"kæ$n!d!/) was produced as ["k%d!] by T1 and ["kæ$nd!] by T2. Goal 2: Position or Sequence Word-initial /s/-clusters: /st/, /sp/, /sn/. Another structural target was /s/-clusters. Neither participant matched any word-initial /s/-clusters in either the pre-baseline or baseline assessments. In the pre-baseline assessment, T1 matched the cluster /sp/ in word-medial position for /"w&sp'/ and T2 matched the same cluster in the baseline assessment. Thus, up to the onset of therapy, there was very little evidence of development of /s/-clusters and none word-initially. After his morphosyntax treatment period (and prior to phonological treatment), T1 matched one word-initial /s/-cluster ([skits] for /ski/), one word-medial /s/-cluster in /"w&sp'/, and one word-  58  final cluster in /v!st/. These matches indicate that some development of /s/-clusters was beginning to occur for T1 prior to phonology treatment. In choosing /s/-clusters as a goal, the findings of Bleile (1995) were taken into consideration. According to Bleile, the two earliest developing cluster types are [-continuant][-consonantal] (/tw/, /kw/) and /s/-clusters (/sp/, /st/, /sk/, /sm/, /sn/, /sw/). Both participants had acquired clusters of the [-continuant]-[-consonantal] type, and therefore /s/-clusters were chosen for intervention. Only three of these six /s/-clusters were chosen in order to observe whether generalization to untreated /s/-clusters might take place. It was decided to pair the /s/ with one nasal (/sn/) and two stops (/st/, /sp/), giving some variation in cluster types. The cluster /sw/ was not chosen due to its similarity to already acquired segments (i.e., /gw/, /kw/, etc.). Goal 3: Segments/Features or Feature Combinations Feature combination [CorLab] (syllable-initial /"/) and [+lateral]. A two-part featural goal involving /"/ and /l/ was also chosen for intervention. In deciding on this goal, all unacquired features and segments were first considered as potential intervention targets. Decisions were made based on frequency of occurrence of segments and stimulability. Furthermore, the unacquired features and segments involved mostly consonants that occur infrequently in English (/#/, /$/, /%/, /&/, and /'/, Shriberg & Kwiatkowski, 1982) and that are late-developing (/l/, /"/, /$/, /%/, /'/, and /(/, Bleile, 1995). In keeping with the Bernhardt and Stemberger (2000) approach, items that were more likely to increase intelligibility (higher frequency) were considered important targets, and thus the more frequently occurring /l/ and /"/ were chosen. Also, while the participants showed no matches for /$/, /%/, /'/, /&/, and /(/ in the baseline assessment, and only T1 showed matches for /$/, /%/, /'/ in the pre-baseline assessment, /l/ and /"/ showed some matches for both twins in both the pre-baseline and baseline assessments, suggesting that these might be more responsive to treatment. Some researchers suggest that selecting unstimulable targets for intervention is the best approach (Gierut,  59  Morrisette, Hughes, & Rowland, 1996), while other suggest selecting stimulable targets (Hodson & Paden, 1991; Powell, Elbert, & Dinnsen, 1991). With the history of slow progress in therapy, it was decided that choosing stimulable targets would be a useful approach for these participants. Because /l/ and /!/ became treatment targets, more detail is provided here on their production. For T1, the feature [+lateral] showed no change prior to treatment. The feature [+lateral] was matched on 2/2 word-initial tokens and 1/2 word-medial tokens in the prebaseline, baseline, and mid point assessments, but was not matched word finally. Similarly, the only change in [+lateral] seen for T2 throughout the baseline period was an increase of one word-medial match. The feature [+lateral] was matched on 2/2 tokens word-initially at both the pre-baseline and baseline assessments, 1/2 tokens word medially in the pre-baseline assessment and 2/2 tokens word medially in the baseline assessment, but was not matched word finally. In terms of /!/ [CorLab] for T1, it was matched in neither word-initial nor word-medial positions across pre-treatment assessments. Word-final matches for /!/ increased from 2/7 at the pre-baseline assessment point to 6/7 at both the baseline and the midpoint assessments. For T2, /!/ was matched in 1/2 tokens word-initially and 1/2 tokens word medially in both pre-treatment assessments. This indicated that no change in the feature combination [CorLab] occurred for T2 over the baseline period. Word-final /!/ was matched on 7/7 tokens at both assessments, meaning that it had been acquired prior to the phonology treatment block. Pre-baseline and baseline data for items containing /!/, /l/ and /w/ indicated that both twins were experiencing difficulty with feature co-occurrence among the approximants (see the Introduction). In order to understand the pattern of substitutions that the twins exhibited, place of articulation of the approximants must be considered in addition to the manner feature [+lateral]. The places of articulation of the four English approximants /!/, /l/, /w/, and /j/, are presented in Table 12 below.  60  Table 12. Place of Articulation (POA) of English Approximants  a  Phoneme  Syllable Initial POA  Syllable Final POA  /!/  [Coronal][Labial] ([Dorsal])  [Coronal] ([Dorsal])  /l/  [Coronal]  [Coronal][Dorsal]a  /w/  [Labial][Dorsal]  ---  /j/  [Coronal][Dorsal]  ---  a  /l/ is velarized in the context of back vowels across word positions.  The dominant mismatch patterns for the approximants were: • Word-initial /!/ ! /w/. This patterns indicated a lack of the feature [Coronal], likely due to difficulty with the co-articulation of [Coronal] and [Labial]. • Word-initial /w/ ! /l/7. This pattern indicated a change of the features [Labial][Dorsal] to [Coronal]. • Word-final /l/ ! /r/. This indicated the appropriate place features [Coronal] and [Dorsal] but a lack of the feature [+lateral]. Some substitution patterns observed less frequently were (1) vowel substitutions (maintaining the [Dorsal] component), (2) deletion of the liquids syllable finally and word medially, and (3) [j] as a substitution (maintaining [Coronal] and [Dorsal], but not [+lateral] or [Labial]). Specific mismatch patterns for approximants are presented in Table 13 below.  7  This substitution and possible coronal assimilation pattern was observed in connected speech.  The word /"w#$%/ was said as [l#$%] and may reflect a cross-vowel sequence constraint NoSequence(Labial…Coronal).  61  Table 13. Mismatch Patterns for Approximants at Pre-Baseline and Baseline Assessment Phoneme /!/  Word-Initial  Word-Medial  Substitutions  Substitutions  T1  T2  T1  T2  [w]  [w]  [j] [w]  [j] [w]  [j]  [j] [u]  /l/ /w/  [l]  Word-Final Substitutions T1  T2  [!] [u] !  [!] !  [l]  /j/ Note that goal 3, [CorLab] and word-final [+lateral], also had a link to structure in the phonological systems of the twins in that [+lateral] only had to be targeted in non-word-initial positions. At baseline, [+lateral] was acquired word-initially for both twins and word medially for T2. As mentioned previously, word-final /!/ is only [Coronal] whereas syllable-initial /!/ is [CorLab]. Therefore, although the acquisition of /!/ showed a positional pattern (i.e., it was acquired word finally by both twins), the feature combination [CorLab] was absent. Summary of Goals and their Constraint Rankings Thus, following principles of non-linear phonological intervention, both structural and segmental goals of the phonological hierarchy were targeted, with the focus on promotion of faithfulness to the adult targets. The segmental goals interacted with word position, and so were also in part structural according to the Bernhardt and Stemberger (2000) framework. Therefore, two structural goals and one goal with aspects of both structure and segment/feature were chosen. The targets identified are summarized below with their relevant constraint rankings outlined. The constraints used here are taken from the list of Bernhardt and Stemberger (1998). (1) Structural goals: Goal 1. Stress patterns: Sww, wSw, wS. This goal addresses these three stress patterns and the constraint ranking NotProminent(Foot,Right) > Survived(Syllable) which leads to  62  deletion of the initial weak syllable in words with iambic stress. This goal also addresses the constraint ranking Binary(Foot,!) > Survived(Syllable), which leads to the deletion of a syllable in three syllable words such as wSw and Sww. NotProminent(Foot,Right): Stressed syllables of a foot must not be on the right. Binary(Foot,!): A foot should have no more than two syllables. Survived(Syllable): Syllables in the underlying representation must be present in the surface pronunciation. Goal 2. /s/-clusters /st/, /sn/, /sp/. This goal addresses consonant sequences and the constraint ranking NoSequence([+cont]-[-cont] Onset ) > Survived(Timing Unit) which leads to the deletion of a consonant in /s/-clusters. NoSequence([+cont]-[-cont] Onset ): No sequences of [+continuant]-[-continuant] in onsets. Survived(Timing Unit): No deletion of the timing unit associated with each segment in a word. (2) Segmental goals, with a link to structure (positional restrictions): Goal 3a. Word-initial /!/ in contrast to /w/ (a main focus for T1 and secondary focus for T2). This goal addresses the missing [Coronal] feature of [CorLab] in missing word positions and the constraint ranking Survived(Labial), NotCo-occurring([Coronal][Labial]) > Survived(Coronal). NotCo-occurring([Coronal][Labial]): The place features [Labial] and [Coronal] may not co-occur. Survived(Coronal): The place feature [Coronal] in the underlying representation must be present in the surface pronunciation and must surface in production. Survived(Labial): The place feature [Labial] in the underlying representation must be present in the surface pronunciation and must surface in production.  63  Goal 3b. Word-final /l/ in contrast to /!/ (a main focus for T2 and secondary focus for T1). This goal addresses [+lateral] in missing word positions and the lack of /!/-/l/ distinction word finally. The constraint ranking that is being addressed here concerns [+lateral] and [+consonantal]. Not([+lateral] Coda ), Survived([+consonantal]) > Survived([+lateral]), Not([+lateral] Coda ): The feature [+lateral] may not occur in syllable-final positions. Survived([+lateral): The feature [+lateral] in the underlying representation must be present in the surface pronunciation. Survived([+consonantal]): The feature [+consonantal] must surface if it exists in the underlying representation. If both liquids are considered [+consonantal] in their systems, the substitution of [!] for /l/ is logical.  Research Questions: Phonological Treatment Effects Returning to the questions stated at the end of Chapter 1, the first concerned the expected effects of phonological intervention, for both treated and untreated elements of the phonological system. In addition to increases in matches for targeted structures and features, increases in matches were also expected for elements closely related to targeted features (direct generalization). Direct generalization was expected in: •  other word positions (i.e., word-medial /!/ and /l/)  •  sequences related to targeted sequences (i.e., other /s/-clusters)  •  sequences containing targeted features (i.e., /!/- and /l/-clusters)  Also, because several of the unacquired elements of the twins’ phonologies were not directly targeted in phonology intervention, this allowed opportunities for evaluating whether or not generalization occurred to untargeted features and structures after treatment (indirect generalization). Indirect generalization would be expected simply because the children were in a focused phonological treatment block. Indirect generalization was thus expected in:  64  •  [-grooved] /!/, /"/  •  [-anterior] /#/, /$/, /%/  •  clusters that did not contain any treated features or feature combinations (i.e., [+nasal][-cont])  Treatment Plan The order of the specific targets for treatment in the treatment cycles was: (1) wordinitial /s/-clusters /st/, /sn/, & /sp/, (2) word-initial /&/ versus /w/ and word-final /l/ versus /&/, and (3) wS, Sww, & wSw stress patterns. A strictly structural goal was chosen to be first in keeping with Von Bremen (1990) who found that the twin assigned to a ‘word structure first’ condition, made greater gains. The method used was a modified cycles approach (Hodson & Paden, 1991) as in Bernhardt (1990, 1992). The first session targeted the first goal, the second session briefly reviewed the first goal and targeted the second goal, the third session briefly reviewed the second goal and targeted the third goal, and the fourth session revisited all three goals (in keeping with the approach of Bernhardt, 1990). The plan was then repeated in sessions 5-8. This schedule allowed for repeated exposure to each goal, follow up on homework from the previous week’s goal, and a predictable and consistent schedule for the child. Each goal was targeted with more awareness activities in the first half of the 8-week block, and more production activities in the second half of the block. In the awareness stage, auditory bombardment and minimal pair (perceptual contrast) activities were used, thus including an aspect of perceptual training. In the production stage, frequent articulatory practice and a gradual increase in complexity was used, as in Van Riper’s (1953) traditional method. This began with the child producing the targeted structure either in isolation or in single words (treatment words) first in imitation, and then in elicited non-imitative contexts. Next, these treatment words were put into increasingly longer phrases. Finally, the treatment words were  65  used in an activity at the conversational level. This progression moved at a different pace for each twin and for each goal. A week-by-week treatment plan is shown in Table 14. Table 14. Week-by-Week Treatment Plan Week  Treatment Plan  Awareness(A)/ Production(P)  T1  T2  T1  T2  1  Main lesson: Word-initial /s/-clusters (45 mins)  AP  AP  2  Review: Word-initial /s/-clusters (15mins)  A  A  Main lesson: Word-initial /!/ (30-35 mins)  AP  AP  Review: Word-final /!/ & /l/ (15 mins)  AP  A  Main lesson: wS, Sww, & wSw stress patterns (30-35 mins)  AP  AP  Review all goals:  P  P  - Word-initial /!/;  P  P  - wS, Sww, & wSw stress patterns (15 mins)  AP  AP  5  Main lesson: Word-initial /s/-clusters (45 mins)  P  P  6  Review: Word-initial /s/-clusters (15 mins)  P  P  Main lesson: Word-initial /!/  Main lesson: Word-final /!/  AP  P  Review: Word-initial /!/  Review: Word-final /!/ & /l/  P  P  (15mins)  (15 mins)  Main lesson: Words with wS, Sww, & wSw stress patterns (30  AP  P  Review all goals:  P  P  P  P  P  P  3 4  -Word-initial /s/-clusters (15 mins) - Word-final /!/ & /l/ (15 mins)  (30 mins) 7  A  & /l/ (30 mins)  mins) 8  A  - Word-initial /s/-clusters (15 mins) - Word-initial /!/ (15 mins)  - Word-final /!/ & /l/ (15 mins)  - Words with wS, Sww, & wSw stress patterns (15 mins)  Because each twin was only seen for one session per week, practice at home and with the school classroom assistant was included in order to provide each participant with more opportunities for practice and repetition. For each participant, treatment activities, at-school  66  practice, and at-home practice were done completely independent of the other twin. This ensured that each participant’s treatment was kept as separate as possible. However, the fact that the twins share the same home environment meant that they were constantly interacting, listening to and potentially reinforcing one another’s speech patterns, and learning from one another. The outside therapy activities included auditory bombardment activities (i.e., storybook or poem), homework sheets with treatment word activities (i.e., separating the words that begin with the target sound from distractors), minimal pair games (i.e., listening to parent productions and sorting matched productions from mismatched productions), and games designed to elicit production (i.e., “go fish”). Taking a psycholinguistic perspective, the treatment words and activities used for intervention were chosen with language, language processing and verbal memory abilities in mind. Early vocabulary and games with familiar or easily explained rules were chosen in order to minimize additional cognitive load and to maximize the resources available for phonological practice. All treatment words chosen for goals 2 and 3 had CVC, CVCVC, CVCVCV(C), CCVC, or CCVCVC structure and any treatment words found to be unfamiliar to the child were eliminated. See Appendix N for a list of treatment words and Appendix O for an example of a treatment plan.  Data Analysis Results were compiled by first transcribing recordings of all 16 CAPES lists (Profile List and List 2 for two participants at four assessment points) and transcribing all 6 probe word recordings (two participants at three assessment points). Reliability of transcription was calculated using the transcriptions of a second transcriber for four of the CAPES lists. Reliability for consonants was 88.4%. Reliability was not calculated for vowels, because they were not the focus of treatment. Analysis of match data was done by entering transcriptions of  67  the 32 lists into the CAPES program. CAPES printouts were compared to manual calculations of the match data for elements that contained mismatches: [+lateral], [CorLab], [-grooved], [-anterior], consonant sequences, stress patterns, vowels, and word shapes. These match data were then organized by twin, assessment, structure, and word position into spreadsheets. This enabled an analysis of changes in matches that occurred after the baseline period, after the first treatment block (morphosyntax for T1 and phonology for T2), and after the second treatment block (phonology for T1 and morphosyntax for T2). Global measures (percent consonant match, percent word shape match, and percent vowel match) were calculated for each assessment point using CAPES Profile List and List 2 data. Because of the low numbers of tokens for each treated element, it was not possible to evaluate treatment effects statistically for individual elements, but only for all treated elements combined. A Wilcoxon statistical analysis was used to address the question: Was there a treatment effect for each twin after each type of treatment? With only one participant receiving each goal attack strategy, it was also not possible to compare the twins’ outcomes statistically. Similarities and differences between the twins were analyzed by calculating a correlation between twin phonologies and looking at twin capability, focus, and treatment outcome. Correlations between the word productions of the twins were calculated pre- and posttreatment using the two CAPES lists. This was calculated by counting the number of words that the twins produced identically and dividing this number by the total number of words in the list. The outcomes of phonological intervention were analyzed using Kwiatkowski and Shriberg’s (1998) capability-focus framework. Participant capability, focus, and treatment outcome were measured as suggested by Kwiatkowski and Shriberg (1998). Capability was measured using consonant inventory based on manner classes. Only children whose inventories contain phonemes in all manner classes are considered to have high capability according to the framework. Focus was measured using the focus scoring system. Fifteen randomly selected  68  video-taped antecedent-response interactions between the child and the clinician were used to code focus as acceptable, questionable, or reduced in three behavioural domains: postural, facial, and verbal. From this, a score out of six was calculated. Three categories of focus were used, i.e., low (0-1.99), medium (2-3.99) and high (4-5.99). Treatment outcome was evaluated by determining if each phonological target had progressed to the next level of the treatment hierarchy in therapy tasks by the end of the treatment phase. Three levels of complexity were used as has been done in previous research (Kwiatkowski & Shriberg, 1998): sound level (clapping for stress patterns), word/phrase level, and generalization. A treatment outcome was considered maximal if 75% of intervention targets had progressed to the next level. This chapter has outlined the methods used in the present study. It has provided information about the dizygotic twin participants and the multiple baseline two-block treatment design. It has also detailed the assessment schedule, assessment procedures, as well as the results of both norm-referenced and descriptive pre-treatment assessments. Pre-treatment phonology results were used to rationalize the non-linear selection of three phonological intervention goals (/s/-clusters, stress patterns, and positional /!/ & /l/). Finally, the treatment plan and the methods used for data analysis have been described.  69  Chapter 3 - Results This chapter will outline the development of the phonological systems of both twin participants over the course of this one-year study. In the first section, results are reported for the global measures: consonant, word shape, and vowel match data. The global measures give an overview of the change that occurred in each twin’s phonological system over the course of the study. Next, match data are presented for all of the structures and features that were considered unacquired at the time of baseline measures. The match data for unacquired elements (both treated and untreated) allow the treatment effects after each block of therapy and over the course of the entire study to be evaluated. The results of the statistical analysis for the development of treated elements are also reported in this section. Following that, data measuring generalization of the phonology goals to conversational speech are presented through single word measures in connected speech. The comments of school personnel regarding the phonological development of each twin at the end of the study are then presented as a second measure of generalization. These combined analyses address the first three research questions about treatment effects after phonology treatment, treatment effects after morphosyntax treatment, and treatment order effects. Finally, results regarding twin similarity are reported followed by results regarding twin differences. These results help to answer the fourth research question, about how twin similarities and differences are related to treatment outcomes. Transcriptions for the CAPES and probe data at all assessment points have been included in Appendices E to J. Mismatch data have been included in Appendices M to S.  Global Measures The global measures used to represent changes in the twins’ phonological systems were percentage match for consonants, word shapes, and vowels. The following global changes were observed over the course of the study:  70  T1: Global Measures - Consonants, Word Shapes, and Vowels Specific match data on global measures for T1 across the four assessment points are shown in Table 15 and diagrammed in Figure 5. The following patterns were observed: Consonant match: !  Increased after the baseline period from 147/231 to 163/232 (63.6% to 70.3%).  !  Remained essentially unchanged after morphosyntax treatment (post-morph tx) at 158/226 (69.9%).  !  Increased again after phonology treatment (post-phon tx) to 174/227 (76.7%), a higher proportion than after baseline.  Word shape match: !  Increased after the baseline period from 61/86 to 68/86 (70.9% to 79.1%).  !  Continued to increase after morphosyntax treatment to 72/86 (83.7%).  !  Continued to increase after phonology treatment to 81/86 (94.2%).  Vowel match: !  Increased after the baseline period from 95/125 to 112/121 (76% to 92.6%).  !  Decreased minimally after morphosyntax treatment to 109/119 (91.6%).  !  Continued to decrease minimally after phonology treatment to 111/124 (89.5%).  Table 15. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match Across Assessment Points for T1 Analysis Consonant Word Shape Vowel  Pre-Baseline  Baseline  Post Morph Tx  Post Phon Tx  147/231  163/232  158/226  174/227  (63.6%)  (70.3%)  (69.9%)  (76.7%)  61/86  68/86  72/86  81/86  (70.9%)  (79.1%)  (83.7%)  (94.2%)  95/125  112/121  109/119  111/124  (76.0%)  (92.6%)  (91.6%)  (89.5%)  71  100% 90% 80% Consonant  70%  Word Shape Vowel  60% 50% Pre-baseline  Baseline  Post Morph Tx Post Phon Tx  Figure 5. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match Across Assessment Points for T1  T2: Global Measures - Consonants, Word Shapes, and Vowels Specific match data on global measures for T2 across the four assessment points are shown in Table 16 and diagrammed in Figure 6. The following patterns were observed: Consonant match: !  Decreased after the baseline period from 160/225 to 144/228 (71.1% to 63.2%).  !  Increased to 174/232 (75.0%) after phonology treatment.  !  Continued to increase after morphosyntax treatment to 184/231 (79.7%).  Word shape match: !  Decreased minimally after the baseline period from 69/84 to 70/86 (82.1% to 81.4%).  !  Increased after phonology treatment to 79/86 (91.9%).  !  Remained at 79/86 (91.9%) after morphosyntax treatment.  72  Vowel match: !  Increased minimally after the baseline period from 92/116 to 98/122 (79.3% to 80.3%).  !  Increased after phonology treatment to 116/123 (94.3%).  !  Decreased minimally to 114/123 (92.7%) after morphosyntax treatment.  Table 16. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match Across Assessment Points for T2 Analysis Consonant Word Shape Vowel  Pre-Baseline  Baseline  Post Phon Tx  Post Morph Tx  160/225  (71.1%)  144/228  (63.2%)  174/232 (75%)  184/231  (79.7%)  69/84  70/86  79/86  79/86  (82.1%)  (81.4%)  (91.9%)  (91.9%)  92/116  98/116  116/123  114/23  (79.3%)  (80.3%)  (94.3%)  (92.7%)  73  100% 90% 80% Consonant  70%  Word Shape Vowel  60% 50% Pre-baseline  Baseline  Post Morph Tx Post Phon Tx  Figure 6. Percent Consonant Match, Percent Word Shape Match, and Percent Vowel Match Across Assessment Points for T2  Unacquired Elements: Treated, Direct Generalization, and Indirect Generalization In order to assess the changes to the phonological system throughout the study, matches to individual unacquired elements on the Profile List and List 2 from CAPES (Masterson & Bernhardt, 2001) were analyzed at each of the four assessment points. The unacquired elements of the phonological system were summarized in Tables 9 and 10 in Chapter 2. This allowed for an evaluation of change after phonology treatment (research question 1), change after morphosyntax treatment (research question 2), and change over both treatment blocks (research question 3). Some of the unacquired elements were chosen as targets for intervention while others were not. The elements that were targeted were: (1) Sww, wSw, wS stress patterns.  74  (2) Word-initial /st, /sn/, /sp/ (in CCVC, CCVCVC). (3) Word-initial [CorLab] and word-final [+lateral]. In addition to evaluating results for the targeted elements over time, potential generalization to untargeted areas of the phonological system were also analyzed in order evaluate specific treatment effects. Because some untreated elements are more closely related to the treated targets than others, two groups of generalization elements were created: direct generalization elements and indirect generalization elements. Direct generalization elements were related to treatment targets due to the presence of treated features [+lateral], [CorLab] or /s/-first sequences. However, none of these generalization targets related directly to stress patterns because all unacquired stress patterns at the word lengths assessed (up to three syllables) fell into the treated group. Therefore, no stress patterns were included in the direct generalization group. Change in matches for these direct generalization targets gave an indication as to whether or not generalization had occurred to structures and features in untreated contexts (hypothesis 3 of research question 1). The direct generalization group contained: (1) Treated targets in untreated word positions: word-medial /l/, word-medial /!/, wordmedial and word-final /sp/, /st/. (2) Untreated /s/-clusters: /sl/, /sw/, /sk/. (3) Consonant sequences that contained /l/ and /!/: /k!/, /b!/, /t!/, /d!/, /p!/, /!b/, /!t/, /!s/, /!d/, /gl/, /bl/, /pl/, /lf/. (4) Untreated word shapes with clusters: CCVCVCC, CVCCVC, CCVCCVC, CVCCVCV. The indirect generalization targets included unacquired elements that did not contain treated features or similar sequences. Measuring changes for these indirect generalization elements evaluated whether or not generalization had occurred from treated structures and  75  features to untreated structures and features (hypothesis 3 of research question 1). The indirect generalization group contained: (1) Untreated features [-anterior], [-grooved]. (2) Unrelated clusters /nd/, /nt/, /!g/, /mpj/, /mp/, /kw/. Percentage match was calculated within these three groups of targets: treated, direct generalization, and indirect generalization. Results across assessment points are presented in Figures 7 and 8. The following were observed:  T1: All Unacquired Elements Treated elements match: !  Decreased after the baseline period from 7/23 to 5/23 (30.4% to 21.7%).  !  Increased after morphosyntax treatment to 8/23 (34.8%).  !  Increased more notably after phonology treatment to 19/23 (82.6%).  Direct generalization elements match: !  Increased after the baseline period from 7/37 to 11/37 (18.9% to 29.7%).  !  Increased after morphosyntax treatment to 15/37 (40.5%).  !  Increased after phonology treatment to 32/37 (86.5%).  Indirect generalization elements match: !  Decreased after baseline from 11/38 to 2/38 (28.9% to 5.3%).  !  Remained at 2/38 (5.3%) after morphosyntax treatment.  !  Increased after phonology treatment to 7/38 (18.4%).  76  100% 90% 80% 70% 60% Treated  50%  Direct Generalization  40%  Indirect Generalization  30% 20% 10% 0% Pre-Baseline  Baseline  Post Morph Tx Post Phon Tx  Figure 7. Percentage Match for Unacquired Elements across Assessment Points for T1  T2: All Unacquired Elements Treated elements match: ! Remained at 11/23 (47.8%) after the baseline period. ! Increased after phonology treatment to 18/23 (78.3%). ! Remained at 18/23 (78.3%) after morphosyntax treatment. Direct generalization elements match: ! Increased after baseline from 6/28 to 9/28 (21.4% to 32.1%). ! Increased after phonology treatment to 14/28 (50.0%). ! Continued to increase after morphosyntax treatment to 21/28 (75.0%). Indirect generalization elements match: ! Remained essentially the same after the baseline period at 3/36 to 3/38 (8.3% and 7.9%). ! Increased after phonology treatment to 5/38 (13.2%).  77  ! Increased minimally after morphosyntax treatment to 6/38 (15.8%). 100% 90% 80% 70% 60% Treated  50%  Direct Generalization  40%  Indirect Generalization  30% 20% 10% 0% Pre-Baseline  Baseline  Post Phon Tx Post Morph Tx  Figure 8. Percentage Match for Unacquired Elements across Assessment Points For T2  The previous sections show the results combined according to treated elements, direct generalization, and indirect generalization. However, when each treated and generalization element is represented individually, some distinct patterns emerge. In the following sections, the percentage match data for each individual element are presented beginning with the treated elements, followed by direct generalization elements, and then indirect generalization elements.  T1: Treated Elements For T1, matches for each treated target are presented in Tables 17 and 18 and figure 9. The pre-baseline includes only CAPES data whereas the other three assessment points include both CAPES data and probe data. Some distinct mismatch patterns appeared in the data. Mismatches to stress patterns included mostly weak syllable deletions (e.g., [g!n] for /"#g!n/ ‘again’). Mismatches for /s/-clusters were /s/-deletions (i.e., [t$%] for /st$%/). Mismatches to the adult target for word-initial /%/ were [w] substitutions (i.e., [w&b"n] for /%&b"n/). Mismatches to  78  the adult target for word-final /l/ were substitution of /!/ (i.e., [w"s#] for /w"s$l/) or vowel substitution (i.e., [siu] for /sil/). The following were observed throughout the study for T1: Sww match: In CAPES: Was not assessed. In probes: •  Remained at 1/1 from the baseline assessment on.  wSw match: In CAPES: !  Decreased after the baseline period from 2/2 to 0/2.  !  Remained at 0/2 after morphosyntax treatment.  !  Increased after phonology treatment back to 2/2.  In probes: !  Increased from 0/1 at baseline to 1/1 after morphosyntax treatment.  !  Remained at 1/1 after phonology treatment.  wS match: In CAPES: !  Decreased after the baseline period from 2/3 to 1/3.  !  Remained at 1/3 after morphosyntax treatment.  !  Increased after phonology treatment to 3/3.  Word-initial (WI) /sn/ match: In CAPES: Was not assessed. In probes: !  Was at 0/2 at baseline and after morphosyntax treatment.  !  Increased after phonology treatment to 1/2.  79  Word-initial (WI) /st/ match: In CAPES: !  Remained at 0/1 after the baseline period.  !  Was not assessed after morphosyntax treatment.  !  Increased after phonology treatment to 1/1.  In probes: !  Was at 0/2 at baseline and after morphosyntax treatment.  !  Increased after phonology treatment to 1/2.  Word-initial (WI) /sp/ match: In CAPES: !  Remained at 0/2 after the baseline period and morphosyntax treatment.  !  Increased after phonology treatment to 1/2.  In probes: •  Remained at 0/2 from the baseline assessment on.  Word shape match for CCVC and CCVCVC: In CAPES: !  Increased after the baseline period from 3/9 to 4/9.  !  Increased after morphosyntax treatment to 7/9.  !  Increased after phonology treatment to 9/9.  Word-initial /!/ match: In CAPES: !  Remained at 0/2 at all assessment points.  In probes: !  Remained at 0/3 from the baseline assessment on.  80  Word-final /l/ match: In CAPES: !  Remained at 0/4 after the baseline period and after morphosyntax treatment.  !  Increased after phonology treatment to 3/4.  In probes: !  Was at 0/1 after baseline and remained at 0/1 after morphosyntax treatment.  !  Increased after phonology treatment to 1/1.  Table 17. Match Data for Treated Structural Elements Across Assessment Points for T1 Element  Data  Post Morph  Pre-Baseline  Baseline  CAPES  -  -  -  -  Probes  -  1/1  1/1  1/1  CAPES  2/2  0/2  0/2  2/2  Probes  -  0/1  1/1  1/1  wS  CAPES  2/3  1/3  1/3  3/3  WI /sn/  CAPES  -  -  -  -  Probes  -  0/2  0/2  1/2  CAPES  0/1  0/1  -  1/1  Probes  -  0/2  0/2  1/2  CAPES  0/2  0/2  0/2  1/2  Probes  -  0/2  0/2  0/2  CAPES  3/9  4/9  7/9  9/9  Sww wSw  WI /st/ WI /sp/ CCVC  CCVCVC  Type  Tx  Post Phon Tx  Table 18. Match Data for Treated Segments Across Assessment Points for T1 Element  Data Type  Pre-Baseline  Baseline  Post Morph Tx  Post Phon Tx  WI /!/  CAPES  0/2  0/2  0/2  0/2  Probes  -  0/3  0/3  0/3  CAPES  0/4  0/4  0/4  3/4  Probes  -  0/1  0/1  1/1  WF /l/  81  100% 90% 80% 70% CorLab  60%  [+lateral]  50%  s-clusters  40%  Word Shapes  30%  Stress  20% 10% 0% Pre-Baseline  Baseline  Post Morph Tx Post Phon Tx  Figure 9. Treated Element Match Data Across Assessment Points for T1 Note. The pre-baseline totals include only CAPES data, whereas the other three points contain CAPES and probe data.  T2: Treated Elements For T2, matches for all treatment targets are presented in Tables 19 and 20 and Figure 10. Both CAPES matches and probe matches have been included in these tables. However, as previously mentioned, no probe data were collected at the first assessment. Some distinct mismatch patterns were seen throughout the study. Mismatches for stress patterns were weak syllable deletions (i.e., [g!n] for /"#g!n/ ‘again’). Mismatches for treated /s/-clusters were all /s/deletions (i.e., [t$%] for /sta%/). Mismatches for word-initial /%/ were [w] substitution (i.e., [w&b"n] for /r&b"n/). Mismatches for word-final /l/ were /%/ (i.e., [w&s"r] for /w&s"l/). The following changes were observed for T2 throughout the study: Sww match: In CAPES: Was not assessed.  82  In probes: !  Was 1/1 from the baseline assessment on.  wSw match: In CAPES: !  Increased after baseline from 0/2 to 1/2.  !  Increased after phonology treatment to 2/2.  !  Remained at 2/2 after morphosyntax treatment.  In probes: !  Was 1/1 from the baseline assessment on.  wS match: In CAPES: !  Remained at 2/3 after the baseline period.  !  Increased after phonology treatment to 3/3.  !  Decreased after morphosyntax treatment to 2/3.  In probes: !  Was 0/2 at baseline and remained at 0/2 after phonological treatment.  !  Increased to 2/2 after morphosyntax treatment.  Word-initial /sn/ match: In CAPES: Was not assessed. In probes: !  Was at 0/2 at baseline and remained at 0/2 after phonology treatment.  !  Increased to 2/2 after morphosyntax treatment.  Word-initial /st/ match: In CAPES: !  Remained at 0/1 after the baseline period.  83  !  Increased after phonology treatment to 1/1.  !  Remained at 1/1 after morphosyntax treatment.  In probes: !  Increased from 0/2 at baseline to 2/2 after phonology treatment.  !  Remained at 2/2 after morphosyntax treatment.  Word-initial /sp/ match: In CAPES: !  Remained at 0/2 after the baseline period.  !  Increased after phonology treatment to 1/2.  !  Remained at 1/2 after morphosyntax treatment.  In probes: !  Increased from 0/2 at baseline to 1/2 after phonology.  !  Increased after morphosyntax treatment to 2/2.  Word shape match for CCVC and CCVCVC: In CAPES: !  Remained at 7/9 after the baseline period.  !  Increased after phonology treatment to 9/9.  !  Remained at 9/9 after phonology treatment.  Word-initial /!/ match: In CAPES: •  Remained at 1/2 at all assessment points.  In probes: !  Decreased from 1/3 at baseline to 0/3 after phonology treatment.  !  Increased after morphosyntax treatment to 3/3.  Word-final /l/ match:  84  In CAPES: !  Remained at 0/4 after the baseline period.  !  Increased after phonology treatment to 1/4.  !  Increased after morphosyntax treatment to 3/4.  In probes: Remained at 0/1 from the baseline assessment on.  •  Table 19. Match Data for Treated Structural Elements Across Assessment Points for T2 Structure Sww wSw wS WI /sn/ WI /st/ WI /sp/ CCVC  CCVCVC  Data  Pre-Baseline  Baseline  Post Phon Tx  Post Morph Tx  CAPES  -  -  -  -  Probes  -  1/1  1/1  1/1  CAPES  0/2  1/2  2/2  2/2  Probes  -  0/1  0/1  0/1  CAPES  2/3  2/3  3/3  2/3  Probes  -  0/2  0/2  2/2  CAPES  -  -  -  -  Probes  -  0/2  0/2  2/2  CAPES  0/1  0/1  1/1  1/1  Probes  -  0/2  2/2  2/2  CAPES  0/2  0/2  1/2  1/2  Probes  -  0/2  1/2  2/2  CAPES  7/9  7/9  9/9  9/9  Type  Table 20. Match Data for Treated Segments Across Assessment Points for T2 Post Morph  Segment  Data Type  Pre-Baseline  Baseline  Post Phon Tx  WI /!/  CAPES  1/2  1/2  1/2  1/2  Probes  -  1/3  0/3  3/3  CAPES  0/4  0/4  1/4  2/4  Probes  -  0/1  0/1  0/1  WF /l/  Tx  85  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  CorLab [+lateral] s-clusters W-Shapes Stress  Figure 10. Treated Element Match Data Across Assessment Points for T2 Note. The pre-baseline totals include only CAPES data, whereas the other three points contain CAPES and probe data.  Statistical Analysis Wilcoxon matched pairs signed ranks tests were completed to evaluate matches to the adult form on all treated phonological elements combined between assessments 1 and 2, between assessments 2 and 3, and between assessments 3 and 4 for each twin. The only significant differences (using a criterion of p < 0.05) found were between assessments 3 and 4 for T1, p< 0.001 (post phonology treatment), and between assessments 2 and 3 for T2, p< 0.02 (post phonology treatment).  T1: Direct Generalization Elements In this section the percentage match will be outlined for each individual item of the direct generalization group. This group included items that contained a treated feature or structure but that were not treated directly. Match data on CAPES and probes for T1 on these  86  items are outlined in Tables 21 and 22. Once again, the pre-baseline assessment does not include probe data. Looking at change in matches for these direct generalization elements was helpful in determining whether or not generalization had occurred from treated contexts to untreated contexts for T1. Some distinct mismatch patterns were observed. They were as follows (1) mismatches to direct generalization /s/-clusters included both /s/-deletion and deletion of the second consonant (i.e., [ki] for /ski/ and [s!"d] for /sl!"d/) (2) mismatches for direct generalization items containing /#/ were glide substitutions (i.e. [g$%w"l$] for /g$%#"l$/) and /#/-deletions (i.e., [%b&v'] for /b#&v'/) (3) mismatches for direct generalization items containing /l/ were glide substitutions (i.e., [bwun] for /b$%lun/) and /l/-deletion (i.e., [gæs"z] for /glæs"z/). For T1, the following were observed: Match for treated /s/-clusters in untreated positions (word-medial (WM) and word-final (WF) /st/ and /sp/): In CAPES: !  Decreased after the baseline period from 1/2 to 0/2.  !  Increased after morphosyntax treatment to 2/2.  !  Remained at 2/2 after phonology treatment.  In probes: •  Remained at 0/1 from the baseline assessment on.  Match to untreated /s/-clusters (/sl/, /sw/, /sk/): In CAPES: !  Remained at 0/3 after the baseline period.  !  Increased after morphosyntax treatment to 1/3.  !  Increased after phonology treatment to 2/3.  87  Match to word shapes CCVCVCC, CVCCVC, CCVCCVC: In CAPES: !  Increased after the baseline from 0/4 to 2/4.  !  Decreased after morphosyntax treatment to 0/4.  !  Increased after phonology treatment to 2/4.  Match to word-medial /!/: In CAPES: •  Remained at 0/2 at all assessment points.  In probes: •  Remained at 0/2 from the baseline assessment on.  Match for C-/!/ sequences (/p!/, /b!/, /t!/, /d!/, /k!/): In CAPES: !  Decreased after the baseline period from 1/7 to 0/7.  !  Increased after morphosyntax treatment to 2/7.  !  Remained at 2/7 after phonology treatment.  In probes: •  Remained at 0/1 from the baseline assessment on.  Match for /!/-C sequences (/!b/, /!t/, /!d/, /!s/): In CAPES: •  Remained at 2/4 after the baseline period and morphosyntax treatment.  •  Increased to 3/4 after phonology treatment.  Match for word-medial /l/: In CAPES: !  Remained at 1/2 after the baseline period and after morphosyntax treatment.  88  !  Increased after phonology treatment to 2/2.  In probes: •  Remained at 1/1 from the baseline assessment on.  Match for C-/l/ sequences (/pl/, /bl/, /gl/): In CAPES: !  Remained at 0/4 after the baseline period and after morphosyntax treatment.  !  Increased after phonology treatment to 1/4.  Match for the /l/-C sequence (/lf/): In CAPES: !  Remained at 0/1 after the baseline period and after morphosyntax treatment.  !  Increased after phonology treatment to 1/1.  Table 21. Match Data for Direct Generalization Structures Across Assessment Points for T1 Structure  Data  Pre-  Baseline  Post Morph  Post Phon  Type  Baseline  WM, WF,  CAPES  1/2  0/2  2/2  2/2  /s/-  /st/, /sp/  clusters  Probes  -  0/1  0/1  0/1  /sl/, /sw/,  CAPES  0/3  0/3  1/3  2/3  CAPES  0/4  2/4  0/4  2/4  /sk/  Word Shapes  CCVCVCC, CVCCVC,  Tx  Tx  CCVCCVC  89  Table 22. Match Data for Direct Generalization Features/Segments Across Assessment Points for T1 Feature WM /!/ [CorLab]  /p!/, /b!/, /t!/, /d!/, /k!/  /!b/, /!t/, /!d/, /!s/ WM /l/ [+lateral]  /pl/, /bl/, /gl/ /lf/  Data  Pre-  Post Morph  Post Phon  Type  Baseline  Tx  Tx  CAPES  0/2  0/2  0/2  0/2  Probes  -  0/2  0/2  0/1  CAPES  1/7  0/7  2/7  2/7  Probes  -  0/1  0/1  0/1  CAPES  2/4  2/4  2/4  3/4  CAPES  1/2  1/2  1/2  2/2  Probes  -  1/1  1/1  1/1  CAPES  0/4  0/4  0/4  1/4  CAPES  0/1  0/1  0/1  1/1  Baseline  T2: Direct Generalization Elements For T2, match data for direct generalization elements on CAPES and probes are outlined in Tables 23 and 24. Once again, the pre-baseline assessment does not include probe data. Looking at change in matches for these direct generalization elements was helpful in determining whether or not generalization had occurred from treated contexts to untreated contexts for T2. Some distinct mismatch patterns were observed throughout the study. Mismatches for direct generalization /s/-clusters were /s/-deletions ([w"p#] for /w"sp#/), deletions of the second consonant ([v$s] for /v$st/), and gliding of /l/ ([swa"d] for /sla"d/). Mismatches for direct generalization items containing /!/ were glide substitutions ([g%&w"l%] for /g%&!"l%/), lateralization ([&bl'v#] for /&b!'(#/), and /!/-deletion ([&g)b"*] for /&g)!b"*/). Mismatches for direct  90  generalization items containing /l/ were gliding ([bwu] for /blu/) and /!/ substitution ([s"!f] for /#"lf/). For T2, the following were observed for direct generalization: Match for treated /s/-clusters in untreated positions (word-medial /sp/ and word-final /st/): In CAPES: !  Increased after the baseline period from 0/2 to 1/2.  !  Remained at 1/2 after phonology treatment and after morphosyntax treatments.  In probes: •  Remained at 0/1 from the baseline assessment on.  Match to untreated /s/-clusters (/sl/, /sw/, /sk/): In CAPES: !  Remained at 0/3 after the baseline period.  !  Increased after phonology treatment to 2/3.  !  Increased after morphosyntax treatment to 3/3.  Match to word shapes CCVCVCC, CVCCVC, CCVCCVC: In CAPES: !  Decreased after the baseline period from 1/4 to 0/4.  !  Increased after phonology treatment to 1/4.  !  Increased after morphosyntax treatment to 3/4.  Match to word-medial /!/: In CAPES: •  Remained at 0/2 at all assessment points.  In probes: !  Increased from 0/2 at baseline to 1/2 after phonology treatment.  !  Remained at 1/2 after morphosyntax treatment.  91  Match for C-/!/ sequences (/k!/, /p!/, /b!/, /t!/, /d!/): In CAPES: !  Increased after the baseline period from 2/7 to 6/7.  !  Decreased after phonology treatment to 4/7.  !  Increased after morphosyntax treatment to 6/7.  In probes: •  Remained at 0/1 from the baseline assessment on.  Match for /!/-C sequences (/!b/, /!t/, /!d/, /!s/): In CAPES: !  Remained at 2/4 after the baseline period.  !  Increased to 3/4 after phonology treatment.  !  Remained at 3/4 after morphosyntax treatment.  Match for C-/l/ sequences (/pl/, /bl/, /gl/): In CAPES: !  Decreased after the baseline period from 1/4 to 0/4.  !  Increased after phonology treatment to 2/4.  !  Increased after morphosyntax treatment to 3/4.  Match for the /l/-C sequence (/lf/): In CAPES: !  Remained at 0/1 after the baseline period and after phonology treatment.  !  Increased after morphosyntax treatment to 1/1.  92  Table 23. Match Data for Direct Generalization Structure Across Assessment Points for T2 Data  Structure /s/clusters  Pre-  Baseline  Post Phon  Post Morph  Type  Baseline  Tx  Tx  WM, WF  CAPES  0/2  1/2  1/2  1/2  /st/, /sp/  Probes  -  0/1  0/1  0/1  /sl/, /sw/, /sk/  CAPES  0/3  0/3  2/3  3/3  CAPES  1/4  0/4  1/4  3/4  CCVCVCC, CVCCVC,  Word  shapes  CCVCCVC, CVCCVCV  Table 24. Match Data for Direct Generalization Features/Segments Across Assessment Points for T2 Feature WM /!/ [CorLab]  /p!/, /b!/, /t!/, /d!/, /k!/  /!b/, /!t/, /!d/, /!s/ WM /l/ [+lateral]  /pl/, /bl/, /gl/ /lf/  Data  Pre-  Post Phon  Post Morph  Type  Baseline  Tx  Tx  CAPES  0/2  0/2  0/2  0/2  Probes  -  0/2  1/2  1/2  CAPES  2/7  6/7  4/7  6/7  Probes  -  0/1  0/1  0/1  CAPES  2/4  2/4  3/4  3/4  CAPES  1/2  Acquired  Acquired  Acquired  CAPES  1/4  0/4  2/4  3/4  CAPES  0/1  0/1  0/1  1/1  Baseline  T1: Indirect Generalization Elements The percentage match for structures and features that were not directly related to the treatment goals also showed some change over the course of the study. Looking at change in matches for these indirect generalization elements was useful in determining whether or not  93  generalization had occurred from treated structures and features to untreated structures and features for T1. This section outlines the development of each unacquired phonological target that was not directly related to treatment goals. Percentage match data on CAPES for T1 are presented in Table 25. Mismatch patterns were as follows (1) mismatches to the feature [-anterior] included stopping (i.e., [dæm] for /!æm/), depalatalization (i.e., ["#mp] for /!#mp/), and deaffrication (i.e. [w$s] for /w$%/) (2) mismatches for [-grooved] were frication for the voiceless fricative (i.e., [f#m] for /&#m/) and stopping for the voiced fricative (i.e., [dæt] for /'æt/). For T1, the following were observed throughout the course of the study: Match to the feature [-anterior]: !  Decreased from 11/21 to 0/21 after the baseline period.  !  Remained as 0/21 after morphosyntax treatment.  !  Increased to 3/21 after phonology treatment.  Match to the feature [-grooved]: !  Remained at 0/9 throughout the course of the study.  Match for the indirect generalization clusters /nd/, /nt/, /(g/, /mpj/, /mp/, /kw/: !  Increased after the baseline period from 0/8 to 2/8.  !  Remained at 2/8 after morphosyntax treatment.  !  Increased after phonology treatment to 4/8.  Table 25. Match Data for Indirect Generalization Elements Across Assessment Points for T1 T1  Pre-Baseline  Baseline  Post Morph Tx  Post Phon Tx  [-anterior]  11/21  0/21  0/21  3/21  [-grooved]  0/9  0/9  0/9  0/9  0/8  2/8  2/8  4/8  /nd/, /nt/, /(g/, /mpj/, /mp/, /kw/  94  T2 Indirect Generalization Elements Percentage match data on CAPES for T2 are presented in Table 26. Looking at change in matches for these indirect generalization elements was useful in determining whether or not generalization had occurred from treated structures and features to untreated structures and features for T2. Mismatches to the feature [-anterior] included stopping (i.e., [dæm] for /!æm/), and deaffrication/ depalatalization (i.e., [w"s] for /w"#/). Mismatches to the feature [-grooved] were frication for the voiceless fricative (i.e., [f$m] for /%$m/) and stopping for the voiced fricative (i.e., [dæt] for /&æt/). The following were observed throughout the course of the study: Match to the feature [-anterior]: !  Decreased after the baseline period from 1/19 to 0/21.  !  Remained at 0/21 after phonology treatment.  !  Increased after morphosyntax treatment to 1/2.  Match to the feature [-grooved]: !  Remained at 0/9 at all assessment points.  Match for the indirect generalization clusters /nd/, /nt/, /'g/, /mpj/, /mp/, /kw/: !  Increased after the baseline period from 2/8 to 3/8.  !  Increased after phonology treatment 2/8 to 5/8.  !  Remained at 5/8 after morphosyntax treatment.  Table 26. Match Data for Indirect Generalization Elements Across Assessment Points for T2 T2  Pre-Baseline  Baseline  Post Phon Tx  Post Morph Tx  [-anterior]  1/19  0/21  0/21  1/21  [-grooved]  0/9  0/9  0/9  0/9  2/8  3/8  5/8  5/8  /nd/, /nt/, /'g/,  /mpj/, /mp/, /kw/  95  Summary of Phonological Changes The above presentation of the data collected through CAPES and probes used individual match scores to outline results. The tables presented in this section (Tables 27 and 28) use the change in number of matches to summarize results. It is useful to see the data presented in this way because the number of tokens elicited in the CAPES assessment for unacquired elements of the phonological system were often small and ranged widely (from 2-21). The same CAPES data presented in Tables 17 to 26 and Figures 9 and 10 above has been represented below in Tables 27 and 28 by reporting the increase in matches directly. This gives a clear diagram of where any new matches to a target form occurred in each area of the phonological system over the course of the study. Only CAPES data have been included in these tables in order to keep the number of tokens assessed at each time point consistent. Thus, an increase in one match, whether the number of tokens was two or 21, is shown as a ‘1,’ recognizing that a change occurred, but showing that this change was not (or may not have been) large. A ‘2’ shows that two new matches occurred and a ‘3+’ shows that three or more new matches occurred. A ‘-’ sign was used to indicate the few instances where a large decrease in matches (three or more) occurred. The decrease is indicated in brackets. In those circumstances where a decrease did occur, later increases in matches were not included in the chart unless they surpassed the initial number of matches. Thus, only matches over and above matches at previous assessments were considered. For example, recall that matches to stress patterns for T1 decreased from 4/5 to 1/5 after the baseline period. An increase in matches was not recorded in this summary chart of match changes until it exceeded the initial total of 4/5. Thus, after the phonology treatment, when matches increased to 5/5, T1 was given a score of ‘1.’ Table 27 shows the increases in number of matches to the adult target that T1 showed after the baseline period, after morphosyntax treatment, and after phonology treatment for treated  96  targets, direct generalization targets, and indirect treatment targets on CAPES. Table 28 shows the same information for T2. Table 27. Match Increase by Phonological Element on CAPES Across Assessment Points for T1 Phonological Element Treated Targets  T1 Baseline  Morphosyntax  Phonology  period  Treatment  Treatment  /st/, /sn/, /sp/ Stress patterns Word shapes  2 - (3) 1  3+ 3+  2  [CorLab] /!/ [+lateral] /l/ Direct  /s/-clusters  Generalization  Word shapes  Indirect  Generalization  3+ 2  1  2  [CorLab] /!/  1  [+lateral] /l/  1  3+  [-grooved] /", #/ [-anterior] /$, %, &/ Other clusters  - (11) 2  2  97  Table 28. Match Increase by Phonological Element on CAPES Across Assessment Points for T2 Phonological Element Treated  T2 Baseline  Phonology  Morphosyntax  period  Treatment  Treatment  /st/, /sn/, /sp/ Stress patterns  2 1  Word shapes  2 2  [CorLab] /!/ [+lateral] /l/ Direct  Generalization  Indirect  Generalization  /s/-clusters  1  1  1  2  1  Word shapes  2  [CorLab] /!/  3+  1  [+lateral] /l/  1  2  1  3+  2  [-grooved] /", #/ [-anterior] /$, %, &/ Other clusters  Tables 27 and 28 provide a slightly different interpretation of the match data in this study. Providing a summary of all changes in matches within one chart for each twin is useful in evaluating phonological change at a glance. It can be seen that T1 showed increases in matches of two or three tokens for many targets after phonology treatment, especially treated targets. For T2, increases of three tokens are few and scattered but many increases of two tokens are seen after phonology treatment, and some after morphosyntax treatment.  Generalization to Connected Speech The CAPES and probe data presented in this chapter provide a measure of subtle changes in different areas of the phonological system at different points throughout this study. However, it is also important to evaluate whether or not the phonology treatment targets had become part of the child’s repertoire in spontaneous connected speech. Generalization of  98  therapy targets to connected speech was calculated for each twin based on single words in connected speech during the narrative elicitation task at the final assessment. T1 matched wordinitial [CorLab] on 0/19 opportunities, word-final [+lateral] on 4/8 opportunities, and /s/clusters on 2/3 opportunities. T2 matched word-initial [CorLab] on 6/8 opportunities, word-final [+lateral] on 19/29 opportunities, and /s/-clusters on 10/14 opportunities. Generalization to the treated stress patterns was not assessed in connected speech because neither twin used any words with the targeted stress patterns in the narrative sample. Table 29. Match Data for Single Words in Narrative Task for T1 and T2  T1  T2  Element  % Match  WI [CorLab]  0.0% (0/21)  WF [+lateral]  50.0% (4/8)  /s/-clusters  50.0% (1/2)  Stress patterns  No data  WI[CorLab]  90.9% (10/11)  WF[+lateral]  65.5% (19/29)  /s/-clusters  77.8% (7/9)  Stress patterns  No data  Qualitative Results Another measure of each child’s use of phonology treated targets in spontaneous speech was based on the comments provided by each boy’s classroom assistant. The assistant in T1’s classroom commented that, for stress patterns, “with prompts, he will now say ‘com-puter’ where previously this had been ‘puter’.” With regard to T1’s lack of distinction within the approximant category, the classroom assistant commented that “[T1] still generally uses ‘l’ for ‘w’ and ‘w’ for ‘r’ although some improvement is intermittently appearing.” For /s/-clusters, the classroom assistant reported that T1 was beginning to self-monitor his production with reminders. “At recess for instance, when he would say, “Time for nack.” I would say, “[T1],  99  time for what?” He would then say, “Time for sssnnack,” in an exaggerated manner.” The classroom assistant wrote that “overall the biggest improvement I see around speech for [T1] is his growing awareness of what we have been working on.” For T2, the classroom assistant also commented on generalization of the phonological goals. “Quite consistent use of s-[clusters], and /!/ and /l/ sounds at the beginning of words” were reported. With regard to stress patterns, the classroom assistant wrote that “[T2] still tends to leave out the quiet syllables in longer words but will self-correct if the incorrect word is repeated to him.” T2’s classroom assistant also wrote that “[T2] is more aware of his speech.”  Similarities and Differences Correlations between twins on CAPES productions and the capability-focus framework analysis (Kwiatkowski and Shriberg, 1998) were used to answer the fourth research question: What are the similarities and differences between the two participants and how do they influence intervention outcomes?  Twin Similarity As was mentioned in Chapter 2, the phonological systems of the two participants were very similar. As a measure of twin similarity, correlations were calculated between the word productions for CAPES for the two boys. Words that the twins produced the same (with identical phonemes) were added up and this number was divided by the total number of words. Correlations were .48 pre-treatment and .59 post-treatment. This acted as a measure of similarity between the two twin participants in phonology.  Twin Difference In addition to understanding similarities, it was also important to understand the children’s differences, because these differences may have been relevant to treatment outcomes. Capability and focus measures were used for this purpose. Capability, focus, and treatment  100  outcome were calculated for each boy using the framework developed by Kwiatkowski and Shriberg (1998) as discussed in the Methods chapter. Both twins were considered to have low capability because their consonant inventories contained phonemes from only five of the six manner classes. Their inventories contained nasals, stops, glides, fricatives, and liquids (although marginal) but not affricates. For T1, focus was calculated to be medium and for T2 focus was calculated to be high. Treatment outcome was evaluated by determining if each phonological target had progressed to the next level of the treatment hierarchy in therapy tasks. For both twins, therapy began with activities at the sound level for [+lateral], the clapping level for stress patterns, and at the word level for [CorLab], and /s/-clusters. For T1, by the end of the treatment block, [+lateral] and stress activities had progressed to the word level, but [CorLab] and /s/-cluster activities had not progressed to the generalization level. Therefore, 2/4 (50.0%) targets had progressed for T1. For T2, [+lateral] and stress activities had progressed to the word level and /s/-cluster activities had progressed to the generalization level, meaning that treatment words were incorporated into unstructured activities that promoted natural conversation. Hence, 3/4 (75%) targets had progressed for T2. A treatment outcome was considered maximal if 75% of targets had progressed to the next level (see Table 30). Therefore, T1’s outcome was considered non-maximal and T2’s was considered maximal according to Kwiatkowski and Shriberg’s (1998) framework. Table 30. Capability-Focus Framework - Outcome Measurement Target  T1 Target Progress  T2 Target Progress  [CorLab]  No  No  [+lateral]  Yes  Yes  Stress patterns  Yes  Yes  /s/-clusters  No  Yes  Total  2/4 = 50%  3/4 = 75%  101  As described in Chapter 1, according to Kwiatkowski and Shriberg’s (1998) model, the typology for the capability-focus framework contains four boxes; high capability - high focus, high capability - low focus, low capability - high focus, and low capability - low focus. In order to discriminate between focus of 3.6 and 5.6, the respective focus scores of T1 and T2, three categories of focus have been used here, low (0-1.99), medium (2-3.99), and high (4-5.99). This gave six types rather than four, high capability - high focus, high capability - medium focus, high capability - low focus, low capability - high focus, low capability - medium focus, and low capability - low focus. This typology is presented in Table 31. Thus, T1 falls into the low capability - medium focus category with a non-maximal outcome and T2 falls into the low capability - high focus category with a maximal outcome. This is a way to represent treatment outcomes that makes a connection to characteristics of the child as an active learner. Table 31. Capability-Focus Typology for T1 and T2 Focus High  Capability Low  4 -5.99  2 -3.99  0 -1.99  High  High  High  Capability  Capability  Capability  High Focus  Medium  Low Focus  Low  Low  Low  Capability  Capability  Capability  High Focus  Focus  Medium  Low Focus  Focus Adapted from Kwiatkowski and Shriberg (1998) T1 – Non-maximal Outcome T2 – Maximal Outcome This results chapter has presented the data of the twin intervention study in order to answer the four research questions posed in Chapter 1. Both global measures of phonology (percent consonant match, percent, word-shape match, and percent vowel match) and match  102  data for individual elements in the phonological systems of the twins were presented across four assessment points. Two evaluations of generalization of treatment targets into connected speech were also given. Finally, this chapter presented the results of two analyses that address similarities and differences, both quantifying the similarities between the two boys’ phonologies and relating treatment outcomes to their individual differences according to the capability-focus framework (Kwiatkowski & Shriberg, 1998).  103  Chapter 4 - Discussion This chapter begins with a discussion of the four research questions that were posed in Chapter 1. The structural and featural changes that occurred in each twin’s phonological system over the course of the study are discussed next with regard to their relevant constraint rankings and possible mechanisms of change. The final sections of this chapter present the limitations of the study, directions for future research, and clinical implications.  Research Questions The four research questions addressed in this study were: (1) Does phonological treatment result in specific changes to either twin’s phonological system and, if so, what type of changes? (2) Does morphosyntax treatment result in specific changes to either twin’s phonological system and, if so, what are these changes? (3) Are morphosyntax-first and phonology-first intervention strategies effective in facilitating phonological development? (4) What are the similarities and differences between the two participants and how do they influence intervention outcomes? Question (1). Did phonological treatment result in specific changes to either twin’s  phonological system and, if so, what type of changes? The first hypothesis stated that changes to the phonological system would be seen following the phonology treatment block. Because the results of this study are based on small numbers of tokens, it is difficult to assess how phonological changes were related to intervention. However, clear phonological change was seen in specific elements of the phonological system after the phonological treatment blocks for both children. Both twins had statistically significant increases in matches for treated phonological elements after phonology  104  treatment. Overall, for global measures of percentage match for consonants, word shapes, and vowels and for individual phonological elements, more change was seen after phonological treatment than after the baseline period or after morphosyntax treatment. This was true for both twins. This finding shows support for the hypothesis that the specificity of treatment during the phonological treatment block would result in changes to the phonological systems of participant children. This is consistent with the findings of Almost and Rosenbaum (1998), Law et al. (2004), and Tyler et al. (2003) and suggests that intervention focused on speech leads to development of the phonological system and that treatment focused on phonology leads to more adult-like speech production. This finding is also consistent with the results of studies that have found that a non-linear approach to goal selection was effective in facilitating phonological growth (Bernhardt & Stemberger, 1998; Bernhardt, 1990; Von Bremen, 1990; Bernhardt & Major, 2005; and Shoaf et al., 2009). Ultimately, the goal of intervention is generalization of new phonology to connected speech. Thus, phonological changes must also be evaluated in these terms. Measures of generalization of specific goals to single words within connected speech indicated that while T1 did not reach the 75% match acquisition criteria for any of the three goals evaluated ([+lateral], [CorLab], or /s/-clusters), T2 reached this criteria for [CorLab] and /s/-clusters but not for [+lateral] (although the 65.5% match for this goal was approaching criterion). Reports from the classroom assistants at the twins’ school were consistent with the connected speech generalization measures. The assistants commented that both boys had increased their awareness of speech, that T1 had begun to self-correct with prompts but was not using the phonological forms targeted in treatment in conversation yet, and that T2 was using /s/-clusters and [CorLab] in treated positions quite consistently but continued to use weak syllable deletion. These reports offer additional evidence that both twins had made some progress toward generalization.  105  The second hypothesis for the first research question stated that treated elements within the phonological system would undergo greater change than untreated elements. As outlined in Chapter 3, for T1, treated elements in both treated and untreated contexts underwent greater change than did untreated elements. For T2, treated elements in treated contexts underwent the greatest change, followed by treated elements in untreated contexts, then by untreated elements. This finding supports the second hypothesis and is consistent with the findings of Von Bremen (1990), whose participants showed minimal development of unrelated segments after intervention. Generalization did occur within voiced or voiceless cognates of treated segments in Von Bremen’s study, but not beyond these segment types. The third part of the hypothesis for the first research question stated that generalization would occur from treated contexts to untreated contexts and from treated structures and features to untreated structures and features. Generalization from treated contexts to untreated contexts can be evaluated by looking at the change in percentage match of direct generalization elements. Both twins showed large changes in percentage match for treated items in untreated contexts after phonology therapy: an increase of 46% match (40.5% - 86.5%) for T1 and an increase of 17.9% match (32.1% – 50%) for T2. The specific features that showed generalization to new contexts for T1 were [+lateral] and [CorLab]. Although matches for /s/-clusters increased for T1, these increases were not greater than those that had occurred after morphosyntax treatment. This implies that generalization to new contexts can occur, and is consistent with what has been reported in the literature (Bankson, Bernthal, & Flipsen, 2009). For T2, the elements that showed generalization to new contexts were /s/-clusters, word shapes containing clusters, [+lateral], and [CorLab] (although not until after morphosyntax treatment). These changes in matches were all greater than changes seen after the baseline period. Generalization from treated structures to untreated structures can be evaluated by looking at the change that occurred throughout the study for clusters whose sequences were not  106  similar to treated clusters and that did not contain any treated features (i.e., indirect generalization clusters). The change in percentage match that was seen for T1 after phonology therapy was not greater than the change seen after the baseline period and therefore is not considered an indication of generalization. For T2, a change of 25% match occurred after phonology treatment, one that was greater than changes at any other point throughout the study. Thus, generalization from treated structures to untreated structures was seen for T2. In order to evaluate generalization from treated features to untreated features, the match data for untreated [–anterior] and [-grooved] can be used. Neither of these features showed any development throughout the study for either twin. Thus generalization from treated features to untreated features did not occur for these participants. However, the fact that the segmental goals selected for intervention were linked to word position meant that no entirely segmental goal was targeted. It is possible that segmental generalization would have been seen if a purely segmental goal had been selected for intervention. Also of interest are the further changes in percentage match that occurred for T2 after the second block of treatment (morphosyntax), which represented a possible delayed generalization of phonological elements. Treatment targets in treated contexts did not experience a further increase in percentage match overall but were nonetheless maintained, indicating that T2 retained what he had learned in treatment. Matches to treated targets in untreated contexts did increase further, however (by 25%). This change may indicate a delayed generalization of treated structures and features to other contexts due to phonological therapy, and supports offering morphosyntax intervention after phonology intervention. This leads to the second research question.  107  Question (2). Does morphosyntax treatment result in specific changes to either twin’s  phonological system and, if so, what are these changes? For T1, there were increases in percentage match seen for some specific elements of the phonological system after morphosyntax treatment (word shapes containing cluster, stress patterns, and some /s/-clusters). However, there were also some specific elements that increased in matches after the baseline period (other word shapes and [+lateral] in some contexts). In terms of global measures, phonological changes seen after morphosyntax treatment (change from 63.3% to 70.3% for PCM) were not greater than those seen after the baseline period (change from 69.9% to 76.7% for PCM). For targets that were to be treated in the phonology block, there was a decrease in matches after this period. Thus, it can be concluded that morphology-only treatment did not lead to gains for T1 in phonology. For T2, changes in matches were greater after morphosyntax treatment than they had been after baseline for many elements (some /s/-clusters, [+lateral] in some contexts, [CorLab], and some word shapes). The opposite was only true for matches to stress patterns. After baseline for the global measures PCM decreased by 7.9% (from 71.1% to 63.2%) and after morphosyntax PCM increased by 4.7% (from 75% to 79.9%), indicating that greater change in matches was seen after morphosyntax treatment as compared to baseline. However, it is important to note that changes after morphosyntax treatment are confounded with any delayed generalization that was occurring from the preceding block of phonology treatment. Thus, it cannot be concluded that change in matches for T2 after morphosyntax treatment were necessarily a product of cross-domain generalization. What can be concluded is that morphosyntax treatment allowed for (did not inhibit) further development of phonology in T2. Past research investigating this type of cross-domain generalization has found conflicting results. The results of the present study do not clearly support findings of Tyler et al. (2002) that treatment focused on morphosyntax leads to gains in phonology. The lack of clear cross-domain  108  effect seen in this study is possibly more in line with the results of Fey et al. (1994) who found that morphosyntactic treatment alone did not result in changes to the phonological system. Although conflicting evidence exists, the findings of the present study suggest that intervention should target phonology directly when phonological development is a priority. The psycholinguistic perspective predicts that stimulation of one language domain can have effects in many other areas of the language system including phonology (Bernhardt et al., 2010). Although clear changes in the phonological system were not seen after morphosyntax treatment alone (in T1), it is possible that subtle changes were occurring in the phonological systems while the morphosyntactic domain was being stimulated. Also, morphosyntax treatment may have led to growth in other areas of the language system that were not investigated in this study (i.e. literacy, vocabulary, phonological awareness). Question (3). Are morphosyntax-first and phonology-first intervention strategies  effective in facilitating phonological development? The fact that specific changes to each twins’ phonological system occurred after treatment suggests that both intervention strategies are effective in facilitating phonological development. This is consistent with the findings of Tyler et al. (2002) that both morphosyntaxfirst and phonology-first goal attack strategies resulted in overall development of the phonological system. Tyler et al. (2003) found that the phonology-first group experienced slightly less phonological change than the morphosyntax-first group; however, these differences were not significant. In contrast to the findings of Tyler et al. (2003), for the twins in the current study, ordering phonology first allowed for further generalization of phonology goals after morphosyntax treatment, indicating that this may be the more efficient intervention strategy for phonological intervention. This topic should be investigated in future research to uncover any subtle differences in how the ordering of phonology and morphosyntax goals can facilitate phonological development.  109  Question (4). What are the similarities and differences between the two participants and  how do they influence intervention outcomes? Overall, results indicate that whereas the twins were very similar in their phonological systems, they were different in their responses to treatment. Thus, this research question can be answered in two parts. First, looking at twin correlations for phonology gives some insight into twin similarities and second, looking at capability, focus, and treatment outcome gives some insight into twin differences. Correlations for twin phonologies on CAPES word productions were .48 pre-treatment and .59 post-treatment. It appears that twin correlations for articulation in this study were similar to what was found by Matheny and Bruggemann (1973). Their estimate was 0.54 to 0.68 in DZ twins. In terms of differences, results of a capability-focus framework analysis (based on Kwiatkowski and Shriberg, 1998) showed that T1 appeared to have low capability – medium focus with a non-maximal treatment outcome, whereas T2 appeared to have low capability – high focus with a maximal outcome. This means that although both twins were equivalent in capability, T1 scored lower in the area of focus and had a less maximal outcome than T2. This is consistent with the predictions of the capability-focus framework that lower capability and/or lower focus are related to non-maximal outcomes and vice versa. This framework places the importance on the child as an involved learner. It has provided insight into why two children with similar phonological systems and receiving similar intervention may respond differently to treatment. Although the evaluation of outcomes was based on changes seen only in phonology, it must be considered that T1 had received morphosyntax treatment immediately prior to phonology treatment whereas T2 had not. It is important to recognize that this could be a confounding factor, influencing the treatment outcomes independently of capability and focus.  110  Discussion of Phonological Changes Now that the research questions have been revisited, treatment outcomes for each boy will be discussed with regard to possible mechanisms of change and corresponding constraint ranking changes, thus reflecting on the sections concerning phonological theory in Chapter 1. Chapter 3 has outlined the phonological changes for each twin, with match data presented in Tables 15 to 26, percentage match for individual elements illustrated in Figures 5 to 10, and increase in matches summarized in Tables 27 and 28. The changes presented in the Results chapter will be reviewed and interpreted here. T1: Change in Stress Patterns For T1, Stress patterns decreased by three matches after the baseline period and did not increase again until after phonology treatment, at which time they increased by four matches. This finding seems to represent a U-shaped learning curve. This shows that the adult-like constraint ranking of Survived(Syllable) > NotProminent(Foot,Right) was optional rather than being firmly established. It showed temporary reversion back to NotProminent(Foot,Right) > Survived(Syllable) during the baseline period. Phonology treatment may then have led to the re-emergence of the adult-like constraint ranking. The adultlike constraint ranking Survived(Syllable) > Binary(Foot,!) seemed to be established for T1 since he matched the stress pattern Sww. T1: Change in /s/-clusters Very little change in percentage match for /s/-clusters occurred prior to phonology treatment. The small change that was seen may have been due to spontaneous development. Small changes seen after morphosyntax treatment may have been due to cross-domain effects or to cross-twin generalization (since T2 was receiving phonology treatment at this time). Increases in matches after phonology treatment were larger indicating that the constraint ordering  111  Survived(TimingUnit) > NoSequence([+cont]-[-cont] Onset ) was becoming established as a response to treatment. T1: Change in Liquids Matches to /!/ in the treated context did not change throughout the study. This indicates that the constraint ranking NotCo-occurring([Coronal][Labial] Onset ) > Survived(Coronal) was not showing signs of re-ordering. Change in matches to /l/ in all contexts was very minimal until after the phonology treatment period, when a large increase occurred. The reordering of the constraints Not(+lateral Coda ) and Survived(+lateral) after phonology treatment resulted in greater faithfulness to the feature [+lateral] not only in the treated word-final context but also in the word-medial position and in consonant sequences. At the baseline assessment, mismatches within the approximant category (word-initial /!/, word-initial /w/, and word-final /l/) indicated that both twins had lack of clear contrasts within that category. Mismatch patterns for /!/ were interpreted as a highly ranked constraint against the co-articulation of the place feature [Coronal] with [Labial]. Mismatch patterns for /l/ were interpreted as a highly ranked constraint against [+lateral].8 T1: Change in [-anterior] and [-grooved] A regression was seen for the feature [-anterior] for T1 after the baseline period and matches remained low throughout the remainder of the study. Unlike the regression seen for stress patterns where the adult-like constraint ranking was re-established after phonology treatment, the adult-like ranking for [-anterior] was not re-established. For the feature  8  Note that a substitution of [l] for /w/ in word-initial position was observed in connected speech  once again after treatment, indicating that constraints disallowing the co-articulation of [Labial] and [Dorsal] were still highly ranked.  112  [-grooved], no change in matches was observed throughout the study. This indicates neither spontaneous development nor generalization for this feature. In summary, throughout the study, increases in matches and thus re-organization of relevant constraints for T1 occurred for treated stress patterns, both treated and untreated /s/clusters, other clusters, treated word shapes, [CorLab] in untreated contexts, and [+lateral] in both treated and untreated contexts. T2: Change in Stress Patterns For T2 increases in matches for stress patterns indicate that some change was occurring spontaneously during the baseline period. This change then accelerated as a response to phonology treatment. Thus, the adult constraint ranking Survived(Syllable) > NotProminent(Foot,Right), Binary(Foot,!) was becoming established. T2: Change in /s/-clusters Little change in matches for /s/-clusters occurred until after phonology treatment. This indicates a change to the adult ranking of NoSequence([+cont]-[-cont] Onset ) > Survived(Timing Unit) as a response to phonology treatment. T2: Change in Liquids The feature combination [CorLab] showed an increase in matches throughout the study. Increases in matches after baseline indicated that some spontaneous re-ranking of NotCooccurring([Coronal][Labial] Onset ) relative to Survived(Coronal). Further increases in matches occurred after the morphosyntax treatment period, which could be a product of either delayed generalization of what was learned in phonology treatment, cross-domain generalization, or cross-twin generalization. For [+lateral], increases in matches occurred after phonology treatment with further increases after morphosyntax treatment. This indicated a re-ranking of Not(+lateral Coda ) and Survived(+lateral) as a response to phonology treatment. Continued change after  113  morphosyntax treatment was either a delayed generalization from the phonology treatment, cross-domain generalization, or cross-twin generalization. T2: Change in [-anterior] and [-grooved] For T2, there was little change in matches to the untreated features relating to palatoalveolars ([-anterior]) and interdentals ([-grooved]), indicating that neither spontaneous development nor generalization occurred for these elements of the phonological system. In summary, increases in matches and thus re-organization of relevant constraints for T2 occurred for treated stress patterns, treated and untreated /s/-clusters, other clusters, treated and untreated word shapes, [CorLab] in treated and untreated contexts, and [+lateral] in treated and untreated contexts. The OT framework thus provides a way to consider phonological change, in terms of the reranking of constraints. The following section presents some final thoughts about the study, i.e., limitations, possible directions for future research, and clinical implications.  Limitations of the Study The study has several limitations. The first is the small sample size and thus the difficulty in relating results to the population at-large (twins or singletons). The second is a limitation of the design, specifically the constraints that this design placed on our ability to evaluate generalization of goals to connected speech for both twins. The third is a limitation in the assessment procedure and the fourth is goal selection. The final limitation that will be discussed is the measurement protocol for evaluating twin capability and treatment outcome using the capability-focus framework. Including a set of twins for this study was advantageous in many ways. It allowed the phonological outcomes of the similar speech and language interventions presented in a different order to be compared in two individuals who shared many similarities. They shared prenatal and home environment, 50% of their genetic information, and similar phonological systems. However, as in all single case research, it is difficult to evaluate how outcomes in these  114  participants relate to the population at-large. Although it can be concluded that intervention led to desired changes in the phonological systems of these two individuals, the findings have little statistical power and therefore generalizations cannot be made. It was possible, however, to determine whether or not intervention produced a general treatment effect in these two children. With regard to the treatment design, it was considered important to offer each twin morphosyntax and phonology intervention in the opposite order in order to evaluate possible order effects of treatments. However, this difference between the two treatment plans meant that the two participants received phonological intervention at different periods. This was a disadvantage in evaluating generalization of the phonology goals since T2 was given a full 3.5 months to consolidate what was learned in phonological therapy before the final assessment whereas T1 was only given a 3.5 week break. This meant that T1’s phonology was not evaluated after an extended period of time following the therapy, thus making it impossible to know whether any further generalization occurred or even if gains were maintained over a more extended period. Another limitation of this study is that assessment measures for phonology were carried out by the same person who had offered intervention in that area. For this reason, scores on those phonological assessments may have been slightly biased, with the clinician’s presence providing a connection to the treatment context. This may have led the child to produce assessment words more carefully and possibly more accurately. The phonological intervention offered in this study focused more heavily on structural goals than on segmental goals. Two structural goals and one segmental goal with a link to structure were chosen for intervention. The lack of development seen in untreated segments throughout the study leads one to ask whether focusing more heavily on segmental aspects of the phonological systems of these participants would have resulted in greater change in segmental areas of their phonological systems. The rationale for goal selection involved  115  choosing stimulable targets instead of non-stimulable targets in order to ensure that the participants, who had experienced slow progress in past therapy, experienced success. Although past research has found conflicting evidence regarding which goal selection approach is best, goal selection in participants with a history of slow progress has not been investigated. This will be discussed further below. With regard to the capability-focus framework used to evaluate how within-child variables were related to treatment outcomes, some specific issues surrounding measurement of particular variables arose. The twins were equivalent on the measure of capability used, which considered only manner class inventory. It is nonetheless important to mention that T1 had lower percentage matches overall for unacquired elements than did T2 at the baseline assessment. Furthermore, T1 showed a greater increase in percentage match for unacquired elements than T2 directly after phonology treatment, although targets had not yet been generalized. Thus, if capability and outcome were to be evaluated based on percentage matches, a different conclusion about the capabilities and outcomes of the twins would be reached. These measurement issues represent a possible drawback of using the capability-focus framework. However, Kwiatkowski and Shriberg (1998) found that using consonant inventory to measure capability and progression in therapy to measure outcome had predictive validity. The limitations of this study mentioned here all lead to further questions about the nature of protracted phonological development, goal selection, and treatment techniques. Several lines of further research are implicated.  Directions for Future Research This section will highlight a few areas in which further research would aid clinicians in making important decisions regarding phonological intervention. These include: methods of goal selection, goal attack, cross-domain effects of treatment, and generalization within the phonological system.  116  Several studies investigating the effectiveness of a non-linear approach to phonological goal selection exist; however, like the present study, they have used relatively small numbers of participants, and have been largely descriptive. Future studies should use a larger n and randomized control and should aim to establish whether or not a non-linear approach to goal selection leads to more desirable outcomes than other approaches. Further research regarding the efficiency of different goal attack strategies is also needed. Tyler et al. (2003) found that although differences in phonological outcomes were minimal across goal attack strategies, a strategy that alternated morphosyntax and phonology goals weekly was the most effective in facilitating morphosyntactic development. More studies are needed in order to further establish which goal attack strategies are most effective for development in different domains of the language system. Research should also investigate which approach to goal selection is implicated for children with different levels of delay and different histories of progress in past therapy. Goal attack strategy has not been explicitly studied in children who have a history of slow progress in therapy such as the participants in this study. Further research in this area would be useful for making goal attack decisions in children with diverse profiles. Further study of how generalization to untreated structure to segment occurs is also needed. This is especially true when targeting multiple goals in therapy is contraindicated, meaning that priorities must be set for goal selection. The inconsistencies across previous studies concerning potential cross-domain effects between morphosyntax and phonology also suggest the need for further research. It is possible that some variables regarding phonological intervention method or child profile are related to cross-domain generalization. Further research in this area can aid clinicians in choosing intervention methods that are most likely to stimulate development in those language areas in which a child has particular needs.  117  Clinical Implications Although much is still unknown regarding therapy for protracted phonological development, the results of this study support a number of principles to follow when making intervention decisions. These are: (a) treating protracted phonological development directly rather than indirectly, (b) selecting intervention targets while considering both structures and segments, and (c) considering a child’s level of focus as an important variable in treatment outcome. This study found little development in the phonological systems of the participants after therapy focused on morphosyntax. This indicates that clinicians should offer treatment to children with protracted phonological development that targets the phonological system directly when phonological development is a priority. Another finding was that generalization to unacquired segments after therapy did not occur for either twin. Because neither twin received intervention for a purely segmental goal, this implied that clinicians should place equal importance on segment and structure when selecting goals for intervention, even if needs fall more heavily into the structural domain (as was the case for the twins in this study). This is consistent with what has been suggested by Bernhardt and Stemberger (1998, 2000) about goal selection. Kwiatkowski and Shriberg’s (1998) capability-focus framework highlights within-child variables and their relation to treatment outcomes in a way that can aid clinicians in determining prognosis. However, the purpose of this framework is not solely to predict progress in therapy. The clinician might consider trying to modify the child’s level of focus if it is thought to be impeding success. Kwiatkowski and Shriberg suggest that modifying focus can be achieved through adjusting the feedback that is given in therapy. This means tapping into the child’s interests, giving verbal feedback, and being creative about motivating the child to work.  118  Conclusions This study has contributed to the growing body of research about the efficacy of nonlinear phonological intervention by showing that intervention focused on phonology facilitates growth of the phonological system in a set of male dizygotic twins. This growth was greater than that seen after morphosyntax treatment alone, indicating that treatment for protracted phonological development should focus directly on phonology. Future research should use experimental control and larger groups of children and compare treatment outcomes of intervention using a non-linear goal selection method with treatment outcomes using other goal selection methods. Further research is also needed in order to understand optimal goal ordering and cross-domain linguistic effects so that treatment can be designed to benefit the entire linguistic system.  119  Works Cited Almost, D. & Rosenbaum, P. (1998). Effectiveness of speech intervention for phonological disorders: A randomized controlled trial. Developmental Medicine and Child  Neurology, 40, 319-325.  Baker, E. (2006). Management of speech impairment in children: The journey so far and the road ahead. Advances in Speech Language Pathology, 8, 156-163.  Baker, E., Croot, K., McLeod, S., & Paul, R. (2001). Psycholinguistic models of speech  development and their application to clinical practice. Journal of Speech, Language, and  Hearing Research, 44, 685-702.  Baker, E. & McLeod, S. (2004). Evidence-based management of phonological impairment in children. Child Language Teaching and Therapy, 20, 261-285.  Bakker, P. (1987). Autonomous languages of twins. Acta Geneticae Medicae Et Gemellologiae:  Twin Research, 36, 233-238.  Bankson, N. W., Bernthal, J. E., & Flipsen, P. (2009). Remediation procedures. In J. E.  Bernthal, N. W. Bankson, & P. Filipsen (Eds.), Articulation and phonological disorders:  Speech sound disorders in children (pp. 251-277). Boston: Pearson Education Inc.  Bankson, N. & Bernthal, J. E. (1990). Bankson-Bernthal Test of Phonology. Austin, TX: Pro Ed.  Barlow, J. A. & Gierut, J. A. (1999). Optimality theory in phonological acquisition. Journal of  Speech, Language, and Hearing Research, 42, 1482-1498.  Barlow, J. A. & Gierut, J. A. (2002). Minimal pair approaches to phonological remediation.  Seminars in Speech and Language, 23, 57-67.  Bernhardt, B. M. (1990). Application of nonlinear phonological theory to intervention with six phonologically disordered children. Dissertation Abstracts International, 52, 10B.  Bernhardt, B. (1992). The application of nonlinear phonological theory to intervention with one phonologically disordered child. Clinical Linguistics and Phonetics, 6, 123-145.  Bernhardt, B. (1994). Phonological intervention techniques for syllable and word structure development. Clinics in Communication Disorders, 4, 54-65.  120  Bernhardt, B., & Major, E. (2005). Speech, language and literacy skills 3 years later: A follow-  up study of early phonological and metaphonological intervention. International Journal  of Language & Communication Disorders, 40, 1-27.  Bernhardt, B. H. & Stemberger, J. P. (1998). Handbook of phonological development from the  perspective of constraint-based nonlinear phonology. San Diego, CA: Academic Press.  Bernhardt, B. H. & Stemberger, J. P. (2000). Workbook in nonlinear phonology for clinical  application. Austin, TX: Pro Ed.  Bernhardt, M., Stemberger, J., & Charest, M. (2010). Intervention for speech production in children and adolescents: Models of speech production and therapy approaches. Introduction to the issue. Canadian Journal of Speech Language Pathology and  Audiology, 34, 157-167.  Bird, J., Bishop, D. V. M., & Freeman, N. H. (1995). Phonological awareness and literacy  development in children with expressive phonological impairments. Journal of Speech  and Hearing Research, 38, 446-462.  Bleile, K. M. (1995). Manual of articulation and phonological disorders. San Diego, CA: Singular Publishing Group.  Bopp, K. D. (1995). The effects of phonological intervention on morphosyntactic development  in preschool children with phonological and morphosyntactic disorders. Unpublished Master of Science, University of British Columbia, Vancouver, Canada.  Bowen, C. & Cupples, L. (1999). Parents and children together (PACT): A collaborative approach to phonological therapy. International Journal of Language and  Communication Disorders, 34, 35-83.  Bracken, B. A. (1984). Bracken Basic Concept Scale. San Antonio, TX: The Psychological Corporation.  Brown, R. (1973). A first language. Cambridge, MA: Harvard University Press. Brown, L., Sherbenou, R. J., & Johnsen, S. K. (1998). Test of Non-Verbal Intelligence (3rd ed.). Austin, TX: Pro Ed.  Burgemeister, B., Blum, B., & Lorge, I. (1972). Columbia Mental Maturity Scale (3rd ed.). New York: The Psychological Corporation.  Chomsky, N. & Halle, M. (1968). Sound pattern of English. New York: Harper & Rowe.  121  Day, E. J. (1932). The development of language in twins: A comparison of twins and single children. Child Development, 3, 179-199.  Dean, E. & Howell, J. (1986). Developing linguistic awareness: A theoretically based approach  to phonological disorders. British Journal of Disorders of Communication, 21, 223-238.  Dell, G. S. (1986). A spreading-activation theory of retrieval in sentence production.  Psychological Review, 93, 283-321.  Doll, E. A. (1965). Vineland Social Maturity Scale: Condensed manual of directions. Circle Pines, MN: American Guidance Service Inc.  Duffy, J. R. (1995). Motor speech disorders. St. Louis, MO: Mosby. Edwards, M. L. & Bernhardt, B. (1973). Twin speech as the sharing of a phonological system. Unpublished manuscript, Stanford University.  Elliot, C. D., Smith, P., & McCulloch, K. (1996). Verbal Comprehension Scale. British Ability  Scales (2nd ed.). Windsor, England: NFER Nelson.  Fey, M. E. (1986). Language intervention with young children. Boston, MA: Allyn and Bacon. Fey, M. E. (1999). PACT: Some comments and considerations. International Journal of  Language and Communication Disorders, 34, 35-83.  Fey, M. E., Cleave, P., Long, S. & Hughes, D. (1993). Two approaches to the facilitation of grammar in children with language impairment. Journal of Speech and Hearing  Research, 36, 141-157.  Fey, M. E., Cleave, P. L., Ravida, A. I., & Long, S. H. (1994). Effects of grammar facilitation on the phonological performance of children with speech and language impairments.  Journal of Speech & Hearing Research, 37, 594-607.  Fey, M. E. & Stalker, C. H. (1986). A hypothesis-testing approach to treatment of a child with  an idiosyncratic (morpho)phonological system. Journal of Speech & Hearing Disorders,  51, 324-336.  Fudala, J. (1970). Arizona Articulation Proficiency Scale: Revised. Los Angeles: Western Psychological Service.  Gathercole, S. E., Willis, C. S., Baddeley, A. D., & Emslie, H. (1994). The Children’s Test of Nonword Repetition: A test of phonological working memory. Memory, 2, 103-127.  122  Gierut, J. A. (1998). Treatment efficacy: Functional phonological disorders in children. Journal  of Speech, Language, and Hearing Research, 41, 85-100.  Gierut, J. A., Morrisette, M. L., Hughes, M. T., & Rowland, S. (1996). Phonological treatment  efficacy and developmental norms. Language, Speech, and Hearing Services in Schools,  26, 215-230.  Gierut, J. A. & Neumann, H. J. (1992). Teaching and learning /!/: A nonconfound. Clinical  Linguistics and Phonetics, 6, 191.  Gierut, J. A. (1992). The conditions and course of clinically induced phonological change.  Journal of Speech & Hearing Research, 35, 1049-1063.  Gierut, J. A. & Morrisette, M. L. (2005). The clinical significance of optimality theory for phonological disorders. Topics in Language Disorders, 25, 266-280.  Girolametto, L., Pearce, P. S., & Weitzman, E. (1997). Effects of lexical intervention on the  phonology of late talkers. Journal of Speech, Language, and Hearing Research, 40, 338348.  Glogowska, M., Roulstone, S., Enderby, P., & Peters, T. J. (2000). Randomized controlled trial  of community based speech and language therapy in preschool children. British Medical  Journal, 321, 923.  Goldman, R. & Fristoe, M. (1986). Goldman Fristoe Test of Articulation. Circle Pines, MN: American Guidance Service. Goldman, R. & Fristoe, M. (2000). Goldman Fristoe Test of Articulation (2nd ed.). Circle Pines, MN: American Guidance Service.  Hay, D. A. & O’Brien, P. J. (1983). The La Trobe twin study: A genetic approach to the  structure and development of cognition in twin children. Child Development, 54, 317330.  Hay, D. A., Prior, M., Collett, S., & Williams, M. (1987). Speech and language development in preschool twins. Acta Geneticae Medicae Et Gemellologiae: Twin Research, 36, 213223.  Hayiou-Thomas, M. (2008). Genetic and environmental influences on early speech, language and literacy development. Journal of Communication Disorders, 41, 397-408.  123  Hayiou-Thomas, M., Kovas, Y., Harlaar, N., Plomin, R., Bishop, D. V. M., & Dale, P. S.  (2006). Common aetiology for diverse language skills in 4 1/2-year-old twins. Journal  of Child Language, 33, 339-368.  Hodson, B. W. (1983). A facilitative approach for remediation of a child’s profoundly unintelligible phonological system. Topics in Language Disorders, 3, 24-34.  Hodson, B. (1986). The Assessment of Phonological Process-Revised. Danville, IL: Interstate Printers and Publishers Inc.  Hodson, B. W. & Paden, E. P. (1991). A phonological approach to remediation: Targeting  intelligible speech (2nd ed.). Austin, TX: Pro Ed.  Hodson, B. W. (1983). A facilitative approach for remediation of a child’s profoundly unintelligible phonological system. Topics in Language Disorders, 3, 24-34.  Howell, J. & Dean, E. (1991). Treating phonological disorders in children: Metaphon - theory to  practice. San Diego, CA: Singular Publishing Group.  Ingram, D. (1976). Phonological disability in children. London: Edward Arnold Ltd. Kirk, C. & Gillon, G. T. (2007). Longitudinal effects of phonological awareness intervention on morphological awareness in children with speech impairment. Language, Speech, and  Hearing Services in Schools, 38, 342-352.  Klein, E. S. (1996). Phonological/traditional approaches to articulation therapy: A retrospective group comparison. Language, Speech, and Hearing Services in Schools, 27, 314-323.  Kovas, Y., Hayiou-Thomas, M., Oliver, B., Dale, P. S., Bishop, D. V. M., & Plomin, R. (2005). Genetic influences in different aspects of language development: The etiology of language skills in 4.5-year-old twins. Child Development, 76, 632-651.  Kwiatkowski, J. & Shriberg, L. D. (1998). The capability-focus treatment framework for child speech disorders. American Journal of Speech Language Pathology, 7, 27-38.  Law, J., Garrett, Z., & Nye, C. (2004). The efficacy of treatment for children with  developmental speech and language delay/disorder: A meta-analysis. Journal of Speech,  Language, and Hearing Research, 47, 924-943.  Lowe, M. & Costello, A. J. (1976). Manual for the Symbolic Play Test. London: NFER Nelson.  124  Major, E. & Bernhardt, B. (1998). Metaphonological skills of children with phonological  disorders before and after phonological and metaphonological intervention. International  Journal of Language and Communication Disorders, 33, 413-444.  Masterson J. & Bernhardt B. H. (2001). CAPES: Computerized Articulation and Phonology  Evaluation System. San Antonio, TX: The Psychological Corporation.  Matheny, A. P. jr. & Bruggemann, C. E. (1973). Children’s speech: Hereditary components and sex differences. Folia Phoniat, 25, 442-449.  Mayer, M. (1969). Frog where are you? New York: Dial Books for Young Readers. Mayer, M. (1973). Frog on his own. New York: Dial Books for Young Readers. McCarthy, D. (1972). McCarthy Scales of Children’s Abilities. New York: The Psychological Corporation.  McCarthy, J. J. & Kirk, S. A. (1961). The Illinois Test of Psycholinguistic Abilities  (Experimental ed.). Urbana, Ill: Institute for Research in Exceptional Children.  McReynolds, L. V. (1975) Distinctive feature analysis of misarticulations. Baltimore, MD: University Park Press.  Miller, J. F., & Iglesias, A. (2006). Systematic Analysis of Language Transcripts (SALT),  English and Spanish (Version 9) [Computer software]. Madison: Language Analysis Lab, University of Wisconsin-Madison.  Mittler, P. (1976). Language development in young twins: Biological, genetic and social aspects. Acta Geneticae, Medicae Et Gemellologiae, 25, 359-365.  Mittler, P. (1970). Biological and social aspects of language development in twins.  Developmental Medicine & Child Neurology, 12, 741-757.  Olswang, L. B. & Bain, B. A. (1985). Monitoring phoneme acquisition for making treatment withdrawal decisions. Applied Psycholinguistics, 6, 17-37.  Pascoe, M., Stackhouse, J., & Wells, B. (2005). Phonological therapy within a psycholinguistic  framework: Promoting change in a child with persisting speech difficulties. International  Journal of Language & Communication Disorders, 40, 189-220.  Pedersen, N. L., Plomin, R., Nesselroade, J. R., & McClearn, G. E. (1992). A quantitative genetic analysis of cognitive abilities during the second half of the life span.  Psychological Science, 3, 346-353.  125  Plomin, R., DeFries, J., McClearn, G. E., & McGuffin, P. (2008). Behavioral genetics. New York: Worth Publishers.  Powell, T. W., Elbert, M., & Dinnsen, D. A. (1991). Stimulability as a factor in the  phonological generalization of misarticulating preschool children. Journal of Speech and  Hearing Research, 34, 1318-1328.  Prezas, R. F. & Hodson, B. W. (2010). The cycles phonological remediation approach. In A. L. Williams, S. McLeod, & R. J. McCauley (Eds.), Interventions for speech sound  disorders in children (pp. 137-177) Paul H. Brookes Publishing Co.  Prince, A. & Smolensky, P. (1993). Optimality theory: Constraint interaction in generative  grammar. Unpublished manuscript.  Renfrew, C. E. (1997a). Action Picture Test (4th ed.). Bicester, Oxon, England: Winslow Press. Renfrew, C. E. (1997b). Bus Story Test (4th ed.). Bicester, Oxon, England: Winslow Press. Reynell, J. (1977). Reynell Developmental Language Scales (revised ed.). London: NFER Publishing.  Rutter, M. & Redshaw, J. (1991). Growing up as a twin: Twin-singleton differences in  psychological development. Journal of Child Psychology and Psychiatry, 32, 885-895.  Rvachew, S. (1994). Speech perception training can facilitate sound production learning. Journal  of Speech & Hearing Research, 37, 347-357.  Rvachew, S. & Bernhardt, B. M. (2010). Clinical implications of dynamic systems theory for  phonological development. American Journal of Speech-Language Pathology, 19, 34-50.  Rvachew, S. & Nowak, M. (2001). The effect of target-selection strategy on phonological learning. Journal of Speech, Language, and Hearing Research, 44, 610-623.  Rvachew, S., Nowak, M., & Cloutier, G. (2004). Effect of phonemic perception training on the speech production and phonological awareness skills of children with expressive  phonological delay. American Journal of Speech-Language Pathology, 13, 250-263. Sadler, C. (2010). Cross-domain effects of phonology and language intervention on the  morphology and syntax of twin boys with protracted speech and language difficulties. Unpublished manuscript.  SALT Reference Database. (n.d.). Retrieved August 21, 2009, from http://www.saltsoftware.com/salt/downloads/referencedatabases.cfm#  126  Semel, E., Wiig, E., & Secord, W. (2003). Clinical Evaluation of Language Fundamentals (4th ed.). San Antonio, TX: The Psychological Corporation.  Shoaf, K. O., Iyer, S. N., & Bothe, K. A. (2009). Using single-subject experimental design to  implement a nonlinear phonology approach to target selection. Contemporary Issues in  Communication Sciences and Disorders, 36, 77-88.  Shriberg, L. D. & Lohmeier, H. L. (2008). The Syllable Repetition Task (SRT). Technical report. No. 14.  Shriberg, L. D., Lohmeier, H. L., Campbell, T. F., Dollaghan, C. A., Green, J. R., & Moore, C. A. (2009). A nonword repetition task for speakers with misarticulations: The Syllable  Repetition Task (SRT). Journal of Speech, Language and Hearing Research, 52, 11891212.  Shriberg, L. D., Austin, D., Lewis, B. A., & McSweeny, J. L. (1997). The Percentage of  Consonants Correct (PCC) metric: Extensions and reliability data. Journal of Speech,  Language, and Hearing Research, 40, 708-722.  Shriberg, L. D. & Kwiatkowski, J. (1982). Phonological disorders: II. A conceptual framework for management. Journal of Speech & Hearing Disorders, 47, 242-256.  Simmerman, I., Steiner, V., & Pond, R. (1992). Preschool Language Scale-3. San Antonio, TX: The Psychological Corporation.  Song, J. Y., Sundara, M., & Demuth, K. (2009). Phonological constraints on children’s  production of English third person singular -s. Journal of Speech, Language, and  Hearing Research, 52, 623-642.  St. Louis, K. O. & Ruscello, M. R. (2000). Oral Speech Mechanism Screening Examination. Austin, TX: Pro Ed.  Stampe, D. (1979). A dissertation on natural phonology. Unpublished manuscript, Indiana University.  Templin, M. C. & Darley, F. L. (1960). The Templin-Darley Test of Articulation. Iowa, IA: Bureau of Educational Research and Service.  Thelen, E. & Bates, E. (2003). Connectionism and dynamic systems: Are they really different?  Developmental Science, 6, 378-391.  127  Tomasello, M., Mannle, S., & Kruger, A. C. (1986). Linguistic environment of 1- to 2-year-old twins. Developmental Psychology, 22, 169-176.  Trouton, A., Spinath, F. M., & Plomin, R. (2002). Twins early development study (TEDS): A multivariate, longitudinal genetic investigation of language, cognition and behavior problems in childhood. Twin Research, 5, 444-448.  Tyler, A. A., Edwards, M. L., & Saxman, J. H. (1987). Clinical application of two  phonologically based treatment procedures. Journal of Speech & Hearing Disorders, 52, 393-409.  Tyler, A. A., Figurski, G. R., & Langsdale, T. (1993). Relationships between acoustically determined knowledge of stop place and voicing contrasts and phonological treatment progress. Journal of Speech & Hearing Research, 36, 746-759.  Tyler, A. A., Lewis, K. E., Haskill, A., & Tolbert, L. C. (2002). Efficacy and cross-domain effects of a morphosyntax and a phonology intervention. Language, Speech, and  Hearing Services in Schools, 33, 52-66.  Tyler, A. A., Lewis, K. E., Haskill, A., & Tolbert, L. C. (2003). Outcomes of different speech and language goal attack strategies. Journal of Speech, Language, and Hearing  Research, 46, 1077-1094.  Van Riper, C. (1953). A case book in speech therapy. New York: Prentice-Hall. Viding, E., Price, T. S., Spinath, F. M., Bishop, D. V. M., Dale, P. S., & Plomin, R. (2003). Genetic and environmental mediation of the relationship between language and  nonverbal impairment in 4-year-old twins. Journal of Speech, Language, and Hearing  Research, 46, 1271-1282.  Viding, E., Spinath, F. M., Price, T. S., Bishop, D. V. M., Dale, P. S., & Plomin, R. (2004). Genetic and environmental influence on language impairment in 4-year-old same-sex and opposite-sex twins. Journal of Child Psychology and Psychiatry, 45, 315-325.  Von Bremen, V. (1990). Nonlinear phonology intervention with twins. Unpublished master of science, University of British Columbia, Vancouver, Canada.  Weiner, F. F. (1981). Treatment of phonological disability using the method of meaningful  minimal contrast: Two case studies. Journal of Speech & Hearing Disorders, 46, 97-103.  Wiig, E. H., Secord, W., & Semel, E. (1992). Clinical Evaluation of Language Fundamentals-  Preschool. San Antonio, TX: The Psychological Corporation.  128  Wilson, R. S. (1975). Twins: Patterns of cognitive development as measured on the Wechsler  Preschool and Primary Scale of Intelligence. Developmental Psychology, 11, 126-134.  Zimmerman, I., Steiner, V., & Pond, R. (1992). Preschool Language Scale-3. San Antonio, Tx: The Psychological Corporation.  129  Appendices Appendix A: Outcomes Survey Please comment on the changes you have seen in Child since the beginning of the school year. Consider speech, language, communication and other areas of development. Please comment on specific speech sounds or grammatical forms if you have noticed any new sounds or sentences. Give examples of how Child’s speech or language has changed throughout this school year.  130  Appendix B – Case History Form Twin Study: Child and Family Background Information This is an optional questionnaire. You are free to fill out all, some or none of it.  Child’s first initial__________________________________ Today’s Date:________________ Early Development  Did your child babble or coo as a baby?______________________________________________ At what age did cooing/babbling start?_______________________________________________  How old was your child when he began to say words?___________________________________  How old was your child when he began putting two or three words together? ________________ At what age did he sit up alone?_______________________ Crawl?______________________  Walk alone?________________________ Pregnancy and Child’s Health  Did mother have any illnesses during pregnancy? If so, what?____________________________________________ ____________________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________________  Were there any complications during pregnancy? If so, what?_____________________________________________  _____________________________________________________________________________________________________________________________ What was the length of the pregnancy?________________________________________________  Were their any complications during delivery of this child? If so what?_________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ Did this child have a normal Apgar score at birth? Yes _____No____If not, what score? _______ How much did the child weigh at birth?_______________ Was this twin born first or second?_______________ How was the child’s health as a newborn?______________________________________________________________________  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ Has the child had any serious illnesses, high fevers, or serious injuries? If so, what?  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ Has the child had ear infections? If so, how many, for how long, and did he need ear tubes?  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ Has your child had his hearing checked?___________What were the results?  131  ____________________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________________ Has the child’s vision been tested?_______________What were the results?  ____________________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________________ Did the child have any problems with feeding as a baby? Please describe.  ____________________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________________ Did the child have any problems with eating as a toddler? Please describe.  ____________________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________________  ____________________________________________________________________________________________________________________________ How would you describe your child’s eating habits now?__________________________________________________  ____________________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________________ Later Development  Did your child attend preschool? ____________From what age? _______________________________________________ Please list all speech and language services that have been accessed, including therapy, camps, parent programs, etc. Service  Date  Description  132  Describe the pre-literacy and literacy activities that you have done with this child in the home? At what age?  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ Has your child had any assessments besides speech and language? (e.g., psychological testing, occupational therapy). Please list and give dates if known.  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________  _____________________________________________________________________________________________________________________________ Does this child have developmental delays in areas outside speech and language? If so, what are they?  _____________________________________________________________________________________________________________________________  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ Family  Please list all languages spoken by the Mother and Father other than English Mother:_____________________________________________________________________ Father: _____________________________________________________________________ What languages have been spoken to this child? _____________________________________ Please list the highest level of educations of each parent.  Mother:_____________________________________________________________________ Father: _____________________________________________________________________  Please list any sibling, parent or other members of the family who has had language, speech, or learning difficulties. Please specify what these difficulties or differences have been. Relationship to child  Description of difficulty or difference  133  Please describe in your own words this child’s strengths: _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________  _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________________ THANK YOU VERY MUCH!  May Bernhardt, Paola Colozzo, Angela Feehan, Charmaine Sadler  134  Appendix C - Match Percentages and Ratios for Unacquired Elements T1 Overall Match for Unacquired Elements across Assessment Points Element type Treated Direct  generalization Indirect generalization  Pre-Baseline  Baseline  Post Morph Tx  Post Phon Tx  30.4% (7/23)  21.7% (5/23)  34.8% (8/23)  82.6% (19/23)  18.9% (7/37)  29.7% (11/37)  40.5% (15/37)  86.5% (32/37)  28.9% (11/38)  5.3% (2/38)  5.3% (2/38)  18.4% (7/38)  T2 Overall Match for Unacquired Elements across Assessment Points Element type  Pre-Baseline  Baseline  Post Phon Tx  Post Morph Tx  Treated  47.8% (11/23)  47.8% (11/23)  78.3% (18/23)  78.3% (18/23)  21.4% (6/28)  32.1% (9/28)  50% (14/28)  75% (21/28)  8.3% (3/36)  7.9% (3/38)  13.2% (5/38)  15.8% (6/38)  Direct  generalization Indirect  generalization  135  Appendix D – Transcriptions of CAPES 9 T1 Productions: CAPES Profile List Word  Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  pig  p!"g  p!"k  p!"k  p!"k  p!"g  toe  t!o#  t!o#  t!o#  t!o#  t!o#  toes  t!o#z  t!o#z  t!o#z  t!o#z  t!o#z  t!$p  t!$p  tub (bathtub)  9  Adult  t!$b (%bæ&'t!$b)  t!$b  %bæf't!$p (bathtub)  duck  d$k  d$k  d$k  d$k  d$k  mommy  m(mi  m(mi  m(mi  m(mi  m(mi  yard  j()d  j*)d  j()t  j()t  j()d  gum  g$m  g$m  g$m  g$m  g$m  nose  no#z  no#+z  no#z  no#z  no#z  fish  f",  f"s  f"s  f"s  f"s  fishing  %f","-  %f"s"n  %f"s"n  %f"s"n  %f"s"n  soup  so#p  so#p  so#p  so#p  so#p  whistle  %w"s.l  %w"sou  %w"s/  %w"s/  %w"s.l  house  ha#s  ha#s  ha#s  ha#s  ha#s  zipper  %z"p/  %s"p.  %sip/  %s"p/  %s"p/  shoe  ,u  ,u  su  su  su  van  væn  væn  væn  væn  væn  TV  't!i%vi  't!i%wi  't!i%wi  't!i%wi  't!i%wi  chicken  %0"kn1  %0"kn1  %t!"kn1  %t!"kn1  %t!"kn1  watch  w(0  w(,  w(s  w(s  w(s  watches  w(0.z  w(,.z  w(siz  w(s.z  w(siz  jam  2æm  2$m  dæm  3æm  2æm  cage  k!e"2  k!e"z  k!e"z  k!e"z  k!e"z  book  b#k  buk  b#k  buk  buk  pages  %p!e"2.z  %p!e"d.z  %p!e"z.z  %p!e"3.z  %p!e"z.z  laugh  læf  l$f  læf  læf  læf  ribbon  %)"bn1  %w"n.n1  %wibn1  %w"bn1  %wibin  Nasalization of vowels before nasals is not marked but is assumed.  136  T1 Productions: CAPES Profile List (continued) Word  Adult  Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  thumb  !"m  f"m  f"m  f"m  f"m  teeth (tooth)  ti! (tu!)  tif  tif  tuf (tooth)  tif  that  #æt  dæt  dæt  dæt  dæt  feather  $f%#&  $f%'w(  $f%v&  $f%v&  $f%v&  star  st)*  t)*  t)*  slide  sla+d  s,a+d  s,a+t  s,a+d  sla+d  swimming  $sw+m+-  $f+m+-  $f+m+-  $f+m+-  $sw+m+-  queen  kwin  gwin  kwin  kwin  kwin  present  $p*%zn.t  $pw%zit  $p%z(t  $p*%zit  $pw%zit  tree  t*i  /i  twi  twi  twi  glove  gl"v  g"v  gw"v  gw"v  gw"v  kangaroo  0k1æ2-g($*u  0k1"g(+$wu  0k1æ2nd($wu  0k1æ2-g($wu  0k1"ndju  computer  k1(m$p1ju3&  k1($p143(  $p143&  $p1+3&  k1(m$p+3&  hand  hænd  hæn  hæn  hænd  hænd  canoe  k1($nu  $k1æ0nu  k1($nu  k1($nu  k1($nu  again  ($g%n  ($g%n  g%n  g%n  ($g%n  balloon  b($lun  bi$jun  bwun  bwun  b($lun  gorilla  g($*+l(  gi$w+l(  $gw"l(  $gw"l(  g($w+l(  muffin  $m"fn.  $m"fin.  $m"fn.  $m"fn.  $m"fn.  Post-  Post-  st)*  T1 Productions: CAPES List 2 Word  Adult  Production  Pre-Baseline  Baseline  Morphosyntax  Phonology  Treatment  Treatment  pick  p1+k  p1+k  p1+k  p1+k  p1+k  dog  d)g  d)g  d)g  d)k  d)g  comb  k1o4m  k1o4m  k1o4m  k1o4m  k1o4m  face  fe+s  fe+s  fe+s  fe+s  fe+s  faces  $fe+s(z  $fe+siz  $fe+s(z  $fe+siz  $fe+siz  coughing  $k1)f+-  $k1o4f+-  $k1)f+n  $k1)f+-  $k1)f+n  soup  sup  5up  sup  sup  sup  chair  /%*  /%u  t1iu  t1%u  /%u  137  T1 Productions: CAPES List 2 (continued) Word  Adult  Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  horse  h!"s  h#"$  h!"s  h!"s  h#"s  zebra  %zib"&  %sib&  %zibw&  %sibw&  %zibw&  shirt  '(t  '(t  s(t  s(t  s(t  vest  v)st  f)s  f)s  w)st  v)st  diving  %da*v*+  %da*f*n  %da*v*n  %da*f*+  %da*v*n  jacket  %,æk*t  %,ægit  %dægit  %,æk*t  %-ækit  beach  bi.  bis  bis  bis  bis  watching  %w#.*+  %w#s*n  %w#s*n  %w#sin  %w#s*n  jump  ,/mp  -/b*n  -/p  ,/p  d/p  (jumping)  (%,/mp*+)  (jumping)  catch  k0æ.  k0æs  k0æs  k0æs  k0æs  bag  bæg  bæg  bæk  bæg  bæg  bridges  %b"*,&z  %b*1&z  %bw*z&z  %bw*-&z  %bw*z&z  leaf  lif  lif  lif  lif  lifs  rabbit  %"æb*t  %w/bid  %wæbit  %wæbit  %wæbit  garbage  %g#"b*,  %g#biz  %g#b*z  %g#b*-  %g#b*z  thing  2*+  fin  f*n  f*n  f*+  mouth  ma32  m)3f  ma3f  ma3f  ma3f  this  4*s  dis  d*s  d*s  d*s  brother  %b"/4(  %b/v(  %b/v(  %b"/v(  %b"/v(  glasses  %glæs*z  %gæs*z  %gæs*z  %gwæsiz  %gwæs*z  shelf  ')lf  '/f  s)wf  s)wf  s)lf  thousand  %2a3zn5d  %fa3zn5z  %fa3sn5  %fa3sn5  %fa3zn5d  planting  %plænt*+  %pwæt*n  %pwænt*n  %pæt*+  %plænt*n  whisper  %w*sp(  (whispering)  (%w*sp(*+)  %wisp&  %w*fp(  crush  k"/'  kw/s  kw/s  kw/s  kw/s  ski (skis)  ski (skiz)  ki  ki  ski6  kiz (skis)  seal  sil  s3w&u  siu  siu  siu  spaceship  %spe*'*p  %pe*'*p  %pe*s*p  %pe*s*p  %pe*'*p  blue  blu  bju  bwu  bwu  bwu  pickle  p0*kl5  p0ik&u  p0*kr5  p0*kr5  p0*kl5  %w*sp(*+ (whispering)  %w*sp(  138  T1 Productions: CAPES List 2 (continued) Word  Adult  Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  spool  spul  piu  pou  pu  spul  dragon  !drægn"  !dre#gin  !dægn"  !dwægn"  !drægin  Post-  Post-  T1 Productions: Probe Word List Word  Adult  Production  Pre-Baseline  Baseline  Morphosyntax  Phonology  Treatment  Treatment  jelly  !$%li  -  !d%li  !d%li  !d%li  rope  &o'p  -  wo'p  wo'p  wo'p  battery  !bæt(i  -  !bæwiwi  !bæ)*wi  !bæ+*wi  store  st,&  -  t,&  t,&  st,&  spill  sp#l  -  p#u  p#u  p#l  crashing  !k&æ-#.  -  !kwæsi  !kwæsin  !kwæ-#n  stove  sto'v  -  to'v  to'v  to'v  splash  splæ-  -  pwæs  bwæs  bwæ-  reading  !&i+#.  -  !wi+#n  !wi+in  !widin  sneeze  sniz  -  niz  !nisin  (sniz#.)  mash  mæ-  -  mæs  mæs  mæs  run  &/n  -  w/n  w/n  w/n  sponge  sp/n$  -  pw/n  p/n0  p/nz  umbrella  *m!b&%l*  -  !bw%l*  *!bw%l*  *m!bw%l*  snoring  !sn,&#.  -  !n,w#.  !n,w#n  !sn,& (snore)  pigeon  !p1#$n"  -  !p1#0n"  !p1#zn"  !p1#zn"  shadow  !-æ+o'  -  !sæ+o'  !sæ+o'  !sædo'  badge  bæ$  -  bæ0  bæs  bæz  ghost  go'st  -  go's  go's  go's  (snore)  (sn,&)  (sneezing)  niz  (sneezing)  139  T2 Productions: CAPES Profile List Word  Adult Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  pig  p!"g  p!"k  p!"k  p!"g  p!"g  toe  t!o#  t!o#  t!o#  t!o#  t!o#  toes  t!o#z  t!o#z  t!o#z  t!o#z  t!o#z  tub  t!$b  t!$p  t!$p  t!$b  t!$b  duck  d$k  d$k  d$k  d$k  d$k  mommy  m%mi  (mom)  (m%m)  m%mi  m%mi  m%m (mom)  m%mi  yard  j%&d  j%&t  j%&t  j%&d  j%&d  gum  g$m  g$m  g$m  g$m  g$m  nose  no#z  no#z  no#s  no#z  no#z  fish  f"'  fis  fis  f"s  fis  fishing  (f"'")  (f"s")  (f"s")  (f"s"n  (f"s"n  soup  so#p  so#p  so#p  so#p  so#p  whistle  (w"s*l  (w"st+  (wis+  (w"s+  (w"s+  house  ha#s  ha#s  ha#s  ha#s  ha#s  zipper  (z"p+  (sip+  (sip+  (s"p+  (s"p+  shoe  'u  su  su  su  su  van  væn  væn  fæn  fæn  væn  TV  ,t!i(vi  ,t!i(vi  ,t!i(wi  ,t!i(wi  ,t!i(vi  chicken  (-"kn.  (t!"kn.  (t!ikin  (t!"kn.  (t!"kn.  watch  w%-  w%s  w%s  w%s  w%s  watches  w%-*z  -  w%siz  w%siz  w%siz  jam  /æm  -  dæm  dæm  dæm  cage  k!e"/  k!e"z  k!e"z  k!e"z  k!e"z  book  b#k  buk  b#k  b#k  buk  p!e"z (page)  (p!e"ziz  (p!e"z*z  (p!e"z (page)  pages (page)  (p!e"/*z (p!e"/)  laugh  læf  læf  læf  læf  læf  ribbon  (&"bn.  (wibin  (wibin  (w"bn.  (w"bn.  thumb  0$m  f$m  f$m  f$m  f$m  teeth  ti0  tif  tif  tif  tif  140  T2 Productions: CAPES Profile List (continued) Word  Adult  Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  that  !æt  dæt  dæt  dæt  dæt  feather  "f#!$  "f#d$  "f#d$  "f#d$  "f#d$  star  st%&  t%&  t%&  st%&  st%&  slide  sla'd  swa'd  s(a'd  sla'd  sla'd  swimming  "sw'm')  "fwimi)  "s'm')  "sw'm')  "sw'm')  queen  kwin  win  kwin  kwin  kwin  present  "p&#zn*t  "p&#zit  "p&#zit  "pw'z+t  "p&#zent  tree  t&i  twi  t&i  twi  t&i  glove  gl,v  gw,v  gw,v  gw,v  gw,v  kangaroo  -k.æ/)g+"&u  -k.æ/)g+"ju  -k.æ/n+"wu  -k.'n+"wu  -k.æ/n+"wu  computer  k.+m"p.ju0$  "p.ju0$  "p.ju0$  k.+m"p.ju0$  k.+m"p.ju0$  hand  hænd  hænd  hænt  hænd  hænt  canoe  k.+"nu  k.+"nu  k.+"nu  k.+"nu  k.+"nu  again  +"g#n  +"g#n  +"g#n  +"g#n  +"g#n  balloon  b+"lun  bjun  blun  b+"lun  blun  gorilla  g+"&'l+  "gw'l+  gwi"l+l+  g+"w'l+  gw+"l'l+  muffin  "m,fn*  "m,fn*  "m,fin  "m,fn*  "m,fn*  Post-  Post-  T2 Productions: CAPES List 2 Word  Adult  Production  Pre-Baseline  Baseline  Morphosyntax  Phonology  Treatment  Treatment  pick  p.'k  p.ik  p.ik  p.'k  p.'k  dog  d%g  d%g  d%k  d%g  d%g  comb  k.o1m  k.o1m  k.o1m  k.o1m  k.o1m  face  fe's  fe's  fe's  fe's  fe's  faces  "fe's+z  "fe's+z  "fe'siz  "fe's+z  "fe's+z  coughing  "k.%f')  "k.%f')  "k.%f')  "k.%f')  "k.%f'n  soup  sup  sup  sup  sup  sup  chair  2#&  t.#&  t.#&  t.#&  t.#&  horse  h3&s  h3&s  h3&s  h3&s  h3&s  zebra  "zib&+  "zibl+  "zib&+  "zib&+  "zib&+  141  T2 Productions: CAPES List 2 (continued) Word  Adult  Production  Pre-Baseline  Baseline  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  shirt  !"t  !"t  s"t  s"t  s"t  vest  v#st  f#s  v#s  f#st  v#s  diving  $da%v%&  $da%n%&  $da%f%n  $da%f%n  $da%v%&  jacket  $'æk%t  $dækit  $dækit  $dækit  $dæk%t  beach  bi(  bis  bis  bis  bi)  watching  $w*(%&  $w*s%&  $w*s%n  $w*s%n  $w*s%&  jump  '+mp  $d+b%&  d+p  d+mp  d+mp  (jumping)  ('+mp%&)  (jumping)  catch  k,æ(  k,æ-  k,æs  k,æs  k,æ)  bag  bæg  bæg  bæk  bæg  bæg  bridges  $b.%'/z  (bridge)  (b.%')  bliz  $b.%siz  b.iz (bridge)  $b.iziz  leaf  lif  lif  lif  lif  lif  rabbit  $.æb%t  $.æbit  $.æbit  $.æbit  $.æbit  garbage  $g*.b%'  $g*biz  $g*biz  $g*b%z  $g*b%z  thing  )%&  fin  f%&  f%n  fin  mouth  ma0)  ma0f  ma0f  ma0f  ma0f  this  1%s  dis  d%s  d%s  d%s  brother  $b.+1"  $bl+v"  $b.+v"  $b.+z"  $b.+v"  glasses  $glæs%z  $gw+siz  $gwæsiz  $gwæsiz  $glæs%z  shelf  !#lf  s#.f  se%.f  s#.f  s#lf  thousand  $)a0zn2d  $fa0sint  $fa0sin  $fa0zn2  $fa0zn2  planting  $plænt%&  $plæti&  $pænt%n  $plænt%&  plænt%n  whisper  $w%sp"  (whispering)  ($w%sp"%&)  $wip"  $w%sp"  crush  k.+!  kw+s  kw+s  kw+s  kw+s  ski (skis)  ski (skiz)  ki  kiz (skis)  kiz (skis)  ski  seal  sil  si.  si.  si.  si.  spaceship  $spe%!%p  $pe%sip  $be%sip  $pe%s%p  $pe%!%p  blue  blu  bju  b.u  blu  blu  pickle  p,%kl2  p,%k"  p,%k"  p,%k"  p,ikl2  spool  spul  pu  bu  spul  spul  $w%s/p%& (whispering)  w%sp"  142  T2 Productions: CAPES List 2 (continued) Word dragon  Adult  Production  Pre-Baseline  !drægn"  !dre#gin  Baseline !drægin  Post-  Post-  Morphosyntax  Phonology  Treatment  Treatment  !drægn"  !drægn"  Post-  Post-  Morphosyntax  Phonology  !d%li  !d%li  T2 Productions: Probe Word List Word  Adult Production  Pre-Baseline  Baseline  Treatment  jelly (jelly  !$%li  beans)  (!$%libinz)  rope  &o'p  -  wo'p  wo'p  wo'p  battery  !bæ()i  -  !bæ(*wi  !bæ()i  !bæ()i  store  st+&  -  t+&  st+&  st+&  spill  sp#l  -  p#&  sp#&  sp#&  crashing  !k&æ,#-  -  !kwæs#-  !kwæs#-  !kwæs#n  stove  sto'v  -  to'v  sto'v  sto'v  splash  splæ,  -  blæs  plæs  splæs  reading  !&i(#-  -  !wi(in  !wi(#n  !&i(#n  sneeze  sniz  -  niz  -  !d%libinz (jelly  Treatment  beans)  !niz#n  (sneezing)  sniz  (sneezing)  (!sniz#-)  mash  mæ,  -  mæs  mæs  mæ,  run  &.n  -  &.n  w.n  &.n  sponge  sp.n$  -  p.n/  p.nz  sp.nz  umbrella  *m!b&%l*  -  !bl%l*  !b%l*  !b%l*  snoring  !sn+&#-  -  !n+#-  !sn+w#-  !sn+w#-  pigeon  !p0#$n"  -  !p0#1n"  !p0#2n"  !b#zn"  shadow  !,æ(o'  -  !sæ(o'  !sæ(o'  !sæ(o'  badge  bæ$  -  bæ3  bæ2  bæz  ghost  go'st  -  go's  go's  go's  143  T1 Minimal Pair Results Minimal Pairs stick – tick  Baseline !!!  Post Morph Tx !!!  Post Phon Tx !!!  snail – nail  !!!  !!!  !!!  shine – sign  !!!  !!!  rip – lip  X!X! !!!  !!!  !!!  ring – wing  !!!  !!!  !!!  jeep – deep  !!!  !!!  !!!  balloon – loon  X!!  !!X  !!!  guitar – tar  !!XX  XXX!  XX!  stick – tick  Baseline !!!!  Post Phon Tx !!!  Post Morph Tx !!!  snail – nail  !!!!  !!!  !!!  shine – sign  !!!!  !!!!  !!!  rip – lip  !!!!  !!!  !!!  ring – wing  !!!!  !!!!  !!!  jeep – deep  !!!!  !!!  balloon – loon  !!!!  !!X !!!!  guitar – tar  !!XX  !X!!  !X!  T2 Minimal Pair Results Minimal Pairs  !!!  144  Appendix E – Summary of Acquired and Unacquired Non-Linear Elements: Pre-Baseline Assessment Pre-Baseline Assessment: Summary of Acquired Structures Acquired Structures  T1  T2  Length  1 and 2-syllable words  1 and 2-syllable words  Stress  Sw  /!"#b$n/  Sw  /!"#b$n/  wS  /k$!nu  wS  /k$!nu/  swS /%kæ&g$!"u/ wSw /g$!"#l$/  Word Shape  Sequences  swS /%kæ&g$!"u/  CVC  CVC  CVCV, CVCVC  CVCV, CVCVC, CCVCVC  1. Clusters: [-cont]-/w/  1. Clusters: [-cont]-  (4/5), CVCCVCV (1/1)  /gw/, /kw/, /pw/  [+son][+continuant]  2. [Coronal]-[Coronal]: /"t/, /"d/  3. Other: /d"/  /gw/, /kw/, /p"/, /d"/, /bl/, /pl/, /bj/, /pj/  2. [Coronal]-[Coronal] /"t/, /"s/  3. [+nasal]-[-cont] /nt/, /nd/, /&g/ Positional Strengths  WI, WM [+voice]-[-cont]  WI, WM [+voice]-[-cont]  WI [-anterior] /'/,/(/, /)/  WF /"/  WI [+lateral] /l/  WM [+lateral] /l/  Pre-Baseline Assessment: Summary of Needed Structures Needed Structures  T1  T2  Length  3+-syllable words  3+-syllable words  Stress  wS - /k$!nu/  wS - /k$!nu/  Word Shape  CCVC, CCVCVC,  CCVC,  CCVCVCC, CCVCCVC,  CCVCVCC, CCVCCVC,  CVCCCVCVC  CVCCCVCVC  CVCCVC, CVCCVCV,  wSw - /k$m!pju*$"/  CVCCVC,  145  Pre-Baseline Assessment: Summary of Needed Structures (continued) Needed Structures  T1  T2  Sequences  1. s-C  1. s-C  /st/, /sn/, /sp/, /sl/, /sw/,  /st/, /sn/, /sp/, /sl/, /sw/,  /sk/  /sk/  2. [-cont]-[+lateral] /gl/, /bl/, /pl/  3. [-cont]-[CorLab] /k!/, /b!/, /p!/, /t!/  4. [+nasal]-[-cont]  /nd/, /nt/, /n"/, /mp/, /mbr/  5. [Coronal]-[+cons] /!b/, /!s/  Positional Restrictions  2. [-cont]-[+lateral] /gl/, /bl/  3. [-cont]-[CorLab] /k!/, /b!/, /t!/  4. [+nasal]-[-cont] /nd/, /nt/, /mp/, mb!/  5. [Coronal]-[-cont] /!b/, /!d/  6. Other: /lf/  6. Other: /lf/  WM, WF [+lateral] /l/  WM, WF [+lateral] /l/  WM, WF [-anterior] /#/,/"/,  WIT, WM /!/  /$/  Pre-Baseline Assessment: Summary of Acquired Features Acquired Features  T1 [+nasal]  T2  [-cont]  /m/, /n/  /t/, /g/  [+nasal] /m/, /n/  [Labial]  /p/, /b/, /m/  [Labial] /p/, /b/, /m/  [Coronal]  /s/, /z/, /t/, /n/  [-cont]  /t/, /g/  [Coronal] /s/, /z/, /t/, /n/  [-grooved] /s/, /z/  [-grooved] /s/, /z/  [Dorsal] (/k/, /g/)  [Dorsal]  Feature combinations  Feature combinations  [+anterior] /t/, /n/  excepting needed features.  [+anterior] /t/, /n/ /k/, /g/  excepting needed features.  Pre-Baseline Assessment: Summary of Needed Features Needed Features  T1  T2  [-grooved]  /%/, /&/  [+lateral]  /l/  [-cont]-[+cont] /#/,/"/ [CorLab] - syllable initial /!/  [-grooved] /%/, /&/  [-anterior] /#/,/"/, /$/ [-cont]-[+cont] /#/,/"/  [CorLab] - syllable initial /!/  146  Appendix F – Summary of Acquired and Unacquired Non-Linear Elements: Baseline Assessment Baseline Assessment: Summary of Acquired Structures Acquired Structures  T1  T2  Length  1 and 2-syllable words  1 and 2-syllable words  Stress  Sw  Sw  Word Shape  CVC  CVC  CVCV, CVCVC, CCVCCVC  CVCV, CVCVC  1. Clusters: [-cont]-/w/  1. Clusters: [-cont]-/w/  2. [Coronal]-[Coronal]  2. [-cont]-/"/  /!"#b$n/  swS /kæ%g$&!"u/  (1/1), CVCCVCV (1/1) Sequences  /gw/, /kw/, bw/, /pw/, /tw/ /"t/, /"s/  /!"#b$n/  swS /kæ%g$&!"u/  /gw/, /kw/  /b"/, /p"/, /t"/, /d"/  3. [Coronal]-[Coronal] /"t/, /"s/  4. [Coronal][+nasal][Coronal][-continuant] /nt/, /nd/ Positional Strengths  WI, WM [+voice]-[-cont]  WI, WM [+voice]-[-cont]  WF /"/  WF /"/  WI, WM [+lateral] /l/  WI, WM [+lateral] /l/  Baseline Assessment: Summary of Needed Structures Needed Structures  T1  T2  Length  3+-syllable words  3+-syllable words  Stress  wS  wS  Word Shape  CCVC, CCVCC, CCCVC  CCVC, CCVCC, CCCVC  CCVCVC, CCVCVCC,  CCVCVC, CCVCVCC,  /k$!nu/  wSw /k$m!pju'$"/)  CVCCVC,  wSw  /k$!nu/)  /k$m!pju'$"/)  CVCCVC, CCVCCVC, CVCCVCV, CVCCCVCVC,  CVCCCVCVC, VCCCVCV  VCCCVCV,  147  Baseline Assessment: Summary of Needed Structures (continued) Needed Structures  T1  T2  Sequences  1. s-C  1. s-C  /st/, /sn/, /sp/, /sl/, /sw/,  /st/, /sn/, /sp/, /sl/, /sw/,  /sk/  /sk/  2. [-cont]-[+lateral] /gl/, /bl/, /pl/  3. [+nasal]-[-cont] (e.g.,  2. [-cont]-[+lateral] /gl/, /bl/, /pl/  3. [+nasal]-[-cont]  /nd/, /nt/, /n!/, /"g/, /mp/,  /nd/, /n!/, /"g/, /mp/,  /mbr/, /mpj)  4. [-cont]-[CorLab]  /k#/, /b#/, /p#/, /d#/, /t#/  mb#/, /mpj/  4. Other: /lf/, /kr/  5. [Coronal]-[-cont] /#b/, /#d/  6. Other: /lf/ Positional Restrictions  WI, WM /#/  WM, WF [+lateral] /l/  WI, WM /#/  WI, WFT [+lateral] /l/  Baseline Assessment: Summary of Acquired Features Acquired Features  T1  T2  [+nasal]  /m/, /n/  [+nasal]  /m/, /n/  [-cont]  /t/, /g/  [-cont]  /t/, /g/  Coronal  /s/, /z/, /t/, /n/  Coronal  /s/, /z/, /t/, /n/  Labial  [-grooved] [+anterior] Dorsal  /p/, /b/, /m/  /s/, /z/  /t/, /n/  /k/, /g/  Labial  [-grooved] [+anterior] Dorsal  /p/, /b/, /m/  /s/, /z/  /t/, /n/  /k/, /g/  Feature combinations  Feature combinations  excepting needed features.  excepting needed features.  Baseline Assessment: Summary of Needed Features Needed Features  T1 [-anterior]  T2 /$/,/!/, /%/  [-grooved] /&/, /'/  [-cont]-[+cont] /$/,/!/  [CorLab] syllable-initial /#/  [-anterior]  [-grooved]  /$/,/!/, /%/ /&/, /'/  [-cont]-[+cont] /$/,/!/  [CorLab] syllable-initial /#/  148  Appendix G – [CorLab] Match Data by Item T1 Match data: [CorLab] List  Word Position  Item  Pre-Baseline  Baseline  Post-Morph  Post-Phon  Tx  Tx  CAPES  WI  !"#bn$  w  w  w  w  CAPES  WI  !"æb#t  w  w  w  w  Probes  WI  "o%p  -  w  w  w  Probes  WI  !"i&#'  -  w  w  w  Probes  WI  "(n  -  w  w  w  CAPES  WM  g)!"#l)  w  w  w  w  CAPES  WM  *k+æ,'g)!"u  w  w  w  j  Probes  WM  !bæt-i  -  w  w  w  Probes  WM  !sn."#'  -  w  w  -  Item  Pre-Baseline  Baseline  T2 Match Data: [CorLab] List  Word  Position  Post-Phon  Post-  Tx  Morph Tx  CAPES  WI  !"#bn$  w  w  w  w  CAPES  WI  !"æb#t  "  "  "  "  Probes  WI  "o%p  -  w  w  w  Probes  WI  !"i&#'  -  w  w  "  Probes  WI  "(n  -  "  w  "  CAPES  WM  g)!"#l)  w  w  w  w  CAPES  WM  *k+æ,'g)!"u  j  w  w  w  Probes  WM  !bæt-i  -  "  "  Probes  WM  !sn."#'  -  w !  w  w  149  Appendix H – [+Lateral] Match Data by Item T1 Match Data: [+lateral] List  Word  Position  Item  Pre-Baseline  Baseline  Post-  Post-Phon  Morph Tx  Tx  CAPES  WM  b!"lun  j  w  w  !  CAPES  WM  g!"#$l!  !  !  !  !  Probes  WM  "%&li  -  !  !  !  CAPES  WF  "w$s!l  u  #  #  !  CAPES  WF  sil  u  u  u  u  CAPES  WF  p'$kl(  u  #  !  CAPES  WF  spul  u  u  # !  Probes  WF  sp$l  -  u  u  !  Item  Pre-Baseline  Baseline  Post-  Post-Phon  !  T2 Match Data [+lateral] List  Word  Position  Morph Tx  Tx  CAPES  WF  "w$s!l  #  #  #  #  CAPES  WF  sil  #  #  #  #  CAPES  WF  p'$kl(  !  WF  spul  # !  #  CAPES  # !  !  !  Probes  WF  sp$l  -  #  #  #  150  Appendix I – Stress Pattern Match Data by Item T1 Match Data: Stress Patterns List  Stress  Pattern  Item  Pre-Baseline  Baseline  Post-  Post-Phon  Morph Tx  Tx  CAPES  wSw  k!"m#p!ju$%  !  Sw  Sw  !  CAPES  wSw  g"#&'l"  !  Sw  Sw  !  Probes  wSw  "m#b&(l"  -  Sw  !  !  CAPES  wS  k!"#nu  Ss  !  !  !  CAPES  wS  "#g(n  !  S  S  !  CAPES  wS  b"#lun  !  S  S  !  Probes  Sww  #bæt%i  -  !  !  !  Item  Pre-Baseline  Baseline  Post-Phon  Post-  T2 Match Data: Stress Patterns List  Stress  Pattern  Tx  Morph Tx  CAPES  wSw  k!"m#p!ju$%  Sw  Sw  !  !  CAPES  wSw  g"#&'l"  Sw  !  !  !  Probes  wSw  "m#b&(l"  -  Sw  Sw  Sw  CAPES  wS  k!"#nu  !  !  !  !  CAPES  wS  "#g(n  !  !  !  !  CAPES  wS  b"#lun  S  S  !  S  Probes  Sww  #bæt%i  -  !  !  !  151  Appendix J – Word Shape Match Data by Item T1 Match Data: Word Shapes Pre-  Post-Phon  Word Shape  Item  CAPES  CCVC  st!"  CVC  CVC  CAPES  CCVC  sla#d  CVC  CVC  CVC  !  CAPES  CCVC  kwin  !  !  !  !  CAPES  CCVC  gl$v  CVC  !  !  !  CAPES  CCVC  kw$%  !  !  !  !  CAPES  CCVC  spul  CVC  CVC  CV  !  Probes  CCVC  st&"  -  CVC  CVC  !  Probes  CCVC  sp#l  -  CVC  CVC  CVC  Probes  CCVC  sto'v  -  CVC  CVC  CVC  Probes  CCVC  sniz  -  CVC  CVCVC  CVC  Baseline  Baseline  Post-  List  Morph Tx  Tx !  CAPES  CCVCVC  (sw#m#)  CVCVC  CVCVC  CVCVC  !  CAPES  CCVCVC  (b"#*+z  CVCVC  !  !  !  CAPES  CCVCVC  (b"$,-  CVCVC  CVCVC  !  !  CAPES  CCVCVC  (glæs#z  CVCVC  CVCVC  !  !  CAPES  CCVCVC  (spe#%#p  CVCVC  CVCVC  CVCVC  CVCVC  CAPES  CCVCVC  (d"ægn.  !  CVCVC  !  !  Probes  CCVCVC  (kræ%#)  -  CCVCV  !  !  Probes  CCVCVC  (sn&"#)  -  CVCVC  CVCVC  !  CAPES  CCVCVCC  (p"/zn.t  CCVCVC  CVCVC  CCVCVC  CCVCVC  CAPES  CVCCVC  (g!"b#*  CVCVC  CVCVC  CVCVC  CVCVC  CAPES  CVCCVC  (w#sp-  CVCCV  !  !  !  CAPES  CCVCCVC  (plænt#)  CCVCVC  !  CVCVC  !  CAPES  CVCCVCV  k0æ)g+("u  CVCVCV  !  !  CVCCCV  CAPES  CVCCCVCVC  k0+m(p0ju1-  CVCVCV  CVCVC  CVCVC  CVCCVCVC  Probes  VCCCVCV  +m(b"/l+  -  CCVCV  VCCVCV  !  Probes  CCVCC  sp$n*  -  CCVC  CVCC  CVCC  Probes  CCCVC  splæ%  -  CCVC  CCVC  CCVC  152  T2 Match Data: Word Shapes Pre-  List  Word Shape  Item  CAPES  CCVC  st!"  CVC  CAPES  CCVC  sla#d  CAPES  CCVC  CAPES  Baseline  Baseline  Post-Phon  Post-  Tx  Morph Tx  CVC  !  !  !  CVC  !  !  kwin  CVC  !  !  !  CCVC  gl$v  !  !  !  !  CAPES  CCVC  kw$%  !  !  !  !  CAPES  CCVC  spul  CV  CV  !  !  CAPES  CCVC  b"#&  -  -  !  -  Probes  CCVC  st'"  -  CVC  !  !  Probes  CCVC  sp#l  -  CVC  !  !  Probes  CCVC  sto(v  -  CVC  !  !  Probes  CCVC  sniz  -  CVC  -  !  CAPES  CCVCVC  )sw#m#*  !  CVCVC  !  !  CAPES  CCVCVC  )b"#&+z  CCVC  !  -  !  CAPES  CCVCVC  )b"$,-  !  !  !  !  CAPES  CCVCVC  )glæs#z  !  !  !  !  CAPES  CCVCVC  )spe#%#p  CVCVC  CVCVC  CVCVC  CVCVC  CAPES  CCVCVC  )d"ægn.  !  !  !  !  Probes  CCVCVC  )kræ%#*  -  !  !  !  Probes  CCVCVC  )sn'"#*  -  CVVC  !  !  CAPES  CCVCVCC  )p"/zn.t  CCVCVC  CCVCVC  CCVCVC  !  CAPES  CVCCVC  )g!"b#&  CVCVC  CVCVC  CVCVC  CVCVC  CAPES  CVCCVC  )w#sp-  CVCVC  !  -  !  CAPES  CCVCCVC  )plænt#*  CCVCVC  CVCCVC  !  !  CAPES  CVCCVCV  k0æ*g+)"u  !  CVCVCV  CVCVCV  CVCVCV  CAPES  CVCCCVCVC  k0+m)p0ju1-  CCVCVC  CCVCVC  !  !  Probes  VCCCVCV  +m)b"/l+  -  CCVCV  CVCV  CVCV  Probes  CCVCC  sp$n&  -  CVCC  CVCC  !  Probes  CCCVC  splæ%  -  CCVC  CCVC  !  153  Appendix K – Cluster Match Data by Item T1 Match Data: Clusters List  Word  Position  Cluster  Item  Pre-  Baseline  Baseline  Post-  Post-Phon  Morph Tx  Tx  CAPES  WI  st  st!"  t  t  -  !  CAPES  WF  st  v#st  s  s  !  !  Probes  WI  st  st$"  -  t  t  !  Probes  WI  st  sto%v  -  t  t  t  Probes  WF  st  go%st  -  s  s  s  CAPES  WI  sl  sla&d  s'  s'  s'  !  CAPES  WI  sw  (sw&m&)  f  f  f  !  CAPES  WI  sp  (spe&*&p  p  p  p  p  CAPES  WI  sp  spul  p  p  p  !  Probes  WI  sp  sp&l  -  p  p  p  Probes  WI  sp  sp+n,  -  pw  p  p  CAPES  WM  sp  w&sp-  !  fp  !  !  CAPES  WI  sk  ski  k  k  !  k  Probes  WI  sn  sniz  -  n  n  n  Probes  WI  sn  (sn$"&)  CVC  n  n  !  Probes  WI  spl  splæ*  -  pw  bw  bw  CAPES  WI  k"  kr+*  kw  kw  kw  !  Probes  WI  k"  (k"æ*&)  -  kw  kw  kw  CAPES  WI  b"  (b"+.-  b  b  !  !  CAPES  WI  b"  (br&,/z  b  bw  bw  bw  CAPES  WM  b"  (zib"/  b  bw  bw  bw  CAPES  WI  t"  t"i  0  tw  tw  tw  CAPES  WI  d"  (d"ægn1  !  d  dw  !  CAPES  WI  p"  (p"#zn1t  pw  p  !  pw  Probes  WM  mb"  /m(b"#l/  -  bw  bw  mbw  CAPES  WI  gl  gl+v  g  gw  gw  gw  CAPES  WI  gl  (glæs/z  g  g  gw  gw  CAPES  WI  bl  blu  bj  bw  bw  bw  CAPES  WI  pl  (plæ2nt&)  pw  pw  p  !  CAPES  WF  nd  hæ2nd  n  n  !  !  CAPES  WF  nd  (3a%zn1d  nz  n  n  !  154  T1 Match Data: Clusters (continued) List  Word  Position  Cluster  Item  Pre-  Baseline  Baseline  Post-  Post-Phon  Morph Tx  Tx  CAPES  WM  nt  !plæ"nt#$  t  !  t  !  CAPES  WF  nt  !p%&zn't  t  t  t  t  Probes  WF  n(  sp)n(  -  n  n*  nz  CAPES  WM  $g  k+æ$g,!%u  g  nd  !  nd  CAPES  WM  mp+j  k+,m!p+ju-.  p  p  p  mp  CAPES  WF  mp  ()mp  b  p  p  p  CAPES  WM  %b  !g/%b#(  b  b  b  b  CAPES  WF  %t  0.t  !  !  !  !  CAPES  WF  %s  h1%s  %2  !  !  !  CAPES  WF  %d  j/%d  !  rt  rt  !  CAPES  WI  kw  kwin  gw  !  !  !  CAPES  WF  lf  0&lf  f  wf  wf  !  Post-Phon  Post-  Tx  Morph Tx  T2 Match Data: Clusters List  Word Position  Cluster  Item  PreBaseline  Baseline  CAPES  WI  st  st/%  t  t  !  !  CAPES  WF  st  v&st  s  s  !  s  Probes  WI  st  st1%  -  t  !  !  Probes  WI  st  sto3v  -  t  !  !  Probes  WF  st  go3st  -  s  s  s  CAPES  WI  sl  sla#d  sw  s4  !  !  CAPES  WI  sw  !sw#m#$  fw  s  !  !  CAPES  WI  sp  !spe#0#p  p  b  p  p  CAPES  WI  sp  spul  p  b  !  !  Probes  WI  sp  sp#l  -  p  !  !  Probes  WI  sp  sp)n(  -  p  p  !  CAPES  WM  sp  w#sp.  p  !  p  !  CAPES  WI  sk  ski  k  k  k  !  Probes  WI  sn  sniz  -  n  n  !  Probes  WI  sn  !sn1%#$  -  n  !  !  Probes  WI  spl  splæ0  -  bl  pl  !  CAPES  WI  k%  kr)0  kw  kw  kw  kw  155  T2 Match Data: Clusters (continued) List  Word  Position  Cluster  Item  Pre-  Baseline  Baseline  Post-Phon Tx  Post-  Morph Tx  Probes  WI  k!  "k!æ#$%  bl  kw !  kw !  kw !  CAPES  WI  b!  "br$&'z  CAPES  WI  b!  "b!()*  bl  !  !  !  CAPES  WM  b!  "zib!'  bl  !  !  !  CAPES  WI  t!  t!i  !  WI  d!  "d!ægn+  !  tw !  !  CAPES  tw !  CAPES  WI  p!  "p!,zn+t  !  !  pw  !  Probes  WM  mb!  'm"b!,l'  -  bl  b  b  CAPES  WI  gl  gl(v  gw  gw  gw  CAPES  WI  gl  "glæs'z  gw  gw  CAPES  WI  bl  blu  b!  CAPES  WI  pl  "plæ-nt$%  bj !  gw !  gw !  p  !  !  CAPES  WF  nd  hæ-nd  !  nt  !  nt  CAPES  WF  nd  ".a/zn+d  nt  CAPES  WM  nt  "plæ-nt$%  t  n !  n !  n !  CAPES  WF  nt  "p!,zn+t  t  t  t  !  Probes  WF  n&  sp(n&  -  n0  nz  nz  CAPES  WM  %g  k1æ%g'"!u  !  n  n  n  CAPES  WM  mp1j  k1'm"p1ju2*  pj  pj  !  !  CAPES  WF  mp  &(mp  b  p  !  !  CAPES  WM  !b  "g3!b$&  b  b  b  b  CAPES  WF  !t  #*t  !  !  !  !  CAPES  WF  !s  h4!s  !  !  !  !  CAPES  WF  !d  j3!d  !t  !t  !  !  CAPES  WI  kw  kwin  w  !  !  !  CAPES  WF  lf  #,lf  !f  !f  !f  !  !  !  156  Appendix L – [-anterior] Match Data by Item T1 Match Data: [-anterior] List  Word  Position  Phoneme  Item  Pre-  Baseline  Baseline  Post-  Post-Phon  Morph Tx  Tx  CAPES  WI  !  "!#kn$  !  t%  t%  t%  CAPES  WI  !  !&'  !  t%  t%  !  CAPES  WM  !  "w(!)z  *  s  s  s  CAPES  WM  !  "w(!#+  s  s  s  s  CAPES  WF  !  w(!  *  s  s  s  CAPES  WF  !  bi!  s  s  s  s  Probes  WF  !  kæ!  s  s  s  s  CAPES  WI  ,  ,æm  !  d  -  !  CAPES  WI  ,  ",æk#t  !  d  !  -  CAPES  WI  ,  ,.mp  -  -  !  d  Probes  WI  ,  ,&li  -  d  d  d  CAPES  WM  ,  "p%e#,)z  d  z  -  z  CAPES  WM  ,  "b'#,)z  /  z  -  z  Probes  WM  ,  "p#,n$  -  -  z  z  CAPES  WF  ,  k%e#,  z  z  z  z  CAPES  WF  ,  "g('b#,  z  -  z  Probes  WF  ,  sp.n,  -  z !  -  z  Probes  WF  ,  bæ,  -  -  s  z  CAPES  WI  *  *u  !  s  s  s  CAPES  WI  *  *0t  !  s  s  s  CAPES  WI  *  *&lf  !  s  s  s  Probes  WI  *  "*æ1o2  -  s  s  s  CAPES  WM  *  "f#*#+  s  s  s  s  CAPES  WM  *  "spe#3*#p  !  s  s  !  Probes  WM  *  "k'æ*#+  -  s  s  !  CAPES  WF  *  f#*  s  s  s  s  CAPES  WF  *  k'.*  s  s  s  s  Probes  WF  *  splæ*  -  s  s  !  Probes  WF  *  mæ*  -  s  s  s  157  T2 Match Data: [-anterior] List  Word  Position  Phoneme  Item  Pre-  Baseline  Baseline  Post-Phon  Post-  Tx  Morph Tx  CAPES  WI  !  "!#kn$  t%  t%  t%  t%  CAPES  WI  !  !&'  t%  t%  t%  t%  CAPES  WM  !  "w(!)z  -  s  s  s  CAPES  WM  !  "w(!#*  s  s  s  s  CAPES  WF  !  w(!  s  s  s  s  CAPES  WF  !  bi!  s  s  s  +  CAPES  WF  !  k%æ!  s  s  s  +  CAPES  WI  ,  ,æm  -  d  d  d  CAPES  WI  ,  ",æk#t  d  d  d  d  CAPES  WI  ,  ,-mp  d  d  d  d  Probes  WI  ,  ,&li  -  d  d  d  CAPES  WM  ,  "p%e#,)z  -  z  z  -  CAPES  WM  ,  "b'#,)z  z  s  -  z  Probes  WM  ,  "p%#,n$  -  .  /  z  CAPES  WF  ,  k%e#,  z  z  z  z  CAPES  WF  ,  "g('b#,  z  z  z  z  Probes  WF  ,  sp-n,  -  0  z  z  Probes  WF  ,  bæ,  -  1  /  z  CAPES  WF  ,  p%e#,  z  -  -  z  CAPES  WF  ,  b'#,  -  -  z  -  CAPES  WI  2  2u  s  s  s  s  CAPES  WI  2  23t  2  s  s  s  CAPES  WI  2  2&lf  s  s  s  s  Probes  WI  2  "2æ4o5  -  s  s  s  CAPES  WM  2  "f#2#*  s  s  s  s  CAPES  WM  2  "spe#62#p  s  s  s  2  Probes  WM  2  "k'æ2#*  -  s  s  s  CAPES  WF  2  f#2  s  s  s  s  CAPES  WF  2  k'-2  s  s  s  s  Probes  WF  2  splæ2  -  s  s  s  Probes  WF  2  mæ2  -  s  s  2  158  Appendix M – [-grooved] Match Data by Item T1 Match Data: [-grooved] List  Word  Position  Phoneme  Item  Pre-  Baseline  Baseline  Post-  Post-Phon  Morph Tx  Tx  CAPES  WI  !  !"m  f  f  f  f  CAPES  WI  !  !#$  f  f  f  f  CAPES  WI  !  %!a&zn'd  f  f  f  f  CAPES  WF  !  t(i!  f  f  f  f  CAPES  WF  !  ma&!  f  f  f  f  CAPES  WI  )  )æt  d  d  d  d  CAPES  WI  )  )#s  d  d  d  d  CAPES  WM  )  f*)+'  ,  v  v  v  CAPES  WM  )  %b+")+'  v  v  v  v  Phoneme  Item  Post-Phon  Post-  Tx  Morph Tx  T2 Match Data: [-grooved] List  Word Position  PreBaseline  Baseline  CAPES  WI  !  !"m  f  f  f  f  CAPES  WI  !  !#$  f  f  f  f  CAPES  WI  !  %!a&zn'd  f  f  f  f  CAPES  WF  !  t(i!  f  f  f  f  CAPES  WF  !  ma&!  f  f  f  f  CAPES  WI  )  )æt  d  d  d  d  CAPES  WI  )  )#s  d  d  d  d  CAPES  WM  )  f*)+'  d  d  d  d  CAPES  WM  )  %b+")+'  v  v  z  v  159  Appendix N – Treatment Word List Treatment Words snack  Magoo  snake  above  stir  below  spider  around  snap  beside  stop  banana - nana  snow  piano  spot –pot  tomato  steam -team  burrito  tail  potato  pull  Canada  nail  coconut  tall  hamburger  ball  elephant  pool  boomerang  doll  peek-a-boo  bell - bear pail - pear eel - ear rock - walk rain - Wayne right – white  160  Appendix O – Sample Treatment Plan Week 2 Treatment Plan: Review of /s/-clusters and introducing /r/ and /l/ Activity 1: Story – Spot and the Steam Engine •  Auditory bombardment of /sp/ and /st/ treatment words  Activity 2: Spot – pot discrimination activity •  Perceptual contrast activity using minimal pairs /sp!t/ and /p!t/.  Activity 3: Story – Ryan and the Rock •  Auditory bombardment of /r/ and /l/ treatment words  Activity 4: Onset – rhyme manipulation •  Introduce manipulating graphemes ‘r’, ‘l’, and ‘w’ with rhyme ‘ock’ to create the words ‘rock’ ‘walk’ and ‘rock’  Activity 5: Rock – walk discrimination activity •  Perceptual contrast activity using minimal pairs /"!k/ and /w!k/.  Activity 5: ‘Talk Like a Pirate’ die game •  Production activity using VC syllable (/!"/) in succession to move the phoneme /"/ to the syllable initial position.  161  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0071464/manifest

Comment

Related Items