Treatment for /s/ Production Errors in Children With Speech Sound Errors

An Alternative Service Delivery Model and Target Selection Approach for the Treatment of /s/ Errors

We want to learn about a new way to help people make the /s/ sound, like in the word "snake." We will be studying a way to teach the /s/ sound that uses very short speech sessions. This study will also collect recordings from people as they say words with the /s/ sound. We will listen to these recordings and check if the words are said clearly.

In the first visit, we'll do a speech lesson on how to make a clear

/s/ sound. We will also have the participant say some words while we record them talking. If they are a good fit for the study, we will invite them to do a second visit. In the second visit, they will name some pictures, answer some questions, repeat words and sentences, and do some mouth movements. After the second visit, if they are a good fit for the study, we will invite them to do the speech lessons. The speech lessons will each take 10 minutes, and they will have lessons 4 times a week for 8 weeks. They will do these lessons on video chat. They will be randomized to either start right away or after 8 weeks. We'll also ask them to participate in 3 other visits to track progress by having them repeat words with /s/.

All sessions (including testing and treatment sessions) will occur over a HIPAA-compliant Zoom account that requires a password for each session.

Prior to the first Zoom visit, parents of potential participants will complete an online prescreener via REDCap. If the participant passes the prescreener, parents will automatically be provided with recording instructions and a link to upload a recording of their child producing a probe list of 30 syllables and 20 words containing /s/. The upload will transmit the file directly into our SU managed REDCap database. Participants must produce at least 3 (but no more than 40) of the 50 items accurately on the probe to move on to the next stage of the eligibility procedures. All probe recordings will be rated by a certified SLP.

Session 1 - Consent/Assent & Dynamic Assessment Visit: (approximately 45 minutes) The goal of Session 1 is to collect consent/assent, answer questions, provide a 30-minute dynamic assessment session, and collect initial recordings. Participants and their parent/guardian will join a private Zoom call with the student researcher, Benedette Herbst. Once oral consent/assent has been collected, the families will be sent a speech/language/health history questionnaire and a home technology set-up questionnaire via email. In the dynamic assessment aspect of Session 1, participants will receive 30 minutes of instruction to orient them to accurate /s/ productions. They will also have the opportunity to practice producing /s/ with maximal feedback and support from the clinician (Benedette Herbst). The orientation and practice aspects of the dynamic assessment portion of Session 1 will include a PowerPoint presentation containing illustrations representing the vocal tract configuration for accurate /s/ production. These images will highlight the fact that correct /s/ productions a raised tongue tip or blade touching the alveolar ridge as well as lateral bracing. The practice aspect will consist of repetitions of syllables containing /s/. The clinician will provide models and specific feedback regarding the participants' productions on every trial. After 30 minutes of the instruction/practice, participants will complete syllable and word level probes for the /s/ sound. To be eligible for this study, participants must demonstrate less than 60% accuracy on /s/ at the word level and an accurate /s/ on at least 5 out of 45 syllables. Additionally, at least 75% of the /s/ errors produced on the syllable and word probes must be dental or lateral distortion errors.

After Session 1, the parents will be sent several questionnaires (participant information questionnaire, speech, language, and health questionnaire, participant demographics questionnaire, and the social-emotional impact questionnaire) via REDCap.

Session 2 - Evaluation Session: (approximately 60 minutes)

The primary goal of Session 2 is to determine eligibility for inclusion in the treatment study. Participants will be asked to complete the following tasks over Zoom:

1. Oral Mechanism Screening (5 minutes) - The structure and function of oral anatomy will be observed using passive (e.g., observe symmetry of facial structures) and active tasks (e.g., wag tongue from side to side). Participants must pass the oral mechanism screening to be included in the study.
2. Goldman Fristoe Test of Articulation - Third Edition (GFTA-3; Goldman & Fristoe, 2015) (approximately 10 minutes) - The Sounds in Words subtest will be administered and scored per standardized clinical procedures. A series of pictures will be presented to the participant and they will be prompted to produce the word depicted in the picture. The accuracy of each target word in the GFTA-3 will be evaluated to generate a norm-referenced snapshot of the participant's speech sound production skills. This assessment will be utilized to confirm the diagnosis of speech sound disorder. Participants must score less than the 7th percentile on the GFTA-3 to be included in this study. Additionally, participants must demonstrate no more than two sounds other than /s/ and /z/ in error on the GFTA-3 to be included in the study.
3. Listening Comprehension and Story Retelling subtests of the Test of Integrated Language and Literacy Skills (TILLS Nelson et al., 2016) (approximately 10-15 minutes) - For the Listening Comprehension subtest, the participants will be asked questions after listening to short stories that use academic language. For the Story Retelling subtest, the participants will be asked to retell a story that was read aloud by the examiner. These subtests will provide norm-referenced information regarding the participants' expressive and receptive language abilities. Participants must receive scaled scores of 5 or higher to be included in the study.
4. Syllable Repetition Task (Shriberg et al., 2009) (5 minutes) - Children repeat nonsense syllables that contain the sounds /ba ma da na/ ranging from 2 to 4 syllables in length (e.g., /daba/ or /mabadana/). The number of items with additions will be recorded. Participants must show less than four additions to be included in the study.
5. Linguisystems Articulation Test Inconsistency Screener (Bowers & Huisingh, 2011) (5 minutes) - Participants will repeat multisyllabic words three times after an examiner prompt (e.g., "Say lifeguard three times"). Productions will be evaluated for consistency. Participants must not demonstrate inconsistent productions on three or more items to be included in the study.

Following Session 2, the participants considered eligible for the treatment study will be randomized into one of four groups: the immediate - both positions group, which will receive 8 weeks of treatment immediately on both /s/ onset and /s/ coda, the immediate - initial only group, which will receive 8 weeks of treatment immediately on only /s/ onset, the delayed - both positions group, which will receive 8 weeks of treatment on both /s/ onset and /s/ coda after an 8-week delay, and the delayed - initial only group, which will receive 8 weeks of treatment on only /s/ onset.

Probe sessions: (30 minutes) Probes will occur at three timepoints: the week before the first treatment session for each participant, then after 8 weeks (once the Immediate Treatment Groups have completed treatment) and after 16 weeks (once the Delayed Treatment Groups have completed treatment). During these sessions, the clinician will administer the stimulability (direct imitation of syllables) probe and word probe. Parents/guardians will record these productions locally using the progress monitoring feature of the Speech Motor Chaining website. This feature allows parents to record the probes through their web browser and store the recordings automatically on the Speech Motor Chaining secure server hosted by SU. The SLP will walk the parents through step-by-step instructions on how to collect the recordings and will be available to help troubleshoot any technical issues. The SLP will provide direct models to the participant via Zoom. The syllable probe includes 45 directly imitated /s/ syllables, and the word probe consists of 75 words containing /s/ in all word positions and in both singletons (e.g., "sock") and clusters (e.g., "scooped").

Following the first probe session, the clinician will select four individualized syllable targets. Treatment targets will include four /s/ onset chains (e.g., "see"- "seat"- "seatbelt" - "wear your seatbelt" - self-generated sentence for "seatbelt") for participants randomized to one of the two Initial Position-Only groups. For participants randomized to one of the two Both-Positions groups, treatment targets will include two /s/ onset chains and two /s/ coda chains (e.g., "ace" - "lace" - "shoelace" - "tie your shoelace" - self-generated sentence for "shoelace"). Targets will be individually selected for each participant when possible so that the chains targeted first in treatment include syllables for which the participant is stimulable.

Treatment sessions (32 sessions, 10 minutes each) The treatment phase of the study is designed to contain 32 ten-minute telepractice sessions, at a frequency of four sessions per week for 8 weeks. Sessions will be facilitated by the Speech Motor Chaining treatment software program (Preston et al., 2019). Speech Motor Chaining was originally encoded into Excel spreadsheets, but it has been since made available as a web-based treatment tool. Client and speech-language pathologist interfaces are accessible through a web browser (https://chaining.syr.edu).

The first treatment session for all participants will consist of 10 minutes of unstructured practice. Participants will be reintroduced to the images within the PowerPoint from the dynamic assessment visit. After reviewing the images and discussing the tongue placement for accurate /s/ production, participants will produce repetitions of the four /s/ syllable sequences that were selected as their initial treatment targets. The clinician will provide models and specific feedback for every trial. In Speech Motor Chaining, treatment stimuli are organized into chains of syllables, monosyllabic words, multisyllabic words, phrases, and sentences. Targets in the chain are presented in order of increasing difficulty. Common to each chain is the foundational syllable target, such that a sample chain for the target /so/ might include /so/, soap, soapy, and soapy water.

Each trial within the treatment sessions will be facilitated by Speech Motor Chaining software (chaining.syr.edu), following the general trial structure common to motor-based practice (i.e., prompt, production, feedback). Trials are organized into linguistically-homogenous treatment blocks of six items. A direct model will be provided to the participant for all six of the items. The clinician will perceptually rate the accuracy of each trial and provide feedback based on the accuracy rating and the type of feedback that was randomized to that trial.

Inclusion Criteria:

1. Age: Participants must be between the ages of 7;0 and 11;11.
2. /s/ Errors: Participants must experience difficulty on /s/.
3. English Language Status: Participants must speak English as a dominant language and must have begun learning English before age 3
4. Hearing Status: Participants must have no reported hearing loss.
5. Neurodevelopmental Status: Participants must have no reported neurodevelopmental diagnosis. A reported diagnosis of ADHD/ADD is admissible if participants meet all other inclusionary criteria.
6. Technology Access: Participants must have access to a Windows or Apple computer with the ability to join a Zoom session. Additionally, participants must have wired broadband internet connection at home.
7. Location: Participants must reside in the United States or Canada.
8. Orthodontic Appliances: Participants must have no orthodontic retainers or appliances that would prevent them from touching the roof of their mouth with their tongue.
9. Prescreener Recordings: Participants must accurately produce at least 3 items on the /s/ prescreener probe list. Additionally, participants must accurately produce no more than 36 items on the /s/ prescreener probe list.
10. Dynamic Assessment Probes: Participants must accurately produce at least 5 out of 45 /s/ syllables correctly and must accurately produce no more than 60% of 75 /s/ words correctly. Additionally, participants must produce at least 75% of /s/ errors as dental or lateral distortion errors.
11. Oral Mechanism Examination: Participants must demonstrate adequate range of motion of the tongue during a brief oral mechanism examination.
12. Goldman Fristoe Test of Articulation - 3rd Edition (GFTA-3): Participants must score below the 7th percentile on the GFTA-3 and must not demonstrate errors on more than two sounds other than /s/ and /z/ on the GFTA-3.
13. Test of Integrated Language and Literacy Skills (TILLS): Must receive a scaled score of at least 5 on the Listening Comprehension and Story Retelling subtests of the Test of Integrated Language and Literacy Skills (TILLS).
14. Linguisystems Articulation Test (LAT) Inconsistency Screener: Participants must produce less than three inconsistencies on the LAT Inconsistency Screener.
15. Syllable Repetition Task (SRT): Participants must produce less than four additions on the SRT.
16. Recording: Consent/assent to recording of assessment and treatment sessions.
17. General consent/assent: Consent/assent provided for study procedures.

Exclusion Criteria:

• 1. Age: Participants less than 7;0 or greater than 11;11. 2. /s/ Errors: Participants who do not experience difficulty on /s/. 3. English Language Status: Participants who do not speak English as a dominant language and/or who began learning English prior to age 3 4. Hearing Status: Participants who have reported hearing loss, per parent report.

  5. Neurodevelopmental Status: Participants who have a reported neurodevelopmental diagnosis. A reported diagnosis of ADHD/ADD is admissible if participants meet all other inclusionary criteria.

  6. Technology Access: Participants who do not have access to a Windows or Apple computer with the ability to join a Zoom session. Additionally, participants who do not have wired broadband internet connection at home.

  7. Location: Participants who do not reside in the United States or Canada. 8. Orthodontic Appliances: Participants who have orthodontic retainers or appliances that would prevent them from touching the roof of their mouth with their tongue.

  9. Prescreener Recordings: Participants who accurately produce less than 3 items on the /s/ prescreener probe list. Additionally, participants who accurately produce more than 36 items on the /s/ prescreener probe list.

  10. Dynamic Assessment Probes: Participants who accurately produce less than 5 out of 45 /s/ syllables correctly and/or who accurately produce more than 60% of 75 /s/ words correctly. Additionally, participants who produce less than 75% of /s/ errors as dental or lateral distortion errors.

  11. Oral Mechanism Examination: Participants who do not demonstrate adequate range of motion of the tongue during a brief oral mechanism examination.

  12. Goldman Fristoe Test of Articulation - 3rd Edition (GFTA-3): Participants who score above the 7th percentile on the GFTA-3 and/or who demonstrate errors on more than two sounds other than /s/ and /z/ on the GFTA-3.

  13. Test of Integrated Language and Literacy Skills (TILLS): Participants who receive a scaled score below 5 on the Listening Comprehension and Story Retelling subtests.

  14. Linguisystems Articulation Test (LAT) Inconsistency Screener: Participants who produce greater than three inconsistencies on the LAT Inconsistency Screener.

  15. Syllable Repetition Task (SRT): Participants who produce greater than four additions on the SRT.

  16. Recording: No consent/assent provided to recording of assessment and treatment sessions.

  17. General consent/assent: No consent/assent provided for study procedures.