introdduction happiness characterics in learning.pdf

Child Language and Happiness Behaviors: Evaluating the Effects of Caregiver

Coaching

Caregivers often serve as their child’s first teacher, creating opportunities for

learning during interactions. Caregivers teach their infant how to communicate during

these interactions by establishing joint attention (e.g., pointing at a toy), modeling (e.g.,

labeling toys during play), and providing feedback (e.g., responding to the child’s

interests) (Owens, 2019). For example, infants initially reflexively cry, which is followed

by a response from their caregiver, such as a feeding or diaper change (Paul & Norbury,

behavior helps explain how behaviors continue or are extinguished by contexts.

Essentially, this means that if the child responds to the caregiver's communication, then

the caregiver's communication is reinforced, and their response back to the child will also

reinforce the child’s responding and/or initiating communication behaviors. Thus, when

practicing language skills during play, ideally, children will be happier and more likely to

reinforce their caregiver’s teaching behaviors as well.

Language development, a communication component, typically emerges later and

may be delayed for some children. Caregivers of children with expressive language

delays may struggle with supporting their child’s language development, especially when

2

children display difficulty reliably responding to their bids for attention. Caregiver

initiations may subsequently decrease because the child responds infrequently, or they

perceive that their child is disinterested. For example, parents and caregivers of children

with autism spectrum disorder (ASD) reported higher levels of stress when attempting to

address their child’s communication and language (Alpern, 2012; Benson & Karlof,

2009; Estes et al., 2009; Hayes & Watson, 2013). When this occurs, caregivers may need

support to know how to encourage their child to communicate using age-appropriate

language (Ledford & Wolery, 2013).

they learn (Sandall et al., 2005). However, caregivers may face barriers when securing

services. Logistical barriers can include location (e.g., living in a rural area),

transportation issues, and lack of childcare for other children. An access barrier is being

on a waitlist for services, especially when caregivers seek behavioral services. Finally, a

participation barrier may also hinder meaningful service delivery; some caregivers

reported feeling like passive participants in their child’s care (Brown et al., 2012; Farmer

et al., 2014) and, as such, are potentially ill-equipped to address their child’s

communication. Potential barriers may decrease opportunities for learning, especially

during critical developmental periods. For children with disabilities, it is crucial that

3

adults, including caregivers, provide an adequate dosage of opportunities to learn and

practice skills within natural routines and activities (Ledford & Wolery, 2013; Warren et

al., 2007). Thus, it is essential that professionals and caregivers collaborate and identify

practical and evidence-based practices in a timely manner to maximize child learning

opportunities (Dunst et al., 2000).

There is a breadth of research demonstrating practitioner success in training

caregivers to implement evidence-based practices in both clinical and home-based

settings (Rocha et al., 2007). Within this context, training refers to teaching a few skills to

components: (a) direct instruction, (b) modeling, (c) role-playing, (d) written/typed

scripts, and (e) feedback in-person or through video (Chaabane, et al., 2009; Cardon,

2012; Kaiser & Roberts, 2013, Lafasakis & Sturmey, 2007; Lane, et al., 2016; see Table 1

for detailed descriptions).

4

Explicit

Instruction)

Scripts

There are a variety of multi-component coaching interventions for caregivers who

want support for intervening on child communication (Ledford et al., 2019). The Family-

5

Guided Routines-based Intervention (FGRBI; Cripe & Venn, 1997; Woods et al., 2004),

Behavioral Skills Training (Shayne & Miltenberger, 2013), and Teach-Model-Coach-

Review (TMCR; Kaiser & Roberts, 2013) are all examples of packaged interventions for

families. FGRBI (Cripe & Venn, 1997; Woods et al., 2004) involves an intensive family-

centered approach (e.g., Problem Solving and Planning, Reflection and Review, etc.)

during weekly visits to help caregivers implement teaching opportunities in their natural

environment and daily family routines; the intervention may take up to 32 weeks for a

caregiver to reach mastery (Woods et al., 2004). BST is a widely used teaching

Roberts, 2013; Wright & Kaiser, 2017). TMCR involves an initial workshop, followed by

approximately 12 weeks (about 3 months) of training, twice a week and for 40 min per

visit (Roberts et al., 2014). Ideally, caregivers will receive coaching for an extended

period, but accessing specialized interventions, as well as barriers, may delay

implementation and be problematic, especially if caregivers lack foundational skills for

encouraging communication and teaching language.

Caregivers of children with delayed language need timely access to services and

supports that can help them provide their children with a rich environment for language

development. A potential option to fill this need is by conducting relatively brief coaching

6

interventions targeting caregivers' foundational skills. An example of this is the rapid

coaching intervention (RCI) described by Lane and colleagues (2016), which found that a

multi-component RCI was effective for teaching caregivers naturalistic language

strategies to use with their child with ASD during play. Naturalistic language strategies

refer to practices caregivers can use to promote language by altering their child’s natural

environment, such as the following skill sets: playing, noticing and responding, providing

linguistic input, and planning communication opportunities. These skills are commonly

recommended to caregivers because they are basic and foundational skills that may help

across contexts that may lead to long-term improvements in communication. This

intervention has been utilized in homes (Zhu et al., 2022), online (Lane et al., 2023), and

in community-based clinics, with support from a speech-language pathologist (Campbell,

2022; Reiss, 2023). In a recent study (Gullett, 2024), caregiver feedback (e.g., watching a

feedback video was nonpreferred) highlighted the importance of potentially tailoring

training and coaching by family instead of an identically packaged intervention. Previous

literature also highlights the importance of tailoring performance feedback systems

(Barton, et al., 2011). This is also an important consideration when further evaluating the

proposed RCI.

7

An important aspect of caregiver-focused interventions, including the RCI, is

considering the impact on the child. For example, communication and language

interventions for young children often include measures of communication, such as

nonverbal behaviors (e.g., gestures), linguistic competence (understanding language), or

linguistic performance (using language) (Lane & Brown, 2016; Owens, 2019). Additional

child behaviors may warrant attention in such studies. For example, child communication

is often measured during play because the context is considered a natural part of early

childhood and may increase a child’s motivation to communicate. Ideally, we can expand

happiness in adults is associated with increased productivity and social engagement

cognitive flexibility (Baas et al., 2008; Bless et al., 1992; Fredrickson & Branigan, 2005;

Ritter & Ferguson, 2017). Given the potential benefits of happiness, expanding measures

to early childhood research may be useful for professionals.

Regarding happiness, increased awareness of the lived experiences of

neurodivergent individuals is important, especially when considering self-reports of what

happiness looks like, which has direct implications for professionals. Happiness is often

defined through either a self-report of well-being or observing a positive affect

8

(emotional) state (Haybron, 2020). Within relatively brief studies, it may be more feasible

and practical to measure happiness as a positive emotional state of an individual, rather

than “well-being” (i.e., long-term measure). Previous researchers have developed the

Indices of Happiness and Unhappiness Questionnaire to assist in making treatment

preference and decision-making for nonverbal and minimally verbal individuals (Parsons

et. al., 2012, Ramey et al., 2023). Individualized indices were created by providing

relevant adults (caregivers, teachers, etc.) with a brief questionnaire about the behaviors

the child exhibits when happy and unhappy. Caregivers completed the questionnaire and

Parsons, et al., 2012, Smith et al., 2005). Therefore, analysis of happiness behaviors may

improve language assessment and treatment by helping practitioners better understand

Additionally, researchers have used measurements of happiness behavior to assist

in determining which teaching interventions to use when working with children with

ASD (Dunlap & Koegel, 1980). However, measuring the children’s indices of happiness

has not previously been used to evaluate the social validity of caregiver training and

intervention services.

9

The purpose of this study was to evaluate the effectiveness of a tailored RCI

(training and coaching) with caregivers of children with developmental disabilities. This

study replicated Lane et al. (2016) and includes components from other studies (e.g.,

Campbell, 2022) and recommendations (Gullett, 2024; Lane, 2023). Like other studies,

this study involved evaluating child-level communication but extended child measures by

including an assessment of happiness behaviors to assess the social validity of caregiver

coaching interventions.

Research Questions

initiations after caregivers learn naturalistic strategies?

Responding to a caregiver was operationally defined as any attempt by the child

to communicate verbally or nonverbally (i.e., point, show, reach) within 5 s of a

caregiver’s question, cue, or prompt. Any verbalization, approximations, and non-word

sounds were considered an attempt to communicate. Only the first instance was counted

if a child repeated the same word or phrase repeatedly unless there was more than a 3-s

delay between the ending of the first word and the beginning of the second one.

Examples included the child imitating a caregiver modeling target language or reaching

towards the item the caregiver modeled, the child saying or approximating the name of a

the caregiver provided a question, cue, or prompt to communicate. Like responding, only

the first vocalization was counted if the child repeated the same word or phrase over and

over unless there were more than 3 s between the ending of one word and the beginning

of another. Examples included the child giving the caregiver a toy to play with, guiding

the caregiver to a toy they cannot reach, or asking to be tickled. Non-examples of

initiation to a caregiver included a child approaching their caregiver after their caregiver

spoke to them or a child handing something to their caregiver when their caregiver

gestured for it.

Happiness Behaviors

Happiness behaviors were defined as observable facial or vocal responses, such as

laughing and smiling, and are unique to all children. Happiness indices were defined

individually for the child, corresponding to standards offered in previous literature

The researcher administered an altered Indices of Happiness and Unhappiness

Questionnaire (Appendix D) to the caregiver and developed an individualized list of

happiness and unhappiness behaviors for the child. The researcher added additional

questions to the questionnaire regarding the child’s joint engagement. Engagement was

and unhappiness behavior from the list was operationally defined and written out for data

collectors (Appendix E). Happiness and Unhappiness behaviors were tracked separately

using momentary time sampling (not mutually exclusive: could happen at different times,

the same time, or neither could be happening).

Since outward expressions of happiness can be fleeting and difficult to capture,

the researcher added a joint-engagement component to the indices of happiness. To better

capture and conceptualize happiness, the researcher provided an operational definition of

joint engagement and no joint engagement for August to the happiness/unhappiness

datasheet (Appendix E). These definitions were created using information collected

during the interview and administration of the altered Indices of Happiness and

Unhappiness Questionnaire (Appendix D). Data collectors used the same momentary

time sampling datasheet to indicate happiness, unhappiness, and joint-engagement

behaviors across all sessions.

Happiness, unhappiness, and joint engagement behaviors were collected and

assessed across all baseline and intervention sessions. Intervals were 10 s long, and all 4-

min sessions had a total of 24 intervals. Potential examples of happiness behavior

included: smiling, laughing, hugging, jumping, giggling, or hand flapping. Potential

complexity of the target behaviors (i.e., discrete skills first, chained skills last). To move

from one tier to the next tier, (1) the adult must first meet the pre-determined criterion,

and (2) baseline data for other skills shows a stable response pattern. Researchers only

intervened on one skill/tier at a time, creating a time lag between intervening on skills

based on the caregiver’s behavior. This design was chosen because the research question

for this study is a demonstration research question that investigated likely non-reversible

behaviors (Lane et al., 2016; Ledford & Gast, 2018, p. 97-131). This design did not

require removing an intervention to demonstrate a functional relation; the caregiver’s

performance was used to make experimental decisions. Additionally, this design allows

the research team to discover any covariation in untreated tiers by having continued

baseline data collection. Since the caregiver was learning to implement various

responsive interaction strategies, monitoring for potential changes to other behaviors/tiers

was vital. Data were collected continuously throughout every session on caregiver and

child behavior. Continuous measurement of pre-intervention behaviors allowed the

research team to examine the data more regularly than a multiple-probe design (Gast et

al., 2018). A multiple-baseline design was used to time-lag the introduction of the RCI

across two caregiver behaviors, each at a different point in time. The researcher will

variables. This design helped detect and control for maturation, history, testing, multi-

treatment interference, and instability by continually collecting data in each condition

until stability was observed in the data path. Secondary observer data collectors were

trained until they reached a pre-determined criterion to control for threats to procedural

infidelity and instrumentation. Functionally independent behaviors were selected to

decrease the possibility of covariation (Gast et al., 2018).

Procedures

Screening & Baseline Condition

Before the first appointment, the caregiver completed the MCDI: Words &

Gestures, Third Edition, using the form corresponding to their child’s age (Marchman et

al., 2023). The purpose of this form was for caregivers to indicate their child’s vocabulary

and related age-expected language. During the initial meeting, the researcher interviewed

the caregiver and administered the altered Indices of Happiness Questionnaire (Appendix

D). Due to a scheduling conflict, the child was unable to attend the first meeting and the

remaining screening/baseline procedures were postponed until the second meeting.

At the beginning of the second meeting, the researcher asked the caregiver how

assess how the child normally communicates. The recording of the language sample was

then transcribed and analyzed (see Appendix F). When measuring vocal communication,

approximations of words/language were included, and approximations were defined as

having at least one phoneme from a word(s) related to the referent (Paul & Norbury,

2012). Results of the language sample indicated that the child said 11 intelligible words

and 5 novel words during the sample. Additionally, the mean length of utterances in

morphemes (MLUm) and words (MLUw) were 1.08, respectively. This score fell into the

1.00-2.00 range and is correlated with the first of Brown’s five stages of development

(Brown, 1973). Children scoring in the first stage of Brown’s language development scale

indicate a child should have or be progressing toward a vocabulary of around 50-60

words (Brown, 1973).

Following the meeting, the implementor met with an SLP to discuss the

assessment data and language sample and to identify child-level instructional targets.

Before beginning coaching sessions, the caregiver was informed of the child’s

communication targets, the rationale behind each target, and how to promote and

reinforce their child’s target communication behaviors.

The 15-min play session was trimmed down to a 12-min sample (three 4-min

Intervention Condition

The implementor, caregiver, and child all met via Zoom together. RCI sessions

occurred approximately once a week for 1-hr. The caregiver received and reviewed an

asynchronous training module on each target behavior before beginning coaching

sessions on the corresponding behavior. At the beginning of the first intervention session,

the researcher explained the child’s language target and a rationale for each target.

Additionally, the researcher asked the caregiver how the child’s day had been. This verbal

caregiver report was collected at the beginning of all intervention meetings to account for

any drastic variability in child behaviors across different meeting sessions. This was

important for happiness data stability because multiple sessions were completed within

each meeting, which could lead to cyclical data or major differences in happiness

behavior across different days.

implementor recorded each coaching session individually and provided live performance

feedback on the caregiver implementing target behaviors (e.g., “Great job adding new

actions to August’s play!” “Awesome job waiting, that was perfect!”). After each session,

the implementor offered to replay parts of the recording to provide positive feedback

regarding the caregiver’s use of the target behavior and identify times they could have

used it. The learning criterion for each skill was the caregiver displaying the behavior at

least four times per session and across three consecutive sessions.

Caregiver Training. The implementor sent an asynchronous training link to the

caregiver via email a few days before the first coaching session for each tier. For each

skill, a link provided the caregiver with a handout about the skill and a short training

video. Each training handout defined a target caregiver skill and provided guidelines for

how often they should engage in the behavior (See Appendix G). Training videos were

each between 5-8 minutes long (Lane 2023) and provided a definition of the skill, a

rationale for using the skill, guidelines for how often to use the skill, and an example of a

research team member modeling the skill with a child.

Coaching Sessions. Right after the implementor reviewed the training materials,

the researcher and caregiver began coaching sessions. The researcher began each session

immediately provided performance feedback. Performance feedback included praising

specific instances of skills, summarizing the live coaching feedback, and providing any

additional commentary (e.g., explaining why a specific instance did not count). An

example of performance feedback included the researcher saying, “You did a great job

noticing and responding when August gestured to go up- it's great that you’re responding

to all of his forms of communication!” (Specific praise), “I noticed that you used the new

skill several times throughout- you did an amazing job!” (Summarizing live coaching

feedback), “Don’t forget to wait for five seconds after your sentence so he has a chance to

respond or mimic you” (additional commentary). Next, the researcher asked the caregiver

how they felt the session went and if they had any questions during the post-session

feedback. After answering any questions, the researcher offered to show the caregiver

specific examples from the video recording. The researcher also offered to model the skill

live and/or rewatch portions of the training video with the caregiver. Performance

feedback is estimated to take 2-3 min (Lane et al., 2016).

Maintenance Condition

Maintenance sessions will occur one week after the caregiver has reached the

mastery criterion for all target behaviors. Maintenance sessions will be conducted via

Social Validity

After the study concludes, social validity will be evaluated by conducting a

caregiver interview. The interview will be conducted by someone unaffiliated with the

intervention team to ensure that the caregiver is comfortable discussing their experiences

in the study. Questions include:

training videos, online coaching, feedback, etc.?

our time together?

Reliability & Fidelity

Interobserver agreement (IOA) and procedural fidelity data were collected for at

least 33% of sessions across every condition. Before assisting with the study, secondary

data collectors were trained on relevant data collection procedures (procedural fidelity,

caregiver behaviors, child happiness behaviors, and child communication behaviors).

stamp video records on caregiver and child behavior. Video records was coded via time-

stamps in an Excel spreadsheet or using datasheets. IOA was analyzed using point-by-

point agreement with timestamps. Time stamps needed to be within 5 s of one another to

count as an agreement for instances of target behavior. The IOA was calculated as the

number of agreements within the time frame divided by the number of agreements plus

disagreements, with the quotient multiplied by 100 (Campbell, 2022; Ledford et al.,

2018). If IOA data had fallen below 80%, data collectors would have been retrained until

they reached the criteria previously listed. IOA for caregiver behaviors was 100 % across

all conditions (baseline and intervention). IOA for child happiness ranged from 94% to

or Tier 1 of the intervention. Data coding and analysis for additional tiers will be

completed as the project progresses.

Discussion

The goal of this study was to replicate and expand the results of Lane et al. (2016)

by evaluating the proposed RCI with a caregiver with a young child with a disability, as

well as to investigate child happiness as a proxy for social validity of the intervention for

the child. The intervention was a multi-component, rapid coaching intervention, which

included an interventionist providing a caregiver with rationales, modeling opportunities,

coaching throughout brief sessions, and performance-based feedback. The caregiver was

trained and coached on naturalistic strategies associated with improved expressive

imitating her child’s play. Thus far, the intervention has led to improvements in two

caregiver behaviors, with no changes observed in untreated tiers until the intervention

was introduced. The caregiver was taught to expand her child’s play behaviors and to

notice and respond to her child’s communication during play-based activities. Thus far,

only one coaching prompt was needed for the caregiver during training sessions (e.g.,

researcher tells the parent, “Now would be a great time to expand on the play”). Next, the

caregiver will be taught how to provide linguistic input (narrate her and her child’s play)

and arrange the environment to provide her child with opportunities to communicate.

There were numerous limitations within this study that should be addressed. The

first limitation of this study is that the study is ongoing. The extent to which this study

replicates findings in previous studies will remain unknown until the study is complete.

Improvements in two caregiver behaviors are promising, but no formal conclusions can

be made regarding the presence or absence of a functional relation. Additionally, child-

level data coding is not complete. Child communication data has not been coded or

evaluated for the post-test condition, and child happiness has not been coded or evaluated

for Tier 2. Once all training tiers and data collection have been completed, the results and

coaching sessions. Further, social validity data have not been collected because the study

is ongoing. However, the caregiver verbally reported that since beginning the training,

she felt that her child was more willing to communicate and engage in play activities with

her.

An additional limitation was conducting online sessions but formally coding data

after the meeting with the caregiver. For example, as evidenced in Tier 1, the caregiver

reached the criterion in Tier 1. Still, additional sessions within the same visit were

conducted, with one of those additional sessions below the criterion (i.e., 3 occurrences

instead of 4). Second, monitoring child behavior was especially difficult using the

online/recorded video format. Although the caregiver tried to stay in the camera frame,

there were times that the child moved away from the camera or was at an angle where it

was difficult to see either one or both participants (e.g., standing directly in front of the

camera with only torso visible, sitting/laying parallel or perpendicular to the camera,

etc.). This made coding happiness particularly difficult and caused the coding to be less

sensitive to some behaviors. For example, one of the main happiness behaviors for the

child was smiling, but there were many times when his face was off-camera completely

or obscured by the angle he was facing. Thus, some happiness/unhappiness behaviors are

are commonly recommended for promoting early communication and language

development in children (Schreibman et al., 2015). Teaching naturalistic strategies to

caregivers has potential benefits over professional-only implemented interventions since

families typically spend the most time with the children (i.e., outside of professional

service settings). If implemented at fidelity and recommended dosages (Bailey, 2024;

Lane et al., 2024; Warren et al., 2007), training caregivers to implement these strategies

can be a time- and cost-effective strategy that can potentially benefit all involved parties.

Since replicating the RCI using an online format was successful, there are

practical implications for practitioners providing online services or with large caseloads.

Families in rural or underserved areas may be more easily able to access services such as

the RCI if they are offered online or remotely. Additionally, training families via video in

their natural environment may be useful for generalization purposes (e.g., practice using

their own toys/space) and be less intrusive than home-based sessions (this is dependent

on family preferences). The results of this study add to existing literature on naturalistic

strategies by demonstrating an online intervention that involves a team of people (i.e.,

different professionals and family members) collaborating to support a child’s acquisition

and generalization of language skills.

consider the utility of including an assessment of happiness during interventions to assess

social validity, especially when participants have limited vocal repertoires. Across all

conditions, the interventionist asked the caregiver to describe the child’s temperament or

mood that day. Although this information was not formally evaluated, this narrative

information informed our analysis of happiness behaviors and could have potentially

explained potential variabilities if data were variable. Future researchers should consider

methods of assessing happiness behavior and possibly ask the caregiver if the indices of

happiness and unhappiness are representative of their child’s most common behaviors.

Throughout this study, a new component was added to the procedures to tailor the

coaching sessions to the caregiver by asking their feedback and training preferences. By

individualizing the coaching sessions to include this feedback, we found anecdotal

evidence that some families may want different levels of feedback and modeling. For

example, previous iterations of the RCI required the interventionist to show the caregiver

a recording of their implementation with their performance feedback, however, in this

study the caregiver did not wish to review any recordings of her sessions.

Individualization of training and coaching procedures may also help improve buy-in from

collaboration between SLP’s and behavior interventionists. Interdisciplinary collaboration

helps focus all adult efforts on teaching important skills across developmental areas and

is crucial for a child’s developmental trajectory (Ronfeldt et al., 2015). This study

contributes to the interdisciplinary literature by modeling a successful collaboration

between a SLP and a behavior interventionist to train caregivers on strategies to promote

expressive communication in their child. Currently, formal data collection of child

communication behaviors is in progress. Communication data will be utilized to analyze

relations between the caregiver’s use of naturalistic strategies and concurrent child

communication (in responses and spontaneous initiations). This data will be collected and

analyzed before, during, and after the collaboration and coaching occurred, to see if there

was any change in the child’s expressive communication.

The purpose of this study was to evaluate the effectiveness of a tailored RCI (training and

coaching) with caregivers of children with developmental disabilities. This study

replicated Lane et al. (2016) and included components from other studies (e.g., Campbell,

2022) and recommendations (Gullet, 2024; Lane, 2023). Additional data are needed to

determine the effectiveness of the tailored RCI for teaching a caregiver to implement

researchers continue to explore the effects of the RCI and how to accurately and

meaningfully collect social validity data from participants with limited or no verbal

repertoires.