M3 Research and Literature
zanuyekeh
Study4FergusonTelehealth.pdf
Evaluating discrete trial teaching with instructive feedback delivered in a dyad arrangement via telehealth
Julia L. Ferguson, Maddison J. Majeski, John McEachin and Ronald Leaf Autism Partnership Foundation
Joseph H. Cihon and Justin B. Leaf Autism Partnership Foundation and Institute of Behavioral Studies, Endicott College
Telehealth research in the field of applied behavior analysis (ABA) has focused on the training and supervision of others to implement various behavior analytic procedures. Current guidelines for practicing telehealth suggest that direct telehealth services may be appropriate for some indi- viduals diagnosed with autism spectrum disorder, but more research is needed. This study evalu- ated the effects of discrete trial teaching with instructive feedback in a dyad arrangement delivered directly via telehealth to teach tact relations to 6 children diagnosed with autism spec- trum disorder. All participants and the experimenter were located in different physical locations. All participants learned their primary and secondary targets, and 5 of the participants acquired the observational primary and secondary targets without direct teaching. Areas of future research and clinical implications are discussed in the context of telehealth service delivery. Key words: autism spectrum disorder, discrete trial teaching, instructive feedback, telehealth
Telehealth is defined as the treatment of health-related conditions through the use of communication technology (Ferguson et al., 2019). Results of a recent literature review on applied behavior analytic (ABA) services deliv- ered via telehealth for children diagnosed with autism spectrum disorder (ASD) indicated that this research has focused solely on training other individuals to implement procedures (Ferguson et al., 2019). This method of telehealth service delivery is different than what many behavior analysts may be providing to families as part of clinical practice, particularly clients who receive services exclusively via telehealth. Depending on the child’s repertoire, some behavior analysts and their supervisees are implementing clinical sessions directly with the child via Zoom™ or other video-based tele- communication platforms. That is, staff are conducting sessions virtually with the child, not
necessarily training parents or another interven- tionist to implement behavior analytic proce- dures (Rodriguez, 2020). Considerations that may be important for
learner success in this format are the client’s attending skills, basic joint attention skills, basic echoic skills, ability to follow one-step instructions, responsiveness to distance pro- mpts, and level of problem behavior (Council of Autism Service Providers, 2020; Rodriguez, 2020). This new practice of directly providing intervention through video confer- encing technology differs from the currently available behavior analytic research on telehealth. Thus, more research is needed on providing commonly implemented ABA proce- dures to clients directly through telehealth. Discrete trial teaching (DTT) is a commonly
implemented ABA procedure used to teach a variety of skills to individuals with ASD (Eikeseth et al., 2014; Leaf et al., 2016). Addi- tionally, an increasing number of studies (e.g., Leaf et al., 2017; Nottingham et al., 2020) have demonstrated the benefits of incorporating
Address correspondence to: Julia L. Ferguson, Autism Partnership Foundation. 200 Marina Dr., Seal Beach, CA 90740. [email protected]
doi: 10.1002/jaba.773
Journal of Applied Behavior Analysis (Fall) 2020, 53, 1876–1888 NUMBER 4
© 2020 Society for the Experimental Analysis of Behavior (SEAB)
1876
instructive feedback into DTT. Instructive feed- back is the practice of including extra nontarget (or secondary targets) stimuli in the instruction or consequence component of DTT (Werts et al., 2003). The learner is not required, or asked, to respond to the additional stimuli, and the instructor does not reinforce any response of the learner to these stimuli. Research on instruc- tive feedback indicates that individuals diagnosed with ASD acquire the secondary responses about 64% of the time (Albarran & Sandbank, 2019). Additional benefits have been documented when instructive feedback is used with dyads or small groups of learners. Research has found that partic- ipants acquired peers’ original targets about 59% of the time and peers’ secondary responses 55% of the time through observational learning (Albarran & Sandbank, 2019). Given the lack of research evaluating the
direct provision of ABA procedures via telehealth and the potential benefits of DTT with instructive feedback, the purpose of this study was to evaluate the effects of this inter- vention delivered via telehealth to teach tact relations to dyads of children with ASD. Specif- ically, the goal was to determine if the partici- pants would acquire secondary and observational responses as demonstrated in pre- vious in-person research. Given potential diffi- culties with providing direct telehealth instruction, data were also collected on partici- pant attending (a potential prerequisite skill for direct telehealth services; Rodriguez, 2020) and engagement (i.e., echoics and comments emit- ted) during teaching sessions.
Method
Participants Six children independently diagnosed with
ASD participated. Participant demographic information is provided in Table 1. All partici- pants had a previous history with DTT and instructive feedback conducted in a group for- mat, but all had limited to no experience with
DTT conducted via telehealth. Four of the six participants (i.e., Carter, Jackson, Ryan, Ethan) had some experience with group instruction delivered via telehealth due to the COVID-19 pandemic. As a result of the pandemic, their ABA-based classroom instruction switched to online instruction 2 weeks prior to this study. The other two participants (i.e., Marshall and Flynn) did not have any prior experience with DTT conducted via telehealth. All participants engaged in reciprocal conversational skills, responded to wh- questions, imitated models, followed instructions, and had a history of praise functioning as a reinforcer. All partici- pants had also demonstrated the acquisition of communication and social skills in group instructional formats. All participants engaged in low to moderate rates of stereotypic behavior (e.g., scripting movie or television show quotes, perseverating on a preferred topic). The partici- pants did not engage in any aggressive or self- injurious behavior, or high rates of noncompliance that interfered with sessions. Carter, Jackson, Ryan, and Ethan were ran- domly divided into dyads and Flynn and Mar- shall were grouped into a dyad because they are brothers and were in the same location.
Setting and Materials All sessions were conducted using the
Zoom™ Video Communications, Inc. (Zoom) platform with a HIPAA-compliant account (Zoom, 2020). Participants and the experi- menter were in their respective homes using various technological devices with video and audio capabilities. Some participants used desk- top or laptop computers (Carter, Jackson, Ryan, Ethan) and other participants used an iPad (Flynn and Marshall). Carter, Jackson, Ryan, and Ethan participated in an ABA-based Kindergarten classroom that had transitioned to online instruction via Zoom™ during the COVID-19 pandemic. For those four partici- pants, the classroom teacher created a breakout
1877Instructive Feedback via Telehealth
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
T ab le 1
Pa rt ic ip an t D em
og ra ph ic In fo rm
at io n
D ya d
N am
e A ge
Se x
R ac e/
E th ni ci ty
D ia gn os is
E du
ca tio
n IQ
Sc or e
V in el an d
E O W PV
T PP
V T
1 Ja ck so n
7 ye ar s 1 m on th
M al e
C hi ne se
A SD
K in de rg ar te n
93 (A ve ra ge )
84 (M
od er at el y Lo
w )
96 (A ve ra ge )
90 (E xp ec te d)
1 C ar te r
7 ye ar s 1 m on th
M al e
K or ea n
A SD
K in de rg ar te n
68 (E xt re m el y Lo
w )
77 (M
od er at el y Lo
w )
66 (V er y Lo
w )
73 (B el ow
E xp ec te d)
2 Fl yn n
5 ye ar s 1 m on th
M al e
B ir ac ia l
A SD
Pr e- K
10 8 (A ve ra ge )
70 (L ow
) 11 0 (A ve ra ge )
10 1
(E xp ec te d)
2 M ar sh al l
3 ye ar s 11
m on th s
M al e
B ir ac ia l
A SD
Pr e- K
10 3 (A ve ra ge )
69 (L ow
) 11 9 (H
ig h A ve ra ge )
10 7
(E xp ec te d)
3 R ya n
7 ye ar s 1 m on th
M al e
C au ca si an
A SD
K in de rg ar te n
10 9 (A ve ra ge )
85 (M
od er at el y Lo
w )
12 9 (S up
er io r)
12 3
(A bo ve
E xp ec te d)
3 E th an
7 ye ar s 0 m on th
M al e
C au ca si an
A SD
K in de rg ar te n
65 (E xt re m el y Lo
w )
75 (M
od er at el y Lo
w )
85 (L ow
A ve ra ge )
71 (B el ow
E xp ec te d)
N ot e. St an da rd
sc or es
an d de sc ri pt or s ar e pr ov id ed
on pa rt ic ip an t fu ll sc al e IQ
sc or e, V in el an d- 3 ad ap tiv e be ha vi or
co m po si te
(S pa rr ow
et al ., 20 16 ), Pe ab od y Pi ct ur e
V oc ab ul ar y T es t (P PV
T -5 ; D un
n 20 19 ) st an da rd
sc or e, an d th e E xp re ss iv e O ne
W or d Pi ct ur e V oc ab ul ar y T es t (E O W PV
T -4 ; M ar tin
& B ro w ne ll,
20 11 ) st an da rd
sc or e. T he
W ec hs le r In te lli ge nc e Sc al e fo r C hi ld re n (W
IS C -V ; W ec hs le r, 20 14 ) w as
us ed
to ob ta in
fu ll sc al e IQ
sc or es
fo r Ja ck so n,
C ar te r, R ya n,
an d E th an
an d th e
W ec hs le r Pr es ch oo la nd
Pr im
ar y Sc al e of
In te lli ge nc e (W
PP SI -I V ;W
ec hs le r, 20 12 ) w as
us ed
to ob ta in
fu ll sc al e IQ
sc or es
fo r Fl yn n an d M ar sh al l.
Julia L. Ferguson et al.1878
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
room in the platform for the experimenter and participants to conduct probe and teaching ses- sions (described later). Jackson and Ryan could independently enter and exit the breakout rooms without adult assistance. Carter and Ethan requested parental assistance to enter and exit the breakout rooms for research ses- sions. Flynn and Marshall were not in the same ABA-classroom as the other participants. Flynn and Marshall were always in the same location together for research sessions, and an adult set up the iPad and ensured that the video, audio, and internet connection for the research session were working appropriately. An adult also helped ensure that Flynn and Marshall could not hear each other and were not in the same room while the experimenter ran probe sessions (described later) with each participant. During probe and teaching sessions
(described later), the experimenter shared her screen, showing a Microsoft® PowerPoint pre- sentation with slides displaying the target stim- uli. Prior to each session, the experimenter ensured the slides were in the correct order based on the datasheet for that session. All ses- sions were recorded using the screen recording function for the purposes of data collection, interobserver agreement, and treatment fidelity. Target stimuli were unknown X-Men® Marvel®
comic book characters. These stimuli were chosen because all participants had current goals targeting superheroes and their corresponding superpowers. Stimuli were taught to each participant in pairs for a total of four stimuli per dyad. Table 2 displays the target stimuli and corresponding instructive feed- back for each participant.
Response Measurement Data were collected on participant acquisi-
tion of primary responses, secondary responses, primary observational responses, and secondary observational responses. Correct primary responses were defined as the participant vocally
stating the name of the stimulus that was directly taught to them in the presence of that stimulus (e.g., saying “Storm” in the presence of the picture of Storm). Correct secondary responses were defined as the participant vocally responding to a question about the superhero’s power (e.g., “What is his power?”) in the pres- ence of the stimulus. A secondary response by the participant did not have to have point-to-point correspondence with the instructive feedback pro- vided by the experimenter during teaching sessions but did need to be a functionally equivalent response. For example, when shown the picture of Beast and asked, “What’s his superpower?”, the secondary response “He transforms into an ani- mal” would be accepted as correct even though the instructive feedback was “He can turn into an animal.” The definitions of correct primary observa- tional responses and secondary observational responses were identical to those of the primary and second- ary responses, but for the stimuli that were primary targets for the other member of the dyad. Data were collected on the percentage of independent correct responses by dividing the number of inde- pendent correct responses by the total number of opportunities to respond in a probe session and multiplying by 100. Observers also collected data on participant
attending, echoics and comments emitted after the experimenter stated the secondary target, and teaching session duration. Attending was defined as the participant orienting, looking, or pointing toward the screen. Partial interval recording was used to collect data on partici- pant attending with each trial of the teaching session considered an interval. The interval began when the experimenter provided the instruction for the trial and ended after the experimenter provided the instructive feedback. Echoics were defined as the participant emitting an auditory vocal response with formal similar- ity to the verbal stimulus (i.e., instructive feed- back). Comments were defined as the participant emitting an auditory vocal response about the target or instructive feedback that
1879Instructive Feedback via Telehealth
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
had no formal similarity to the instructive feedback.
Interobserver Agreement and Treatment Fidelity A second independent observer recorded par-
ticipant responding during probe sessions from the screen-recorded sessions. Interobserver agreement was calculated by taking the number of agreements trial by trial, dividing them by the total number of trials, and multiplying by 100. These data were collected for 35.3% of baseline probe sessions, 45.7% of intervention probe sessions, and 33.3% of maintenance probe sessions. Mean agreement was 99.5% (range, 93.7-100%) across all dyads. The same independent observer also
recorded the experimenter’s implementation during probe and teaching sessions to assess treatment fidelity. During probe sessions, the independent observer collected data on the experimenter (a) showing the target stimulus, (b) providing an instruction, (c) allowing up to 10 s for the participant to respond, (d) providing neutral feedback, (e) providing instructive feedback, (f) allowing up to 10 s for the participant to respond, and (g) providing neutral feedback. During teaching sessions, the independent observer collected data on the experimenter (a) showing the target stimulus,
(b) stating the child’s name, (c) providing an instruction, (d) allowing up to 10 s for the par- ticipant to respond, (e) providing praise if the participant was correct or providing corrective feedback if the participant was incorrect, and (f) providing instructive feedback. Treatment fidelity was calculated by taking the number of trials in which the experimenter implemented all steps correctly, dividing by the total number of trials per session, and multiplying by 100. Data on treatment fidelity were collected for 29.4% of baseline probe sessions, 34.2% of probe sessions during intervention, 22.2% of probe sessions during maintenance, and 31.5% of teaching sessions across dyads. Mean treat- ment fidelity was 99.5% (range, 93.8- 100%) for probe sessions during baseline, intervention, and maintenance and 97.9% (range, 93.8-100%) during teaching sessions.
General Procedures Sessions were conducted 5 days per week.
The study had three conditions: baseline, inter- vention, and maintenance. During baseline and maintenance, the experimenter only conducted probes. During intervention, the experimenter conducted one probe and one teaching session twice per day, for a total of two probe sessions and two teaching sessions each day. Daily pro- bes always occurred prior to teaching sessions,
Table 2
Target Stimuli and Instructive Feedback
Dyad Participant Stimulus Instructive Feedback
Dyad 1 Jackson Colossus Gambit
Super strong Makes things explode
Carter Beast Rogue
Turns into an animal Takes other people’s powers
Dyad 2 Flynn Storm Gambit
Controls the weather Makes things explode
Marshall Beast Rogue
Turns into an animal Takes other people’s powers
Dyad 3 Ryan Storm Gambit
Controls the weather Makes things explode
Ethan Beast Rogue
Turns into an animal Takes other people’s powers
Julia L. Ferguson et al.1880
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
followed by a 1 to 5 min break. Sessions lasted between 2-5 min. The mastery criterion was 100% correct responses during probe sessions for the primary response across three consecutive sessions. Similar to Leaf et al. (2017), if one of the participants in the dyad reached the mastery criterion for their primary targets before the other par- ticipant, no other probes were conducted for that participant, but they continued to participate in teaching sessions until the other member of the dyad reached the mas- tery criterion.
Probes Probe sessions occurred in a one-to-one
format with each participant during all condi- tions (i.e., baseline, intervention, maintenance). Jackson, Carter, Ryan, and Ethan entered a breakout room for the probe session. For Flynn and Marshall, an adult ensured that only one of them was in the room for the probe session; the other participant was in another room and could not hear the responses. The experimenter conducted a total of 16 trials per probe session, presenting each stimulus four times. The exper- imenter asked the participant the character’s name (i.e., primary response, primary observa- tional response) during each trial and gave the participant an opportunity to provide the information delivered as part of instructive feedback during intervention for each stimulus (i.e., secondary response, secondary observa- tional response). The order of stimuli was ran- domly determined prior to each session. Probe sessions began with the experimenter sharing their screen which contained a PowerPoint pre- sentation with 16 slides (i.e., one slide per trial). Each slide was numbered, had a white background, and contained the target stimulus for each trial (e.g., a picture of an X-Man). Each trial began by the experimenter presenting the stimulus on their screen and providing an instruction to respond (e.g., “What is his
name?”) and waiting up to 10 s for the partici- pant to respond. Regardless of the participant’s response, the experimenter provided neutral feedback (e.g., “okay,” “alright,” “thanks”). After providing neutral feedback, the stimulus remained on the screen and the experimenter then asked another question relating to the instructive feedback (e.g., “What is his super- power?”), gave the participant up to 10 s to respond, and provided neutral feedback. This continued until all 16 trials were completed. The experimenter provided no prompting, feedback, or programmed reinforcement for correct or incorrect responses during pro- bes, with the exception of praise and feed- back for compliance, attending, and general learning-to-learn behaviors (e.g., sitting still, looking at the screen, not touching the computer or iPad).
Intervention Teaching sessions occurred in a dyad
arrangement. Sessions for Jackson, Carter, Ethan, and Ryan occurred in a breakout room. Flynn and Marshall sat in chairs in front of the iPad so that they could both see the screen and the experimenter could see both participants. The experimenter conducted a total of 16 teach- ing trials, four trials per target stimulus, for a total of eight trials per participant. The order of teaching trials was randomized prior to sessions. Identical to the probe sessions, the experi- menter shared their screen with the PowerPoint slides containing the target stimulus. The experimenter then stated the participant’s name in the instruction to indicate which participant should respond (e.g., “Flynn this one is for you. What is her name?”) and gave the partici- pant up to 10 s to respond. If the participant responded correctly, the experimenter provided praise and instructive feedback (e.g., “Yep you got it! And she can control the weather.”). If the participant responded incorrectly, the experimenter provided corrective feedback and
1881Instructive Feedback via Telehealth
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
instructive feedback (e.g., “No, this is Gambit. And he can make things explode.”). The exper- imenter ignored any responses made by partici- pants after providing instructive feedback and presented the next trial. If the other member of the dyad attempted to respond on a trial that was not for them, the experimenter provided feedback (e.g., “Marshall, remember to only answer when it’s your turn”) and provided praise for waiting appropriately in future trials. This only occurred once with dyad 2. The experimenter ended the teaching session after presenting all 16 teaching trials. Teaching ses- sions concluded when both participants in the dyad had reached the mastery criterion during probe sessions.
Maintenance Maintenance probes occurred 3, 7, and
9 days after the performance of both partici- pants in the dyad reached the mastery criterion. One probe session occurred per day and was identical to baseline and intervention probe ses- sions. This maintenance schedule was selected based on that used in a previous instructive feedback study (i.e., Leaf et al., 2017).
Experimental Design A nonconcurrent multiple baseline design
across dyads (Watson & Workman, 1981) was used to assess the effects of DTT with instruc- tive feedback delivered via telehealth on the acquisition of tact relations.
Results
Figure 1 displays the percentage of correct primary (left panel) and secondary responses (right panel) during probe sessions for all par- ticipants. During baseline, all participants responded incorrectly on all probe trials. For dyad 1, Carter reached the mastery criterion for primary responses in six sessions and Jackson reached the mastery criterion in seven sessions. Both Jackson and Carter began to respond
correctly in 100% of trials for secondary responses after three teaching sessions. For dyad 2, Flynn reached the mastery criterion in five sessions and Marshall reached the mastery crite- rion in seven sessions. Flynn and Marshall began to respond correctly in 100% of trials for secondary responses after two teaching sessions, with Marshall engaging in more variable responding throughout the rest of intervention. For dyad 3, both Ryan and Ethan met the mas- tery criterion for primary responses in five ses- sions. Ryan responded correctly in 100% of trials for his secondary responses after one teaching session and Ethan responded correctly in 100% of trials for his secondary responses after four teaching sessions. Across maintenance probes, all participants responded correctly dur- ing 100% of trials for their primary responses and demonstrated high levels of responding for secondary responses. Figure 2 displays the percentage of correct
primary observational responses and secondary observational responses. During baseline, all participants responded incorrectly in all trials. For dyad 1, Jackson began to respond correctly in 100% of trials for primary observational responses after two teaching sessions and Carter began to respond correctly in 100% of trials for primary observational responses after four teaching sessions. For the secondary observa- tional responses, Jackson began to respond cor- rectly in 100% of trials after two teaching sessions and Carter began to respond correctly in 100% of trials after four teaching sessions. For dyad 2, Flynn responded correctly in 100% of trials for the primary observational responses after three teaching sessions and Mar- shall responded correctly in 100% of trials after six teaching sessions. Flynn responded correctly in 100% of trials for the secondary observa- tional response after one teaching session. However, Marshall required seven teaching ses- sions (i.e., up to the last teaching session for this dyad) before he responded correctly in 100% of trials for his secondary observational
Julia L. Ferguson et al.1882
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
responses. For dyad 3, Ryan began to respond correctly in 100% of trials for the primary observational response after two teaching ses- sions. Ethan never reached 100% correct responding for the primary observational responses. After the first teaching session, Ethan responded correctly in 50% of the trials for the primary observational responses and
remained at this level for the remainder of the intervention and maintenance probe sessions. Ryan responded correctly in 100% of trials for the secondary observational responses after one teaching session. However, Ethan never dem- onstrated acquisition of his secondary observa- tional responses. All participants but Ethan engaged in high levels of their primary and
Figure 1 Participant Responding for Primary and Secondary Responses across Dyads
1883Instructive Feedback via Telehealth
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
secondary observational responses during the maintenance probes. Ethan’s responding remained at baseline levels (i.e., 0% correct responding) during maintenance. Total teaching time to mastery was 20 min,
12 s and 17 min, 48 s for Jackson and Carter (dyad 1), respectively; 18 min, 30 s and 23 min, 47 s for Flynn and Marshall (dyad 2),
respectively; and 15 min, 46 s for both Ryan and Ethan (dyad 3). All participants engaged in high rates of attending during teaching sessions (100% of intervals for Jackson and Carter, 99.1% of intervals [range, 93.8-100%] for Flynn, 98.2% of intervals [range, 93.8-100%] for Marshall, 96.3% of intervals [range, 93.8-100%] for Ryan, and 98.8% of intervals
Figure 2 Participant Responding for Primary and Secondary Observational Responses across Dyads
Julia L. Ferguson et al.1884
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
[range, 93.8-100%] for Ethan). Tables 3 and 4 display the number of echoics and comments that each participant emitted after the experi- menter delivered instructive feedback. Jackson emitted at total of four echoics and 31 com- ments, and Carter emitted a total of three echoics and 57 comments across all teaching sessions. Flynn emitted a total of 17 echoics and 27 comments, and Marshall emitted a total of 10 echoics and 27 comments. Ryan emitted a total of 17 echoics and 20 comments, and Ethan emitted a total of 17 echoics and 9 comments.
Discussion
All six participants acquired their respective primary and secondary responses, and five of six participants acquired the primary and sec- ondary observational responses while receiving DTT with instructive feedback in a dyad arrangement directly via telehealth. Perfor- mance also maintained for up to 9 days follow- ing the training. The intervention was both effective and efficient. Jackson, Carter, Flynn, Marshall, and Ryan required less than 24 min of teaching to acquire eight responses, and Ethan required less than 16 min to acquire five responses. These outcomes are similar to those demonstrated in previous studies on DTT with instructive feedback delivered in person (Albarran & Sandbank, 2019). Results extend the literature on ABA telehealth services by
evaluating the effectiveness of interventions delivered directly to learners with minimal adult assistance. Although results indicate that the direct pro-
vision of DTT via telehealth is both effective and efficient, it is important to consider how participant characteristics may relate to the out- comes. All participants demonstrated high levels of attending (i.e., 93.8-100% of trials per session) and engagement (i.e., echoics, com- ments) during teaching sessions beyond responding to the experimenter’s instruction on their respective trials. This level of attending and engagement may be important for clients to learn through direct telehealth intervention and may explain the acquisition of the observa- tional responses and those related to the instructional feedback. Nottingham et al. (2020) discussed the potential role of a bidirectional naming repertoire in the acquisition of second- ary target responses without direct teaching. A bidirectional naming repertoire is a higher- order operant in which there is a two-way rela- tionship between speaker and listener behaviors in which teaching one of the behaviors is enough to establish both repertoires without direct training (Horne & Lowe, 1996; Miguel, 2016). All participants emitted overt echoics immediately after the instructive feed- back, but the majority of participants emitted comments more frequently. Comments about targets or the instructive feedback ranged from short statements like “cool” or “yeah, she can”
Table 3
Echoics Emitted after Instructive Feedback
Dyad 1 Dyad 2 Dyad 3
Teaching Session Jackson Carter Flynn Marshall Ryan Ethan
1 3 1 1 0 1 0 2 1 1 2 1 2 4 3 0 1 2 3 4 6 4 0 0 3 2 10 6 5 0 0 5 1 0 1 6 0 0 4 2 -- -- 7 0 0 0 1 -- --
1885Instructive Feedback via Telehealth
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
to longer comments such as “oh yeah, he’s got gains” (referring to the character’s muscles) or “like a volcano” (referring to Gambit’s super- power). Ethan’s behavior differed from the other participants in that he emitted more echoics than comments during teaching ses- sions. In addition, he emitted only one echoic that related to a secondary observational response. Thus, although participants engaged in behaviors consistent with a bidirectional naming repertoire, overt echoics did not occur at high rates consistently across trials or sessions. Rodriguez (2020) suggested assessing client’s
attending, responsiveness to prompting, and frequency and intensity of challenging behavior to determine if a client would be appropriate for direct telehealth services. Participants in this study demonstrated high levels of attending during teaching trials (i.e., 93.8-100% per ses- sion), responded to vocal corrective feedback, and did not engage in frequent challenging behavior. Therefore, the participants demon- strated behaviors that align with Rodriguez’s telehealth model selection assessment. Nonetheless, Ethan acquired only one pri-
mary observational response (i.e., Storm), and he did not learn any of the secondary observa- tional responses. It is unclear if use of a telehealth modality was responsible for this out- come because the experimenters did not
evaluate the effects of in-person DTT with instructive feedback in a dyad arrangement prior to this study. Although Ethan demon- strated at least one behavior related to attend- ing in the majority of trials, he may have been distracted by other variables in his home when it was Ryan’s turn to answer during teaching sessions. A limitation of services delivered via telehealth is that the interventionist has less control over the environment than when pro- viding in-person intervention. Televisions may be on, siblings may walk through the room, phones can ring, and other distractions can occur that are outside of the control of the interventionist. While these types of distrac- tions can still occur during intervention pro- vided in-person, the effects can be more difficult to mitigate when services are provided via telehealth. For example, if a sibling entered into the room during a session provided in- person and the child was distracted, the inter- ventionist would be able to see that the sibling entered the room and could wait to issue fur- ther instructions until the sibling leaves the room or until the child is no longer distracted by their sibling before resuming instruction. During telehealth sessions, the interventionist is only able to see a portion of the child’s envi- ronment and may be unaware of distractions present in the child’s environment and there- fore unable to alter their behavior to be
Table 4
Comments Emitted after Instructive Feedback
Dyad 1 Dyad 2 Dyad 3
Teaching Session Jackson Carter Flynn Marshall Ryan Ethan
1 7 11 8 5 8 4 2 7 8 1 4 5 1 3 5 9 5 5 4 1 4 4 5 5 5 2 2 5 4 12 4 2 1 1 6 2 7 3 3 -- -- 7 2 5 1 3 -- --
Julia L. Ferguson et al.1886
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
responsive to those distractions. Ethan’s full- scale IQ score was also lower than that of the other participants and differed from the IQ scores reported for participants in previous group instructive feedback studies (e.g., Leaf et al., 2017). Anecdotally, Ethan also did not seem as interested as the other participants in learning about the superheroes. The other par- ticipants made comments about wanting to know the superheroes’ names and powers in baseline while Ethan never made any com- ments or showed any interest in learning about the targets prior to intervention. One factor Rodriguez (2020) did not men-
tion is the type of reinforcement used during direct telehealth services. In this study, the experimenter delivered only verbal praise to reinforce participant responding. Ethan’s per- formance may have improved if the experi- menter had provided more potent reinforcers (e.g., tangibles, edibles) for correct responding. Assessing prerequisite skills for observational learning (e.g., generalized imitation) over a telehealth platform prior to selecting this type of instructional arrangement could also be important to determine if small group or dyad DTT would be appropriate for a client. An additional variable that could be impor-
tant is a client’s independent use of the telehealth platform. Some participants in this study could navigate the Zoom™ platform without adult assistance, while others needed adult assistance to enter/exit breakout rooms or when technology issues arose. Technology issues were rare but did occur on a couple of occasions. Carter’s computer froze and the Zoom™ program shut down when he attempted to enter a breakout room. He required his parent’s help to restart the com- puter and re-enter the meeting. For Marshall and Flynn, connectivity issues interfered with one session, causing a lag on their end between slide transitions in the PowerPoint presenta- tion. All of these issues were easily remedied with adult assistance and communication
between the experimenter and caregivers, but it is another factor to consider when implementing telehealth services directly. Although this study adds to the current
behavior analytic literature on telehealth, more research is needed. This study evaluated only one type of behavioral intervention (i.e., DTT with instructive feedback in a dyad arrangement) and one type of skill. Fur- ther research should evaluate the effectiveness of direct intervention via telehealth for teach- ing a variety of skills, such as play, social, and listener behavior. This study also only used praise as a reinforcer for correct responses. Future researchers should conduct research with learners that may require other types of reinforcers (e.g., tangibles, edibles, token economies) to see if the same effects can be replicated via telehealth. Teaching sessions were also of very short duration (i.e., 2-5 min sessions), which differs from those of most students with ASD, who may receive up to 8 hr of services per day. It is important to continue to maximize
the effects of behavioral intervention even when ABA service delivery is provided via telehealth. This study demonstrated that one way to maximize learning for individuals diagnosed with ASD is to implement instruc- tive feedback in a dyad arrangement via Zoom™ (or other video teleconferencing platforms), which allows learners to not only be directly taught new tact relations, but acquire additional information through instructive feedback and by observing the instruction of their peers.
REFERENCES
Albarran, S. A., & Sandbank, M. P. (2019). Teaching non-target information to children with disabilities: An examination of instructive feedback literature. Journal of Behavioral Education, 28, 107–140. https://doi.org/10.1007/s10864-018-9301-3.
Council of Autism Service Providers (2020). Practice parameters for telehealth-implementation of applied
1887Instructive Feedback via Telehealth
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
behavior analysis: Continuing of care during COVID-19 pandemic. Author.
Dunn, D. M. (2019). Peabody picture vocabulary test (5th ed.) Pearson Assessments.
Eikeseth, S., Smith, D. P., & Klintwall, L. (2014). Dis- crete trial teaching and discrimination training. In J. Tarbox, D. Dixon, P. Sturmey, & J. L. Matson (Eds.), Handbook of early intervention for autism spec- trum disorders (1st ed., pp. 229–253). https://doi.org/ 10.1007/978-1-4939-0401-3_12.
Ferguson, J., Craig, E. A., & Dounavi, K. (2019). Telehealth as a model for providing behaviour ana- lytic interventions to individuals with autism spec- trum disorder: A systematic review. Journal of Autism and Developmental Disabilities, 49, 582–616. https:// doi.org/10.1007/s10803-018-324-5.
Horne, P. J., & Lowe, C. F. (1996). On the origins of naming and other symbolic behavior. Journal of the Experimental Analysis of Behavior, 65(1), 185–241. https://doi.org/10.1901/jeab.1996.65-185.
Leaf, J. B., Cihon, J. H., Alcalay, A., Mitchell, E., Townley-Cochran, D., Miller, K., … McEachin, J. (2017). Instructive feedback embedded within group instruction for children diagnosed with autism spec- trum disorder. Journal of Applied Behavior Analysis, 50(2), 304–316. https://doi.org/10.1002/jaba.375.
Leaf, J. B., Cihon, J. H., Leaf, R., McEachin, J., & Taubman, M. (2016). A progressive approach to dis- crete trial teaching: Some current guidelines. Interna- tional Electronic Journal of Elementary Education, 9(2), 361–372.
Martin, N. A., & Brownell, R. (2011). Expressive one- word picture vocabulary test (4th ed.). Academic Ther- apy Publications.
Miguel, C. F. (2016). Common and intraverbal bidirec- tional naming. The Analysis of Verbal Behavior, 32, 125–138. https://doi.org/10.1007/s40616-016- 0066-2.
Nottingham, C. L., Vladescu, J. C., DeBar, R. M., Deshais, M., & DeQuinzio, J. (2020). The influence of instructive feedback presentation schedule: A repli- cation with children with autism spectrum disorder. Journal of Applied Behavior Analysis. Advance online publication. https://doi.org/10.1002/jaba.706.
Rodriguez, K. A. (2020). Maintaining treatment integrity in the face of crisis: A treatment selection model for transitioning direct ABA services to telehealth. Behav- ior Analysis in Practice, 13(2), 291–298. https://doi. org/10.1007/s40617-020-00429-8.
Sparrow, S. S., Cicchetti, D. V., & Saulnier, C. A. (2016). Vineland-3: Vineland adaptive behavior scales (3rd ed.). Inc: NCS Pearson.
Watson, P. J., & Workman, E. A. (1981). The non- concurrent multiple baseline across-individuals design: an extension of the traditional multiple base- line design. Journal of Behavior Therapy and Experi- mental Psychiatry, 12(3), 257–259. http://doi.org/10. 1016/0005-7916(81)90055-0.
Wechsler, D. (2012). Wechsler preschool and primary scale of intelligence-fourth edition- technical manual and interpretive manual (4th ed.) Pearson Assessments.
Wechsler, D. (2014). Wechsler intelligence scale for children- fifth edition-technical manual and interpretive manual (5th ed.) Pearson Assessments.
Werts, M. G., Caldwell, N. K., & Wolery, M. (2003). Instructive feedback: Effects of a presentation variable. The Journal of Special Education, 37(2), 124–133. https://doi.org/10.1177/00224669030370020601.
Zoom. (2020, May). HIPAA compliance guide. https:// zoom.us/docs/doc/Zoom-hipaa.pdf
Received May 11, 2020 Final acceptance August 24, 2020 Action Editor, Dorothea Lerman
Julia L. Ferguson et al.1888
19383703, 2020, 4, D ow
nloaded from https://onlinelibrary.w
iley.com /doi/10.1002/jaba.773 by W
estfield State U niversity, W
iley O nline L
ibrary on [20/06/2023]. See the T erm
s and C onditions (https://onlinelibrary.w
iley.com /term
s-and-conditions) on W iley O
nline L ibrary for rules of use; O
A articles are governed by the applicable C
reative C om
m ons L
icense
