A Critical Review of Social Narratives

Abstract

Social narratives, or story-based interventions, are defined as stories that describe social situations, appropriate social behaviors to display, and when to display the specified behaviors. Social narratives are a commonly implemented and empirically evaluated procedure used to improve social behavior and decrease the probability of aberrant behavior for individuals diagnosed with autism spectrum disorder. Although social narratives are a commonly implemented and evaluated procedure, recommendations about their use and effectiveness is conflicting. This paper reviews six interventions that fit the definition of social narratives (i.e., Social Stories™/social stories, social scripts, cartooning, comic strip conversations, power cards, and social autopsies). Fifteen articles were analyzed across multiple methodological dimensions to determine the level of evidence (i.e., convincing, partial, or not convincing). Results of the analysis indicated that the majority of social narrative studies did not demonstrate convincing evidence. Recommendations for clinicians and future research are discussed based on the results of the literature review.

Full text

REVIEW ARTICLE
A Critical Review of Social Narratives
Justin B. Leaf
1,2
&Julia L. Ferguson
1
&Joseph H. Cihon
1,2
&Christine M. Milne
1,2
&
Ronald Leaf
1
&John McEachin
1
#Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
Social narratives, or story-based interventions, are defined as stories that describe social
situations, appropriate social behaviors to display, and when to display the specified
behaviors. Social narratives are a commonly implemented and empirically evaluated
procedure used to improve social behavior and decrease the probability of aberrant
behavior for individuals diagnosed with autism spectrum disorder. Although social
narratives are a commonly implemented and evaluated procedure, recommendations
about their use and effectiveness is conflicting. This paper reviews six interventions that
fit the definition of social narratives (i.e., Social Stories™/social stories, social scripts,
cartooning, comic strip conversations, power cards, and social autopsies). Fifteen
articles were analyzed across multiple methodological dimensions to determine the
level of evidence (i.e., convincing, partial, or not convincing). Results of the analysis
indicated that the majority of social narrative studies did not demonstrate convincing
evidence. Recommendations for clinicians and future research are discussed based on
the results of the literature review.
Keywords Social narratives .Social stories .Story based .Story based intervention
Individuals diagnosed with autism spectrum disorder (ASD) and other
neurodevelopmental disorders have a right to effective treatment (e.g., Van Houten
et al. 1988). The effectiveness of behavior analytic interventions is also one of the
dimensions of applied behavior analysis (Baer et al. 1968). Van Houten and colleagues
(Van Houten et al. 1988)statedthat“behavior analysts have an obligation to only use
techniques that have been demonstrated by research to be effective, to acquaint
consumers and the public with the advantages and disadvantages of these techniques,
and to search continuously for the most optimal means of changing behavior”(p. 383).
Journal of Developmental and Physical Disabilities
https://doi.org/10.1007/s10882-019-09692-2
*Justin B. Leaf
Jblautpar@aol.com
1
Autism Partnership Foundation, 200 Marina Dr., Seal Beach, CA 90740, USA
2
Endicott College, Beverly, MA, USA

Furthermore, certifying boards have stated that it is the ethical responsibility of
behavior analysts to have a reliance on scientific knowledge and provide intervention
based upon scientific research (Behavior Analyst Certification Board 2014). Not only
are behavior analysts obligated to use techniques backed by scientific research, but
educators (Wong et al. 2013), speech pathologists (Dollaghan, 2007), and other service
providers working with individuals diagnosed with ASD are also required to use
practices that have been demonstrated to be effective in the scientific literature.
When determining what interventions to implement there are several criteria that
should be evaluated. First, individuals should ensure that interventions are evi-
dence-based, meaning they rely on clinical expertise, patient preference, and most
importantly the best empirical evidence (National Autism Center 2015a). Further-
more, one should evaluate more stringent definitions of what is considered to be
evidence-based (e.g., Cook and Cook 2011) to ensure the interventions they en-
dorse, recommend, or implement meet the criteria to be evidence-based. Consider-
ing a more stringent criterion ensures interventionists are looking at all three prongs
of evidence-based practice as opposed to just one or two. Second, one should ensure
the procedures they implement have empirical support. This would mean that there
have been empirical studies conducted on the interventions and/or the principles
behind these interventions that support their effectiveness. The quality of empirical
evidence can vary greatly and when evaluating the published, peer-reviewed re-
search behind an intervention, individuals providing intervention for individuals
diagnosed with ASD should look for important quality indicators including clear
operational definitions, treatment fidelity data, and a functional relationship dem-
onstrated through the correct implementation of the experimental design (Horner
et al. 2005). Finally, the evidence found from any comparative research studies
should be evaluated to ensure that the targeted intervention is the most effective
procedure available.
To help determine which procedures are effective, evidence-based, and have empir-
ical support, professionals have written commentaries (e.g., Horner et al. 2005), review
papers (Wong et al. 2015), meta-analyses (Roth et al. 2014), and standards reports (e.g.,
National Autism Center 2015b). Additionally, researchers have conducted experimental
analyses to evaluate the effectiveness of a variety of interventions. Sometimes these
papers have resulted in universally accepted recommendations for clinicians. For
example, reviews, commentaries, and standards projects have universally concluded
that facilitated communication and rapid prompting method are not evidence-based and
should not be implemented (e.g., National Autism Center 2015b). On other occasions,
professionals have reached differing conclusions about the effectiveness of various
interventions. For example, there have been several review papers which have warned
against the implementation of Social Stories™due to the lack of empirical evidence
(e.g., Leaf et al. 2015; Reynhout and Carter 2011; Styles 2011). Yet, other commen-
taries and standards projects state that professionals could implement Social Stories™
as they meet the criteria to be evidence based (e.g., National Autism Center 2015b).
When conflicting recommendations such as these exist, behavior analysts may be
uncertain how to proceed.
Social narratives, also known as story-based interventions (referred to as social
narratives throughout this paper), is an intervention that is implemented for individuals
diagnosed with ASD (e.g., National Autism Center 2015b). Social narratives have been
Journal of Developmental and Physical Disabilities

defined as stories, represented visually, that describe various social situations, appro-
priate social behaviors to be displayed, and when these behaviors should be displayed
(Zimmerman and Ledford 2017). Given this broad definition there are numerous
interventions that meet this criterion including social scripts (Loveland and Tunali
1991), Social Stories™(Gray and Garand 1993), cartooning (Coogle et al. 2017),
Comic Strip Conversations™(Gray 1994a), power cards (Campbell and Tincani 2011),
and social autopsies (Bieber 1994). This broad definition has resulted in professionals
providing different operational definitions for what constitutes a social narrative. In
some instances, social narratives are used synonymously with Social Stories™(e.g.,
Zimmerman and Ledford 2017), a trademarked intervention where professionals write
a story following guidelines that have been created by Carol Gray (Gray 1994b). In
other instances, social narratives are used to define non-trademarked social stories (i.e.,
stories that are written but that do not follow the guidelines laid out by Gray; e.g.,
Adams et al. 2004). In other instances, social narratives have also been used as a
hypernym that includes all types of visually based stories to teach skills to individuals
diagnosed with ASD (e.g., Social Stories™, comic strip conversations, and social
autopsies; Wong et al. 2013). The lack of a clear operational definition may lead to
confusion of what actually constitutes a social narrative.
For the purpose of this paper we define social narratives as a hypernym for six
specific interventions: (1) Social Stories™/ social stories, (2) social scripts; (3) car-
tooning; (4) comic strip conversations; (5) power cards; and (6) social autopsies. Social
narratives, as defined here, have been described as an evidence-based intervention by
the National Autism Standards Project Phase 2 (National Autism Center 2015b)andby
the National Professional Development Center on ASD (Wong et al. 2013). Yet other
professionals have warned against the implementation of some of the interventions
(e.g., Social Stories™) that fall under the hypernym of social narratives (e.g.,
Zimmerman and Ledford 2017). Furthermore, there may be little to no empirical
research and/or reviews on other interventions that fall under the hypernym “social
narratives”(e.g., comic strip conversations, social autopsies, and power cards).
Therefore, the purpose of this paper is to provide an evaluation and discussion of
social narratives as they relate to interventions for individuals diagnosed with ASD
and other neurodevelopmental disorders. In doing so, we describe each of the
interventions that fall under the umbrella of social narratives. Next, we conduct a
systematic search of the literature. Then we evaluate the studies found on quality
indicators of experimental control (e.g., objective measurement, correct implementa-
tion of experimental designs) to classify each study as providing convincing evidence,
partial convincing evidence, or not convincing evidence for the social narrative
intervention. After experimental control has been evaluated and the articles have been
classified, we then assess the quantity of empirical studies that were found to have
convincing evidence to conclude if social narrative interventions would be considered
to be evidence based. In this review, we include five of the six social narrative
interventions (i.e., social scripts, cartooning, comic strip conversations, power cards,
and social autopsies). It should be noted, we are not reviewing Social Stories™/social
stories as there have been several reviews on that specific type of social narratives
(e.g., Reynhout and Carter 2011;Styles2011). While Social Stories™/social stories
were not included in the literature review we do include them in our discussion of
social narratives more generally.
Journal of Developmental and Physical Disabilities

Social Narrative Interventions
Social Scripts
Social scripts are a basic social narrative strategy in which the interventionist
provides the learner with verbal statements or questions that they can use during
social situations (Loveland and Tunali 1991). Within social scripts the intervention-
ist typically has the learner practice using the script in contrived/analogue learning
sessions as a way for the learner to become more fluent with the script. Once the
learner displays the script appropriately in the analogue setting then the interven-
tionist trains for generalization for the learner to display the skill in novel situations
or with novel people.
Cartooning
Cartooning is a form of a social narrative where the interventionist draws a cartoon to
help explain the social situation and what behaviors the learner should engage in during
different social situations (Coogle et al. 2017). Within cartooning, the interventionist
also draws thought bubbles to display what the cartoon characters are feeling and
thinking.
Comic Strip Conversations™
Comic Strip Conversations™are similar to cartooning in that the interventionist
draws a series of cartoons to display the social situation and behaviors the learner
should display. Comic Strip Conversations™are based on the belief that visual
supports help aid understanding and comprehension for individuals diagnosed
with ASD. Comic Strip Conversations ™are an illustrative conversation between
two or more people and are drawn out by at least two people together (Gray
1994a). Comic Strip Conversations™use a set of eight symbols to denote basic
conversation skills and also use specific colors to represent how others may feel
about certain comments, thoughts, and questions used in the Comic Strip Conver-
sation ™(Gray 1994a).
Power Cards
Power cards are a type of social narrative that use a child’s special interest or
obsession as a motivator to teach the desired social behavior through a rule
governed statement (Campbell and Tincani 2011). There are four steps when using
the power card strategy. First, the child’s interests and problem behaviors are
determined through observation and interview with individuals that know the child
well. Second, a functional behavior assessment is conducted to understand the
contingencies surrounding the problem behavior. Third, the power card and sce-
nario are developed based on the child’s special interest, current comprehension
level, and current reading level. The power card is then implemented with the child
and the intervention is evaluated (Campbell and Tincani 2011).
Journal of Developmental and Physical Disabilities

Social Autopsies
Social autopsies are one format of social narratives where the primary teaching
modality is more vocal-verbal then visual. Social autopsies occur after the learner
engages in an incorrect social behavior. At this point the interventionist has the learner
identify the incorrect social behavior, who was harmed by the incorrect social behavior,
how to correct the social mistake, and develop a plan to ensure that it does not occur in
the future (Bieber 1994).
Methods
Inclusion Criterion
To be included in this review an article had to meet the following criteria. First, articles
had to be published in a peer reviewed journal between January 1950 and November
2018. Second, the studies had to have either social scripts, cartooning, comic strip
conversations, power cards, or social autopsies as the independent variable. Third, the
study had to include at least one participant diagnosed with ASD, DD, or ID. Fourth,
articles had to be accessible either through a university online library, or available for
purchase through the journal. Finally, articles had to be written or available in English.
Search Procedure
We conducted a systematic search of the social narratives in accordance with the
PRISMA guidelines (Moher et al. 2009;seeFig.1). The PRISMA guidelines consists
of four broad levels: (1) identification of articles, (2) screening of articles; (3) eligibility
of articles, and (4) inclusion of articles. Two reviewers (referred to as primary and
secondary reviewer hence forward) were used across all levels.
The primary database that was searched was PsychINFO from the years of January
1950 to November 2018. The search terms were: [“power cards”OR “social autopsies”
OR “comic strip conversations”OR “cartooning”OR “social articles”OR “social
narratives OR “story based interventions”OR “social scripts”] all paired with “autism
spectrum disorder”or “developmental disabilities.”PsychINFO search yielded 262
articles. Two reviewers read the title and abstract of the 262 articles and retrieved the
full text of any article (n= 40) which appeared to implement one of the social narrative
interventions. Interrater agreement was taken for which articles were going to be
excluded and which articles were going to be assessed for eligibility. To conduct
interrater agreement, we calculated the number of agreements (articles chosen to assess
eligibility or excluded) over the total number of articles screened. Interrater agreement
for this measure was 100%.
Next, two reviewers evaluated the 40 articles to identify if the studies met the
inclusion criterion stated above. A total of 15 articles were deemed relevant; while
25 were deemed not to meet the inclusion criterion. Interrater agreement was taken on
which articles were going to be excluded and which articles were going to be included
for further analysis. For interrater agreement we calculated the number of agreements
Journal of Developmental and Physical Disabilities

(articles included or excluded) over the total number of articles screened. Interrater
agreement on this measure was 100%.
Measures
Experimental Control Our first measure was to evaluate the articles to assess if the
researchers implemented an experimental design that met previously described stan-
dards (Campbell and Stanley 1963; Kazdin 2011;Kennedy2005). To evaluate exper-
imental designs, we modified an evaluation of experimental designs previously imple-
mented by Leaf and colleagues (Leaf et al. 2015). That is, we used specific criteria
developed for each type of design that implemented in the research articles (i.e., AB
case design, reversal design, multiple baseline design, group design). The criteria used
to evaluate each study was based on the research design implemented. This evaluation
method was chosen because this method analyzes the experimental rigor of the design
implemented in more ways than other evaluation methods have done. For example, the
National Autism Center (2015a,b) used the Scientific Merit Rating Scale to evaluate
research articles for interventions implemented with individuals with ASD. Although
Records idenfied through database
searching
(n = 262)
ScreeningIncluded Eligibility noitacifitnedI
Addional records idenfied through
other sources
(n = 0)
Duplicate records removed
(n = 262)
Records screened
(n = 262)
Records excluded
(n = 222)
Full-text arcles assessed for
eligibility
(n = 40)
Full-text arcles excluded,
with reasons
(n = 25)
13 not a Social Narrave
intervenon
3 not published in English
9 not objecve data (e.g.,
commentary or review)
Studies included in
qualitave synthesis
(n = 15)
Studies included in
quantave synthesis (meta-
analysis)
(n = 15)
Fig. 1 PRISMA diagram
Journal of Developmental and Physical Disabilities

the Scientific Merit Rating Scale takes the research design into account (i.e., group
design, single subject designs except alternating treatment design, and alternating
treatment design), they only consider some important variables when analyzing a
research design. For example, when analyzing research using a single-subject design
other than an alternating treatment design (e.g., reversal design) a rating of 5 would be
received if the study had a minimum of three comparisons of intervention and control
conditions, if the number of data points per condition was greater than 5, the number of
participants was more than three, and there was no data loss. Although these are critical
variables to a single-subject research design, it does not evaluate other important
aspects to a correctly implemented research design such as ensuring baseline data is
trending in the correct way prior to implementing the intervention, demonstrating clear
behavior change (i.e., non-overlapping data) in the intervention condition, and the
immediacy of the effect of intervention. Below are the criteria and definitions used to
evaluate each research design.
Case Study Tab le 1displays the criteria used to evaluate and classify studies that
implemented a case study design. First, the type of data collected was evaluated (i.e.,
objective, subjective). Next, the length of baseline was evaluated. The baseline trend
was also evaluated. Stability in a baseline trend was defined as (a) two consecutive
sessions trending in the right direction, or (b) two consecutive days of stability, or (c)
three out of four sessions at a stable rate without the last data point heading in the
wrong direction, and (d) criteria had to be met across all applicable skills and/or
participants. Next, the immediacy of the effect of intervention was evaluated. Immedi-
acy of effect included evaluating if the behavior change occurred within three sessions
or after three sessions and also included an evaluation of (a) if two of the three data
points were higher (for increasing behaviors) or lower (for decreasing behaviors) than
all data points during baseline or (b) if the third data point of intervention was higher
than all data points during baseline. Overlapping data was also assessed through the
percentage of sessions where an intervention data point was the same as any baseline
Table 1 Criteria for studies using a case study design
Level of
Demonstration
Type of
Data
Length of
Baseline
Baseline Trending Effect
Immediate
Overlapping
Data
Other
Procedures
Convincing Objective 3 or more
sessions
Stable or trending
in correct
direction
Behavior
change
within 3
sessions
20–0%
overlap-
ping data
(ALL)
Not
Combined
with other
proce-
dures
Partial Objective 1 to 3
sessions
Stable or trending
in correct
direction
Behavior
change
within 3
sessions
40–21%
overlap-
ping data
(ALL)
Combined
with other
Proce-
dures
Not
Convincing
Subjective 0 sessions
or not
reported
No stability or not
trending in the
correct direction
Behavior
change
occurring
after 3
sessions
100 to 41%
overlap-
ping data
(ALL)
Combined
with other
Proce-
dures
Journal of Developmental and Physical Disabilities

level data points, Finally, case study designs were evaluated to see if the social narrative
intervention was combined with any other procedures.
Reversal Tabl e 2displays the criteria used to evaluate and classify studies that imple-
mented a reversal design. First, the type of data collected was evaluated (i.e., objective,
subjective). Next, the baseline trend was evaluated prior to implementing intervention.
This was defined as (a) two consecutive sessions trending in the right direction, or (b)
two consecutive sessions of stability, or (c) three out of four sessions at a stable rate
without the last data point heading in the wrong direction, and (d) criteria had to be met
across all participants and/or skills. The intervention condition trend was also assessed
based on this criteria: (a) the last two data points trending in the correct direction and
higher than 85% of all baseline points, or (b) the last two data points are stable and
higher than 85% of baseline sessions, or (c) three out of four sessions are at a stable
rate, are higher than 85% of all baseline sessions, without the last data point heading in
the wrong directions, and (d) criteria was met across all skills and/or participants. Next,
the behavior change was assessed. This was defined as (a) 75% of all data points in a
condition are higher (for increasing behaviors) or lower (for decreasing behaviors) than
all of the baseline data points or (b) clear level change through visual analysis. Finally,
each study that implemented a reversal design was assessed to see if the social narrative
intervention was combined with any other procedures.
Multiple Baseline Design Table 3displays the criteria used to evaluate and classify
studies that implemented a multiple baseline design. First, the type of data collected
was evaluated (i.e., objective, subjective). Next the baseline trend prior to intervention
was evaluated. This was defined as (a) two consecutive sessions trending in the right
direction, or (b) two consecutive sessions of stability, or (c) three out of four sessions at
a stable rate without the last data point heading in the wrong direction, and (d) criteria
had to be met across all participants and/or skills. Next the staggering of the interven-
tion was evaluated. The definition for staggering correctly was defined as (a) trending
correct was on the panel directly above without the previous two data points trending
Table 2 Criteria for studies using a reversal design
Level of
Demonstration
Type of
Data
All Baseline
Tren ding th e
Correct Way
Prior to Reversal
All Intervention
Trending the
Correct Way Prior
to Reversal
Clear Behavior
Change
Other
Procedures
Convincing Objective 100% of all
baseline
condition
100% of all
intervention
conditions
100 to 80% of all
intervention
conditions
Not Combined
with other
Procedures
Partial Objective 99 to 50% of all
baseline
conditions
99 to 50% of all
intervention
conditions
79 to 50% of all
intervention
conditions
Combined with
other
procedures
Not
Convincing
Subjective 49% to 0% of all
baseline
conditions
49 to 0% of all
baseline
conditions
49 to 0% of all
intervention
conditions
Combined with
other
procedures
Journal of Developmental and Physical Disabilities

the incorrect direction and (b) stability on the last two data points were higher than 80%
of the baseline data points. Next, the behavior change was assessed. This was defined
as (a) 75% of all data points in a condition are higher (for increasing behaviors) or
lower (for decreasing behaviors) than all of the baseline data points or (b) clear level
change through visual analysis. Each study that implemented a multiple baseline design
was also assessed to see if the social narrative intervention was combined with any
other procedures. Finally, studies using a multiple baseline design were evaluated on
the number of legs implemented in the multiple baseline design.
Group Design Table 4displays the criteria used to evaluate and classify studies that
implemented a group design. First the type of data collected was evaluated (i.e.,
Table 3 Criteria for studies using a multiple baseline design
Level of
Demonstration
Type of
Data
All baseline trending
the correct way prior
to intervention
Staggering
Correct
Clear
Behavior
Change
Other
Procedures
Number
of Legs
Convincing Objective 100% of all stable 100 to 75%
correctly
staggered
100 to 80%
of all
condi-
tions
Not Combined
with other
Procedures
3or
More
Legs
Partial Objective 99 to 67% stable 74 to 50%
correctly
staggered
79 to 50%
of all
condi-
tions
Combined
with other
procedures
2 Legs
Not
Convincing
Subjective 66 to 0% stable 49 to 0%
correctly
staggered
49 to 0% of
all
condi-
tions
Combined
with other
procedures
Less
than 2
legs
Table 4 Criteria for studies using a group design
Level of
Demonstration
Type of Data Groups Randomized Evaluator Pre-Test Post Test
Convincing Objective Control
Group or
2nd
Treatment
Group
Randomized,
Quasi
Randomized, or
Match to
Sample
Blind
Evaluator
Occurred
Both
Groups
Occurred
Both
Groups
Partial Standardized
Assessment
N/A N/A Research or
Teacher
Occurred
with
Treat-
ment
Group
Occurred
with
Treatment
Group
Not
Convincing
Subjective No Second
Group
No Randomized,
Quasi
Randomized, or
Match to
Sample
Child or
Caregiver
No
Pre-test
No Post-Test
Journal of Developmental and Physical Disabilities

objective, standardized assessment, subjective). Next the study was evaluated for
the groups included (i.e., control group or 2nd treatment group, no second group).
The randomization of the groups was also evaluated (i.e., randomized, quasi
randomized, or match to sample vs. no randomization, quasi randomization, or
match to sample). Studies implementing a group design were also assessed on the
type of evaluator they used and whether it was a blind evaluator, a researcher or
teacher, or a child or caregiver. Finally, each study that implemented a group design
was evaluated for if a pre-test and post-test occurred for both groups, if a pre-test
and post-test only occurred with the treatment group, or no pre-test and post-test
occurred.
Level of Demonstration Based on the criteria described above, a study was determined
as either having convincing evidence, partial evidence, or not convincing evidence. The
definitions for convincing evidence, partial evidence, and not convincing evidence for
each type of research design are displayed in Tables 1,2,3,and4. A study was
classified as convincing, partially convincing, or non-convincing based upon their
lowest score across the variables evaluated. Interrater agreement was conducted on
40% of articles evaluated for study classification. Interrater agreement was 100% for
experimental control study classification.
Quantity of Evidence Base Once the quality of evidence was assessed, our second
measure was to evaluate if the studies that had convincing evidence met the basic
quantity requirement to establish the intervention as an evidence-based practice laid out
by Horner et al. (2005). Specifically, Horner and colleagues’criteria included a
minimum of five experimental studies published in peer-reviewed journals, conducted
by at least three separate research groups, and including at least 20 participants
collectively across the studies. Thus, we evaluated the number of studies that were
found to have convincing evidence that were conducted for each social narrative
intervention published in a peer-reviewed journal. Second, we evaluated the number
of different research groups that conducted the convincing research studies on the
interventions and if they met the minimal threshold of at least three separate research
groups conducting the research studies. Finally, we evaluated the number of partici-
pants per convincing study and the total number of participants across all convincing
studies for each type of social narrative intervention.
Results
Tab le 5provides an overview of the 15 studies which were evaluated in this review and
denotes the type of social narrative intervention evaluated, the research design used, the
quality rating, and why that quality rating was given to each study. If studies were
found to have convincing evidence then the total number of studies found to be
convincing, the participants in each study, and the number of research groups
conducting the research was evaluated.
Journal of Developmental and Physical Disabilities

Table 5 Overview of studies
Authors Intervention Number of
Participants
Research Design Level of
Convincing
Evidence
Why
Designation
of Evidence
Ahmed-Husain and
Dunsmuir
(2014)
Comic Strip
Conversations™
8 Multiple baseline Not convincing Trending
baseline
levels
Angell et al. (2011) Power cards 3 Reversal Not convincing Trending
baseline
levels
Campbell and
Tincani (2011)
Power cards 3 Multiple baseline Partial convincing
evidence
Trend ing
baseline
levels
Daubert et al.
(2015)
Power cards 2 Multiple baseline Partial convincing
evidence
Trend ing
baseline
levels
Davis et al. (2010) Power cards 3 Multiple baseline Convincing
evidence
N/A
Ganz et al. (2008) Social scripts 3 Multiple baseline Not convincing Trending
baseline
levels
Hundert et al.
(2014)
Social scripts 3 Multiple baseline Partial convincing
evidence
Staggering
correctly &
other
procedures
Hutchins and
Prelock (2006)
Comic Strip
Conversations™
2 Case study Not convincing Type of data,
immediate
effect, &
overlapping
data.
Hutchins and
Prelock (2008)
Comic Strip
Conversations™
1 Reversal Not convincing Type of data,
clear
behavior
change,
Hutchins and
Prelock (2012)
Comic Strip
Conversations™
17 Multiple baseline Not convincing Type of data &
staggering
correctly
Keeling et al.
(2003)
Power cards 1 Multiple baseline Not convincing Trending
baseline
levels
Loveland and
Tunal i (1991)
Social scripts 13 Group design Partial convincing
evidence
Evaluator
Parker and Kamps
(2011)
Social scripts 2 Multiple baseline Partial convincing
evidence
Trending
baseline
levels
Pierson and Glaeser
(2005)
Comic Strip
Conversations™
4 Case study Not convincing Type of data, &
data not
reported
Pierson and Glaeser
(2007)
Comic Strip
Conversations™
3 Case study Not convincing Length of
baseline,
baseline
trend, no
immediate
effect
Journal of Developmental and Physical Disabilities

Social Scripts There was a total of four publications on social scripts published by four
different research groups found through the literature search. When evaluating the four
social script studies with respect to experimental control, three of the studies demon-
strated partially convincing evidence (i.e., Hundert et al. 2014; Loveland and Tunali
1991; Parker and Kamps 2011) while one study showed no convincing evidence (i.e.,
Ganz et al. 2008). Since none of the social script studies were found to have a
convincing level of evidence, the quantity of studies that could contribute to the
establishment of an evidence-based practice could not be assessed. Therefore, there
were zero social scripts studies to meet the quantity standards for an evidence-based
practice according to Horner and colleagues (Horner et al. 2005).
Comic Strip Conversations™A total of six publications were found from the litera-
ture search on Comic Strip Conversations™. When evaluating the six publications
with respect to experimental control (i.e., Leaf et al. 2015), all six of the studies
demonstrated no convincing evidence (i.e., Ahmed-Husain and Dunsmuir 2014;
Hutchins and Prelock 2006,2008,2012; Pierson and Glaeser 2005,2007). Since
none of the studies evaluated demonstrated convincing evidence, Comic Strip
Conversations™did meet the quantity standards to be considered an evidence-
based practice according to Horner et al. (2005).
Power Cards There were a total of five publications found on power cards. When
evaluating the five publications with respect to experimental control, only one
study was found to demonstrate a convincing level of evidence (i.e., Davis et al.
2010), two studies were found to have partial convincing evidence (i.e.,
Campbell and Tincani 2011; Daubert et al. 2015), and two were found to have
no convincing evidence (i.e., Angell et al. 2011; Keeling et al. 2003)thatpower
cards were responsible for the behavior change. The one publication that was
found to have convincing evidence was conducted by one research group and
had a total of 3 participants. Thus, there was an insufficient number of power
card studies to meet the quantity standards for an evidence-based practice
according to Horner et al. (2005).
Social Autopsies There were no peer-reviewed empirical studies that evaluated social
autopsies on changing behavior for individuals diagnosed with ASD, DD, or ID. Since
we found no peer-reviewed published studies it was impossible to determine the quality
of experimental control as outlined by Leaf et al. 2015. Therefore, social autopsies did
not meet the quantity standards to be considered an evidence-based practice according
to Horner et al. (2005).
Cartooning There were no peer-reviewed studies published evaluating cartooning
with individuals diagnosed with ASD, DD, or ID. Since we found no peer-
reviewed published studies it was impossible to determine the quality of
experimental control as outlined by Leaf et al. 2015. Thus, cartooning did
not meet the minimum quantity standards to be considered evidence-based
practice according to Horner et al. (2005).
Journal of Developmental and Physical Disabilities

Discussion
The purpose of this review was to evaluate procedures that are commonly defined
under the hypernym of social narratives. While conducting the review we assessed if
the research conducted demonstrated convincing levels of experimental control be-
tween the intervention implemented and the corresponding change in behavior as
defined by Leaf et al. (2015). We also assessed if the studies found to have convincing
levels of evidence resulted in the minimum quantity standards to be considered an
evidence-based practice as defined by Horner et al. (2005). A total of 15 studies were
found that evaluated social narratives. When evaluating if sufficient experimental
control was established across the 15 studies, only 1 study was determined to show
convincing evidence, 5 studies (i.e., 33.3%) were determined to have partially con-
vincing evidence, and 9 studies (i.e., 60%) were determined to show no convincing
evidence. Since only one study demonstrated convincing evidence, none of the inter-
ventions under the hypernym of social narratives (i.e., social scripts, Comic Strip
Conversations™, power cards, social autopsies, cartooning) met the quantity standards
to be considered an evidence-based practice as defined by Horner et al. (2005).
Therefore, the results show the interventions under the hypernym “social narratives”
do not meet the basic quantity criteria to be considered an evidence-based practice and
the majority of studies did not demonstrate a convincing functional relationship
between the intervention and corresponding change in behavior.
We did not evaluate Social Stories, a type of social narrative that is widely used,
because there have been numerous reviews that have already been written (e.g., Leaf
et al. 2015;ReynhoutandCarter2011;Rhodes2014;Styles2011). Although there is
sufficient research on Social Stories to meet the quantity criteria defined by Horner
et al. (2005), previous reviews have shown that the majority of studies only showed
partially convincing evidence (41.5% of studies) or no convincing evidence (51.2% of
studies; Leaf et al. 2015).
This review also has implications for interventionists, researchers, and entities
using and producing evidence-based standards. Interventionists have a legal,
ethical (e.g., Behavior Analyst Certification Board 2014), and moral obligation
to implement only interventions that are considered to be evidence-based and
which have been shown to be effective for the clients whom they serve, especially
when effective alternatives are available. Failure to do so could result in a waste
of time, money, and could be emotionally taxing for families to spend valuable
time on non-evidence-based interventions. Interventionists should avoid the use of
social narratives given the lack of empirical research documenting that the
interventions are responsible for observed changes. In that vein, interventionists
should implement interventions that meet the minimum quality and quantity
standards to be considered an evidence-based practice and convincing levels of
experimental control (e.g., video modeling, discrete trial teaching, behavioral
skills training).
Although, the current research has not convincingly shown that social narrative
interventions are responsible for behavior change, researchers should continue to
evaluate social narrative interventions. In doing so, researchers should first and
Journal of Developmental and Physical Disabilities

foremost ensure appropriate experimental rigor by ensuring the research design is
employed correctly to demonstrate a functional relationship between the independent
and dependent variable. Single-subject research methodology may be best suited to
evaluate the conditions under which social narrative interventions are and are not
effective. Individual differences could be clearly identified and further evaluated. If
the use of single-subject designs are effective at identifying certain conditions under
which social narratives are effective, researchers could begin to conduct comparative
studies between social narrative interventions and other established interventions (e.g.,
video modeling). The results of this line of research could be greatly beneficial for
interventionists designing and implementing interventions. That is, this research could
inform when and when not to use social narrative interventions.
Finally, this review and previous reviews also have implications for entities that
provide standards of practice (e.g., National Autism Center 2015b). First, we encourage
that these entities not only evaluate minimum quantity standards that constitute an
evidence-based practice (e.g., number of participants, number of studies) but also
evaluate if these studies have demonstrated a convincing functional relationship be-
tween the independent and dependent variables by evaluating the trends in the data
during baseline and intervention, the immediacy of the effect, and if clear behavior
change was demonstrated. Second, while we understand the potential rationales for
using hypernyms when defining evidenced or best practices, there are potential chal-
lenges. For instance, lacking clear definitions, criteria, or parameters for which proce-
dures or interventions fall within the hypernym may result in interventionists finding it
difficult to distinguish which procedures or interventions constitute the evidence base.
This could also lead to a misunderstanding about whether a procedure is or is not
considered to be evidence based. If hypernyms are unavoidable, our recommendation is
to provide clear definitions, criteria, and parameters for the intervention or procedure
andtoupdatethemfrequently.
We conducted this review on interventions that have been considered to be social
narrative interventions. While this review focused on social narratives, it is possible that
several other perceived evidence-based practices could be evaluated in a similar vein.
For instance, “Social Skills Package”is a common term to define a broad range of
intervention packages. It is possible that there are interventions that consider them-
selves or are marketed as a “Social Skills Package”with varying levels of convincing
evidence of their effectiveness. It is our hope that researchers continue to evaluate these
types of interventions, and that clinicians utilize critical thinking when evaluating the
evidence base for any purported established intervention.
Compliance with Ethical Standards
Ethical Approval This article does not contain any studies with human or animal participants performed by
any of the authors.
Informed Consent As such no informed consent was needed in this study.
Conflict of Interest None of the authors have any conflict of interests with the information presented within
this article.
Journal of Developmental and Physical Disabilities

References
Adams, L., Gouvousis, A., VanLue, M., & Wladron, C. (2004). Social story intervention: Improving
communication skills in a child with autism spectrum disorder. Focus on Autism and Other
Developmental Disabilities, 19(2), 87–94.
Ahmed-Husain, S., & Dunsmuir, S. (2014). An evaluation of the effectiveness of comic strip conversations in
promoting the inclusion of young people with autism spectrum disorder in secondary schools.
International Journal of Developmental Disabilities, 60(2), 89–108.
Angell, M.E.,Nicholson, J. K., Watts, E. H., & Blum, C. (2011). Using a multicomponent adapted power card
strategy to decrease latency during interactivity transitions for three children with developmental disabil-
ities. Focus on Autism and Other Developmental Disabilities, 26(4), 206–217.
Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis.
Journal of Applied Behavior Analysis, 1(1), 91–97.
Behavior Analyst Certification Board. (2014). Professional and ethical compliance code for behavior
analysts. Littleton: Author.
Bieber, J. (1994). Learning disabilities and social skills with Richard LaVoie: Last one picked...First one
picked on. Washington, DC: Public Broadcasting Service.
Campbell, D. T., & Stanley, J. (1963). Experimental and quasi-experimental designs for research. Chicago:
Rand McNally.
Campbell, A., & Tincani, M. (2011). The power card strategy: Strength-based intervention to increase
direction following of children with autism spectrum disorder. Journal of Positive Behavior
Interventions, 13(4), 240–249.
Coogle, C. G., Ahmed, S., Aljaffal, M. A., Alsheef, M. Y., & Hamdi, H. A. (2017). Social narrative strategies
to support children with autism spectrum disorder. Early Childhood Education Journal, 46, 445–450.
https://doi.org/10.1007/s10643-017-0873-7.
Cook, B. G., & Cook, S. C. (2011). Unraveling evidence-baased practices in special education. The Journal of
Special Education, 47(2), 71–82.
Daubert, A., Hornstein, S., & Tincani, M. (2015). Effects of a modified power card strategy on turn taking and
social commenting of children with autism spectrum disorder playing board games. Journal of
Developmental and Physical Disabilities, 27,93–110 .
Davis, K. M., Boon, R. T., Cihak, D. F., & Fore, C. (2010). Power cards to improve conversational skills in
adolescents with Asperger syndrome. Focus on Autism and Other Developmental Disabilities, 25(1), 12–
22.
Dollaghan, C. A. (2007). The handbook for evidence-based practice in communication disorders. Baltimore,
MD: Brookes
Ganz, J. B., Kaylor, M., Bourgeois, B., & Hadden, K. (2008). The impact of social scripts and visual cues on
verbal communication in three children with autism spectrum disorder. Focus on Autism and Other
Developmental Disabilities, 23(2), 79–94.
Gray, C. (1994a). Comic strip conversations: Illustrated interactions that teach conversation skills to students
with autism and related disorders. Jenison: Jenison Public Schools.
Gray, C. (1994b). Social stories. Arlington: Future Horizons.
Gray, C. A., & Garand, J. D. (1993). Social stories: Improving responses of students with autism with accurate
social information. Focus on Autistic Behavior, 8,1–10.
Horner, R. H., Carr, E. G., Halle, J., McGee, G., Odom, S., & Wolery, M. (2005). The use of single-subject
research to identify evidence-based practice in special education. Exceptional Children, 71(2), 165–179.
Hundert, J., Rowe, S., & Harrison, E. (2014). The combined effects of social script training and peer buddies
on generalized peer interaction of children with ASD in inclusive classrooms. Focus on Autism and Other
Developmental Disabilities, 29(4), 206–215.
Hutchins, T. L., & Prelock, P. A. (2006). Using social stories and comic strip conversations to promote socially
valid outcomes for children with autism. Seminars in Speech and Language, 27(1), 47–59.
Hutchins, T. L., & Prelock, P. A. (2008). Supporting theory of mind development: Considerations and
recommendations for professionals providing services to individuals with autism spectrum disorder.
Topics in Language Disorders, 28(4), 340–364.
Journal of Developmental and Physical Disabilities

Hutchins, T. L., & Prelock, P. A. (2012). Parents’perspective of their children’s social behavior: The social
validity of social stories™and comic strip conversations. Journal of Positive Behavior Interventions,
15(3), 156–168.
Kazdin, A. E. (2011). Single-case research designs: Methods for clinical and applied settings (2nd ed.). New
York: Oxford University Press.
Keeling, K., Myles, B. S., Gagnon, E., & Simpson, R. L. (2003). Using the power card strategy to teach
sportsmanship skills to a child with autism. Focus on Autism and Other Developmental Disabilities,
18(2), 105–111.
Kennedy, C. H. (2005). Single-case designs for educational research. Boston: MA: Pearson Education, Inc..
Leaf, J. B., Oppenheim-Leaf, M. L., Leaf, R. B., Taubman, M., McEachin, J., Parker, T., Waks, A. B., &
Mountjoy, T. (2015). What is the proof? A methodological review of studies that have utilized social
stories. Education and Training in Autism and Developmental Disabilities, 50(2), 127–141.
Loveland, K. A., & Tunali, B. (1991). Social scripts for conversational interactions in autism and down
syndrome. Journal of Autism and Developmental Disorders, 21(2), 177–186.
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for systematic reviews
and meta-analyses: The PRISMA statement. PLoS Medicine, 6(7), e1000097. https://doi.org/10.1371
/journal.pmed.1000097.
National Autism Center. (2015a). Evidence-based practice and autism in the schools (2nd ed.). Randolph:
Author.
National Autism Center. (2015b). Findings and conclusions: National standards project, phase 2. Randolph:
Author.
Parker, D., & Kamps, D. (2011). Effects of task analysis and self-monitoring for children with autism in
multiple social settings. Focus on Autism and Other Developmental Disorders, 26(3), 131–142.
Pierson, M. R., & Glaeser, B. C. (2005). Extension of research on social skills training using comic strip
conversations to students without autism. Education and Training in Developmental Disabilities, 40(3),
279–284.
Pierson, M. R., & Glaeser, B. C. (2007). Using comic strip conversations to increase social satisfaction and
decrease loneliness in students with autism spectrum disorder. Education and Training in Developmental
Disabilities, 42(4), 460–466.
Reynhout, G., & Carter, M. (2011). Evaluation of the efficacy of social stories™using three single subject
metrics. Research in Autism Spectrum Disorders, 36,445–469.
Rhodes, C. (2014). Do social stories help to decrease disruptive behaviour in children with autistic spectrum
disorders? A review of the published literature. Journal of Intellectual Disabilities, 18,35–50.
Roth, M. E., Gillis, J. M., & DiGennaro Reed, F. D. (2014). A meta-analysis of behavioral interventions for
adolescents and adults with autism spectrum disorders. Journal of Behavioral Education, 23(2), 258–286.
Styles, A. (2011). Social stories™: Does the research evidence support the popularity? Educational
Psychology in Practice, 27, 415–436.
V
an Houten, R., Axelrod, S., Bailey, J. S., Favell, J. E., Foxx, R. M., Iwata, B. A., & Lovaas, O. I. (1988). The
right to effective behavioral treatment. Journal of Applied Behavior Analysis, 21(4), 381–384.
Wong, C., Odom, S. L., Hume, K., Cox, A. W., Fettig, A., Kucharczyk, S., et al. (2013). Evidence-based
practices for children, youth, and young adults with autism Spectrum disorder. Chapel Hill: The
University of North Carolina, Frank Porter Graham Child Development Institute, Autism Evidence-
Based Practice Review Group.
Wong, C., Odom, S. L., Hume, K. A., Cox, A. W., Fettig, A., Kucharczyk, S., Brock, M. E., Plavnick, J. B.,
Fleury, V. P., & Schultz, T. R. (2015). Evidence-based practices for children, youth, and young adults with
autism spectrum disorder: A comprehensive review. Journal of Autism and Developmental Disorders,
45(7), 1951–1966. https://doi.org/10.1007/s10803-014-2351-z.
Zimmerman, K. N., & Ledford, J. R. (2017). Beyond ASD: Evidence for the effectiveness of social narratives.
Journal of Early Intervention, 39(3), 199–217. https://doi.org/10.1177/105381511770900.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Journal of Developmental and Physical Disabilities