See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/285754576

Treatment of Echolalia in Individuals with Autism Spectrum Disorder: a

Systematic Review

Article · December 2015

DOI: 10.1007/s40489-015-0067-4

CITATIONS READS

4 3,477

5 authors, including:

Leslie Neely Stephanie Gerow

University of Texas at San Antonio Baylor University
35 PUBLICATIONS   315 CITATIONS    21 PUBLICATIONS   74 CITATIONS   

SEE PROFILE SEE PROFILE

Mandy Rispoli Russell Lang

Purdue University Texas State University
125 PUBLICATIONS   2,708 CITATIONS    197 PUBLICATIONS   4,025 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Neurodevelopmental diseases: Assessment & Interventions View project

All content following this page was uploaded by Russell Lang on 08 March 2016.

The user has requested enhancement of the downloaded file.

Rev J Autism Dev Disord (2016) 3:82–91
DOI 10.1007/s40489-015-0067-4

REVIEW PAPER

Treatment of Echolalia in Individuals with Autism Spectrum

Disorder: a Systematic Review
Leslie Neely 1 & Stephanie Gerow 2 & Mandy Rispoli 3 & Russell Lang 4 & Nathan Pullen 4

Received: 15 October 2015 / Accepted: 25 November 2015 / Published online: 4 December 2015
# Springer Science+Business Media New York 2015

Abstract Echolalia can lead to communication breakdowns conclusive levels of evidence included cues-pause-point, dif-
that increase the likelihood of social failure and stigmatization ferential reinforcement of lower rates of behavior, script train-
in children with autism spectrum disorder (ASD). In an effort ing plus visual cues, and verbal modeling plus positive rein-
to facilitate evidenced-based intervention and inform future forcement for appropriate responses. Implications for practi-
research, this systematic review analyzes peer-reviewed stud- tioners and directions for future research are offered.
ies involving the treatment of echolalia in individuals with
ASD. Using predetermined inclusion criteria, a total of 11
studies were identified, reviewed, and summarized in terms Keywords Autism spectrum disorder . Echolalia .
of the following: (a) participant characteristics (e.g., verbal Treatment . Systematic review
and cognitive functioning), (b) type of echolalia (e.g., delayed
or immediate), (c) intervention procedures, (d) intervention
outcomes, (e) maintenance and generalization of outcomes, Autism spectrum disorder (ASD) is a developmental disorder
and (f) research design and other indicators of rigor (i.e., cer- characterized by deficits in social communication skills and
tainty of evidence). Nine studies successfully reduced echola- excesses in repetitive and restrictive patterns of behaviors
lia in a total of 17 participants. However, only six of those nine (Diagnostic and Statistical Manual of Mental Disorders [5th
studies met criteria to be classified as providing the highest ed.; DSM-5]; American Psychiatric Association [APA] 2013).
level of certainty (i.e., conclusive). The findings of this review The combination of behavioral excesses and deficits can man-
suggest that a number of treatment options can be considered ifest as a repetitive, restricted pattern of vocal behavior called
promising practices for the treatment of echolalia in children echolalia (Stribling et al. 2007). Echolalia is typically defined
with ASD. Although no single treatment package can be de- as the socially awkward or inappropriate verbatim repetition
scribed as well-established evidence-based practice, all 11 of part or all of a previously spoken utterance (Karmali et al.
studies involved behavior analytic intervention components, 2005; Stribling et al. 2007; Valentino et al. 2012). The initial
suggesting strong support for operant-based treatments. In utterance, that is then repeated, may come from another per-
particular, behavior analytic interventions demonstrating son in the environment or from a recording (television or
audio source) and maybe immediate or delayed. Immediate
echolalia occurs when the latency between initial utterance
* Leslie Neely
[email protected]
and repetition is within a few seconds, whereas delayed
echolalia occurs when the time between the initial utterance
and the repetition involves longer durations, inclusive of
1
Department of Educational Psychology, The University of Texas at repetitions occurring days after the initial utterance being
San Antonio, 501 W. Cesar E. Chavez Blvd., San
Antonio, TX 78207-4415, USA
echoed (Foxx et al. 2004; Hetzroni and Tannous 2004).
2
Another form of echolalia is palilalia. Palilalia involves the
Texas A&M University, College Station, TX, USA
repeating of one’s own words in a quiet whispered voice
3
Purdue University, West Lafayette, IN, USA immediately following the initial typical volume utterance
4
Texas State University, San Marcos, TX, USA (Karmali et al. 2005).
Rev J Autism Dev Disord (2016) 3:82–91 83

Although language repetition is part of typical child Kavon and McLaughlin’s (1995) review categorized the
development (Howlin 1982). some children with ASD en- remaining interventions as verbal prompting interventions.
gage in echolalia that persists past the early childhood Studies in that category used a combination of reinforcement,
developmental period (Barrera and Sulzer-Azaroff 1983; prompting, and error correction to reduce echolalia but did not
Neely 2014; Fay 1969). In addition, children with ASD utilize the more specific sequence of cues-pause-point (e.g.,
often engage in echolalia that lacks social context Lovaas Freeman et al. 1975; Lovaas 1977). For example, Freeman
et al. 1973) and occurs at a higher rate than in typically et al. (1975) used positive reinforcement for correct
developing children (Fay 1973). responding to questions and an error prevention procedure
Echolalia may (a) complicate educational programs de- (consisting of interrupting the echolalia) to treat the echolalia
signed to improve speech, (b) contribute to communication of a 5-year-old male with autism. The intervention produced
breakdowns, (c) increase the likelihood of social failure or decreases in echolalia that maintained following the with-
stigmatization, and (d) increase the risk of challenging behav- drawal of intervention. Although the review by Kavon and
ior (Light et al. 1998; Valentino et al. 2012). For example, McLaughlin provides evidence in support of these interven-
Valentino et al. (2012) identified a 3-year-old male with tions, additional studies have emerged over the last 20 years
ASD who repeated the instruction Bsay^ during echoic train- and an updated systematic review appears warranted.
ing. The immediate echolalia was interfering with instruction Therefore, the purpose of this review is to update and ex-
and complicating the educational program aimed at teaching tend the previous review by Kavon and McLaughlin by (a)
the child to tact. Previous research aimed at identifying the utilizing broader inclusion criteria not limited to behavioral
operant function of echolalia suggests that the complete range (operant) approaches, (b) conducting a systematic review of
of functions found to maintain other behaviors (e.g., automatic the literature, (c) rating each included study’s certainty of ev-
reinforcement, socially mediated positive reinforcement, and idence (quality of research design and controls) so results can
socially mediated negative reinforcement) may also reinforce be considered in light of each study’s methodological rigor,
and maintain echolalia (Goren et al. 1977; Healy and Leader and (d) identifying advances in treatment that may have de-
2011). veloped since the previous review. A review of this nature is
A descriptive review by Kavon and McLaughlin (1995) intended to offer directions for future research and to provide
identified two interventions with preliminary support for the guidance to practitioners interested in the use of evidence-
treatment of echolalia (i.e., cues-pause-point and more general based treatments for echolalia in children with ASD.
verbal prompting interventions). Cues-pause-point is a behav-
ioral intervention that has been evaluated for the treatment of
immediate echolalia. Cues-pause-point was introduced by Method
McMorrow and Foxx (1986) in their treatment of a 21-year-
old male with ASD. The cues-pause-point intervention con- Search Procedures
sists of a trainer providing a visual cue to the learner to remain
silent (cue). The trainer then maintains the visual cue while Four electronic databases were searched to identify potential
providing instructions about the upcoming teaching session. studies for this review: ERIC (EBSCO), Medline, Psychology
The trainer poses a question and provides a short pause fol- and Behavioral Sciences Collection, and PsycINFO. There
lowing the question (pause). Finally, the trainer points to a were no limitations on publication year, but results were lim-
card to prompt the learner to verbalize the answer to the ques- ited to English language, peer-reviewed research. Terms to
tion (point). For example, to teach the individual with ASD to describe individuals with an ASD were combined with terms
respond appropriately to the question, BWhat is your name?,^ to describe echolalia. The terms for individuals with an ASD
the trainer held up an index finger to cue silence (cue), stated, included BAsperger,^ Bautis*,^ Bdevelopmental disab*,^
BI’m going to ask you some questions, do your best to answer BASD,^ and BPDD-NOS.^ The terms searched to describe
them correctly^ (pause), asked the question, pointed to a card echolalia included Becholal*,^ Brepetitive speech,^ Brepetitive
with the person’s name prompting the correct response verbal*,^ Brepetitive talking,^ Brepetitive communication*,^
(point), and then provided reinforcement contingent on the and Bpalilalia.^ Following the initial search, the last name of
correct response. Following the initial study by McMorrow the first author of each included study was also entered into
and Foxx, follow-up studies extended the procedure to indi- PsychINFO to identify any other potentially relevant studies
viduals with less developed language skills (McMorrow et al. that had been published by that author. Finally, the reference
1987) and individuals with intellectual disabilities (Foxx, list of Kavon and McLaughlin (1995) was examined for ad-
Faw, McMorrow, Kyle, & Bittle 1988) and then demonstrated ditional studies meeting inclusion criteria.
that reductions of echolalia following cues-pause-point could These search procedures were conducted in May 2014,
be maintained up to 57 months following the cessation of the updated in April 2015, and yielded a total of 568 articles
intervention (Foxx and Faw 1990). (534 from the original search and 34 from the updated search).
84 Rev J Autism Dev Disord (2016) 3:82–91

The title and abstracts of the 568 articles were screened using intervention outcomes, (e) maintenance and generalization
the predetermined inclusion criteria (see BInclusion Criteria^ of outcomes, and (f) research design and other indicators of
section) to identify articles for potential inclusion in this re- rigor (i.e., certainty of evidence). Participant description in-
view. Following this screening of title and abstracts, a total of cluded the number of participants with ASD, their ages, and
46 articles were identified for further review. gender. Participant verbal and cognitive functioning was cod-
ed using reported standardized assessments or was gleaned
Inclusion Criteria from detailed descriptions of participant functioning.
Echolalia was coded as either immediate, delayed, or palilalia,
The 46 articles were then downloaded and evaluated based on and when noted in the reviewed study, the operant function of
the pre-set inclusion criteria. Studies were included if they (a) echolalia was noted. Various procedural aspects were coded to
included a participant diagnosed with ASD or was described identify intervention protocols or components (e.g., cues-
as an individual with Bautistic-like behaviors^ (included due pause-point protocol, script training, or reinforcement
to the age of the literature base), (b) implemented an interven- procedures)
tion and reported outcomes for echolalia (inclusive of palilalia Intervention outcomes were summarized and coded as nega-
and defined as repetition of a previously spoken word or tive, mixed, or positive. As all 11 studies employed single-case
phrase) as a dependent variable, (c) employed an experimental research designs, study outcomes were determined based on vi-
design (inclusive of single-case and group experimental de- sual analysis criteria for single-case research outlined by
signs), and (d) echolalia outcomes for the individual with Kennedy (2005). A study was rated as having negative results
ASD could be disaggregated from participants without ASD if there was no reduction observed in echolalia as indicated by a
and target behaviors other than echolalia. Studies which im- flat or increasing trend in the intervention phase as compared to
plemented interventions for individuals with ASD who uti- the baseline phase. Studies were coded as having mixed results if
lized echolalic speech but did not present outcomes related some, but not all, of the participants demonstrated a reduction in
to the echolalia were excluded (e.g., Barrera and Sulzer- echolalia during the intervention phase relative to the baseline
Azaroff 1983; Charlop-Christy and Kelso 2003; Charlop phase. Positive results indicated that echolalia decreased in all
1983). Studies which evaluated echolalia under different con- participants during intervention phase as relative baseline.
ditions but did not implement an intervention to address echo- The study’s capacity to provide a certainty of evidence was
lalia were also excluded (e.g., Rydell and Mirenda 1994; rated as suggestive, preponderant, or conclusive, with conclu-
Violette and Swisher 1992). In addition, studies which imple- sive being the highest rating (Schlosser 2009; Simeonsson and
mented interventions to treat other repetitive speech (i.e., Bailey 1991; Smith 1981). Studies rated as conclusive had the
noncontextual vocal stereotypy, such as a sound rather than a following: (a) an experimental design capable of establishing
word or phrase) were excluded (e.g., Mancia et al. 2000; Ahearn experimental control (e.g., ABAB, multiple-baseline design,
et al. 2000; Taylor et al. 2005). Studies excluded because data alternating treatments design), (b) sufficient interobserver
on echolalia were not disaggregated from other outcomes in- agreement (IOA) collected on the observed educator behav-
volving other topographies of behavior were Arntzen et al. iors (i.e., agreement coefficients above 80 % and IOA collect-
(2006) and Mancia et al. (2000). For example, Arntzen et al. ed for a minimum of 20 % of the sessions), (c) intervention
(2006) taught a 44-year-old woman functional verbal responses procedures detailed enough to promote replication of the pro-
and tracked subsequent decreases in aberrant verbal behavior. cedures, (d) operationalized descriptions of the dependent var-
Although aberrant verbal behavior included repetitive echolalic iable, and (e) demonstrated convincing effects of the interven-
responses, the aberrant verbal behavior also included tion for every participant (i.e., received a rating of positive
Bpsychotic^ verbalizations and results for the two were results). A study rated as preponderant met most of the criteria
collapsed into one dependent variable. Finally, Cohen (1981) for a Bconclusive^ study, but results may have demonstrated
was excluded because the figure referenced in the article was Bmixed^ effects of the intervention for some or all of the
not included in the article and was not accessible to the authors participants with ASD. Any study that (a) lacked an experi-
after multiple attempts to locate the figure through university- mental design capable of establishing experimental control,
based library services. Ultimately, a total of 11 studies met (b) did not meet the minimum IOA criterion, (c) did not
inclusion criteria and were included in this review. operationally define the intervention procedures, (d) or did
not operationally define the dependent variable were auto-
Descriptive Synthesis matically rated as offering suggestive evidence.

Included studies were reviewed and summarized based on the Interrater Reliability
following categories: (a) participant characteristics (e.g., ver-
bal and cognitive functioning), (b) type of echolalia (e.g., de- Inclusion Criteria To ensure accurate application of the in-
layed or immediate), (c) intervention procedures, (d) clusion criteria, two raters reviewed each of the 46 articles,
Rev J Autism Dev Disord (2016) 3:82–91 85

resulting from the systematic search and initial title/abstract information regarding the language functioning of partici-
review, for potential inclusion. Agreement was reached on pants, all of which suggested below-average verbal abilities
whether to include or exclude a study on 100 % of the articles. for 16 of the 25 participants (64 %; Ganz et al. 2008; Hetzroni
and Tannous 2004; Karmali et al. 2005; Laski et al. 1988;
Descriptive Synthesis To establish interrater reliability (IRR) Valentino et al. 2012).
for the data summaries, two independent raters coded five of
the 11 included articles (46 %). A third rater reviewed the Type of Echolalia
independent data summaries and made a decision as to wheth-
er the summaries agreed. IRR was calculated based on wheth- Across the 11 studies, five targeted immediate echolalia only
er the two raters agreed on the extracted data. There were a (45 %; Foxx et al. 2004; McMorrow and Foxx 1986;
total of 30 items in which there could be agreement or dis- Nientimp and Cole 1992; Palyo et al. 1979; Valentino et al.
agreement (i.e., five studies with six data categories each). 2012), two targeted delayed echolalia only (18 %; Ganz et al.
IRR was calculated using percent agreement by dividing the 2008; Handen et al. 1984), and three studies (27 %) targeted
total number of agreements by the sum of the agreements and both immediate and delayed echolalia (Freeman et al. 1975;
disagreements and multiplying by 100 % to convert to a per- Hetzroni and Tannous 2004; Laski et al. 1988). Finally, one
centage. Initial agreement for the coding of studies was 90 %. study targeted palilalia (Karmali et al. 2005). No study report-
In instances of disagreement, the raters discussed until 100 % ed operant functions of target behaviors.
agreement was reached.
Intervention Procedures

Results All of the 11 studies employed an intervention with behavioral

analytic components (e.g., differential reinforcement, model-
Table 1 created from the coded study summarizes and displays ing, prompting). Five of the 11 studies (45 %) evaluated the
each study according to the following: (a) participant charac- effects of specific treatment package on echolalia (i.e., cues-
teristics (e.g., verbal and cognitive functioning), (b) type of pause-point; Natural Language Paradigm, and computer-
echolalia (e.g., delayed or immediate), (c) intervention proce- based intervention). Cues-pause-point was the most frequent
dures, (d) intervention outcomes, (e) maintenance and gener- treatment package evaluated (n=3; 27 %; Foxx et al. 2004;
alization of outcomes, and (f) research design and other indi- McMorrow and Foxx 1986; Valentino et al. 2012). For exam-
cators of rigor (i.e., certainty of evidence). ple, Valentino and colleagues (2012) evaluated the use of
cues-pause-point to treat a 3-year-old male child who echoed
Participant Characteristics the instruction say during echoic training. Decreased echolalia
and increase appropriate responding (e.g., repeating the target
The 11 studies included a total of 25 participants with ASD. word without echoing say) were noted.
Ten of the 11 studies reported the gender of their participants Laski et al. (1988) trained parents to implement the Natural
with 17 male and 5 female participants. One study did not Language Paradigm within a clinical setting and then assessed
report participants’ gender (Laski et al. 1988). All of the stud- the effects of the parent implemented program on child com-
ies reported the participants’ ages, with a mean reported age of munication outcomes. Parents were taught to use direct rein-
8 years (range 3–21 years) across studies. forcement of verbal attempts, to promote turn-taking with play
Ten studies (90 %) reported information regarding partici- items, to vary stimuli and exemplars, and to utilize shared
pants’ verbal or cognitive functioning. Four studies (36 %) control (i.e., rotating between child-led and parent-led activi-
reported results from standardized cognitive assessments ties). Although appropriate vocalizations were the primary
(i.e., Stanford-Binet, Merrill-Palmer, and Peabody Picture dependent variable, child engagement in echolalia was mea-
Vocabulary Test (PPVT)) with three studies including four sured as an ancillary variable with mixed results noted for the
participants with extremely low cognitive functioning (16 %; effects of the Natural Language Paradigm on echolalia.
Handen et al. 1984; McMorrow and Foxx 1986; Nientimp and The final treatment package was a computer program enti-
Cole 1992) and one study including one participant with tled BI Can Word It Too^ that was available in both Arabic and
below-average to average cognitive functioning (4 %; Hebrew (Hetzroni and Tannous 2004). The program presented
Freeman et al. 1975). Three studies (27 %) reported results participants with a simulated situation in which a parent asked
from standardized language assessments (i.e., Alpern-Boll the participant a question. The participant would then choose
and PPVT-III) indicating extremely low verbal abilities and the appropriate sentence or question option, and an animation
below-average cognitive functioning for eight of the 25 par- of their choice would be played. For example, if the question
ticipants (32 %; Foxx et al. 2004; Karmali et al. 2005; Palyo was Bwhat would you like to play with^ and the participant
et al. 1979). Five of the studies reported descriptive selected the option BI want to play ball with you,^ an
 86

Table 1 Interventions studies to decrease echolalia for individuals with ASD

Article Participant characteristics Type of echolalia Intervention procedures Intervention Maintenance and generalization Certainty of evidence
outcomes of outcomes

Foxx et al. (2004) n=2; males; 5 and 6 years Repeated words contained Cues-pause-point Positive Positive using a gradual Conclusive
P1: PPVT-III 40 (standard score); in an asked question; intervention fading
EOWVT 55 (standard score); immediate echolalia procedure)/ positive across
CARS 37; GARS Autism novel setting, and novel
Quotient 97 trainer
P2: PPVT-III 40 (standard score);
EOWVT 58 (standard score);
CARS 40; GARS Autism
Quotient 83
Freeman et al. n=1; male; 5 years Immediate and delayed Positive reinforcement Positive Not reported/not reported Suggestive; ABA
(1975) No standardized language echolalia for correct answers design; IOA was
assessment; IQ assessments and error correction not assessed
ranged from 81 (Stanford-Binet) procedure to block
to 120 (Merrill-Palmer). echolalia
Ganz et al. n=2 males; 7 and 12 yearsb P1: Repeated phrases from Script training and Positive Not reported/not reported Conclusive
(2008) No standard language assessments. television and video visual cues
No cognitive assessments. games; delayed echolalia
P1: difficulty with Wh- questions; P2: Repeated phrases from
age-appropriate phonological songs and books; delayed
and semantic speech echolalia
P2: Age-appropriate phonologic
and semantic skills; rarely
initiated conversation.
Handen et al. n=1; male; 16 years Repeated statements and/or Differential reinforcement Positive Positive at 9 and 14 months Conclusive
(1984) No standardized language asking same question of lower rates follow-up/not reported
assessment; mental age of multiple times a day;
5 years 11 months delayed echolalia
(Stanford-Binet).
Hetzroni and n=5; 3 males and 2 females; Immediate and delayed echolalia Software program Mixed Not reported/results Suggestive; one leg
Tannous (2004) 7.8, 8, 8.5, 11.5, and 12.5 years (I Can Word it Too) generalized to of every participant’s
No standardized language authentic settings MBD did not
assessments. No cognitive for some participants demonstrate effects;
assessments. additional information
necessary for replication
of intervention
Karmali et al. (2005) N=5; 4 male and 1 female; Palilalia was all related to Tact modeling plus Positive Not reported/positive Conclusive
3 to 4 years children’s movies or positive reinforcement across settings
Norm-referenced assessments songs; delayed palilalia of appropriate responses
indicated speech delays for all
participants; no cognitive
assessments reported
Laski et al. (1988) n=3; gender not specified; Immediate and delayed echolalia Natural Language Mixed Not reported/not reported Suggestive (echolalia was
5.8, 6.2, and 8.11 yearsa Paradigm an ancillary dependent
variable)
Rev J Autism Dev Disord (2016) 3:82–91
Table 1 (continued)

Article Participant characteristics Type of echolalia Intervention procedures Intervention Maintenance and generalization Certainty of evidence
outcomes of outcomes

No standardized language
assessments; P1: could imitate
sounds and a few words on
request, rarely initiated, and
receptive vocabulary less than
15 words. P6 and P8: Blarger
vocabularies^ and used short
phrases; rarely spoke
Rev J Autism Dev Disord (2016) 3:82–91

spontaneously.
McMorrow n=1; male; 21 years Repeated statements; Experiment 1: cues-pause- Positive Positive at 57 months follow-up Conclusive
and Foxx (1986) No standardized language immediate echolalia point and then pause only (Foxx and Faw 1990)/results
& Foxx and Faw assessment. No verbal Experiment 2 and 3: modeling did not generalize to new
(1990) initiations; did not respond to question (cues-pause-point
questions; nearly all not used in generalization
verbalizations were echolalic. probes)
IQ of 40 using the PPVT.
Nientimp and n=3; 2 males and 1 female; 12, Repeated all or part of a Constant time delay and Positive Positive for two of the Suggestive; pre-
Cole 1992 12.8, 13.4 years greeting; immediate contingent verbal praise participants during experimental design
No standardized language echolalia immediate withdrawal of (AB design; one
assessment. All described as intervention/mixed with participant) and ABA
verbal but prompt dependent. generalization to novel peers designa (two participants)
P1: IQ 38 and P2: IQ 32 for two of three participants
(Stanford-Binet). No IQ on
female participant
Palyo et al. (1979) n=1; female; 5.7 yearsb Immediate echolalia Punishment for echolalia, Positive Positive at 12 months follow- Suggestive; pre-
Vineland Social Quotient 63; positive reinforcement up/positive generalization experimental design
Alpern-Boll IQ 61; Alpern- for appropriate responses, across settings and stimuli (AB design)
Boll communication age prompting using tape (untrained questions)
equivalent 2 years recording of appropriate
response
Valentino et al. n=1; male; 3 years Repeated Bsay^ during Cues-pause-point Positive Positive at 3 months Conclusive
(2012) Tact repertoire included 75 echoics training; follow-up/positive with
common words; intraverbals immediate echolalia faster demonstration
included fill-ins to songs, of effects to novel
animal sounds, and the stimuli
function of some items.

P participant, IOA interobserver agreement, MBD single-subject multiple baseline design, AB single-subject design consisting of a baseline phase (BA^) and intervention phase (BB^)
a
ABA single-subject design consisting of two baseline phases (BA^) and one intervention phase (BB^)
b
Only participants who met the pre-set criteria were included in this review
87
88 Rev J Autism Dev Disord (2016) 3:82–91

animation of a father and child playing ball would appear. treatment means for echolalia, with no differences noted in
The participant was then observed in their classroom, and one setting (i.e., the break room). The second study
data were collected on appropriate and inappropriate verbal (Hetzroni and Tannous 2004) utilized a multiple baseline de-
behavior. Results were mixed with some participants sign across settings to evaluate the effects of their technology-
demonstrating improvement in echolalia and some demon- based intervention on participant echolalia. However, de-
strating no improvement. creases in echolalia were not demonstrated for all three set-
The remaining six studies employed a variety of behavior tings for any of the participants undercutting the experimental
analytic interventions to treat echolalia. Behavioral compo- control of the multiple baseline design. The two studies with
nents included error correction and differential reinforcement mixed results represented 8 of the 25 participants.
(n=2; Freeman et al. 1975; Palyo et al. 1979). differential
reinforcement of lower rates (n=1; Handen et al. 1984).
modeling and positive reinforcement (n=2; Karmali et al. Maintenance and Generalization
2005; Palyo et al. 1979). modeling (n=1; McMorrow and
Foxx 1986). time delay and differential reinforcement (n=1; Five of studies assessed maintenance of behavior change
Nientimp and Cole 1992). and visual cues with differential (Foxx et al. 2004; Handen et al. 1984; Nientimp and Cole
reinforcement (n = 1; Ganz et al. 2008). For example, 1992; Palyo et al. 1979; Valentino et al. 2012). and one study
Handen and colleagues (1984) implemented differential rein- was published as a long-term follow-up to the McMorrow and
forcement of lower rates (DRL) of behavior to decrease the Foxx study (1986; Foxx and Faw 1990). The timing of the
echolalia of a 16-year-old male with ASD. The intervention collection of maintenance data ranged from immediately fol-
occurred over an 18-month time frame and involved pro- lowing the conclusion of the intervention (Foxx et al. 2004;
viding the participant with tokens for engaging in lower McMorrow and Foxx 1986; Nientimp and Cole 1992) to
rates of echolalia than a predetermined criterion. When 57 months after the intervention (Foxx and Faw 1990). All
the participant engaged in echolalia below the target studies reported that echolalia levels at maintenance were be-
rate, he exchanged the tokens for a tangible item from low baseline levels. Seven studies assessed stimulus general-
his reinforcement menu. The intervention was effective ization (Foxx et al. 2004; Hetzroni and Tannous 2004;
in reducing the participant’s engagement in echolalia; Karmali et al. 2005; McMorrow and Foxx 1986; Nientimp
however, after intervention was removed, the partici- and Cole 1992; Palyo et al. 1979; Valentino et al. 2012) in-
pant’s echolalia returned to baseline levels. cluding generalization across settings, people, materials, and
In another study, Ganz et al. (2008) taught two children different preceding utterances (questions). Four studies report-
with ASD who engaged in echolalia to engage in reciprocal ed positive results for generalization (Foxx et al. 2004;
social-communicative responses (e.g., compliments, ques- Karmali et al. 2005; Palyo et al. 1979; Valentino et al.
tions, and statements corresponding to the current activity). 2012). Two studies reported that generalization occurred for
Responses were taught by providing visual scripts of the target some participants but not for all (Hetzroni and Tannous 2004;
response and systematically fading scripts over three phases. Nientimp and Cole 1992). One study found that results did not
To reduce echolalia, a visual cue was presented which sig- generalize for the participants (McMorrow and Foxx 1986).
naled to the participant that they should cease talking (i.e., a
3″×3″ line drawing of a face with a finger in front of the
mouth indicating Bquiet^). This visual cue was introduced Certainty of Evidence
only if the participant engaged in echolalia. Results indicated
clear decreases in echolalia. Six of the studies were categorized as offering a conclusive
level of evidence with positive results, sufficient research de-
Intervention Outcomes sign and IOA data, and detailed procedural descriptions (Foxx
et al. 2004; Ganz et al. 2008; Handen et al. 1984; Karmali et al.
The data from nine of the studies indicated that the results 2005; McMorrow and Foxx 1986; Valentino et al. 2012). Five
were positive for all participants (Foxx et al. 2004; Freeman studies were categorized as suggestive (Freeman et al. 1975;
et al. 1975; Ganz et al. 2008; Handen et al. 1984; Karmali et al. Hetzroni and Tannous 2004; Laski et al. 1988; Nientimp and
2005; McMorrow and Foxx 1986; Nientimp and Cole 1992; Cole 1992; Palyo et al. 1979). Of the five studies, three did not
Palyo et al. 1979; Valentino et al. 2012). Data from two of the demonstrate experimental control (Freeman et al. 1975;
studies suggested mixed results with some participants dem- Nientimp and Cole 1992; Palyo et al. 1979). two studies had
onstrating improved behavior and some demonstrating no im- mixed results (Hetzroni and Tannous 2004; Laski et al. 1988).
provement (Hetzroni and Tannous 2004; Laski et al. 1988). and one study did not assess IOA (Freeman et al. 1975). None
The first study( Laski et al. 1988) measured echolalia as an of the studies was classified at the preponderant level of
ancillary dependent variable and provided pre- and post- evidence.
Rev J Autism Dev Disord (2016) 3:82–91 89

Discussion preliminary support for operant-based behavioral interven-

tions in general and cues-pause-point in particular.
The purpose of this review was to identify promising practices As the interventions evaluated in these studies align with
for decreasing echolalia in individuals with ASD. This sys- the operant conditioning paradigm, it is alarming that none of
tematic literature review synthesized 11 studies which the studies assessed the operant function of echolalia. In other
employed a variety of behavioral interventions. Of the 11 forms of challenging behavior (e.g., aggression, property de-
studies, nine reported positive results for 17 participants, and struction, and self-injury), identifying the function of the be-
two of the studies reported mixed results for two participants. havior via a functional analysis procedure (Iwata et al.
When examining the quality of the literature base, six of the 11 1982/1994; Lydon et al. 2012) is associated with better re-
studies were classified as providing conclusive evidence. sponse to treatment (e.g., Didden et al. 2006). Of particular
Ultimately, the findings of this review indicate that the litera- concern is that echolalia may serve various social as well as
ture base cannot conclusively support any one approach for non-social communicative functions for individuals with ASD
the treatment of echolalia in individuals with ASD. (Goren et al. 1977; Healy and Leader 2011). If echolalia is
reduced during treatment but a functionally equivalent re-
Implications for Research placement behavior is not taught, this could limit the mainte-
nance and generalization of the behavioral change
The first purpose of this review was to update the previous (Schreibman and Carr 1978). In addition, as there is diver-
review by Kavon and McLaughlin (1995) to identify effective gence within the field regarding whether echolalia is nonfunc-
interventions for the treatment of echolalia in individuals with tional (e.g., Lovaas, Schreibman, and Koegel 1974). a neces-
ASD. Although no single intervention procedure or package sary part of developing functional communication (e.g.,
met any of the commonly used criteria for classification as a Roberts 2014). or serves a social function (e.g., Prizant and
well-established or evidence-based practice (e.g., Chambless Duncan 1981). identifying the function of echolalia would
and Holland 1998; Odom and Wolery 2003). a number of help guide future research and practice.
themes emerged. First, of the six studies classified as conclu- One potential strength of this literature base is the assess-
sive, all contained behavioral analytic intervention compo- ment of maintenance of behavioral change following cessa-
nents (e.g., programmed reinforcement contingencies) sug- tion of the intervention phase. Of the six articles that assessed
gesting support for operant-based treatments in the reduction the maintenance of behavioral change, all reported positive
of echolalia for individuals with ASD. When considering im- results indicating levels of echolalia maintained below base-
mediate echolalia, the cues-pause-point intervention was iden- line levels during follow-up sessions. However, although sev-
tified as effective in three studies (Foxx et al. 2004; en of the studies did evaluate the generalization of the results
McMorrow and Foxx 1986; Valentino et al. 2012). This con- across stimuli, people, and settings, results were mixed with
clusion supports the previous descriptive review identifying only four reporting successful generalization. Future re-
cues-pause-point as a potentially effective intervention for im- searchers might consider evaluating interventions to promote
mediate echolalia. Of note, only two studies have evaluated sustained behavioral change (e.g., fading, multiple exemplar
the effects of cues-pause-point since the previous review training; Valentino et al. 2012).
(Kavon and McLaughlin 1995) highlighting the need for more
research in this area. Implications for Practice
Conclusive studies evaluating interventions for delayed
echolalia also utilized behavioral analytic components. Three A second purpose of this review was to offer guidance to
interventions, DRL of behavior (Handen et al. 1984). script practitioners interested in the use of evidence-based treatments
training plus visual cues (Ganz et al. 2008). and tact modeling for individuals with ASD. Given the limited number of con-
plus positive reinforcement for appropriate responses clusive studies, recommendations as to an evidence-based in-
(Karmali et al. 2005). were all identified as effective for de- tervention for decreasing echolalia cannot be drawn.
layed echolalia. Of particular interest is the study by Ganz However, the results of this review did suggest that cues-
et al. (2008) which utilized visual cues to signal to the partic- pause-point, which was investigated by three different studies
ipant that they should cease talking. In addition, visual scripts with positive results and conclusive levels of evidence, is po-
were provided to prompt the target response. Although not tentially effective for immediate echolalia (Foxx et al. 2004;
inclusive of all the elements of cues-pause-point, Ganz et al. McMorrow and Foxx 1986; Valentino et al. 2012). Although
did implement a cue to remain silent and a point to prompt the these results support the conclusions from previous research
correct answer. In addition, the DRL of behavior intervention that cues-pause-point is potentially effective (Kavon and
by Handen et al. (1984) utilized differential reinforcement McLaughlin 1995). practitioners should use caution when
which was a contingency in effect in the cues-pause-point implementing this intervention due to the limited number of
interventions. Overall, the research combined provides studies. In particular, practitioners should closely monitor the
90 Rev J Autism Dev Disord (2016) 3:82–91

intervention effects through ongoing progress monitoring and References

rely on objective data to evaluate the effectiveness of the
intervention. *Indicates studies which were included in this review

Limitations of this Review

There are a couple of limitations of this review to consider. Ahearn, W. H., Clark, K. M., Macdonald, R. P., & Chung, B. I. (2000).
First, the definition of echolalia used by the authors was Assessing and treating vocal stereotypy in children with autism.
Journal of Applied Behavior Analysis, 40, 263–275. doi:10.1901/
intended to distinguish echolalia from vocal stereotypy.
jaba.2007.30-06.
However, it was difficult to identify a definition of echolalia American Psychiatric Association. (2013). Diagnostic and statistical
that was accepted throughout the literature base. In addition, manual of mental disorders (5th ed.). Arlington, VA: American
as none of the studies reported the function of the target echo- Psychiatric Publishing.
lalia behavior, it is uncertain whether echolalia was isolated Arntzen, E., Tonnessen, I. R., & Brouwer, G. (2006). Reducing aberrant
verbal behavior by building a repertoire of rational verbal behavior.
from other forms of vocal stereotypy. A second limitation is Behavioral Interventions, 21, 177–193. doi:10.1002/bin.220.
the age of the literature base. Of the 11 studies reviewed, six of Barrera, R. D., & Sulzer-Azaroff, B. (1983). An alternating treatment
the studies were published over 20 years ago. As research comparison of oral and total communications training programs
quality indicators have evolved dramatically in the past with echolalic autistic children. Journal of Applied Behavior
Analysis, 16, 379–394. doi:10.1901/jaba.1983.16-379.
20 years, the age of this literature base may have been a factor Chambless, D. L., & Holland, S. D. (1998). Defining empirically sup-
in the conclusiveness of the evidence. In addition, since the ported therapies. Journal of Consulting and Clinical Psychology, 66,
review by Kavon and McLaughlin (1995). only five additional 7–18.
studies have been published on this topic. Therefore, there is a Charlop-Christy, M. H., & Kelso, S. E. (2003). Teaching children with
autism conversational speech using cue card/written script program.
need to update and expand this literature base to promote the Education and Treatment of Children, 26(2), 108–127. Retrieved
use of evidence-based practices in the treatment of echolalia from: http://www.educationandtreatmentofchildren.net/.
for individuals with ASD. A third limitation is the procedures Charlop, M. H. (1983). The effects of echolalia on acquisition and gen-
used to code intervention outcomes as applied to this literature eralization of receptive labeling in autistic children. Journal of
Applied Behavior Analysis, 16, 111–126. doi:10.1901/jaba.1983.
base. Study results were rated as Bpositive,^ mixed, or 16-111.
Bnegative^ with mixed indicating that some but not all partic- Cohen, M. (1981). Development of language behavior in an autistic child
ipants demonstrated improvements in behavior. Five of the 11 using total communication. Exceptional Children, 47, 379–381.
included studies contained only one subject with ASD, there- Retrieved from: http://journals.cec.sped.org/ec/.
Didden, R., Korzilius, H. K., Oorsouw, W. V., & Sturmey, P. (2006).
fore restricting the rating of the outcomes to either negative or Behavioral treatment of challenging behaviors in individuals with
positive. Although restricted codes could have negatively im- mild mental retardation: meta-analysis of single-subject research.
pacted intervention outcome ratings, all of the studies with one American Journal of Mental Retardation, 111, 290–298.
subject received positive ratings. However, the limited num- Fay, W. H. (1969). On the basis of autistic echolalia. Journal of
Communication Disorders, 2, 31–41. doi:10.1016/0021-9924(69)
ber of subject limits the external validity of the conclusions. A
90053-7.
fourth limitation is the absence of large-scale randomized con- Fay, W. H. (1973). On the echolalia of the blind and of autistic children.
trol trials in the resulting literature base. Although this might The Journal of Speech and Hearing Disorders, 38, 478–489.
be a reflection of the exceptionality of the population, the Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/4754881.
*Foxx, R. M., & Faw, G. D. (1990). Long-term follow-up of echolalia
exclusive use of single-subject designs limits the external va-
and question answering. Journal of Applied Behavior Analysis, 23,
lidity of this literature base. A fifth limitation of this review is 387–396. doi: 10.1901/jaba.1990.23-387.
the focus on individuals with ASD. Future researchers might Foxx, R. M., Faw, G. D., McMorrow, M. J., Kyle, M. S., & Bittle, R. G.
consider expanding to include other developmental disabil- (1988). Replacing maladaptive speech with verbal labeling
responses: an analysis of generalized responding. Journal of
ities to ensure a more comprehensive review. Such an analysis
Applied Behavior Analysis, 21, 411–417
might allow for distinctions relevant to the presentation and/or *Foxx, R. M., Schreck, K. A., Garito, J., Smith, A., Weisenberger, S.
treatment of echolalia in children with ASD relative to indi- (2004). Replacing the echolalia of children with autism with func-
viduals with other forms of disability. tional use of verbal labeling. Journal of Developmental and Physical
Disabilities, 16, 307–320. doi: 10.1007/s10882-004-0688-5.
*Freeman, B. J., Ritvo, E., Miller, R. (1975). An operant procedure to
Compliance with Ethical Standards teach an echolalic autistic child to answer questions appropriately.
Journal of Autism and Childhood Schizophrenia, 5, 169–176. doi:
Funding The authors report no funding for this manuscript. 10.1007/BF01537933.
*Ganz, J., Kaylor, M., Bourgeois, B., Hadden, K. (2008). The impact of
Conflict of Interest The authors report no conflicts of interest. social scripts and visual cues on verbal communication in three
children with autism spectrum disorders. Focus on Autism and
Ethical Approval This article does not contain any studies with human Other Developmental Disabilities, 23, 79–94. doi: 10.1177/
participants performed by any of the authors. 1088357607311447.
Rev J Autism Dev Disord (2016) 3:82–91 91

Goren, E. R., Romanczyk, R. G., & Harris, S. L. (1977). A functional responses to questions. Journal of Applied Behavior Analysis, 19,
analysis of echolalic speech. Behavior Modification, 1, 481–498. 289–297. doi: 10.1901/jaba.1986.19-289.
doi:10.1177/014544557714003. McMorrow, M. J., Foxx, R. M., Faw, G. D., & Bittle, R. G. (1987). Cues-
*Handen, B. L., Apolito, P. M., Seltzer, G. S. (1984). Use of differential pause-point language training: teaching echolalics functional use of
reinforcement of low rates of behavior to decrease repetitive speech their verbal labeling repertoires. Journal of Applied Behavior
in an autistic adolescent. Journal of Behavior Therapy and Analysis, 20, 11–22. doi:10.1901/jaba.1987.20-11.
Experimental Psychiatry, 15, 359–364. doi: 10.1016/0005- Neely, L. C. (2014). Echolalia. In C. R. Reynolds, K. J. Vannest, & E.
7916(84)90102-2. Fletcher-Janzen (Eds.), Encyclopedia of special education: a refer-
Healy, O., & Leader, G. (2011). Assessments of rituals and stereotypy. In ence for the education of children, adolescents, and adults with
J. L. Matson & P. Sturmey (Eds.), International handbook of autism disabilities and other exceptional individuals (4th ed.). Hoboken,
and pervasive developmental disorders (pp. 233–245). New York: NJ: John Wiley & Sons.
Springer. *Nientimp. E. G., & Cole, L. (1992). Teaching socially valid social inter-
Hetzroni, O. E., & Tannous, J. (2004). Effects of a computer-based inter- action responses to students with severe disabilities in a school set-
vention program on the communicative functions of children with ting. Journal of School Psychology, 30, 343–354. doi: 10.1016/
autism. Journal of Autism and Developmental Disorders, 34, 95– 0022-4405(92)90002-M.
113. doi:10.1023/B:JADD.0000022602.40506.bf. Odom, S. L., & Wolery, M. (2003). A unified theory of practice in early
Howlin, P. (1982). Echolalic and spontaneous phrase speech in autistic intervention/early child special education: evidence-based practices.
children. Journal of Child Psychology and Psychiatry, 23, 281–293. Journal of Special Education, 37, 164–173.
doi:10.1111/j.1469-7610.1982.tb00073.x. *Palyo, W. J., Cooke, T. P., Schuler, A. L., Apolloni, T. (1979).
Iwata, B. A., Dorsey, M. F., Slifer, J. J., Bauman, K. E., & Richman, G. S. Modifying echolalic speech in preschool children: training and gen-
(1994). Toward a functional analysis of self-injury. Journal of eralization. American Journal of Mental Deficiencies, 83, 480–489.
Applied Behavior Analysis, 27, 197–209. doi:10.1901/jaba.1994. Prizant, B. M., & Duncan, J. F. (1981). The functions of immediate
27-197 (Reprinted from Analysis and Intervention in echolalia in autistic children. The Journal of Speech and Hearing
Developmental Disabilities, 2, 3–20, 1982). Disorders, 46, 241–249. Retrieved from: http://www.ncbi.nlm.nih.
*Karmali, I., Greer, R. D., Nuzzlo-Gomez, R., Ross, D. E., Rivera- gov/pubmed/7278167.
Valdes, C. (2005). Reducing palilalia by presenting tact corrections Roberts, J. M. A. (2014). Echolalia and language development in children
to young children with autism. Analysis of Verbal Behavior, 21, with autism. In J. Arciuli & J. Brock (Eds.), Communication in
145–153. Retrieved from: http://www.ncbi.nlm.nih.gov/pmc/ autism (pp. 53–74). Philadelphia: John Benjamins Publishing
journals/609/ Company.
Rydell, P. J., & Mirenda, P. (1994). Effects of high and low constraint
Kavon, N. M., & McLaughlin, T. F. (1995). Interventions for echolalic
utterances on the production of immediate and delayed echolalia in
behaviour for children with autism: a review of verbal prompts and
young children with autism. Journal of Autism and Developmental
the cues pause point procedure. B.C. Journal of Special Education,
Disorders, 24, 719–735. doi:10.1007/BF02172282.
19(2–3), 39–45. Retrieved from: http://eric.ed.gov/?id=EJ519927.
Schlosser, R. (2009). The role of evidence-based journals as evidence-
Kennedy, C. H. (2005). Single-case designs for educational research.
based information sources. Applied Behavior Analysis International
Boston, MA: Allyn and Bacon.
Annual Convention (Conference). Phoenix: Az.
*Laski, K. E., Charlop, M. H., Schriebman, L. (1988). Training parents to
Schreibman, L., & Carr, E. G. (1978). Elimination of echolalic
use the natural language paradigm to increase their autistic chil-
responding to questions through the training of a generalized verbal
dren’s speech. Journal of Applied Behavior Analysis, 21, 391–400.
response. Journal of Applied Behavior Analysis, 11, 453–463. doi:
doi: 10.1901/jaba.1988.21-391.
10.1901/jaba.1978.11-453.
Light, C., Roberts, B., Dimarco, R., & Greiner, N. (1998). Augmentative Simeonsson, R., & Bailey, D. (1991). Evaluating programme impact:
and alternative communication to support receptive and expressive levels of certainty. In D. Mitchell & R. Brown (Eds.), Early inter-
communication for people with autism. Journal of Communication vention studies for young children with special needs. New York:
Disorders, 31, 153–180. doi:10.1016/S0021-9924(97)00087-7. Chapman and Hall.
Lovaas, O. I., Koegel, R. L., Simmons, J. Q., & Long, J. S. (1973). Some Smith, N. (1981). The certainty of evidence in health evaluations.
generalization and follow-up measures on autistic children in behav- Evaluation and Program Planning, 4, 273–278. Retrieved from:
ior therapy. Journal of Applied Behavior Analysis, 6, 131–166. doi: http://www.journals.elsevier.com/evaluation-and-program-
10.1901/jaba.1973.6-131. planning/.
Lovaas, O. I., Schreibman, L., & Koegel, R. L. (1974). A behavior mod- Stribling, P., Rae, J., & Dickerson, P. (2007). Two forms of spoken rep-
ification approach to the treatment of autistic children. Journal of etition in a girl with autism. International Journal of Language &
Autism and Developmental Disabilities, 4, 111–129. doi:10.1007/ Communication Disorders, 42, 427–444. doi:10.1080/
BF02105365. 13682820601183659.
Lovaas, O. I. (1977). The autistic child: language development through Taylor, B. A., Hoch, H., & Weissman, B. (2005). The analysis and treat-
behavior modification. New York: Irvington. ment of vocal stereotypy in a child with autism. Behavioral
Lydon, S., Healy, O., O’Reilly, M. F., & Lang, R. (2012). Variations in Interventions, 20, 239–253. doi:10.1002/bin.200.
functional analysis methodology: a systematic review. Journal of *Valentino, A. L., Schillingsburg, M. A., Conine, D. E., & Powell, N. M.
Developmental and Physical Disabilities, 24, 301–326. (2012). Decreasing echolalia of the instruction Bsay^ during echoic
Mancia, C., Tankersley, M., Kamps, D., Kravits, T., & Parrett, J. (2000). training through use of the cues-pause-point procedure. Journal of
Brief report: reduction of inappropriate vocalizations for a child with Behavioral Education, 21, 315–328. doi: 10.1007/s10864-012-
autism using a self-management treatment program. Journal of 9155-z.
Autism and Developmental Disorders, 30, 599–606. doi:10.1023/ Violette, J., & Swisher, L. (1992). Echolalic responses by a child with
A:1005695512163. autism to four experimental conditions of sociolinguistic input.
*McMorrow, M. J., & Foxx, R. M. (1986). Some direct and generalized Journal of Speech and Hearing Research, 35, 139–147. doi:10.
effects of replacing an autistic man’s echolalia with correct 1044/jshr.3501.139.

View publication stats