Content uploaded by Gabriela Walker
Author content
All content in this area was uploaded by Gabriela Walker on Feb 25, 2015
Content may be subject to copyright.
ORIGINAL PAPER
Constant and Progressive Time Delay Procedures for Teaching
Children with Autism: A Literature Review
Gabriela Walker
Published online: 2 June 2007
Springer Science+Business Media, LLC 2007
Abstract A review of 22 empirical studies examining the
use of constant (CTD) and progressive (PTD) time delay
procedures employed with children with autism frames an
indirect analysis of the demographic, procedural, method-
ological, and outcome parameters of existing research.
None of the previous manuscripts compared the two re-
sponse prompting procedures. This review suggests that the
effectiveness of the two procedures is similar on some
variables. However, the CTD procedure resulted in more
errors to criterion, a greater magnitude of procedural
modifications, and in a delayed moment of transfer of
stimulus control than in the PTD studies. Conclusions may
influence clinical and educational practices and indicate a
need for research.
Keywords Constant time delay Progressive time delay
Time delay Autism Autistic-like Review
Introduction
After the first clinical description of the syndrome (Kanner,
1943) and prior to 1990, relatively few individuals were
identified with Autism Spectrum Disorders (ASD). Since
then a higher percentage of children have been correctly
diagnosed with ASD. Possible reasons for an increased
prevalence include (a) increased public awareness of the
existence, characteristics, symptomatology, and prevalence
of ASD, (b) more refined diagnostic criteria and sub clas-
sification categories within ASD, (c) empirical evidence
that autism could coexist with other conditions (intellectual
disability, psychiatric disorders, medical conditions), (d)
development of specialized services, (e) increased advo-
cacy, (f) increased funds allocated to ASD research have
led to increased research on teaching interventions for
individuals with autism, and (g) a possible true increase in
the number of individuals displaying autistic symptoms
(Bristol et al., 1996; Kabot, Masi, & Segal, 2003; Wing &
Potter, 2002).
ASD, also known as pervasive developmental disorders
(PDD), are neuro-developmental disorders characterized
by a triad of symptoms: (a) impaired reciprocal social
interaction, (b) impaired verbal and nonverbal communi-
cation, and (c) stereotyped behaviors restricted to certain
interests (DSM IV-TR, American Psychiatric Association,
2000; ICD-10, 1993). These symptoms are both quantita-
tive and qualitative in nature, i.e., some symptoms may be
manifested in a delayed fashion and others in a deviant
fashion. Autistic-specific characteristics are not limited,
however, to these three core features. Rather, symptoms
appear intertwined with other impairments on a continuum.
The National Research Council (2001) reported that one-
third to one-half of individuals diagnosed with autism do
not make use of functional speech. Many children with
autism develop problem behaviors as a form of commu-
nicative expression (Buschbacher & Fox, 2003). It has been
reported that approximately 60% of the children with ASD
function with an IQ range below 50, and about 20% have
an IQ between 50 and 70 (Campbell, Morgan, & Neighbors,
2004; Ritvo & Freeman, 1977). Bodfish, Symons, Parker,
and Lewis (2000) found that repetitive behavior occurred at
higher rates and with greater severity in individuals with
G. Walker
Department of Communication Sciences and Special Education,
University of Georgia, Athens, GA, USA
G. Walker (&)
Department of Global Policy Studies in Education, University of
Illinois, 405 E. Burkwood Ct. W., Urbana, IL 61801, USA
e-mail: gab_rielaa@yahoo.com
123
J Autism Dev Disord (2008) 38:261–275
DOI 10.1007/s10803-007-0390-4
autism than in individuals with intellectual disabilities.
Individuals with autism also exhibit impairments in exec-
utive functioning, including deficits in planning, organi-
zation, flexibility of mental functioning, and self-regulation
(Ozonoff, 1998). Other impairments in individuals with
autism include perseveration, resistance to change, sensory
overarousal and underarousal (e.g., fluorescent lights,
sounds, smells, pain, etc.), reluctance to share, attachment,
theory of mind, social judgment and perception, a distorted
sense of time, self-stimulatory behaviors, self-injurious
behaviors, and psychomotor coordination problems which
result in limited verbal and motor imitation (Cohen &
Volkmar, 1997; Dawson et al., 2004). These behavior
challenges may interfere with learning, and prevent
acquisition of adaptive behavior, since they can compete
with socially appropriate behavior. They can potentially (a)
cause the learners to harm themselves and others; (b) re-
duce the locus of attention and the ability to shift attention
within and extra stimulus (stimulus overselectivity); (c)
restrict motivation; (d) constrain ability to attend to stimuli
and instruction or to make conditional discriminations
across sensory modalities; and (e) confine ability of
expression. Teachers working with children with ASD
characteristics should therefore be prepared to work with
learners who display a wide range of diverse abilities.
Time delay strategies are categorized as response
prompting procedures. In response prompting procedures,
the teacher initially provides the student with assistance
following a stimulus, and continues until the learner emits
a target response (Morse & Schuster, 2004). The prompts
are then faded or removed in subsequent presentations of
the stimulus as instruction progresses. Presentation of the
prompts before the student responds reduces errors and
provides more opportunities for reinforcement. Prompting
can occur simultaneous to the stimulus (simultaneous
prompting) or with a short delay (time delay).
A variant of time delay instructional strategies, also
known as ‘‘prompt delay’’ or ‘‘delay’’ procedures, was
introduced by Touchette (1971). The method was used to
teach a form discrimination task to children with intellec-
tual disabilities. Time delay is a method for transfer of
stimulus control in which varying amounts of time are
inserted between the natural cue or task direction and a
controlling prompt (i.e., a prompt that ensures that the
learner performs the correct response). With this technique,
the stimulus control is transferred from the prompt to the
target natural discriminative stimulus (Schuster, Gast,
Wolery, & Guiltinan, 1988). The CTD and PTD response
prompting procedures may be considered two of the most
structured interventions in terms of ‘‘the extent to which
they influence control over students’ interactions with the
environment’’ (Wolery & Schuster, 1997, p. 68). During
initial time delay trials, the discriminative stimulus and the
controlling prompt are presented simultaneously. Since
there is no delay between the two, these trials are called
zero-second (0-s) delay trials. Following a number of 0-s
delay trials, the delay between the natural cue and the
controlling prompt is increased (Gast, Wolery, Ault, Doyle,
& Alig, 1988), hence the instructional prompt is faded on a
time dimension. Time delay strategy may hold great po-
tential as a training procedure with individuals with ASD
since they do not require the prerequisite of imitation, but
essentially the following prerequisite skills: (a) perceive
the instructional prompt; (b) wait for a certain number of
seconds (if the ability to wait is not already in the learner’s
repertoire, a wait training can be performed); (c) perform
task-specific response; (d) increase positive behavior in the
presence of the reinforcers (participant’s history with
reinforcing systems); and (e) follow one step directions.
Handen and Zane (1987) reviewed 26 investigations
published between 1971 and 1986. The authors found both
graduated delay (PTD) and standard delay (CTD) to be
efficient teaching procedures for people with various dis-
abilities, although not all participants had benefited from
the procedures. More recently, Wolery et al. (1992) re-
viewed the use of the CTD strategy for teaching discrete
responses to participants with various diagnoses. Finally,
Schuster et al. (1998) reviewed studies that used CTD with
chained tasks. Both of the latter reviews found that pro-
gressive and constant time delay procedures ‘‘(a) have
been effective with a wide range of students, ages, and
disabilities, (b) have been used effectively in a variety of
settings with an assortment of persons serving as trainers,
(c) have been successful in both individual and group
arrangements, (d) have employed a variety of prompts and
other procedural parameters when delivering instruction,
and (e) have been used in studies that are considered
methodologically sound’’ (Schuster et al., 1998, p. 104).
None of the reviews of the CTD and PTD studies that
were published before 2006 compared, either directly or
indirectly, the two response prompting procedures. Fur-
thermore, none of the earlier reviews investigated the use
of time delay procedures for teaching only individuals with
ASD. By focusing on a specific population, researchers
may be better able to make adaptations of existent
instructional methodologies based on unique characteristics
of various subpopulations of students. The future of the
field of autism research looks promising primarily because
of the recent increased prevalence of ASD diagnoses in the
population. Early intervention and naturalistic procedures
appear to be effective with the ASD population. As pre-
viously stated, CTD and PTD may be considered two of the
most structured interventions because of their influence on
student-environment interactions. At the same time, CTD
and PTD are flexible enough to be used naturalistically and
to be implemented as early as possible. The purpose of the
262 J Autism Dev Disord (2008) 38:261–275
123
current review is to provide an updated summary of re-
search that has been conducted on both the CTD and PTD
procedures used for teaching individuals with ASD.
Articles published during the preceding 20 years are
included because: (a) the ASD diagnosis has been better
described as empirical research data have broadened; (b)
the need to find empirically-based educational strategies
for people with ASD increased as their number increased;
and (c) the body of published research is so small that it
was necessary to consider as many available studies as
possible. The CTD and PTD procedures are becoming
increasingly important, as it is shown by the increasing
number of studies using time delay strategies with indi-
viduals with ASD characteristics, also because of the
increasing number of individuals that are diagnosed with
ASD, and because they were shown to be both effective
and efficient (e.g., Bennett, Gast, Wolery, & Schuster,
1986; Keel & Gast, 1992). Evaluating the variables in-
cluded in these procedures may help both researchers and
practitioners to manipulate different parameters to optimize
the education of students diagnosed with ASD. This review
is organized around demographic variables, procedural
parameters, methodological adequacy, and outcome mea-
sures, since these are the primary characteristics of an
instructional method, as identified in the empirical litera-
ture (Schuster et al., 1998; Wolery et al., 1992).
Methodology
Three types of literature review methods were used in order
to identify published empirical studies that investigated the
effectiveness and efficiency of time delay strategies for
teaching persons with ASD. First, electronic searches on
ERIC, PsycINFO, PsycARTICLES, Professional Devel-
opment Collection, and Education Full Text library data-
bases were conducted. Descriptors for the electronic search
included ‘‘time delay’’, ‘‘prompt delay’’, ‘‘stimulus de-
lay’’, ‘‘delay procedure’’, ‘‘constant time delay’’, ‘‘pro-
gressive time delay’’, and the names of researchers known
to have published articles that assessed effectiveness of the
time delay strategy. These key terms were used in con-
junction with terms such as ‘‘PDD’’, ‘‘ASD’’, ‘‘autism’’,
‘‘autistic’’, ‘‘autistic-like’’, and ‘‘Asperger Syndrome’’.
The text of each result was investigated. Second, a manual
search of the Journal of Applied Behavior Analysis and the
Journal of Autism and Developmental Disorders, due to
their reputation in the field of behavioral and autism re-
search, revealed additional articles pertaining to this topic.
Third, references cited in identified and recent articles were
used to obtain additional target studies. The search was
narrowed to studies that met the following criteria: (a) a
time delay strategy was used; (b) participants were
identified as having a PDD, ASD, or ‘‘autistic-like’’ con-
dition; (c) the studies were published in a peer-reviewed
journal; (d) the article was published between January 1985
and April 2005; and (e) the investigations were published
in English. Inclusion criteria and potential papers for this
review were discussed with two other experienced faculty
members from the University of Georgia. Once the
manuscripts were collected, each one of them was coded
according to specific target demographic, procedural,
methodological, and outcome parameters. This review is an
analysis and an indirect comparison of these variables, as
resulted from both CTD and PTD studies.
Results
This search resulted in identification of 10 studies pub-
lished in five different journals that reported outcomes of
the CTD procedure and 12 studies from five journals that
used the PTD procedure: J. of Applied Behavior Analysis
(8), J. of Autism and Developmental Disorders (1), J. of
Behavioral Education (2), J. of Developmental and Physi-
cal Disabilities (2), Early Childhood Research Quarterly
(1), Education and Training in Mental Retardation and
Developmental Disabilities (2), Education and Treatment
of Children (1), Exceptional Children (1), Preventing
School Failure (1), Research in Developmental Disabilities
(1), J. of School Psychology (1), J. of Special Education
Technology (1). The 22 articles are indicated in the refer-
ence section with an asterisk. Tables 1 and 2 summarize
the main descriptors for each of the studies identified. Of
the 10 reports on CTD, there were 4 treatment packages
that combined time delay with other instructional proce-
dures (Dipipi, Jitendra, & Miller, 2001; Gardill & Browder,
1995; Morse & Schuster, 2000; Winterling, Gast, Wolery,
& Farmer, 1992), and one comparison of CTD with the
system of least prompts (Ault, Wolery, Gast, Doyle, &
Eizenstat, 1988). Three of the twelve studies compared the
effectiveness of the PTD procedure with other instructional
procedures, including the system of least prompts strategy
(Godby, Gast, & Wolery, 1987), a visual fading procedure
(Matson, Sevin, Box, Francis, & Sevin, 1993), and the
most-to-least instructional strategy (Heckaman, Alber,
Hooper, & Heward, 1998).
Demographic Variables
The demographic variables analyzed included gender, age,
level of functioning, history with the time delay procedure,
training settings, teachers, instructional arrangements, and
the target dependent variables.
J Autism Dev Disord (2008) 38:261–275 263
123
Table 1 Constant Time Delay (CTD)
Citation Population Diagnosis Setting Behavior Procedure Controlling
prompt
Error correction Design Effectiveness
No. &
Gender
Age
(years-
months)
Name Type
Ault et al. (1988)1M
1M
8-5
8-3
Autism
PDD
Cl 16 numeral
identification
(8 per
procedure)
D CTD 4-s
SLP
CTD: Verbal Mo
SLP:
1Task direction
2TD, SD
3TD, number set
4TD, number
name
‘‘No’’ (VI) + wait
for
the rest of the
intertrial interval
Parallel
treatment
design
+2
CTD more
efficient
Nietimp & Cole (1992)1M
1M
1F
12-0
13-4
12-8
Autism or
PDD
Cl Responses to
social
greetings
D CTD 5-s VI ‘‘No’’ + verbal
model
ABA + 2
~ 1(not ASD)
Winterling et al. (1992)2M
2F
(at least 1
with ASD)
17–21 Mo, Sev MR,
& Autism
Cl Safety skills Ch CTD 5-s VI + Mo Interruption + VI +
model
MP · subj and
replicated ·
tasks
+1
Hughes, Schuster &
Nelson (1993)
1M
1M
10
11
CP, Sev MR
Sev MR,
autistic
charact
Cl Dressing skills Ch CTD 5-s FP + verbal
prompt
Teacher performs
step for the
student
MP ·
behaviors
+2
Gardill & Browder
(1995)
1F
1F
1F
12
13
12
Autism
Social deficits
Fragile X
Cl+
Com
Independent
purchasing
D CTD 4-s PP Controlling prompt MB · subjects + 2; 1 not
complete
training
(not ASD)
Wall & Gast (1997)1M
1M
1M
1M
42-3
37-2
15-5
15-7
Down S
Mo MR
Autism
Autism
Com Leisure skills Ch CTD 4-s VI + G
(pointing)
Interruption +
controlling
prompt
MP · behaviors
replicated ·
subjects
+4
Wall, Gast, & Royston,
(1999)
1M
1M
1M
1M
20-7
21-10
19-11
20-0
Prof MR
Sev Autism
Sev MR
Sev MR
Cl Leisure skills Ch CTD 4-s VI + FP/PP or
model
Interruption +
Con-trolling
prompt
MP · beh
and
replicated ·
subj
+4
264 J Autism Dev Disord (2008) 38:261–275
123
Table 1 continued
Citation Population Diagnosis Setting Behavior Procedure Controlling
prompt
Error
correction
Design Effectiveness
No. &
Gender
Age (years-
months)
Name Type
Morse & Schuster
(2000)
1F
1M
1M
1M
1M
1F
1M
1M
1M
1F
12
10
12
12
8
9
9
5
6
6
FMD
FMD
FMD
FMD
FMD
Autism
Down S
FMD
Autism
Autism
Com Shop for
groceries
Ch CTD 4-s Model + VI PP MB ·
subjects
+6
2 improved
(not ASD)
2 not under
interv
(participants
with ASD)
Dipipi et al. (2001) 1F 18 Mi MR,
seizure
disorder,
autistic
behavior
Home Social
phrases
D CTD 5-s
Self-monitoring
DRO
Model N/A AB + 1
Norman, Collins, &
Schuster, (2001)
1F
1M
1M
8-1
9-8
12-3
Mo MR
Down S
Mi MR
Down S
Autism,
Mo MR
Cl Self-help
skills
Ch CTD 5-s + Video
instruction
Videotape
model + VI
Interrup + pause
videotape + PP
MP ·
behaviors
+3
Legend: Prof MR = Profound Mental Retardation; Sev MR = Severe Mental Retardation; Mo MR = Moderate Mental Retardation; Mi MR = Mild Mental Retardation; PDD = Pervasive
Developmental Disorders; Dev Delay = Developmental Delay; Down S = Down Syndrome; FMD = Functional Mental Disability; Com = Community; Cl = Classroom; Exp
Room = Experimental Room; D = Discrete behaviors; Ch = Chained behaviors; PTD = Progressive Time Delay; CTD = Constant Time Delay; SLP = System of Least Prompts;
LTM = Least to Most Prompts; DRO = Differential Reinforcement of Other Behavior; TD = Task Direction; SD = Discriminative Stimulus; VI = Verbal Instruction; II = Indirect Verbal
Instruction; G = Gestural prompt; PP = Partial Physical Prompt; FP = Full Physical Prompt; MB = Multiple Baseline design; MP = Multiple Probe design; ABA = reversal design;
AB = baseline and intervention design; · = across; N/A = not reported
Note: Underlined participants were included in this study
J Autism Dev Disord (2008) 38:261–275 265
123
Table 2 Progressive Time Delay (PTD)
Citation Population Diagnosis Setting Behavior Procedure Controlling
Prompt
Error
Correction
Design Effectiveness
No. &
Gender
Age
(years-
months)
Name Type
Charlop,
Schreibman, and
Thibodeau (1985)
7M 109, 63,
115,
63, 51,
65, 69
Autism Cl + Exp
Room
Spontaneous
verbalization
D PTD 0-10-s
2-s increment
Verbal Mo ‘‘No’’ (VI) + SD
removed (time-
out)
MB · subjects + 7
Charlop & Walsh
(1986)
1M
1M
1M
1M
6-0
8-9
7-11
8-6
Autism
Autism
Autism
Autism
Com
(homeout-
door)
Spontaneous
verbalization
D PTD 0-10-s
2-s increment
Verbal Mo Time-out for
the rest of the
intertrial time
MB · subjects
and settings
+4
Godby et al. (1987)1F
1M
1F
8-3
16
11
Autism
Sev MR
Cerebral
Palsy
Cl Objects
identification
D PTD 0-7-s
1-s increment
Gestural Mo 10-s in-seat
time-out
Parallel
treatments
design
+3
PTD superior
to SLP
System of Least
Prompts
1 VI, G; 2 VI, Mo;
3 VI, PP; 4 VI, FP
Wolery, Gast, Kirk,
and Schuster (1988)
1M
1M
1M
7
9
7
Autism
Autism
Autism
Cl Occupational
labels
D PTD 0-10-s
1-s increment
Verbal Mo 5-s in-seat
time-out
MP · behaviors + 3
Matson Sevin,
Friedley, and Love
(1990
)
1F
1M
1M
11
11
9
Autism
Autism
Autism
Exp
Room
Spontaneous
verbalization
D PTD 0-10-s
2-s increment
Verbal Mo Controlling
prompt
MB · behavior + 3
Charlop & Trasowech
(1991)
1M
1M
1M
7-9
8-7
7-11
Autism
Autism
Autism
Com Spontaneous
speech
D PTD 0-10-s
2-s increment
Verbal Mo No reinforcement MB · subjects
and settings
+3
Ingenmey & Van
Houten (1991)
1M 10 Autism Home Spontaneous
speech
D PTD 0-10-s
2-s increment
Verbal Mo Controlling
prompt
MB · behavior + 1
Matson et al. (1993)1M
1M
1M
5
4
4
Autism
Autism
Autism
Clinic
Home
Cl
Self-initiated
verbalizations
D PTD 0-10-s
2-s increment
Verbal Mo No reinforcement MB · behaviors + 3
PTD superior
to fading
Visual fading 1 cue card;
2card + nonverbal
SD; 3 card size
½ + nonverbal
SD; 4 card size ¼;
5 nonverbal SD
266 J Autism Dev Disord (2008) 38:261–275
123
Table 2 continued
Citation Population Diagnosis Setting Behavior Procedure Controlling
Prompt
Error
Correction
Design Effectiveness
No. &
Gender
Age
(years-
months)
Name Type
Venn et al.
(1993)
1M
1M
1M
4-9
3-4
4-10
Autism
Dev
Delay
Autistic
features
Cl
(inclusive)
Imitation during
art activity
D PTD 0-6-s
2-s increment
PP No
reinforcement
MP ·
subjects
+3
Leung (1994)1M
1M
1M
55
11
7
Autism
Autism
Autism
Home Spontaneous
requests
D PTD 0-10-s
2-s increment
Verbal Mo SD re-moved MB ·
subjects
+3
Taylor & Harris
(1995)
1F
1M
1M
9
5
9
Autism
Autism
Autism
Cl Spontaneous
question
D PTD 0-10 s
2-s increment
Verbal Mo
(+ G)
No reinforcement +
same delay
on next trial
MB ·
subjects
+3
Heckaman
et al.(1998)
4 M 9, 6,
7, 6
Mo to
Sev
Autism
Cl Disruptive
behavior;
Identification
of sight
words
D PTD 0-5-s
1-s increment
One of the six
levels of
hierarchy,
depending
upon
participant
‘‘No’’ (VI) +
controlling
prompt + return to
previous delay
level
Alternating
treatment design
replicated · subj:
1easy vs. difficult
task;
+4
1 difficult tasks
correlated w/
disruptive
behavior
for 2 subj
LTM 1TD
2IVI
3VI
4 partial
model
5 model
6PP
2LTM vs.
PTD
2 PTD
more
efficient
vs. LTM
Legend: Prof MR = Profound Mental Retardation; Sev MR = Severe Mental Retardation; Mo MR = Moderate Mental Retardation; Mi MR = Mild Mental Retardation; PDD = Pervasive
Developmental Disorders; Dev Delay = Developmental Delay; Down S = Down Syndrome; FMD = Functional Mental Disability; Com = Community; Cl = Classroom; Exp
Room = Experimental Room; D = Discrete behaviors; Ch = Chained behaviors; PTD = Progressive Time Delay; CTD = Constant Time Delay; SLP = System of Least Prompts; LTM = Least
to Most Prompts; DRO = Differential Reinforcement of Other Behavior; TD = Task Direction; SD = Discriminative Stimulus; VI = Verbal Instruction; IVI = Indirect Verbal Instruction;
G = Gestural prompt; PP = Partial Physical Prompt; FP = Full Physical Prompt; MB = Multiple Baseline design; MP = Multiple Probe design; ABA = reversal design; AB = baseline and
intervention design; · = across
J Autism Dev Disord (2008) 38:261–275 267
123
CTD Procedure
Results of a total of 36 subjects with various disabilities were
reported in the 10 articles, of which at least 16 participants
were identified with ASD (14) or autistic characteristics (2).
Winterling et al. (1992) did not specify which of the four
participants was identified with ASD, so data from this study
are presented separately under each analyzed variable, since
we know that at least one of the participants was diagnosed
with ASD or autistic-like behaviors.
From the 16 participants, 10 were male and 5 were female.
In Winterling et al. (1992), two males and two females re-
ceived instruction. Eight participants were functioning in the
moderate cognitive range (IQ = 40–55), six were function-
ing in the severe and profound intellectual range (IQ < 40),
and one was functioning in the mild cognitive range
(IQ = 55–70). The participants in Winterling et al. (1992)
study were functioning in the moderate range of intellectual
disabilities. Eight participants were adolescents (12–
18 years), six were elementary-aged students (6–11 years),
and one young adult 21 years old. Winterling et al. (1992)
reported that their four participants were between 17 and
21 years of age. Nine individuals had minimal verbal abili-
ties (i.e., some functional language), two had average verbal
abilities, and four were nonverbal. Verbal abilities were not
reported in Winterling et al. (1992). Only three studies re-
ported participants’ previous history with a time delay pro-
cedure, and in these studies there were a total of five
individuals with ASD. Only two studies reported that par-
ticipants with ASD were diagnosed according to DSM III
(1980) criteria (Wall & Gast, 1997; Wall et al., 1999), and in
one study participants were diagnosed based on clinical
interviews by psychiatrists (Nietimp & Cole, 1992). Only 8
studies reported what type of personnel delivered instruc-
tion: 6 classroom teachers and paraprofessionals, 4 graduate
students, 2 caregivers, and 1 research or university associ-
ated staff member in total. CTD instruction was conducted
most often during one-on-one sessions, but two investiga-
tions used small group arrangements of three students
(Nietimp & Cole, 1992; Norman et al., 2001). Four of the ten
studies targeted discrete behaviors (i.e., those which required
one unit of behavioral response), and six targeted chained
behaviors (i.e., those which required sequenced steps or units
of behavior to form a complex skill) (Wolery, Ault, & Doyle,
1992). There were two studies each teaching leisure skills,
shopping skills, self-help skills, and social interaction skills,
one study was found on teaching safety skills, and one on
numeral identification.
PTD Procedure
Of the 40 participants involved in the 12 studies, 37 (34
males and 3 females) were diagnosed with ASD (see
Table 2). Six participants were preschool aged (3–5 years),
thirty-three participants were elementary school-aged
(6–11 years), and one was an adolescent (12–18 years).
There were approximately 21 participants with minimal
expressive language abilities, 7 participants who were
nonverbal, and 6 who had ‘‘average’’ expressive verbal
skills. Only 6 of the 12 studies reported the assessment tool
used to diagnose ASD in participants: (a) the Childhood
Autism Rating Scale (Heckaman et al., 1998), (b) the DSM
III-R, 1980 (Leung, 1994; Matson et al. 1990; Matson
et al., 1993; Taylor & Harris, 1995), and (c) and the Na-
tional Society for Autistic Children—1978 test (Charlop &
Walsh, 1986). Matson et al. (1990) reported one girl being
treated with 25 g of fenfluramine daily to control hyper-
activity, and this was the only study in which the use of
medication during the investigation was mentioned. Per-
sonnel delivering instruction included classroom staff
(Godby et al., 1987; Heckaman et al., 1998; Taylor &
Harris, 1995; Venn et al., 1993; Wolery et al.,
1988), re-
search or university associated staff (Charlop et al., 1985;
Charlop & Walsh, 1986; Ingenmey & Houten, 1991;
Leung, 1994; Matson et al., 1990; Matson et al., 1993), and
parents (Charlop & Trasowech, 1991). All 12 studies were
conducted in a 1:1 instructional arrangement, and partici-
pants were taught discrete behaviors. Spontaneous speech
was taught in 8 studies, while object identification, occu-
pational labels, imitation of art activities, and sight words
were targeted in the remaining studies.
Procedural Variables
The procedural parameters analyzed include attending
cues, task directions or discriminative stimuli, number of
sessions at 0-s delay, frequency of instruction, duration of
delay interval, response intervals, presentation of control-
ling prompts, format of task analyses presentation, nature
of trial sessions, consequences, and procedural modifica-
tions.
CTD Procedure
Attending cues and responses were used in order to ensure
a student’s attention and focus on the critical features of the
target stimulus (Wolery et al., 1992). Three studies (30%)
used general active attentional cues (Hughes, Schuster, &
Nelson, 1993; Morse & Schuster, 2000; Winterling et al,
1992), specifically ‘‘Are you ready to work?’’, ‘‘Are you
ready to shop for (item)?’’, and ‘‘(Name), look’’. Specific
attending cues required task-specific responses such as
matching (Ault et al., 1988) and pointing (Wall & Gast,
1997; Wall et al., 1999). In most studies, discriminative
stimuli were accompanied by the presence of a target
268 J Autism Dev Disord (2008) 38:261–275
123
stimulus. The number of sessions at 0-s delay ranged from
1 (Winterling et al., 1992) to as many as necessary until
one session of 100% prompted corrects was recorded (Ault
et al., 1988; Norman et al., 2001). Two 0-s delay sessions
were employed in three studies (Morse & Schuster, 2000;
Nietimp & Cole, 1992; Wall & Gast, 1997), three 0-s
sessions were conducted by Wall et al. (1999), four ses-
sions by Gardill and Browder (1995), and ten by Hughes
et al. (1993). Dipipi et al. (2001) used 0-s delay trials
conducted at the beginning of each of the 6 CTD sessions.
On average, 3.4 0-s sessions were used. The length of the
delay interval used was 4-s in five of the studies and 5-s in
the remainder.
Controlling prompts are prompts that ensure correct stu-
dent responses. Because the participant in the Hughes et al.
(1993) study was not imitative, a verbal prompt was paired
with a full physical prompt. Overall, 8 studies reported the
use of a model. Six investigations that taught chained tasks
used total task presentation. Seven studies used a distributed
trial format, two used massed and distributed trials (Gardill
& Browder, 1995; Norman et al., 2001), and one used
massed trials (Ault et al., 1988). Only 5 studies reported the
length of the response interval that the student was given to
complete the correct response. Target response intervals
included 3-s to 5-s (Ault et al., 1988), 10-s to 2 min (Morse
& Schuster, 2000), 15-s (Norman et al., 2001), 20-s (Hughes
et al., 1993), and 75-s (Winterling et al., 1992). Seven of the
investigations described the frequency of the training, which
ranged from 1 or 2 sessions per day, 3 times per week, to
2 days per week with 2 trials per session. On average there
were 1 to 2 sessions a day, with 7 to 8 trials per session.
In six investigations, only descriptive verbal praise was
used as a reinforcer. The use of differential reinforcement
was reported in three studies, and Dipipi et al. (2001) did
not specify the nature of the reinforcer used. Error cor-
rection procedures consisted equally of interruption of the
response and the delivery of the controlling prompt. Four
investigations reported the use of procedural modifications:
(1) task analyses and leisure equipment were adapted to the
individual physical characteristics, and additional physical
assistance was added (Wall et al., 1999); (2) the controlling
prompt was changed to a physical prompt and 34 additional
0-s delay trials were conducted on two steps of the task
analysis for 1 participant (Winterling et al., 1992); (3)
delivery of one 0-s delay trial with each student prior to the
generalization probe sessions (Morse and Schuster, 2000);
and (4) addition of differential reinforcement and massed
trials (Norman et al., 2001).
PTD Procedure
Only two studies used general active attentional cues
(Charlop & Walsh, 1986; Godby et al., 1987). The
discriminatory stimuli varied, such as the use of a direct
task direction in eight studies (Charlop & Walsh, 1986;
Godby et al., 1987; Heckaman et al., 1998
; Taylor &
Harris, 1995; Venn et al., 1993; Wolery et al., 1988), and
the simple presentation of the discriminative stimulus in
the remaining studies. The number of 0-s delay sessions
were reported in five studies and ranged from one session
to as many as necessary until correct imitations we re-
corded for three consecutive sessions (Charlop & Tra-
sowech, 1991; Godby et al., 1987; Leung, 1994; Venn
et al., 1993; Wolery et al., 1988). The delay interval in-
creased from 1-s increments in two studies (Godby et al.,
1987; Wolery et al., 1988), to 2-s increments in nine
studies. Heckaman et al. (1998) initially increased the de-
lay interval from 0-s to 0.5-s, then to 1-s, and continued to
increase the delay interval by 1-s thereafter. Six studies
reported the use of a criterion in order to increase the delay
interval to the next level: (a) 4 out of 5 trials over two
consecutive sessions (Charlop & Walsh, 1986; Charlop &
Trasowech, 1991; Matson et al., 1990; Matson et al.,
1993); (b) 90% correct responses for two consecutive
sessions (Heckaman et al., 1998), and (c) every 8 trials
(Ingenmey & Houten, 1991). The rest of the studies did not
report the use of a criterion when increasing the delay
interval. The ceiling of the delay interval ranged from 5-s
to 10-s, with a mean of approximately 9-s across all stud-
ies. The frequency of instruction was reported in nine
studies and varied from as often as 3 sessions per day to 2
per week, with a mean of 5.7 sessions per week. A verbal
model was used as a controlling prompt in nine studies.
The remaining studies used gestural model and physical
prompts. Natural positive reinforcers were used in all the
studies. The error correction procedures employed con-
sisted mainly of a time-out period and deprivation of
reinforcement (i.e., the teacher did not attend to the student
for a specified interval of time and he/she did not deliver
the reinforcement), and less frequently of an interruption of
the response and the delivery of the controlling prompt.
Only two studies mentioned the use of different conse-
quences for incorrect responses after the prompt and for
lack of a response (Taylor & Harris, 1995; Venn et al.,
1993). Procedural modifications such as the addition of
another reinforcer (Ingenmey & Houten, 1991; Matson
et al., 1990; Venn et al., 1993) or special discriminative
training (Charlop et al., 1985) were used in four studies.
Outcomes and Outcome Measures
The parameters analyzed in this section included effec-
tiveness of the time delay procedure, treatment packages
and comparisons, efficiency, maintenance, and general-
ization measures.
J Autism Dev Disord (2008) 38:261–275 269
123
CTD Procedure
This strategy was effective in teaching targeted skills to
criterion levels to all 14 participants to whom CTD training
was introduced. Due to time constraints, Morse and
Schuster (2000) were unable to complete training of 2
participants or begin training with 2 other participants. In
Dipipi et al. (2001) and Gardill and Browder (1995) it was
difficult to determine a criterion for the CTD phase of
intervention on the basis of the reported data. One partic-
ipant did not show improvement with the CTD or system of
least prompts (SLP) procedures and he was not included in
the Ault et al. (1988) study. Various indices of efficiency
(errors, sessions, and instructional time to criterion) were
reported, but only error percentages were consistently re-
ported. This efficiency parameter was reported in all six
studies that targeted chained tasks, and only in one study
that targeted discrete tasks. The mean error percentage
across the seven studies was 9.95% errors toward criterion,
with a range from 1.35% to 32.50% errors to criterion.
Across five studies, the moment of transfer of stimulus
control (i.e., the point at which the percentage of un-
prompted correct responses exceeded the percentage of
other responses) occurred within a range from 2 to 17
sessions, and at a mean of 5.57 sessions. In general,
behaviors acquired were maintained during follow-up
probes, as reported in nine studies.
PTD Procedure
The PTD strategy was effective with all 37 participants.
The mean error percentage averaged across three studies
was 2.8 errors towards criterion with a range from 0.3% to
5.3% errors (Godby et al., 1987; Heckaman et al., 1998;
Venn et al., 1993; Wolery et al., 1988.) Maintenance data
were collected during 9 of the 12 studies, except in Charlop
et al. (1985), Godby et al. (1987), and Venn et al. (1993).
All acquired behaviors were maintained over time. Except
in Matson et al. (1990) and Godby et al. (1987), the
remaining 10 studies collected generalization data across
persons, settings, and materials. Transfer of stimulus con-
trol was reported in four studies (Charlop & Trasowech,
1991; Taylor & Harris, 1995; Venn et al., 1993; Wolery
et al., 1988), and ranged from two to seven sessions, with a
mean of 4.0 sessions. The mean length of the PTD inter-
vention across 11 studies was 13.93 sessions per behavior,
ranging from 4.5 to 32.2 sessions. In the Charlop and
Trasowech (1991) study, the length of the intervention was
difficult to calculate due to the visual representation of the
intervention (graph), but the authors describe that the PTD
treatment lasted from 1 month to several months. In the
three comparison studies, the PTD strategy was more
effective in terms of efficiency measures than the system of
least prompts (SLP), least-to-most (LTM) prompts, and
visual fading procedures (see Table 3).
Methodological Adequacy
The variables used for the analysis of the methodological
adequacy include the types of experimental designs used,
type and percentage of interobserver agreement, procedural
fidelity measures, and social validity measures.
CTD Procedure
All studies used single-subject experimental designs
(Tawney & Gast, 1984). Adequate experimental control
was established in 9 of the 10 studies. In Dipipi et al.
(2001), the study involved only 1 participant and therefore,
replications and experimental control did not exist. Point-
by-point interobserver agreement data (i.e., percent of
agreements divided by percent of agreements plus dis-
agreements multiplied by 100) were collected in all studies
and ranged from 94% to 100%, with an average of 98.0%
during instructional sessions. Procedural fidelity, i.e., the
extent to which the intervention was implemented as
planned, was reported in six studies (Ault et al., 1988;
Hughes et al.,
1993; Morse & Schuster, 2000; Wall &
Gast, 1997; Wall et al., 1999; Wintering et al., 1992) and
ranged from 89% to 100%, with an average of 97.4%. Only
five studies mentioned the nature of the procedural errors.
The reported procedural errors included the teacher’s fail-
ure to: (a) wait for the delay interval and to provide con-
sequences (Winterling et al., 1992); (b) record a stop time
at the end of the session and provide more frequent rein-
forcement (Wall et al., 1999); (c) secure a student’s
attention and deliver the consequences (Wall & Gast,
Table 3 Treatment comparisons, according to data reported in the
studies
Ault et al. (1988) Mean sessions per student Error
percentage
CTD 9.1 10.8
PTD 3.7 11.6
Godby et al. (1987) No. of sessions Error
percentage
PTD 120 (6 h) 4.5
SLP 141 13.0
Matson et al. (1993) No. of sessions Error
percentage
PTD 47.3 Not reported
Visual fading 56.7 Not reported
Heckaman et al. (1993) Mean sessions per student No. of errors
PTD – 47
LTM – 609
270 J Autism Dev Disord (2008) 38:261–275
123
1997); (d) present the correct prompt (Hughes et al., 1993);
and (e) present stimulus and correct consequence (Ault
et al., 1988). Social validity data on the importance of
acquiring target behaviors (Nietimp and Cole, 1992; Wall
et al., 1999) on methods of instruction and on outcomes
(Morse & Schuster, 2000; Norman et al., 2001; Winterling
et al., 1992) were reported in five studies, and were gath-
ered from participants’ caregivers, general and special
education teachers, school administrators, participants
themselves, and typical peers.
PTD Procedure
All 12 studies used single-subject research designs to
evaluate the effectiveness of the PTD strategy for teaching
children with ASD or autistic-like characteristics. All
studies employed adequate experimental control and re-
ported point-by-point interobserver agreement percentages,
with a mean of 95.2% and range from 80% to 100%.
Procedural fidelity data were collected in five studies
(Godby et al., 1987; Heckaman et al., 1998; Matson et al.,
1990; Venn et al., 1993; Wolery et al., 1988) and ranged
from 95% to 100%, with an overall mean of 97.3%. Social
validation measures were reported across objectives,
methods, and outcomes in five studies (Charlop & Tra-
sowech, 1991; Charlop & Walsh, 1986; Matson et al.,
1990; Matson et al., 1993; Wolery et al., 1988) and in-
cluded ratings of video tapes, questionnaires, Likert-scales,
individual educational plan objectives, non-disabled peers’
recordings, and t-tests.
Conclusions and Discussion
Conclusions
An analysis of the demographic, procedural, outcomes, and
methodological variables found in 22 empirical studies was
used to summarize research on the effectiveness of the
CTD and PTD procedures as methods for teaching indi-
viduals identified with ASD symptoms. Both procedures
have been shown to be effective with persons identified
with autistic characteristics, in various settings, in different
instructional arrangements, and with adult instructors,
regardless of their level of cognitive functioning, gender, or
age. For a detailed comparison between the CTD and PTD
procedures, refer to Table 4. The extent of the average
number of 0-s delay sessions, behavior maintenance and
generalization over time, and procedural reliability suggest
that the two procedures are similar on these variables.
However, the CTD procedure resulted in more errors to
criterion (10.0%) than PTD (2.6%), the transfer of stimulus
control occurred later in the CTD studies (after 5 sessions)
than in the PTD studies (after 4 sessions), and a greater
magnitude of procedural modifications was required in the
CTD studies. A possible explanation is that the CTD pro-
cedure was implemented with more children functioning in
the severe to profound cognitive, hence other abilities,
range (11.3% of the 53 total sample) than PTD (3.8% of the
total sample). For a more accurate comparison of the
variables of these two teaching techniques, direct com-
parison studies are recommended. For conclusions of this
review, see Table 5.
Limitations
Taking a broad view at the sample of publications in this
review, one notices that there are limitations which suggest
researchers and practitioners caution in interpreting the
results and careful programming when working with per-
sons with ASD. The authors of the articles suggested some
potential limitations of these studies. First, there is the
difficulty to generalize the results of individual studies into
general conclusions because of the size of the sample (Wall
& Gast, 1997). Second, the researchers should be specific
about taking pre-instructional data on participants’ verbal
abilities because of the possibility that verbalizations may
have been conditioned by other variables, such as day-time
cues (Charlop & Trasowech, 1991). Third, the probability
that verbalizations could be imitations of trainers’ original
comments should be taken into consideration (Ingenmey &
Houten, 1991
). Fourth, procedural modifications may
interfere with the effects of the time delay procedure
(Matson et al., 1990). Fifth, there is the possibility that the
training of the target responses in the same order for all
children may result in order effects (Matson et al., 1990).
Sixth, it is impossible to separate treatment variables in the
case of intervention packages (Dipipi et al., 2001; Gardill
& Browder, 1995; Morse & Schuster, 2000; Nietimp &
Cole, 1992; Winterling et al., 1992). Last, there is a lack of
generalization training across conditions, settings, and
persons (Gardill & Browder, 1995; Ingenmey & Houten,
1991; Johnston, Nelson, Evans, & Palazolo, 2003; Win-
terling et al., 1992).
Other limitations that emerged during the reviewing
process include: (a) inconsistent reporting of diagnosis and
diagnostic criteria; (b) since acquiring speech was targeted
in 45% of the studies, failure to specify participants’ verbal
abilities as a predictor of future performance (Cohen &
Volkmar, 1997), one cannot appreciate the amount of
individual progress due to intervention; (c) failure to
specify whether a pharmacological treatment was being
administered during the study; (d) lack of reporting on
efficacy and efficiency measures; (e) poor graphic displays
where transfer of stimulus control is not visible, i.e., only
J Autism Dev Disord (2008) 38:261–275 271
123
the unprompted correct responses were shown on the
graph; and, (f) lack of detail in the description of treatment
variables.
Future Research
More studies which will evaluate the relative efficacy of
the CTD and PTD procedures compared to each other and
to other teaching methods, within a frame of direct
comparison studies, are needed. Also, there is a need for
more studies that assess the effectiveness of time delay
methods for individuals with various sub-classifications of
ASD. Replications of effective treatment packages and
isolation of intervention variables and more studies which
include preschoolers and adults with ASD will contribute
greatly to the empirical data in this field. Authors will
need to clarify the criteria used in the diagnosis of the
persons included in future studies. Specific data on the
type and dosage of the medication the participants receive
during the study will help researchers and practitioners
differentiate between the results of a medicated and a
drug-free sample of participants. More specific data on
superimposed, or co-occurring, conditions, e.g., dual
diagnosis, would be of help in identifying results of
instructional strategies on low-incidence populations.
More studies on generalization programming, studies that
include caregivers as instructors, and studies that use the
PTD intervention with chained behaviors are needed. A
comprehensive literature review on the effectiveness of
the PTD strategy with chained tasks will be informative
to both practitioners and researchers. Finally, if detailed
data in describing the variables of the intervention used
(e.g., if the number of sessions in which a delay of a
specific duration was used would be reported), then a
better correlation between the level of functioning and an
optimum length of delay interval would be established.
Table 4 Data comparison between the CTD and the PTD variables
Variable CTD PTD
Total no. of studies (Jan.
1985–April 2005 )
10 12
No. of participants with ASD
out of total number
16 out of 36 37 out of 40
Participants’ IQ functioning 1 Mi, 9 Mo, 6 Sev 3 Mi, 17 Mo, 2 Sev, 1 no impairment, 13
N/A
Type of target behavior Discrete—4 studies,
Chained—6 studies
Discrete—12 studies
Attending cues General—3 studies General—2 studies
Specific—3 studies
Mean no. of 0-s delay sessions ~ 3.4 ~ 2.5
Effectiveness 87.5% reached criterion 100% reached criterion
12.5% CTD not introduced to
Mean of transfer of stimulus
control
By 5.6th session across 5 studies By 4.0th session across 4 studies
Percentage of errors to
criterion
10.0% across 7 studies 2.7% across 3 studies
Procedural modifications Adaptation of task analysis and equipment, addition of physical
prompt, addition of 0-s delay trials, differential reinforcement,
addition of massed trials, and skipping the task
Addition of picture book as reinforcer, of
edible as reinforcer, and special
training
Behaviors were maintained YES YES
Design (all single-subject) 5 MP, 2 MB, 1 withdrawal, 1 parallel treatments, 1 AB 8 MB, 2 MP, 1 parallel treatments, 1
alternating treatments
Mean of point-by-point
interobserver agreement
across studies
97.9% 95.2%
Procedural reliability 97.4% (6 studies) 97.3% (5 studies)
Social validity collection In 50.0% of the studies In 41.7% of the studies
Comparison studies CTD more efficient than SLP PTD more efficient than SLP, LTM, and
visual fading
Legend: Mi = Mild; Mo = Moderate; Sev = Severe; MB = Multiple Baseline; MP = Multiple Probe; SLP = System of Least Prompts; LTM
= Least-to-Most Prompt; N/A = data Not Available
272 J Autism Dev Disord (2008) 38:261–275
123
Recommendations
Procedural recommendations for practitioners may include
the following. The teacher should choose a controlling
prompt with consideration to the abilities and limitations of
individuals diagnosed with ASD, who are known to have
difficulties with imitation, especially verbal imitation.
Also, consider the fact that verbal behaviors are difficult to
be prompted. Since the most common procedural modifi-
cation used in the studies was an addition of a reinforcer,
plan for differential reinforcement and fading early in the
intervention to prevent prompt dependency. The interven-
tionist should use social validation measures to select the
target behaviors to ensure that the dependent variable is
functional for the participant. When participants with lower
cognitive functioning are involved in the study, the teacher
should use attentional responses, since in the CTD studies,
which included lower functioning individuals that the PTD
studies, the teachers delivered more attentional cues. The
practitioner may find that an increase in the delay interval
at each step of the task analysis, as it is the case with the
PTD procedure, is cumbersome, since the PTD studies
were implemented with discrete behaviors only. However,
the gradual increase in the delay interval works as a fading
procedure, naturally embedded into the PTD strategy,
which leads to a decreased percentage of errors to criterion
and to a more accelerated transfer of the stimulus control.
The practitioners from these studies found single-subject
research design comfortable, so it may be suitable for fu-
ture studies, too. The teacher should be aware as much as
possible of the student’s preferences for certain stimuli
involved in the study, such as materials, day of time, trainer
(see for example Charlop et al., 1985). Whenever possible,
the teacher should implement a direct comparison of
instructional procedures, as it helps the practitioner to find
out what method of instruction the student is more
responsive to (e.g., more or less structured) and it will also
add to the body of research in the field.
Table 5 General conclusions
No. Conclusions
1. The majority of the articles (36%) were published in the J. of Applied Behavior Analysis
2. Of the 53 total participants diagnosed with ASD, 83% were male
3. The participants diagnosed with autism or autistic-like symptoms were in majority elementary aged (73.6%)
4. The majority of the participants (49.0%) were functioning mainly within a moderate range of cognitive abilities
5. Classroom personnel and university staff mainly conducted the studies
6. Instruction was mostly conducted in typical settings (classroom and community settings)
7. Instruction was mostly conducted in one-on-one instructional format
8. More discrete (72.8%) than chained skills were taught with these procedures
9. Both procedures had a mean of approximately 2 sessions of 0-s delay
10. CTD was implemented with 4-s and 5-s delay intervals, equally
11. PTD had mostly 1-s and 2-s increments in the delay interval, with the highest ceiling of the delay interval at 10-s
12. The behaviors taught were mostly functional behaviors
13. Both procedures used mostly verbal and gestural models as controlling prompts
14. As error consequence for unprompted errors, in the CTD studies the teachers mostly interrupted the response and delivered the controlling
prompt, while in PTD studies the consequences were time-out and deprivation of reinforcement
15. The most frequent type of reinforcer was the natural consequence of performing the target behavior and it was often associated with
descriptive verbal praise
16. In general, the instruction took place one to two times per day
17. Procedural modifications were described in 4 studies per each procedure, but the modifications that were required in the CTD studies were
of a greater magnitude;
18. CTD resulted in more errors to criterion (10.0%) than PTD (2.6%)
19. The transfer of stimulus control occurred more rapidly when using the PTD procedure (after 4 sessions) than when using the CTD procedure
(after 5.5 sessions)
20. All learned behaviors were maintained and generalized with both procedures
21. In terms of efficiency when compared to another instructional procedure, the CTD procedure was shown superior to the SLP procedure, and
the PTD procedure was shown superior to the SLP, LTM, and visual fading procedures
22. All studies used single-subject experimental designs, especially multiple baseline (45.5%) and multiple probes (32.0%)
23. Interrater agreement and procedural fidelity means were similar and high for both procedures
24. Social validity was collected for almost half of the studies
25. Both CTD and PTD procedures, including treatment packages, were shown to be effective in teaching individuals with ASD
J Autism Dev Disord (2008) 38:261–275 273
123
In summary, this review seeks to identify the best
instructional parameters within the time delay procedures,
and to reveal possible directions for future research, so that
special education practitioners can improve their teaching
methods by being aware of the results of these studies. The
CTD and PTD procedures were procedures have been
shown to be both effective and efficient for use with
students with ASD.
Acknowledgments The author gives special thanks to Dr. David L.
Gast, Dr. Karen Braxley, Dr. David R. Walker, and Dr. John Matson, as
well as the reviewers of the Journal of Autism and Developmental
Disorders for their insight and suggestions regarding this manuscript.
The results of this review were summarized and presented at the
Georgia Council for Exceptional Children Conference in January, 2006.
References
American Psychiatric Association. (2000). Autistic disorder. In
Diagnostic and statistical manual of mental disorders, (4th ed.,
text revision, (pp. 69–75)). Washington, DC: Author.
Ault, M. J., Gast, D. L., & Wolery, M. (1988). Comparison of
progressive and constant time delay procedures in teaching
community-sign word reading. American Journal of Mental
Retardation, 93, 44–56.
*Ault, M. J., Wolery, M., Gast, D. L., Doyle, P. M., & Eizenstat, V.
(1988). Comparison of response prompting procedures in
teaching numeral identification to autistic participants. Journal
of Autism and Developmental Disorders, 18, 627–636.
Bennett, D. L., Gast, D. L., Wolery, M., & Schuster, J. (1986). Time
delay and system of least prompts: A comparison in teaching
manual sign production. Education and Training of the Mentally
Retarded, 21, 117–129.
Bodfish, J. W., Symons, F. J., Parker, D. E., & Lewis, M. H. (2000).
Varieties of repetitive behavior in autism: Comparisons to
mental retardation. Journal of Autism and Developmental
Disorders, 30, 237–243.
Bristol, M. M., Cohen, D. J., Costello, E. J., Denkla, M., Eckberg, T.
J., Kallen, R., et al. (1996). State of the science in autism: Report
to the National Institutes of Health. Journal of Autism and
Developmetal Disorders, 26, 121–154.
Buschbacher, P. W., & Fox, L. (2003). Understanding and intervening
with challenging behavior of young children with Autism
Spectrum Disorder. Language, Speech, and Hearing Services
in Schools, 34, 217–227.
Campbell, J. M., Morgan, S. B., & Neighbors, J. W. (2004). Autism
spectrum disorders and mental retardation. In R. T. Brown (Ed.),
Handbook of pediatric psychology in school settings. Mahwah,
NJ: Lawrence Erlbaum.
*Charlop, M. H., Schreibman, L., & Thibodeau, M. G. (1985).
Increasing spontaneous verbal responding in autistic children
using a time delay procedure. Journal of Applied Behavior
Analysis, 18, 155–166.
*Charlop, M. H., & Trasowech, J. E. (1991). Increasing autistic
children’s daily spontaneous speech. Journal of Applied Behav-
ior Analysis, 24, 747–761.
*Charlop, M. H., & Walsh, M. E. (1986). Increasing autistic
children’s spontaneous verbalizations of affection: An assess-
ment of time delay and peer modeling procedures. Journal of
Applied Behavior Analysis, 19, 307–314.
Cohen, D. J., & Volkmar, F. R. (1997). Handbook of autism and
developmental disorders. New York: John Wiley & Sons, Inc.
Dawson, G., Toth, G., Abbott, R., Osterling, J., Munson, J., Estes, A.,
& Liaw, J. (2004). Early social attention impairments in autism:
Social orienting, joint attention, and attention to distress.
Developmental Psychology, 40, 271–183.
*Dipipi, C. M., Jitendra, A. K., & Miller, J. A. (2001). Reducing
repetitive speech: Effects of strategy instruction. Preventing
School Failure, 45, 177–181.
Gast, D. L., Wolery, M., Ault, M. J., Doyle, P. M., & Alig, C. (1988).
How to use time delay. Comparison of Instructional Strategies
Project. Unpublished manuscript, University of Kentucky.
*Gardill, M. C., & Browder, D. M. (1995). Teaching stimulus classes
to encourage independent purchasing by students with severe
behavior disorders. Education and Training in Mental Retarda-
tion and Developmental Disabilities, 254–264.
*Godby, S., Gast, D. L., & Wolery, M. (1987). A comparison of
time delay and system of least prompts in teaching object
identification. Research in Developmental Disabilities, 8, 283–
306.
Handen, B. L., & Zane, T. (1987). Delayed prompting: A review of
procedural variations and results. Research in Developmental
Disabilities, 8, 307–330.
*Heckaman, K. A., Alber, S., Hooper, S., & Heward, W. L. (1998). A
comparison of least-to-most prompts and progressive time delay
on the disruptive behavior of students with autism. Journal of
Behavioral Education, 8, 171–201.
*Hughes, M. V., Schuster, J. W., & Nelson, C. M. (1993). The
acquisition of independent dressing skills by students with
multiple disabilities. Journal of Developmental and Physical
Disabilities, 5
, 233–295.
*Ingenmey, R., & Houten, R. V. (1991). Using time delay to promote
spontaneous speech in an autistic child. Journal of Applied
Behavior Analysis, 24, 591–596.
Kabot, S., Masi, W., & Segal, M. (2003). Advances in the diagnosis
and treatment of autism spectrum disorders. Professional
Psychology: Research and Practice, 34, 26–33.
Kanner, L. (1943). Autistic disturbances of affective contact. Nervous
Child, 2, 217–250.
Keel, M. C., & Gast, D. L. (1992). Small-group instruction for
students with learning disabilities: Observational and incidental
teaching. Exceptional Children, 58(4), 1–12.
*Johnston, S., Nelson, C., Evans, J., & Palazolo, K. (2003). The use of
visual supports in teaching young children with Autism Spec-
trum Disorder to initiate interactions. Augmentative and Alter-
native Communication, 19, 86–103.
*Leung, J. P. (1994). Teaching spontaneous requests to children with
autism using a time delay procedure with multi-component toys.
Journal of Behavioral Education, 4, 21–31.
*Matson, J. L., Sevin, J. A., Box, M. L., Francis, K. L., & Sevin, B.
M. (1993). An evaluation of two methods for increasing self-
initiated verbalizations in autistic children. Journal of Applied
Behavior Analysis, 26, 389–398.
*Matson, J. L., Sevin, J. A, Fridley, D., & Love, S. R. (1990).
Increasing spontaneous language in three autistic children.
Journal of Applied Behavior Analysis, 23, 227–233.
*Morse, T. E., & Schuster, J. W. (2000). Teaching elementary
students with moderate intellectual disabilities how to shop for
groceries. Exceptional Children, 66, 273–288.
Morse, T. E., & Schuster, J. W. (2004). Simultaneous prompting: A
review of the literature. Education and Training in Develop-
mental Disabilities, 39(2), 153–168.
National Research Council. (2001). Educating children with autism.
Washington, DC: National Academy Press.
*Nientimp, E. G., & Cole, C. L. (1992). Teaching socially valid social
interaction responses to students with severe disabilities in an
integrated school setting. Journal of School Psychology, 30,
343–354.
274 J Autism Dev Disord (2008) 38:261–275
123
*Norman, J. M., Collins, B. C., & Schuster, J. W. (2001). Using an
instructional package including video technology to teach self-
help skills to elementary students with mental disabilities.
Journal of Special Education Technology, 16(3), 5–18.
Ozonoff, S. (1998). Assessment and remediation of executive
dysfunction in autism and Asperger Syndrome. In E. Schopler,
G. B. Mesibov, & L. Kunce (Eds.), Asperger Syndrome and high
functioning autism. Plenum Press.
Ritvo, E. R., & Freeman, B. J. (1977). National society for autistic
children definition of the syndrome of autism. Journal of
Pediatric Psychology, 2, 146–148.
Schuster, J. W., Gast, D. L., Wolery, M., & Guiltinan, S. (1988). The
effectiveness of a constant time-delay procedure to teach chained
responses to adolescents with mental retardation. Journal of
Applied Behavior Analysis, 21, 169–178.
Schuster, J. W., Morse, T. E., Ault, M. J., Doyle, P. M., Crawford, M.
R., & Wolery, M. (1998). Constant time delay with chained
tasks: A review of the literature. Education and Treatment of
Children, 21, 74–107.
Tawney, J. W., & Gast, D. L. (1984). Single participant research in
special education. Columbus, OH: Charles Merill.
*Taylor, B. A., & Harris, S. L. (1995). Teaching children with autism
to seek information: Acquisition of novel information and
generalization of responding. Journal of Applied Behavior
Analysis, 28, 3–14.
Touchette, P. E. (1971). Transfer of stimulus control: Measuring the
moment of transfer. Journal of Experimental Analysis of
Behavior, 15, 347–354.
*Venn, M. L., Wolery, M., Werts, M. G., Morris, L. D., DeCesare, L.
D., & Cuffs, M. S. (1993). Embedding instruction in art activities
to teach preschoolers with disabilities to imitate their peers.
Early Childhood Research Quarterly, 8, 277–294.
*Wall, M. E. & Gast, D. L. (1997). Caregivers’ use of constant time
delay to teach leisure skills to adolescents or young adults with
moderate or severe intellectual disabilities. Education and
Training in Mental Retardation and Developmental Disabilities,
340–356.
*Wall, M. E., Gast, D. L., & Royston, P. A. (1999). Leisure skills
instruction for adolescents with severe and profound develop-
mental disabilities. Journal of Developmental and Physical
Disabilities, 11, 193–219.
Wing, L., & Potter, D. (2002). The epidemiology of autism spectrum
disorders: Is the prevalence rising? Mental Retardation and
Developmental Disabilities, 8, 151–161.
*Winterling, V., Gast, D. L., Wolery, M., & Farmer, J. A. (1992).
Teaching safety skills to high school students with moderate
disabilities. Journal of Applied Behavior Analysis, 25, 217–227.
Wolery, M., Ault, M., & Doyle, P. M. (1992). Teaching students with
moderate to severe disabilities: Use of response prompting
strategies. White Plains, NY: Longman Publishing Group.
*Wolery, M., Gast, D. L., Kirk, K., & Schuster, J. (1988). Fading
extra-stimulus prompts with autistic children using time delay.
Education and Treatment of Children, 11, 29–44.
Wolery, M., Holcombe, A., Cybriwsky, C., Doyle, P. M., Schuster, J.
W., Ault, M. J., et al. (1992). Constant time delay with discrete
responses: A review of effectiveness and demographic, proce-
dural, and methodological parameters. Research Developmental
Disabilities, 13, 239–266.
Wolery, M, & Schuster, J. W. (1997). Instructional methods with
students who have significant disabilities. Journal of Special
Education, 31, 61–79.
World Health Organization (WHO). (1993). The ICD-10 classifica-
tion of mental and behavioural disorders: Diagnostic criteria for
research. Geneva: Author.
*References marked with an asterisk indicate studies included in
the analysis.
J Autism Dev Disord (2008) 38:261–275 275
123
