Focus on Autism and Other

The Role of the Reflexive-Conditioned Developmental Disabilities

25(2) 110­–124
© 2010 Hammill Institute on Disabilities
Motivating Operation (CMO-R) During Reprints and permission:
[Link]/[Link]

Discrete Trial Instruction of Children DOI: 10.1177/1088357610364393

[Link]

With Autism

Vincent J. Carbone,1 Barry Morgenstern,2 Gina Zecchin-Tirri,3

and Laura Kolberg3

Abstract
The principle of motivation has resurfaced as an independent variable in the field of behavior analysis over the past
20 [Link] increased interest is the result of refinements of the concept of the motivating operation and its application to
the learning needs of persons with developmental disabilities. Notwithstanding the increased emphasis upon modification
of motivating operations to reduce problem behavior, there is limited recognition of this important behavioral variable
in autism treatment literature. An overview of antecedent-based instructional modifications that lead to a reduction of
escape and avoidance behavior of children with autism during instruction is provided. An analysis of these instructional
methods as motivating operations is proposed. A conceptually systematic analysis of the influence of instructional methods
is offered as a tool for improving the selection and implementation of effective teaching procedures.

Keywords
motivating operations, establishing operations, autism, escape and avoidance behavior, discrete trial instruction

Comprehensive intensive treatment based upon the Green, & Foxx, 2001; Maurice, Green, & Luce, 1996) for
application of behavior analytic principles has proven to be parents and practitioners are available to provide a sum-
an effective form of intervention for children with autism mary of the effective teaching methods discovered through
(Green, 1996). Researchers have demonstrated the superi- controlled studies. These manuals provide a user-friendly
ority of behavior analytic programs over other approaches method of disseminating effective behavior analytic meth-
to autism treatment or differing levels of intensities of ods for teaching children with autism. The result may be
services (Birnbrauer & Leach, 1993; Cohen, Amerine- greater acceptance and widespread application of behavior
Dickens, & Smith, 2006; Eikseth, Smith, Jahr, & Eldevik, analytic methods with children with autism.
2007; Lovaas, 1987; Sallows & Graupner, 2005; Remington Much of the research and all of the manualized treat-
et al., 2007; T. Smith, Groen, & Wynne, 2000). These ment packages emphasize the importance of motivating
researchers have provided clear evidence that intensive children to respond to teacher-directed instructional tasks.
intervention guided by behavior analytic principles can pro- R. L. Koegel, Carter, and Koegel (1998) and L. K. Koegel,
duce substantial benefits for children with a disorder that Koegel, Shoshan, and McNerney (1999) suggested that
was once thought to be resistant to all forms of treatment. motivation is pivotal to teaching of children with autism
There are reports of children with autism entering regular because its creation is critical to the development of a wide
education classrooms, achieving substantial cognitive gains, range of skills. Moreover, given the tendency of these
and developing age-appropriate social skills after many
years of intensive behavioral intervention (Lovaas, 1987). 1
Carbone Clinic, Valley Cottage, NY
Recently, evidence has been gathered that suggests school, 2
Institute of Professional Practice, Inc., Woodbridge, CT
community, and home applications of intensive behavioral 3
Carbone Clinic, Valley Cottage, NY
intervention can be equally successful (Eikseth, Smith,
Corresponding Author:
Jahr, & Eldevik, 2002; Howard, Sparkman, Cohen, Vincent J. Carbone, The Carbone Clinic, 614 Corporate Way, Suite # 1,
Green, & Stanislaw, 2005). At least five published manuals Valley Cottage, NY 10989, USA
(Leaf & McEachin, 1999; Lovaas, 1981, 2003; Maurice, Email: drvjc@[Link]
Carbone et al. 111

children to engage in high rates of escape and avoidance Risley (1968, p. 96) stated that practitioners within a scien-
behaviors (R. L. Koegel, Koegel, Frea, & Smith, 1995) tific discipline require more than a “collection of tricks” as
within instructional demand settings, methods that increase the source of their procedures. Extension to new areas is
the motivation to respond may be essential for producing accomplished only through the understanding of how pro-
long-term positive outcomes. The ultimate outcome for cedures work in terms of basic principles. In the case of
many children with autism may depend at least partially discrete trial instruction of children with autism, practitio-
upon their learning to attend to teacher-directed activities ners may benefit from a conceptually systematic analysis of
and respond correctly and quickly for reasonable periods motivation when conducting training, applying the princi-
of time each day (Drash & Tudor, 1993). This is especially ples to new problems, generally reducing the aversiveness
important for children with autism because they frequently of teaching environments, and decreasing reliance on escape
fail to learn through exposure to typical social environ- extinction. Moreover, improved selection of appropriate
ments (T. Smith, 2001). As an alternative to mere exposure instructional methods may be facilitated.
to everyday experiences, the method of discrete trial
instruction (Lovaas, 1981, 1987; T. Smith, 2001) has been
demonstrated to be one of the most effective instructional The Establishing Operation
tools for teaching important language, social, and cogni- Michael (1993, p. 192) stated the establishing operation (EO)
tive skills to children with autism as a component of a
comprehensive program of intervention. The method is is an environmental event, operation, or stimulus con-
modeled after Skinner’s (1968) three-term contingency dition that affects an organism by momentarily altering
arrangement whereby a stimulus is presented by a teacher, (a) the reinforcing effectiveness of other events and
a response occurs, and a consequence follows the response (b) the frequency of occurrence of that part of the
to strengthen or weaken the likelihood that it will occur organism’s repertoire relevant to those events as
again under similar conditions. consequences.
When discrete trial instruction has been used as a com-
ponent of a comprehensive program of intensive intervention To paraphrase Michael (2004), EOs make someone
for children with autism, long-term benefits have been “want something” and lead to the actions that have produced
achieved with many children (Lovaas, 1987; McEachin, what is now “wanted.” Food deprivation makes you “want”
Smith, & Lovaas, 1993; T. Smith, 1999). Notwithstanding food and therefore leads to actions that have produced
the benefits of this method, its proper implementation food ingestion in the past, such as making a sandwich. A
presents substantial challenges to practitioners. The imple- headache makes you “want” pain relief and therefore leads
mentation of discrete trial instruction may conflict with the to actions that reduce pain, such as swallowing an aspirin.
learning history of children with autism related to escape A significant portion of tantrums and generally disruptive
and avoidance behavior. In other words, the high demand behavior in children with autism during instruction may
requirements of discrete trial instruction are the same con- result from strong motivation for something (EO), such as
ditions that typically evoke problem behavior in the form of task removal, a toy, or attention.
tantrumming, flopping, high rates of stereotypies, aggres- The term EO has been considered awkward because it
sion, and self-injury. Smith (T. 2001, p. 89) explains: “. . . implies only an increase in reinforcing or punishing
children with autism may attempt to escape or avoid almost effectiveness. Therefore, Laraway, Syncerski, Michael, and
all teaching situations, as well as any requests that adults Poling (2003) recommended replacing the term with MO.
make of them.” Consequently, a thorough conceptual under- Within the remainder of this article, MO will be used rather
standing and practical repertoire related to the modification than EO.
of instructional variables that reduce escape-maintained Michael (1993, 2004, 2007) provided descriptions of
and avoidance-maintained problem behavior of children several unconditioned and conditioned MOs. A full descrip-
with autism appears essential. tion of each is beyond the scope of this article. Instead, an
This article was generated to provide an overview of analysis of problem behavior during discrete trial instruc-
the behavioral analysis of motivation during discrete trial tion will be provided, utilizing the relevant concept of the
instruction and a re-interpretation of the effects of ante- CMO-R. Methods will be suggested that appear to abolish
cedent variables as motivating operations (MO), and, more the CMO-R, leading to reductions in problem behavior
specifically, the reflexive-conditioned motivating operation within the context of demand-related instructional activities
(CMO-R). No new methods are presented. Instead, this with persons with developmental disabilities and autism.
interpretation is offered to help practitioners and teachers Despite the fact that several researchers have demonstrated
understand why varieties of procedures that have been a reduction in escape-motivated behavior without acknowl-
reported in the literature are effective. Baer, Wolf, and edging the role of the CMO-R, an increasing number of
112 Focus on Autism and Other Developmental Disabilities 25(2)

studies (Iwata, Smith, & Michael, 2000) implicating this children often throw tantrums when demands are placed on
important motivational variable seems to suggest a previ- them. Their tantrums may interfere seriously with their
ously unrecognized role. The CMO-R has been implicated learning of more appropriate behaviors.” Other researchers
directly in several studies as an independent variable that also have documented the negative role that escape and
affects the occurrence of problem behavior (Crockett & avoidance behavior plays in the teaching and acquisition of
Hagopian, 2006; DeLeon, Neidert, Anders, & Rodriguez- important skills of children with autism. R. L. Koegel et al.
Catter, 2001; Ebanks & Fisher, 2003; Lalli et al., 1999; (1998, pp. 167–168) claim that:
McComas, Hoch, Paone, & El-Roy, 2000). The presentation
of instructional demands in all these studies implicated the It is well documented that children with autism fail to
CMO-R as the potential mechanism that accounted for the respond to and avoid many types of language and aca-
reported behavioral effects. Michael (1993, p. 203) defined demic interactions . . . failure to respond to everyday
the CMO-R as: environmental stimuli, which appears as a widespread
motivation problem, may not only have an impact on
Any stimulus condition whose presence or absence a child’s communicative and scholastic activities but
has been positively correlated with the presence or also can be profoundly detrimental to a child’s social
absence of any form of worsening will function as a development.
CMO in establishing its own termination as effective
reinforcement and in evoking any behavior that has Sundberg (1993) suggested that the teaching of language
been so reinforced. and other skills is often complicated when instructional
stimuli act as a CMO-R. This conclusion is particularly
The CMO-R is an environmental event that ultimately problematic because one of the most frequently imple-
increases the value of conditioned negative reinforcement mented behavior analytic methods, discrete trial instruc-
and therefore evokes any behavior that has led to a tion, includes the presentation of frequent teacher-initiated
reduction in the current aversive condition. In the case of academic demands. T. Smith (2001, p. 86) suggests “As a
the CMO-R specifically, the conditioned aversive stimulus result, these children are likely to experience frustration in
is the onset of the very stimulus whose offset would teaching situations. . .. They may react to such frustrations
function as a form of conditioned reinforcement. For with tantrums and other efforts to escape or avoid future
example, when teaching children with autism, the mere failures.” Smith suggests that providers of these services
delivery of an instructional demand may establish its must be equipped with the skills necessary to reduce these
removal as a reinforcer. Therefore, the offset of the stimulus problem behaviors during teaching sessions. Some inves-
will act as a reinforcer for any response that removes the tigators have concluded that the best outcome for children
instructional demand. In other words, if instructional with autism may be related to the teacher’s or parent’s skill
demands and the setting in which they are presented in reducing disruptive behavior and developing learner
“signals” or warns of any type of worsening situation cooperation during instruction (Lovaas, 2003). Given the
(reduced reinforcement, difficult instructional demands, fact that there is evidence that instructional and other types
many instructional demands, high rate of errors, etc.), of demands delivered to children with autism during teach-
responses that remove the warning signal will be evoked. ing sessions (and at other times) might well function as
Within this context, instructional demands act as aversive CMO-Rs for some children (R. Smith & Iwata, 1997), a
stimuli and therefore evoke problem behavior that has led comprehensive understanding of how this independent
to the removal of the demands in the past. variable affects learning, and information on how to
weaken its control over problem behavior appears essen-
tial for teachers and others who guide programs for children
The CMO-R and Teaching Children With Autism with autism.
Responding maintained by escape and avoidance of instruc- To facilitate an understanding of CMO-R, an example
tional and other types of demands accounts for between 33% from the laboratory setting is offered. Figure 1 illustrates the
and 48% of self-injurious and aggressive behaviors of per- development of the CMO-R and the development of the
sons with developmental disabilities (Derby et al., 1992; escape and avoidance behavior it evokes in a laboratory
Iwata et al., 1994). The behavior analytic research literature environment. The operant experimental preparation that has
is replete with interventions for escape-motivated behavior yielded high rates of escape and avoidance behavior is referred
including but not limited to functional communication train- to as the discriminated avoidance paradigm (Hoffman, 1966).
ing (FCT) plus extinction (Hanley, Iwata, & McCord, 2003) In a laboratory example, rats subjected to painful shock that
and noncontingent escape (J. E. Carr & LeBlanc, 2006). was preceded by and positively correlated with the sound of a
Lovaas (1981, p. 29) suggested “Developmentally disabled neutral tone learned to terminate the tone and avoid the shock
Carbone et al. 113

In general, it is recommended that many children with

Development of the CMO-R in the Laboratory
autism receive as much as 25 to 40 hrs per week of intensive
Neutral “Painful Stimulation” Effects
Stimulus behavioral intervention (Green, 1996; Leaf & McEachin,
(Tone) Time
Termination of
Worsening
1999; National Research Council, 2001). An important com-
(Shock)
Presentation of Worsening Set Condition ponent of the intensive treatment model is the use of discrete
Stimulus Object of Conditions is a Reinforcer &
or Event Evokes Behavior trial instruction. Within this approach, behavioral tasks are
that has been so
Reinforced divided into component activities. While the instructor is
sitting at a child-sized table, he or she usually presents
After repeated correlations of the events in the above sequence……..
an instructional demand, waits for or prompts the correct
response, provides a consequence for the child’s response,
Warning Stimulus Effects
(CMO-R) and then pauses for a few seconds before presenting the next
Establishes
(Tone) Termination of instructional demand (Anderson, Taras, & O’Malley-
Warning Stimulus
Presentation of
Stimulus Object (tone) as a Cannon, 1996). The daily activities may alternate between
Reinforcer and
or Event
Evokes Behavior structured and unstructured, with opportunities for inciden-
that has led to its
Termination
tal teaching (Leaf & McEachin, 1999). Many programs
combine discrete trial instruction sessions with natural envi-
Figure 1. Illustrative diagram of the development of the ronment teaching (Sundberg & Partington, 1998). Whatever
reflexive-conditioned motivating operation (CMO-R) in the format is chosen, all behavioral treatment programs for chil-
laboratory. dren with autism emphasize active learner responding to
high rates of teacher-presented instructional demands with
the degree of learner cooperation affecting the benefit
by pressing a metal bar. In this experiment, after repeated achieved.
exposures to the tone–shock pairings, the mere presentation As Figure 2 suggests, the presence of the teacher, display
of the tone established its removal as a reinforcer and evoked of the materials, and requests to move to the instructional
behavior that in the past had resulted in its termination, such environment all may have been correlated with later stages
as bar pressing. Notice how the tone presentation met the of the instructional setting when the “worsening set of con-
two-part definition of the MO in terms of value-altering ditions” became increasingly potent. All of the instructional
and behavior-altering effects. Also note the termination of the activities listed in the worsening conditions column in
tone acted as a conditioned reinforcer for the bar pressing. Figure 2 have been identified in the behavioral literature as
Within the behavioral literature, the onset of a stimulus potentially aversive conditions that may occur during the
like the tone has been identified as a discriminative stimulus instruction of children with autism (Langthorne, McGill, &
(SD) for the behavior of bar pressing. Michael’s (1982, 2007) O’Reilly, 2007; McGill, 1999; R. Smith & Iwata, 1997;
reinterpretation of the difference between discriminative Wilder & Carr, 1998). In this way, the activities at the
stimuli and MOs leads to the conclusion that the tone onset beginning of the session serve as a warning signal of move-
acts as a CMO-R. In addition, the reinforcer for the bar press ment toward the later stages of the instructional session and
has typically been identified as avoidance of the shock, not therefore establish removal of any and all signs of instruc-
the termination of the tone. Michael (2004, p. 71) suggested tion as a reinforcer and evoke problem behavior, such as
from a molecular perspective this does not seem reasonable aggression, self-injury, and tantrumming that have histori-
since, “Something not happening does not easily qualify as cally produced task removal (Michael, 2000). In this case,
the kind of event that can function as an immediate response the teacher, the materials, the teacher’s voice, and the actual
consequence.” Michael’s (1982, 1988, 1993, 2000, 2004, demands all may begin to function as a CMO-R because of
2007) refinements of the concept of the CMO has added their correlation with instructional activities that represent a
greatly to our understanding of this behavioral variable. worsening set of conditions.
Failure to properly identify these events in terms of their The worsening set of conditions in the instructional
functional relations to behavior may lead to imprecise and example is only metaphorically referred to as “painful stim-
ineffective control of behavior in the laboratory, and worse, ulation.” Conditions or stimuli that warn of a decrease in the
to poorly designed and implemented treatment programs for rate of reinforcement, decrease in the amount of rein-
children with autism in classrooms and other settings. forcement, less immediate reinforcement, greater response
Now consider the same arrangement as it relates to requirement, greater response effort, and so forth are all
the instruction of children with autism within a discrete trial worsening conditions that can act as reinforcers for behavior
instruction format. Figure 2 illustrates the same arrangement that terminates them (Michael, 2004). Failure to recognize
of behavior analytic variables described in the laboratory the contribution of the CMO-R to the development of escape
example provided in Figure 1. and avoidance behavior during the instruction of children
114 Focus on Autism and Other Developmental Disabilities 25(2)

Development of the CMO-R in the Classroom

Neutral “Painful Stimulation” Effects

Stimulus Worsening Set of Conditions

Presentation of • Session Begins with Removal Termination of

Instructional Time of Positive Reinforcement Worsening

=
Demands, • Low value Positive Reinforcement Condition
Instructional • Low rate of Positive Reinforcement is a Reinforcer &
Materials and • Frequent Social Disapproval Evokes Behavior
Presence of • Effortful Responses Required that has been so
Teacher • Difficult Responses Required Reinforced
• High Rate of Demands
• Frequent Learner Errors
• Delayed Positive Reinforcement
• Low magnitude Positive
Reinforcement

After repeated correlations of the events in the above sequence……..

Warning Stimulus Effects

(CMO-R)
Presentation of
Instructional Establishes Termination of
Demands, the Warning Stimuli as a

=
Instructional Reinforcer
Materials and AND
Presence of Evokes all Responses that
Teacher have led to their Removal

Figure 2. Illustrative diagram of the development of the reflexive-conditioned motivating operation (CMO-R) in the classroom.

with autism may reduce the likelihood that the instruc- this distinction, behavior analysts typically have been
tional methods necessary to weaken its effects will be trained to classify all antecedent evocative stimuli as
implemented. discriminative stimuli (Schlinger, 1993). This set of cir-
cumstances “. . . leaves a gap in our understanding of
operant functional relations” (Michael, 1993, p. 191).
Differentiating SDs from MOs Moreover, Michael (1996) suggests that being able to talk
An issue central to this topic is the difference between the about these different variables is essential to being able to
SD and the MO. The fact is these two antecedent stimuli analyze them effectively during instructional sessions.
share several structural and functional characteristics Therefore, when analyzing the evocative effects of demands
including the fact that both are antecedent variables, are on problem behavior with children with autism, reliance on
learned, and evoke and abate behavior. SD control is identi- the concept of the MO may lead to practice that is more
fied frequently as the source of behavior change that is effective.
more properly ascribed to the effects of CMO-R. “Whereas Because instructional demands do not “signal” the
the discriminative stimulus derives control over responding availability of reinforcement for problem behavior but
through a special historical relationship with behavioral instead make negative reinforcement in the form of task
consequences, Skinner’s account of other antecedents sug- removal valuable, they are best identified as an MO. This
gests a different source of influence between some is the critical property that differentiates an SD from a
antecedent stimuli and behavior” (R. Smith & Iwata, 1997, CMO-R. “In short, EOs change how much people want
p. 346). In this quote, Smith and Iwata are referring to the something; SDs change their chances of getting it” (McGill,
MO as the “different source of influence.” Notwithstanding 1999, p. 395).
Carbone et al. 115

Differentiating the CMO-R From Other MOs removal as reinforcers. As pointed out by R. Smith and
Iwata (1997), however, few of the earlier researchers relied
Different MOs acquire their control over behavior through on the concept of the MO. Instead, they attributed the results
different mechanisms and histories. Unconditioned MOs to the structural variables of setting events and contextual
have unique histories related to the species phylogeny. variables or improperly to the effects of stimulus control.
Conditioned MOs have unique histories related to an indi- Recognition of the role of the MO has been obscured by the
vidual’s ontogeny. In other words, the histories that have fact that a conceptually systematic approach that focuses on
led to the development of the many unconditioned and the functional relations among environmental stimuli and
conditioned MOs are remarkably different. Moreover, the behavior has not been the general practice in the field. “In
mechanisms that account for their effects are all different. fact, a criticism of applied behavior analysis is a perceived
Consequently, practitioner efforts to abolish the effects and failure to relate the many procedures generated for changing
abate behavior related to any of the unconditioned or socially significant behavior to basic behavioral principles”
conditioned MOs would require substantially different (R. Smith & Iwata, 1997, p. 343).
environmental manipulations specific to each type of moti- Michael (2000, 2007) provided a conceptual analysis of
vating operation. As a result, Michael (1993, 2007) provided the modification of the CMO-R as a guide to practitioners
specific labels for each MO as a way of acknowledging the serving persons with autism and developmental disabilities.
different histories that have led to their control over behavior. He adopted the notion of increasing the effectiveness of
Moreover, he identified different forms of unpairing that instruction as a unifying concept under which motivational
can be used to decrease behavior evoked by conditioned antecedent variables, previously identified as setting events
MOs. Practitioners who are aware of these differences will or contextual variables, could be classified as motivating
certainly be more effective in controlling behavior than operations. Within his analysis, Michael rejected the idea of
those who are unaware. merely removing the CMO-R (e.g., instructional demands)
The CMO-R is the only MO that is engendered with to reduce problem behavior because presentation of fre-
evocative control over behavior through a history of cor- quent instructional demands is a necessary condition for
relation with a worsening setting of conditions. As a result learning within discrete trial instruction methodology.
of this unique history, the mere presentation of this type of Additionally, he agreed that the function-altering effects of
stimulus event immediately establishes its removal as a extinction could reduce problem behavior but would leave
form of reinforcement. Methods to reduce the effects of the the CMO-R in place and therefore would be a practical
CMO-R are procedurally distinct from unconditioned as solution only if there could be no reduction in the aversive
well as other conditioned MOs (i.e., surrogate, transitive). nature of the demands as CMO-Rs. He concluded that in
Michael (2000, p. 402) highlighted the importance of this most cases the CMO-R could be abolished by altering the
distinction by claiming “. . . to say that thinking of two instructional practices so that “instruction results in less
evocative variables with such different histories and impli- failure, more frequent social and other forms of rein-
cations for prediction and control as though they were forcement, and other general improvements in the demand
the same would surely result in theoretical and practical situation to the point at which it may not function as
ineffectiveness.” a demand but rather as an opportunity” (Michael, 2000,
p. 409). Michael identifies a heretofore largely overlooked
independent variable (or class of motivational variables)
Re-Interpreting Existing Treatments that needs to be considered during discrete trial instruction
From a CMO-R Perspective of children with autism.
Iwata et al. (2000) suggested researchers have demonstrated McGill (1999) provided additional support for Michael’s
the value of modifying MOs to increase or decrease problem recommendation related to instructional modification. He
behavior. The authors of all three major reviews of the topic stated that merely reducing the problem behavior while
(McGill, 1999; R. Smith & Iwata, 1997; Wilder & Carr, leaving the aversive nature of the demand situation unre-
1998) devoted sections of their articles to the modification solved is an unsatisfactory solution. He suggested that not
of MOs as independent variables. They all subdivided this only are practitioners obligated to reduce problem behavior
section into the MO modifications that were effective in but also to alter the challenging environment encountered by
reducing problem behavior maintained by positive, nega- most persons with autism and developmental disabilities.
tive, and automatic reinforcement. The modification of McGill (p. 406) agrees with Durand (1990) that problem
antecedent motivation variables to reduce problem behavior behaviors are at least partially the result of poorly arranged
maintained by negative reinforcement was analyzed in terms environments and that the CMO-R “. . . is a reflection of
of the CMO-R. All authors cited studies in which investiga- aberrant environmental characteristics (such as inappropri-
tors implemented procedures to reduce the value of task ate demands).” McGill (p. 406) goes on to say that failure to
116 Focus on Autism and Other Developmental Disabilities 25(2)

manipulate the CMO-R may raise ethical concerns “. . . behavioral analysis may have important practical implica-
because it leaves a counterhabilitative environment in place tions for persons who instruct children with autism.
and may be limited in its effectiveness because the circum- Many behavior analytic practitioners have made use of
stances evoking problem behavior still exist.” Moreover, he the evidenced-based procedures described in the following
states that FCT without extinction, punishment, and/or use section. No new procedures are offered. What follows is a
of antecedent modifications generally is ineffective in reduc- discussion of some of the evidenced-based instructional
ing behavior maintained by negative reinforcement. This practices that have been demonstrated to reduce problem
contention is supported empirically by Fisher et al. (1993) behavior during instruction along with a re-interpretation of
and Hagopian, Fisher, Sullivan, Acquisto, and LeBlanc the effects and benefits of these methods in terms of altering
(1998). Finally, McGill concluded that merely teaching a the function of CMO-Rs.
functionally equivalent response may not be sufficient to
reduce problem behavior without modification of the value
of the reinforcer that has led to the acquisition and mainte- Methods to Reduce the Effects
nance of the response. of the CMO-R During Discrete
Trial Instruction
Treatments Designed Programming Competing Reinforcers
to Abolish the CMO-R Researchers of several studies with persons with disabili-
Many effective antecedent modifications to reduce prob- ties demonstrated that problem behavior evoked by a
lem behavior have been demonstrated, often under the CMO-R and reinforced through termination of the demand
heading of curricular revisions (G. Dunlap, Foster-Johnson, situation can be reduced without controlling the negative
Clarke, Kern, & Childs, 1995; G. Dunlap & Kern, 1993, reinforcing consequence that has maintained the behavior
1996; G. Dunlap et al., 1993; G. Dunlap, Kern-Dunlap, (Call, Wacker, Ringdahl, Cooper-Brown, & Boeltric, 2004;
Clarke, & Robbins, 1991; Kern, Childs, Dunlap, Clarke, & Lalli & Casey, 1996; Parrish, Cataldo, Kolko, Neef, &
Falk, 1994; Kern & Dunlap, 1998) or antecedent interven- Egel, 1986; Piazza et al., 1997; Russo, Cataldo, & Cushing,
tions (Miltenberger, 2006). Many of these researchers 1981). In other words, behavior maintained by negative
have tested the effectiveness of treatment packages. Typi- reinforcement can be weakened by programming positive
cally, variables related to choice of task, task variation, reinforcement for an alternative compliant response or by
pace of instruction, interspersal of high-probability tasks, delivering it noncontingently during high demand situa-
partial-task versus whole-task instruction, task difficulty, tions. This can be accomplished without eliminating the
reducing learner errors, and so forth, have been included response–reinforcer relation in some cases (Lalli et al.,
in the treatment packages to reduce escape-motivated 1999). The effects of positive and negative reinforcement
problem behavior (Munk & Repp, 1994). Although these were studied in a series of investigations with participants
reports provided useful descriptions of behavior change whose problem behavior had been acquired and main-
methods, they failed to analyze them in terms of basic tained through task removal (Lalli & Casey, 1996; Lalli
behavioral principles. Failure to provide a behavioral et al., 1999; Piazza et al., 1997). By programming concur-
analysis of the effects of antecedent manipulations leaves rent schedules of reinforcement in which compliance with
the practitioner without the information necessary to ana- task demands was positively reinforced (e.g., with food,
lyze complex and novel cases. Notwithstanding this issue, praise) and problem behavior resulted in task termination,
many of the antecedent behavior reduction procedures the competing effects of positive and negative reinforce-
recommended to reduce escape-motivated behavior can ment could be assessed.
be re-interpreted in terms of modification of the CMO-R. These researchers demonstrated that introduction of
Such an analysis suggests that the antecedent variables positive reinforcement for responses that were alternatives
identified in the curricular revision literature acted as to the negatively reinforced problem behavior reduced the
abolishing operations to the extent that they decrease the problem behavior without modification of the maintaining
value of the reinforcer that is maintaining the problem contingency, and in some cases without the use of extinc-
behavior and therefore abated the responses that they pre- tion for problem behavior. In the Lalli et al. (1999) study,
viously controlled. A re-interpretation of the curricular the results were achieved when the programmed schedule
revision research findings will reduce their explanatory of reinforcement actually favored responses that produced
mechanisms to a handful of behavioral principles and pro- task removal (i.e., negative reinforcement). The authors
vide a conceptually systematic approach to the treatment concluded that the presentation of the positive reinforcer
of escape-motivated problem behaviors of children with abolished the CMO-R or value of task removal as a rein-
autism during discrete trial instruction. This type of forcer and abated the class of responses that had produced
Carbone et al. 117

that reinforcer in the past. In a follow-up study by DeLeon Embedding reinforcing activities in a context of instruc-
et al. (2001), the competing effects of positive and negative tional demands has been shown to reduce behavior evoked
reinforcement on problem behavior maintained by task by instructional demands. Studies by E. G. Carr and Carlson
removal were investigated with a chained schedule. A child (1993) and Kemp and Carr (1995) demonstrated that
with autism was provided the opportunity to choose a posi- demand-related problem behavior during community activ-
tive reinforcer (i.e., potato chip) or negative reinforcer ities and in employment settings could be reduced by
(i.e., break) after completing a scheduled number of embedding reinforcing activities. E. G. Carr, Newsome, and
responses. When the number of demands was relatively Binkoff (1980) found that activities such as storytelling
low, the participant reliably chose the positive reinforcer. reduced escape-motivated responses and increased compli-
It appeared that the presence of the positive reinforcer ance with demands. Kennedy, Itkonen, and Lindquist (1995)
decreased the value of task termination as a reinforcer. demonstrated that merely embedding social comments prior
However, the participant’s preference switched to the to low probability demands decreased noncompliance in
break when the number of tasks required for reinforcement students with severe disabilities.
increased to more than 10. The authors concluded that the
switch to the preference for a break when demands were
increased indicated the demands had returned to their ini- Errorless Instruction
tial status as a CMO-R and therefore the value of task Several researchers have demonstrated that when students
removal increased and evoked the participant’s choice make frequent errors during instructional sessions, levels
behavior of a break. of problem behavior are high (E.G. Carr & Durand, 1985;
As demonstrated by Kennedy (1994) and then again Ebanks & Fisher, 2003; Heckaman, Alber, Hooper, &
by Call et al. (2004), the addition of a positive reinforcer Heward, 1998; Weeks & Gaylord-Ross, 1981). Instructional
delivered during instruction reduced the escape-motivated methods that reduce the frequency of errors have been
noncompliant behavior of some participants. Call et al. demonstrated to reduce the level of problem behavior.
(2004, p. 155) concluded “. . . the addition of an arbitrary An analysis of these results in terms of motivational vari-
positive reinforcer can sometimes be sufficient to reduce ables suggests that errors may function as an MO and
problem behavior that is maintained, partially or solely, by increase the reinforcing value of task removal or termination.
negative reinforcement.” These authors and others sug- If the instructor prevents or at least minimizes errors during
gested that this effect is the result of lessening the aversive instruction (i.e., errorless learning), the CMO-R is abolished
context of the instructional setting by the delivery of a and students engage in fewer problem behaviors. For
competing positive reinforcer. These results appear consis- example, Heckaman et al. (1998) demonstrated that when
tent with Michael’s (2000) analysis of how the function of instructors used response prompts with a progressive time
demands may be altered from an aversive stimulus to an delay and students made very few errors, levels of disruptive
opportunity for the delivery of reinforcement. behavior were dramatically reduced. In comparison, when a
least-to-most prompting strategy was used the student made
many more errors and had higher levels of disruptive
Pairing and Embedding the Instructional behavior.
Environment With Positive Reinforcement In a similar manner, Ebanks and Fisher (2003) reduced
McGill (1999) recommends several methods for weakening escape-motivated destructive behavior by providing ante-
the value of the CMO-R to reduce escape-motivated prob- cedent prompting to reduce errors and by interspersing
lem behavior during instructional sessions with persons easy tasks with the more difficult demands. This interven-
with developmental disabilities and autism. He suggests tion resulted in zero levels of destructive behaviors. Weeks
both consequence and antecedent modifications that may and Gaylord-Ross (1981) found that students had higher
be effective. In any case, presentation of the stimuli that levels of problem behavior during difficult as opposed to
have evoked negatively reinforced problem behavior easy tasks. Almost no problem behavior occurred when
without presentation of the worsening condition that has students were making correct responses. Errorless instruc-
typically accompanied them will reduce the value of the tion dramatically reduced problem behavior and increased
CMO-R and abate problem behavior. One method of learning.
accomplishing this outcome is to pair and embed the teach- These finding suggest the importance of minimizing
ing context, personnel, materials, and so forth with an learner errors through antecedent prompting methods. The
“improving set of conditions” through the delivery of posi- reduction in errors probably functioned as an abolishing
tive reinforcers. In this way, the aversiveness of the teaching operation that reduced the effectiveness of escape as a
environment is reduced and therefore less likely to evoke reinforcing consequence and, as a result, reduced escape-
escape and avoidance responses (Kemp & Carr, 1995). motivated problem behavior.
118 Focus on Autism and Other Developmental Disabilities 25(2)

Stimulus Demand Fading and an adult with autism. They compared a condition in
which the same task was presented on every trial to a condi-
Instructional demands have been implicated as a CMO-R in tion in which tasks were varied frequently. The task variation
several studies (DeLeon et al., 2001; Ebanks & Fisher, 2003; condition produced less problem behavior. They demon-
Lalli et al., 1999; McComas et al., 2000). Researchers have strated that increased skill acquisition occurred with the task
shown that escape-motivated problem behavior can be virtu- variation approach in a second study with an adult with
ally eliminated by removing demands (E. G. Carr & Durand, autism. These results were probably obtained because task
1985; E. G Carr et al., 1980). However, this approach is variation functioned as an abolishing operation that reduced
impractical for teaching children with autism because the value of escape from tasks. To use everyday language,
failure to present instructional demands virtually eliminates doing the same task over and over again is boring. These
learning opportunities. As a result, several researchers have findings suggest that mixing and varying instructional tasks
shown that it is possible to alter the demands along a vari- during discrete trial instruction may function as an abolish-
ety of dimensions, including task difficulty (Cameron, ing operation and decrease the effectiveness of escape as
Ainsleigh, & Bird, 1992; Weeks & Gaylord-Ross, 1981), a reinforcer.
number of low probability requests (Ducharme & Worling,
1994), response effort (Horner & Day, 1991; Richman,
Wacker, & Winborn, 2001; Wacker et al., 1990; Weld & Pace of Instruction
Evans, 1990), and number or rate of instructional trials Researchers have evaluated the effects of pace of instruction
(Kennedy, 1994; Zarcone, Iwata, Smith, Mazaleski, & on acquisition and problematic behavior in different types
Lerman, 1994; Zarcone, Iwata, Vollmer et al., 1993). For of learners (Carnine, 1976; Tincani, Ernsbarger, Harrison, &
example, Pace, Iwata, Cowdery, Andree, and McIntyre Heward, 2005). For example, R. L. Koegel, Dunlap, and
(1993) used a combination of extinction and fading Dyer (1980) and G. Dunlap, Dyer, and Koegel, (1983)
instructional demands to reduce escape-motivated problem demonstrated that short intertrial intervals (ITI) reduced ste-
behaviors. Initially the instructor simply sat with the child reotypic behavior in children with autism when compared to
until she or he completed a session with no problem behav- long ITIs. In addition, children achieved higher rates of cor-
ior. Then, the instructor delivered one instructional demand rect responding during the short ITI condition. In general,
at about the midpoint of the session. Over successive ses- children exhibited less off-task behavior and acquired more
sions, more demands were faded into the session. The results skills during brisk-paced instruction. Pace of instruction
suggested that the fading procedures accelerated the behav- probably functions as an abolishing operation, reducing the
ior reduction effects of extinction. These results probably value of escape and avoidance as reinforcers. Specifically,
were obtained because the original task demands functioned during the ITI, reinforcement is not available and with
as a CMO-R that increased the value of escape-motivated longer, as compared to shorter intervals, the child receives a
problem behavior. Removal of demands weakened the MO lower rate of reinforcement for instructional sessions of
and decreased escape-motivated problem behaviors. Their equal duration. Roxburgh and Carbone (2007) investigated
gradual re-introduction did not create enough of a CMO-R this issue directly and found that during instruction of
to increase escape-motivated problem behaviors. children with autism, shorter ITIs produced a higher rate of
Modifying the rate, difficulty, and effort of responses reinforcement and therefore less problem behavior. During
during discrete trial instruction appears to reduce escape- long ITIs, the learner likely receives automatic rein-
motivated and avoidance-motivated problem behaviors. forcement for stereotypic behavior. In contrast, instructional
Over time, instructors may be able to fade in the rate, diffi- demands delivered at a brisk pace reduce the rate of rein-
culty, and effort of demands until high levels of instructional forcement available through automatic reinforcement and
participation are reached without problem behavior. increases the rate of socially mediated positive reinforce-
ment available. Children who do not engage in off-task
behavior and are impulsive (i.e., respond too quickly) are
Task Variation unlikely to benefit from fast-paced instruction (Dyer,
Some investigators have found that mass trialing (i.e., con- Christian, & Luce, 1982). However, it appears that these
stantly presenting the same stimulus over consecutive trials) children are less likely to engage in escape-motivated
may increase problematic behavior during instructional ses- problem behavior in the first place.
sions for persons with autism (G. Dunlap, 1984; G. Dunlap, In contrast, a few researchers suggest that a faster pace
Dyer, & Koegel, 1980; L. K. Dunlap & Dunlap, 1987; of instruction is related to increases in escape-motivated
McComas et al., 2000). For example, Winterling, Dunlap, problem behavior (Zarcone et al., 1994; Zarcone, Iwata,
and O’Neill (1987) demonstrated that task variation dra- Vollmer et al., 1993). In these studies, when the pace of
matically reduced the levels of problem behavior for children the instruction was increased, the number of tasks the
Carbone et al. 119

individuals were required to complete also was increased. of understanding the concept of the CMO-R in reducing
For example, in the study by R. Smith, Iwata, Goh, and problem behavior.
Shore (1995), the two conditions were a high-rate condition
in which 30 trials were presented during the 15-min session
and a low-rate condition in which 10 trials were presented Choice Making
during the 15-min session. The low-rate condition always Choice making may function as an abolishing operation
produced lower rates of self-injurious behaviors. Because and reduce the value of escape from tasks (McComas et al.,
the number of instructional demands delivered is con- 2000; Vaughn & Horner, 1997). For example, Dyer, Dunlap,
founded with pace in this experiment, it is not possible to and Winterling (1990) found problem behavior was dra-
separate out the effects of pace with the effects of the matically reduced when students were offered choices of
number of instructional demands. The authors of this study activities and reinforcers during instructional sessions. The
discussed the difficulty of attempting to study pace of choice condition dramatically reduced problem behavior in
instruction without confounding variables of differences in all participants. Choice likely functions as an abolishing
reinforcement amount, rate, and ITIs. operation for escape-motivated problem behavior because
Overall, it has been found that pace of instruction is an the child has the opportunity to specify the current motiva-
important variable that might serve as an abolishing opera- tion. Because the child could stop an activity at any time
tion that reduces the effectiveness of escape as a reinforcer. and choose a new activity, there is limited possibility of cre-
But as mentioned above, there are some exceptions to this ating a CMO-R for escape-maintained problem behavior.
finding. First, pace of instruction is not likely to be an effec- Many children will benefit from the opportunity to make
tive abolishing operation if the number of demands or the choices regarding activities within discrete trial instruction
duration of the session also is increased. Second, if a child sessions.
does not engage in escape-motivated problem behavior or
engages in quick responding, she or he is less likely to ben-
efit from a fast pace of instruction. For a comprehensive Interspersal Instruction
discussion of variables related to pace of instruction see Several researchers have demonstrated that problem behav-
Tincani et al. (2005). ior can be reduced when easy tasks are interspersed with
difficult tasks (E. G. Carr et al., 1980; Harchik & Putzier,
1990; Horner, Day, Sprague, O’Brien, & Healthfield, 1991;
Neutralizing Routines Mace & Belfiore, 1990; Mace et al., 1988; Neef, Iwata, &
Several researchers have demonstrated that variables beyond Page, 1980; Singer, Singer, & Horner, 1987). In two studies,
the control of the instructor may establish CMO-R during similar effects were found when interspersing social com-
planned instructional sessions. Occurrences such as sleep ments with instructional demands (Kennedy, 1994; Kennedy
deprivation (Kennedy & Meyer, 1996; O’Reilly, 1995), et al., 1995). Problem behavior may have been reduced with
otitis media (O’Reilly, 1997), and cancellation of preferred the use of these procedures because the interspersal of easy
activities (Horner, Day, & Day, 1997) have increased tasks functions as an abolishing operation reducing the value
problem behavior during instructional sessions that have of escape as a reinforcer. Difficult tasks probably function as
followed them. Horner et al. (1997) demonstrated that it a CMO-R because they are correlated with a worsening set
may be possible to create an abolishing operation or “neu- of conditions related to low rates of reinforcement, high
tralizing routine” that reduces the effectiveness of the value rates of error, and higher rates of social disapproval. By
of instructional demands as CMO-R following unplanned interspersing easy tasks with more difficult tasks, the value
daily occurrences. In this study, two students engaged in of the CMO-R is reduced. The recommendation is to com-
problem behavior contingent on error corrections when the bine extinction with interspersal instruction to ensure its
additional CMO-R of having a planned activity cancelled or effectiveness (Zarcone, Iwata, Hughes, & Vollmer, 1993).
delayed occurred. The implementation of a neutralizing rou- It is also important to avoid presenting easy tasks immedi-
tine substantially reduced problem behavior. The neutralizing ately following problem behavior. If this were to occur,
routines used in this study consisted of the students engag- problem behavior would likely be strengthened by negative
ing in highly preferred activities 30 to 40 min prior to the reinforcement (Sailor, Guess, Rutherford, & Baer, 1968).
instructional session. Students emitted zero levels of prob- Despite the data suggesting the negative effects of this prac-
lematic behavior during the neutralizing routine condition. tice (Sailor et al., 1968), many educators remove difficult
Some individuals will benefit from high periods of tasks contingent upon problem behavior and present alterna-
dense reinforcement and low demand activities prior to tive maintenance or more easily mastered tasks. In any case,
instructional sessions especially after the denial of other children with autism may benefit from interspersal of easy
reinforcers. These researchers demonstrate the importance and target skills during discrete trial instruction.
120 Focus on Autism and Other Developmental Disabilities 25(2)

Task Novelty escape and avoidance behavior exhibited by a large percent-

age of children with autism and related disabilities. A
The simple presentation of a novel task may serve as CMO-R conceptually systematic approach to determining the influ-
for some students and increase the value of task removal as ence of antecedent motivational variables will equip
a reinforcer. R. Smith et al. (1995) demonstrated this effect instructional decision makers with a wider range of choices
by introducing new tasks each time self-injurious behavior of teaching methods and, maybe more importantly, will pro-
(SIB) reached low levels. Following introduction of the vide a natural science approach to analyzing and modifying
novel task, SIB increased, leading to the identification of the instructional methods when the performance of learners with
novel tasks as MOs. Simple exposure to the task over sev- autism does not result in expected outcomes.
eral sessions may reduce the value of this CMO-R. It is
probably important to introduce novel tasks gradually, Acknowledgments
because introducing high rates of novel stimuli will likely The authors would like to thank Drs. Jack Michael and Timothy
serve as an MO, increasing the effectiveness of escape as a Vollmer for their comments on an earlier version of this manu-
reinforcer. Gradual introduction may be effective in keeping script and Dr. Patrick McGreevy for helping the first author to
the value of task removal as a reinforcer low. Simple expo- discover what he had to say on this topic.
sure to novel stimuli may benefit some children and reduce This paper was previously published in the Journal of Early
escape-motivated problem behavior. Intensive Behavior Intervention (JEIBI),Volume 4, 2007, now
Journal of Behavior Assessment and Intervention in Children,
(JBAIC). The authors would like to express their appreciation to
Session Duration the editorial staff of JBAIC for their permission to reprint this
The length of the treatment session may serve as a CMO-R paper.
that increases the value of escape. R. Smith et al. (1995)
found idiosyncratic differences among participants in how Declaration of Conflicting Interests
session duration may serve as an MO. The authors clearly The authors declared no conflicts of interest with respect to the
considered the passage of time as a behavioral variable. authorship and/or publication of this article.
Some participants had little or no problematic behavior early
in the session, but high rates later in the session suggested Funding
that the passage of time in the demand condition may have The authors received no financial support for the research and/or
functioned as a CMO-R. Other participants engaged in a authorship of this article.
relatively high rate of problem behavior early in the session,
but the rate decreased over the length of the session. This References
implies that the actual presentation of the demand condition Anderson, S. P., Taras, M., & O’ Malley-Cannon, B. (1996).
may have served as the MO. The authors make treatment Teaching new skills to young children with autism. In C. Maurice,
recommendations based on this analysis. Specifically, for G. Green, & S. C. Luce (Eds.), Behavioral intervention
learners who engage in problematic behavior late in the for young children with autism (pp. 181–194). Austin, TX:
session, it may be best to arrange several sessions of short PRO-ED.
durations. For students who engage in the most problem Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current
behavior at the start of the session, it may be advantageous dimensions of applied behavior analysis. Journal of Applied
to have relatively long instructional sessions, but fewer per Behavior Analysis, 1, 91–97.
day. These treatment recommendations are directly related Birnbrauer, J. S., & Leach, D. J. (1993). The Murdoch Early
to an analysis of the behavior based on session duration Intervention Program after 2 years. Behaviour Change, 10,
functioning as an MO that may either establish or abolish 63–74.
the reinforcing value of escape from tasks. Call, N. A., Wacker, D. P., Ringdahl, J. E., Cooper-Brown, L. J., &
Boeltric, E. W. (2004). An assessment of antecedent events influ-
encing noncompliance in an outpatient clinic. Journal of Applied
Conclusions Behavior Analysis, 37, 145–158.
A thorough understanding of the principle of motivation and Cameron, M. J., Ainsleigh, S. A., & Bird, F. L. (1992). The acqui-
an analysis of instructional methods as MOs can provide sition of stimulus control of compliance and participation
behavior analysts with a powerful technology for reducing during an ADL routine. Behavioral Residential Treatment, 7,
problem behavior during any instruction, including discrete 327–340.
trial instruction. With knowledge of the concept of the Carnine, D. W. (1976). Effects of two teacher presentation rates on
CMO-R, practitioners may be better equipped to evaluate, off-task behavior, answering correctly and participation. Journal
select, and implement instructional methods that reduce of Applied Behavior Analysis, 9, 199–206.
Carbone et al. 121

Carr, E. G., & Carlson, J. I. (1993). Reduction of severe behav- D. P. Wacker (Eds.), Communication and language interven-
ior problems in the community using a multicomponent tion series: Vol. 3. Communicative alternatives to challenging
treatment approach. Journal of Applied Behavior Analysis, behavior: Integrating functional assessment and intervention
26, 157–172. strategies (pp. 177–203). Baltimore: Brookes.
Carr, E. G., & Durand, V. M. (1985). Reducing behavior prob- Dunlap, G., & Kern, L. (1996). Modifying instructional activi-
lems through functional communication instruction. Journal of ties to promote desirable behavior: A conceptual and practical
Applied Behavior Analysis, 18, 111–126. framework. School Psychology Quarterly, 11, 297–312.
Carr, E. G., Newsome, C. D., & Binkoff, J. A. (1980). Escape as Dunlap, G., Kern-Dunlap, L., Clarke, S., & Robbins, F. R. (1991).
a factor in the aggressive behavior of two retarded children. Functional assessment, curricular revisions and severe behav-
Journal of Applied Behavior Analysis, 13, 101–117. ior problems. Journal of Applied Behavior Analysis, 24,
Carr, J. E., & LeBlanc, L. A. (2006). Noncontingent reinforce- 387–397.
ment as antecedent behavior support. In J. K. Luiselli (Ed.), Dunlap, G., Kern, L., dePerczel, M., Clarke, S., Wilson, D.,
Antecedent assessment & intervention: Supporting children & Childs, K. E., White, R., . . . Falk, G. (1993). Functional anal-
adults with developmental disabilities in community settings ysis of classroom variables for students with emotional and
(pp. 147–164). Baltimore: Brookes. behavioral challenges. Behavioral Disorders, 18, 275–291.
Cohen, H., Amerine-Dickens, M., & Smith, T. (2006). Early inten- Dunlap, L. K., & Dunlap, G. (1987). Using task variation to moti-
sive behavioral treatment: Replication of the UCLA model in a vate handicapped students. Teaching Exceptional Children, 19,
community setting. Journal of Developmental and Behavioral 16–19.
Pediatrics, 27, S145–S155. Durand, V. M. (1990). Severe behavior problems: A functional
Crockett, J. L., & Hagopian, L. P. (2006). Prompting procedures communication instruction approach. New York: Guilford.
as establishing operations for escape-maintained behavior. Dyer, K., Christian, W. P., & Luce, S. C. (1982). The role of
Behavioral Interventions, 21, 65–71. response delay in improving the discrimination performance
DeLeon, I. G., Neidert, P. L., Anders, B. M., & Rodriguez-Catter, V. of autistic children. Journal of Applied Behavior Analysis, 15,
(2001). Choice between positive and negative reinforcement dur- 231–240.
ing treatment for escape-maintained behavior. Journal of Applied Dyer, K., Dunlap, G., & Winterling, V. (1990). Effects of choice
Behavior Analysis, 34, 521–525. making on the serious problem behaviors of students with
Derby, K. M., Wacker, D. P., Sasso, G., Steege, M., Northup, J., severe handicaps. Journal of Applied Behavior Analysis, 23,
Cigrand, K., & Asmus, J. (1992). Brief functional assessment 515–524.
techniques to evaluate aberrant behavior in an outpatient Ebanks, M. E., & Fisher, W. W. (2003). Altering the timing of
setting: A summary of 79 cases. Journal of Applied Behavior academic prompts to treat destructive behavior maintained by
Analysis, 25, 713–721. escape. Journal of Applied Behavior Analysis, 36, 355–359.
Drash, P. W., & Tudor, R. M. (1993). A functional analysis of ver- Eikseth, S., Smith, T., Jahr, E., & Eldevik, S. (2002). Intensive
bal delay in preschool children: Implications for prevention behavioral treatment at school for four to seven year old chil-
and total recovery. The Analysis of Verbal Behavior, 11, 19–29. dren with autism: A one-year follow-up. Behavior Modification,
Ducharme, J. M., & Worling, D. E. (1994). Behavioral momentum 26, 49–68.
and stimulus fading in the acquisition and maintenance of child Eikseth, S., Smith, T., Jahr, E., & Eldevik, S. (2007). Outcomes for
compliance in the home. Journal of Applied Behavior Analysis, children with autism who began intensive behavioral treatmet
27, 639–647. between ages 4 and 7. Behavior Modification, 31, 264–278.
Dunlap, G. (1984). The influence of task variation and maintenance Fisher, W., Piazza, C., Cataldo, M., Harrell, R., Jefferson, G., &
tasks in the learning and affect of autistic children. Journal of Conner, R. (1993). Functional communication training with
Experimental Child Psychology, 37, 41-64. and without extinction and punishment. Journal of Applied
Dunlap, G., Dyer, K., & Koegel, R. L. (1980). Motivating autistic Behavior Analysis, 26, 23–36.
children through stimulus variation. Journal of Applied Behavior Green, G. (1996). Early behavioral intervention for autism: What
Analysis, 13, 619–627. does research tell us? In C. Maurice, G. Green, & S. C. Luce
Dunlap, G., Dyer, K., & Koegel, R. L. (1983). Autistic self- (Eds.), Behavioral intervention for young children with autism
stimulation and intertrial interval. Journal of Mental Defi- (pp. 181–194). Austin, TX: PRO-ED.
ciency, 88, 194–202. Hagopian, L. P., Fisher, W. W., Sullivan, M. T., Acquisto, J., &
Dunlap, G., Foster-Johnson, L., Clarke, S., Kern, L., & LeBlanc, L. A. (1998). Effectiveness of functional communi-
Childs, K. E. (1995). Modifying activities to produce func- cation training with and without extinction and punishment:
tional outcomes: Effects on the disruptive behaviors of students A summary of 21 inpatient cases. Journal of Applied Behavior
with disabilities. Journal of the Association for Persons with Analysis, 31, 211–235.
Severe Handicaps, 20, 248–258. Hanley, G. P., Iwata, B. A., & McCord B. E. (2003). Functional
Dunlap, G., & Kern, L. (1993). Assessment and intervention for analysis of problem behavior: A review. Journal of Applied
children within the instructional curriculum. In J. Reichle & Behavior Analysis, 36, 147–185.
122 Focus on Autism and Other Developmental Disabilities 25(2)

Harchik, A. G., & Putzier, V. A. (1990). The use of high-probability Kern, L., & Dunlap, G. (1998). Curricular modification to
requests to increase compliance with instructions to take medi- promote desirable classroom behavior. In J. K. Luiselli &
cation. The Journal of the Association for Persons with Severe M. J. Cameron (Eds.), Antecedent control (pp. 289–308).
Handicaps, 15, 40–43. Baltimore: Brookes.
Heckaman, K., Alber, S., Hooper, S., & Heward, W. (1998). Koegel, L. K., Koegel, R. L., Shoshan, Y., & McNerney, E. (1999).
A comparison of least to most and progressive time delay on Pivotal response intervention II: Preliminary long term out-
the disruptive behavior of students with autism. Journal of come data. Journal of the Association for Persons with Severe
Behavioral Education, 8, 171–202. Handicaps, 24, 186–198.
Hoffman, H. S. (1966). The analysis of discriminated avoidance. Koegel, R. L., Carter, C. M., & Koegel, L. K. (1998). Setting
In W. K. Honig (Ed.). Operant behavior: Areas of research and events to improve parent – teacher coordination with autism.
application (pp. 499–530). Norwalk, CT: Appleton-Century- In J. K. Luiselli & M. J. Cameron (Eds.), Antecedent control
Crofts. (pp. 167–187). Baltimore: Brookes.
Horner, R. H., & Day, H. M. (1991). The effects of response effi- Koegel, R. L., Dunlap, G., & Dyer, K. (1980). Intertrial interval
ciency in functionally equivalent competing behaviors. Journal duration and learning in autistic children. Journal of Applied
of Applied Behavior Analysis, 24, 719–732. Behavior Analysis, 13, 91–99.
Horner, R. H., Day, H. M., & Day, J. R. (1997). Using neutral- Koegel, R. L., Koegel, L. K., Frea, W. D., & Smith, A. E. (1995).
izing routines to reduce problem behaviors. Journal of Applied Emerging interventions for children with autism: Longitudi-
Behavior Analysis, 30, 601–614. nal and lifestyle implications. In R. L. Koegel & L. K. Koegel
Horner, R. H., Day, H. M., Sprague, J. R., O’Brien, M., & (Eds.), Teaching children with autism: Strategies for initiating
Heathfield, L. T. (1991). Interspersed requests: A nonaver- positive interactions and improving learning opportunities
sive procedure for reducing aggression and self-injury during (pp. 1–15). Baltimore: Brookes.
instruction. Journal of Applied Behavior Analysis, 24, 265–278. Lalli, J. S., & Casey, S. (1996). Treatment of multiplied controlled
Howard, J. S., Sparkman, C. R., Cohen, H. G., Green, G., & problem behavior. Journal of Applied Behavior Analysis, 29,
Stanislaw, H. (2005). A comparison of intensive behavior ana- 391–396.
lytic and eclectic treatments for young children with autism. Lalli, J. S., Vollmer, T., Progar, P. R., Wright, C., Borrero, J.,
Research in Developmental Disabilities, 26, 359–383. Dency, D., . . . May, W. (1999). Competition between positive
Iwata, B. A., Pace, G. M., Dorsey, M. F., Zarcone, J. R., and negative reinforcement in the treatment of escape behavior.
Vollmer, T. R., Smith, R. G., . . . Mazalesk, J. L. (1994). The Journal of Applied Behavior Analysis, 32, 285–296.
functions of self-injurious behavior: An experimental- Langthorne, P., McGill, P., & O’Reilly, M. O. (2007) Incorpo-
epidemiological analysis. Journal of Applied Behavior Analysis, rating motivation into the functional analysis of challenging
27, 215–240. behavior. Behavior Modification, 31, 466–487.
Iwata, B. A., Smith, R. G., & Michael, J. (2000). Current research Laraway, S., Snycerski, S., Michael, J., & Poling, A. (2003).
on the influence of establishing operations on behavior in Motivating operations and terms to describe them: Some fur-
applied settings. Journal of Applied Behavior Analysis, 33, ther refinements. Journal of Applied Behavior Analysis, 36,
401–410. 407–414.
Kemp, D. C., & Carr, E. G. (1995). Reduction of severe problem Leaf, R., & McEachin, J. (1999). A work in progress: Behavior
behavior in community employment using a hypothesis-driven management strategies and a curriculum for intensive behav-
multicomponent intervention approach. Journal of the Asso- ioral treatment of autism. New York: DRL Books, L.L.C.
ciation for Persons with Severe Handicaps, 20, 229–247. Lovaas, O. I. (1981). Teaching developmentally disabled children:
Kennedy, C. H. (1994). Manipulating antecedent conditions to alter The ME book. Austin, TX: PRO-ED.
the stimulus control of problem behavior. Journal of Applied Lovaas, O. I. (1987). Behavioral treatment and normal educational
Behavior Analysis, 27, 161–170. and intellectual functioning in young autistic children. Journal
Kennedy, C. H., Itkonen, T., & Lindquist, K. (1995). Comparing of Consulting and Clinical Psychology, 55, 3–9.
interspersed requests and social comments as antecedents for Lovaas, O. I. (2003). Teaching individuals with developmental
increasing student compliance. Journal of Applied Behavior delays: Basic intervention techniques. Austin, TX: PRO-ED.
Analysis, 28, 97–98. Mace, F. C., & Belfiore, P. (1990). Behavioral momentum in the
Kennedy, C. H., & Meyer, K. A. (1996). Sleep deprivation, allergy treatment of escape-motivated stereotypy. Journal of Applied
symptoms and negatively reinforced problem behavior. Journal Behavior Analysis, 23, 507–514.
of Applied Behavior Analysis, 29, 133–135. Mace, F. C., Hock, M. L., Lalli, J. S., West, B. J., Belfiore, P.,
Kern, L., Childs, K. E., Dunlap, G., Clarke, S., & Falk, G. D. Pinter, E., & Brown, D. K. (1988). Behavioral momentum in
(1994). Using assessment-based curricular intervention to the treatment of noncompliance. Journal of Applied Behavior
improve the classroom behavior of a student with emotional Analysis, 21, 123–141.
and behavioral challenges. Journal of Applied Behavior Analysis, Maurice, C., Green, G., & Foxx, R. M. (2001). Making a difference:
27, 7–19. Behavioral intervention for autism. Austin, TX: PRO-ED.
Carbone et al. 123

Maurice, C., Green, G., & Luce, S. C. (1996). Behavioral inter- O’Reilly, M. F. (1997). Functional analysis of episodic self-injury
ventions for young children with autism: A manual for parents correlated with recurrent otitis media. Journal of Applied
and professionals. Austin, TX: PRO-ED. Behavior Analysis, 30, 165–167.
McComas, J., Hoch, H., Paone, D., & El-Roy, D. (2000). Escape Pace, G. M., Iwata, B. A., Cowdery, G .E., Andree, P. J., &
behavior during academic tasks: A preliminary analysis of idio- McIntyre, T. (1993). Stimulus (instructional) fading during
syncratic establishing operations. Journal of Applied Behavior extinction of self-injurious escape behavior. Journal of Applied
Analysis, 33, 479–493. Behavior Analysis, 26, 205–212.
McEachin, J. J., Smith, T., & Lovaas, O. I. (1993). Long-term out- Parrish, J. P., Cataldo, M. F., Kolko, D. J., Neef, N. A., &
come for children with autism who received early intensive Egel, A. L. (1986). Experimental analysis of response covaria-
behavioral treatment. American Journal on Mental Retardation, tion among compliant and inappropriate behaviors. Journal of
97, 359–372. Applied Behavior Analysis, 19, 241–254.
McGill, P. (1999). Establishing operations: Implications for the Piazza, C. C., Fisher, W. W., Hanley, G. P., Remick, M. L.,
assessment, treatment and prevention of problem behaviors. Contrucci, S. A., & Aitken, T. L. (1997). The use of positive
Journal of Applied Behavior Analysis, 32, 389–418. and negative reinforcement in the treatment of escape-maintained
Michael, J. (1982). Distinguishing between the discriminative and destructive behavior. Journal of Applied Behavior Analysis,
motivational functions of stimuli. Journal of the Experimental 30, 279–298.
Analysis of Behavior, 37, 149–155. Remington, B., Hastings, R. P., Kovshoff, H., degli Espinoza, F.,
Michael, J. (1988). Establishing operations and the mand. The Jahr, E., Brown, T., . . . Ward, N. (2007). Early intensive behav-
Analysis of Verbal Behavior, 6, 3–9. ioral intervention: Outcomes for children with autism and their
Michael, J. (1993). Establishing operations. The Behavior Analyst, parents after two years. American Journal on Mental Retardation,
16, 191–206. 112, 418–438.
Michael, J. (1996). What every student of behavior analysis ought Richman, D. M., Wacker, D. P., & Winborn, L. (2001). Response
to learn: A system for classifying the multiple effects of behav- efficiency during functional communication instruction: Effects
ioral variables. The Behavior Analyst, 18, 273–284. of effort on response allocation. Journal of Applied Behavior
Michael, J. (2000). Implications and refinements of the establish- Analysis, 34, 73–76.
ing operation concept. Journal of Applied Behavior Analysis, 33, Roxburgh, C., & Carbone, V. J. (2007). The effects of varying teacher
401–410. presentation rate on responding during discrete trial instruction
Michael, J. (2004). Concepts and principles of behavior analysis. of two children with autism. Manuscript in preparation.
Kalamazoo, MI: Society for the Advancement of Behavior Russo, D. C., Cataldo, M. F., & Cushing, P. J. (1981).Compliance
Analysis. instruction and behavioral covariation in the treatment of mul-
Michael, J. (2007). Motivating Operations. In J. O. Cooper, tiple behavior problems. Journal of Applied Behavior Analysis,
T. E. Heron, & W. L. Heward (Eds.), Applied behavior analysis 14, 209–222.
(pp. 374–391). Upper Saddle River, NJ: Prentice Hall/Merrill Sailor, W., Guess, D., Rutherford, G., & Baer, D. M. (1968).
Miltenberger, R. G. (2006). Antecedent interventions for challeng- Control of tantrum behavior by operant techniques during
ing behaviors maintained by escape from instructional activities. extinction of self-injurious escape behavior. Journal of Applied
In J. K. Luiselli (Ed.), Antecedent assessment & intervention: Behavior Analysis, 1, 237–243.
Supporting children & adults with developmental disabilities Sallows, G. O., & Graupner, T. D. (2005). Intensive behavioral
in community settings (pp. 101–124). Baltimore, MD: Brookes. treatment for children with autism: Four year outcome and
Munk, D., & Repp, A. (1994). The relationship between instruc- predictors. American Journal on Mental Retardation, 6,
tional variables and problem behavior: A review. Exceptional 417–438.
Children, 60, 390–401. Schlinger, H. D. (1993). Establishing operations: Another step
National Research Council. (2001). Educating children with toward a functional taxonomy of environmental events. The
autism. Committee on Educational Interventions for Children Behavior Analyst, 16, 207–209.
with Autism. Division of Behavioral and Social Sciences and Singer, G. H., Singer, J., & Horner, R. H. (1987). Using pretask
Education. Washington, DC: National Academy Press. requests to increase the probability of compliance for students
Neef, N. A., Iwata, B. A., & Page, T. (1980). The effects of inter- with severe disabilities. Journal of the Association for Persons
spersal instruction versus high density reinforcement on spelling with Severe Handicaps, 12, 287–291.
acquisition and retention. Journal of Applied Behavior Analysis, Skinner, B. F. (1968). The technology of teaching. Upper Saddle
13, 153–158. River, NJ: Prentice Hall.
O’Reilly, M. F. (1995). Functional analysis and treatment of Smith, R., & Iwata, B. (1997). Antecedent influences on behav-
escape-maintained aggression correlated with sleep depriva- ior disorders. Journal of Applied Behavior Analysis, 30,
tion. Journal of Applied Behavior Analysis, 21, 225–226. 267–278.
124 Focus on Autism and Other Developmental Disabilities 25(2)

Smith, R. G., Iwata, B. A., Goh, H. L., & Shore, B. A. (1995). Winterling, V., Dunlap, G., & O’Neill, R. E. (1987). The influence
Analysis of establishing operations for self-injury maintained of task variation on the aberrant behaviors of autistic students.
by escape. Journal of Applied Behavior Analysis, 28, 515–536. Education and Treatment of Children, 10, 105–119.
Smith, T. (1999). Outcome of early intervention for children with Zarcone, J. R., Iwata, B. A., Hughes, C. E., & Vollmer, T. R.
autism. Clinical Psychology: Research and Practice, 6, 33–49. (1993). Momentum versus extinction effects in the treatment
Smith, T. (2001). Discrete trial instruction in the treatment of of self-injurious escape behavior. Journal of Applied Behavior
autism. Focus on Autism and Other Developmental Disabilities, Analysis, 26, 135–136.
16, 86–92. Zarcone, J. R., Iwata, B. A., Smith, R. G., Mazaleski, J. L., &
Smith, T., Groen, A. D., & Wynne, J. W. (2000). Randomized trial Lerman, D. C. (1994). Reemergence and extinction of self-
of intensive early intervention for children with pervasive devel- injurious escape behavior during stimulus (instructional)
opmental disorder. American Journal on Mental Retardation, fading. Journal of Applied Behavior Analysis, 27, 307–316.
105, 269–285. Zarcone, J. R., Iwata, B. A., Vollmer, T. R., Jagtiani, S.,
Sundberg, M. (1993). The application of establishing operation. Smith, R. G., & Mazaleski, J. L. (1993). Extinction of self-
The Behavior Analyst, 16, 215–218. injurious escape behavior with and without instructional fading.
Sundberg, M. L., & Partington, J. W. (1998). Teaching language Journal of Applied Behavior Analysis, 26, 353–360.
to children with autism or other developmental disabilities.
Pleasant Hills, CA: Behavior Analysts, Inc. About the Authors
Tincani, M., Ernsbarger, S., Harrison, T. J., & Heward, W. L. (2005). Vincent J. Carbone, EdD, BCBA-D, is director of the Carbone
Effects of two instructional paces on pre-k children’s participa- Clinic and adjunct faculty member at Penn State University and
tion rate, accuracy, and off task behavior in the language for Simmons College, Boston, MA. His current interests include the
learning program? Journal of Direct Instruction, 5, 97–109. application of behavior analytic principles to the instructional
Vaughn, B. J., & Horner, R. H. (1997). Identifying instructional needs of persons with developmental disabilities and autism and
tasks that occasion problem behaviors and assessing the effects the teaching of verbal behavior.
of student versus teacher choice among these tasks. Journal of
Applied Behavior Analysis, 30, 299–312. Barry D. Morgenstern, PhD, BCBA-D, is Connecticut Director
Wacker, D. P., Steege, M. W., Northup, J., Sasso, G., Berg, W., of Research for the Institute of Professional Practice. His current
Reimers, T., . . . Donn, L. (1990). A component analysis of func- interests include parent training and instructional design for chil-
tional communication instruction across three topographies of dren with autism.
severe behavior problems. Journal of Applied Behavior Analysis,
23, 417–429. Laura Kolberg, BA, BCaBA, is an outreach consultant at the
Weeks, M., & Gaylord-Ross, R. (1981). Task difficulty and aberrant Carbone Clinic, Valley Cottage, NY. Her current interests include
behavior in severely handicapped students. Journal of Applied the application of behavioral principles in classrooms for children
Behavior Analysis, 14, 449–463. with developmental disabilities and autism.
Weld, E. M., & Evans, I. M., (1990). Effects of part versus whole
instructional strategies on skill acquisition and excess behavior. Gina Zecchin-Tirri, MSEd, BCBA, is an outreach consultant at
American Journal of Mental Retardation, 4, 377–386. the Carbone Clinic, Valley Cottage, NY. Her interests include
Wilder, D. A., & Carr, J. E. (1998). Recent advances in the modi- classroom applications of behavioral principles and staff training
fication of establishing operations to reduce aberrant behavior. and development in school programs for children with develop-
Behavioral Interventions, 13, 43–59. mental disabilities and autism.