Athens y Vollmer, Refuerzo Diferencial

7/27/2019 Athens y Vollmer, Refuerzo Diferencial

1/21

AN INVESTIGATION OF DIFFERENTIAL REINFORCEMENT OFALTERNATIVE BEHAVIOR WITHOUT EXTINCTION

ELIZABETH S. ATHENS

MARCUS AUTISM CENTER

AND

TIMOTHY R. VOLLMER

UNIVERSITY OF FLORIDA

We manipulated relative reinforcement for problem behavior and appropriate behavior usingdifferential reinforcement of alternative behavior (DRA) without an extinction component.Seven children with developmental disabilities participated. We manipulated duration(Experiment 1), quality (Experiment 2), delay (Experiment 3), or a combination of each(Experiment 4), such that reinforcement favored appropriate behavior rather than problem

behavior even though problem behavior still produced reinforcement. Results of Experiments 1to 3 showed that behavior was often sensitive to manipulations of duration, quality, and delay inisolation, but the largest and most consistent behavior change was observed when severaldimensions of reinforcement were combined to favor appropriate behavior (Experiment 4).Results suggest strategies for reducing problem behavior and increasing appropriate behaviorwithout extinction.

Key words: attention deficit hyperactivity disorder, autism, concurrent schedules, differentialreinforcement, extinction, problem behavior

_______________________________________________________________________________

Differential reinforcement is a fundamentalprinciple of behavior analysis that has led to thedevelopment of a set of procedures used astreatment for problem behavior (Cooper,Heron, & Heward, 2007). One of the mostfrequently used of these procedures is thedifferential reinforcement of alternative behav-ior (DRA). DRA typically involves withholdingreinforcers following problem behavior (extinc-tion) and providing reinforcers following ap-

propriate behavior (Deitz & Repp, 1983).Pretreatment identification of the reinforcersthat maintain problem behavior (i.e., functionalanalysis) permits the development of extinctionprocedures, which, by definition, must match

the function of problem behavior (Iwata, Pace,Cowdery, & Miltenberger, 1994). In addition,the reinforcer maintaining problem behaviorcan be delivered contingent on the occurrenceof an alternative, more appropriate response.Under these conditions, DRA has been success-ful at reducing problem behavior (Dwyer-Moore & Dixon, 2007; Vollmer & Iwata,1992).

Although extinction is an important and

powerful component of DRA, it is, unfortu-nately, not always possible to implement it(Fisher et al., 1993; Hagopian, Fisher, Sullivan,

Acquisto, & LeBlanc, 1998). For example, acaregiver may be physically unable to preventescape with a large or combative individual,leading to compromises in integrity of escapeextinction. It would also be difficult to withholdreinforcement for behavior maintained by

attention in the form of physical contact ifphysical blocking is required to protect theindividual or others. For example, if anindividuals attention-maintained eye gouging

Address correspondence concerning this article toElizabeth Athens, who is now at ABA Learning Centre,#10021320 Gordon Way, Richmond, British ColumbiaV6W 1J8, Canada (e-mail: [email protected]).

doi: 10.1901/jaba.2010.43-569

We thank Brian Iwata, Lise Abrams, and Stephen Smithfor their comments on an earlier draft of this manuscript.Portions of this manuscript were included as part of thedissertation of the first author at the University of Florida.

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2010, 43, 569589 NUMBER4 (WINTER2010)

569


2/21

is a threat to his or her eyesight, intervention isnecessary to protect vision.

Several studies have found that DRA is lesseffective at decreasing problem behavior when

implemented without extinction (Volkert, Ler-man, Call, & Trosclair-Lasserre, 2009). Forexample, Fisher et al. (1993) evaluated func-tional communication training (FCT; a specifictype of DRA procedure) without extinction,with extinction, and with punishment contin-gent on problem behavior. Results showed thatwhen FCT was introduced without an extinc-tion or punishment component for problembehavior, the predetermined goal of 70%reduction in problem behavior was met withonly one of three participants. FCT was moreeffective at reducing problem behavior whenextinction was included, and the largest andmost consistent reduction was observed whenpunishment was included.

Hagopian et al. (1998) conducted a replica-tion of the Fisher et al. (1993) study and foundthat a predetermined goal of 90% reduction in

problem behavior was not achieved with any of11 participants exposed to FCT withoutextinction. When FCT was implemented withextinction, there was a 90% reduction inproblem behavior for 11 of 25 applications,with a mean percentage reduction in problembehavior of 69% across all applications.

McCord, Thomson, and Iwata (2001) foundthat DRA without extinction had limited effects

on the self-injurious behavior of two individu-als, one whose behavior was reinforced byavoidance of transition and another whosebehavior was reinforced by avoidance oftransition and avoidance of task initiation. Inboth cases, DRA with extinction and responseblocking produced sustained decreases in self-injury. These examinations of research on DRAwithout extinction have shown a bias inresponding toward problem behavior when therate and immediacy of reinforcement ofproblem and appropriate behavior are equiva-lent.

When considering variables that contributeto the effectiveness (or ineffectiveness) of DRAwithout extinction as a treatment for problembehavior, it is helpful to conceptualize differ-

ential reinforcement procedures in terms of aconcurrent-operants arrangement (e.g., Fisher etal., 1993; Mace & Roberts, 1993). Concurrentschedules are two or more schedules in effectsimultaneously. Each schedule independentlyarranges reinforcement for a different response(Ferster & Skinner, 1957). The matching lawprovides a quantitative description of respond-ing on concurrent schedules of reinforcement(Baum, 1974; Herrnstein, 1961). In general,the matching law states that the relative rate ofresponding on one alternative will approximatethe relative rate of reinforcement provided onthat alternative. Consistent with the predictionsof the matching law, some studies have reportedreductions in problem behavior without extinc-tion when differential reinforcement favorsappropriate behavior rather than problembehavior (Piazza et al., 1997; Worsdell, Iwata,

Hanley, Thompson, & Kahng, 2000).For example, Worsdell et al. (2000) exam-

ined the effect of reinforcement rate on responseallocation. Five individuals whose problembehavior was reinforced by social positivereinforcement were first exposed to an FCTcondition in which both problem and appro-priate behavior were reinforced on fixed-ratio(FR) 1 schedules. During subsequent FCT

conditions, reinforcement for problem behaviorwas made more intermittent (e.g., FR 2, FR 3,FR 5), while appropriate behavior continued tobe reinforced on an FR 1 schedule. Four of theparticipants showed shifts in response allocationto appropriate behavior as the schedule ofreinforcement for problem behavior becamemore intermittent. There were several limita-tions to this research. For example, reinforce-ment rate was thinned in the same order foreach participant such that reductions in prob-lem behavior may have been due in part tosequence effects. In addition, the reinforcement

570 ELIZABETH S. ATHENS and TIMOTHY R. VOLLMER


3/21

schedule was thinned to FR 20 for twoindividuals. For these two participants, problembehavior rarely contacted reinforcement. Theschedule in these cases may have been func-

tionally equivalent to extinction rather thanintermittent reinforcement. Nevertheless, theseresults suggest that extinction may not be anecessary treatment component when the rateof reinforcement favors appropriate behaviorrather than problem behavior.

In another example of DRA without extinc-tion, Piazza et al. (1997) examined the effects ofincreasing the quality of reinforcement forcompliance relative to reinforcement associatedwith problem behavior. Three individualswhose problem behavior was sensitive tonegative reinforcement (break from tasks) andpositive reinforcement (access to tangible items,attention, or both) participated. Piazza et al.systematically evaluated the effects of reinforc-ing appropriate behavior with one, two, or threeof the reinforcing consequences (a break,tangible items, attention), both when problem

behavior produced a break and when it did not(escape extinction). For two of the threeparticipants, appropriate behavior increasedand problem behavior decreased when appro-priate behavior produced a 30-s break withaccess to tangible items and problem behaviorproduced a 30-s break. The authors suggestedthat one potential explanation for these findingsis that the relative rates of appropriate behavior

and problem behavior were a function of therelative value of the reinforcement produced byescape. It is unclear, however, whether theintervention would be effective with individualswhose problem behavior was sensitive to onlyone type of reinforcement.

Together these and other studies have shownthat behavior will covary based on rate, quality,magnitude, and delay of reinforcement. Re-sponding will favor the alternative associatedwith a higher reinforcement rate (Conger &Killeen, 1974; Lalli & Casey, 1996; Mace,McCurdy, & Quigley, 1990; Neef, Mace, Shea,

& Shade, 1992; Vollmer, Roane, Ringdahl, &Marcus, 1999; Worsdell et al., 2000), greaterquality of reinforcement (Hoch, McComas,

Johnson, Faranda, & Guenther, 2002; Lalli et

al., 1999; Neef et al.; Piazza et al., 1997),greater magnitude of reinforcement (Catania,1963; Hoch et al., 2002; Lerman, Kelley,Vorndran, Kuhn, & LaRue, 2002), or moreimmediate delivery of reinforcement (Mace,Neef, Shade, & Mauro, 1994; Neef, Mace, &Shade, 1993; Neef, Shade, & Miller, 1994).

Although previous research suggests thatextinction may not always be a necessarycomponent of differential reinforcement treat-ment packages, as described above there werecertain limitations inherent in previous investi-gations. In addition, there has not been acomprehensive analysis of several differentreinforcement dimensions both singly and incombination. The current study sought toextend this existing research by examining theinfluence of multiple dimensions of reinforce-ment and by incorporating variable-interval

(VI) reinforcement schedules.Interval schedules are less likely than ratio

schedules to push response allocation exclusivelytoward one response over another. Under ratioschedules, reinforcer delivery is maximizedwhen responding favors one alternative (Herrn-stein & Loveland, 1975). Under intervalschedules, reinforcer delivery is maximized byvarying response allocation across alternatives

(MacDonall, 2005). If responding favors oneresponse alternative over another under aninterval schedule, this would indicate a bias inresponding that is independent of the scheduleof reinforcement. This bias would not be aseasily observable during ratio schedules ofreinforcement. In the current application, aninterval schedule allowed us to identify poten-tial biases in responding that were independentof the reinforcement schedule. In addition, theapplication of a VI schedule mimics, to adegree, the integrity failures that could occur inthe natural environment.

DIFFERENTIAL REINFORCEMENT 571


4/21

In the natural environment, caregivers maynot always implement extinction proceduresaccurately. They also may fail to implementreinforcement procedures accurately (Shores et

al., 1993). Therefore, it may be important toidentify a therapeutic differential reinforcementprocedure that is effective despite intermittentreinforcement of both appropriate and problembehavior. The use of concurrent VI schedules inthe current experiments allowed the examina-tion of the effects of failure to withholdreinforcement following every problem behav-ior and failure to reinforce every appropriatebehavior in a highly controlled analogue setting.

We evaluated several manipulations thatcould be considered in the event that extinctioneither cannot or will not be implemented. InExperiments 1 to 3, we manipulated a singledimension of reinforcement such that reinforce-ment favored appropriate behavior along thelines of duration (Experiment 1), quality(Experiment 2), or delay (Experiment 3). InExperiment 4, we combined each of these

dimensions of reinforcement such that rein-forcement favored appropriate behavior.

GENERAL METHOD

Participants and Setting

Seven individuals with developmental disor-ders who engaged in severe problem behaviorparticipated. These were the first seven individ-uals who engaged in problem behavior sensitive

to socially mediated reinforcement (as identifiedvia functional analysis) and were admitted to anoutpatient clinic (Justin, Henry, Corey, Ken-neth, Lana) or referred for behavioral consulta-tion services at local elementary schools(George, Clark). (See Table 1 for each partic-ipants age, diagnosis, problem behavior, andappropriate behavior.) We selected the targetedappropriate behavior for each participant based

on the function of problem behavior. Forexample, if an individual engaged in problembehavior to access attention, we selected a mandfor attention as the appropriate behavior.

Targeted response forms were in the partici-pants repertoires, although the behavior typi-cally occurred at low rates.

Session rooms in the outpatient clinic (3 m

by 3 m) were equipped with a one-wayobservation window and sound monitoring.

We conducted sessions for George and Clark ina classroom at their elementary schools. Therooms for all participants contained materialsnecessary for a session (e.g., toys, task materi-als), and the elementary school classroomscontained materials such as posters and tables(George and Clark only). With the exception ofthe final experimental condition assessinggenerality, no other children were in the roomduring the analyses with George and Clark.

Trained clinicians served as therapists andconducted sessions 4 to 16 times per day, 5 daysper week. Sessions were 10 min in duration, andthere was a minimum 5-min break betweeneach session. We used a multielement designduring the functional analysis and a reversaldesign during all subsequent analyses.

Response Measurement andInterobserver Agreement

Observers were clinicians who had receivedtraining in behavioral observation and hadpreviously demonstrated high interobserveragreement scores (.90%) with trained observ-ers. Observers in the outpatient clinic sat behinda one-way observation window. Observers inthe school sat out of the direct line of sight of

the child. All observers collected data ondesktop or laptop computers that providedreal-time data and scored events as eitherfrequency (e.g., aggression, disruption, self-injury, and screaming) or duration (e.g.,delivery of attention, escape from instructions;see Table 1 for operational definitions ofbehavior). Observations were divided into 10-sbins, and observers scored the number (or

duration) of observed responses for each bin.The smaller number (or duration) of observedresponses within each bin was divided by thelarger number and converted to agreement



5/21

percentages for frequency measures (Bostow &Bailey, 1969). Agreement on the nonoccurrenceof behavior within any given bin was scored as100% agreement. The agreement scores for bins

were then averaged across the session.Two independent observers scored the target

responses simultaneously but independentlyduring a mean of 37% of functional analysissessions (range, 27% to 49%) and 29% ofexperimental analysis sessions (range, 25% to32%). We assessed interobserver agreement forproblem behavior (aggression, disruption, inap-propriate sexual behavior) and appropriate

behavior (compliance and mands) of all partic-ipants and for the therapists behavior, whichincluded therapist attention, delivery of tangibleitems, and escape from demands.

For Justin, mean agreement was 98% foraggression (range, 87% to 100%), 96% fordisruption (range, 85% to 100%), 100% forinappropriate sexual behavior, and 98% for

compliance (range, 86% to 100%). For Henry,mean agreement was 100% for aggression,99.9% for disruption (range, 99.7% to100%), and 97% for mands (range, 95% to99%). For Corey, mean agreement was 100%for aggression and disruption and 97% formands (range, 95% to 100%). For Kenneth,mean agreement was 98% for aggression (range,94% to 100%), 99% for disruption (range,

97% to 100%), and 99% for mands (range,95% to 100%). For Lana, mean agreement was99% for aggression (range, 99% to 100%) and100% for mands. For George, mean agreement

Table 1

Participants Characteristics

Name Age (years) Diagnosis Problem behavior Appropriate behavior

Justin 7 Attention deficithyperactivity disorderinstructional (ADHD)

Aggression: forcefully hitting, kicking,biting others body parts, pinching skinbetween fingers, scratching others with nails,forceful pushing, and head head buttingothers. Behavior drew blood or caused bruiseson his victims. Disruption: forcefullythrowing objects and hitting walls.Inappropriate sexual behavior: touchinghimself or the therapist in a sexual way bycontact of the hand to the torso, bottom, orgenitals.

Compliance with demands suchas fold the clothing or pickup the trash.

Henry 8 Autism Aggression: forcefully hitting and kicking others resulting in bruising his victims.Disruption: forcefully throwing objects.

Exchange of a picture card

Corey 9 Autism and ADHD Aggression: forcefully hitting, biting,spitting, and kicking resulting inbruising or bleeding of victims.Disruption: forcefully throwingobjects around room and at people,tearing paper materials.

Vocal request (May I have mytoy please?)

Kenneth 6 Autism Aggression: forcefully hitting, scratching,and pinching resulting in bleeding orbruising of victims. Disruption:throwing objects around room andat people.


Lana 4 Autism Aggression: forcefully hitting, kicking, andscratching resulting in bruising orbleeding in victims.

Sign language (sign for play)

George 10 Autism Aggression: forcefully hitting, kicking, andbiting resulting in bruising or bleedingvictims. Disruption: throwing objectsaround the room and at people.


Clark 12 Autism Aggression: hitting, kicking, and scratching resulting in bruising or bleeding of victims.

Vocal request (toy please)



6/21

was 99% (range, 98% to 100%) for aggression,99% for disruption (range, 98% to 100%), and93% for mands (range, 88% to 99%). Meaninterobserver agreement scores for 39% of all

sessions was 100% for therapist attention,99.9% for access to tangible items (range,99% to 100%), and 100% for escape frominstructions.

Stimulus Preference Assessment

We conducted a paired-stimulus preferenceassessment for each participant to identify ahierarchy of preferred items for use in thefunctional analysis (Fisher et al., 1992). In

addition, for those participants whose problembehavior was reinforced by tangible items(Corey, Lana, and Clark), a multiple-stimulus-without-replacement (MSWO) preference as-sessment (DeLeon & Iwata, 1996) was con-ducted immediately prior to each session of thetreatment analyses. We used informal caregiverinterviews to select items used in the preferenceassessments, and a minimum of six items were

included in the assessments.

Functional Analysis

We conducted functional analyses prior tothe treatment evaluation. Procedures weresimilar to those described by Iwata, Dorsey,Slifer, Bauman, and Richman (1982/1994)with one exception to the procedures forGeorge. His aggression was severe and primarilydirected toward therapists heads; therefore, a

blocking procedure was in place throughout thefunctional analysis for the safety of thetherapist. Blocking consisted of a therapistholding up his arm to prevent a hit fromdirectly contacting his head. During thefunctional analysis, four test conditions (atten-tion, tangible, escape, and ignore) were com-pared to a control condition (play) using amultielement design.

Figure 1 shows response rates of problembehavior during the functional analyses forJustin, Corey, Kenneth, and Henry. Wecollected data for aggression and disruption

separately and obtained similar results for eachtopography for all participants; therefore, bothtopographies were combined in these datapresentations. We obtained similar results for

inappropriate sexual behavior for Justin, whichwe combined with aggression and disruption.

Justin engaged in the highest rates ofproblem behavior in the escape condition.

Although the overall trend in the escapecondition is downward, inspection of the datashowed that he was becoming more efficient inescape behavior by responding only when thetherapist presented demands. Corey engaged inthe highest rates of problem behavior during thetangible condition. Kenneth engaged in thehighest rates of aggression and disruptionduring the attention and escape conditions.Henry displayed the highest rates of aggressionand disruption in the escape condition.

Figure 2 shows the results of the functionalanalyses for Lana, Clark, and George. Lana andClark displayed the highest rates of aggressionduring the tangible condition. George engaged

in the highest rates of aggression and disruptionduring the attention condition.

Baseline

During baseline and all subsequent condi-tions of Experiments 1 to 4, equal concurrentVI schedules of reinforcement (VI 20 s VI 20 s)were in place for both problem and appropriatebehavior. A random number generator selectedintervals between 1 s and 39 s, with a mean

interval length of 20 s, and the programmedintervals for each session were available on acomputer printout. A trained observer timedintervals using two timers set according to theprogrammed intervals. The first instance ofbehavior following availability of a reinforcerresulted in delivery of the reinforcer for 30 s(for an exception, see Experiment 1 involvingmanipulations of reinforcer duration). When

reinforcement was available for a response (i.e.,the interval elapsed) and the behavior occurred,the observer discreetly tapped on the one-waywindow from the observation room (clinic) or



7/21

briefly nodded his head (classroom) to promptthe therapist to reinforce a response. After 30 sof reinforcer access (or the pertinent durationvalue in Experiment 1), the therapist removedthe reinforcer and reset the timer for thatresponse. The VI clock for one response (e.g.,appropriate behavior) stopped while the partic-

ipant consumed the reinforcer for the otherresponse (e.g., problem behavior). The therapistreinforced responses regardless of the interval oftime since the last changeover from the other

response alternative. The reinforcer identifiedfor problem behavior in the functional analysisserved as the reinforcer for both responsesduring baseline. In Experiments 2 and 4, whichinvolved manipulations of quality, participantsreceived the same high-quality toy contingenton appropriate or problem behavior during

baseline.We conducted each baseline in the experi-ment as described but labeled them differentlyin order to highlight the dimensions of

Figure 1. Response rates during the functional analysis for Justin, Corey, Kenneth, and Henry.



8/21

reinforcement that varied across experiments.For example, in Experiment 1 we manipulatedduration of reinforcement, and baseline is

labeled 30-s/30-s dur to indicate that reinforce-ment was provided for 30 s (duration) followingproblem and appropriate behavior. In Experi-ment 2, we manipulated quality of reinforce-ment, and baseline is labeled 1 HQ/1 HQ toindicate that a high-quality reinforcer wasdelivered following appropriate and problembehavior. In Experiment 3, we manipulateddelay to reinforcement, and baseline is labeled

0-s/0-s delay. In Experiment 4 we manipulatedduration, quality, and delay in combination,and baseline is labeled 30-s dur 1 HQ 0-s delay/30-s dur 1 HQ 0-s delay.

EXPERIMENT 1: DURATION

Method

The purpose of Experiment 1 was to examine

whether we could obtain clinically acceptablechanges in behavior by providing a longer dura-tion of access to the reinforcer following appro-priate behavior and shorter duration of access tothe reinforcer following problem behavior.

30-s/10-s dur. Justin and Lana participated inthe 30-s/10-s dur condition. For Justin, appro-priate behavior produced a 30-s break frominstructions. Problem behavior produced a 10-s

break from instructions. For Lana, appropriatebehavior produced access to the most preferredtoy for 30 s, and problem behavior producedaccess to the same toy for 10 s.

Figure 2. Response rates during the functional analysis for Lana, Clark, and George.



9/21

45-s/5-s dur. Justin participated in the 45-s/5-s dur condition during which the duration ofreinforcement was more discrepant across

response alternatives. Appropriate behaviorproduced a 45-s break from instructions, andproblem behavior produced a 5-s break frominstructions.

Results and Discussion

Figure 3 shows the results for Justin andLana. For Justin, during the 30-s/30-s durbaseline condition, problem behavior occurred

at higher rates than appropriate behavior. In the30-s/10-s dur condition, there was a slightdecrease in the rate of problem behavior, andsome appropriate behavior occurred. Because

problem behavior still occurred at a higher ratethan appropriate behavior, we conducted the45-s/5-s dur condition. In the last five sessions

of this condition, problem behavior decreasedto low rates, and appropriate behavior in-creased. In a reversal to the 30-s/30-s durbaseline, problem behavior returned to levelshigher than appropriate behavior. In thesubsequent return to the 45-s/5-s dur condition,the favorable effects were replicated. Respond-ing stabilized in the last five sessions of thiscondition, with problem behavior remaining

low and appropriate behavior remaining high.In a reversal to the 30-s/30-s dur baseline,however, there was a failure to replicate previousbaseline levels of responding. Instead, low rates

Figure 3. Justins and Lanas response rates during the duration analysis for problem behavior and appropriate behavior.



10/21

of problem behavior and high rates of appro-priate behavior occurred.

During the 30-s/30-s dur baseline, Lanasproblem behavior occurred at higher rates than

appropriate behavior. During the 30-s/10-s durcondition, appropriate behavior occurred athigher rates, and problem behavior decreasedto zero. The effects of the 30-s/30-s dur baselineand the 30-s/10-s dur condition were replicatedin the final two conditions.

In summary, the duration analysis indicatedthat for both participants, the relative rates ofproblem behavior and appropriate behavior

were sensitive to the reinforcement durationavailable for each alternative in four of the fiveapplications in which duration of reinforcementwas unequal. This finding replicates thefindings of previous investigations on the effectsof reinforcement duration on choice responding(Catania, 1963; Lerman et al., 2002; Ten Eyck,1970).

There were several limitations to this exper-iment. For example, the participants did not

show sensitivity to the concurrent VI scheduleswhen both the rate and duration of reinforce-ment were equal. Under this arrangement, theparticipants would have collected all of theavailable reinforcers had they distributed theirresponding roughly equally between the tworesponse options. The failure to distributeresponding across responses indicates a biastoward problem behavior. Additional research

into this failure to show sensitivity to theconcurrent VI schedules is warranted but wasoutside the scope of this experiment.

With Justin, we were unable to recapturebaseline rates of problem and appropriatebehavior in our final reversal to the 30-s/30-sdur baseline. This failure to replicate previousrates of responding may be a result of his recenthistory with a condition in which reinforcement

favored appropriate behavior (i.e., the 5 s/45-sdur condition). Nevertheless, this lack ofreplication weakens the demonstration ofexperimental control with this participant. With

both participants, there was a gradual change inresponding in the condition that ultimatelyproduced a change favoring the alternativebehavior, which is not surprising given that

extinction was not in place. Responding underintermittent schedules of reinforcement can bemore resistant to change (Ferster & Skinner,1957).

EXPERIMENT 2: QUALITY

The purpose of Experiment 2 was to examinewhether we could obtain clinically acceptablechanges in behavior by providing a higher

quality reinforcer following appropriate behav-ior and lower quality reinforcer followingproblem behavior.

Method

Reinforcer assessment. We conducted a rein-forcer assessment using procedures described byPiazza et al. (1999) before conducting thequality analysis with Kenneth. The assessment

identified the relative efficacy of two reinforcers(i.e., praise and reprimands) in a concurrent-operants arrangement. During baseline, thetherapist stood in the middle of a room thatwas divided by painters tape and provided nosocial interaction; toy contact (e.g., playing withgreen or orange blocks on either side of thedivided room) and problem behavior resulted inno arranged consequences. Presession prompt-ing occurred prior to the beginning of the initial

contingent attention phase and the reversal(described below). During presession prompt-ing, the experimenter prompted Kenneth tomake contact with the green and orange toys.Prompted contact with green toys resulted inpraise (e.g., Good job, Kenneth, delivered ina high-pitched, loud voice with an excitedtone). Prompted contact with the orange toysresulted in reprimands (e.g., Dont play with

that, delivered in a deeper pitched, loud voicewith a harsh tone). Following presessionprompting, we implemented the contingentattention phase. The therapist stood in the



11/21

middle of a room divided by painters tape anddelivered the consequences to which Kennethhad been exposed in presession prompting. Thetherapist delivered continuous reprimands or

praise for the duration of toy contact andblocked attempts to play with two different-colored toys simultaneously.

During the second contingent attentionphase, we reversed the consequences associatedwith each color of toys such that green toys wereassociated with reprimands and orange toyswith praise. The different-colored toys werealways associated with a specific side of theroom, and the therapist ensured that theyremained on that side. Kenneth selected thecolored toy associated with praise on a mean of98% of all contingent attention sessions.

1 HQ/1 LQ. For Justin, problem behaviorproduced 30 s of escape with access to one low-quality tangible item identified in a presessionMSWO. Appropriate behavior produced 30 s ofescape with access to one high-quality tangibleitem identified in a presession MSWO. Al-

though the variable that maintained his prob-lem behavior was escape, we used disparatequality toys as a way of creating a qualitativedifference between the escape contingencies forappropriate and problem behavior.

For Kenneth, problem behavior producedreprimands (e.g., Dont do that, I really do notlike it, and you could end up hurtingsomeone), which the reinforcer assessment

identified as a less effective form of reinforce-ment than social praise. Appropriate behaviorproduced praise (e.g., Good job handing methe card; I really like it when you hand it to meso nicely.), which was identified as a moreeffective form of reinforcement in the reinforcerassessment.

3 HQ/1 LQ. For Justin and Kenneth,problem behavior did not decrease to therapeu-

tic levels in the 1 HQ/1 LQ condition. ForJustin, within-session analysis showed that assessions progressed during the 1 LQ/1 HQcondition, he stopped playing with the toy and

showed decreases in compliance, possibly due toreinforcer satiation. Unfortunately, we did nothave access to potentially higher quality toysthat Justin had requested (e.g., video game

systems). Given this limited access, we increasedthe number of preferred toys provided contin-gent on appropriate behavior as a way ofaddressing potential satiation with the toys.

We provided three toys selected most frequentlyin presession MSWO assessments. Therefore,for Justin, in the 3 HQ/1 LQ condition,appropriate behavior produced 30 s of escapewith access to three high-quality toys. Problembehavior produced 30 s of escape with access toone low-quality tangible item.

For Kenneth, anecdotal observations be-tween sessions showed that he frequentlyrequested physical attention in the forms ofhugs and tickles by guiding the therapists handsaround him or to his stomach. Based on thisobservation, we added physical attention to thesocial praise available following appropriatebehavior. Therefore, during the 3 HQ/1 LQ

condition, appropriate behavior producedpraise and the addition of physical attention(e.g., Good job handing me the card,hugs and tickles). Problem behavior producedreprimands.


During the 1 HQ/1 HQ baseline condition,Justin (Figure 4, top) engaged in higher rates ofproblem behavior than appropriate behavior. In

the 1 HQ/1 LQ condition, rates of problembehavior decreased, and appropriate behaviorincreased. However, toward the end of thephase, problem behavior increased, and appro-priate behavior decreased. Lower rates ofproblem behavior than appropriate behaviorwere obtained in the 3 HQ/1 LQ condition.During the subsequent 1 HQ/1 HQ baselinereversal, there was a failure to recapture previous

rates of problem and appropriate behavior.Instead, problem behavior occurred at a lowerrate than appropriate behavior. Despite this,problem behavior increased relative to what was



12/21

observed in the immediately preceding 3HQ/1 LQ condition. Problem behavior de-

creased, and appropriate behavior increased tohigh levels during the return to the 3 HQ/1 LQcondition.

Kenneth (Figure 4, bottom) engaged inhigher rates of problem behavior than appro-priate behavior in the 1 HQ/1 HQ baseline. Inthe 1 HQ/1 LQ condition, rates of problembehavior decreased, and appropriate behaviorincreased. During the last five sessions, re-

sponding shifted across response alternativesacross sessions. During a replication of 1 HQ/1HQ baseline, we observed high rates of problembehavior and relatively lower rates of appropri-

ate behavior. During a subsequent replication ofthe 1 HQ/1 LQ condition, slightly higher rates

of problem behavior than appropriate behaviorwere obtained, with responding again shiftingacross response alternatives across sessions. In areplication of 1 HQ/1 HQ baseline, high ratesof problem behavior and lower rates ofappropriate behavior were obtained. Followingthis replication, we conducted the 3 HQ/1 LQcondition, and problem behavior decreased torates lower than observed in previous conditions

and appropriate behavior increased to highrates. The effects of the 1 HQ/1 HQ baselineand the 3 HQ/1 LQ condition were replicatedin the final two conditions.

Figure 4. Response rates for Justin and Kenneth during the quality analysis for problem behavior andappropriate behavior.



13/21

In summary, results of the quality analysesindicated that for both participants, the relativerates of both problem behavior and appropriatebehavior were sensitive to the quality of

reinforcement available for each alternative.These results replicate the findings of previousinvestigations on the relative effects of quality ofreinforcement on choice responding (Conger &Killeen, 1974; Hoch et al., 2002; Martens &Houk, 1989; Neef et al., 1992; Piazza et al.,1997).

One drawback to this study was themanipulation of both magnitude and qualityof reinforcement with Justin. Given the cir-cumstances described above, a greater numberof higher quality toys were provided contingenton appropriate behavior relative to problembehavior prior to obtaining a consistent shift inresponse allocation.

As in Experiment 1, the failure to replicateprior rates of appropriate behavior in our finalreversal to the 1 HQ/1 HQ baseline weakenedexperimental control with Justin. Again, base-

line levels of behavior were not recaptured afteran intervening history in which the reinforce-ment quality and magnitude favored appropri-ate behavior.

EXPERIMENT 3: DELAY

Method

The purpose of Experiment 3 was to examinewhether we could produce clinically acceptable

changes in behavior by providing immediatereinforcement following appropriate behaviorand delayed reinforcement following problembehavior.

0-s/30-s delay. Corey and Henry participatedin the 0-s/30-s delay condition. For Corey,appropriate behavior produced 30-s immediateaccess to a high-quality toy (selected from apresession MSWO). Problem behavior pro-

duced 30-s access to the same high-quality toyafter a 30-s unsignaled delay. For Henry,appropriate behavior produced an immediate30-s break from instructions. Problem behavior

produced a 30-s break from instructions after a30-s unsignaled delay. With both participants,once a delay interval started, additional instanc-es of problem behavior did not reset the

interval. When problem behavior occurred,the data collector started a timer and signaledthe therapist to provide reinforcement when thetimer elapsed by a discreet tap on the one-waywindow. If a participant engaged in appropriatebehavior during the delay interval for problembehavior, the therapist immediately deliveredthe reinforcer for appropriate behavior (asprogrammed), and the delay clock for problembehavior temporarily stopped and then resumedafter the reinforcement interval for appropriatebehavior ended.

0-s/60-s delay. When the initial delay intervaldid not result in therapeutic decreases inproblem behavior for Corey, we altered thedelay interval such that problem behaviorproduced 30-s access to a high-quality toy(selected from a presession MSWO) after a 60-sunsignaled delay. Appropriate behavior contin-

ued to produce 30-s immediate access to thesame high-quality toy. For Henry, problembehavior produced a 30-s break from instruc-tions after a 60-s unsignaled delay, andappropriate behavior continued to produce animmediate 30-s break.


During the 0-s/0-s delay baseline, Corey(Figure 5, top) engaged in higher rates of

problem behavior than appropriate behavior.In the 0-s/30-s delay condition, problembehavior continued to occur at a higher ratethan appropriate behavior. Given this, the 0-s/60-s delay condition was implemented, and agradual decrease in problem behavior andincrease in appropriate behavior was obtained.During a reversal to the 0-s/0-s delay baseline,there was an increase in problem behavior and a

decrease in appropriate behavior. In the finalreversal to the 0-s/60-s delay condition, Coreybecame ill with strep throat. His caregivercontinued to bring him to the clinic and did not



14/21

inform us until after he began treatment. (Wehave indicated this period on the graph.)Following his illness, problem behavior ceased,

and appropriate behavior increased to high,steady rates.

During the 0-s/0-s delay baseline, Henry(Figure 5, bottom) engaged in higher rates ofproblem behavior than appropriate behavior. Inthe 0-s/30-s delay condition, Henry continuedto engage in a higher rate of problem behaviorthan appropriate behavior. In a reversal to 0-s/0-s delay baseline, there was a slight increase in

problem behavior from the previous conditionand a decrease in appropriate behavior. Duringthe 0-s/60-s delay condition, there was adecrease in problem behavior to zero rates and

an increase in appropriate behavior to steadyrates of two per minute (perfectly efficientresponding given 30-s access). These results

were replicated in the reversals to 0-s/0-s delaybaseline and 0-s/60-s delay condition.

In summary, results of the delay analysisindicate that the relative rates of problembehavior and appropriate behavior were sensi-tive to the delay to reinforcement followingeach alternative. These results replicate thefindings of previous investigations on the effectsof unsignaled delay to reinforcement (Sizemore

& Lattal, 1978; Vollmer, Borrero, Lalli, &Daniel, 1999; Williams, 1976). For example,Vollmer et al. showed that aggression occurredwhen it produced immediate but small rein-

Figure 5. Coreys and Henrys response rates during the delay analysis for problem behavior andappropriate behavior.



15/21

forcers even though mands produced largerreinforcers after an unsignaled delay. In theirstudy, participants displayed self-control whentherapists signaled the delay to reinforcement.

It is important to note that the programmeddelays were not necessarily those experienced bythe participant. The occurrence of problembehavior started a timer that, when elapsed,resulted in delivery of reinforcement. Additionalproblem behavior during the delay did not addto the delay in order to prevent extinction-likeconditions. It was therefore possible that prob-lem behavior occurred within the delay interval

and resulted in shorter delays to reinforcement.This rarely occurred with Henry. By contrast,Coreys problem behavior sometimes occurred inbursts or at high rates. In these cases, problembehavior was reinforced after delays shorter thanthe programmed 30 s or 60 s. Nevertheless, thedifferential delays to reinforcement followinginappropriate and appropriate behavior eventu-ally shifted allocation toward appropriate re-sponding. One way to address this potential

limitation would be to add a differentialreinforcement of other behavior (DRO) compo-nent with a resetting reinforcement interval. Theresetting feature would result in the occurrenceof problem behavior during the interval resettingthe interval and therefore delaying reinforce-ment. With high-rate problem behavior, thisDRO contingency would initially result in verylow rates of reinforcement, making the condition

similar to extinction. We did not add a DROcomponent in the current experiment becauseour aim was to evaluate treatments withoutextinction.

Another potential limitation to the currentexperiment was the possibility of adventitiousreinforcement of chains of problem andappropriate behavior. For example, whenappropriate behavior occurred during the delay

interval for problem behavior and the VIschedule indicated reinforcement was availablefor that response, there was immediate rein-forcement of appropriate behavior. This rein-

forcement could have strengthened a chain ofproblem and appropriate behavior. Althoughthis did not seem to be a concern in the currentexperiment, one way to control for this

limitation would be to add a changeover delay(COD). A COD allows a response to bereinforced only if a certain interval has passedsince the last changeover from the otherresponse alternative. The COD could preventadventitious reinforcement of problem andappropriate behavior and result in longerperiods of responding on a given alternativeand thus greater control by the relativereinforcement available for those alternatives(Catania, 1966).

EXPERIMENT 4: DURATION, QUALITY,AND DELAY

The purpose of Experiment 4 was to evaluatethe effects of delivering immediate, longerduration access to high-quality reinforcementfollowing appropriate behavior and delayed,

shorter duration access to low-quality reinforce-ment following problem behavior. We observedgradual treatment effects in the previousexperiments. This was to be expected, becauseboth types of responding were reinforced, but isnot an ideal clinical outcome. In addition,experimental control was not clear in several ofthe cases, and none of the experiments clearlydemonstrated how reinforcement that favoredappropriate behavior could be used in a

practical manner as a treatment for problembehavior. The focus of Experiment 4, therefore,was to combine all the variables and examinewhether clinically acceptable changes in behav-ior could be produced by making reinforcementfor appropriate behavior greater along severaldimensions. We also assessed the maintenanceand generality of treatment effects. George andClark participated in Experiment 4.

Method

Reinforcer assessment. Before conducting theexperimental analyses with George, we con-



16/21

ducted a reinforcer assessment using proceduressimilar to those described in Experiment 2. Wecompared the reinforcing efficacy of praise (e.g.,Good job, George) and physical contact (e.g.,

high fives, pats on the back) with reprimands(e.g., Dont do that) and physical contact(e.g., therapist using his hands to blockaggression from George for safety reasons).George allocated a mean of 96% of hisresponses to the colored toys that resulted inpraise and physical contact.

30-s dur HQ 0-s delay/5-s dur LQ 10-s delay.As in previous experiments, equal concurrent VIschedules of reinforcement (VI 20 s VI 20 s) werein place for both problem and appropriatebehavior throughout the experiment. ForGeorge, appropriate behavior immediately pro-duced 30 s of high-quality attention in the formof social praise and physical attention (e.g., highfives, pats on the back). Problem behaviorproduced 5 s of low-quality attention in theform of social disapproval and brief blocking ofaggression after a 10-s unsignaled delay. For

Clark, appropriate behavior produced 30 s ofimmediate access to a high-preference toy.Problem behavior produced 5 s of access to alow-preference toy after a 10-s unsignaled delay.The therapist timed delays to reinforcement inthe same manner as described in Experiment 3.

We assessed maintenance of treatment effectsand extended treatment across therapists withboth participants. Georges participation con-cluded with a 1-month follow-up to evaluate the

maintenance of treatment effects. His teacherconducted the final three sessions of thiscondition. Clarks participation concluded witha 2-month follow-up during which his teacherconducted sessions. Teachers received writtendescriptions of the protocol, one-on-one trainingwith modeling of the procedures, and feedbackafter each session regarding the accuracy of theirimplementation of the procedures.


During the 30-s dur 1 HQ 0-s delay/30-s dur1 HQ 0-s delay baseline, George (Figure 6, top)

engaged in higher rates of problem behaviorthan appropriate behavior. In the 30-s dur HQ0-s delay/5-s dur LQ 10-s delay condition, therewas a decrease in problem behavior and an

increase in appropriate behavior. In a reversal tobaseline, there was an increase in problembehavior and a decrease in appropriate behavior.In the final reversal to the 30-s dur HQ 0-sdelay/5-s dur LQ 10-s delay condition, therewas a further decrease in problem behavior andan increase in appropriate behavior. At the 1-month follow-up, no problem behavior oc-curred, and appropriate behavior remainedhigh.

During the 30-s dur 1 HQ 0-s delay/30-s dur1 HQ 0-s delay baseline, Clark (Figure 6,bottom) engaged in higher rates of problembehavior than appropriate behavior. In theinitial 30-s dur HQ 0-s delay/5-s dur LQ 10-sdelay condition, there was a decrease inproblem behavior and an increase in appropri-ate behavior. In a reversal to baseline, there wasan increase in problem behavior and a decrease

in appropriate behavior. In a reversal to the 30-sdur HQ 0-s delay/5-s dur LQ 10-s delaycondition, there was a further decrease inproblem behavior and an increase in appropri-ate behavior. At the 2-month follow-up, noproblem behavior occurred, and appropriatebehavior remained high.

In summary, results of the combined analysesindicate that for these participants the relative

rates of problem behavior and appropriatebehavior were sensitive to a combination ofthe quality, delay, and duration of reinforce-ment following each alternative. Compared tothe first three experiments, Experiment 4resulted in clear experimental control; therewere rapid changes in response allocation acrossconditions and consistent replications of re-sponding under previous conditions, despite the

fact that we did not include an extinctioncomponent.There were several limitations to this exper-

iment. We did not conduct within-subject



17/21

comparisons of manipulating single versus

multiple dimensions of reinforcement. Inaddition, the response blocking included inGeorges case limits conclusions regardingefficacy of treatments that do not includeextinction because response blocking mayfunction as either extinction or punishment(Lerman & Iwata, 1996). Unfortunately,Georges aggression tended to cause substantialharm to others and warranted the use of the

briefest sufficient block to prevent harm. Theblocking used during treatment was the same asthat used in the functional analysis. Theblocking response did not serve to suppress

aggression in the functional analysis, and it is

doubtful that it exerted any such suppressiveeffects during the intervention. We did attemptto control for the addition of physical contactrequired following problem behavior by addingphysical contact contingent on appropriatebehavior.

A potential strength of this investigation wasthat we assessed both maintenance and gener-ality of the procedures in a 1-month follow-up,

with Georges and Clarks teachers serving astherapists in several of the sessions. Georgesteacher reported that he had a history ofattacking peers, making his behavior too severe

Figure 6. Georges and Clarks response rates for problem behavior and appropriate behavior.



18/21

to ignore. His teacher also indicated that thepresence of four other children in the roomlimited the amount of attention she coulddeliver following appropriate behavior. Clarks

behavior was so severe that prior to thisinvestigation, he had been moved to a classroomin which he was the only student; he returned toa small-size (four peers) classroom followingthis investigation. The current procedure iden-tified an effective treatment in which teachersdelivered a relatively long duration of high-quality reinforcement immediately followingsome appropriate behavior and brief, low-quality reinforcement after a short delayfollowing some problem behavior. Our specificrecommendation to both teachers was to followthe procedures in Experiment 4 to the best oftheir abilities, with the caveat that each shouldimmediately intervene for aggression that wasdirected toward peers or was likely to causesevere harm.

GENERAL DISCUSSION

The current experiments attempted toidentify differential reinforcement proceduresthat were effective without extinction bymanipulating several dimensions of reinforce-ment. We sought to extend prior research thatfocused solely on multiply maintained problembehavior (Piazza et al., 1997) and examinedonly single manipulations of reinforcement(Lalli & Casey, 1996; Piazza et al.). The

present studies showed the effectiveness ofDRA that provided some combination of moreimmediate, longer duration, or higher qualityof reinforcement for appropriate behaviorrelative to reinforcement for problem behavior.In cases in which extinction is not feasible, thecurrent studies offer a method of decreasingproblem behavior and increasing appropriatebehavior without the use of extinction. For

example, if problem behavior is so severe (e.g.,severe aggression, head banging on hardsurfaces) that it is not possible to withhold oreven delay reinforcement, it may be possible to

manipulate other parameters of reinforcementsuch as duration and quality to favor appro-priate behavior. If attention maintains problembehavior in the form of severe self-injury, for

example, problem behavior could result in briefsocial attention and appropriate behavior couldresult in a longer duration of attention in theform of praise, smiles, conversation, laughter,and physical attention such as hugs andtickling.

One potential contribution of the currentexperiments was procedural. The use of inter-mittent schedules of reinforcement in the

treatment of problem behavior had severalbenefits. For example, these schedules likelymimic to a degree the schedules of reinforce-ment in the natural environment. It is unlikelythat at home or school, for example, eachinstance of behavior produces reinforcement. Itis likely, however, that variable amounts ofappropriate and problem behavior are rein-forced or that varying amounts of time passbetween reinforced episodes. Further, concur-rent VI arrangements allow comparisons to andtranslations from experimental work on thematching law.

One limitation of these experiments is thebrevity and varying length of the conditions.In a laboratory, it may be possible toconduct conditions until meeting a stabilitycriterion (e.g., a difference of less than 5%between data points); however, in a clinical

setting, it is not always possible to bringeach condition to stability before exposingbehavior to another condition (i.e., Coreyand Kenneth).

A second potential limitation to the currentexperiment is the difference in obtained versusprogrammed schedules of reinforcement. VIschedules of reinforcement involve delivery of areinforcer for the first response after an average

length of time has passed since the lastreinforcer. Participants did not always respondimmediately after the required length of timeelapsed, resulting at times in a less dense



19/21

reinforcement schedule than programmed. Thedifferences in obtained versus programmedreinforcement schedules were neither large norconsistent, however.

Our study suggests several areas for futureresearch. These experiments included concur-rent schedules of VI 20-s reinforcement forproblem and appropriate behavior. Futureresearch may involve similar analyses usingconcurrent-schedules arrangements based onnaturalistic observations. The extent to whichrelative response allocation is similar underdescriptive and experimental arrangements may

suggest values of reinforcement parameters thatmay increase both the acceptability and integ-rity of treatment implementation by caregivers.For example, researchers could conduct descrip-tive analyses (Bijou, Peterson, & Ault, 1968)with caregivers and analyze the results usingreinforcers identified in a functional analysiswith procedures similar to those described byBorrero, Vollmer, Borrero, and Bourret (2005).If descriptive analysis data show that problembehavior is reinforced on average every 15 s andappropriate behavior is reinforced on averageevery 30 s, treatment might involve reinforcingappropriate behavior every 15 s and problembehavior every 30 s.

Investigations similar to the current exper-iments could further explore the dimensions ofquality, duration, and delay with moreparticipants and with additional values of

these dimensions. In addition, future research-ers could investigate the effect of concurrentmanipulations of the dimensions of reinforce-ment as treatment for problem behavior. Forexample, when it is not possible to withholdreinforcement for problem behavior, it may bethat the rate of reinforcement can continue tofavor problem behavior if several dimensionsof reinforcement, such as magnitude, quality,

and duration, favor appropriate behavior. Thisarea of research may result in the developmentof more practical and widely adopted inter-ventions for problem behavior.

REFERENCES

Baum, W. M. (1974). Choice in free-ranging wildpigeons. Science, 185, 7879.

Bijou, S. W., Peterson, R. F., & Ault, M. H. (1968).A method to integrate descriptive and experimentalfield studies at the level of data and empiricalconcepts. Journal of Applied Behavior Analysis, 1,175191.

Borrero, C. S. W., Vollmer, T. R., Borrero, J. C., &Bourret, J. (2005). A method for evaluating thedimensions of reinforcement in parent-child interac-tions. Research in Developmental Disabilities, 26,577592.

Bostow, D. E., & Bailey, J. B. (1969). Modification ofsevere disruptive and aggressive behavior using brieftimeout and reinforcement procedures. Journal of Applied Behavior Analysis, 2, 3137.

Catania, A. C. (1963). Concurrent performances: Abaseline for the study of reinforcement magnitude.Journal of the Experimental Analysis of Behavior, 6,299300.

Catania, A. C. (1966). Concurrent operants. In W. K.Honig (Ed.), Operant behavior: Areas of research andapplication (pp. 213270). Englewood Cliffs, NJ:Prentice Hall.

Conger, R., & Killeen, P. (1974). Use of concurrentoperants in small group research. Pacific SociologicalReview, 17, 399416.

Cooper, J. O., Heron, T. E., & Heward, W. L. (2007).

Applied behavior analysis (2nd ed.). Upper SaddleRiver, NJ: Prentice Hall.

Deitz, D. E. D., & Repp, A. C. (1983). Reducingbehavior through reinforcement. Exceptional Educa-tion Quarterly, 3, 3446.

DeLeon, I. G., & Iwata, B. A. (1996). Evaluation of amultiple-stimulus presentation format for assessingreinforcer preferences. Journal of Applied BehaviorAnalysis, 29, 519532.

Dwyer-Moore, K. J., & Dixon, M. R. (2007). Functionalanalysis and treatment of problem behavior of elderlyadults in long-term care. Journal of Applied Behavior

Analysis, 40, 679683.Ferster, C. B., & Skinner, B. F. (1957). Schedulesof reinforcement. New York: Appleton-Century-Crofts.

Fisher, W., Piazza, C. C., Bowman, L. G., Hagopian, L.P., Owens, J. C., & Slevin, I. (1992). A comparisonof two approaches for identifying reinforcers forperson with severe and profound disabilities. Journalof Applied Behavior Analysis, 25, 491498.

Fisher, W., Piazza, C. C., Cataldo, M. F., Harrell, R.,Jefferson, G., & Conner, R. (1993). Functionalcommunication training with and without extinctionand punishment. Journal of Applied Behavior Analysis,26, 2336.

Hagopian, L. P., Fisher, W. W., Sullivan, M. T.,Acquisto, J., & LeBlanc, L. A. (1998). Effectivenessof functional communication training with and



20/21

without extinction and punishment: A summary of 21inpatient cases. Journal of Applied Behavior Analysis,31, 211235.

Herrnstein, R. J. (1961). Relative and absolute strength ofresponse as a function of frequency of reinforcement.Journal of the Experimental Analysis of Behavior, 4,267272.

Herrnstein, R. J., & Loveland, D. H. (1975). Maximizingand matching on concurrent ratio schedules. Journalof the Experimental Analysis of Behavior, 24, 107116.

Hoch, H., McComas, J. J., Johnson, L., Faranda, N., &Guenther, S. L. (2002). The effects of magnitude andquality of reinforcement on choice responding duringplay activities. Journal of Applied Behavior Analysis,35, 171181.

Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., &Richman, G. S. (1994). Toward a functional analysisof self-injury. Journal of Applied Behavior Analysis, 27,197209. (Reprinted from Analysis and Intervention inDevelopmental Disabilities, 2, 320, 1982)

Iwata, B. A., Pace, G. M., Cowdery, G. E., &Miltenberger, R. G. (1994). What makes extinctionwork: An analysis of procedural form and function.Journal of Applied Behavior Analysis, 27, 131144.

Lalli, J. S., & Casey, S. D. (1996). Treatment of multiplycontrolled problem behavior. Journal of AppliedBehavior Analysis, 29, 391395.

Lalli, J. S., Vollmer, T. R., Progar, P. R., Wright, C.,Borrero, J., Daniel, D., et al. (1999). Competitionbetween positive and negative reinforcement in the

treatment of escape behavior. Journal of AppliedBehavior Analysis, 32, 285296.

Lerman, D. C., & Iwata, B. A. (1996). A methodology fordistinguishing between extinction and punishmenteffects of response blocking. Journal of AppliedBehavior Analysis, 29, 231233.

Lerman, D. C., Kelley, M. E., Vorndran, C. M., Kuhn, S.A. C., & LaRue, R. H. J. (2002). Reinforcementmagnitude and responding during treatment withdifferential reinforcement. Journal of Applied BehaviorAnalysis, 35, 2948.

MacDonall, J. S. (2005). Earning and obtaining reinforc-

ers under concurrent interval scheduling. Journal of the Experimental Analysis of Behavior, 84, 167183.

Mace, F. C., McCurdy, B., & Quigley, E. A. (1990). Acollateral effect of reward predicted by matchingtheory. Journal of Applied Behavior Analysis, 23,197205.

Mace, F. C., Neef, N. A., Shade, D., & Mauro, B. C.(1994). Limited matching on concurrent-schedulereinforcement of academic behavior. Journal of Applied Behavior Analysis, 24, 719732.

Mace, F. C., & Roberts, M. L. (1993). Factors affectingselection of behavioral interventions. In J. Reichle &

D. P. Wacker (Eds.), Communicative alternatives tochallenging behavior: Integrating functional assessmentand intervention strategies (pp. 113134). Baltimore:Brookes.

Martens, B. K., & Houk, J. L. (1989). The application ofHerrnsteins law of effect to disruptive and on-task behavior of a retarded adolescent girl. Journalof the Experimental Analysis of Behavior, 51, 1727.

McCord, B. E., Thomson, R. J., & Iwata, B. A. (2001).Functional analysis and treatment of self-injuryassociated with transitions. Journal of Applied BehaviorAnalysis, 34, 195210.

Neef, N. A., Mace, F. C., & Shade, D. (1993).Impulsivity in students with serious emotionaldisturbance: The interactive effects of reinforcer rate,delay, and quality. Journal of Applied BehaviorAnalysis, 26, 3752.

Neef, N. A., Mace, F. C., Shea, M. C., & Shade, D.(1992). Effects of reinforcer rate and reinforcerquality on time allocation: Extensions of matchingtheory to educational settings. Journal of AppliedBehavior Analysis, 25, 691699.

Neef, N. A., Shade, D., & Miller, M. S. (1994). As-sessing influential dimensions of reinforcers onchoice in students with serious emotional distur-bance. Journal of Applied Behavior Analysis, 27,575583.

Piazza, C. C., Bowman, L. G., Contrucci, S. A.,Delia, M. D., Adelinis, J. D., & Goh, H. (1999).An evaluation of the properties of attention asreinforcement for destructive and appropriate be-havior. Journal of Applied Behavior Analysis, 32,437449.

Piazza, C. C., Fisher, W. W., Hanley, G. P., Remick,M. L., Contrucci, S. A., & Aitken, T. L. (1997).The use of positive and negative reinforcement inthe treatment of escape-maintained destructive be-havior. Journal of Applied Behavior Analysis, 30,279298.

Shores, R. E., Jack, S. L., Gunter, P. L., Ellis, D. N.,DeBriere, T. J., & Wehby, J. H. (1993). Classroominteractions of children with behavior disorders.Journal of Emotional and Behavioral Disorders, 1,2739.

Sizemore, O. J., & Lattal, K. A. (1978). Unsignalled delay

of reinforcement in variable-interval schedules. Jour-nal of the Experimental Analysis of Behavior, 30,169175.

Ten Eyck, R. L., Jr. (1970). Effects of rate of reinforce-ment-time upon concurrent operant performance.Journal of the Experimental Analysis of Behavior, 14,269274.

Volkert, V. M., Lerman, D. C., Call, N. A., & Trosclair-Lasserre, N. (2009). An evaluation of resurgenceduring treatment with functional communicationtraining. Journal of Applied Behavior Analysis, 42,145160.

Vollmer, T. R., Borrero, J. C., Lalli, J. S., & Daniel, D.(1999). Evaluating self-control and impulsivity inchildren with severe behavior disorders. Journal of Applied Behavior Analysis, 32, 451466.



21/21

Vollmer, T. R., & Iwata, B. A. (1992). Differentialreinforcement as treatment for behavior disorders:Procedural and functional variations. Research inDevelopmental Disabilities, 13, 393417.

Vollmer, T. R., Roane, H. S., Ringdahl, J. E., & Marcus,B. A. (1999). Evaluating treatment challenges withdifferential reinforcement of alternative behavior.Journal of Applied Behavior Analysis, 32, 923.

Williams, B. A. (1976). The effects of unsignalled delayedreinforcement. Journal of the Experimental Analysis ofBehavior, 26, 441449.

Worsdell, A. S., Iwata, B. A., Hanley, G. P., Thompson,R. T., & Kahng, S. W. (2000). Effects of continuousand intermittent reinforcement for problem behaviorduring functional communication training. Journal of Applied Behavior Analysis, 33, 167179.

Received December 30, 2008Final acceptance November 18, 2009Action Editor, Rachel Thompson


Athens y Vollmer, Refuerzo Diferencial

Documents

Transcript of Athens y Vollmer, Refuerzo Diferencial