manipulating establishing operations to verify … · manipulating establishing operations to...

MANIPULATING ESTABLISHING OPERATIONS TO VERIFY ANDESTABLISH STIMULUS CONTROL DURING MAND TRAINING

ANIBAL GUTIERREZ, JR.

UNIVERSITY OF MIAMI

TIMOTHY R. VOLLMER AND CLAUDIA L. DOZIER

UNIVERSITY OF FLORIDA

JOHN C. BORRERO

UNIVERSITY OF THE PACIFIC

JOHN T. RAPP

ST. CLOUD STATE UNIVERSITY

JASON C. BOURRET

NEW ENGLAND CENTER FOR CHILDREN

AND

DANA GADAIRE

UNIVERSITY OF FLORIDA

Acquisition of verbal behavior is a major goal of interventions for children with developmentaldisabilities. We evaluated the effectiveness of manipulation of an establishing operation forfunctional discriminated mands. Four individuals with developmental disabilities participated ina training procedure designed to teach two separate mands for two separate preferred items.Participants were taught to mand using picture cards. Following training, the manipulation ofthe establishing operation was used to assess and establish discriminated manding. Thismanipulation involved providing free access to one of the preferred items, such that there shouldbe no motivation to ask for it, while motivation to ask for the other item remained in place.Three of the 4 participants acquired discriminated manding using topographically similarresponses (picture cards). One participant did not acquire a discriminated mand untiltopographically distinct mands were taught (vocal and picture card). Results suggest thatdiscrimination training is not necessarily sufficient to teach discriminated manding when morethan one picture card showing preferred items is used. In addition, manipulation of theestablishing operation served as an appropriate assessment tool for the verification ofdiscriminated manding as well as a possible training tool to establish discriminated manding.

DESCRIPTORS: autism, picture cards, discrimination training

_______________________________________________________________________________

The acquisition of language for children withdevelopmental disabilities is a major goal oftraining programs. A large number of young

children with autism or related disabilitiesenter school programs without speech or othercommunicative behavior (Bondy & Frost,

This research partially fulfilled dissertation requirementsfor the first author. We thank Maureen Conroy, Brian A.Iwata, Scott A. Miller, and Henry S. Pennypacker for theircontributions to this article.

Address correspondence to Anibal Gutierrez, Universityof Miami, 5665 Ponce de Leon Boulevard, Miami, Florida33124 (e-mail: [email protected]).

doi: 10.1901/jaba.2007.645–658

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2007, 40, 645–658 NUMBER 4 (WINTER 2007)

645

1994a). In the population at large, languagedelays are common for children under the ageof 3 years (10% to 15%), and by school age it isestimated that 3% to 7% experience a languagedisorder (American Psychiatric Association,2000). For children with developmental dis-abilities to learn to gain access to desired itemsappropriately, communicate their needs anddesires, request information, obtain others’attention, and generally control their environ-ments, it is important that they acquirea functional form of communication (McCoy& Buckhalt, 1990). In addition, developmentof functional communication skills has alsobeen shown to reduce problem behavior byindividuals with developmental disabilities(Carr & Durand, 1985; Richman, Wacker, &Winborn, 2001; Winborn, Wacker, Richman,Asmus, & Geier, 2002).

The use of a picture-card exchange system(Bondy & Frost, 1993, 1994b; Liddle, 2001) isa widely used intervention for children withcommunication deficits (Charlop-Christy, Car-penter, Le, LeBlanc, & Kellet, 2002), and therehave been systematic programs designed withthese techniques (e.g., Bondy & Frost, 1993,1994b). Several published empirical studieshave focused on procedures designed to teachcommunication using picture cards (Bondy &Frost, 1993; Liddle, 2001; Schwartz, Garfinkle,& Bauer, 1998). Although establishing discrim-inated manding was not the sole purpose ofthese studies, each incorporated procedures toteach mands similar to those described byBondy and Frost (1994b).

Bondy and Frost (1993) reported on thesuccessful implementation of picture-card mandtraining for 74 students with communicationdelays. Of those students, 18 learned to makediscriminations from a group of pictures. Bondyand Frost (1994a) reported that of the 85 childrenwho had been taught to use picture cards at theDelaware Autism Project, over 95% learned touse two or more pictures to request itemsfollowing the discrimination training phase.

Schwartz et al. (1998) taught 31 childrenwith disabilities to request items using picturecards. Results showed that children completedthe discrimination phase after 7 months oftraining. In a similar study, Liddle (2001)taught 21 children with language delays to usepicture cards to request items in their class-room. Results showed that 16 of the 21 childrencompleted the discrimination training phase, inwhich they were taught to request preferreditems.

Although the literature supports the use ofpicture cards to teach mands (Bondy & Frost,1993, 1994b; Liddle, 2001; Schwartz et al.,1998), an empirical question remains regardingthe effectiveness of training procedures toproduce appropriate or discriminated mandingfor preferred items. At times, the success ofmand training is evaluated based on theparticipant’s ability to make the desired com-municative response (e.g., hand a picture card).In other words, if a child hands a ‘‘cookie’’ cardto the therapist and then gains access to thecookie, it is presumed that the child appropri-ately manded for a cookie. However, on closerexamination, the appropriateness of the indi-vidual’s mand cannot be entirely presumedbased on his or her response. Rather, it remainsunclear if the child is actually requestinga cookie or another item but accepts andconsumes a cookie because it is a suitablereinforcer. It is also difficult to determine theappropriateness of the mand response becausethe response of handing a card, in general, hasbeen sufficiently reinforced with access topreferred items such that an individual mayhave learned to hand any card to gain access toreinforcement.

Because picture-card training involves teach-ing individuals to hand picture cards, thetopography of the response is the same fromone occurrence of the behavior to another. Inother words, regardless of which item theindividual requested, the form of the responseis the same (i.e., handing a picture card). This is

646 ANIBAL GUTIERREZ, JR., et al.

contrasted with other forms of communicationtraining that involve teaching topographicallydifferent responses that correspond to each item(e.g., sign language, vocal responding). Thisdistinction between response forms is referredto as topography-based and selection-basedverbal behavior (Michael, 1985). Michaelargued that responding using picture cards isa form of selection-based verbal behavior, andas such, the individual must learn to makea conditional discrimination to mand for anitem. The individual must make a discrimina-tion from among the visual stimuli that is notrequired when using topography-based verbalbehavior. Michael suggested that this additionaldiscrimination requirement may influence fac-tors such as ease of acquisition and susceptibilityto interference during training. Studies evaluat-ing the two types of verbal behavior have shownthat topography-based verbal behavior can beacquired in fewer training trials and results inmore accurate performance than selection-basedverbal behavior (Sundberg & Sundberg, 1990;Wraikat, Sundberg, & Michael, 1991), suggest-ing that discriminating between picture cardsmay be difficult skill for some individuals toacquire.

Bondy and Frost outlined methods forteaching mands using picture cards that includea correspondence check during the discrimina-tion phase designed to detect and correctdiscrimination problems (Bondy & Frost,1994b, 2001, 2002; Frost & Bondy, 1994).Correspondence checks involve placing twoitems as well as the two corresponding picturecards in front of the individual. When theindividual hands the therapist a card, thetherapist allows the individual to pick up theitem he or she has requested. If the individualselects the appropriate item (i.e., the item thatcorresponds to the selected picture card), thetherapist allows the individual to consume orengage with the item. If the individual attemptsto select an item that does not correspond to theselected picture card, the therapist corrects the

individual and prompts him or her to select thepicture card corresponding to the selected item.

The correspondence check (Bondy & Frost,1994b) is advantageous because it is practicaland allows systematic correction if errors aremade. It may be limited, however, because itsutility does not include direct assessment of theestablishing operation (EO) that evokes themand. According to Skinner (1957) a mand isunder the control of relevant conditions ofdeprivation. Thus, to completely assess theoccurrence of a mand, it may be necessary toinclude an assessment of the EO that evokes theverbal response.

Interpreting the correspondence check re-quires the inference that the EO that evoked theselection of a particular reinforcer was the sameEO that evoked the verbal response. However,the correspondence check does not involvea direct manipulation of the EO prior to theresponse. It may be important to test moredirectly whether the response emitted is func-tionally a mand. It may also be necessary toevaluate the extent to which the selection ofa specific picture card is a discriminated re-sponse to assess the effectiveness of mandtraining using picture cards. If individuals donot engage in a discriminated response withpicture cards, then training has not resulted ina functional, discriminated mand response.

The purpose of the current study was toevaluate the extent to which mand trainingusing picture cards will result in discriminatedmands in the context of two picture cards andto determine the extent to which manipulatingEOs will function to verify the existence of andestablish discriminated manding. Manipulationof the EO may have an advantage over thecorrespondence check, at least in some cases,because it directly manipulates the controllingvariables for mands. This procedure moredirectly tests whether the response emitted isfunctionally a mand by manipulating priorcontrolling variables. Although the relativemerits of picture-card systems can be debated,

ESTABLISHING OPERATIONS 647

additional research is warranted due to wide-spread use of these systems.

METHOD

Participants

Four individuals who did not ordinarilyrequest items either vocally or nonvocallyparticipated in the study. Mario was a 13-year-old boy who had been diagnosed withmental retardation (although not formallydiagnosed as autistic, his behavior was charac-teristic of autism), Millie was a 4-year-old girl,Will was a 5-year-old boy, and Malcolm wasa 6-year-old boy. All had been diagnosed withautism. Participants had minimal exposure topicture cards prior to the study. Picture cardshad been placed in some classrooms at theirschools (e.g., picture card for bathroom outsidethe bathroom), but they were not used explicitlyto teach mands. Formal training of a picture-card communication system had not taken placefor any participant prior to the study. Teachersor staff referred all participants following anannouncement at the school that the study wasunderway to teach and evaluate communicationusing picture cards. These were the first 4individuals referred whose parents agreed togive informed consent.

Setting

All sessions were conducted at the partici-pant’s school. Blocks of sessions lasted forapproximately 30 min and were conducted 3 to5 days per week. During sessions, each partic-ipant was seated at a table across from thetherapist and data collectors. Sessions wereconducted in a small room reserved for thestudy.

Except for the preference assessment, sessionsconsisted of 10 trials for Mario, Will, andMalcolm and 10 to 15 trials for Millie. Duringeach trial the participant was presented with theopportunity to emit a mand and receive accessto preferred items. After the participant wasallowed access to items for approximately 30 s

or to consume the food items, the card positionwas counterbalanced quasirandomly in an arrayof two and the cards were presented to theparticipant to start the next trial (specificprocedures varied by condition; see below).

Data Collection and Response Definition

Trained graduate and undergraduate obser-vers recorded correct and incorrect responses.During training, correct responses were definedas handing the therapist the picture card.Incorrect responses were defined as handingthe therapist the distracter card. Duringmanipulation of the EO (Michael, 1982)(explained below), observers recorded the cardthat was handed to the therapist. During theEO manipulation with topographically differentresponses (explained below), observers recordedthe response emitted by the participant, whichcould be handing a picture card or emittinga vocal response.

Interobserver Agreement

During all phases, two trained observerscollected data simultaneously and independent-ly during 34% of the sessions. Interobserveragreement was calculated by dividing thenumber of agreements by the number ofagreements plus disagreements in each sessionand multiplying by 100%. An agreement wasdefined as a trial in which both observersrecorded the same occurrence of behavior.Mean agreement for all sessions was 99%(range, 80% to 100%).

Materials

Two picture cards and two distracter cardswere used during training. The picture cardscorresponded to preferred items that wereidentified for each participant during a prefer-ence assessment. Distracter cards displayed an Xor an O.

Procedure

Participation in each phase of the study wasdetermined by participant performance in the


prior phase. As participants completed eachphase, they either moved on to the next phase ofmand training or moved to a supplementaltraining phase. Figure 1 shows the order ofphases.

Phase 1: Preference assessment. The purpose ofthis phase was to identify items to be used asreinforcers. A free-operant preference assess-ment (Roane, Vollmer, Ringdahl, & Marcus,1998) was conducted to identify highly pre-ferred items for each participant. Each partic-ipant was presented with an array of food andleisure items during one or two 5-min sessions.

Separate assessments were conducted for Willand Millie to identify preferred items oractivities as well as consumable items. Thetwo separate preference assessments were con-ducted to avoid a possible displacement ofnonconsumable items by consumable items(DeLeon, Iwata, & Roscoe, 1997). Datacollectors recorded the amount of time theparticipant spent interacting with each item.Items with which the participant spent the mosttime were chosen as preferred items to be usedduring mand training.

Phase 2: Single card. The purpose of thisphase was to establish a card-handing repertoire.The participant was seated with a single picturecard in front of him or her. Handing thetherapist the picture card resulted in access tothe item for approximately 30 s or (when foodor drink was the preferred item) a small amountof the consumable item. Participants weretaught to use picture cards to mand forpreferred items using a four-step sequence (5-spause, verbal instruction, physical model,physical guidance).

Phase 3: Distracter-card probe. We conducteda probe for all participants in which a picturecard and a distracter card were presented. Thepurpose of this phase was to evaluate whetherparticipants were more likely to select a picturecard, as a point of comparison to a subsequentdifferential reinforcement phase. The partici-pant was seated with the two cards in front ofhim or her. Responding using either cardresulted in access to the preferred item. Afterthe participant responded, the item was de-livered and both cards were removed. Access tothe preferred item lasted for approximately 30 s(in the case of consumable items, a smallamount was delivered). After each trial, theposition of the cards was reassigned quasiran-domly, the cards were again presented to theparticipant, and the next trial began.

Phase 4: Differential reinforcement (mandtraining). The purpose of this phase was todifferentially reinforce responses using picture

Figure 1. Flow chart depicting the order of phasesduring the study.


cards, to establish a simple discriminationbetween the target card and distracter cards orat least to provide a history of reinforcement forthe picture-card response. There was no in-tention to show acquisition, because in somecases there was a bias towards selecting the cardwith the picture (see the results). Duringdifferential reinforcement, only responses usingthe picture card resulted in reinforcement, andresponses using the distracter card resulted inextinction. The experimental design was a con-current-schedules design, in which experimentalcontrol is demonstrated by differential respond-ing across choice alternatives (Sidman, 1960).During these sessions, the participant was seatedwith the two cards placed in front of him or her.The item itself was present but out of reach.Requests using the picture card resulted inaccess to the item for approximately 30 s (ora small amount of the edible item), and bothcards were removed. Requests using the dis-tracter card resulted in extinction (and a brieftime-out), during which the therapist removedthe cards and turned away from the participantfor approximately 5 s. After each trial, theposition of the cards was reassigned quasiran-domly and the trial was restarted. Participantswere taught to use picture cards to mand fortwo different preferred items. Participants whopassed this phase (at least two sessions with 80%correct) were deemed to have completed theinitial discrimination screening and had dem-onstrated performance of the discriminationskills necessary to move to Phase 5.

Phase 5: Manipulation of the EO. Thepurpose of this phase was to evaluate theeffectiveness of the initial mand training toestablish complex manding in the presence oftwo picture cards, each corresponding to pre-ferred items, as well as to serve as an additionaltraining phase to establish discriminated mands.All 4 participants participated in this phase. TheEO manipulation phase differed from thedifferential reinforcement phase in two respects.First, the two previously trained picture cards

were placed in front of the participant, andthere were no distracter cards present. This wasthe first time the participants were given bothpicture cards for which handing had beendifferentially reinforced. Second, the participanthad free access to one of the items he or she hadbeen previously taught to mand for duringdifferential reinforcement. The EO phase wasconducted using a combination of a concurrent-schedules and multielement design (Sidman,1960) in which we compared the effects of theEO (EO present vs. EO absent) on manding.During EO manipulation sessions, the partic-ipant was allowed free access to one of thepreviously taught items, and access to a seconditem was restricted. For example, if a participanthad been was previously taught to mand for theradio and a drink, during this phase theparticipant was allowed free access to the radio(radio EO absent) and access to the drink wasrestricted (drink EO present) or vice versa.Sessions in which one EO was present or absentwere alternated. Also, requests using the picturecard corresponding to the restricted itemresulted in access to that item for approximately30 s (or a small amount of the consumableitem), and both cards were removed. Requestsusing the picture card corresponding to the itemto which the participant already had accessresulted in the therapist removing the cards forapproximately 30 s (but the participant main-tained access to the item). After each trial theposition of the cards was reassigned quasiran-domly, and a new trial was started.

Phase 6a: Topographically different responsetraining (vocal and card). The purpose of thisphase was to develop two distinct mand forms(vocal and card). Only Millie participated inthis phase because she was the only participantwho showed no sensitivity to the EO manip-ulation in Phase 5. It is possible that thetopographical similarity between handing re-sponses may have impeded discriminationduring the EO manipulation (Michael, 1985).One picture card and two distracter cards were


used during the training phase. The picture cardcorresponded to a preferred item, and distractercards displayed an X or an O. The participantwas also presented with vocal mand trainingdesigned to produce differential respondingduring the EO phase. The participant wastaught to request an item using a picture cardand was taught to request a second item bysaying the word corresponding to the item (e.g.,saying ‘‘chips’’). Millie was taught to respondvocally by first presenting a vocal prompt (e.g.,‘‘say ‘chips’’’) and delivering the reinforcer(chips) following the imitative response(‘‘chips’’), and prompts were gradually faded(Bourret, Vollmer, & Rapp, 2004). During thevocal manding sessions, Millie was seated ata table and no cards were present. Requestsusing the vocal mand resulted in access toa small amount of the item. Vocalization of anyother word resulted in extinction during whichthe therapist turned away from her forapproximately 5 s. She was taught to use thepicture card to mand for the item correspond-ing to the picture (music) and a vocal mand tomand for the item corresponding to the vocalmand (chips). She was taught to use the picturecard to mand for the preferred item usinga three-prompt sequence (verbal instruction,physical model, physical guidance) followed byaccess to the item. She was taught to mandvocally using prompts and modeling followedby access to the item.

Phase 6b: EO manipulation with topographi-cally different responses (vocal and card). Thepurpose of this phase was to test for complexdiscriminations using two distinct responseforms (vocal and card) as well as to serve asan additional training phase to establishdiscriminated mands using two distinct re-sponse forms. Two topographically differentresponses were used to overcome the possiblediscrimination problem caused when topo-graphically similar responses were used. ForMillie, the EO manipulation was similar to thatused in Phase 5; however, only one picture card

was placed in front of her. There were nodistracter cards present, and she had free accessto one of the items she had been previouslytaught to mand for (vocally or handinga picture) during Phase 6a, and access toa second item was restricted. The mand forthe second item had also been taught pre-viously. Requests using the mand correspondingto the restricted item resulted in access to musicfor approximately 30 s or a small amount of thechip (i.e., one chip), and the card was removed.Requests using the mand corresponding to theitem to which the participant already had accessresulted in the therapist removing the card forapproximately 30 s (while the participantmaintained access to the item).

RESULTS

Phase 1: Preference Assessment

For Mario, the free-operant preference as-sessment identified a radio as the most preferreditem, and multiple observations not related tothis study also identified soda as a preferreditem. For Will, the assessment identifiedcookies and a radio as the most preferred items.For Millie, the items identified as most pre-ferred were music and chips. For Malcolm, themost preferred items were television and chips.

Phase 2: Single Card

All participants acquired the single-cardresponse within one very brief session (datanot shown). In all cases, the last severalresponses occurred with no prompting.

Phases 3 and 4: Distracter-Card Probe andMand Training

Figure 2 shows the percentage of trials withcards selected for Mario and Will during thedistracter-card probe and differential reinforce-ment (mand training) phases. Recall that duringthe distracter-card probe, responding usingeither the picture card or the distracter cardresulted in reinforcement. During the differen-tial reinforcement phase, only responses using


the picture card resulted in reinforcement, andresponses using the distracter card resulted inextinction. During the distracter-card probe,Mario selected both the picture card and thedistracter card during both radio and drinksessions but showed a slight bias for the picturecard during the drink (M 5 63%) and radio (M5 68%) sessions. During differential reinforce-ment, Mario selected the distracter card moreoften than the picture card when differentialreinforcement was implemented during theradio sessions and drink sessions (M 5 60%).During this phase, Mario did not look at thecards; instead, he handed the therapist a cardwithout looking down. We then gave Marioinstructions during the radio sessions to handthe picture card to gain access to the radio.However, this manipulation was not effective atproducing 100% correct responding (M 5

73%), and Mario was still not looking at thecards consistently. To ensure that Mario lookedat the cards before handing them to thetherapist, we placed the cards in a small photoalbum to create a communication book. Thisrequired that Mario open the book and find thecorrect card. This intervention was immediatelyeffective at producing 100% correct respondingduring the radio and drink sessions.

During the distracter-card probe, Will re-sponded using the distracter card almostexclusively during the cookie (M 5 95%)sessions and showed a bias for the distractercard during the radio (M 5 73%) sessions.When differential reinforcement was imple-mented, Will responded more often using thepicture card rather than the distracter cardduring both cookie (M 5 85%) and radio (M5 86%) sessions.

Figure 3 shows the percentage of trials withcards selected by Malcolm and Millie duringdistracter-card probe and differential reinforce-ment phases. During the distracter-card probe,Malcolm showed a bias for the picture card. Heresponded using the television card more often(M 5 72%) than the distracter card during thetelevision sessions and selected the chips cardalmost exclusively (M 5 92%) during the chipssessions. When differential reinforcement wasimplemented, Malcolm responded using thepicture card for the majority of the trials duringboth the television (M 5 92%) and chips (M 5

93%) sessions.During the distracter-card probe, Millie

responded using the chips card almost exclu-sively during the chips sessions (M 5 95%) andthe music card exclusively during the musicsessions (M 5 100%). When differentialreinforcement was implemented, Millie re-sponded using the picture card rather than thedistracter card during both the chips (M 5

95%) and music (M 5 97%) sessions.In summary, when differential reinforcement

was implemented for responding using picturecards, all 4 participants responded using the

Figure 2. Percentage of trials with picture card anddistracter card selected during the distracter-card probeand differential reinforcement phases for Mario and Will.


picture card to gain access to preferred items.This was not intended to be a demonstration ofacquisition (recall that 2 individuals selectedpicture cards prior to differential reinforce-ment). Rather, differential responding wasa prerequisite to the EO phase. All 4 individualsresponded differentially and demonstrated theprerequisite skills necessary for inclusion inPhase 5.

Phase 5: Manipulation of the EO

Figure 4 shows the percentage of trials withthe picture card selected for Mario and Willduring the EO phase. Responses using thepicture card when the EO was present resultedin reinforcement and removal of the cards, andresponses using the picture card when the EOwas absent resulted in the therapist removingthe cards for approximately 30 s (but the

participant maintained access to the item).During the radio EO-present sessions, Marioresponded using the radio card 100% of thetime. Mario responded using the drink cardduring the drink EO-present sessions approxi-mately 80% of the time during the first eightsessions and 100% of the time the remainder ofthe sessions (overall M 5 89%), and 0% of thetime in sessions when the EO was absent(overall M 5 0%).

Initially, during the cookie EO-presentsession, Will responded using the cookie cardslightly more often when the EO was presentthan when the EO was absent. However, duringthe last five sessions, he responded using thepicture card only when the EO was present(overall M 5 92%). During the radio EO-present sessions, Will responded using the radiocard most of time (M 5 88%) and rarely duringthe radio EO absent.

Figure 3. Percentage of trials with picture card anddistracter card selected during the distracter-card probeand differential reinforcement phases for Malcolmand Millie.

Figure 4. Percentage of trials with picture card selectedduring manipulation of the EO for Mario and Will.


Figure 5 shows the percentage of trials withthe picture card selected for Malcolm and Millieduring the EO phase. During the televisionEO-present sessions, Malcolm responded usingthe television card more often (M 5 83%) thanwhen the EO was absent. Malcolm respondedusing the chips card during the chips EO-present sessions more often (M 5 83%) thanwhen the EO was absent. During the chips EOpresent, Millie responded using the chips cardslightly more often (M 5 55%) than duringchips EO absent (M 5 44%) and the musiccard more often (M 5 82%) when the EO wasabsent. Millie failed to emit a response duringeach trial during some sessions; therefore, thepercentages do not always add up to 100%.

In summary, results of this phase showed that3 of the 4 participants accurately manded fortwo different items using picture cards in thecontext of two picture cards. Three participants

(Mario, Malcolm, and Will) were able torespond using a picture card representing anitem when the EO for that item was present anddid not typically respond using the picture cardfor an item when the EO was absent. Millie didnot learn to mand for two items using picturecards in the context of two picture cards. Sheresponded using a picture card representing anitem when the EO was present and absent. Thispattern of responding suggested that Millie hadnot acquired the complex discrimination be-tween the two picture cards that was necessaryfor discriminated manding in the context of twopicture cards. It was necessary to teach hera discriminated manding response for her toeffectively mand for two different items.Therefore, she participated in Phase 6, in whichshe was taught to mand using two topograph-ically different responses to teach a discriminatedmand response in the context of two differentitems.

Phase 6a: Topographically Different ResponseTraining (Vocal and Mand)

We taught Millie a topographically differentresponse to establish the discrimination betweenmands for preferred items. Figure 6 shows thevocal mand training data for Millie. Millieresponded independently using the vocal mand‘‘chips’’ as training progressed (M 5 66%), andby the last four sessions she was respondingindependently an average of 85% of the time(100% in the final session).

Figure 5. Percentage of trials with picture card selectedduring manipulation of the EO for Malcolm and Millie.

Figure 6. Percentage of trials with responses emittedduring vocal mand training for Millie.


Phase 6b: Manipulationof the EO withTopographically Different Responses (Vocal andCard)

Figure 7 shows the data for Millie during theEO with two topographically different re-sponses. The Xs represent responses duringbooster sessions. Booster sessions consisted ofsupplemental mand training (of the sort done inPhase 4 for cards and Phase 6a for vocals) andwere conducted when Millie returned tosessions after having missed school for a sub-stantial amount of time. In the first sessions,Millie responded by saying ‘‘chips’’ and usingthe music card more often when the EO wasabsent, but as sessions continued she respondedmore often when the EO was present for boththe chips (M 5 77%) and the music (M 5

84%).In summary, results of this phase showed that

Millie accurately manded for two differentitems using a picture card and a vocal responsein the context of two preferred items. Millie was

able to respond using a picture card represent-ing an item or a vocal response for an itemwhen the EO for that item was present andrarely responded using the picture card or thevocal response for an item when the EO for thatitem was absent.

DISCUSSION

The results of this study showed that for 3participants (Mario, Malcolm, and Will), mandtraining using picture cards resulted in complexdiscriminated mands using picture cards. Inaddition, the implementation of an EO manip-ulation yielded additional validity to theappropriateness of discrimination training inteaching individuals to mand for preferreditems. Conversely, 1 individual (Millie) didnot acquire complex discrimination usingpicture cards. For Millie, discrimination be-tween the picture cards and distracter cards wasestablished in training, but did not result indiscriminated responding when presented withthe EO manipulation. This warrants furtherinterpretation.

A possible explanation for Millie’s undiffer-entiated responding during the EO phase is thatshe may not have been emitting the verbaloperant programmed by the experimenter. Thistype of responding during the EO manipulationwas initially interpreted as inability to mandusing the picture card. It is possible, however,that Millie may have been tacting (labeling)items she had by using the picture card. Thiswould explain why during the EO manipulationshe most frequently handed the picture cardcorresponding to the item she already had accessto (EO absent) rather than manding for the itemshe did not have access to (EO present). A third,but similar, possibility is that Millie was emittinga mand response but may have been manding formore of the item she had access to (as whena restaurant customer asks for a refill before hisdrinking glass is empty).

The results of this study have severalimplications for the practical use of mand

Figure 7. Percentage of trials with responses emittedduring manipulation of the EO with different topogra-phies (vocal and mand) for Millie.


training using picture cards. At present, picture-card training has been used successfully to treatlanguage delays and increase communicativebehavior (Bondy & Frost, 1993; Liddle, 2001;Schwartz et al., 1998). Despite the literature onthe application of picture-card training, far lessresearch has been focused on factors related tothe effectiveness and efficiency with whichpicture-card training produces discriminatedmands. In one case (Millie), modifications tothe initial training procedure were made thatincluded the addition of topographically dis-similar responses (picture cards and vocal).From a practical standpoint, clinicians whouse picture cards in isolation may fail toproduce discriminated manding with someindividuals. Rather than focusing on extensive,repeated exposure to picture cards, mandingmay be more quickly and accurately establishedwith sign language or vocalizations, becauseverbal units of sign language or vocal speech areinherently topographically distinct.

The use of a posttraining assessment, such asthe EO manipulation, may be of practical use indetermining whether the desired response hasbeen established. For Millie, the use of a bookfull of picture cards may have misled others intobelieving that she was successfully makingcomplex discriminations given that she likelywould have pulled out cards and handed themto an adult. The EO manipulation showed thatmore training was needed. Because the EOmanipulation examines responding as a functionof that manipulation, the procedure may helpclinicians identify true, functional mands. Inaddition, the EO manipulation may also be ofpractical use as an additional training phase.Correct responding increased during the EOmanipulation for Mario, Will, and Malcolm,suggesting that further learning may have takenplace as a result of the reinforcement con-tingencies in place for responding using theappropriate picture card.

The results of this study systematicallyreplicate the findings of Brown et al. (2000),

who evaluated functional communication train-ing (FCT; Carr & Durand, 1985) when theEOs were present and absent. Participants weretaught to mand for reinforcers that were eitherrelevant or irrelevant based on the function ofproblem behavior identified through brieffunctional analyses (Northup et al., 1991).Results showed that during the EO-presentcondition (relevant to function of problembehavior), relevant mands occurred often andirrelevant mands rarely occurred. Results alsoshowed that relevant and irrelevant mandsrarely occurred in the EO-absent condition.Data from the Brown et al. study, along withthose from the current study, provide supportfor the use of the EO as an effective assessmentand training component of mand training andFCT treatment packages. In addition, thesedata collectively indicate that additional re-search is needed to identify methods with whichto assess the effectiveness of communicationtraining programs.

The results of this study contribute to theliterature on communication training for indi-viduals with language delays. However, thereare potential limitations of the study that shouldbe taken into account when interpreting theresults. One potential limitation is that the EOmanipulation involved differential conse-quences for selecting the card for the stimulusto which the participant did not have access. Ifthe participant selected the card for the item towhich he or she had access, extinction was ineffect. If the participant selected the card for theitem to which he or she did not have access, thatresponse was reinforced with access to the item.Although the EO manipulation was designed totest for the presence of discriminated manding,the data suggest that responding may have beenshaped by the differential consequences in effectduring this phase. In addition, the isolatedoccurrence of a participant (Millie) whodemonstrated incorrect responding during theEO manipulation should be considered asa potential limitation. The fact that only 1 of


4 participants displayed opposite effects in thisphase makes it difficult to determine the extentto which such responding may occur in others.Millie’s response patterns in the EO phase wereinterpreted as a potential need to conduct a largenumber of assessments (similar to the EOphase) to determine if the participant acquireddiscriminated manding using the discriminationtraining.

Finally, future research may address thelimitation noted above regarding differentialconsequences in effect during the EO phase byconducting an EO test phase in extinction. A testphase conducted in extinction may allowresearchers to attribute more conclusively theresponding using the restricted picture card to theEO manipulation and rule out any learning thatmay have taken place during the EO test phase.Also, future research may extend the results of thepresent study by evaluating whether discriminat-ed mand responses acquired through discrimina-tion training persist under less controlled settings.An extension of the present study may includegeneralization probes in more natural settings(e.g., classroom, home). The EO manipulationcould be easily replicated in natural settings byproviding access to one of the items whilewithholding access to the other. Then, in thepresence of two or more picture cards, if theindividual hands over the picture card of theunavailable item, generalization outside theexperimental setting would be demonstrated.

REFERENCES

American Psychiatric Association. (2000). Diagnostic andstatistical manual of mental disorders (4th ed.).Washington, DC: Author.

Bondy, A. S., & Frost, L. A. (1993). Mands across water:A report on the application of the picture-exchangecommunication system in Peru. The Behavior Analyst,16, 123–128.

Bondy, A. S., & Frost, L. A. (1994a). The Delawareautistic program. In S. Harris & J. Handleman (Eds.),Preschool education programs for children with autism(pp. 37–54). Austin, TX: Pro-Ed.

Bondy, A. S., & Frost, L. A. (1994b). The pictureexchange communication system. Focus on AutisticBehavior, 9, 1–19.

Bondy, A., & Frost, L. (2001). The picture exchangecommunication system. Behavior Modification, 25,725–745.

Bondy, A., & Frost, L. (2002). The picture exchangecommunication system training manual. Newark, NJ:Pyramid Educational Products.

Bourret, J., Vollmer, T. R., & Rapp, J. T. (2004).Evaluation of a vocal mand assessment and vocalmand training procedures. Journal of Applied BehaviorAnalysis, 37, 129–244.

Brown, K. A., Wacker, D. P., Derby, K. M., Peck, S. M.,Richman, D. M., Sasso, G. M., et al. (2000). Eval-uating the effects of functional communication trainingin the presence and absence of establishing operations.Journal of Applied Behavior Analysis, 33, 53–71.

Carr, E. G., & Durand, V. M. (1985). Reducing behaviorproblems through functional communication train-ing. Journal of Applied Behavior Analysis, 18,111–126.

Charlop-Christy, M. H., Carpenter, M., Le, L., LeBlanc, L.A., & Kellet, K. (2002). Using the picture exchangecommunication system (PECS) with children withautism: Assessment of PECS acquisition, speech,social-communicative behavior, and problem behavior.Journal of Applied Behavior Analysis, 35, 213–231.

DeLeon, I. G., Iwata, B. A., & Roscoe, E. M. (1997).Displacement of leisure reinforcers by food preferenceassessments. Journal of Applied Behavior Analysis, 30,475–484.

Frost, L. A., & Bondy, A. S. (1994). The picture exchangecommunication system training manual. Cherry Hill,NJ: PECs.

Liddle, K. (2001). Implementing the picture exchangecommunication system (PECS). International Journalof Language and Communication Disorders, 36,391–395.

McCoy, J. F., & Buckhalt, J. A. (1990). Languageacquisition. In J. L. Matson (Ed.), Handbook ofbehavior modification with the mentally retarded (pp.445–466). New York: Plenum.

Michael, J. (1982). Distinguishing between discriminativeand motivational functions of stimuli. Journal of theExperimental Analysis of Behavior, 37, 149–155.

Michael, J. (1985). Two kinds of verbal behavior plusa possible third. The Analysis of Verbal Behavior, 3,1–4.

Northup, J., Wacker, D. P., Sasso, G. M., Steege, M.,Cigrand, K., Cook, J., et al. (1991). A brief functionalanalysis of aggressive and alternative behavior in anoutclinic setting. Journal of Applied Behavior Analysis,24, 509–522.

Richman, D. M., Wacker, D. P., & Winborn, L. (2001).Response efficiency during functional communicationtraining: Effects of effort on response allocation.Journal of Applied Behavior Analysis, 34, 73–76.

Roane, H. S., Vollmer, T. R., Ringdahl, J. E., & Marcus,B. A. (1998). Evaluation of a brief stimulus preferenceassessment. Journal of Applied Behavior Analysis, 31,605–620.


Schwartz, I. S., Garfinkle, A. N., & Bauer, J. (1998). Thepicture exchange communication system: Communi-cative outcomes for young children with disabilities.Topics in Early Childhood Special Education, 18,144–159.

Sidman, M. (1960). Tactics of scientific research: Evaluatingexperimental data in psychology. Boston: AuthorsCooperative.

Skinner, B. F. (1957). Verbal behavior. New York:Appleton-Century-Crofts.

Sundberg, C. T., & Sundberg, M. L. (1990). Comparingtopography-based verbal behavior with stimulusselection-based verbal behavior. The Analysis of VerbalBehavior, 8, 31–41.

Winborn, L., Wacker, D. P., Richman, D. M., Asmus, J.,& Geier, D. (2002). Assessment of mand selection forfunctional communication training packages. Journalof Applied Behavior Analysis, 35, 295–298.

Wraikat, R., Sundberg, C. T., & Michael, J. (1991).Topography-based and selection-based verbal behav-ior: A further comparison. The Analysis of VerbalBehavior, 9, 1–17.

Received November 29, 2005Final acceptance April 10, 2007Action Editor, Louis Hagopian


manipulating establishing operations to verify … · manipulating establishing operations to...

Documents