ERIC EJ758917: A Preliminary Evaluation of the Emergence of Novel Mand Forms

2007·0.46 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2007, 40, 137–156 NUMBER1 (SPRING2007)
A PRELIMINARY EVALUATION OF THE
EMERGENCE OF NOVEL MAND FORMS
EMMA HERNANDEZ, GREGORY P. HANLEY, EINAR T. INGVARSSON,
AND JEFFREY H. TIGER
UNIVERSITYOFKANSAS
Strategies that produce generalized responding are valuable, especially with regard to language
acquisition,becauserelativelylittletrainingmayresultinlargebehaviorchanges.Conditionsthat
resultingeneralizedmandingwereanalyzedinthecurrentstudy.Wedemonstratedinreversal
designs that undesirable or single-word responses were the predominant mand forms of 3
preschool children. Multiple baseline designs with 2 participants and a reversal design with 1
participant were then used to demonstrate the extent to which differential reinforcement of
single-wordmands(e.g.,‘‘cars’’)orframedmands(e.g.,‘‘Iwantthecars,please’’)wouldresultin
theemergenceofothersingle-wordandframedmandsfordifferentitems(e.g.,mandsformusic,
puppets, or puzzles). Results showed that prompting and differential reinforcement of one or
twomandframes resultedin theemergence ofother framed mandsfor allparticipants.
DESCRIPTORS: differential reinforcement, generalization, language training, mands,
preschool children, tacts,verbal behavior
_______________________________________________________________________________
Over1 millionchildrencategorizedashaving Wacker et al., 1990). In essence, both the
speech and language delays were served by problem and the alternative behavior are
public school special education programs dur- controlled by the same establishing operations
ing the 2000–2001 academic year (National andconsequences,andbothmaybeunderstood
Dissemination Center for Children with Dis- as mands. Skinner (1957) used the term mand
abilities, 2004). Problem behaviors are more to describe a verbal operant that is evoked by
prevalent among children with language delays deprivationoraversivestimuli(e.g.,establishing
than typically developing children (Willinger et operations) and is reinforced by specific con-
al., 2003). It is therefore not surprising that sequences directly related to the response. For
previousresearchhasshownthatmanyformsof example, restricted access to a preferred toy
problem behavior serve a communicative (so- truck is likely to function as an establishing
cial) function (e.g., Carr & Durand, 1985; operation for saying ‘‘truck,’’ which would be
Winborn, Wacker, Richman, Asmus, & Geier, reinforced with the specific consequence of
2002). Consequently, strategies have been de- access to the truck.
veloped that replace problem behavior that Single words (e.g., ‘‘truck’’) are often sug-
serves a social function with an acceptable gested as target mands for children with
alternative (e.g., Carr & Durand; Hagopian, language deficits due to the ease of production
Fisher, Sullivan, Acquisto, & LeBlanc, 1998; in comparison to framed responses (e.g., ‘‘I
want the truck, please’’; Horner & Day, 1991;
Musselwhite & St. Louis, 1988), and several
Thisprojectwascompletedaspartialfulfillmentofthe
requirements for a Master of Arts degree by the first studies have shown that prompting and differ-
author. We thank Rachel H. Thompson and James A. ential reinforcement of single-word mands
Shermanfortheirhelpfulcommentsonanearlierversion
result in a reliable increase in these responses
ofthismanuscript.
Reprints may be obtained from Gregory P. Hanley, (Gobbi, Cipani, Hudson, & Lapenta-Neudeck,
Applied Behavioral Science Department, University of 1986;Winbornetal.,2002;Yoderetal.,1995).
Kansas, 4023 Dole, Lawrence, Kansas 66045 (e-mail:
Targeting single-word responses may be effi-
[email protected]).
doi:10.1901/jaba.2007.96-05 cient in the initial stages of mand teaching, and
137
138 EMMA HERNANDEZ et al.
may be especially important if those mands are animportantresearchareainvolvesunderstand-
being taught as functionally equivalent replace- ing the conditions under which generalization
ments for problem behavior (Carr & Durand, between verbal operants (e.g., tacts and mands)
1985; Fisher et al., 1993; Horner & Day; will occur or, from an applied perspective,
Wacker et al., 1990). However, the same identifying the procedures that can produce
strategy may be inefficient in the long term, generalization between verbal operants.
because it is likely that each new mand would Wallace, Iwata, and Hanley (2006) recently
requireseparateinstructionpriortoacquisition. showed that transfer from tact to mand
Thus, it may be important that mand training, functions was dependent, at least in part, on
whether as an intervention for a preexisting the valueof theitems to be tacted and manded.
pattern of socially mediated problem behavior That is, if the items were shown to be highly
or as a means to preempt the development of preferred, mands for these items occurred and
this sort of behavior, contain features that allow were maintained following tact training. An
the development of mands that are not directly additional strategy for promoting transfer
trained. among verbal operant classes was offered by
Strategies to promote generalization were Skinner (1957), who suggested that the use of
first described by Stokes and Baer (1977), who a frame might facilitate generalization to un-
suggested that generalization should be directly trainedresponses(e.g.,themandframe,‘‘Iwant
assessed and carefully programmed. Although the —,’’ or the tact frame, ‘‘That is a —’’). A
a variety of procedures have been found to be mand frame may serve to bridge tacts and
effectiveforstrengtheningverbalrepertoires(see mands. That is, a mand frame may serve to
Warren & Reichle, 1992, for a review), rela- mediategeneralization(Stokes&Baer,1977)in
tively few studies have determined the teaching that the response, ‘‘I want the —, please,’’
conditions that will result in generalization. would be taught in the presence of a strong
One possible reason for this relative lack of establishing operation. This frame may thus be
research is that similar responses may have evoked in the presence of other similar estab-
dissimilar controlling variables. For instance, lishing operations, thus resulting in generaliza-
several researchers have shown that mands and tion.Forexample,achildmaybetaughttosay,
other verbal operants (e.g., tacts) are function- ‘‘I want juice, please’’ when thirsty and may
ally independent (Hall & Sundberg, 1987; subsequently use a frame to mand for food
LaMarre & Holland, 1985; Sigafoos, Doss, & when hungry (e.g., ‘‘I want cookies, please’’).
Reichle, 1989), and that teaching a child to Such outcomes were suggested by Hart and
label (i.e., tact) an item will not necessarily Risley (1980), who showed that an experimen-
result in that child being able to request (i.e., tal group who were taught to use compound
mand) that item when it may be important for sentences to access restricted toys used more
the child to do so. In contrast, more recent complex vocabulary and sentence structure that
research (Petursdottir, Carr, & Michael, 2005) had not been directly taught than the compar-
has shown that mands and tacts are not ison group did.
necessarily independent. Petursdottir et al. Simic and Bucher (1980) provided an
showed that mand training tended to lead to example of teaching mands within frames when
emergent tacting, and tact training resulted in they taught 5 children with mental retardation
some emergent manding. Although this study the partial frame, ‘‘I want a —.’’ Their data
suggests that mand training leads to emergent showed that mand frames were acquired by
tacting, variables that effect transfer between the participants, but the benefit of gener-
mandsandtactshavenotbeenidentified.Thus, alized manding via framed mands was not
NOVEL MAND FORMS 139
demonstrated. Therefore, further analysis of the participated in this study. All 3 children
conditions in which teaching framed mands attended a classroom in a full-day inclusive
results in generalization to untrained mands preschool program. They were selected to
seems warranted. participate in this study because they made
In the current study we first determined the few vocal requests in the classroom, and
predominant mand forms for 3 preschool-aged teachers reported that their communication
childrenwhoappearedtobeatriskforlanguage consisted primarily of gestures, crying, and
delays. We then determined the direct and taking items from children and adults.
indirect effects of providing differential re- The preference assessments and mand form
inforcement for relatively smaller or larger and generalization analyses were conducted in
topographical mand classes. More specifically, aroom(2.3 mby2.7 m)thatcontainedachild-
once we identified a variety of preferred items sized table, two child-sized chairs, five plastic
anddeterminedthattactingrepertoiresforthese opaque containers, and two bookshelves. The
items were intact, we evaluated the extent to tact assessment and training were conducted in
which training both single-word mands (e.g., a classroom (12 m by 7 m) that contained
‘‘truck’’) and framed mands (e.g., ‘‘I want the a variety of recreational activities.
truck,please’’)wouldresultintheemergenceof
Data Collection, Target Behaviors, and
novel mands.
Interobserver Agreement
During the preference assessment and tact
GENERAL METHOD
assessment, trial-by-trial data were collected on
Assessment and Training Overview the child’s approach toward an item or vocal
A preference assessment (Fisher et al., 1992) approximation of the name of the item,
was conducted with each participant to identify respectively. An approach was defined as reach-
preferred items that would maximize the ing toward the item with one or both hands.
motivating conditions of the mand analysis During the tact assessment, a correct response
(Wallace et al., 2006). A tact assessment was was scored when the child’s vocal response
conducted to ensure that the participants were approximated the agreed-upon name of the
able to tact each of the items that were to be item;anincorrectresponsewasscoredwhenthe
includedin themand analysis,and tacttraining child did not make a vocal response within 5 s
was provided when the participant was not able orwhenthevocalresponsedidnotresemblethe
to tact a preferred item. Thus, items that the name of the item. The relevant tacts for each
children preferred and could tact were included child are listed in Table 1.
in the mand analysis. The initial mand analysis During each preference and tact assessment,
identified the particular response forms that data were collected using paper and pencil and
functioned as mands (referred to as the mand were summarized as the percentage of trials on
form analysis). These data were then used as which an item was selected (preference assess-
baselines to evaluate the direct and indirect ment) or the number of correct responses over
effects of teaching single-word and framed the total number of trials conducted (tact
mands (referred to as the generalization analy- assessment and training). Interobserver agree-
sis). ment was assessed by having two observers
simultaneously but independently record selec-
Participants and Setting tions (preference assessment) and correct or
Two 4-year-old-boys (Tom and Jason) with incorrect responses (tact assessment) on a trial-
nonspecific developmental delays and 1 1.7- by-trial basis. Agreements were defined as both
year-old typically developing girl (Ana) observersscoringthesameresponseduringeach
140 EMMA HERNANDEZ et al.
Table 1
TactsandItem Descriptions
Child Tacts Items
Ana Frog,froggies Fourbeanbagfrogsandarubberfrog
Babies,girl,doll,bottle One20-in.doll,one8-in.cryingandlaughingdollwithabottleandblanket,
andone5-in.musicaldoll
Phone,telephone Sixplasticphoneswithsoundsandmusicbutton
Dinosaur,dino PlasticLegoHdinosaursandaplasticdinosaurwithsoundsandalight
Cars,train,bus Plasticcarswithsoundsandlights,abusandatrain
Tom Cars Fivecarsofvariouscolors
Dress-up Scarves,necktie,necklaces,avest,dresses,bracelets,glasses,twohats,aring,andshoes
DisneyH PlasticAladdinHandPoohBearHmoviecharacters
Books PaperbackpreschoolbookssuchasTonkaTrucksHandabook
Dinosaurs PlasticLegoHdinosaursandaplasticdinosaurwithsoundsandalight
Jason Music Drum,metaltriangle,xylophone,clapper,cymbals,andrhythmsticks
Puppet Animalhandpuppets
Bears Colorfulplasticstackablebears
Dinosaurs PlasticLegoHdinosaursandaplasticdinosaurwithsoundsandalight
Puzzles Electronicinsetpuzzlewithsounds
trial. Agreement for each session was calculated development of their vocal verbal repertoires.
by dividing the total number of agreements by Because single-word mands, as opposed to
the total number of agreements and disagree- undesirable behavior, were found to be the
ments and multiplying by 100%. Agreement predominant mand form for Jason, undesirable
was assessed during at least 50% of preference responses are not reported in the generalization
assessment sessions across all participants and analysis. Single-word responses were defined as
averaged 98% (range, 96% to 100%). Agree- vocalizations thatincluded anapproximationof
ment was assessed during at least 75% of tact the name of the target item (e.g., ‘‘dinosaurs,’’
assessment and training sessions across all ‘‘dino,’’ or ‘‘want dino’’) and did not include
participants and averaged 94% (range, 73% to a mand frame. Framed responses were defined
100%). as vocalizations taking the form of ‘‘I want
The frequency of each participant’s undesir- [name of the object], please,’’ ‘‘May I have
able, single-word, and framed responses was [name of the object], please?’’ or ‘‘Can I have
recorded during the mand form and general- [name of the object], please?’’ Single-word
ization analyses. Undesirable responses for Ana responses were not scored in the same instance
were crying, yelling, and two rudimentary signs a framed response occurred (i.e., single and
defined as placing an open palm on her chest framed responses were mutually exclusive
and moving it from side to side (similar to the categories).
American sign language sign for ‘‘please’’) or During the mand form and the generaliza-
extending her arms toward the items while tion analyses, data on child behavior were
making contact with the fingertips of both collected using handheld computers during
hands (similar to the sign for ‘‘more’’). The continuous 10-s intervals and were summarized
undesirable response form for Tom and Jason as the number of responses per minute.
was pointing, which was defined as extending Interobserver agreement was assessed in at least
the index finger in the direction of the item. 49% of mand form analysis sessions across all
These responses were considered undesirable participants and averaged 98% (range, 87% to
because all participants showed evidence of 100%). Agreement was assessed in at least 22%
somewhat intact vocal verbal repertoires; thus, of the generalization analysis sessions across all
theseresponsesmayhavecompetedwithfurther participants and averaged 98% (range, 64% to
NOVEL MAND FORMS 141
100%). Exact agreement scores were calculated dividing the number of times the child selected
on an interval-by-interval basis by dividing the an item by the total number of trials in which
number of intervals with agreements by the theitemwaspresentedandthenmultiplyingby
number of intervals with agreements and 100%.
disagreements; these fractions were then aver- Tact assessment and training. Each child’s
aged and multiplied by 100%. An agreement ability to tact each of the 10 items was assessed
was scored for each interval if both observers by asking the child, ‘‘What is this?’’ in the
recorded the same number of responses. presence of each item. Following a correct
response, the therapist provided social praise
PREFERENCE ASSESSMENTAND TACT ASSESSMENT (e.g., ‘‘That’s right!’’). Incorrect responses
AND TRAINING resulted in a brief 2-s time-out from interaction
Thepurposeofthepreferenceassessmentand followed by the presentation of the next item.
the tact assessment and training was to identify The therapist presented each of the 10 items in
five items to be included in the mand analysis. a random order, and each item was presented
Items to be included met three criteria: (a) three times (for a total of 30 tact trials per
When provided with the item following an child). If the child did not accurately tact the
approach response during the preference assess- target items in all three of the opportunities,
ment, the child engaged with the items for the tact training was conducted.
entire 30-s period in a developmentally appro- Tact training was conducted in a manner
priatemanner(showingtheabsolutevalueofan
similar to the tact assessment except that if the
item). (b) The child could tact the items in
child did not respond within 5 s or responded
three consecutive opportunities during the tact
incorrectly, the therapist repeated the question
assessment or following tact training (showing
and simultaneously provided a model prompt.
a minimal tact repertoire with respect to the
For example, the therapist asked, ‘‘What is
items). (c) The items were selected over other
this?’’ and then immediately stated, ‘‘This is
items on at least some of opportunities
a car, say car.’’ If the child imitated the
(showing the relative value of the item).
vocalization within 5 s of the model, social
praise was provided. On the next trial, the
Procedure
prompt was delayed by 2 s. For example, the
Preference assessment. Interviews of parents
therapist said, ‘‘What is this?’’ and waited 2 s,
and teachers yielded a list of 10 preferred items
before stating, ‘‘This is a car, say car.’’ The
for each child. A paired-choice preference
modelpromptwasdelayedbyincrementsof2 s
assessment (Fisher et al., 1992) was conducted
following each correct response (Ault, Gast, &
with each participant to identify five items for
Wolery, 1988). Tact training was terminated
use during the mand analysis. Each item was
when the child independently (i.e., in the
paired with every other item once during the
absenceofamodelprompt)respondedcorrectly
assessment. This occurred by placing two bins,
for three consecutive trials.
each containing different sets of items, on the
table simultaneously. The child’s approach
toward one of the two bins (e.g., a bin MAND FORM ANALYSIS
containing cars and a bin containing dinosaurs) The purpose of the mand form analysis was
resultedin30-saccesstothatbin.Whenachild to identify the response forms that functioned
approached two bins simultaneously, the ther- as mands for each child. Five 3-min sessions
apist blocked access to the bins, and the bins were conducted daily, 4 to 5 days per week.
werepresentedagain.Thepercentageoftrialsin Each of the 3-min sessions involved one of the
which items were selected was calculated by five sets of items that the child was able to tact
142 EMMA HERNANDEZ et al.
and had been selected on over 20% of single-word, or framed responses resulted in
opportunities. The presentation of the items brief (e.g., 30-s) access to the bin of items.
was conducted in random order for Tom.
However, to determine if a temporal general- Experimental Design
ization gradient of novel mand forms would A reversal design was used to identify the
develop (i.e., if untrained mands that were response forms maintained by access to the
assessedcloserintime todirectlytrainedmands target items (i.e., mands). For Ana, this analysis
would emerge sooner than mands that were began with the DRA condition due to signif-
assessed later), the order in which the items icant changes in the definition of her undesir-
were presented was held constant for Ana and able responses that occurred in previous (un-
Jason. Brief breaks were provided between each reported) observations during NCR and DRA
sessionsuchthatthetotalamountoftimespent sessions.
in sessions each day was approximately 20 min
for all participants. GENERALIZATION ANALYSIS
Procedure
Procedure The purpose of the generalization analysis
Noncontingent reinforcement (NCR). During wastodetermineifdifferentialreinforcementof
NCR, one of the five sets of items was freely single-word and framed mands would result in
accessible for 3 consecutive minutes. At the end generalization to single-word or framed mands
ofthe3-minsession,theitemswerereturnedto with no explicit training. The NCR and DRA
the bin and placed out of view. A total of five conditions from the mand form analysis served
sessions (i.e., one session per toy set) were as baselines for all 3 children for the general-
conducted consecutively, and only one set of ization analysis.
itemswaspresentedatatime.Thetherapistdid Differential reinforcement of single-word and
not respond in any way to the target responses framed responses (DRSF). This condition was
or other behaviors during these sessions. implemented with Ana and Tom only because
Because Ana rarely experienced periods in their mand form analysis showed that undesir-
which an adult who was in close proximity able mands were their predominant mand
did not interact with her for minutes at a time, forms. Thus, the DRSF condition restricted
she was also provided with 10 s of attention the contingency class that received reinforce-
every 30 s, independent of her behavior. ment to single-word and framed responses. In
During this condition, the establishing opera- other words, this condition was similar to the
tionformandingforthetoysetswaseliminated originalDRAconditiondescribedabove,except
or at least minimized by providing continuous that access to the visible item was not provided
access to the preferred items. Therefore, this if the child engaged in an undesirable response.
condition served as a control for the effects of A model of the single-word mand was provided
differential reinforcement. after the initial 30 s had expired following the
Differential reinforcement of undesirable, sin- presentation of the item or when the child had
gle-word, or framed mands (DRA). The DRA not emitted a single-word or framed response
condition was similar to the NCR condition for 5 s following removal of the items. Prompt
except that the items were placed on the table delays were then used to eliminate the
but out of the child’s reach, and access to the vocal model prompts. One second was added
items was provided for 30 s following any to the 5-s delay following a correct response
target response. That is, to increase the rate of and 1 s was subtracted from the delay
the target response that functioned as the following an undesirable response. When vocal
participant’s predominant mand, undesirable, model prompts had been eliminated for three
NOVEL MAND FORMS 143
consecutive sessions (i.e., independent respond- indicating that some items were more preferred
ing occurred), vocal mand models were not thantheotheritemsincludedintheassessment.
provided in subsequent sessions. Results obtained from the tact assessment
Differential reinforcement of framed responses and training are shown above the bars on
(DRF). Differential reinforcement of framed Figure 1. Ana tacted four of the items (tele-
responseswasimplementedwhenAnaandTom phones, dinosaurs, frogs, and babies) on all
reliably engaged in single-word mands on one three trials and appropriately interacted with
of the baselines that had previously been thesematerialsduringthepreferenceassessment
exposed to the DRA or DRSF conditions, and (data available from the second author);
theindirect effectsofthatbehaviorchange were therefore, these items were included in her
incomplete (e.g., undesirable behavior persisted mand analysis. Ana did not engage with the
on the remaining baselines). DRF was imple- pensappropriately,sotheywerenotincludedin
mented with Jason to increase the developmen- themandanalysis.Shetactedcarsinatleastone
tal appropriateness of his manding repertoire. of three trials; therefore, cars were selected for
This condition was similar to the DRSF inclusionasherfifthtoyset.Thirteentrialswere
condition, except that a framed mand model required to teach Ana to tact cars on three
was provided after the initial 30 s had expired, consecutive opportunities (data not shown).
and access to the preferred item was provided Tom tacted cars, dinosaurs, and books in at
only following framed responses (i.e., undesir- least one of the three trials and appropriately
able responses and single-word mands were engaged with these materials during the prefer-
placed on extinction). Prompting, prompt ence assessment; therefore, these items were
delays, and mand models were arranged as included in his mand analysis. Three, 11, and
described above. 15 trials were required to teach Tom to tact
cars, dinosaurs, and books, respectively, on
Experimental Design three consecutive opportunities (data not
Thedirect effects of therestricted differential shown). Tom appropriately engaged with the
reinforcement contingencies were demonstrated DisneyH and dress-up toys; therefore these
in multiplebaseline designs acrossbehaviors for items were also included as the fourth and fifth
Ana and Tom and in a reversal design with toy sets. Fifteen and 18 trials were required to
Jason.Althoughmultiplebaselineswereused to teach Tom to tact DisneyH and dress-up,
observe the indirect effects of the restricted respectively, on three consecutive opportunities
differential reinforcement procedures for all (data not shown). Jason tacted puppets, bears,
participants, experimental control over the puzzles, and dinosaurs on all three trials and
indirect effects of the differential reinforcement appropriately engaged with these materials
contingencies was shown only with Jason in during the preference assessment; therefore,
a reversal design. these items were included in his mand analysis.
Jason did not engage with the necklaces or
RESULTSAND DISCUSSION books appropriately; therefore, music was
Preference Assessment and Tact Assessment selected for inclusion as his fifth toy set. Six
and Training trials were required to teach him to tact this
Items included in the mand analysis met item on three consecutive opportunities (data
three criteria described above. Results obtained not shown).
from the preference assessment are shown in
Figure 1 as the percentage of trials in which Mand Form Analysis
each item was selected. In general, preference The results of the mand form analysis for
assessment data show a descending slope, thus Ana are shown in Figure 2. The mean rate of
144 EMMA HERNANDEZ et al.
Figure1. PercentageoftrialsinwhichitemswereselectedduringthestimuluspreferenceassessmentforAna(top),
Tom(middle), andJason (bottom).
undesirable responses was highest during the items (M 5 0.3). The rate of undesirable
first DRA condition (M 5 0.9) across all items responses increased during the return to DRA
(i.e., cars, dinosaurs, frogs, babies, and phones). (M 5 0.8 across all items), and single-word
Single-word responses were also observed responses persisted at similar, albeit highly
during DRA (M 5 0.4 across all items), but variable, rates (M 5 0.7 across all items).
to a much lesser extent than undesirable Framed responses were not observed in any
responses. During the NCR condition, no condition with Ana.
undesirable responses occurred, but single-word Low and inconsistent rates of undesirable
responses occurred intermittently across all five (M 5 0.3) and single-word (M 5 0.1)
NOVEL MAND FORMS 145
Figure2. Independenttargetresponsesperminutefor Figure3. Independenttargetresponsesperminutefor
Anaduring the mandformanalysis. Tomduring themand formanalysis.
responses occurred during the initial NCR NCR, undesirable responses were zero in 14
condition for Tom across each of the five of the 15 sessions across the five baselines,
baselines (i.e., cars, dress-up, DisneyH, dino- and single-word responses occurred at low
saurs, and books) (Figure 3). During and intermittent rates (M 5 0.1). Upon the
DRA, high and persistent levels of undesirable return to the DRA condition, undesirable
responses occurred (M 5 1.4), whereas single- responses increased (M 5 1) and single-word
word responses occurred during only one (M 5 0.1) and framed (M 5 0) responses
of the 20 sessions. During the return to did not.
146 EMMA HERNANDEZ et al.
NCR resulted in either low or zero rates of
all target responses across each of the five
baselines (i.e., music, puppets, bears, dinosaurs,
and puzzles) for Jason (Figure 4). That is,
the mean rate of responding for undesirable,
single-word, and framed responses was 0, 0.3,
and 0 responses per minute, respectively.
During the DRA condition, single-word
responses occurred consistently in all sessions
(M 5 1.7), undesirable responses occurred at
variable rates in just over half of the sessions
(M 5 0.8), and framed responses were not
observed. When NCR was reinstated, the rate
of undesirable responses continued to be vari-
able (M 5 1.8), whereas the consistent rate of
single-word responses observed during DRA
was disrupted (M 5 0.6). The return to DRA
resulted in stable rates of single-word responses
across all baselines (M 5 1.9) and inconsistent,
and at times high, rates of undesirable behavior
(M 5 1.7).
The results of the mand form analysis
showed that Ana’s and Tom’s preexisting
and predominant mand forms were undesirable
responses, given that they appeared to be
most sensitive to the changes in the relevant
contingencies. By contrast, single-word re-
sponses appeared to be most sensitive to the
condition changes for Jason and appeared to
be his predominant mand form. Although
all target responses would have resulted in
access to the toy bins during the DRA
condition, less desirable mand topographies
emerged for each child: rudimentary signs
for Ana, pointing for Tom, and uttering single
words for Jason. That is, by programming
a broad differential reinforcement contingency
(i.e., one that provided reinforcement for
Figure4. Independenttargetresponsesperminutefor
several categories of responding sequenced on a Jason during themand formanalysis.
developmental continuum), the predominant
mand form was identified for each child across Although reliable patterns of one of the
all item sets. For all participants, the respective topographical classes were evident from the
mand forms possessed strong generality as mand form analysis for each child, there was
similar topographies of manding were observed some variability in the data that deserves
in the DRA condition across all baselines. comment. The establishing operation for