The role of specific consequences in the maintenance of three types of q's williams
JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2003, 36, 285–296 NUMBER 3 (FALL 2003) THE ROLE OF SPECIFIC CONSEQUENCES IN THE MAINTENANCE OF THREE TYPES OF QUESTIONS GLADYS WILLIAMS APPLIED BEHAVIORAL CONSULTANT SERVICES, NEW YORK LUIS ANTONIO PEREZ-GONZALEZ ´ ´ UNIVERSITY OF OVIEDO, SPAIN AND KIM VOGT DAVID GREGORY SCHOOL, PARAMUS, NEW JERSEY This research replicated and extended a study by Williams, Donley, and Keller (2000). In that study, children with autism received a box with an object inside and learned to ask ‘‘What’s that?,’’ ‘‘Can I see it?,’’ and ‘‘Can I have it?’’ to have the name of the object, to see the object, and to get the object, respectively. The purpose of the present research was to determine if the three questions (a) were three independent repertoires of behavior, (b) constituted three instances of a single functional response class, or (c) belonged to a chain of behavior. The 3 boys with autism who participated responded independently to each question when the consequences for each question were altered. This indicates that the three target responses were three independent repertoires of behavior, each one re- inforced and maintained with its speciﬁc consequences. Thus, this procedure serves to teach children with autism to ask questions with ﬂexibility according to a variable context. DESCRIPTORS: question asking, verbalizations, autism, language acquisition, ver- bal behavior, establishing operations Asking questions is a complex verbal be- ‘‘Where is it?,’’ and ‘‘Can I see it?’’ (Koegel,havior that for most children with autism Camarata, Valdez-Menchaca, & Koegel,does not emerge until it is explicitly taught. 1998; Shabani et al., 2002; Sundberg, 2000;Many researchers have been successful at Taylor & Harris, 1995; Williams, Donley, &teaching these children to ask questions such Keller, 2000). Williams et al. taught 2 chil-as ‘‘What’s that?,’’ ‘‘What’s inside the box?,’’ dren with autism to ask three questions, in a stepwise fashion, about a closed box with Partial support for this research comes from a grantfrom Cajastur to the ﬁrst two authors. We thank Jo- a toy inside. The experimenter ﬁrst showedseph Spradlin and Jennie Keller for reviewing previous the closed box to the child. When the childversions of the manuscript; all the teachers who par- asked the ﬁrst question (‘‘What is that?’’),ticipated in this research; the principals of the School the experimenter said the name of the hid-Santa Marıa del Naranco in Oviedo, Spain, and the ´David Gregory School in Paramus, New Jersey; and den object. When the child asked the secondthe children’s parents. Details of the procedure and question (‘‘Can I see it?’’), the experimenterdata-collection protocol are available upon request to showed it to him. When the child asked thethe ﬁrst author. Address correspondence to Gladys Williams, Ap- third question (‘‘Can I have it?’’), the exper-plied Behavioral Consultant Services, 66 Re- imenter gave the object to the child.gency Circle, Englewood, New Jersey 07631 (e-mail: The study by Williams et al. (2000) email@example.com) or to Luis A. an important question: Did these three typesPerez-Gonzalez, Departamento de Psicologıa, Univer- ´ ´ ´sity of Oviedo, Plaza Feijoo s/n, Oviedo 33003, As- of questions have the same functions as theyturias, Spain (e-mail: firstname.lastname@example.org). have in everyday interactions? In a typically 285
286 GLADYS WILLIAMS et al.developing person, question asking has con- velop strategies to teach these children tosequences that vary according to the situa- produce each type of question with enoughtion. Therefore, it is important to know if variability, according to the speciﬁc states ofthis type of teaching leads to establishing a deprivation, and according to the particularrepertoire of three mands, each one main- consequences those responses have for typi-tained by its proper functions in a varied cally developing individuals.context. The main purpose of the present research The results of Williams et al. (2000) may was to study further the variables that inﬂu-have been due to three possible factors: First, ence question asking. We initially taughtit is possible that the responses to the three children the three questions used by Wil-questions were independent of the other re- liams et al. (2000). Once the children begansponses, each one maintained by its speciﬁc to ask the three questions, we manipulatedconsequence. That is what presumably oc- two variables: First, we provided no conse-curs in everyday life, because typically de- quences for the second question (‘‘Can I seeveloping children may ask only one of these it?’’). Second, we introduced unpleasant ob-questions according to the situation. Second, jects. Children’s performance should indicateeach response could have been established whether each question serves a distinct op-because initially each one produced access to erant function, whether the three responsesthe item. In fact, during the initial stages of formed a single response class, or whetherteaching, the object followed each question the three operants formed a response se-at the moment of teaching it. According to quence.this hypothesis, any of the three responseswould produce the ﬁnal item and, therefore, METHODwould be equivalent to each other: Theywould have formed three response forms of Participantsa single operant response class (e.g., Dona- The participants were 3 boys who hadhoe & Palmer, 1994). Third, it is possible been diagnosed with autism by several in-that this type of teaching led to a sequence dependent professionals and ofﬁcials of theof responses in which each question was educational system. Jim, 2 years 9 monthsmaintained by the onset of the next question old, received intensive behavioral in-homeor the terminal reinforcer (the object). The teaching. Erik, 4 years 5 months old, at-object, thus, would ultimately reinforce the tended a special needs behavioral program.entire sequence. Therefore, the teacher’s re- Jim and Erik were U.S. residents and thussponses to the ﬁrst and second questions received instruction in English. Dario, 9would reinforce each question by presenting years 4 months old, attended a special needsthe opportunity to ask the following ques- behavioral program in Spain and thus histion, as in a response chain (Skinner, 1934). instruction was given in Spanish. All theIf that happens, the interruption of the se- children displayed mands (e.g., ‘‘I want toquence by removing the consequence could eat candy’’) and tacts (e.g., ‘‘This is a cat’’;result in the absence of any further behavior. they could tact at least 100 objects); repeated These hypotheses have programmatic im- statements of at least four words (e.g., ‘‘It isplications: Due to the lack of spontaneity cold today’’); responded to basic social ques-and language production shown by children tions such as ‘‘What is your name?’’; andwith autism, it is important to know the de- selected objects and pictures, without errors,gree of variability and ﬂexibility these ques- to the instruction ‘‘Give me the [name oftions have. Moreover, it is important to de- the object].’’ All 3 students could say ‘‘No!’’
ASKING QUESTIONS 287and rejected items they disliked by pushing never again prompted, even if the child nev-them away with their hands or feet, or by er said it again. If the child asked ‘‘Can I seeturning away or covering their faces. None it?’’ before asking ‘‘What’s in the box?’’ theof them, however, asked questions prior to experimenter said ‘‘Sure, you can see it,’’ andthe onset of this study. showed the object in the box to the child but did not give it to him. Or if the childMaterials asked ‘‘Can I have it?’’ before asking the ﬁrst The experimenters provided about 80 two questions, the experimenter said ‘‘Sure’’small boxes varying in shape, color, and tex- and gave him the box with the object inside.ture, with attractive objects hidden inside Baseline. The experimenter presented one(e.g., a sparkling spinning wheel, a shiny red box at a time, 10 times, for 20 s. The ex-race car). The objects were novel for each perimenter did not explicitly prompt any be-child but were similar to other objects pre- havior, but opened the box with an objectviously used as reinforcers to teach other inside and made a comment about the ob-skills. The unpleasant objects were selected ject (e.g., ‘‘Oh, this one is great!’’ or ‘‘I reallybased on information from the child’s moth- like this one.’’). The experimenter did noter, who in some instances provided some un- show the object to the child. She changedpleasant objects from home (e.g., a used cu- to a new box if 20 s passed with no responseneiform sponge from the bathroom, a slimy from the child.jelly-like form toy). We also used live insects Teaching the ﬁrst response form (‘‘What’s inobtained outdoors (a worm, a spider, a snail) the box?’’). The experimenter held the boxin some trials with all children. and made a comment about the object in- side the box. She then prompted the childProcedure to repeat the question ‘‘What’s in the box?’’ Overview. The treatment package, a rep- by modeling the question in a ﬁrm tone oflication with an extension of Williams et al. voice. When the child repeated the question,(2000), was evaluated with a multiple base- the experimenter told the child what was in-line design across the three response forms. side the box and gave the box with the ob-When all of the children acquired the ques- ject inside to the child. The child could playtions, we varied the consequences to the with the toy for approximately 20 s; then, aquestions in several phases. The child sat new box was presented. When the child re-across from the experimenter in the class- peated the question correctly for two con-room or in the child’s teaching room. One secutive trials, the experimenter faded thetrial consisted of presenting one box with a echoic prompt by providing a partial prompthidden object for 20 s. The presentation of for two more consecutive trials. For example,one box and the eventual production of one, the experimenter provided the wordtwo, or three questions constituted one trial. ‘‘What’s’’ instead of providing the wholeOne session consisted of 10 trials, that is, question. The prompt was gradually reducedthe presentation of 10 boxes, one at a time. until the child asked ‘‘What’s in the box?’’Usually, we conducted one session per day. without any prompt. If the child producedWe recorded data on each question in each three consecutive errors (saying the questiontrial. The prompted and self-initiated re- incorrectly or saying nothing within 20 s ofsponses were recorded separately, and only presenting the box), the experimenter pro-the self-initiated responses were considered vided the full echoic prompt again. Thiscorrect. After the child had met criterion on procedure was repeated until transfer of con-each speciﬁc question, that question was trol from the echoic prompt to the box itself
288 GLADYS WILLIAMS et al.was accomplished. That is, the echoic tion correctly, the experimenter said, ‘‘Ofprompt was eliminated when the child ini- course, I can give this toy to you,’’ and gavetiated the question without any prompt each the box to the child. The echoic prompt wastime a box was presented. Criterion consist- presented only after the child emitted theed of 90% or better correct unprompted re- second response (‘‘Can I see it?’’). The pro-sponses in two consecutive sessions. At this cedure to fade and ﬁnally eliminate thepoint, the teaching of the second response echoic prompt was the same as for the ﬁrstform began, and no more prompts were pro- and second response forms. Once this thirdvided for the ﬁrst question. question reached criterion, the experimenter Teaching the second response form (‘‘Can I gave the child the object after the childsee it?’’). This procedure started the same way asked this question, even if the child skippedas the ﬁrst response form. However, when the other two questions. The child couldthe child self-initiated ‘‘What’s in the box?,’’ emit any of the three questions during thethe experimenter told him the name of the 20-s trials. After criterion had been obtainedobject inside the box but neither showed it for the third response, all the prompts werenor gave it to him. The experimenter then eliminated for the rest of the study. Thisprompted the child to repeat ‘‘Can I see it?’’ phase was extended several more sessions toWhen the child repeated the question cor- ensure stability.rectly, the experimenter said ‘‘Sure, I can ‘‘No!’’ to the question ‘‘Can I see it?’’ Theshow it to you,’’ and gave the box to the purpose of this phase was to see if the otherchild. The procedure to fade and eliminate two operants would be maintained if one ofthe echoic prompt and the criterion to move the operants (‘‘Can I see it?’’) was eliminat-on to teach the next response form were the ed. Immediately after the child asked ‘‘Cansame as in the teaching of the ﬁrst response I see it?’’ the experimenter said ‘‘No!’’ andform. During this phase, the echoic prompt did not show the object to the child. Whenwas presented only after the child emittedthe ﬁrst response (‘‘What’s in the box?’’). But the child asked ‘‘Can I have it?’’ the experi-once the second question reached criterion, menter gave him the box. No prompts werethe child’s second question resulted in the provided in this phase, and the contingen-experimenter saying ‘‘Sure, I can show it to cies of reinforcement were the same as in theyou’’ and giving the box to the child, even previous phases (except that for the secondif the child skipped the ﬁrst question (this question the experimenter said ‘‘No!’’) evenhappened only on two occasions with 1 if the child skipped one or two questions. Ifchild). the child did not ask any question, the ex- Teaching the third response form (‘‘Can I perimenter changed to a new box after 20 s.have it?’’). The procedure started the same If the three questions were a response chain,way as the ﬁrst and second response forms. the two responses emitted before removingWhen the child self-initiated ‘‘What’s in the the reinforcer (in this case, the questionsbox?,’’ the experimenter told him the name ‘‘What’s in the box?’’ and ‘‘Can I see it?’’)of the object inside but neither showed it should be extinguished and the questionnor gave it to him. When the child self-ini- emitted after the reinforcer should remaintiated ‘‘Can I see it?,’’ the experimenter (e.g., Skinner, 1934). If, alternatively, allshowed him the object inside the box but three response forms decreased or weredid not give it to him. The experimenter maintained together, this would indicatethen prompted the child to repeat ‘‘Can I that the three responses are instances of ahave it?’’ When the child repeated the ques- single response class. If the children respond-
ASKING QUESTIONS 289ed with an alternative pattern, that would response, even if the child skipped any ques-indicate alternative sources of control. tion; no prompts were provided. ‘‘No!’’ to the question ‘‘Can I see it?’’ and ‘‘No!’’ to the question ‘‘Can I have it?’’ Thishiding the box contingently. Hiding the box phase was introduced to study further thewas introduced when the response ‘‘No!’’ to role of this consequence on the recovery ofthe question ‘‘Can I see it?’’ did not decrease the question ‘‘Can I see it?’’ When the childthat question. After the child asked ‘‘Can I asked ‘‘Can I have it?,’’ the instructor saidsee it?’’ the experimenter said ‘‘No!’’ and hid ‘‘No!’’ and waited to see if the response ‘‘Canthe box behind her back. No prompts were I see it?’’ reemerged within the remaininggiven in this phase, and the contingencies of time. When the child asked ‘‘Can I see it?’’reinforcement remained the same for the experimenter said ‘‘Sure!’’ and gave the‘‘What’s in the box?’’ and ‘‘Can I have it?’’ box with the object inside to the child. Return to original contingencies of reinforce- These contingencies of reinforcement,ment. Conditions returned to the same con- slightly different from previous phases, weretingencies of reinforcement as in the phase applied even if the child skipped the ﬁrstof teaching the third question, but no question (‘‘What’s in the box?’’). Noprompts were provided. The child’s respons- prompts of any kind were given in thises were reinforced accordingly, even if he phase, and no additional time was given.skipped questions. For example if the child Measurement and Interobserver Agreementsaid ‘‘Can I have it?’’ before asking ‘‘What’sin the box?’’ the experimenter told him The ﬁrst self-initiated question of each re-‘‘Sure, you can have it’’ and gave him the sponse form constituted the behavior for that particular trial. The child, however, hadbox. If the child did not emit any question the opportunity to ask three questions perduring the 20-s period, the experimenter trial. The experimenter did not count thechanged the box and started a new trial. subsequent questions that sometimes fol-This phase was repeated intermittently lowed each response form (repetitions suchthroughout this investigation because we as ‘‘Can I see it? Can I see it?’’). The datawanted to know if the questions would re- on each of the questions were recorded sep-cover under the initial conditions of teach- arately by the experimenter and an observer.ing. An agreement was scored when both record- Some unpleasant objects. Conditions re- ed the same response or responses in all con-mained the same as in the previous phase, ditions within the 20-s trial. A disagreementwith the exception that the experimenter was scored when the experimenter and thepresented, in a random fashion, ﬁve trials observer recorded one or more different re-with unpleasant objects in one session (10 sponses. The observer was present on 31%trials). This phase was introduced to explore of the sessions for Jim, 31% of the sessionsfurther the conditions that maintained re- for Erik, and 45% of the sessions for Dario.sponding to the three questions. Interobserver agreement was 100%. All unpleasant objects. The experimenterpresented all 10 trials with unpleasant ob-jects to show the child. The purpose of this RESULTSphase was to see clearly the effects intended The 3 children learned to ask the threein the previous phase. The contingencies of questions. Although the various conditionsreinforcement for each question were the did not always produce the same effects insame as in the phase of teaching the third each child, all of them recovered and main-
290 GLADYS WILLIAMS et al.tained the three questions throughout the stopped asking the question (‘‘Can I see it?’’)study. This ﬁnding replicates that of Wil- in two sessions, whereas he continued to askliams et al. (2000). the other two questions. A return to the Figure 1 shows the results for Jim. During original contingencies did not recover thebaseline, Jim asked no questions in any of second question (‘‘Can I see it?’’). Then, thethe three forms. During the intervention to experimenters introduced a new phase toteach the ﬁrst question, he asked ﬁve to 10 study whether the child would ask this ques-questions in three sessions and maintained tion again. When the child asked ‘‘Can Ithe 10 questions (the maximum he could have it?,’’ the experimenter said ‘‘No!’’ andask in a given session) for seven more ses- did not give the object to the child. If hesions. He met acquisition criterion for the asked ‘‘Can I see it?’’ the experimenter toldsecond question rather fast, in that he went him ‘‘Sure, you can see it,’’ and gave himfrom one to 10 in two sessions. He acquired the object. In this phase, Erik again askedthe third question in two sessions. Thereaf- the second question (immediately reachedter, he continued asking these three ques- the maximum level of 10 questions per ses-tions at very high levels (nine or 10 ques- sion); moreover, he continued asking thetions per session). During the phase of third question (‘‘Can I have it?’’). In the‘‘No!’’ to the question ‘‘Can I see it?,’’ the phase with some unpleasant objects, Erikquestion decreased gradually to zero in the asked fewer questions for ‘‘Can I have it?’’fourth session, while the other two questions (six per session) and also fewer questions forremained at high levels (near 10). A return ‘‘Can I see it?’’ (from zero to four). He con-to the original contingencies brought back tinued to ask the ﬁrst question (‘‘What’s inthe second question to previous levels (near the box?’’) in every trial. The next phase10 questions per session) and maintained the consisted of showing only unpleasant ob-other two questions at similar levels. In the jects. Erik continued to ask the second ques-next phase, with some unpleasant objects, tion (‘‘Can I see it?’’) at a low frequency, andJim decreased the questions ‘‘Can I see it?’’ he almost stopped asking the third questionand ‘‘Can I have it?,’’ but he asked the ﬁrst (‘‘Can I have it?’’). In the return to the phasequestion (‘‘What’s in the box?’’) on all trials. with the original contingencies, Erik did notIn the next phase, the experimenter present- ask the second question, but he asked theed only unpleasant objects. Jim almost third question again, as he did in the pre-stopped asking the second and third ques- vious phase with the original contingencies.tions at the same time that he maintained For this reason, we again introduced thethe ﬁrst one in a consistent manner. A return same phase of ‘‘No!’’ to the question ‘‘Can Ito normal conditions brought back all three have it?’’ In only one session in this phase,questions to 10. Erik acquired the second question again. He Figure 2 shows results for Erik. Erik did asked the ﬁrst question (‘‘What’s in thenot ask any questions during baseline, but box?’’) consistently throughout all the phas-he learned to ask the three questions to reach es.levels of 10. He met acquisition criteria for Figure 3 shows Dario’s results. Duringthe ﬁrst question in eight sessions, the sec- baseline, Dario asked no questions in any ofond question in three sessions, and the third the three forms. Some difﬁculty was notedquestion in four sessions. The three question with Dario’s intelligibility; for this reason,forms were maintained for two more sessions we taught him to say ‘‘What’s that?’’ insteadat levels of 10 (or near 10). In the phase of of ‘‘What’s in the box?’’ During the inter-‘‘No!’’ to the question ‘‘Can I see it?’’ Erik vention, he met acquisition criterion for
ASKING QUESTIONS 291 Figure 1. Frequency of questions asked by Jim during the presentation of 10 boxes with hidden objectsinside during baseline, teaching of the three response forms, maintenance, removal of consequences for asking‘‘Can I see it?,’’ and the introduction of unpleasant objects (see text for details).
292 GLADYS WILLIAMS et al. Figure 2. Frequency of questions asked by Erik during the presentation of 10 boxes with hidden objectsinside during baseline, teaching of the three response forms, maintenance, removal of consequences for asking‘‘Can I see it?,’’ and the introduction of unpleasant objects (see text for details).
ASKING QUESTIONS 293 Figure 3. Frequency of questions asked by Dario during the presentation of 10 boxes with hidden objectsinside during baseline, teaching of the three response forms, maintenance, removal of consequences for asking‘‘Can I see it?,’’ and the introduction of unpleasant objects (see text for details).
294 GLADYS WILLIAMS et al.‘‘What’s that?’’ in eight sessions, ‘‘Can I see sult in identical effects in the three classes,it?’’ in four sessions, and ‘‘Can I have it?’’ in either maintenance or a reduction of thesix sessions. Subsequently, he asked these three responses at the same time. This effectthree questions in almost every trial. During did not occur. On the contrary, the secondthe subsequent condition (‘‘No!’’ to the response form (‘‘Can I see it?’’) decreasedquestion ‘‘Can I see it?’’), the response ‘‘Can (extinguished by the lack of the former con-I see it?’’ was not affected initially, as he con- sequence, punished by ‘‘No!,’’ or some com-tinued asking this question. We introduced bination), while the other two responsea new phase in which the experimenter said forms were maintained. Thus, it indicates‘‘No!’’ and hid the box immediately after he that this question was not in the same re-asked ‘‘Can I see it?’’ Asking this question sponse class as the other two questions. Thisdecreased, and it did not occur in the last manipulation, however, did not tell us any-three sessions. He continued to ask the other thing about whether the ﬁrst and the thirdtwo questions. In the subsequent phase response forms (‘‘What’s in the box?’’ and(some unpleasant objects), the ﬁrst question ‘‘Can I have it?’’) are members of the same(‘‘What’s that?’’) was maintained (at the fre- response class.quency of 10 questions in 10 trials). Mean- This performance also weakens the hy-while, he stopped asking the second (‘‘Can pothesis that the three responses were aI see it?’’) and the third (‘‘Can I have it?’’) chain of responses. Had they constituted aquestions. The experiment was then discon- response chain, the ﬁrst response (‘‘What’stinued for external reasons. in the box?’’) would have decreased as a re- sult of removing the reinforcer that main- tained the subsequent responses in the chain. DISCUSSION Thus, maintenance of that question suggests The 3 children learned to ask the three that the three questions were not membersquestions. These results replicate those of of a response chain.Williams et al. (2000). Once all the children The second manipulation consisted of in-acquired the three questions, we applied dif- troducing unpleasant objects on some or allferent consequences to two questions to see trials. In those conditions, the children askedhow the other questions were affected. ‘‘Can I see it?’’ and ‘‘Can I have it?’’ less andWhen we responded ‘‘No!’’ to ‘‘Can I see less frequently. Still, they asked ‘‘What’s init?,’’ 2 children gradually decreased asking the box?’’ when new boxes were presented.this question, whereas they continued to ask This response pattern rules out the possibil-the other two questions. The other child de- ity that the ﬁrst and third questions formedcreased this question after an additional pro- a single response class, because one of themcedure was implemented in a subsequent remained at a high frequency while the otherphase (the experimenter said ‘‘No!’’ and hid dropped (for the same reasons as in the ﬁrstthe box). He also continued to ask the other manipulation). This response pattern alsotwo questions. rules out the possibility that the three re- The response pattern displayed when the sponses were members of a chain. Had theyexperimenter said ‘‘No!’’ to the question been members of a chain, the three respons-‘‘Can I see it?’’ weakens the hypothesis that es would have been maintained by the ﬁnalthe three question types constitute a re- reinforcer (the object). Then, the removal ofsponse class. Had they constituted a re- the object should have resulted in the ex-sponse class, removing the consequence (the tinction of the three responses. This did notputative reinforcer) of a response should re- occur. Thus, this response pattern in one
ASKING QUESTIONS 295sense rules out the possibility that the three maintained the second question. The thirdresponses were members of a response chain. response form (‘‘Can I have it?’’) was prob- Another way to interpret the results is that ably maintained by the access to the object.question asking served the function ascribed The antecedent stimuli of the three ques-by the contingencies of reinforcement in any tions also changed. The antecedent stimulusgiven condition. For example, it may be that for the ﬁrst question was the presentation ofsaying ‘‘Can I see it?’’ originally was rein- the box. The antecedent stimulus for theforced by gaining access to the item, but second question could have been the exper-eventually it was reinforced by merely seeing imenter’s response to the ﬁrst question,the item. These differential contingencies which established the occasion to reinforcewere programmed and, hence, the operant that response with the view of the object.function of the questions necessarily shifted The antecedent stimulus for the third ques-in response to newly arranged contingencies tion could have been the response to the sec-of reinforcement. ond question. However, when the second The results observed after the teaching question was no longer reinforced, the thirdphases suggest that each question was main- response was emitted just after the experi-tained by its own consequence. The ﬁrst menter responded to the ﬁrst question. Thisquestion (‘‘What’s in the box?’’) was main- fact indicates that the antecedent stimulustained by the answer the child received to for the third response was not precise. In-that question. The experimenter’s responses stead, it seems that the response to the ﬁrstmaintained the question independently of question established the conditions for thethe fact that the experimenter said the name onset of the second and third questions.of a pleasant object (a potential reinforcer) In summary, these data show that each ofor an unpleasant object. We may suppose the three questions ultimately was main-that ‘‘What’s in the box?’’ would have been tained by different consequences thatextinguished if the experimenter discontin- changed over conditions according to chang-ued responding to that question by not giv- es in consequences. This fact indicates thating the name of an object (either pleasant or the questions did not belong to a responseunpleasant). We did not manipulate this var- class, because not all the questions wereiable in the current study. maintained by the same reinforcer—in this The second response form (‘‘Can I see case, access to the object. The three ques-it?’’) could be maintained by the view of the tions were not a response chain becauseobject and the possibility of asking the third when one question decreased, regardless ofquestion. The view of the object played a which one, the other two were maintainedrole. In fact, when the consequence in that at high levels. Given the ﬂexibility of re-phase was not to have the opportunity to sponding according to a variable environ-view the object, the question decreased. ment, the teaching methods used here ap-However, the children still had the oppor- pear to be sufﬁciently versatile to teach chil-tunity to ask the third question (‘‘Can I have dren with autism question asking thatit?’’). The 3 children continued to ask this matches the functions of question asking inquestion, even though they did not ask the everyday life.second one. This fact suggests that the con-sequence that maintained the second ques-tion was the view of the object. Also, it sug- REFERENCESgests that the opportunity to ask the third Donahoe, J., & Palmer, D. (1994). Learning andquestion was not the consequence that complex behavior. Boston: Allyn and Bacon.
296 GLADYS WILLIAMS et al.Koegel, L. K., Camarata, S. M., Valdez-Menchaca, M., ence of the Association for Behavior Analysis, & Koegel, R. L. (1998). Setting generalization of Washington, DC. question asking by children with autism. American Taylor, B. A., & Harris, S. L. (1995). Teaching chil- Journal on Mental Retardation, 102, 346–357. dren with autism to seek information: AcquisitionShabani, D. B., Katz, R. C., Wilder, D. A., Beau- of novel information and generalization of re- champ, K., Taylor, C. R., & Fischer, K. J. (2002). sponding. Journal of Applied Behavior Analysis, 28, Increasing social initiations in children with au- 3–14. tism: Effects of a tactile prompt. Journal of Applied Williams, G., Donley, C., & Keller, J. (2000). Teach- Behavior Analysis, 35, 79–83. ing children with autism to ask questions aboutSkinner, B. F. (1934). The extinction of chained re- hidden objects. Journal of Applied Behavior Anal- ﬂexes. Proceedings of the National Academy of Sci- ysis, 33, 627–630. ences, 20, 532–536.Sundberg, M. (2000, May). Teaching mands for in- Received June 18, 2002 formation by manipulating the establishing opera- Final acceptance March 26, 2003 tion. Paper presented at the 26th annual confer- Action Editor, Timothy R. Vollmer STUDY QUESTIONS 1. What explanations did the authors propose to account for the ﬁndings of the Williams et al. (2000) study, in which children asked a series of related questions about something? 2. Describe the general procedure used to train participants to use the initial questions. 3. What was the rationale for the condition in which the experimenter responded ‘‘No!’’ to the question ‘‘Can I see it?’’ 4. What was the rationale for the condition in which unpleasant items were placed in the boxes? 5. Describe the general effects observed on question asking during (a) the condition in which the therapist responded ‘‘No!’’ to the question ‘‘Can I see it?’’ and (b) the condition in which some unpleasant items were placed in the boxes. 6. Erik was the only participant exposed to the condition in which the experimenter said ‘‘No!’’ to ‘‘Can I have it?,’’ and he continued to ask this question. How would you account for these results? 7. What conclusions did the authors draw regarding maintaining variables for the three ques- tions? 8. What implications do the test conditions and results of the curent study have for clinical use? Questions prepared by Carrie Dempsey and Stephen North, The University of Florida