THE ROLE OF NAMING IN STIMULUS CATEGORIZATION BY PRESCHOOL CHILDREN

CAIO F. MIGUEL, ANNA I. PETURSDOTTIR, JAMES E. CARR, AND JACK MICHAEL

WESTERN MICHIGAN UNIVERSITY

The purpose of the current study was to assess whether children would categorize pictures when taughtthe relevant listener and speaker behaviors separately. A category-sort test was used to assess emergentconditional relations. Category-sort trials consisted of looking at (Test 1) or tacting/labeling (Test 2) asample stimulus and selecting the appropriate comparison stimuli. In Experiment 1, 4 children (3–5 years) were taught to tact pictures of six U.S. state maps as either north or south. An assessment wasconducted to determine whether they would (1) correctly categorize or sort when presented with avisual sample and (2) select the correct stimuli when hearing their category names (listener behavior).Two of the children categorized the pictures during Posttest 1 after the initial (pairwise) tact training.The other 2 categorized after receiving additional tact training with all pictures presented together.However, one of them categorized only during Posttest 2. In Experiment 2, 4 children (3–5 years) weretaught to select pictures when hearing their category names. An assessment was conducted to determinewhether they would (1) correctly categorize or sort and (2) tact the stimuli (speaker behavior). Onechild categorized the pictures during Posttest 1, and two during Posttest 2. The other child requiredadditional training with all pictures grouped together. When participants failed to categorize, they alsofailed to tact the pictures accurately. Taken together, results from Experiments 1 and 2 show that bothspeaker and listener behavior play an important role in stimulus categorization.

Key words: naming, categorization, stimulus equivalence, verbal behavior, preschool children

_______________________________________________________________________________

The process of determining how to groupobjects or events together is usually calledcategorization or classification, while thoseobjects or events that cohere may be regardedas a category or a class. It often is assumed thatthe categorization process is dependent uponthe acquisition of specific ‘‘concepts’’ (Barsa-lou, 1992). These concepts are said to be unitsof mental representation and seem to existindependently of any behavior–environmentrelation (Zentall, Galizio, & Critchfield, 2002).

In contrast, a concept also could be defined asa group of objects (e.g., stimuli, actions) thatcontrol similar responses. When an individualresponds similarly to each object in a group ofobjects, these objects constitute a class, whichcan then be called a concept (Keller &Schoenfeld, 1950). As a result, concepts maybe equated to stimulus classes.

Interesting examples of stimulus classes arethose whose members do not share physicalsimilarity. Examples of such classes include therelation among pictures, printed words, andspoken words. These stimuli may becomesubstitutable for each other under specificconditions. When asked to point to ‘‘ball,’’someone might point to an actual ball, thepicture of a ball, or the printed word ‘‘ball.’’However, if asked to kick the ball, only onestimulus (i.e., the actual ball) would functionas an effective discriminative stimulus for sucha response (Green & Saunders, 1998). Inter-estingly, after being taught to respond to somemembers of a class, humans may behavesimilarly in the presence of other classmembers without being directly trained to doso. The search for an understanding of thevariables responsible for this emergent reper-toire is what drives research in the area ofstimulus equivalence (Green & Saunders; Sid-man, 1994).

Caio Miguel is now at California State University,Sacramento. Anna Petursdottir is now at Texas ChristianUniversity.

This study is based on a dissertation submitted by thefirst author to the Department of Psychology at WesternMichigan University in partial fulfillment of the require-ments for the Ph.D. degree. This research was funded bythe Society for the Advancement of Behavior Analysis –Sidney and Janet Bijou Fellowship, and the WesternMichigan University Graduate Research Fund, bothawarded to the first author. We thank Deborah Johnstonand the staff at Little Lambs Childcare Center inKalamazoo, MI for their invaluable on-site support, as wellas Carol Pilgrim and Bill Ahearn for their comments onearlier versions of the manuscript. We also thank AmandaMahoney, Gina Savo, Tyler Weeks, and Kaneen Geiger fortheir assistance with data collection.

Address correspondence to Caio Miguel, Department ofPsychology, California State University, Sacramento, 6000 JSt., Sacramento, CA 95819-6007 (e-mail: miguelc@csus.edu).

doi: 10.1901/jeab.2008-89-383

JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR 2008, 89, 383–405 NUMBER 3 (MAY)

383

Research on stimulus equivalence has beenprevalent in the experimental analysis ofbehavior for years, but there is still noconsensus regarding the mechanism responsi-ble for the emergent behavior observed duringequivalence tests (Saunders, Williams, & Spra-dlin, 1996). According to Horne and Lowe(1996), the formation of equivalence classes isheavily dependent upon the subject’s verbalrepertoire, or more specifically, upon what isdescribed as the naming relation. Horne andLowe consider naming to be a higher-orderclass of behavior that involves a bidirectionalrelation between ‘‘a class of objects and eventsand the speaker–listener behavior they occa-sion’’ (p. 200). The necessary conditions forthe development of naming are presentduring typical child–caregiver interactions.When naming is acquired, the presence ofone member of the class evokes a tact (i.e., averbal response evoked by a nonverbal SD), theproduct of which (e.g., auditory stimulus) inturn evokes the listener behavior of reorient-ing and selecting other members that are partof the same name relation. It follows that toestablish arbitrary stimulus classes, one wouldneed only to learn how to name each memberof the designated class.

Despite suggestions that the development ofequivalence classes may not depend on theability of humans to name or label stimuli(e.g., Sidman, Cresson, & Wilson-Morris, 1974;Sidman, Kirk, & Wilson-Morris, 1985; Sidman& Tailby, 1982; Sidman, Wilson-Morris, & Kirk,1986), and the evidence for the developmentof stimulus equivalence in nonhuman subjectsand individuals with minimal verbal reper-toires (e.g., Brady & McLean, 2000; Carr,Wilkinson, Blackman, & McIlvane, 2000; Shus-terman & Kastak, 1993), it is not unreasonableto suggest that the development of equiva-lence classes may be influenced by language.Some reports have suggested a positive corre-lation between verbal competence and suc-cessful performance on equivalence tests (e.g.,Barnes, McCullagh, & Keenan, 1990; Devany,Hayes, & Nelson, 1986; Eikeseth & Smith,1992). Additionally, research has demonstrat-ed that the development of equivalence may atleast be influenced, if not determined, bylanguage (e.g., Eikeseth & Smith; Goyos, 2000;Horne, Lowe, & Randle, 2004; Lowe, Horne,Harris, & Randle, 2002; Mandell & Sheen,1994; Randell & Remington, 1999, 2006). It is

possible that equivalence class formation inhumans could be either verbally mediated(e.g., rule-governed) or contingency-shaped(de Rose, 1996). Hence, if verbal behavior canmediate the formation of stimulus classes, itsdirect manipulation should generate or atleast improve performance on tests for equiv-alence or categorization.

Lowe et al. (2002) used a stimulus-sortingprocedure to evaluate the development ofequivalence classes. They demonstrated thatstimulus classes could be established withoutthe use of conventional matching-to-sampleprocedures by teaching children to tact eachmember of the target stimulus class. In theirfirst experiment, 9 typically developing 2- to 4-year-old children learned a common vocalresponse (‘‘zag’’) for three arbitrary shapes,and another (‘‘vek’’) for three additionalshapes. Following tact training, children wereexposed to a categorization test. During thesecategory-sort trials, each child was presentedwith all six shapes; the experimenter selectedone of them and asked the child to ‘‘give theothers’’. A correct category sort was scored if,when presented with a sample stimulus (eithera ‘‘zag’’ or a ‘‘vek’’), the child selected the twopositive comparisons. Lowe et al. reported thatafter training, 4 children passed the categori-zation test (passing criterion was four out ofnine correct sorts per common tact category).In the second experiment, the remaining 5children were exposed to another categoriza-tion test in which they were required to tactthe sample before selecting the appropriatecomparisons. All 5 participants passed thesecond categorization test. Moreover, 2 chil-dren demonstrated class expansion to two 6-member classes via tact training. The authorsalso attempted to control for the possibilitythat having all stimuli presented togetherduring tact training facilitated stimulus classformation. Three children were exposed totact training in pairwise trials only. Once thetraining criterion was reached, the childrenwere tested for categorization. Two childrenpassed the first categorization test and onepassed the second. Additionally, when probedfor listener behavior, all children from thesecond experiment were able to select thestimuli even though they had never beentrained to do so.

In a follow up study, Horne et al. (2004)taught 9 children (1 to 4 years old) to select

384 CAIO F. MIGUEL et al.

specific arbitrary shapes in the presence of thewords ‘‘zog’’ and ‘‘vek’’ in a tabletop two-comparison auditory–visual matching pre-paration. After this training (i.e., listenertraining), none of the children categorizedthe pictures when asked to look at the samplebefore selecting two of the five comparisons, asin Lowe et al. (2002). When probed for tacts, 7children failed to tact the stimuli correctly.However, after being directly trained to tactthe stimuli, 2 of these children passed catego-rization tests when required to look at thesample and 3 passed categorization tests whenrequired to tact the sample. Two children weretrained on a second stimulus set. After listenertraining, one child passed the first categoriza-tion test, and the other passed the secondcategorization test only after direct tact train-ing.

A series of recent studies (Horne, Hughes, &Lowe, 2006; Horne, Lowe, & Harris, 2007;Lowe, Horne, & Hughes, 2005) replicated andextended previous findings by showing thatcommon naming was effective in 1) establish-ing arbitrary stimulus classes in children asyoung as 1 year and 7 months, and 2) bringingabout transfer of function among members ofa class. In the Lowe et al. (2005) study, 9children (1 to 3 years old) learned commontacts for two experimentally defined three-member classes. They also were taught to clapor wave to one of the stimuli from each class.Participants demonstrated listener behavior,categorization, and emitted the specifiedmotor response in the presence of other classmembers without direct training. When thesesame motor responses were presented by theexperimenter, participants were able to selectthe corresponding stimuli. Furthermore, whenmotor responses were trained in the presenceof new sets of stimuli, children correctlynamed, selected, and categorized the corre-sponding stimuli without training.

In Horne et al. (2006), 14 children (1 to3 years old) learned to select specific arbitraryshapes in the presence of the words ‘‘zog’’ and‘‘vek’’. When tacts were probed, 4 childrenfailed to tact the stimuli correctly. As in theprevious study, participants learned to clap orwave in response to one of the stimuli fromeach class. Participants who demonstratedboth listener and speaker (tact) behavior, wereable to categorize the shapes accurately, as wellas emit the specified motor response in the

presence of other class members. Childrenwho failed to tact, in contrast, were only able tocategorize and emit motor responses in thepresence of other class members once tactswere directly trained.

In Horne et al. (2007), 8 children (2 to4 years old) were able to categorize afterlearning to emit a common manual sign inthe presence of each member of the stimulusclass. In addition, transfer of function wasdemonstrated when after learning to tact oneexemplar of each class, children correctlytacted the remaining stimuli. The authorsconcluded that together, these studies addsupport to the notion that it is only whenchildren acquire both speaker and listenerrelations that they are able to categorize.

Given the recent literature on the impor-tance of verbal behavior in the development ofequivalence classes, and the availability of newmeasures of stimulus categorization (e.g.,Horne et al., 2004, 2006, 2007; Lowe et al.2002, 2005; Pilgrim & Galizio, 1996), thecurrent study was designed to replicate andextend the work of Horne et al. (2004) andLowe et al. (2002) by assessing whether slightlyolder, typically developing children (3 to5 years old) could categorize unfamiliar pic-tures when taught listener and speaker behav-iors separately. Training and testing proce-dures were similar to those of previous studies,but there were several important methodolog-ical differences that are worth mentioning.First, whereas previous studies used completelyabstract stimuli, the current study employedunfamiliar two-dimensional pictures (maps)and their corresponding category names.These stimuli and categories were selected tosimulate what might be taught to preschoolchildren within a natural context. Second, amultiple-baseline design was used to controlfor exposure to experimental stimuli prior tothe onset of the study, and history of exposureto the experimental stimuli during probes.Third, listener training and testing was con-ducted with three, rather than two comparisonstimuli to 1) decrease the likelihood thatconditional discriminations would be con-trolled by the incorrect comparison (i.e., rejectrelation), and 2) decrease the probability thathigh accuracy was a product of chanceresponding. Fourth, the categorization–tacttest (tact sample match-to-others) was con-ducted with all participants to assess whether

NAMING 385

vocal tacts could have been emitted duringcategorization tests.

Two experiments were conducted. In thefirst experiment, children were taught todirectly tact (using a common name) individ-ual members of specific categories (categoriesincluded maps of northern and southern U.S.states) and were assessed for the ability to (1)sort them into groups without training and (2)select the correct pictures when given thecommon names (categories). In the secondstudy, children were directly taught to selectpictures when given their common names andwere assessed to see whether they were able to(1) sort them into groups without directtraining and (2) correctly tact them.

EXPERIMENT 1METHOD

Participants, Setting, and Materials

Four typically developing children, Tom,Adam, Rita, and John participated in Experi-ment 1. Their ages were 5 years 3 months,4 years 6 months, 4 years 2 months, and3 years 1 month, respectively, at the beginningof the study. Participants had no knownlearning or developmental disabilities andpossessed verbal skills that were appropriateto their age. Sessions were conducted behind adivider in a quiet area at the children’spreschool. During each session, the child andexperimenters sat on the same side of a smalltable to avoid any inadvertent cueing by theexperimenter. Sessions lasted approximately10 min and were conducted twice each week-day.

Materials included six unfamiliar pictures(state maps) and six familiar pictures. Familiarpictures were 5 cm by 5 cm color photographsof objects on a white background, obtainedfrom the Picture ThisE CD-ROM. Each pic-ture was encased in a transparent hard-plasticcover measuring 7.5 cm by 10 cm. Unfamiliarpictures were 6.5 cm by 6.5 cm laminatedblack-and-white maps on a white background,obtained online from the MicrosoftE Clip-art database. Each trial was presented byplacing the pictures horizontally on thetable. As each trial was completed, theexperimenter rearranged the pictures on thetable and presented the next trial. Figure 1depicts the unfamiliar pictures used duringtraining.

Dependent Variables and Data Collection

The main dependent measure was thepercentage of correct category sorts out of sixtrials. A correct sort was scored when in thepresence of the sample (e.g., N11), the childselected the two correct comparisons (e.g., N2and N3) from a five-stimulus array (e.g., N2,N3, S1, S2, S3). Selection during categoriza-tion consisted of removing the picture fromthe table and giving it to the experimenter.Data on the number of correct category sortswere collected during probe sessions.

Additional dependent measures included(1) selecting a stimulus in response to aninstruction given by the experimenter (i.e.,listener behavior) and (2) participants’ vocal–verbal descriptions of their performance.Listener tests were administered prior to tacttraining and following categorization tests.Vocal–verbal behavior was assessed duringsessions via digital recording, as well as whileanswering questions following successful com-pletion of categorization tests. The experi-menter asked participants to answer open-ended questions regarding the categorizationtask. During all training and probe sessions, adigital voice recorder was positioned close tothe children in an unobtrusive manner. Dataon vocal–verbal behavior were later tran-scribed from the digital recorder.

Interobserver Agreement (IOA)

A second observer was present for at least37% of all categorization tests, listener tests,and tact training sessions to assess IOA. Foreach category-sort and tact trial, either anagreement or a disagreement on the depen-dent measures between the two observers wasscored. Point-by-point agreement was calculat-ed by dividing the number of agreements bythe sum of agreements and disagreementsmultiplied by 100%. IOA across conditionsaveraged 98% (range 83–100%) for Tom, 99%(range 83–100%) for Adam, 99% (range 83–100%) for Rita, and 99% (range 87–100%) forJohn.

Experimental Design

A nonconcurrent multiple-baseline designacross 2 participants (Watson & Workman,

1 N and S refer to the experimenter-defined categoriesNorth and South, respectively. Exemplars from eachcategory were numbered (1, 2, and 3).

386 CAIO F. MIGUEL et al.

1981) was used to assess the effects of theindependent variable on the percentage ofcorrect categorizations. Much like a concur-rent multiple-baseline design, each pair ofparticipants started the experiment at approx-imately the same time and sessions wereconducted with each participant twice a day.The design was chosen so that individualparticipant absenteeism, as well as individualdifferences in training (i.e., lengthy trainingphases with one participant but not the other)would not affect the other participant’s pro-gress through the experiment.

Two different categorization tests were ad-ministered after training criterion was achievedin experimental conditions. Test 1 was a look-at-sample categorization test, and Test 2 consisted of atact-sample categorization test. A description ofeach of these tests is provided below.

The order of conditions was as follows:Categorization 1 and 2 pretests, listener-1

pretest, tact-1 training, categorization posttests1 and 2, listener-1 posttest. For those partici-pants who failed to categorize after tact-1training, the following additional conditionswere presented: listener-2 pretest, tact-2 train-ing, categorization 1 and 2 posttests, listener-2posttest. Interviews were conducted followingsuccessful categorization posttests.

Prior to baseline, participants were exposedto pretraining with familiar stimuli to controlfor the possibility that subsequent failure tocategorize with arbitrary stimuli was due tolack of instructional control. Table 1 summa-rizes the experimental conditions.

Pretraining

Tact-1 pretraining. During tact-1 pretrain-ing, the experimenter randomly separated thefamiliar pictures into three training pairs.Each pair contained a picture of a bird and apicture of an insect. Pictures were placed

Fig. 1. Unfamiliar pictures. The North, South, and distracter maps are displayed in the top, middle, and bottompanels, respectively.

NAMING 387

horizontally on the table. The experimenterpresented the first pair to the child and whilepointing to one of the stimuli said, ‘‘What isthis?’’ If the child produced the correct tactresponse for the category (i.e., saying ‘‘bird’’or saying ‘‘insect’’), the experimenter deliv-ered praise (e.g., ‘‘Good job!’’). If the childresponded incorrectly or did not respond, theexperimenter pointed to the stimulus andsaid, ‘‘This is a bird [insect], can you say it?’’If the child repeated the name of the category,the experimenter said, ‘‘That’s right’’, or‘‘OK’’ and began the next trial. On any giventrial, only one picture in the pair was targeted.Pictures of birds and insects were presentedunsystematically either on the right or on theleft of the child in 8-trial blocks (each picturewas targeted twice on the left and twice on theright). The training criterion for each pair wascorrectly tacting all pictures in two consecutive8-trial blocks. After training pair 1, theprocedure was repeated for pairs 2 and 3.Tact-1 pretraining was completed when re-sponding met criterion for all pairs.

Tact-2 pretraining. During this condition,training was conducted with all pictures atonce. The three pairs of pictures presentedduring Tact-1 pretraining were combined in asix-picture array. The pictures were randomlyordered in a row and placed in front of the

child. The experimenter pointed to one of thepictures and asked, ‘‘What is this?’’ If the childproduced the correct tact response, theexperimenter delivered praise. If the childresponded incorrectly or did not respond, theexperimenter pointed to the stimulus andsaid, ‘‘This is a bird [insect], can you say it?’’If the child repeated the name of the category,the experimenter acknowledged the correctresponse and pointed to the next picture.After all pictures were targeted (six trials), theexperimenter showed them in a differentorder and started the training again. Tact-2training was completed when the child cor-rectly tacted all pictures in two consecutivearrays (six-trial blocks) with no corrections.

Listener-1 pretraining. Listener training re-fers to the behavior of selecting/pointing tovisual stimuli in the presence of an instructionprovided by the experimenter. This task hasalso been referred to as receptive discrimina-tion (e.g., Miguel, Petursdottir, & Carr, 2005).During this condition, the experimenter ran-domly separated the pictures into three pairs.Each pair contained a picture of a bird and apicture of an insect. Three additional picturesfrom different categories were used as distrac-ters (e.g., tools, kitchen items). One distracterpicture was randomly selected and added toeach pair as the third comparison during each

Table 1

Experimental Phases – Experiment 1

Phase Task Stimuli per trial Trials per block Training Criterion

Pretraining Tact-1 (pairs) 2 8 2 blocks at 100%Pretraining Tact-2 (all stimuli) 6 6 2 blocks at 100%Pretraining Listener-1 3 6 2 blocks at 100%Pretraining Listener-2 6 6 2 blocks at 100%Pretraining Categorization Pretraining 6 6 2 blocks at 100%1 Categorization Pretest 1 6 6 N/A

Categorization Pretest 2 6 6 N/A2 Listener-1 Pretest 3 6 N/A3 Tact-1 Training (pairs) 2 8 3 blocks at 100%4 Cat. Pretraining Review 6 4 1 block at 100%5 Categorization Posttest 1 6 6 N/A

Categorization Posttest 2 6 6 N/A6 Listener-1 Posttest 3 6 N/A7a Listener-2 Pretest 6 6 N/A8a Tact-2 Training (all stimuli) 6 6 3 blocks at 100%9a Cat. Pretraining Review 6 4 1 block at 100%10a Categorization Posttest 1 6 6 N/A

Categorization Posttest 2 6 6 N/A11a Listener-2 Posttest 6 6 N/A12b Posttest Interview N/A N/A N/A

a Children were exposed to these phases only if they failed the previous categorization test.b Children were exposed to this phase immediately after successful categorization tests.

388 CAIO F. MIGUEL et al.

trial. The three-choice matching procedurewas chosen to decrease the probability ofconditional discriminations coming undercontrol of an incorrect comparison (i.e., areject relation), which could result in failureon subsequent equivalence tests (de Rose,1996).

The experimenter presented the first threecomparisons (set 1) to the child and asked,‘‘Can you give me the bird [insect]?’’ If thechild selected the correct picture, and handedit to the experimenter, praise was delivered. Ifthe child responded incorrectly or did notrespond, the experimenter pointed to thecorrect picture and said, ‘‘This is the correctone’’. Training for each set was arranged insix-trial blocks in which pictures of birds andinsects were presented in a quasirandomorder, each picture serving as the correctcomparison once to the right, once in themiddle, and once to the left of the child. Thetraining criterion was achieved when a childresponded correctly to all bird and insectcomparisons in two consecutive six-trial blocks.After training set 1, the procedure was repeat-ed for sets 2 and 3. Listener-1 training wascompleted when responding was at criterionfor all trained sets.

Listener-2 pretraining. During this condition,the three pairs of pictures presented duringlistener-1 pretraining were combined in a six-picture array, and selection responses weretrained in this context. Distracters were omittedduring this phase. The pictures were randomlyordered in a row and placed in front of thechild. The experimenter said to the child, ‘‘Canyou give me all the birds [insects]?’’ If the childselected the three correct comparisons (eitherthe three pictures of birds or the three picturesof insects), the experimenter provided praise.Selecting the three correct comparisons wasconsidered a correct response. If the childselected an incorrect comparison stimulus, theexperimenter pointed to the correct compari-sons and said, ‘‘These are the correct ones’’. Ifthe child selected some but not all three correctcomparisons, the experimenter prompted thenext response by saying, ‘‘Are there any more?’’When the child selected the correct compari-son(s) after the prompt, the experimenterdelivered praise and then provided the promptagain to avoid the possibility that the number ofpictures selected was being controlled by theprompt. When the child did not respond after

the prompt, the experimenter pointed to theremaining stimuli and said, ‘‘These are thecorrect ones’’. Bird and insect trials wereinterspersed in six-trial blocks. Training con-tinued until the child performed at 100%(responded correctly in all trials) in twoconsecutive blocks.

Categorization pretraining. During this con-dition, participants were taught to respond totwo different tasks that were later used to testcategorization with unfamiliar pictures. Thetwo tasks were a look-at-sample categorizationtest and a tact-sample categorization test.

During the look-at-sample categorizationtest, the experimenter placed all six picturesin a predetermined location in front of thechild, picked one at random (randomizationwas predetermined and indicated on thedatasheet), and asked the child, ‘‘Look at this,can you give me the others?’’ If the childselected the two correct comparisons (twobirds or two insects), the experimenter pro-vided verbal praise. Selecting the two correctcomparisons was considered a correct catego-ry-sort trial. If the child selected an incorrectcomparison stimulus, the experimenter point-ed to the comparison and said, ‘‘This is aninsect [a bird].’’ If during category-sort trialsthe child only selected one correct compari-son, the experimenter prompted the nextresponse by saying, ‘‘Are there any more?’’ Ifthe child selected the remaining correctcomparison after the prompt, the experiment-er delivered verbal praise. After each prompt-ed response, the experimenter provided theprompt again. Trials of birds and insects wereinterspersed in six-trial blocks in which eachpicture served as a sample once. Trainingcontinued until the child responded correctlyon all trials for two consecutive blocks.

During the tact-sample categorization test,the experimenter selected a picture and asked,‘‘What is this?’’ If the child produced thecorrect tact response by saying either ‘‘bird’’or ‘‘insect,’’ the experimenter continued byasking, ‘‘Can you give me the others?’’Consequences for correct and incorrect selec-tions were the same as in the look-at-sampletest. Training continued until the child per-formed at 100% in two consecutive blocks.

Experimental Procedures

Categorization tests. This condition was de-signed to assess conditional relations among

NAMING 389

stimuli labeled as ‘‘north’’ and ‘‘south.’’Categorization tests were administered duringbaseline (pretests) and immediately aftertraining (posttests). Prior to categorizationposttests, participants were reexposed to cate-gorization training with familiar pictures untilthey performed correctly on four consecutivetrials. This review was conducted to ensure thatperformance on categorization tests was underappropriate instructional control.

During test 1, the experimenter used thelook-at-sample instruction. All six pictureswere placed in a predetermined prerando-mized location in front of the child. Theexperimenter selected one picture and said,‘‘Look at this. Can you give me the others?’’ Ifthe child selected the two correct compari-sons, the experimenter waited for five secondsand if no additional pictures were selected, theresponse was recorded as correct, but noconsequences were provided. If during a trialthe child selected only one correct compari-son, the experimenter prompted anotherresponse by saying, ‘‘Are there any more?’’After each prompted response, the experi-menter repeated the prompt. If the childselected the correct stimuli, the trial wasscored as correct. If the child selected all fivecomparison stimuli, the experimenter provid-ed feedback to the child by saying, ‘‘I don’twant all of the pictures, just some of them’’,and scored that trial as incorrect. Testing wasconducted in six-trial blocks, with each pictureserving as a sample once per block. Thelocation of comparison stimuli was counter-balanced across sessions. In each trial, theexperimenter showed a different picture untilall pictures had served as a sample. Thecriterion for ‘‘passing the test’’ was set at threeconsecutive blocks at 66% (four out of sixcorrect categorizations) or higher. The criteri-on for failure was set at three consecutiveblocks at 33% (two out of six correct catego-rizations) or below.

During test 2, the experimenter used thetact-sample instruction. All procedures werethe same as in test 1 except for the instruction.The experimenter asked the child, ‘‘What isthis?’’ and waited 5 s for the child to tact thesample stimulus. Regardless of the child’sresponse, the experimenter proceeded byasking, ‘‘Can you give me the others?’’ Trialswere scored as correct when the child selectedthe two correct comparison stimuli that were

consistent with the sample regardless of hertact performance. Categorization passing andfailure criteria were the same as in test 1.

Listener-1 pretest/posttest. The purpose ofthis condition was to assess whether partici-pants were able to select a specific picture inresponse to a verbal stimulus specifying thecategory to which that picture belonged. Theexperimenter randomly separated the picturesinto three pairs. Three additional maps ofCanadian provinces that were not part oftraining were used as distracters (see Fig-ure 1). Each trial included a map of anorthern state, a map of a southern state,and a distracter as comparisons. The experi-menter presented the first set to the child andasked, ‘‘Give me the North [South]?’’ Re-sponses were recorded, but no consequenceswere provided. Testing for each set wasarranged in a six-trial block in which picturesof northern and southern states were present-ed in a quasirandom order, each pictureserving as the correct comparison once onthe right, once in the middle, and once on theleft of the child. After testing set 1, theprocedure was repeated with sets 2, 3, andwith the pictures reassigned to new groups(Mixed).

Tact-1 training. The purpose of this condi-tion was to train participants to tact thecategories to which pictures belonged. Thecondition was similar to tact pretraining.Pictures were separated into three trainingpairs (N1S1, N2S2, N3S3). The experimenterpresented the first pair to the child, and whilepointing to one of the stimuli asked, ‘‘What isthis?’’ When the child produced the correctvocal response, the experimenter deliveredpraise. When the child responded incorrectlyor did not respond, the experimenter pointedto the stimulus and said, ‘‘This is North[South], can you say it?’’ If the child repeatedthe name of the category, the experimenteracknowledged the correct response and pre-sented the next trial. Only one picture in thepair was targeted per trial. Training wasarranged in eight-trial blocks in which Northand South pictures were presented unsystem-atically either on the right or left of the child.Training criterion for each pair was correctlytacting all pictures in three consecutive eight-trial blocks (each picture being targeted twiceon the left and twice on the right of the child).After training pair 1, the procedure was

390 CAIO F. MIGUEL et al.

repeated for pairs 2 and 3. When respondingwas at criterion for all trained pairs, the sixstimuli were randomly reassigned to three newmixed training pairs. Training criterion for thenew pairs was the same as for the initial pairs.When criterion was met, the probability ofreinforcement was reduced from 100% to 50%(responses on every other correct trial werereinforced). When the 50% criterion was met,the reinforcement probability was reduced to0% (responses on correct trials were no longerreinforced). Once performance was main-tained without reinforcement, training wasterminated and categorization tests were ad-ministered. Probability of reinforcement wasreduced so that performance observed duringsubsequent tests (conducted under extinc-tion) could not be attributed to a suddenchange in the frequency of reinforcement.

Listener-2 pretest/posttest. During this condi-tion, participants were exposed to all picturesat once. The North and South stimuli werecombined in a six-picture array, and selectionresponses to each picture were recorded. Thepictures were randomly ordered in a row andplaced in front of the child. The experimenterasked the child, ‘‘Can you give me all of theNorth [South]?’’ If the child selected the threecorrect comparisons, the trial was scored ascorrect, but no consequences were delivered.No additional instructions or prompts wereprovided. Testing was arranged in six-trialblocks in which North and South questionswere presented in a quasirandom order threetimes each. A minimum of two blocks or 12trials were administered.

Tact-2 training. In this condition, the sixNorth and South pictures were randomlyordered in a row and placed in front of thechild. The experimenter pointed to eachpicture and asked, ‘‘What is this?’’ When thechild produced the correct vocal response, theexperimenter delivered praise. When the childresponded incorrectly or did not respond, theexperimenter pointed to the stimulus andsaid, ‘‘This is North [South], can you say it?’’If the child repeated the name of the category,the experimenter acknowledged the correctresponse and pointed to the next picture.After responding to each picture had occurred(six trials), the experimenter arranged them ina different order and repeated the training.The training criterion was met when the childcorrectly tacted all pictures in three consecu-

tive six-trial blocks without prompting. Whencriterion was met, the probability of reinforce-ment was reduced from 100% to 50% to 0%.Once the child correctly tacted all pictures inthree consecutive arrays without any prompt-ing or reinforcement, training was terminatedand categorization tests were administered.

Posttest interviews. Following the comple-tion of successful categorization tests, theexperimenter exposed the child to an addi-tional categorization test and asked questionswhile the child was selecting the pictures.These questions were: ‘‘How did you do that?’’and ‘‘How did you know that these gotogether?’’ Vocal responses were digitallyrecorded for later analysis. Although suchverbal reports are likely to be under controlof variables other than the experimentalcontingencies, such data may still provideadditional information regarding participants’verbal repertoires.

Independent-Variable Integrity

Independent-variable integrity (IVI) wasassessed for at least 34% of tact-1 and tact-2training sessions by an independent observer.Sessions used in the calculation of IVI wererandomly selected. IVI was calculated bydividing the number of correctly implementedtrials by the total number of trials conductedby the experimenter. Trials were scored asentirely correct or incorrect based on thefollowing categories: (1) Reinforcement—praise was delivered for all correct trials duringthe 100% praise condition, for every othercorrect trial during the 50% praise condition,and not delivered during the 0% praisecondition; and (2) Correction—the correctionprocedure had to be correctly implemented ifthe trial was marked as incorrect. IVI averaged100% for Tom, 99% (range 87–100%) forAdam, 99% (range 87–100%) for Rita, and100% for John. Additionally, experimenterswere trained to avoid cueing during testsessions by not engaging in any specific motormovements. They also were taught to avoidproviding any additional instructions or feed-back other than what was previously described.Experimenters were always seated in a chairpositioned next to the child on the same sideof the table (as opposed to in front of thechild) to prevent eye contact or any othersubtle cue (i.e., specific hand movementstowards the correct comparisons). Finally,

NAMING 391

experimenters were trained not to expect anyspecific outcomes, and were observed forcompliance to the protocol by one of the firsttwo authors for at least 60% of the sessions.

RESULTS

Pretraining with Familiar Pictures

Tom, Adam, Rita, and John required 182,196, 222, and 250 trials, respectively, to reachthe training criterion. All participants maderelatively few errors during pretraining. Thisrapid acquisition of tacts, selection responses,and categorizations indicated that the instruc-tions and consequences provided were ade-quate to teach and test these repertoires.

Unfamiliar Pictures (maps)

Tom and Adam. Figure 2 depicts the per-centage of correct category sorts (circles) and

tacts (bars) of the sample for Tom (upperpanel) and Adam (lower panel). Tom failedcategorization pretests 1 and 2. Once criterionon tact-1 training was reached, Tom selectedthe two correct comparisons in the presence ofa sample in almost all trials on categorizationposttest 1 (look at sample). When required totact the sample (posttest 2), Tom continued toselect the correct comparisons. Additionally,he tacted the samples correctly 100% of thetime, suggesting that the pairwise trainingproduced reliable tacts.

Adam was exposed to additional categoriza-tion pretest probes (Figure 2, lower panel).This served to assess the possibility that extend-ed exposure to testing conditions would resultin any observed increase in the percentage ofcorrect category sorts. Adam failed both pre-tests. After tact-1 training (pairwise), Adam did

Fig. 2. Percentage of correct responses (circles) during categorization pre- and posttest conditions and tacts (bars) tosample for Tom (upper panel) and Adam (lower panel).

392 CAIO F. MIGUEL et al.

not categorize the pictures into North or Southin either of the categorization posttest condi-tions. Although he tacted some of the samplescorrectly during posttest 2 (tact sample), histacting performance was inconsistent. BecauseAdam did not categorize following tact-1training, the tact-2 training procedure with allpictures presented at once was implemented.After meeting criterion during tact-2 training,Adam selected the correct comparisons on100% of trials during posttest-1 and 80% orbetter of posttest-2 probes.

Table 3 displays data on listener tests for allparticipants. During pretests, both Tom andAdam responded at or below chance levels.After tact-1 training, Tom selected the correctcomparison 83%, 100%, 100%, and 100% ofthe time when presented with the targets N1S1,N2S2, N3S3, and Mixed, respectively, suggest-ing that tact-1 training generated accuratelistener behavior. For Adam, by contrast, tact-1training did not generate accurate listenerbehavior. He selected the correct comparison50%, 50%, 16%, and 66% of the time whenpresented with the targets N1S1, N2S2, N3S3,and Mixed, respectively, on the posttest.

Since Adam had to be exposed to tact-2training, listener-2 pretests and posttests wereconducted. Table 3 also displays the data fromlistener-2 tests for Adam. During pretest,Adam selected the correct comparisons in 4out of 12 trials. After tact-2 training, Adamselected the correct comparisons during lis-tener-2 posttests on 92% of trials.

Rita and John. Figure 3 depicts data oncorrect category sorts (circles) and tacts (bars)to the sample for Rita (upper panel) and John(lower panel). During categorization pretests,Rita never selected the two correct comparisonsin the presence of a sample. After reachingcriterion on tact-1 training, Rita selected the

two correct comparisons in the presence of asample 100% of the time. Her accurateperformance was maintained when categoriza-tion posttest 2 (tact sample) was implemented.

John also failed both pretests, and for him,pairwise tact training (tact-1) did not generatecorrect category sorts. His performance was at0% through posttest 1(look at sample). Eventhough John was not required to tact thesamples during posttest 1, he spontaneouslytacted a sample once. During posttest 2 (tactsample), John correctly categorized the pic-tures in only one out of six trials (16%);however, he tacted all samples correctly. Be-cause John did not categorize after tact-1training, tact-2 training was implemented. Aftermeeting criterion on tact-2 training, John’sperformance still met the failing criterionduring the posttest-1 condition. On categoriza-tion posttest 2, however, John correctly catego-rized the pictures most of the time. Thissuggests that John categorized correctly onlyafter additional tact training (tact 2), and whenrequired to tact the sample (test 2).

Table 3 shows that after tact-1 training, Ritaselected the correct comparisons 100%, 83%,66%, and 100% of the time when presentedwith the targets N1S1, N2S2, N3S3, and Mixed,respectively. In contrast, John’s performanceon the listener posttest after tact-1 training wasfar from accurate. He selected the correctcomparisons 83%, 66%, 50%, and 50% of thetime when presented with the targets N1S1,N2S2, N3S3, and Mixed, respectively. Prior totact-2 training, John selected the correctcomparisons 75% of the time (9 out of 12trials). Correct listener behavior remained thesame following tact-2 training.

Trials to criterion. The number of trialsrequired to reach the training criterion ontact-1 training was inversely correlated with the

Table 3

Percentage of correct responses during listener tests for Experiment 1.

Listener-1 Test (6-trial blocks)Listener-2 Test(12-trial blocks)

N1S1 N2S2 N3S3 Mixed

Pre Post Pre Post Pre Post Pre Post Pre Post

Tom 33% 83% 16% 100% 16% 100% 33% 100% - -Adam 50%–0% 50% 0% 50% 16% 16% 0% 66% 33% 92%Rita 50%–30% 100% 16% 83% 33% 66% 16% 100% - -John 50% 83% 16% 66% 0% 50% 0% 50% 75% 75%

NAMING 393

participants’ ages. Tom (5 years 3 months),Adam (4 years 6 months.), Rita (4 years2 months), and John (3 years 1 month) re-quired 448, 536, 640, and 880 trials to reachcriterion for tact-1 training, respectively. How-ever, John (the youngest participant) requiredfewer trials to master tact-2 training than hisolder peer, Adam.

Vocal–Verbal Behavior

During sessions, Tom tacted the stimulicorrectly as the experimenter was placing themon the table on several occasions before thesession began, particularly during tact training.Adam tacted one of the samples (N1) as ‘‘book’’and another (S3) as ‘‘tree’’ during baselinecategorization tests. During the listener-1 pre-

test, Adam tacted the sample S3 as ‘‘chickenbone’’, and during tact-1 training, tacted thesample N3 as ‘‘cat’’. Although not required,Adam correctly tacted the samples and thecomparisons during categorization posttests.

Rita and John emitted few vocalizationsduring the sessions. Rita correctly tacted asample once during categorization posttest 1and John correctly tacted different samplesthree times during categorization posttest 1.John also referred to some pictures as ‘‘north’’and ‘‘south’’ during listener-1 posttest, mostlyin response to the experimenter’s instruction,‘‘point to north [south]’’.

Duringposttest interviews,onlyAdamprovideda verbal description of his selections. When askedby the experimenter, ‘‘Why did you give me this

Fig. 3. Percentage of correct responses (circles) during categorization pre- and posttest conditions and tacts (bars) tosample for Rita (upper panel) and John (lower panel).

394 CAIO F. MIGUEL et al.

one?’’ after successfully categorizing the ‘‘north’’pictures, Adam said, ‘‘Because it is north’’. BothTom and John justified their selections byaffirming that the stimuli selected ‘‘matched’’with the sample. Tom also said, ‘‘They gotogether’’ when asked, ‘‘Why did you pick thesetwo pictures?’’ Of note, the experimenters neverused the word ‘‘match’’ during sessions.

DISCUSSION

Experiment 1 replicated the results of Loweet al. (2002) in that teaching a common vocalresponse (i.e., tact) to each unfamiliar pictureestablished those pictures as a category or aclass. All participants were able to categorizeafter tact training, although only Tom and Ritapassed categorization tests immediately afterpairwise tact training (tact 1). These 2 partic-ipants made few errors during the listener-1posttest, suggesting that for Tom and Rita, tacttraining generated accurate listener behavior.

Both Adam and John required additional tacttraining with all pictures (tact-2 training) beforedemonstrating stimulus class formation. John,who also failed categorization tests, demonstrat-ed inaccurate listener behavior after tact-1training. As was the case for Adam, John passedcategorization tests only after being trained totact in the presence of all pictures simultaneous-ly (tact 2); however, this additional training didnot seem to improve his listener performance,and John passed categorization tests eventhough he was still making errors during listenertests. The reasons for the facilitative effects ofadditional tact training over categorizationperformance will be discussed later.

In summary, data obtained in Experiment 1support the assumption that the acquisition ofboth speaker and listener repertoires may havefacilitated stimulus class formation. Experi-ment 2 was designed to train participants torespond as listeners. The goal of Experiment 2was to observe whether increases in speakerbehavior (tacts) would accompany increases incategorization performances after participantshad been directly trained to receptively dis-criminate pictures (listener behavior).

EXPERIMENT 2METHOD

Participants, Setting, and Materials

Four typically developing children—James,Maria, David, and Pam—participated in Ex-

periment 2. Their ages were 4 years 0 months,4 years 5 months, 3 years 5 months, and4 years 8 months, respectively, at the begin-ning of the experiment. Sessions were con-ducted as in Experiment 1.

Dependent Variables and Data Collection

The main dependent measure was correctcategory sorts as described in Experiment 1.Additional dependent variables included (1)tacts of the stimuli, and (2) the participants’vocal–verbal descriptions of their perfor-mance. Tact tests were administered prior tolistener training and following categorizationtests.

Interobserver Agreement

A second observer recorded data on at least29% of all categorization tests, tact tests, andlistener training sessions. IOA was calculatedas described in Experiment 1. IOA across allconditions averaged 99% (range 87–100%) forJames, 99% (range 83–100%) for Maria, 99%(range 83–100%) for David, and 100% forPam.

Experimental Design

A nonconcurrent multiple-baseline designacross 2 participants was used as described inExperiment 1. The order of conditions was asfollows: baseline categorization pretests, tact-1pretest, listener-1 training, categorization post-tests, and tact-1 posttest. For the participantwho failed to categorize after listener-1 train-ing, the following additional conditions werepresented: tact-2 pretest, listener-2 training,categorization posttests, and tact-2 posttest.Interviews were conducted after successfulcategorization tests. Prior to training withunfamiliar stimuli, participants were exposedto pretraining conditions with familiar stimuli,as described in Experiment 1. Table 2 sum-marizes the experimental conditions.

Pretraining

Pretraining was conducted as described inExperiment 1.

Experimental Procedures

Categorization tests. Categorization testswere conducted as described in Experiment 1.

Tact-1 pretest/posttest. Pictures were separat-ed into three training pairs (N1S1, N2S2,

NAMING 395

N3S3) and reassigned to three new mixedpairs (Mixed). The experimenter presentedthe first pair to the child and while pointing toone of the stimuli asked, ‘‘What is this?’’ If thechild produced the correct tact response, theexperimenter scored the trial as correct. If thechild produced an incorrect tact or did notrespond for 10 s, the next trial was presented.No additional instructions or prompts wereprovided. Only one picture in the pair wastargeted per trial. Sessions were arranged ineight-trial blocks in which North and Southpictures were presented unsystematically ei-ther on the right or left of the child.Participants were exposed to one block ofeach pair and one block of mixed pairs.

Listener-1 training. During this condition,the experimenter randomly separated thepictures into three training pairs. Threeadditional maps of Canadian provinces wereused as distracters. Each trial included a mapof a northern state, a map of a southern state,and a map of a Canadian province ascomparisons. The experimenter presentedthe first set to the child and said, ‘‘Can yougive me the North [South]?’’ If the childselected the correct picture, the experimenterdelivered praise (e.g., ‘‘Good job!’’). If thechild selected the incorrect picture or did not

respond within 10 s, the experimenter pointedto the correct stimulus and said, ‘‘This iscorrect!’’ If the child pointed to the correctpicture after the prompt, the experimenteracknowledged the correct response and said,‘‘That’s right’’ but scored that trial as incor-rect. Sessions were arranged in 6-trial blocks inwhich pictures of northern and southern stateswere presented in a quasirandom order, witheach picture serving as the correct comparisononce on the right, once in the middle, andonce on the left of the child. After training set1, the procedure was repeated with sets 2 and3. The training criterion was three consecutiveblocks at 100%. When responding was atcriterion for all trained sets, the stimuli wererandomly reassigned to three new mixed sets.The training criterion for the new sets was thesame as for the initial sets. Once criterionperformance was attained, the probability ofreinforcement was reduced from 100% to50%. If performance was maintained at 50%reinforcement, probability was reduced to 0%.This condition was terminated when perfor-mance was maintained with no reinforcement.

Tact-2 pretest/posttest. This condition wassimilar to the tact-1 pretest/posttest describedfor Experiment 1. The pictures presentedduring this condition were combined in one

Table 2

Experimental Phases – Experiment 2

Phase TaskNumber of

stimuliTrials

per block Training Criterion

Pretraining Tact-1 (pairs) 2 8 2 blocks at 100%Pretraining Tact-2 (all stimuli) 6 6 2 blocks at 100%Pretraining Listener-1 3 6 2 blocks at 100%Pretraining Listener-2 6 6 2 blocks at 100%Pretraining Categorization Pretraining 6 6 2 blocks at 100%1 Categorization Pretest 1 6 6 N/A

Categorization Pretest 2 6 6 N/A2 Tact-1 Pretest 2 8 N/A3 Listener-1 Training 3 6 3 blocks at 100%4 Cat. Pretraining Review 6 4 1 block at 100%5 Categorization Posttest 1 6 6 N/A

Categorization Posttest 2 6 6 N/A6 Tact-1 Posttest 2 8 N/A7a Tact-2 Pretest 6 6 N/A8a Listener-2 Training 6 6 3 blocks at 100%9a Cat. Pretraining Review 6 4 1 block at 100%10a Categorization Posttest 1 6 6 N/A

Categorization Posttest 2 6 6 N/A11a Tact-2 Posttest 6 6 N/A12b Posttest Interview N/A N/A N/A

a Children were exposed to these phases only if they failed the previous categorization test.b Children were exposed to this phase immediately after successful categorization tests.

396 CAIO F. MIGUEL et al.

array. The pictures were randomly ordered ina row and placed in front of the child. Theexperimenter pointed to each picture in aprespecified order and asked, ‘‘What is this?’’If the child produced the correct tact re-sponse, the experimenter scored the trial ascorrect. No additional instructions or promptswere provided. No consequences were provid-ed. After all pictures were presented (sixtrials), the experimenter showed them in adifferent order. Two 6-trial blocks were pre-sented.

Listener-2 training (all stimuli). In this con-dition, pictures presented during listener-1training (with the exception of the distracters)were combined in a six-picture array, andselecting responses were assessed. The pictureswere randomly ordered in a row and placed infront of the child. The experimenter asked thechild, ‘‘Can you give me all of the North[South]?’’ Selecting the three correct compar-isons was considered a correct response. Whena correct picture was selected the experiment-er acknowledged the response by saying,‘‘That’s right’’. Enthusiastic praise was onlyprovided when all three correct pictures hadbeen selected independently. If the childselected an incorrect comparison stimulus,the experimenter said, ‘‘No’’ and waitedapproximately 10 s for the child to select acorrect comparison. If a correct selection wasmade, the experimenter acknowledged theresponse by saying, ‘‘That’s right’’, and waitedfor other selections. If the child selected asecond incorrect comparison, the experiment-er said, ‘‘No’’, pointed to one of the correctcomparisons and said, ‘‘This is correct’’. Ifthere was still a correct comparison to beselected, the experimenter waited approxi-mately 5 s for the child to select it. Selectingthe remaining correct comparison was ac-knowledged. If the child again selected anincorrect comparison, the correction proce-dure described above was repeated.

When the child selected only one or twocorrect comparisons, the experimenter waited5 s to prompt the next response by saying,‘‘Are there any more?’’ If the child selectedthe remaining correct comparison(s) after theprompt, the experimenter acknowledged theresponse, and provided the prompt again.Training was arranged in six-trial blocks inwhich North and South questions were pre-sented in a quasirandom order during three

trials each (a six-trial block). Training contin-ued until the child selected all pictures in bothcategories three times in a row (three 6-trialblocks at 100%) When criterion was met, theprobability of reinforcement was reduced asdescribed for listener-1 training. This condi-tion was terminated once performance wasmaintained with no reinforcement.

Posttest interviews. Interviews were conduct-ed as described in Experiment 1.

Independent Variable Integrity

IVI was assessed by an independent observerfor at least 29% of the listener-1 and listener-2training conditions for all participants. Ses-sions used in the calculation of IVI wererandomly selected. IVI was calculated bydividing the number of correctly implementedtrials by the total number of trials conductedby the experimenter. Trials were scored ascorrect or incorrect based on the followingcategories: (1) Reinforcement—praise had tobe delivered for all correct trials during the100% praise condition, for every other correcttrial during the 50% praise condition, and notdelivered during the 0% praise condition; and(2) Correction—the correction procedure hadto be correctly implemented if the trial wasmarked as incorrect. IVI averaged 100% forJames, 98% (range 66–100%) for Maria, 99%(range 83–100%) for David, and 100% forPam. As described in Experiment 1, experi-menters were extensively trained to avoidinadvertent cueing during test sessions.

RESULTS

Pretraining with Familiar Pictures

Pam, Maria, James, and David required 156,200, 154, and 190 trials to reach the trainingcriterion, respectively. As in Experiment 1, therapid acquisition during pretraining indicatedthat instructions and consequences providedduring listener, tact, and categorization train-ing were adequate to teach these skills.

Unfamiliar Pictures (maps)

James and Maria. Figure 4 depicts data oncorrect category sorts (circles) and tacts (bars)to the sample for James (upper panel) andMaria (lower panel). During categorizationpretests 1 and 2, James never selected the twocorrect comparisons in the presence of asample nor tacted the samples. Although

NAMING 397

James’s performance met the failure criterionon posttest 1 (three consecutive sessions at33% or below), he performed above baselinelevel. When required to tact the sample(posttest 2), James passed the categorizationtest. During this test, James responded incor-rectly once and incorrectly tacted the sample.

Maria did not meet performance crite-rion on posttest 1. When required to tact

the samples (posttest 2), her performanceimproved considerably. Interestingly, Maria’stact responses during these probes were notreliably correct; in every trial in which thesample was tacted incorrectly, categoriza-tion performance was also incorrect. As thenumber of correct tacts to the sample in-creased, so did the number of correct catego-rizations.

Fig. 4. Percentage of correct responses (circles) during categorization pre- and posttest conditions and tacts (bars) tosample for James (upper panel) and Adam (lower panel).

398 CAIO F. MIGUEL et al.

Table 4 displays data on tact-1 tests forJames and Maria. During the pretest, Jamesdid not tact any of the pictures when present-ed with the targets, suggesting the absence ofspeaker behavior with regard to the trainingstimuli. After listener-1 training, James tactedthe pictures 100% of the time. During pretest,Maria never tacted the pictures correctly, butafter listener-1 training, she tacted the picturescorrectly most of the time. Most of theincorrect tacts occurred in the presence ofthe N3 and S3 pictures. The lack of stimuluscontrol exerted by these two stimuli contrib-uted to the inaccuracies observed when shewas required to tact the pictures in thereassigned (mixed) pairs.

David and Pam. Figure 5 depicts data oncorrect category sorts (circles) and tacts (bars)to the sample for David (upper panel) andPam (lower panel). During pretests 1 and 2,David never categorized correctly. After reach-ing criterion for listener-1 training, Davidfailed posttest 1. When exposed to categoriza-tion posttest 2, David continued to selectincorrect comparisons, even though he wasable to correctly tact some of the samples.Because David did not categorize after listener-1 training during either categorization testcondition, listener-2 training was implement-ed. After meeting criterion in listener-2 train-ing, David successfully passed categorizationposttest 1. Under categorization posttest 2,when David incorrectly tacted the sample, healso categorized the pictures incorrectly.

Categorization data for Pam are displayedon the lower panel of Figure 5. During pretests1 and 2, Pam never categorized correctly. Hercategorization performance seemed to havebeen acquired during posttest. After beingexposed to six probes, Pam successfully passedcategorization posttest 1. When required to

tact (posttest 2), Pam continued to categorize,as well as to correctly tact all samples.

As seen in Table 4, during tact-1 pretest,David did not correctly tact the pictures. Afterlistener-1 training, David correctly tacted 75%,0%, 100%, and 62% of the time whenpresented with the targets N1S1, N2S2, N3S3,and Mixed, respectively. Because David had tobe exposed to listener-2 training, tact-2 pre-tests and posttests also were conducted.During tact-2 pretest, David tacted correctly58% of the time (7 out of 12 trials). Afterlistener-2 training, he tacted the picturescorrectly 100% of the time. Pam never tactedthe pictures correctly during tact-1 pretest, butmade almost no errors on tacts followinglistener-1 training.

Trials to criterion. There was no apparentrelation between the number of trials to reachcriterion and the participants’ age. Pam andJames, who were approximately the same age,required 198 and 162 trials, respectively, toreach criterion on listener-1 training. Mariaand David were separated in age by over a yearbut required approximately the same numberof trials to reach criterion on listener-1training: 1,116 and 1,062, respectively.

Vocal–Verbal Behavior

Neither James nor Pam emitted any un-prompted vocalizations during the sessions.Maria tacted the north and south stimuli onceafter the experimenter implemented thecorrection procedure during listener-1 train-ing. On another trial, Maria said, ‘‘South is thehighway’’ after hearing the experimenter say,‘‘Point to south’’. David correctly tacted thecomparisons five times during listener-1 train-ing trials, and on a trial in which his selectionresponse was scored as incorrect he tacted N2as ‘‘South’’, and S1 as ‘‘North’’. Additionally,

Table 4

Percentage of correct responses during tact tests for Experiment 2.

Tact-1 Test (8-trial blocks)

Tact-2 Test (12-trial blocks)N1S1 N2S2 N3S3 Mixed

Pre Post Pre Post Pre Post Pre Post Pre Post

James 0% 100% 0% 100% 0% 100% 0% 100% - -Maria 0% 100% 0% 100% 0% 75% 0% 87% - -David 0% 75% 0% 0% 0% 100% 0% 62% 58% 100%Pam 0% 100% 0% 87% 0% 87% 0% 100% - -

NAMING 399

in the presence of S3, David said, ‘‘It looks likean L’’, and in the presence of N1, ‘‘It lookslike a D’’. David correctly tacted one sampleand one comparison on different trials duringthe categorization-1 posttest.

During posttest interviews, Maria andDavid provided verbal descriptions of their

selections. When asked by the experi-menter, ‘‘Why did you give me this one?’’after successfully categorizing, Maria said,‘‘Because it is north [south]’’. Similarly,David pointed to the sample and comparisonswhile labeling them (e.g., ‘‘north, north,north’’).

Fig. 5. Percentage of correct responses (circles) during categorization pre- and posttest conditions and tacts (bars) tosample for David (upper panel) and Pam (lower panel).

400 CAIO F. MIGUEL et al.

DISCUSSION

During Experiment 2, 3 out of 4 participants( James, Maria, and Pam) were able to correctlycategorize the pictures into north and southafter being exposed to listener-1 training. Twoof these 3 participants ( James and Maria) wereonly able to do so during categorization posttest2, although Maria seemed to have learned thetask during the categorization test. AlthoughPam successfully categorized during categoriza-tion posttest-1, like Maria, this occurred onlyafter repeated exposure to the test.

In addition to passing categorization testsafter listener-1 training, James, Maria, and Pamshowed near perfect scores on tact-1 posttests,suggesting an increase in both categorizationand tacts as a function of training the listenerrepertoire. The participant who did not catego-rize after listener-1 training, David, did notaccurately tact. After being trained on listener-2,David passed both categorization tests, andcorrectly tacted all of the stimuli during thetact-2 posttest, showing that the additionallistener training with all pictures present at thesame time 1) improved his tact performance,and 2) facilitated stimulus class formation.

Pam’s categorization performance appearedto increase as a result of being exposed to thecategorization posttest. It is possible that,initially, Pam was not tacting the samples,given that she was not required to do so, andthis may have resulted in her not being able tocorrectly categorize the pictures.

Maria’s categorization performance alsoincreased as a result of repeated exposure tocategory-sort trials, but not until categorizationposttest 2. Her tact performance also increasedduring posttest 2. It is possible that Maria wasnot tacting the samples during categorizationposttest 1, and this contributed to her poorcategorization performance. When asked totact the sample (posttest 2), she initially didnot categorize, but as the number of correcttacts emitted by Maria increased, so did thenumber of correct categorizations.

Results obtained with James suggest that hiscategorization performance increased as soonas he was required to tact. As mentionedbefore, in most cases, when participants failedto correctly tact the sample, they also failed tocorrectly select the pictures. Nevertheless,there were trials in which participants correctlytacted the sample, but still failed to select thecorrect pictures.

It is important to note that two of theparticipants from Experiment 2 required over1,000 trials to reach mastery criterion forlistener-1 training. It is possible that the lackof a specifically trained observing responsecould have contributed to the delayed acqui-sition of listener behavior.

Results from Experiment 2 replicate thoseobtained by Horne et al. (2004) in that theparticipants who responded accurately as bothlisteners and speakers (tact) also passedcategorization tests.

GENERAL DISCUSSION

The purpose of the current study was toassess whether children could categorize unfa-miliar pictures when taught listener and speak-er behaviors separately. Results suggest thatchildren between 3 and 5 years of age were onlyable to categorize when they behaved both asspeakers and listeners with regard to the visualstimuli. Moreover, some children categorizedonly when tacts to the sample were required.These results were consistent with the resultsfrom Horne et al. (2004, 2006, 2007) and Loweet al. (2002, 2005) who found that youngerchildren categorized dissimilar objects whenable to behave as speakers and listeners withregard to those objects.

Taken together, results from Experiments 1and 2 appear to support the assumption thatnaming can play an important role in thedevelopment of stimulus classes and categori-zation by typically developing children. Basedon the naming analysis, during categorizationtests children had to tact the sample (speakerbehavior), either overtly or covertly, producinga stimulus (e.g., the sound heard), which inturn controlled responses of selecting thecorrect comparisons (listener behavior). Par-ticipants in the current study who successfullypassed categorization also demonstrated thedevelopment of the full name relation (speak-er + listener). However, as discussed below, theresults are not unequivocal and alternativeinterpretations are possible.

Three aspects of the current study warrantfurther discussion. First, children who failedcategorization tests after tact-1 training (Ex-periment 1) or listener-1 training (Experiment2) later learned to categorize after trainingwith all pictures grouped together. Second,interdependence between listener and speaker

NAMING 401

repertoires was observed across all partici-pants. Third, there was some evidence thatchildren were not only tacting the samples, butalso the comparisons. These three findings arediscussed in the following sections.

Pairwise versus Group Stimuli Training

Tact-2 and listener-2 training may havefacilitated categorization by promoting stimu-lus control topography coherence (McIlvane &Dube, 2003) and generalization between train-ing and testing conditions. During tact-1 andlistener-1 training, children learned to tact orselect pictures when they were presented inpairs or groups of three, respectively, but neverwhen they were grouped together. During tact-1 and listener-1 training, a correct responsecould have come under control of the pres-ence or absence of a specific stimulus feature.During tact-1 training, for instance, a mapcontaining a round corner could have evokedthe response ‘‘north.’’ This would have al-lowed positive results on pairwise (tact-1)training, given that stimuli in the pair couldhave become discriminable based on thisspecific feature. Similarly, the children’s be-havior may have come under control of thepresence or absence of stimulus featuresduring listener-1 training.

When stimuli were grouped together duringcategorization posttests, features that earlierserved as the basis for discrimination (duringpairwise or training with three pictures) mayhave been shared by exemplars belonging todifferent stimulus classes. For example, if the‘‘round corner’’ controlled the tact ‘‘north,’’pictures containing round corners would haveevoked the same tact, regardless of thecategory to which they belonged. If this werethe case, additional tact (or listener) trainingwith all stimuli grouped together may haveencouraged control by the experimenter-spec-ified (more complex) stimulus differences.Given that pictures were similarly groupedduring tact-2 and listener-2 training andcategorization tests, stimulus generalizationbetween training and testing conditions mayhave occurred.

Interdependence between Speaker andListener Repertoires

In Experiment 1, all participants showedemergence of the listener repertoire after tact

training. Lowe et al. (2002) suggested thatwhen teaching tacts to typically developingchildren, listener behaviors are usually learnedconcurrently. This is because children’s ownutterances may continually precede (listener)behaviors of (re)orienting to the tacted object,establishing the auditory stimulus produced byvocal responses as a discriminative stimulus forthese behaviors.

In contrast to the results obtained by Horneet al. (2004), all participants in Experiment 2showed the emergence of the speaker reper-toire after listener training. Because thechildren in the current study were older thanthe ones in the Horne et al. study, it seems thatthis emergence of tacts after listener trainingwas dependent upon their more sophisticatedverbal repertoires including echoic, tacts, andlistener behaviors. However, specific contin-gencies of reinforcement for the directtraining of tacts can be identified in theprocedure used in Experiment 2. It is possiblethat during listener training, when required toselect pictures in the presence of an auditorystimulus (e.g., ‘‘Give me north [south]’’),participants oriented toward the correct com-parison while engaging in covert self-echoicbehavior (repeating the name of the categoryspoken by the experimenter). When partici-pants looked and selected the correct compar-ison, the experimenter who was deliveringreinforcement contingent upon the correctselection also may have reinforced children’ssubvocal echoic behavior in the presence ofthe comparison, resulting in the emergenceof a tact.

So far, specific conditions responsible forthe transfer between speaker and listenerrepertoires are unknown (e.g., Wynn & Smith,2003). Researchers may face procedural limi-tations when studying this phenomenon, giventhat teaching one repertoire gives rise tocontingencies that may shape the other.

Joint Control

There were several instances in whichparticipants from both experiments per-formed incorrectly on a specific category-sorttrial despite the fact that they had beenobserved to correctly (1) tact the sample, and(2) select the pictures given their names. Onother occasions, participants performed cor-rectly on a specific category-sort trial despitethe fact that they had been observed to

402 CAIO F. MIGUEL et al.

incorrectly tact the sample. These outcomesseem inconsistent with Horne and Lowe’s(1996) analysis, which would predict thatcategorizations would occur as a function of(1) tacting the sample, and (2) responding asa listener to the response product generatedby this tact. Results obtained with one partic-ipant in Experiment 1 (John) also seemed todeviate from predictions based on the naminghypothesis given that he correctly categorizedthe pictures even though he was observed tooccasionally perform inaccurately as a listener.

The notion of joint control may serve toexplain some of the results that were inconsis-tent with the naming hypothesis. According toLowenkron (1998), joint control can bedefined as ‘‘a discrete event, a change instimulus control that occurs when a responsetopography, evoked by one stimulus (e.g., thesample) and preserved by rehearsal, is emittedunder the additional (and thus joint) controlof a second stimulus, (e.g., the comparison)’’(p. 332). In the current study, the tact evokedby the presence of a sample (e.g., the vocalresponse ‘‘north’’) would be preserved byrehearsal (e.g., the child would covertly echothe word ‘‘north’’), and emitted under theadditional (and thus joint) control of a correctcomparison (e.g., a north picture). Theselection response (i.e., categorization) wouldbe evoked by the occurrence of joint controlover the topography rehearsed as an echoicbehavior (e.g., ‘‘north’’) (Lowenkron, 1998).

Based on this analysis, the children may havebeen unable to categorize when they failed totact the comparisons. A failure to correctly tactthe comparisons would have prevented theadditional stimulus (the auditory stimulusproduced by tacting the comparison) fromjointly controlling the specific topography,evoking the categorization response. In con-trast, the participant who categorized despitehis lack of accurate listener behavior ( John)may have done so by tacting the sample,covertly rehearsing this topography, and tactingeach of the comparisons, as required toproduce joint control. In the joint controlanalysis, listener behavior seems to play a minorrole in the formation of arbitrary stimulusclasses. Unfortunately, Horne and Lowe’s(1996, 1997), and Lowenkron’s (1996, 1997,1998) analyses describe processes that are notdirectly observable, but inferred from what iscurrently known about stimulus control.

Alternatively, weak tact performance duringcategorization tests could implicate poor stim-ulus discrimination. If this were the case, thenadditional discrimination training would besufficient to produce consistent correct per-formance, despite participants’ speaker andlistener repertoires. This far more parsimoni-ous interpretation dispenses with the use ofunobservable processes to interpret categori-zation and suggests that the listener andspeaker repertoires are corollary to discrimi-nation training. The specific verbal processesinvolved, if any, remain unclear and should bethe focus of future investigations.

Limitations

At least three limitations of the currentstudy are noteworthy. First, it is possible thatcategorization performance was influenced bythe specific instructions provided. On catego-rization tests, the experimenter showed apicture to the child and asked, ‘‘Look, canyou give me the others?’’ (Test 1), or ‘‘What isthis? Can you give me the others?’’ (Test 2).During this condition, the children wereexpected to tact the sample by saying (overtlyor covertly) ‘‘north’’ or ‘‘south’’. The stimulusproduced by this tact would in turn evoke thelistener behavior of reorienting and selectingother members of the same class. However,immediately before selections had to occur,children heard the experimenter say, ‘‘Canyou give me the others?’’ The auditory stimulus‘‘others,’’ instead of the name of the category(‘‘north’’ or ‘‘south’’) may have controlledchildren’s behavior of selecting all picturesfrom the array. Thus, the instructions providedmay have contributed to some of the errorsobserved in category-sort trials. Future replica-tions should try to eliminate this form ofinstruction, guaranteeing that the last soundheard by the child is the auditory product ofher tact.

A second limitation was that correlationsbetween participants’ performance and theirverbal abilities could not be observed giventhat participants’ language skills were notdirectly assessed via standardized measures.O’Donnell and Saunders (2003) suggestedthat better documentation of participants’characteristics, including preexperimental ver-bal skills, would enhance contributions to thestimulus equivalence literature. Future re-search on naming should attempt to test

NAMING 403

participants’ verbal skills prior to the onset ofthe study.

A third limitation consists of the fact thatthe maps used as distracters during listenertraining and testing never served as correctcomparisons. Although the task could havebeen procedurally defined as three-choicematching-to-sample, it was functionally a two-choice matching-to-sample. Therefore, it ispossible that discriminations made by partici-pants were under control of reject relations.Future replications should attempt to establishthree as opposed to two equivalence classes toreduce the likelihood of spurious stimuluscontrol.

Future Directions

The current study succeeded in assessingstimulus class formation by using an alterna-tive measure to the standard matching-to-sample procedure; the categorization test.During categorization tests each pictureserved, at some point, as a sample and as oneof the positive comparisons. Future research-ers should better evaluate the appropriatenessof the categorization test used as a measure ofstimulus equivalence, as well as explore othertests involving conditions similar to thoseencountered in a child’s natural environment.

As previously discussed, results from thecurrent study could be used to support boththe naming (Horne & Lowe, 1996) and joint-control accounts (Lowenkron, 1998). Futureresearchers should attempt to evaluate the rolethat the verbal processes described by each ofthese accounts have on stimulus-class forma-tion. Researchers should evaluate the effectsthat tacting the comparisons (vocal as well asnonvocal tacts) has on accurate category-sorttrials. Given the similarities between thenaming and joint-control hypotheses, theexperimental analysis of verbal behavior hasmuch to gain by a closer examination of dataindependently gathered by these two distinctlines of research.

Conclusion

The results from the current study seem toadd to the current literature on stimuluscategorization by replicating previous findingssuggesting that the speaker and listenerrepertoires may contribute to categorizing bycapably speaking children. Results also suggest

that the specific procedures, as well as thesingle-subject experimental design (i.e., multi-ple baseline) used, may have been adequate tostudy stimulus categorization.

It is important to note that the repertoiredescribed as important for the development ofcategorization (i.e., naming) was not directlyobserved, but inferred. Thus, it is alwayspossible that something else was responsiblefor the observed performances. Further clari-fications will depend on additional demon-strations that equivalence classes only emergewhen the repertoires described by Horne andLowe (1996) are present. Unfortunately, re-search on naming suffers from a majormethodological limitation, namely, the impos-sibility of direct measurement of the putativecontrolling variable (Pilgrim, 1996). Althoughit may be impossible to directly assess covertverbal behavior, it is hoped that futureresearchers will concentrate on proceduralrefinements for the study of naming. Ifnothing else, research on the naming relationwill contribute to a better understanding oflanguage development, in particular, theinteraction among the various verbal operantsdescribed by Skinner (1957).

REFERENCES

Barnes, D., McCullagh, P. D., & Keenan, M. (1990).Equivalence class formation in non-hearing impairedand hearing impaired children. The Analysis of VerbalBehavior, 8, 19–30.

Barsalou, L. W. (1992). Cognitive psychology: An overview forcognitive scientists. Hillsdale, NJ: Erlbaum.

Brady, N. C., & McLean, L. K. S. (2000). Emergentsymbolic relations in speakers and nonspeakers.Research in Developmental Disabilities, 21, 197–214.

Carr, D., Wilkinson, K. M., Blackman, D., & McIlvane, W. J.(2000). Equivalence classes with individuals withminimal verbal repertoires. Journal of the ExperimentalAnalysis of Behavior, 74, 101–114.

de Rose, J. C. (1996). Controlling factors in conditionaldiscriminations and tests of equivalence. In T. R.Zentall, & P. M. Smeets (Eds.), Stimulus class formationin humans and animals (pp. 253–277). Amsterdam,Holland: Elsevier.

Devany, J. M., Hayes, S. C., & Nelson, R. O. (1986).Equivalence class formation in language able andlanguage disabled children. Journal of the ExperimentalAnalysis of Behavior, 46, 243–257.

Eikeseth, S., & Smith, T. (1992). The development offunctional and equivalence classes in high-function-ing autistic children: The role of naming. Journal of theExperimental Analysis of Behavior, 58, 123–133.

Goyos, C. (2000). Equivalence class formation via commonreinforcers among preschool children. The Psycholog-ical Record, 50, 629–654.

404 CAIO F. MIGUEL et al.

Green, G., & Saunders, R. R. (1998). Stimulus equivalence.In K. A. Lattal, & M. Perone (Eds.), Handbook ofresearch methods in human operant behavior (pp. 229–262). New York: Plenum Press.

Horne, P. J., Hughes, J. C., & Lowe, C. F. (2006). Namingand categorization in young children: IV: Listenerbehavior training and transfer of function. Journal ofthe Experimental Analysis of Behavior, 85, 247–273.

Horne, P. J., & Lowe, C. F. (1996). On the origins ofnaming and other symbolic behavior. Journal of theExperimental Analysis of Behavior, 65, 185–241.

Horne, P. J., & Lowe, C. F. (1997). Toward a theory ofverbal behavior. Journal of the Experimental Analysis ofBehavior, 68, 271–296.

Horne, P. J., Lowe, C. F., & Harris, F. D. A. (2007). Namingand categorization in young children: V. Manual signtraining. Journal of the Experimental Analysis of Behavior,87, 367–381.

Horne, P. J., Lowe, C. F., & Randle, V. R. L. (2004).Naming and categorization in young children: II.Listener behavior training. Journal of the ExperimentalAnalysis of Behavior, 81, 267–288.

Keller, F. S., & Schoenfeld, W. N. (1950). Principles ofpsychology: A systematic text in the science of behavior. NewYork: Appleton-Century-Crofts.

Lowe, C. F., Horne, P. J., Harris, F. D. A., & Randle, V. R. L.(2002). Naming and categorization in young children:Vocal tact training. Journal of the Experimental Analysis ofBehavior, 78, 527–549.

Lowe, C. F., Horne, P. J., & Hughes, J. C. (2005). Namingand categorization in young children: III. Vocal tacttraining and transfer of function. Journal of theExperimental Analysis of Behavior, 83, 47–65.

Lowenkron, B. (1996). Joint control and word-objectbidirectionality. Journal of the Experimental Analysis ofBehavior, 65, 252–255.

Lowenkron, B. (1997). The role of joint control in thedevelopment of naming. Journal of the ExperimentalAnalysis of Behavior, 68, 244–247.

Lowenkron, B. (1998). Some logical functions of jointcontrol. Journal of the Experimental Analysis of Behavior,69, 327–354.

Mandell, C., & Sheen, V. (1994). Equivalence classformation as a function of the pronounceability ofthe sample stimulus. Behavioural Processes, 32, 29–46.

McIlvane, W. J., & Dube, W. V. (2003). Stimulus controltopography coherence theory: Foundations and ex-tensions. The Behavior Analyst, 26, 195–213.

Miguel, C. F., Petursdottir, A. I., & Carr, J. E. (2005). Theeffects of multiple-tact and receptive-discriminationtraining on the acquisition of intraverbal behavior.The Analysis of Verbal Behavior, 21, 27–41.

O’Donnell, J., & Saunders, K. J. (2003). Equivalencerelations in individuals with language limitations andmental retardation. Journal of the Experimental Analysisof Behavior, 80, 131–157.

Pilgrim, C. (1996). Can the naming hypothesis be falsified?Journal of the Experimental Analysis of Behavior, 65,284–286.

Pilgrim, C., & Galizio, M. (1996). Stimulus equivalence: Aclass of correlations or a correlation of classes. In T. R.Zentall, & P. M. Smeets (Eds.), Stimulus class formationin humans and animals (pp. 173–195). Amsterdam,Holland: Elsevier.

Randell, T., & Remington, B. (1999). Equivalence rela-tions between visual stimuli: The functional role ofnaming. Journal of the Experimental Analysis of Behavior,71, 395–415.

Randell, T., & Remington, B. (2006). Equivalence rela-tions, contextual control, and naming. Journal of theExperimental Analysis of Behavior, 86, 337–354.

Saunders, K. J., Williams, D. C., & Spradlin, J. E. (1996).Derived stimulus control: Are there any differencesamong procedures and processes? In T. R. Zentall, &P. M. Smeets (Eds.), Stimulus class formation in humansand animals (pp. 93–109). Amsterdam, Holland:Elsevier.

Shusterman, R. J., & Kastak, D. (1993). A California sealion (Zalophus californianus) is capable of formingequivalence relations. The Psychological Record, 43,823–839.

Sidman, M. (1994). Equivalence relations and behavior: Aresearch story. Boston: Authors Cooperative.

Sidman, M., Cresson, O., & Wilson-Morris, M. (1974).Acquisition of matching to sample via mediatedtransfer. Journal of the Experimental Analysis of Behavior,22, 261–273.

Sidman, M., Kirk, B., & Wilson-Morris, M. (1985). Six-member stimulus classes generated by conditional-discrimination procedures. Journal of the ExperimentalAnalysis of Behavior, 43, 21–42.

Sidman, M., & Tailby, W. (1982). Conditional discrimina-tion vs. matching to sample: An expansion of thetesting paradigm. Journal of the Experimental Analysis ofBehavior, 37, 23–44.

Sidman, M., Wilson-Morris, M., & Kirk, B. (1986).Matching-to-sample procedures and the developmentof equivalence relations: The role of naming. Analysisand Intervention in Developmental Disabilities, 6, 1–19.

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

Watson, P. J., & Workman, E. A. (1981). The nonconcur-rent multiple baseline across individuals design: Anextension of the traditional multiple baseline design.Journal of Behavior Therapy and Experimental Psychiatry,12, 257–259.

Wynn, J. W., & Smith, T. (2003). Generalization betweenreceptive and expressive language in young childrenwith autism. Behavioral Interventions, 18, 245–266.

Zentall, T. R., Galizio, M., & Critchfield, T. S. (2002).Categorization, concept learning and behavior analy-sis: An introduction. Journal of the Experimental Analysisof Behavior, 78, 237–248.

Received: August 9, 2005Final acceptance: December 7, 2007

NAMING 405