THE EFFECTS OF ERRORLESS LEARNING ANDBACKWARD CHAINING ON THE ACQUISITION OF

INTERNET SKILLS IN ADULTS WITH DEVELOPMENTAL DISABILITIES

JARED JEROME, ERIC P. FRANTINO, AND PETER STURMEY

QUEENS COLLEGE AND THE GRADUATE CENTER OF THE CITY

UNIVERSITY OF NEW YORK

An important area in the learning and development of individuals with disabilities is theacquisition of independent, age-appropriate leisure skills. Three adults with autism and mentalretardation were taught to access specific Internet sites using backward chaining and most-to-least intrusive prompting. The number of independent steps completed in the task analysisincreased following training.

DESCRIPTORS: autism, computer use, leisure skills

_______________________________________________________________________________

Age-appropriate leisure skills are importantand valued for all individuals, and the use ofpersonal computers has become an importantform of leisure activity for many, includingthose with developmental disabilities. Severalprior investigations have demonstrated effectivemethods for training leisure skills to individualswith developmental disabilities. For example,Luyben, Funk, Morgan, Clark, and Delulio(1986) trained 3 adults with mild mentalretardation a side-of-the-foot soccer pass usingchaining and prompting. The pass was analyzedinto nine steps that were taught sequentiallythrough forward chaining with a varying degreeof prompts as training progressed. At first,verbal instruction plus a physical prompt wereused. These were succeeded by imitativeprompts, gestural prompts, and finally, verbalprompts alone. Eventually the target behavior ofside passing the soccer ball was learned withoutprompts for all participants.

Frank, Wacker, Berg, and McMahon (1985)taught 5 individuals with mental retardation toperform two computer tasks. There were 32 stepsrequired to initiate and terminate a spellingprogram and 23 steps required to initiate andterminate a clock program on the computer. Bothskills were evaluated in a combined multiple

baseline (across students) and sequential with-drawal design. After the first two training sessionsin which picture prompts were used, thepercentage of correct steps completed increasedacross sessions for each participant. When thepicture prompts were removed in the secondbaseline condition, however, the percentage ofcorrect steps decreased. The posttest and follow-up procedures showed a return to the highpercentages found after the first two trainingsessions with picture prompts. Thus, Frank et al.demonstrated that picture prompts were veryeffective in the training of computer skills toindividuals with developmental disabilities.

Although these studies show that adults withdevelopmental disabilities can learn both leisureand computer skills using prompting, forwardand backward chaining, and differential reinforce-ment, no previous research has shown that adultswith developmental disabilities can be taught touse the Internet to access age-appropriate adultleisure activities. Thus, the aim of the currentstudy was to teach adults with autism and mentalretardation to access age-appropriate Web sites onthe Internet using a combined errorless learningand backward chaining procedure.

METHOD

Participants and SettingsChris and Mark were 32-year-old and 24-year-

old men, respectively, and both had been

Requests for reprints should be sent to Jared Jerome,67-41 Burns St. Apt. L7, Forest Hills, New York 11375.

doi: 10.1901/jaba.2007.41-06

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2007, 40, 185–189 NUMBER 1 (SPRING 2007)

185

diagnosed with autism and mild mental re-tardation. Ethan was a 25-year-old man who hadbeen diagnosed with mild mental retardation anddeafness. Each man was chosen to be a participantbecause, when asked ‘‘Do you want to play on thecomputer?’’ he got up and walked to thecomputer and sat next to it within 10 s on fiveconsecutive trials, and if he turned on the powerbutton on either the computer or the monitor onfive consecutive trials when prompted.

All sessions were conducted in a day-habil-itation center for adults with mental retardationand autism. The sessions took place either inthe participant’s classroom or in a separateclassroom with a different computer. Across allsessions, the participant was seated in a chairapproximately 1 m away from and facing thecomputer.

Response Measurement and Reliability

Across all conditions, data were collected onthe number of independent tasks completed.The frequency of steps completed per sessionwas based on a 13-step task analysis (describedbelow), and the data were analyzed based on thenumber of steps completed independentlyrelative to the total number of steps. Interob-server agreement was assessed on 53% of allsessions by having a second observer simulta-neously but independently collect data on thecompletion of each step of the task analysis. Anagreement was defined as both observersmarking a check when a step of the task analysiswas performed or marking an X when a stepwas not performed. Agreement was calculatedby dividing the number of steps with agree-ments by the number of steps with agreementsplus the number of steps with disagreementsmultiplied by 100%. Agreement averaged100% for task completion throughout baselineand postteaching.

Procedure

The following 13-step task analysis wasconducted to develop the requisite skillsnecessary to access a specific Web site:

1. Press the computer power button.

2. Press the monitor power button.

3. Place hand on the mouse.

4. Move the cursor with the mouse until it

points to the Internet ExplorerH icon.5. Double click the Internet ExplorerH

icon.

6. Move the cursor with the mouse to the

GoogleH search box.7. Left click in the box.

8. Type in the search topic of interest.

9. Place hand back on mouse.

10. Move cursor to the box labeled ‘‘search.’’

11. Single click the box.

12. Move the cursor with the mouse down to

the Web site of choice.

13. Single click the Web site of choice.

Across all steps, clicking was defined aspushing down with the right index finger onthe front part of the mouse.

Prior to each day’s sessions, a stimuluspreference assessment (based on DeLeon &Iwata, 1996) was conducted to determine theitems to be presented as reinforcers during theteaching sessions. For all participants, smalledible items (e.g., jelly beans) were used asreinforcers.

Preferred online games or Web sites for eachparticipant were determined before baseline byasking program staff what topics interested theparticipants and then observing the participantsengage in the game or Web site for a minimumof 5 min when the Web sites had been accessedby the experimenters. Staff informed theexperimenters that the participants often usedthose Web sites, but the staff had to access theWeb sites for them. To gain access to thepreferred Internet activity, participants neededto double click the link to the Web site froma search engine (i.e., the GoogleH homepage;www.google.com). Chris and Ethan worked toaccess an online pinball game, and Markworked to access a Web site that played musicvideos (specific links are available from the firstauthor).

186 JARED JEROME et al.

Baseline. During baseline, each session beganwith the experimenter saying, ‘‘[name], do youwant to play on the computer?’’ After theparticipant sat down, the experimenter stood1 m behind the participant. The experimentersaid nothing else and did not deliver any otherprompts during the baseline condition. Whenthe participant either looked away from thecomputer or did not engage in the initial step orany subsequent step of the task for 2 min, thetrial was terminated and the experimenteraccessed the target online game or Web siteand allowed the participant to interact with theonline game or Web site for 5 min. All baselinesessions lasted between 2 and 5 min, dependingon the participant’s behavior.

Teaching. In the teaching condition, theexperimenter stood 1 m behind the seatedparticipant throughout the session. When eachteaching session began, however, the initial 12steps of the task analysis were complete suchthat the participant was only required to clickon the link for their preferred internet activitypresent on the GoogleH homepage. Theexperimenter then said, ‘‘[name], begin playingon the computer.’’ If the participant correctlycompleted this step, access to the preferredinternet activity was delivered for 5 min.

If the participant did not click the mouse after3 s, an errorless learning procedure was used toclick the GoogleH link. A most-to-least intrusiveprompting procedure (i.e., hand-over-handguidance, followed by hand-over-wrist guid-ance, then hand-over-elbow guidance, andfinally hand-over-shoulder guidance) was useduntil all prompts were faded and the participantindependently clicked on the GoogleH link.Experimenters progressed to a less intrusiveprompt after the participant performed the taskwith the previous prompt on two consecutivetrials. Thus, the participants were not given theopportunity to perform an incorrect task on thecomputer. With the exception of the most-to-least intrusive prompting procedure, no otherverbal prompts were delivered.

The prompting procedure continued untilthe participant independently completed eachof the 13 task-analysis steps two times consec-utively. After mastery of each step, training onthe previous step was added. For example, oncethe participant independently completed the13th step on two consecutive trials, theprompting procedure was applied to the 12thstep and so on, based on a backward chainingprocedure. When all 13 steps were completedindependently for three consecutive sessions,teaching was considered to be completed.

Throughout the teaching condition, theexperimenters delivered edible items after theparticipant completed each step of the taskanalysis, for both prompted and independenttask completion. Edible items were deliveredas a form of immediate reinforcement tomaintain behavior in the absence of thedelayed reinforcement provided by accessingthe Internet activity. In addition, the participantreceived 5-min access to the preferredInternet activity after completion of the finalstep in the task analysis, regardless of whethercompletion of that step was prompted oroccurred independently. Each teaching sessionlasted for a maximum of 40 min based on themaximum time available on the participants’schedules.

Postteaching. Postteaching sessions were con-ducted in an identical manner to the baselinecondition; that is, edible reinforcers andprompts were not delivered. The only differencewas if the participants did not complete the 13-step chain, the session was terminated withoutaccess to the Web site. Participants wererequired to progress from one step to the nextwithin the same 2-min time frame that existedduring baseline.

Generalization probes. Beginning in the post-teaching phase, a second computer becameavailable in a separate classroom, locatedapproximately 8 m from the computer usedduring baseline. The generalization computerand monitor were similar to the training

INTERNET SKILL DEVELOPMENT 187

Figure 1. Number of steps of task analysis completed for 3 participants across all trials.

188 JARED JEROME et al.

computer in size and general layout (e.g., powerbutton, mouse type). The same 13-step taskanalysis allowed access to the Web site ofchoice, and the same procedure as baseline andpostteaching was used. Generalization probeswere conducted on this second computer on52%, 58%, and 38% of postteaching sessionsfor Chris, Mark, and Ethan, respectively.

RESULTS AND DISCUSSION

The results are shown in Figure 1. Chris hada range of one to four steps completed inbaseline. He met criterion for all 13 steps in one40-min teaching session. In postteaching, hecompleted all 13 steps of the task analysisduring every session. Mark completed zero toone steps in baseline. He met criterion after five40-min teaching sessions. In postteaching, hecompleted 1 to 13 steps, and he completed all13 steps in 10 of 12 sessions. Ethan completedzero to five steps in baseline. He met criterion inone 40-min teaching session. In postteaching,he completed all 13 steps in every session. Thus,the number of steps completed independentlyincreased after teaching for each participant. Inaddition, participants’ skills generalized toa novel computer.

Previous research has demonstrated the use oftask analyses and errorless learning to teacha variety of nonleisure computer skills and otherleisure activities. The present study combinedthe efforts of this previous research by teachingleisure skills on the computer to adults withdevelopmental disabilities. A limitation of thepresent study was that it taught access to onlytwo Web sites, an online game and a musicWeb site that had been determined to bepreferred prior to the study. Future researchshould extend this method to include choiceamong a variety of available Web sites, becausechoosing among multiple activities may resultin higher levels of task engagement (Tiger,Hanley, & Hernandez, 2006). A secondlimitation to the study was that only one

generalization computer was used and was usedonly during postteaching and not duringbaseline. Future research should evaluate stim-ulus generalization across different computersand locations as well as response generalization,such as to other Internet skills. Also, during thebaseline condition participants were given onlythe following discriminative stimulus: ‘‘[Name],do you want to play on the computer?’’ It ispossible that if participants were given moredetailed instructions, they would have been ableto perform more task-analysis steps. Likewise, itis possible that the delivery of edible reinforcersin the teaching condition may have resulted inincreased task completion relative to baseline.Finally, although the participants completed the13-step chain during the postteaching andgeneralization conditions without any prompt-ing, it was not determined whether they learnedto approach a computer independently to accessa Web site of choice. Future research shouldobserve whether participants would indepen-dently initiate trained leisure skills while notunder the control of a specific discriminativestimulus.

REFERENCES

DeLeon, I. G., & Iwata, B. A. (1996). Evaluation ofa multiple-stimulus presentation format for assessingreinforcer preferences. Journal of Applied BehaviorAnalysis, 29, 519–532.

Frank, A. R., Wacker, D. P., Berg, W. K., & McMahon,C. M. (1985). Teaching selected microcomputer skillsto retarded students via picture prompts. Journal ofApplied Behavior Analysis, 18, 179–185.

Luyben, P. D., Funk, D. M., Morgan, J. K., Clark, K. A.,& Delulio, D. W. (1986). Team sports for theseverely retarded: Training a side-of-the-foot soccerpass using a maximum-to-minimum prompt reduc-tion strategy. Journal of Applied Behavior Analysis, 19,431–436.

Tiger, J. H., Hanley, G. P., & Hernandez, E. (2006). Anevaluation of the value of choice with preschoolchildren. Journal of Applied Behavior Analysis, 39,1–16.

Received March 16, 2006Final acceptance September 12, 2006Action Editor, Henry Roane

INTERNET SKILL DEVELOPMENT 189