Running Head: AUDITORY PREFERENCE ASSESSMENT THE EFFECTS OF HIGHLY PREFERRED AUDITORY STIMULI IDENTIFIED

Transcription

1 Auditory Preference Assessment 1 Running Head: AUDITORY PREFERENCE ASSESSMENT THE EFFECTS OF HIGHLY PREFERRED AUDITORY STIMULI IDENTIFIED FROM PAIRED STIMULUS PREFERENCE ASSESSMENTS ON CORRECT RESPONDING FOR ADOLESCENTS WITH DEVELOPMENTAL DISABILITIES by Erin Horrocks A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Education Approved: Thomas S. Higbee Committee Chair Kimberly Snow Committee Member David Forbush Commitee Member Thomas L. Kent Dean of Graduate Studies UTAH STATE UNIVERSITY Logan, Utah 2005

2 Auditory Preference Assessment 2 Abstract Stimulus preference assessment (SPA) procedures are used to identify salient reinforcers for individuals with developmental disabilities. Previous researchers have used SPA methods to identify reinforcers including tangible items, leisure items, edibles, and olfactory reinforcers. The purpose of this study is to extend research on the paired stimulus preference assessment using auditory stimuli. In the present study, SPA procedures are used to identify potential auditory reinforcers for individuals with severe disabilities. A reinforcing stimulus must also be associated with an increase in the frequency of a response on which it is contingent; therefore, the present study will also determine the reinforcement value of preferred and non-preferred auditory stimuli. This preliminary investigation will include data and suggestions for future research utilizing auditory stimuli. (35 pages)

3 Auditory Preference Assessment 3 CONTENTS ABSTRACT... 2 INTRODUCTION. 4 REVIEW OF LITERATURE... 4 Single Stimulus Presentation.. 5 Paired Stimulus Presentation.. 6 Multiple Stimulus Presentation.. 9 Previous Paired Presentation Studies. 11 Rationale for Investigating Auditory Stimuli. 17 Research Questions. 19 Current Study Overview. 19 METHODS - EXPERIMENT 1 20 Settings and Participants 20 Materials 20 Target Behaviors 21 Measures 21 Procedures Analysis of Results 23 METHODS- EXPERIMENT 2 24 Setting and Participants 24 Materials 24 Target Behaviors 25 Measures 25 Procedures. 26 Analysis of Results. 27 REFERENCES.. 29 APPENDICES 31 Appendix A: PS Data Sheet 32 Appendix B: Correct Responding Data Sheet 35

4 Auditory Preference Assessment 4 Introduction Positive reinforcement is used in most behavioral acquisition and reduction procedures for individuals with developmental disabilities. Contingent application of positive reinforcement lies at the heart of operant procedures used to treat behavioral deficits. In applying operant techniques to establish or maintain socially desirable outcomes, significant emphasis is placed on the selection of salient reinforcers. However, the identification of reinforcers for the population of individuals with severe disabilities may be particularly difficult because of developmental or physical disabilities, limited verbal repertoires, or other factors. Review of Literature All literature reviewed was found through the ERIC database. The search terms stimulus preference assessments; single stimulus preference assessment, paired stimulus preference assessment, and multiple stimulus preference assessment were used. Five single-stimulus presentation studies were reviewed, and four were excluded. Five pairedstimulus presentation studies were reviewed, and all five are included. Five multiplestimulus presentation studies were reviewed, and four were excluded. Because there are possible difficulties in identifying reinforcers for individuals with developmental disabilities, assessment methodologies have been developed for this purpose (Ivancic & Bailey, 1996; Pace, Ivancic, Edwards, Iwata, & Page, 1985; Rotatori, Fox, & Switzky, 1997). These methods include personal interviews or nominations, which involves questioning parents, caretakers, and others that are knowledgeable about the individual with a disability to determine what serves as reinforcers for the individual.

5 Auditory Preference Assessment 5 Reinforcer surveys are another method of determining reinforcing stimuli; this is where the individual is provided with a written or verbal survey that permits the individual to choose various items that are highly favored. Still, stimulus preference assessment (SPA) methods have been found to more accurately predict reinforcer effectiveness than student, teacher, or parent reports. SPA methods allow the reinforcing effects of new stimuli to be evaluated, the reinforcing effects of various stimuli to be compared, and individuals with and without vocal communication skills to actively choose reinforcers. Single Stimulus Presentation The first SPA method developed is the single-stimulus presentation, which was developed by Pace, et al. (1985). This procedure includes placing items in front of the individual one at a time, and measuring whether or not the individual approaches each item. Each item is presented several times, and the measure of preference is the percentage of times an item is approached when it is available. Advantages of this method include placing items into general categories of preferred and non-preferred, as well as successfully identifying reinforcers for individuals with severe disabilities. Disadvantages include the tendency to overestimate the individual s preferences, the various stimuli cannot be compared to one another, and the procedure is very time consuming. An example of the single-stimulus presentation was illustrated by Pace, et al. (1985) in which the investigators examined the use of a single-stimulus presentation procedure for identifying potential reinforcers for six individuals with profound mental retardation. In experiment 1, the six participants were repeatedly exposed to 16 stimuli and approach behaviors to each stimulus were used to identify preferred and nonpreferred

6 Auditory Preference Assessment 6 stimuli. The data was reported using a bar graph to show percent approach responses for each of the subjects and each stimulus. Preferred stimuli were defined as those approached on at least 80% of the trials and nonpreferred stimuli were defined as those stimuli approached on 50% or less of the trials. In experiment 2, they examined the reinforcing properties of preferred and nonpreferred stimuli by delivering them contingently on the occurrence of randomly selected responses. Results demonstrated that the preferred stimulus conditions typically produced higher rates of responding than did either the baseline or the nonpreferred stimulus conditions, suggesting that the procedure can be used to assess reinforcer value for individuals with limited behavioral repertoires. The experimental design used in this experiment involved baseline, preferred, and nonpreferred conditions that were arranged in a reversal design, with the order of conditions varied across subjects. The single-stimulus preference assessment used in this study was very time-consuming, requiring 2 hours per participant to complete. This study served as a basis for future researchers to build other reinforcement assessment techniques that are more time efficient and effective. Paired Stimulus Presentation Another SPA method is the paired-stimulus presentation, or forced choice method, developed by Fisher et al. (1992). This procedure involves presenting two stimuli simultaneously and allowing participants to choose one. The first item touched is scored as the selection. Assessment continues until each item has been paired with every other item at least twice. The measure of preference is the percentage of times an item is selected when it is available for selection. Advantages of the paired stimulus preference assessment include it is more precise than the single-stimulus method, it allows items to

7 Auditory Preference Assessment 7 be compared to one another and ranked according to preference, and it is the most accurate SPA method. Disadvantages include that the procedure can be time consuming and it may be impractical to use on a frequent basis. Fisher, et al. (1992) used a concurrent operants paradigm to compare the Pace et al. (1985) procedure with a modified procedure where participants were presented with two stimuli simultaneously and were given access only to the first stimulus approached. In Phase 1 of this investigation, the stimuli used in the Pace et al. (1985) study were compared using the stimulus preference procedure and the forced choice method of stimulus presentation. Sixteen stimuli were individually presented a total of 10 times each over eight sessions. Within a session, four stimuli were presented five times each in a counterbalanced order. Approach behavior to the stimuli resulted in access to the stimuli for 5 seconds. During the forced-choice assessment, the same 16 stimuli were presented in pairs. Each stimulus was paired once with every other stimulus, in a random order, for a total of 120 stimulus-pair presentations. For each trial, two stimuli were placed in front of the participant. Approaches to one of the stimuli resulted in access to that stimulus for 5 seconds and removal of the other stimulus. The results of Phase 1 produced high-high stimuli (stimuli approached on at least 80% of trials on both stimulus preference and forced choice assessments) and SP-high stimuli (stimuli approached on at least 80% of stimulus preference trials and 60% or fewer forced-choice trials). However, no FC-high stimuli (stimuli approached on at least 80% of forced-choice trials and 60% or fewer stimulus preference trials) were produced in phase 1, as a result, only high-high and SP-high stimuli were compared. There were nine stimuli identified as highly preferred by both assessments for the 4 participants. These were the only nine stimuli

8 Auditory Preference Assessment 8 identified as highly preferred by the stimulus-choice assessment. There were 27 additional stimuli identified as highly preferred by the stimulus-choice assessment and 19 of which met criteria for an SP-high stimulus (approached on 80% or greater stimulus preference trials and 60% or fewer forced-choice trials). In phase 2, the initial and return to baseline sessions consisted of a therapist being present in the treatment room, but no interaction with the participants. None of the stimuli were presented during baseline. Following baseline, the participant received training trials to teach him or her to gain access to the stimuli being assessed. A training trial consisted of placing the stimuli in a box (or chair) in front of the participant. The participant was given 5 seconds to move into the box or chair independently and then given sequential verbal and physical prompts every 5 seconds until the participant displayed the target response. Immediately after displaying the target response, the participant was given access to the stimuli for 10 seconds. During treatment, sessions were conducted in a manner identical to baseline except that two stimuli were positioned in each of the boxes on the floor or directly next to each of the chairs. The two high-high stimuli were placed with one box or chair, and the two SP-high stimuli were placed with the other box or chair. The participant could gain access to high-high or SP-high stimuli by engaging in independent in-square or inchair behavior in the corresponding box or chair on which those stimuli were placed. For example, if the participant exhibited in-square or in-chair behavior, the therapist presented the stimuli associated with that square or chair for 3 seconds. In phase 1, a comparison of two stimulus assessments showed that all of the items identified as highly preferred by the forced-choice assessment were also identified as highly preferred on the stimulus preference assessment. In addition, on all the stimuli for which the two

9 Auditory Preference Assessment 9 assessments disagreed, the stimulus preference assessment identified the item as highly preferred and the forced-choice assessment identified the item as low to moderate. In Phase 2, stimuli on which the two procedures agreed and disagreed were compared using a concurrent operants paradigm. For all 4 clients, greater increases in responding occurred with high-high stimuli (stimuli approached on at least 80% of trials on both stimulus preference and forced choice assessments) than with SP-high stimuli (stimuli approached on at least 80% of stimulus preference trials and 60% or fewer forced-choice trials), indicating that the forced-choice assessment better predicted which stimuli would function as more potent reinforcers. The results revealed that this forced choice stimulus preference assessment resulted in greater differentiation among stimuli and better predicted which stimuli would result in higher levels of responding when presented contingently in a concurrent operants arrangement. These results also suggest that the forced-choice assessment has good concurrent validity and that the single-item stimulus preference assessment may identify stimuli as highly preferred too often. This may occur because stimuli are presented individually and alternative stimuli are not available to the participants. Multiple Stimulus Presentation The last method used is the multiple-stimulus without replacement (MSWO) procedure that was developed by DeLeon and Iwata (1996). In this method, all items to be assessed are presented simultaneously in an array, and each time an individual chooses one of the items from the array, that item is removed and the remaining items are presented to the individual. The process continues until all items have been selected or the individual does not select any item within a set amount of time. Items are ranked in

10 Auditory Preference Assessment 10 order of preference by the percentage of times each item is selected when it is available. Advantages of this method are the results produced are similar to the paired stimulus method and it can be completed in less time than the paired stimulus assessment. Disadvantages include that it can be difficult to manage multiple stimuli and it is not quite as precise as the paired stimulus method. DeLeon and Iwata (1996) compared three methods for presenting stimuli during reinforcer-preference assessments. The first experiment compared three methods for presenting stimuli during reinforcer-preference assessments; a paired-stimulus format (PS), a multiple-stimulus format with replacement of the selected stimulus (MSW), and a multiple-stimulus format without replacement (MSWO). The experimental design that was used in this experiment was an alternating treatment design. The alternating treatments were the MSWO, PS, and MSW procedures and each individual s specific reinforcers. The second experiment in this study was conducted to verify predictions about stimuli that were not selected in the MSW procedure. Reinforcement effects were examined by organizing a schedule of contingent delivery for four of these items to determine if they could maintain levels of responding above baseline. This experiment included four participants from the first experiment. Three of the four participants showed an increase in responses per minute over baseline, when using the reinforcers identified during the PS and MSWO assessments, but not the MSW assessment. Results for the 7 participants with profound developmental disabilities showed moderate to high rank-order correlations between the MSWO and PS procedures and a similar number of identified reinforcers. Also, the time to administer the MSWO was comparable to that required for the MSW method and less than half that required to administer the PS

11 Auditory Preference Assessment 11 procedure. Overall, the three assessment formats created similar results in identifying the most preferred stimuli. However, the MSWO and PS presentations generally produced more consistent rankings. In addition, the results from this experiment suggest that in a least some cases, items that remained identified as reinforcers in the MS procedure did function effectively as reinforcers, and that the MSWO and PS procedures more readily identified those items as reinforcers. Previous Paired Presentation Studies The current study will use a paired stimulus preference assessment to determine potential reinforcing auditory stimuli. Several paired stimulus preference assessment methods have been used in previous studies. The first study reviewed was conducted by Dattilo et al. (1987), in which an investigation was carried out to display the application of a computerized assessment procedure for determining the leisure preferences for persons with severe disabilities. Three children with severe disabilities who were enrolled in a local public school served as the participants. The experimental equipment consisted of a switch- activated microcomputer and a computer program designed to interpret the switch activations to generate a selection of leisure activities. This provided immediate and contingent sensory activities to the participants in order to systematically assess their individual leisure preferences. A single subject research method using a multiple treatment design involving five conditions across three subjects was used. Each subject participated in a series of three-minute sessions during each condition. Each of the five conditions provided students with a choice between two of the following leisure activities: listening to music, watching action videos, activating a blender and drinking a portion of a milkshake, watching a slide show, and feeling the vibrations from a vibrating

12 Auditory Preference Assessment 12 pad. The number of switch activations for each of the three participants was automatically tabulated, analyzed, and plotted by the computer program. The results suggested that all three individuals sampled each activity at least once per condition, rather than making one choice and then perseverating on that choice. The results of the investigation verified that the leisure preferences of people with severe disabilities can be assessed systemically. This study relates to the current study in that a paired stimulus preference assessment will be used, in addition, listening to music was one leisure activity that was offered to participants in this study. In the current study; however, auditory stimuli will be the only type of stimuli offered to participants. Another study using the paired stimulus presentation was conducted by Parsons et al. (1990). In this study, a program was evaluated for assessing choice making skills to provide opportunities for persons with severe disabilities to express food and drink preference. Two experiments took place in this study. The major purpose of Experiment 1 was to develop and evaluate a way of assessing choice-making skills for expressing food and drink preferences for persons with severe disabilities. A second purpose of Experiment 1 was to compare the results of a structured behavioral assessment to a more traditional caregiver survey for determining participant food and drink preference. A third purpose of Experiment 1 was to socially validate the importance of providing persons with severe disabilities with more choice-making opportunities regarding food and drink consumption. In Experiment 1, the assessment procedure involving repeated, paired-item presentation, which resulted in active choice making and the identification of preferences for all 5 participants. Results indicated that caregiver opinion was not predictive of participant food and drink preferences. Based on 80% selection criterion,

13 Auditory Preference Assessment 13 results indicated that all participants demonstrated a preference for at least one item when presented with item pairs. Although each participant expressed at least one preference, only 2 participants expressed the same preference, representing the fact that food and drink preferences are very much individualized. Results of the staff survey indicated that staff opinion of participant preferences did not consistently correspond with the results of the systematic assessment. The results concerning staff knowledge of participant preferences seemed contradictory. Particularly, staff members may have been knowledgeable about a participant s strongest or most general preference in terms of favorite food or drinks, but may have not been knowledgeable about preferences concerning a comparison between any two items. Fourteen adolescents and adults from one unit of the same residential living area and educational program as the participants in Experiment 1 participated in Experiment 2. The same criteria used in Experiment 1 to select food and drink items were used in Experiment 2. Nine food and drink pairs were assessed across 14 participants. Results indicated that the five staff members who were the routine day treatment providers for the participants were able to conduct the behavioral assessment procedures with appropriate supervision. Each participant was provided with an opportunity to make a choice, and 13 of 14 participants expressed a distinct food or drink preference in direct response to the paired-item assessment procedure. Overall, these results were encouraging, because all 14 participants made clear choices and the assessment was conducted reasonably quickly. In Experiment 2, the practicality of the assessment procedure was supported by demonstrating that routine caregivers could apply the procedure with proper supervision to provide choice opportunities. Results of the procedure were predictive of participant choices when a less

14 Auditory Preference Assessment 14 structured and more normal opportunity to express a preference was provided during regular mealtimes. Results of the two experiments suggest a way in which persons with severe disabilities may participate more actively in events that are likely to affect their quality of life, such as at mealtimes. The results also demonstrated that the type of systematic preference and reinforcer assessment technology developed through previous research in leisure and vocational areas can be applied to a third important area, choice making at mealtime. Across the two experiments, the assessment procedure demonstrated that all 19 non-verbal participants expressed clear preferences. The importance of these results seems huge, when considering that the more traditional approach of questioning direct care staff did not reliably predict what the participant s would really choose. Furthermore, demonstration that the participants routine day treatment providers could successfully conduct assessment sessions with appropriate supervision supports the practicality of the procedure. This study relates to the current study in that a paired stimulus method was used, and the importance and appropriateness of choice making for individuals with severe disabilities is emphasized. Another study utilizing paired stimulus preference assessments was conducted by Belfiore et al. (1994). Choice and preference in four adults with severe disabilities was examined within the context of a community-based and center-based location. Prior to the preference analysis, an independent selecting response was empirically assessed and validated for each adult. Results of the preference analysis showed that in the community site there was a higher cumulative frequency of choice-making responses towards beverages for all participants, and an indication of beverage preference for two of the four participants. Conversely, in the center-based setting, there were fewer responses

15 Auditory Preference Assessment 15 made towards the beverages, and no indication of preference for any of the four adults. The cumulative frequency and ration data showed that all four participants engaged in a higher frequency of beverage selecting behavior in the community setting than in the center-based setting, and two participants showed a repeated preference for one of the two beverages presented. These results support previous findings on the ability of persons with profound disabilities to demonstrate choice-making and preference. Although, previous research showed choice-making and preference occurred in vocational or school locations, this study found choice-making and preference were not observed in such traditional instructional locations. Rather, for the participants in this study, choice-making and preference were observed only in the community setting. Derby et al. (1995) conducted a study in which a forced-choice preference assessment was used and two dependent measures were used to select preferred stimuli by approach responding and latency to the first aberrant response. Two individuals diagnosed with severe developmental disabilities participated in this experiment. Sessions were videotaped and data was collected on approach to stimuli and latency to the first abnormal response. The participants were exposed to four stimuli, such as a ball, vibrator, radio, and stuffed animal. A modification of the forced-choice method described by Fisher et al. (1992) was used. At least 15 presentations of each stimulus were completed. The participants were provided with access to the stimuli they approached until either an aberrant response occurred or 5 minutes elapsed. For both participants, treatment conditions were developed using stimuli identified as preferred using both dependent measures. Treatment effectiveness was then tested across a series of analogue conditions. For one participant, treatment conditions consisted of the

16 Auditory Preference Assessment 16 noncontingent presentation of each stimulus with and without parental interaction in the form of neutral comments within a multi-element design. For the other participant, two differential reinforcement of other behavior treatment conditions were compared to an ignore condition and a non-contingent attention condition within a multi-element design. Stimuli identified as preferred based on both dependent measures were then evaluated during treatment. The results suggested that latency may be a useful measure in the selection of preferred stimuli during forced-choice procedures. For both participants, a different stimulus was identified as most preferred based on each dependent measure. In both cases, the stimulus that was approached most often also resulted in the shortest latency. The latest investigation using a paired stimulus preference assessment was conducted by Higbee et al. (2003). In this study, one participant was involved in a paired-stimulus preference assessment to evaluate the participant s preference for six olfactory stimuli. The participant was presented with two different scents for approximately two seconds. Then, the participant was instructed to pick the one he wanted the most, and was allowed to choose between the two stimuli. Immediately following the participant s choice, he was allowed access to the selected stimulus for 10 seconds. This process continued until all 30 stimulus trials were completed. Stimuli were ranked from most to least preferred. To analyze the reinforcing effectiveness of the stimuli assessed in the paired stimulus assessment, a reinforcer evaluation assessment was conducted. Three stimuli, composed of one high ranked, one medium ranked, and one low ranked stimulus were evaluated using a multi-element experimental design. The target response was the pressing of a button switch which was connected to an electronic

17 Auditory Preference Assessment 17 counter that recorded responses. Following three baseline sessions, where no systematic consequences were delivered for responding, one stimulus per session was delivered contingent on the target response on a fixed ratio reinforcement schedule. During the sessions, each time the schedule requirement was met, the participant was provided with 10 seconds of access to the stimulus for that session. The results from this study indicated that olfactory stimuli identified as preferred by a paired-stimulus preference assessment may function as reinforcers. Rationale for Investigating Auditory Stimuli Most of the stimuli used in the studies previously discussed consisted of edible and tangible (e.g., toys or manipulatives) items. Some of the studies reviewed, including the Pace, et al. (1985) study, the Fisher, et al. (1992) study, and the DeLeon and Iwata (1996) study used a type of auditory stimuli, such as music, clappers, or beeps in their studies, in addition with other various types of stimuli. The current study will make use of only auditory stimuli in the array of reinforcers. The reasons for making use of edible or tangible stimuli are evident, because they are easily accessible, as well as frequently used in behavioral interventions. However, other types of reinforcers, such as auditory stimuli, may be functional and appropriate for individuals with developmental disabilities. Yet, there is a lack of research in the area of stimulus preference assessments utilizing auditory stimuli. Using auditory stimuli as reinforcers would expand the assortment of prospective reinforcers available for use in educational settings for individuals with developmental disabilities. Auditory stimuli would have the benefit of being easily accessible, simple to deliver, and appropriate for all ages.

18 Auditory Preference Assessment 18 One prospective reason for investigating auditory stimuli as reinforcement is for use in circumstances where edible reinforcers are not permissible because of health concerns. Overusing edible reinforcers may also cause individuals with disabilities to become overweight or unhealthy. Determining other reinforcers, such as auditory stimuli, for individuals with health issues may be beneficial. Another reason for investigating auditory reinforcers is that music falls into this category. Music is a leisure activity that many typical developing individuals of all ages enjoy. Using music as reinforcement for individuals with disabilities is an ageappropriate reinforcement procedure for all age ranges. A third prospective reason for identifying auditory reinforcers is the accessibility, variety, and simplicity of delivering auditory reinforcers in educational settings. Music and other auditory stimuli are easily accessible and can be purchased, downloaded from the internet, or accessed freely through radio stations. In addition, there is a huge variety of different types of music available, which means that individuals with disabilities are offered a variety of choices in determining which auditory stimuli is reinforcing. Because of this wide range of music types available, individuals with disabilities are likely to find at least one, if not more auditory stimuli favored from among the substantial selection that is available. Music and other auditory stimuli are very easy to deliver to students with disabilities. Most educators have access to CD players, computers, or tape players to deliver the auditory stimuli to students. Headphones may also be used to keep the noise level and distractions down for other students. The ease of delivering auditory stimuli as a reinforcer may be appealing in an educational setting.

19 Auditory Preference Assessment 19 Knowing the possible benefits of identifying auditory reinforcers, the purpose of this study is to extend research on the paired stimulus preference assessment by applying this assessment methodology to auditory stimuli. This study will investigate the effectiveness of auditory stimuli on correct responding, using the highest ranked selection as the reinforcer. The paired stimulus preference assessment method will be used in this study, because this procedure generally produces consistent rankings across administrations, and because of the difficulty of presenting more than two auditory stimuli at the same time. Research Questions The primary research questions to be addressed in this study are: 1) what are the effects of a paired-stimulus preference assessment used to determine preferences for various types of auditory stimuli for individuals with developmental disabilities? 2) what are the effects of access to preferred auditory stimulation on the academic behavior of students with disabilities? and 3) what are the effects of auditory stimuli determined to be either high- or low-preference through the preference assessment on levels of academic responding that correspond to the degree of preference? Current Study Overview The current study will consist of two experiments. Experiment 1 will consist of the SPA presentation, in which all stimuli will be presented with a paired stimulus. Each participant will select one of the two auditory stimuli choices and have access to the music for a specific amount of time. This continues until all of the stimuli have been

20 Auditory Preference Assessment 20 presented together as pairs. This procedure determines the music stimuli that are preferred and nonpreferred for each participant, and also establishes a ranking system for the auditory stimuli for each participant. A reinforcing stimulus must also be associated with an increase in the frequency of a response on which it is contingent; therefore, experiment 2 will determine the reinforcement value of preferred and nonpreferred music. Experiment 1 Methods Settings and Participants Four to six adolescents, ranging from years of age, will participate in the study. Participants are sixth to eighth graders that receive special education services in a self-contained classroom in a public middle school. Each participant is classified with a disability as defined by IDEA, and each of the six participants fall under the categories of autism, intellectual disability, or multiple disabilities. All participants have current Individual Education Plan s (IEP s), and have needs which require them to receive special education services in a self-contained setting. Participants do not have any hearing problems that interfere with the ability to listen to auditory stimuli. In addition, existing classroom staff has reported that participants seem to like auditory stimuli. Prerequisite skills for each participant include the ability to differentiate between different auditory stimuli and the ability to select a choice after a verbal prompt is given. Preference assessment sessions will be conducted in a small, separate room within the middle school, to minimize distractions and noise levels for other classmates. The

21 Auditory Preference Assessment 21 participant, the experimenter, and the inter-observer will be the only individuals present in the room. Materials An array of 6 various types of music will be used in the form of compact discs. Initial interviews with parents, caregivers, and participants will be used to select the 6 music types for each individual. Music selections will vary depending on each participant s preferences in music. Two CD players will be used for playing the two different types of music. In addition, two burned CD s with the various music selections determined by an initial interview with a caregiver will be prepared for each participant. Each song title on both CD s will be assigned a letter and number, such as 1-a, which corresponds to a particular music selection on the data sheets. Burned CD s will be prepared in advance for each participant and will contain all music types on one CD in the correct order to be paired with each of the other music types on the other CD. Burned CD s will be used for cueing different music selections more effectively and stopping and starting music selections more quickly. Target Behaviors The dependent variable for experiment 1 is the selection of preferred and nonpreferred music types. The method for assessing stimulus preference will consist of measuring the selection of each of the 6 auditory stimuli types as a percentage. A selection response will be recorded when the participant makes physical contact with either the left or right CD player.

22 Auditory Preference Assessment 22 Measures The dependent measure in experiment 1, which is selection of music type, is recorded using a paired procedure. A percentage of selected auditory stimuli score will be calculated for each stimulus by dividing the number of times it was selected by the number of times it was available for selection. That number will be multiplied by 100 to produce a percentage. Auditory stimuli will then be ranked from most-to least-preferred based on this selection percentage. The experimenter and an independent observer will record data on all preference trials using a data sheet. Interobserver agreement will be calculated by dividing the number of agreements by the number of agreements plus disagreements and multiplying that by 100 to produce a percentage. An agreement will be scored when both observers score the same item as being selected on a given preference trial. Procedures A paired-stimulus (PS) preference assessment, comparable to procedures used by Fisher et al. (1992) will be used to assess the participant s preference for the auditory stimuli. The paired-stimulus method will be used, because of the tendency that the single-item method has to overestimate preference (Fisher et al., 1992) and the complexity of presenting more than two auditory stimuli concurrently. An initial interview with parents, caregivers, and participants will be used to identify the 6 music types that will be used in the paired stimulus preference assessment for each participant. All auditory stimuli will be presented twice to each participant, once on the left and once on the right, to control for potential sequence effects or side preferences for a total of 30 preference trials. Before each preference trial, the participant will be allowed to listen to

23 Auditory Preference Assessment 23 each auditory stimulus for approximately seconds each. The two CD players will be placed approximately 1 m apart and will be centered in front of the participant. The auditory stimulus on the left will always be activated first. After the participant has listened to each stimulus, he or she will be given the instruction, Touch the one you like the most, and then each participant will be allowed to choose between the two music selections. The participant will indicate a selected music choice by touching one of the CD players. Following the selection response, the participant will be given access to the selected stimulus for 15 seconds. At the end of the 15 seconds, the music will be stopped, and the next trial will begin. This process continues until each of the six participants complete all 30 preference trials. A percentage of selected auditory stimuli score will be calculated for each stimulus by dividing the number of times it was selected by the number of times it was available for selection. That number will then be multiplied by 100 to produce a percentage. Auditory stimuli will then be ranked from most-to least-preferred based on this selection percentage. All observers will receive training on specific data sheets, how to record data accurately, and will be provided with examples and non-examples of approach behavior during the preference assessment and the various tasks involved in the study. Interobserver agreement levels will need to be at 80% or better. If the 80% criteria level is not met, retraining of observers will be completed. This will consist of additional training, including modeling of examples and non-examples of correct responding.

24 Auditory Preference Assessment 24 Analysis of Results It is expected that the Paired-Stimulus preference assessment procedure will be an effective procedure in identifying reinforcing auditory stimuli for individuals with severe disabilities. The data will be analyzed by first, graphing the selection percentages of each auditory stimuli on a bar graph. By doing so, preferred and non-preferred music stimuli will be easily distinguished and assessed for use in Experiment 2. Stimuli with selection percentages of 80% or greater will be classified as preferred. Experiment 2 Methods Settings and Participants The same four to six adolescents that participated in Experiment 1 will participate in experiment 2. Each of the six participants range from ages 13-15, and receive special education services in a self-contained classroom in a public middle school. The setting for Experiment 2 also remains the same as that in Experiment 1. All experimental procedures will be carried out in a small separate room within the middle school, and all tasks involved in Experiment 2 will be in alignment with each child s IEP goal and objectives. Prerequisite skills for each participant will vary, in that each participant is functioning at a different ability level; therefore, each task will be chosen based on the participant s functioning level and needs. All tasks in Experiment 2 will be pulled from each participant s ability level, current curriculum and programming, and IEP goals. Examples of potential tasks include sorting items such as crayons or checkers by color, sorting containers by size, sorting school supplies (pencils, paper, paper clips, rubber

25 Auditory Preference Assessment 25 bands, etc.) by category, or sorting and putting candy in plastic baggies according to color. Materials Materials used in assessing the music s potency as a reinforcer will be selected based on each participant s IEP goals and objectives. These materials will include items such as sorting tasks and bins, matching tasks and bins, flashcards, and puzzles. Each of these will be free-operant tasks, in which the participant is not restricted, and may complete as much or as little work as desired. A stopwatch will also be used to time the experimental sessions, which will each be 5 minutes in length. In addition, the stop watch will be used during sessions, when the schedule requirement is met, to time the 15 seconds of access to the auditory stimulus for that session. Target Behaviors The dependent variable in Experiment 2 is rate of correct responding on a given task. These tasks will consist of adaptive behaviors that the participants exhibit at low rates prior to the study, as well as tasks that the participants are not highly motivated to engage in. These will serve as the target behaviors to be increased. The dependent measures will be selected from each participant s IEP goals, which fall into the category of free-operant tasks exhibited at a low rate due to motivational problems. Existing classroom data will be used to determine and select tasks that fall below at least 70% correct responding. The selection of tasks to be used in the experiment will be determined by the experimenter. The independent variable will be the contingent

26 Auditory Preference Assessment 26 presentation of the highly preferred music, identified by the paired stimulus preference assessment procedure. Measures The target behavior in Experiment 2, which is rate of correct responding, will be will be recorded as correct response per minute. Responding will be coded separately when music is on and correct responding is taking place, and when music is off, and correct responding is taking place. The tasks selected for each participant will be freeoperant tasks; therefore, no restrictions will be placed on the participant. Procedures In order to analyze the reinforcing effectiveness of the stimuli assessed in the paired stimulus preference assessment, a reinforcer evaluation assessment will be conducted. Preferred and non-preferred music selections will be used to compare the effectiveness of the two stimuli. Preferred music stimuli will be defined as music types that are ranked first, or rated with the highest percentage of selections during Experiment 1. Non-preferred music stimuli will be defined as music selections ranked last, or rated with the lowest percentage of selections. An alternating treatments (multielement) design will be the method used to analyze the reinforcing effectiveness of the auditory stimuli. The basic strategy involved in this design is the rapid alteration of two or more treatments for each participant. The treatments to be alternated for each participant will include baseline (no auditory stimuli), preferred auditory stimuli, and non-preferred auditory stimuli. The order of the introduction of treatments will be randomized to control for sequential confounding, or the possibility that introducing Treatment A first,

27 Auditory Preference Assessment 27 for example, would bias the results in favor of Treatment A. The first phase will consist of baseline sessions in which no systematic consequences or no auditory stimuli will be delivered for responding. Baseline sessions will continue until stable responding is observed, which will be a minimum of three sessions. The fixed ratio reinforcement schedule for each participant will be determined from baseline performance. Following baseline, the alternating treatments condition will begin in which baseline, most-preferred auditory stimuli, and least-preferred auditory stimuli will alternate from session to session in a semi-random fashion. Experimental sessions will last 5 minutes each. During sessions, other than baseline, each time the schedule requirement is met, the participant will be provided with 15 seconds of access to the stimulus for that session. At the beginning of each session, the experimenter will present a verbal direction to engage in the predetermined task. For example, the experimenter will say, Sort the silverware. Specific tasks for each participant will be operationally defined to easily determine occurrences and non occurrences of the selected behaviors. The alternating treatments phase will continue until stable trends are visible for each condition. Interobserver agreement will be assessed on the occurrence of the target responses across all participants on at least 50% of all the sessions. Interobserver agreement will be evaluated using the total agreement method in which the lower number of recorded responses will be divided by the higher number of correct responses and the result multiplied by 100 to produce a percentage agreement score. All observers will receive training on specific data sheets, how to record data accurately, and will be provided with examples and nonexamples of approach behavior during the preference assessment and the various tasks involved in the study. Interobserver agreement levels will need to be at 80% or better. If

28 Auditory Preference Assessment 28 the 80% criteria level is not met, retraining of observers will be completed. This will consist of additional training, including modeling of examples and non-examples of correct responding. Analysis of Results The contingent use of preferred auditory stimuli is expected to increase the occurrence of the target behavior; whereas, the contingent use of non-preferred music stimuli and baseline conditions are expected to decrease the occurrence of the target behaviors, or remain at baseline levels. The results from Experiment 2 will be analyzed using a line graph. Responses will be expressed and analyzed as the number of correct responses per minute for baseline, high preference stimuli, and low preference stimuli.

29 Auditory Preference Assessment 29 References Belfiore, P.J., Browder, D.M., & Mace, C. (1994). Assessing choice-making and Preference in adults with profound mental retardation across community and center-based settings. Journal of Applied Behavioral Education, 4, Dattilo, J., & Mirenda, P. (1987). An application of a leisure preference assessment protocol for persons with severe handicaps. Journal of the Association for Persons With Severe Handicaps, 12, DeLeon, I. G., & Iwata, B.A. (1996). Evaluation of a multiple-stimulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29, Derby, K.M., Wacker, D.P., Andelman, M., Berg, W., Drew, J., Asmus, J., Prouty, A., & Laffey, P. (1995). Two measures of preference during forced-choice assessments. Journal of Applied Behavior Analysis, 28, Fisher, W., Piazza, C.C., Bowman, L.G., Hagopian, L.P., Owens, J.C., & Slevin, I. (1992). A comparison of two approaches for identifying reinforcers for persons with severe and profound disabilities. Journal of Applied Behavior Analysis, 25, Higbee, T., Haymes, L., Kenyon, S., (2003). Preference Assessment of Olfactory Stimuli: A Preliminary Investigation. Ivancic, M.T., & Bailey, J.S. (1996). Current limits to reinforcer identification for some persons with profound multiple disabilities. Research in Developmental Disabilities, 17, Pace, G. M., Ivanic, M.T., Edwards, G. L., Iwata, B.A., & Page, T.A. (1985). Assessments of stimulus preference and reinforcer value with profoundly retarded

30 Auditory Preference Assessment 30 individuals. Journal of Applied Behavior Analysis, 18, Parsons, M.B., & Reid, D.H. (1990). Assessing food preference among persons with Profound mental retardation: Providing opportunities to make choices. Journal of Applied Behavior Analysis, 23, Rotatori, A.F., Fox, B., & Switzky, H. (1979). An indirect technique for establishing Preferences for categories of reinforcement for severely and profoundly retarded Individuals. Perceptual and Motor Skills, 48,

31 Appendices Auditory Preference Assessment 31

32 Auditory Preference Assessment 32 Appendix A: PS Data Sheet Student Name: Date: By: Trial 21 Hip Hop Disney 1 Hard Rock Classical 22 Disney Country 2 Soft Rock Classical 23 Asian Disney 3 Country Classical 24 Classical Disney 4 Latin Classical 25 Hard Rock Hip Hop 5 Hip Hop Classical 26 Soft Rock Hip Hop 6 Disney Classical 27 Country Hip Hop 7 Asian Classical 28 Latin Hip Hop 8 Classical Hard Rock 29 Hip Hop Latin 9 Hard Rock Asian 30 Disney Hip Hop 10 Soft Rock Asian 31 Asian Hip Hop 11 Country Asian 32 Classical Hip Hop 12 Latin Asian 33 Hard Rock Latin 13 Hip Hop Asian 34 Soft Rock Latin 14 Disney Asian 35 Country Latin 15 Asian Soft Rock 36 Latin Hip Hop 16 Classical Asian 37 Hip Hop Latin 17 Hard Rock Disney 38 Disney Latin 18 Soft Rock Disney 39 Asian Latin 19 Country Disney 40 Classical Latin 20 Latin Disney 41 Hard Rock Country