Research Study on Language Skills & Data Collection

• This study is an evaluation of choice on instructional efficacy and individual preferences among children with autism.

Key Objectives

• Compare the differing effects of choice and no-choice reinforcement conditions regarding skill acquisition.
• Assess preference for choice-making opportunities among 3 children with autism using a concurrent-chains procedure.
• Replicate the experiment with 2 participants.

Key Findings

• Choice-making opportunities increased treatment efficacy for 2 of the 3 participants.
• All 3 participants showed a preference for choice-making opportunities.

Background

• Providing students with choice is beneficial during instructional programs.
• Choice can decrease problem behavior and increase academic engagement.
• Choice increases the frequency of academic responding.
• Choice has been shown to increase academic engagement but not skill acquisition.
• One study evaluated whether choice influenced the rate of skill acquisition and disruptive behavior during discrete-trial instruction for children with autism.
• Participants could select the task order and reinforcer (choice) or the experimenter selected both (no choice).
• Results demonstrated no difference in instructional efficacy.
• The effects of choice on acquisition may be more pronounced by providing opportunities to choose during the consequence portion of instruction rather than before.
• This study aims to evaluate the effects of choice-making when they are provided as a consequence.
• The analysis will replicate and extend previous findings that demonstrate preschool children with and without developmental disabilities prefer choice-making opportunities

Methods

• Three preschool-aged children with autism participated in the evaluation at a university-based early intervention program.
• Samuel was 5 years old and used simple sentences to mand and frequently engaged in social exchanges.
• Ethan was 4 years old, used short phrases to label and request items, and engaged in limited social conversation.
• Patrick was 3 years old and used a combination of short phrases and simple sentences to engage in tacting, manding, and social exchanges.
• All sessions were conducted in the participants' typical learning environment, a room (5 m by 5 m) with a child-sized table and chairs.
• Three sessions were conducted daily for 3 to 4 days per week.
• The child sat at the table with the experimenter next to them, and a secondary observer sat behind and to one side.
• Data was recorded on a trial-by-trial basis on unprompted correct responses, defined as vocalization of the targeted response before the delivery of the model prompt.
• Data was also collected on initial-link selections, defined as participants touching a card, during the concurrent-chains arrangement.
• A second observer independently recorded data for a minimum of 32% of all sessions.
• Observers recorded the same target responses in a trial.
• Agreement for initial-link selections was scored if both the experimenter and the second observer independently recorded the same initial-link selection.
• Interobserver agreement was calculated by dividing the number of agreements by the total number of trials.
• Mean agreement for unprompted correct responses was 99% for Samuel and 100% for both Ethan and Patrick.
• Initial-link selection agreement was 100% across all participants.
• Procedural integrity data was collected for a minimum of 22% of sessions.
• Procedural integrity was calculated for each session by dividing the number of correctly implemented trials by the total number of trials.
• Mean procedural integrity score was 99.4% for Samuel, 100% for Ethan, and 99.8% for Patrick.
• A paired-item preference assessment was conducted with small edible items and larger items broken into small pieces.
• The top three items associated with the highest selection percentage were used.
• A preference assessment for colors was conducted, and the three colors associated with the lowest selection percentage served as initial-link stimuli.
• An evaluation of the effect of contingent choice on correct responses was conducted using a nonconcurrent multiple baseline design embedded within a multiple schedule design. Instructional stimuli previously assigned to the control condition served as acquisition targets during the preference phase. Twice the experiment was conducted with Samuel and Patrick for replication.
• Six targets existed in each condition in the initial evaluation, and each target was presented twice.
• For Patrick and Samuel, targets consisted of intraverbal responses to common “wh–?” questions and questions about functions of objects.
• Ethan's targets were tacting pictures of common objects.
• During replication, 20 presented targets consisted of tacting pictures of popular people or characters.
• An attempt was made to equate targets across conditions by creating sets that contained targets with a similar number of syllables.
• During the efficacy portion of the evaluation, three sessions per day were conducted.
• The mastery criterion was set at two consecutive sessions with correct, unprompted responses at or above 90%.
• Conditions were presented in a pseudorandom and counterbalanced order, with all participants experiencing a choice session before a no-choice session during the first series of the evaluation.
• Reinforcers selected in the choice condition were yoked to the subsequent no-choice condition.
• At least 10 min were arranged between each condition to account for satiation associated with food items.
• Each instructional condition was paired with a colored index card, which served as the discriminative stimulus.
• Colored cards later served as initial links of a concurrent-chains arrangement in the instructional preference evaluation.
• An effective method for identifying consumer preference for instructional conditions is to measure selection among concurrently available responses that determine the contingencies in place in the terminal link.

Efficacy Evaluation

• Baseline: The experimenter presented the antecedent stimulus and allowed the participant 5s to respond, without consequences for correct or incorrect responses.
• Choice: The experimenter physically guided the participant to touch the relevant initial-link stimulus and then presented the antecedent stimulus. After a correct response, an array of three edible items was presented, and the participant was prompted to select one. The experimenter blocked attempts to select multiple items. Instructions used a progressive prompt delay to a vocal model prompt, with prompt delays at 0 s, 2 s, 5 s, 7 s, and 10 s. All participants first experienced trials at a 0-s prompt delay, and the second session at a 2-s delay. The prompt delay was increased (5 s, 7 s, 10 s) if at least 50% of unprompted incorrect responses were errors of omission. Reinforcement delivered after both unprompted and prompted correct responses until demonstrated two consecutive sessions with at least 50% unprompted correct responses. Thereafter, reinforcement was provided only for unprompted correct responses, continuing until performance reached mastery criterion.
• No choice: Identical to the choice condition except that after a correct response, the experimenter delivered an edible item yoked to selections in the previous choice condition. In the event of more reinforced trials in a no-choice session than in the previous choice session, the experimenter repeated the delivery of items based on the sequence of item selections.
• Control: The control condition was identical to baseline except that the therapist presented the relevant initial-link stimulus.

Preference Evaluation

• During preference evaluation, the experimenter placed the three initial-link stimuli in front of the participant with the instruction, "pick one." After initial-link selection, the experimenter presented the antecedent stimulus, and responding resulted in access to the terminal-link contingencies for that session.

Results and Discussion

• All participants showed low levels of correct responding during baseline sessions, and correct responding increased for all participants after implementing the instructional procedures.
• Samuel demonstrated mastery-level responding in the choice condition in five training sessions but in 14 training sessions in the no-choice condition.
• Differential efficacy was replicated in the choice condition versus the no-choice condition during Samuel's second evaluation.
• Ethan met mastery criterion within an equal number of choice and no-choice sessions.
• Patrick's correct responding reached the mastery criterion in fewer training sessions in the choice condition than in the no-choice condition in both evaluations.
• Two of three participants required fewer instructional sessions to reach the mastery criterion under choice conditions.
• All participants favored choice, Samuel exclusively preferred the choice condition, Ethan preferred it in seven of eight sessions and Patrick exclusively selected it in both evaluations.
• A yoking procedure attempted to isolate choice as a variable and control for item preference.
• The differential consequences associated with choosing can influence results.
• Individuals with typical and atypical development prefer choice-making opportunities.
• Providing choice of reinforcement is one way to provide choice-making opportunities and promote personal liberties for individuals with autism and related developmental disabilities.

This study focuses on language skills and data collection methods. It covers procedural integrity, preference assessments, and interobserver agreement. The study also involves evaluating and teaching language skills such as tacting.