Autism Constant vs. Varied Reinforcers on Preferences and Resistance to Change

Autism: Constant vs. Varied

Varied reinforcers have been shown repeatedly to have many benefits over constant reinforers. But real life environmental challenges can preclude reliable measurement results, making it less clear how the response patterns contribute to subject preferences and their resistance to change. The study article by Miros et al. utilize a fixed-ration assessment methodology grounded in an additional level of variability that fluctuates in their testing methodology between the subject's preference assessment and the baseline and reinforcement tests of resistance. The expectation is to address findings in other studies that have limited the generality of results because ordinary disruptive challenges are confronted in applied settings. As such, this study moves the discussion forward and allows for a greater reliability of the role of variation in reinforcers in affirming and sustaining prosocial outcomes.

In the Miros study four young male subjects between the ages of six and 10 were selected. Each had different levels of communication and interactive abilities, and as such required the use of targeted ways that some methodological techniques were used. The authors reviewed these requirements and then detailed the configuration of the study space and the layout of the assessment and monitoring equipment, therein confirming how each subject used these setups using other recognized measures of interaction with constant and varied reinforcers (foods). The same setups were then used again with matched and appropriate disruptive operants in the form of videos aligned with subject likes.

As an example of the techniques used, the Reinforcer Assessment for Individuals with Severe Disabilities (RAISD) determine the top 10 food items that were preferred for each subject. Physical modeling preliminary testing using a "Do This" methodology was implemented, setting up the circumstances where each subject could undergo three F1 (fixed ratio varied or fixed ratio constant with continuous reinforcement) preference assessments -- the orientation and sessions counterbalanced across participants to add further to the reliability of the results (Cosgrove, 2011). Various randomized F10 baseline conditions were then established for testing preferences for choosing specific reinforcers. These results showed a clear and consistent pattern favoring varied rewards in a way consistent with past studies. Statistic tests were conducted to ensure that the results were significant.

Next, a series of distractions were introduced using similar baseline methodologies. Preferred videos were identified for each subject in accordance with their communication limitations. Other levels of environmental variation occurred because of holiday breaks for two subjects (Peter and Dean), each of whom underwent assessments on three distraction conditions. The other two subjects (Sammy and Frank) interacted with four distraction settings.

Two of the boys, Frank and Sammy, showed greater deviations in some of their responses, though statistical analyses indicated they were not beyond what was expected. Both of these boys used American Sign Language as part of their preferred methods of communication, and were on the younger end of the age spectrum. Frank is deaf. During the distraction session, Sammy's involvement with the video brought about aggressive and disruptive behavior. Peter and Dean, on the other hand, experienced somewhat shorter distraction assessments, but none of these circumstances interfered with the impact of the preference for varied reinforcers and contributed to the conclusion that the methods allowed for a better understanding of the outcomes in applied settings.

Other studies have indicated lesser degrees of uncertainty about the impact of varied and constant reinforcers. Older assessments of older, more fixed studies showed that the results could well be related to how individuals react over the duration of the study itself, or because of outside or environmental facts (Bowman, Piazza, Fisher, Hagopian & Jeffry, et al. 1997). More recent reviews, such as Assessment and Treatment of Problem Behavior Occasioned by Interruption of Free Operant Behavior by Hogapian, et al. (2005), further refined how interruptions in reinforcers can impact the results. Hogapian's 2005 work thus confirms even more how methodological changes can complicate understanding but may be more relevant in conditions where varied reinforces are further influenced by relatively uncontrolled outside sources common to applied settings. Miros et al. further this progression toward validating varied preferences and consistency of expectations.