Discussion and Conceptual Analysis of Four Group Contingencies for Behavioral Process Improvement in an ABA Service Delivery Quality Framework

Abstract

Applied behavior analysis (ABA) organizations that provide services to children with autism can apply contingencies to improve employee performance or behavioral processes. Such contingencies may be especially important for maximizing ABA service delivery quality (ASDQ). For some behavioral processes, group contingencies applied to the behavior of individuals within the process may be more appropriate than individualized contingencies. In the history of the profession, behavior analysts have used group contingencies at the operant level of selection, such as independent, interdependent, and dependent group contingencies. However, recent experimental work in culturo-behavioral science suggests the metacontingency, an analogue of the operant contingency at the cultural level of selection, can also control the behavior of individuals in a group. This article discusses how such group-oriented contingencies could be used by managers in organizations in behavioral process improvement efforts to target key performance indicators of quality in an ASDQ framework. The paper ends with a discussion of limitations and future research.

The year 2021 was significant in the applied behavior analysis (ABA) service delivery industry (i.e., “the industry”) for service quality and outcomes. Autism spectrum disorder (American Psychiatric Association, 2013) therapy organizations and insurance companies in the industry are becoming increasingly focused on clinical standards, and clinical outcomes, and the quality of ABA autism services. ABA service delivery quality (ASDQ) was defined at the organizational level with a focus on standards (including clinical outcomes) and financial health, for the first time in the scientific literature, by Silbaugh and El Fattal (2021). In that same year, TRICARE’s revised Autism Care Demonstration began to require that ABA autism therapy organizations report multiple valid outcome measure scores of beneficiaries in order for Health Net Federal Services to process authorization requests for ABA services (Health Net Federal Services, 2021). The industry continues to wrestle with the best ways to measure and maximize clinical outcomes for individuals with autism, with some promising new advances. For example, the International Consortium for Health Outcomes Measurement (ICHOM, 2021) recently released the Autism Spectrum Disorder Standard Set of outcome measures as the next step in shifting to a value-based care model potentially selective for autism care value. In September 2021, the Behavioral Health Center of Excellence (BHCOE), a major accrediting body for ABA autism therapy organizations, released a “BHCOE ABA Outcomes Framework” ABA autism therapy organizations can use as a tool to promote “adoption of common measures to establish the value of ABA treatment” (BHCOE, 2021a). This document describes a set of measurement instruments and a process in which they can be used to assess outcomes. Better outcomes are reasonable to expect from ABA autism therapy organizations that more effectively meet industry standards and deliver valuable high-quality ABA services, but the relationship among standards attainment, outcomes, and service quality remains an unanswered empirical question. The impact of practica experiences of students pursuing behavior analyst certification on service quality also remains an empirical question. In a press release on October 19, the BHCOE announced new on-site practica accreditation standards for, “universities who seek to ensure the quality of the practica experiences that ABA trainees receive and to align on-site learning with best practices in supervision” (BHCOE, 2021b).

The rising demand for certified behavior analysts suggests the demand for ABA services for individuals with autism continues to rise nationwide according to a report commissioned by the Behavior Analyst Certification Board (BACB, 2021). A report from Verified Market Research (2021) suggested the value of the autism spectrum disorders market (i.e., a part of the health-care sector) is projected to grow by 4.27% to $4.53 billion by 2027. The increasing value of the autism therapy market could attract additional private equity investment in the near future, which has been criticized by experts in economics and public policy for its ability to destabilize the health-care industry by reducing competition with rapid consolidation and focusing on short-term revenue generation (Scheffler et al., 2021). It seems reasonable to assume that destabilization of an industry is a threat to service quality, although the relationship between private equity investment and ABA service quality is also an unanswered empirical question.

The year 2022 is shaping up to have a big impact on quality, too. On January 31, the Council of Autism Service Providers (CASP) issued a press release announcing the availability of new organizational guidelines for business operations, clinical operations, and risk management. These guidelines provide “recommendations for best practices regarding organizational policies, procedures, and infrastructure necessary for individual practitioners to consistently deliver high-quality applied behavior analysis (ABA) services” (CASP, 2022a). On May 2, CASP announced it had launched the Autism Commission on Quality, which includes the ABA Standards Committee originally founded through CASP, to “develop quality and performance standards for ABA services and implement a meaningful accreditation review process to ensure applicants meet these standards” (CASP, 2022b). On May 16, a CASP press release announced the publication of “Applied Behavior Analysis Telehealth Parameters Best Practices” to help organizations, “improve both the quality of telehealth care and providers’ understanding of quality telehealth services” (CASP, 2022c). In April (BHCOE, 2022a) and May (BHCOE, 2022b), the BHCOE announced partnerships with health-care organizations to promote quality by advancing the measurement of health outcomes for autism treatment essential for value-based health care and reimbursement systems that reward organizations for delivering better outcomes at a lower cost per patient (Porter & Teisberg, 2006).

These events suggest it has never been more important to objectively define ABA service quality, evaluate the variables in service delivery settings that control quality, and take full advantage of behavioral technology to improve and maintain high quality services over time. In their conceptualization of ABA service quality at the organizational level, Silbaugh and El Fattal (2021) drew attention to limitations of relying on external audits for determining service quality and suggested organizations should focus more on the role contingencies within organizations play in determining ASDQ. However, they stopped short of discussing the kinds of continencies organizations could arrange to improve ASDQ through process improvement. Through conceptual analysis, this article draws comparisons between metacontingency-based interventions and operant group contingency interventions in ABA autism therapy settings, introduces a potentially useful new terminological distinction, and discusses how managers could use four group-oriented contingencies for behavioral process improvement within organizations to empirically promote ASDQ.

ASDQ and the Evidence-Based Practice of ABA at the Organizational Level

Silbaugh and El Fattal (2021) defined ASDQ as, “the extent to which an organization’s ABA products, services, and outcomes meet standards determined by professionals and consumers, over time, in response to changes in a receiving system, while maximizing the financial health of the organization” (p. 8). Leaders in ABA autism therapy organizations implement change initiatives to improve performance and processes such as clinical supervision, the intake process, or treatment planning believed to affect service quality. To systematically evaluate the impact of change initiatives on ASDQ, ABA autism therapy organizations could engage in the evidence-based practice of ABA at the organizational level. By analogy to the evidence-based practice of ABA at the practitioner level (Slocum et al., 2014), at the organizational level evidence-based practice is conceptualized as “organizational decision making based on the best available evidence, consumer values/context, ... and operational expertise” (Silbaugh & El Fattal, 2021). More conceptual analysis can clarify exactly what constitutes “operational expertise” in that framework. However, at the practitioner level clinical expertise is said to include knowledge of the research literature and data-based decision making (Slocum et al., 2014). Thus, for the present paper, behavior analysts can assume operational expertise might include the skilled application of well-established systems design and process improvement tools in organizational behavior management such as behavioral systems analysis (e.g., Sigurdsson & McGee, 2015), as representative of the use of relevant research literature knowledge and data-based decision making at the organizational level.

Behavioral Systems Analysis

Within an ASDQ framework, the ultimate quality goal of an organization that provides ABA therapy to children with autism can be viewed as attainment of their internal professional and consumer standards (i.e., which the organization believes defines high ASDQ) and strong financial health, resulting from recurring pro-quality practices exhibited by employees (Silbaugh & El Fattal, 2021) in a total performance system (e.g., Brethower, 1982; Diener et al., 2009) of interrelated behavioral processes.

Brodhead et al. (2018) provide a succinct and highly accessible overview of behavioral systems analysis for behavior analysts. Malott and Garcia (1987) defined a system as, “an organized, integrated, unified set of components, accomplishing a particular set of ultimate goals or objectives” (p. 127) and a behavioral system as, “one in which the principle components are organisms, usually human beings, working together to accomplish some set of ultimate goals or objectives” (p. 128). Behavioral systems are therefore comprised of behavioral processes subject to control by principles that govern human behavior. Behavioral systems analysis is an approach to process improvement that takes into consideration the interrelatedness of multiple levels of organizational complexity and their component processes, within the context of a general organizational framework such as the total performance system model described below (Sigurdsson & McGee, 2015). A wide variety of tools, such as the Behavioral Systems Analysis Questionnaire (Diener et al., 2009) have been developed for each of the organizational, process, and performer levels (Sigurdsson & McGee, 2015).

Total Performance Systems

Brethower’s (1982) total performance system was based on a general behavioral approach to solving organizational problems. A given performance within a system or process was specified and the analyst determined whether the problem with performance was due to inadequate job design, tools, or materials; inadequate knowledge or skills; or ineffective reinforcement contingencies. The analyst then addressed the problem by adjusting the job, providing training, or introducing or strengthening reinforcement contingencies; then the analyst monitored progress and adjusted the intervention until the desired performance was achieved and maintained. Brethower (1982) and colleagues visually depicted the total performance system with a simple diagram comprised of seven components: (1) inputs; (2) processing system; (3) processing system feedback; (4) outputs; (5) receiving system; (6) receiving system feedback; and (7) goal(s). By viewing the organization as comprised of these seven components, it was easier to, “deal with difficulties encountered in organizational behavior management” (p. 350) and manage complexity by, “coordinating key resources in response to changing needs and demands” (p. 350).

Clinical outcome standards are an essential element of standards targeted in an ASDQ framework (Silbaugh & El Fattal, 2021). Clinical outcomes of ABA services for individuals with autism are the result of a complex interconnected set of processes including direct treatment usually delivered by registered behavior technicians (RBT). Thus, ABA autism therapy organizations have behavioral processes (i.e., within the total performance system) for preparing new hire behavior technicians to pass the RBT certification exam and demonstrate basic clinical competence delivering direct treatment. In the event this process adversely affects ASDQ by failing to ensure RBTs meet professional and consumer standards assumed important for meeting clinical outcome standards, the organization can use behavioral systems analysis targeting key performance indicators (KPI) in the evidence-based practice of ABA at the organizational level to improve it.

Key Performance Indicators

One definition of KPI is a measure of an aspect of organizational performance that is critical for the future success of the organization because it helps managers make decisions that positively affect business goals (Parmenter, 2015). KPIs can benefit organizations in multiple ways, such as aligning everyday actions of employees to critical success factors, improvements in performance, and a more widespread sense of ownership, empowerment, and fulfillment throughout the organization (Parmenter, 2015). Silbaugh and El Fattal (2021) suggested KPIs could play an important role in using evidence-based practices at the organizational level to promote ASDQ in ABA organizations serving individuals with autism. When KPIs measure aspects of quality in an organization, Silbaugh and El Fattal (2021) suggested organizations refer to them as quality-dependent KPIs (QD-KPI). Variables thought to affect KPIs can be modified or controlled. KPI data must be valid, reliable, and accurate. Parmenter (2015) suggested that KPI are (1) nonfinancial; (2) measured frequently; (3) acted on by the chief executive officer and senior management team; (4) clearly indicate what action is required of staff; (5) measures that tie responsibility down to a team or cluster of teams; (6) have a significant impact on the organization; and (7) encourage appropriate action. When an indicator is not crucial to the organization’s survival but nevertheless is important and can effectively guide decision making, Parmenter (2015) recommends referring to it as a performance indicator rather than a KPI. For simplicity, all indicators in this article are referred to as KPIs.

Contingency-Based Behavioral Process Improvement in an ASDQ Framework

As discussed in the introduction, organizations now have access to multiple sets of standards and guidelines for quality ABA autism service delivery. The suitability of a given set of standards or guidelines may vary across organizations. As such, organizations should adopt professional and consumer standards and guidelines that are sensitive to the cultures and communities they serve and aligned with the organization’s vision, mission, and core values. For example, leaders in an organization that places a high value on ethical practices might choose to select the BHCOE’s 2021 Full Accreditation Standard A1.0 as part of a comprehensive set of professional standards they adopt in their ASDQ framework. The standard reads: “The organization has a designated ethics officer or ethics committee to address ethical issues such as patient programming, and organizational, staff, and patient concerns” (BHCOE, 2022c). What an ethics committee is, and how to set one up in an organization, has been described (Cox, 2020). ABA autism therapy organizations with sufficient resources might also consider establishing an ethics network with the potential for an even broader impact on the organization’s culture (LeBlanc et al., 2020).

An ethics committee is a group of people, presumably knowledgeable about ethics relevant to their organization, who must coordinate their behavior to address ethics concerns from staff, consumers, and perhaps even members of the community or other stakeholders (Cox, 2020). Because ethics expertise doesn’t lay with a single individual within the committee (Cox, 2020), neither do the committee’s decisions.

Activities common to ethics committees in health-care institutions include, “educating health care professionals about applied ethics relevant to their workplace settings, drafting and reviewing organizational policy, providing ethics case consultation, providing a forum for employees to voice ethical challenges they face on the job, and reviewing potential research projects within the organization” (Cox, 2020, p. 941). In an ABA organization serving children with autism, an ethics committee’s basic responsibilities might include addressing ethics concerns within the organization, writing and implementing ethics policy, and arranging contingencies that consistently yield timely and satisfactory resolutions of concerns from caregivers, staff, or other stakeholders brought to the committee.

The recurring activities and results of an ethics committee can be viewed as comprising a behavioral process within a total performance system. In this view, standards attainment by an ethics committee can be evaluated and improved over time with an ASDIER approach to behavioral systems analysis targeting ethics KPIs in the evidence-based practice of ABA at the organizational level.

ASDIER is one approach to process improvement in behavioral systems analysis, originally referred to by Malott (1974) as a universal systems-design approach. The six steps of this approach are, (1) analyze; (2) specify; (3) design; (4) implement; (5) evaluate; and (6) recycle. The ASDIER behavioral approach to systems design or process improvement has some features in common with a frequently used continuous improvement method rooted in the scientific method called the “Plan-Do-Check-Act” cycle (PDCA; a.k.a. Shewhart Cycle or Deming Cycle) established in the 1950s by W. Edwards Deming and later modified by Japanese manufacturers (Deming Cycle (PDCA), 2000; Moen & Norman, 2010). In ASDIER, analyze means identify the controlling contingencies relevant to the system. Specify means operationally defining target behaviors within the system and determining performance objectives for those behaviors. Design means to develop an intervention on target behaviors within the system, which includes observation, measurement, and arranging contingencies. Evaluate means to empirically (and experimentally if possible) determine whether the intervention is having the desired effect. Recycle means to engage in a continuous improvement process based on data (i.e., sometimes referred to as continuous quality improvement) until the system is performing as desired.

In the analyze phase, an organization could identify quantifiable KPIs in their ASDQ framework to assess not just attainment of the standard (e.g., establish an ethics committee), but to raise the rigor of the standard by setting expected KPI levels (i.e., relative to industry benchmarks when available) in the specify phase to enable monitoring of the extent to which the standard is attained and maintained over time. In the design phase, the organization would arrange contingencies to ensure the committee consistently meets the relevant ASDQ standards. For example, the organization might collect baseline data on KPIs such as the percentage of concerns that staff and/or patients rated as having been satisfactorily resolved, the percentage of ethical concerns that did not result in an interruption in services, or perhaps the average time it takes the committee to bring an ethical issue to a resolution; then redefine the standard in terms of each expected KPI level. Such as, consistently reducing the average time to satisfactorily resolve ethical issues to two weeks and maintaining target KPI levels over time. Then antecedents and consequences could be arranged and implemented to ensure improvements in KPIs over time. Methods should be employed to evaluate whether programmed contingencies are having the desired effect on KPIs, and the member(s) of the organization responsible for the committee’s ASDIER process would recycle through this process until target KPI levels are met.

The performance of the committee as a behavioral process and its interactions with other processes within the organization’s total performance system are clearly dependent on two things. First, they depend on individual and group contingencies for employee performances represented by KPIs arranged in the design phase of the ASDIER process. Second, they depend on how data are used with the evidence-based practice at the organizational level in all phases of the ASDIER process to ensure contingencies are effective.

Operant Contingencies

Nearly a century of basic and applied research has shown us that behavior analysts can use applied behavioral technology based on the four-term contingency to change behavior at the level of the individual. The four-term contingency describes the functional relations between a motivating operation (first term), a discriminative stimulus class (SD; second term), one or more members of a functional response class (third term), and a controlling consequent stimulus class (fourth term; Michael, 1982; Miguel, 2013). The four-term contingency can also be applied in behavioral interventions to teach new skills or influence characteristics (e.g., dimensional quantities) of socially significant behavior or employee behavior important to organizations (Daniels & Bailey, 2014). For example, within a motivating context (i.e., conditioned establishing operation for positive reinforcement) that makes a small monetary bonus (i.e., generalized conditioned positive reinforcement) effective as a reinforcer, small bonuses can be delivered to an ABA autism therapy organization’s intake coordinator for target behaviors they exhibit related to producing expected levels of a target KPI (e.g., scores on customer satisfaction surveys) for ABA service referrals received (discriminative stimulus for generalized conditioned positive reinforcement). When arranging contingencies to influence the results or products of behavioral processes within an organization, group contingencies may in some cases be more feasible, effective, or socially acceptable to peers (i.e., avoid perceived favoritism or bias), depending on the context. For example, bonuses could instead be delivered to the whole intake team contingent on customer satisfaction scores exceeding a target KPI level.

Group Contingencies

Two people comprise the smallest possible group. Cooper et al. (2020) define a group contingency as one in which “a common consequence is contingent on the behavior of one member of the group, the behavior of part of the group, or the behavior of everyone in the group” (p. 664). That the contingency is applied to behavior, not the results or accomplishments of behavior (Gilbert, 1978; Lindsley, 1991), is notable and important in pursuit of a more versatile group contingency technology. Hayes (1976) suggested applied behavior analytic work in group contingencies can be traced to an early experimental study of the effects of edible reinforcement on cooperation in children conducted by Azrin and Lindsley (1956). In their experiment, pairs of children played a game in which each child earned edible reinforcement for inserting a stylus into the same holes on a board centered between them. The results showed that the contingency increased cooperative behavior relative to an extinction period. Benefits or advantages of group contingencies relative to individual contingencies include that they can (1) save time on administration; (2) reduce staffing or staff workload; (3) address problems in settings where multiple individualized contingencies are not feasible; (4) occasion opportunities for peers to function as change agents; and (5) capitalize on the use of peer-monitoring for progress monitoring (Cooper et al., 2020).

Group contingencies have been applied in a diverse array of settings to produce socially significant outcomes such as smoking cessation (e.g., Meredith & Dallery, 2013) and improvements in classroom management (e.g., Bowman-Perrott et al., 2016). For some uses, the application of group contingencies is an evidence-based practice. For example, the use of group contingencies in classrooms to support students with challenging behavior is an evidence-based practice based on What Works Clearinghouse standards and procedures for evaluating single-subject research (Maggin et al., 2012).

Types of Group Contingencies

In theoretical terms, group contingencies can be based on one or more functional consequences such as conditioned positive reinforcement, conditioned negative reinforcement, conditioned positive punishment, conditioned negative punishment or unconditioned positive punishment. Types of group contingencies are based on which group members must meet the performance criterion and which group members receive the programmed consequence when the criterion is met (Cariveau et al., 2020). Three prototypical group contingencies are independent, dependent, and interdependent (Litoe & Pumroy, 1975).

For the remainder of this article, the alternative terms “group-oriented contingencies” and “operant group contingencies” (OGC) are used for the following reasons. In the history of the profession the concept of the “group contingency” in ABA has been applied only to OGC, thereby excluding group contingencies at the cultural level of selection (i.e., metacontingencies; e.g., Cooper et al., 2020). This practice has potentially limited the kinds of group level interventions leaders in ABA organizations have considered in their efforts to improve behavioral processes. The term “group-oriented contingencies” can be used as suggested by Litoe and Pumroy (1975) to refer to any contingency applied to affect behavior of individuals in groups. This terminology avoids implying that groups behave (i.e., only individual organisms behave) and allows behavior analysts to inclusively refer to group contingencies based on operant selection (i.e., independent, dependent, and interdependent) and cultural selection (i.e., metacontingencies) described in detail below. The term OGC can then be more deliberately used to specify that the terms independent, dependent, and interdependent group contingencies are only used to refer to operant contingencies applied to the same behavior (not the results of the behavior) of two or more individuals.

Independent OGC

In an independent OGC based on positive reinforcement, the target behavior and reinforcer are the same for all group members, and the reinforcer is delivered on an individual basis only to group members who meet the performance criterion (e.g., Sloman et al., 2014).

Sloman et al. (2014) evaluated the effects of an independent group contingency on inappropriate behavior exhibited by offenders with developmental disorders at a secure residential facility. The contingency was that each resident could access an enriched special activity program consisting of special lunches, music, and socialization opportunities if they exhibited no more than one instance of minor inappropriate behavior and no instances of severe inappropriate behavior. The contingency was superimposed on existing level and token systems and behavior was monitored via observers and recorded video surveillance for the duration of the study. Weeks with and without the contingency were alternated in a multielement design. The intervention was an independent OGC because all group members (i.e., residents) had the opportunity to earn the same reinforcer, the target behavior (i.e., inappropriate behavior, although topographies likely differed across residents) was the same for all group members, and the reinforcer was delivered to each group member only if they individually met the inappropriate behavior criterion. Levels of inappropriate behavior were lower overall when the independent group contingency was in effect.

For example, a clinical BCBA supervisor in an ABA autism therapy organization could give each RBT a reinforcer contingent on demonstrating quality indicators (i.e., skills) of early intensive behavioral intervention (e.g., Langh et al., 2021) to criterion. It should be noted that in such an arrangement two or more RBTs can earn the same reinforcer, the reinforcer is contingent on RBT behavior, not client behavior, and the RBTs do not need to interact with each other to produce the reinforcer.

Dependent OGC

In a dependent OGC based on positive reinforcement, the reinforcer is delivered to all group members only if one member or a subset of group members meet the performance criterion (Cooper et al., 2020). Heering and Wilder (2006) used a dependent group contingency to increase third and fourth grade students’ on-task behavior in two classrooms during mathematics in a private elementary school. First the researchers conducted stimulus preference assessments to identify common reinforcers all students could earn by meeting the group contingency. The researchers also provided all class members with a description of the intervention. For the intervention, during class periods the researcher observed a randomly selected row of students at predetermined intervals and recorded whether all students in the row were on task. At the end of the class period, the researchers gave all students in class access to the reinforcers if students were on-task for 75% or more of the selected intervals. This was a dependent group contingency because all students had the opportunity to earn the same reinforcers, the target behavior was the same for all group members, and the reinforcer was delivered to all students contingent on a subset (or subsets) of students meeting the on-task criterion. The results demonstrated that the dependent group continency increased on-task behavior in both classrooms and researchers obtained positive social validity assessment results.

An example of a dependent OGC in an ABA autism therapy organization is if one RBT or a subset of RBTs can earn the reinforcer for a larger team of RBTs by demonstrating the target early intensive behavioral intervention quality indicator. It is notable that like the independent OGC, two or more RBTs can earn the same reinforcer, the reinforcer is contingent on RBT behavior not client behavior, and the RBTs do not need to interact with each other to produce the reinforcer. Readers interested in settings, populations, outcomes measures, intervention components, and procedural parameters of dependent OGC can learn about those characteristics in the recent systematic review by Page et al. (2021) spanning 1970–2019.

Interdependent OGC

In an interdependent group contingency based on positive reinforcement, no group member can produce the reinforcer alone. The reinforcer is delivered to all group members contingent on all group members meeting the performance criterion. This group contingency is also known as the “all or none” group contingency. Variations on this type of operant group contingency include delivering reinforcement based on a group mean score or on the results of the good behavior game or good student game (Cooper et al., 2020).

The most well-known application of the interdependent OGC in the scientific literature is the Good Behavior Game (e.g., Bowman-Perrott et al., 2016). In their groundbreaking study, Barrish et al. (1969) observed a disruptive classroom, and divided the classroom into teams who worked towards access to common classroom privileges. When a student on a team engaged in disruptive behavior, the teacher marked a tally on the whiteboard. At the end of the period, teams who received the fewest tallies, or fewer than a target number of marks, would win access to the privileges. The results showed the intervention significantly reduced talking out and out-of-seat disruptive behavior. The study authors suggested that the Good Behavior Game creates social strain among the team members to act appropriately, although it is unclear exactly what social strain is. It should be noted that, based on the description of the procedures, team members were not required to interact with each other to reduce disruptive behavior. In Bowman-Perrott et al. (2016) the results of their meta-analysis suggested a significant decrease in undesirable behavior across many Good Behavior Game studies. Interdependent OGC have been demonstrated to improve other socially important behaviors in other populations too such as smoking abstinence in adults (e.g., Meredith & Dallery, 2013)

Consider an ABA autism therapy organization example in which all five RBTs in a clinic are required to meet the early intensive behavioral intervention quality indicators criterion for all of the RBTs to access the reinforcer. Once again, two or more RBTs can earn the same reinforcer, the reinforcer is contingent on RBT behavior not client behavior, and the RBTs do not need to interact with each other to produce the reinforcer.

Units of Analysis and Units of Measurement

Groups do not behave. Groups are comprised of individuals engaging in behavior. Any given OGC consists of a dependency between the behavior of one or more individuals in the group and an operant consequence (e.g., reinforcer or punisher) available to everyone in the group. Thus, although the effects of group contingencies sometimes are evaluated at the level of the individual, and other times at the group level (e.g., Page et al., 2021), the unit of analysis (i.e., that part of the phenomenon being measured) in OGCs is the operant response class targeted by the group contingency. The unit of measurement is the group or individual level dependent measure used to evaluate the effects of the contingency (Johnston & Pennypacker, 1993, p. 66) on the unit of analysis. It is a measure of change in the phenomenon over the course of the intervention. For example, during the independent contingency condition in Meredith and Dallery (2013) the unit of analysis was submission of video samples of negative breath carbon monoxide measurements, and the units of measurement were the percentage of negative carbon monoxide samples and the most consecutive negative carbon monoxide samples. The probability of the target operant response class being exhibited by group members is controlled by processes of reinforcement or punishment, depending on the OGC. Because OGC are dependencies between programmed consequences and the behavior of individual group members, they do not require functional interdependence between behaviors emitted by multiple group members. That is, under an OGC, the contingency does not seem to require individuals to interact or exert control over one another’s behavior to meet the performance criterion. Because they rely on the use of operant contingencies, OGCs are procedures that enable researchers and practitioners to evaluate the effects of operant selection on groups of individuals’ behaviors. Given those additional considerations, behavior analysts can classify the three prototypical OGCs as types of group-oriented contingencies and add a fourth to the set: the metacontingency.

Metacontingency

The metacontingency is an analogue of the operant contingency conceptualized by Sigrid Glenn in the 1980s (e.g., Glenn, 1986). Researchers use metacontingencies to experimentally evaluate environmental variables that control group member behavior through cultural selection. In this experimental research, cultural practices and their products are dependent variables and selecting environmental conditions called cultural consequences (CC) are independent variables (e.g., Vichi et al., 2009). Just as immediate consequences can increase or decrease the future probability of a response reoccurring in the future based on operant selection, CC increase or decrease the future probability of group cultural practices recurring in the future based on cultural selection (Glenn et al., 2016).

The definition of the metacontingency has undergone much scrutiny and has evolved for decades. The most recent and perhaps now widely accepted definition of the metacontingency is, “A contingent relation between 1) recurring interlocking behavioral contingencies having an aggregate product and 2) selecting environmental events or conditions” (Glenn et al., 2016). This two-term metacontingency is comprised of a recurring culturant (first term) and a cultural consequence (CC; second term). A contingency diagram illustrates the metacontingency in Fig. 1.

Fig. 1.

Generic diagram of a metacontingency. AP aggregate product, CC cultural consequence, IBC interlocking behavioral contingencies

Culturant (Hunter, 2012) is a label for a dependency between interlocking behavioral contingencies (IBC) and their aggregate products (AP). This dependence is analogous to the interaction between a rat’s lever press and the switch closure produced by lever pressing (Glenn et al., 2016). Except, instead of switch closure produced by a single rat pressing a lever, a product (i.e., accomplishment or result) is generated by the functionally interdependent behavior of two or more individuals in a group. In contrast to OGC, which rely on the response class as a unit of analysis, Baia and Sampaio (2019) suggested that in a metacontingency the culturant is the unit of analysis. IBCs are comprised of behaviors exhibited by individual group members who interact with the behavior of other group members through operant functional relations. Each group member’s behavior serves as controlling antecedents or consequences for other group members’ behavior. Thus, unlike the behavior of individuals in OGC, the behavior of individuals which comprise a culturant are functionally interdependent, meaning they are causally related and depend on one another to recur over time.

For example, in team sports, each player’s behavior is influenced by every other player’s behavior, and team members interact with each other in ways that are necessary to play and win games. The team’s score is an example of an AP. An AP is an outcome, result, or accomplishment of IBCs (Glenn et al., 2016). No individual member of the group can produce the AP. Behaviors of those members must interact. AP as a product measure of IBC is the basis for selection by CC, which is analogous to reinforcement delivered contingent on a rat’s lever presses producing switch closures (Glenn et al., 2016). For example, in team sports, a higher team score results in the consequent designation of “winning team,” which is contacted by all team members. Being designated as the winning team is contingent on the team producing the higher score, and to the extent that a metacontingency is in effect, the “win” is a CC that will select those IBCs that produced the win to reoccur in future games.

Over repeated cycles of variation, selection, and differential replication of IBC that produce AP on which the CC is contingent, selective effects of CC cause some variations in IBC to increase in future probability and others to decrease in future probability. In a recent review of experimental analysis of cultures published in English and Portuguese, the authors found that culturants can maintain across generations (i.e., as participants come and go over time), selection operates on culturants under conditions in which communication is restricted, culturants can be controlled by aversive stimuli, culturant responding is sensitive to analogue schedules of reinforcement and extinction, culturants are sensitive to CC even when operants and CCs are nonconcurrent, and CC can select culturants even when concurrent operant consequences produce relatively higher magnitude consequences (Cihon et al., 2020).

It is important to distinguish the selective effects of CC and consequences produced by behavior. Consequences produced by behavior can reinforce or punish that behavior. But CC cannot be produced by the behavior of any one individual because CC are contingent on AP and delivered by a selecting or receiving system (Glenn et al., 2016). For example, no one basketball player on a team can produce the win. Multiple team members must work together to score points throughout the game and based on which team has the highest score (i.e., higher rate of the target AP), the game officials deliver the “win.” In the future, the performance of the team as a whole (i.e., their future IBCs) can be explained at the cultural level of selection as due to a history of differential CC for certain variations in IBC that were more or less successful at scoring points (Glenn et al., 2016).

Thus, the mechanisms by which culturants are selected by CC appear to be different than the mechanisms by which behavior is controlled by consequences. IBCs are comprised of operant behaviors interacting with each other as controlling stimuli and responses. If IBCs vary, behavior analysts can infer the variables in the environment controlling the corresponding operants have been altered. It follows then that the contingency between a culturant (i.e., IBC-> AP relation) and CC may have function-altering effects, dimension-altering effects, value-altering effects, or some combination of these, on the operants within IBCs. Within the context of arranging contingencies to improve processes that affect ASDQ within a total performance system, it follows that metacontingencies may have different effects than other group-oriented contingencies on the performance of individuals within a behavioral process, and therefore may yield different products from processes than OGC. This is also an empirical question ripe for investigation.

Preliminary Summary and Conclusions

The ABA industry serving individuals with autism (i.e., autistic people) has expanded and increased its focus on service quality and outcomes. This trend suggests an urgent need for a better understanding of the contingencies that control behavioral processes associated with ASDQ. Organizations can promote high ASDQ by using the evidence-based practice of ABA at the organizational level with well-established organizational behavior management tools such as behavioral systems analysis in a total performance system to intervene on KPIs. The contingencies controlling behavioral processes within the behavioral systems analysis literature have received little attention (Fryling et al., 2022), however, this article argues that managers can design and implement interventions using group-oriented contingencies to improve behavioral processes and promote standards attainment. A lack of applied research on metacontingencies in ABA organizations suggests behavior analysts in those settings may have (1) underutilized contingencies of cultural selection to improve behavioral systems and processes; (2) applied organizational change initiatives using metacontingency-based interventions without recognizing it; or (3) intentionally used such interventions, but without documenting it in the scientific record. When metacontingencies are considered a fourth type of prototypical group-oriented contingency and the four types are compared, clear difference emerge, which may have important implications for the effectiveness of interventions on KPIs for promoting high ASDQ.

Key Differences between Group-Oriented Contingencies

Some key differences between reinforcement-based OGC and metacontingencies are summarized in Table 1. In OGCs, the consequence for a group member engaging in the response targeted for operant selection is a putative reinforcer (e.g., Heering & Wilder, 2006). In other words, a consequence produced by the target response that increases the future probability the response will repeat. In a metacontingency, the consequence controlling the IBCs and AP produced by the group is not a reinforcer, but a CC selective for the relation between IBCs and AP. In OGCs, the behaviors of individuals in the group that produce the putative reinforcer can be functionally independent, meaning the behavior is not required to exert antecedent or consequent control over the same target behavior of other group members for those members to access the reinforcer. In contrast, in a metacontingency members of the group produce the AP that results in access to the CC only if group members engage in responses that exert antecedent or consequent stimulus control over the behavior of other group members (e.g., Vichi et al., 2009). In an OGC, when emission of the target behavior is required of more than one group member, that target behavior is the same (e.g., Barrish et al., 1969). In contrast, in a metacontingency members may sometimes need to emit different behaviors in IBCs to produce the target AP on which the CC is contingent (e.g., Glenn, 2004). In an OGC, the unit of analysis is an operant response class, and the unit of measurement is a dimensional quantity or dimensionless quantity (Johnston & Pennypacker, 1993) of the behavior of one group member, a subset of group members, or everyone in the group. But in a metacontingency, the unit of analysis is a culturant (Baia & Sampaio, 2019) and the unit of measurement is an aspect of the IBCs producing the AP (e.g., Carvalho et al., 2017) or an aspect of the AP (e.g., Vichi et al., 2009). The reinforcing effects of OGCs are dimension-altering (e.g., increase a dimensional or dimensionless quantity of the target response). However, the mechanisms by which CC affect IBCs and AP production are not well-understood. If CCs can differentially select variations on an AP, and variability in ICBs is required to produce variations on the AP, then it is reasonable to speculate that differential selection of AP by CC have function-altering, dimension-altering, and/or value-altering effects on interlocked behaviors and their controlling operant stimuli. In OGC, group members need not coordinate their behaviors to produce the reinforcer. However, by definition group members must engage in coordinated IBCs to produce the AP on which the CC is contingent. In other words, it seems metacontingencies require that coordination and cooperation occur in order for the group to produce the CC. However, although such behaviors might occur in OGCs, such contingencies may not be arranged in a manner that makes coordination and cooperation necessary in order for group members to produce the programmed consequence. Lastly, the level of selection in an OGC is operant, but the level of selection by metacontingencies is cultural (e.g., Glenn, 2004).

Table 1.

Summary of differences between operant group contingencies and metacontingencies

	Operant Group Contingencies	Metacontingencies
Consequence	Everyone in the group can earn the same reinforcer (Cooper et al., 2020)	Everyone in the group can earn the same cultural consequence (Glenn et al., 2016)
Contingency	Reinforcer is delivered contingent on behavior (Cooper et al., 2020)	A cultural consequence is contingent on an aggregate product, with IBCs often free to vary (Glenn et al., 2016)
Functional interdependence	The behaviors of individuals in the group can be functionally independent	The behaviors of individuals in the group are functionally interdependent in relation to the aggregate product; the behaviors of each group member serve as antecedents and/or consequences for the behaviors of other group members (e.g., Malott & Glenn, 2006)
Heterogeneity of target behavior	Members must emit the same target behavior to produce the reinforcer (Cooper et al., 2020)	Members may sometimes need to emit different behaviors in IBCs to produce target APs on which the cultural consequence is contingent (e.g., Glenn, 2004)
Unit of analysis	Operant response class (Johnston & Pennypacker, 1993)	Culturant (Baia & Sampaio, 2019)
Unit of measurement	A measure of the behavior of one group member, a subset of group members, or everyone in the group	A measure of the aggregate product (e.g., Vichi et al., 2009)
Effects	Dimension-altering through reinforcement or punishment	Potential function-altering, dimension altering, and/or value-altering effects on interlocking behaviors and related controlling operant stimuli
Coordination	Group members need not coordinate	Coordination is required (e.g., Vichi et al., 2009)
Level of selection	Operant	Cultural (e.g., Glenn et al., 2016)

These potential differences between OGC and metacontingencies highlight the complexity of designing group level interventions with specificity and precision in organizations, and may provide a springboard for advancing research and development of a more versatile and precise technology of group contingencies.

The remainder of this article provides multiple examples to illustrate how organizations might arrange each type of group-oriented contingency targeting QD-KPIs to increase ASDQ.

Hypothetical Examples Intervening on KPI with OGC and Metacontingencies

The extent to which the ADSQ framework proposed by Silbaugh and El Fattal (2021) has been applied in ABA autism therapy organizations is unknown. However, ABA organizations that utilize an ASDQ framework adopt professional and consumer standards and design and improve processes throughout their organization in service of maximizing ASDQ (Silbaugh & El Fattal, 2021). Through conceptualizing the organization as a total performance system, leaders can list and define processes within the system and identify a target process to design or intervene on in service of maximizing ASDQ. They can set professional and consumer standards applicable to that target behavioral process (e.g., the relevant BHCOE standard), and determine KPI values for those standards (e.g., reduce the average time to satisfactory resolution of ethical issues to 2 weeks). Changes in KPI values can be evaluated through progress monitoring (e.g., data collected on the time it takes for an ethical issue to resolve satisfactorily) and intervened on empirically (e.g., using single-subject design methodology or other methods). In doing so, managers can determine whether the change initiatives implemented with the evidence-based practice of ABA at the organizational level are effective. Change initiatives can include antecedent, consequent, or a combination of antecedent or consequence manipulations to improve the QD-KPI by changing the behavior of performers within the behavioral process.

Ethics Committee Example

If an established ethics committee’s underperformance (i.e., what may also be referred to in behavioral systems analysis as a critical business issue; Diener et al., 2009) warrants intervention to accomplish its goals, contingencies can be manipulated to improve the process. Contingency manipulations can consist of (1) operant contingencies individualized to the specific behavior(s) of one member of the committee; (2) OGC applied to the same behavior expected of every member on the committee; (3) a metacontingency arranged for an AP produced by the committee regardless of the IBCs that produce the target aggregate product1; or (4) a combination of these.

Once an ethical concern is submitted to the committee, one step in the process might include all the committee members convening to discuss and vote on whether the concern warrants a full case review or expedited review. In this hypothetical example behavior analysts can consider the existence of the ethics committee itself a professional standard within the organization. The KPI is the time it takes to convene and vote on full or expedited review, and the target value set for this KPI is 2 business days following submission of the concern to the committee. Because the same behavior (i.e., voting for full or expedited review) is required of all committee members, a group-oriented contingency may be feasible and appropriate. An independent OGC doesn’t seem warranted because the vote requires cooperation (i.e., everyone must discuss the concern prior to voting to conduct an expedited or full review) and independent OGCs don’t require functional interdependence across group members. If an assessment determines that one committee member consistently delays these meetings, an operant contingency (reinforcer delivered only to that member) or a variation of the dependent OGC could be applied to their on-time meeting attendance and voting, the latter contingency potentially benefiting from peer pressure. However, because voting requires cooperation, and the latency between the submission of the ethics concern and the group’s decision to conduct a full or expedited review is a product of their discussion and voting, a group-oriented intervention in which reinforcers are provided to all committee members for a final vote within 2 days of the ethical concern submission is an option. Because committee members must cooperate to discuss the concern and cast their vote in the meetings, their behaviors are interlocked, meaning that their behaviors are functionally interdependent. In interdependent OGCs, functional interdependence across group members’ behavior is not required to produce the reinforcer, so reinforcement delivered to all committee members contingent on meeting the target KPI value is more precisely characterized as a metacontingency. If a representative of the group records the latency between the submission of the concern and the decision to conduct a full or expedited review is documented in a data collection system (e.g., in a shared Excel file), latencies that meet the target KPI value could be differentially selected with cultural consequences delivered by a manager in the organization tasked with oversight of the ethics committee. Figure 2 is a diagram illustrating such a metacontingency-based intervention. The manager could deliver a CC to the entire committee for discussing and voting within 2 days, collecting the data, and submitting the permanent product of the data (i.e., the excel file). The manager would withhold the CC if their data showed vote latency longer than 2 days.

Fig. 2.

Theoretical metacontingency implemented by clinical or operations manager to help an ethics committee meet a target key performance indicator

Discrete Trial Teaching Quality Example

Leaders in ABA autism therapy organization can also apply OGC to improve the quality of teaching during direct treatment sessions using instruments that measure treatment quality. However, research on such instruments is sorely lacking. One instrument, the York Measure of Quality of Intensive Behavioral Intervention (YMQI) is an empirically supported performer level questionnaire used to assess the quality of early intensive behavioral intervention (e.g., Perry et al., 2008). Research has shown that supervision facilitated by the YMQI outperformed supervision uniformed by the YMQI (Langh et al., 2022) and quality scores from the YMQI are predictive of clinically significant positive outcomes of early intensive behavioral intervention services for preschool children with autism (Langh et al., 2021). The YMQI consists of 21 items grouped into nine categories, and YMQI ratings are determined using two 5-min video sequences (randomly selected from 20-min videos) of the therapist during therapy sessions. Each item is scored on a 5-point Likert scale. A summary score of less than 2.1 on the YMQI represents poor quality and a score of 2.5–3 excellent quality (Perry et al., 2008). Unfortunately, the YMQI is not commercially available. However, organizations that provide early intensive behavioral interventions to children with autism and adopt an ASDQ framework could develop in-house tools like the YMQI to arrange contingencies in process improvement to control clinical quality at the performer level.

For example, an ABA autism therapy organization could develop an in-house direct treatment quality control instrument based on guidelines for a progressive “high-quality” approach to discrete trial teaching described by Leaf et al. (2016). A clinical manger could list and operationally define each of the guidelines (e.g., “select trial targets and placement of stimuli based upon the learner” and “use natural language instructions when possible”) in a word document, assign a Likert scale (e.g., 1–5) and points (e.g., up to 5 points each, for a total of 40 possible points) to each guideline, and ask clinical supervisors to administer this improvised discrete trial teaching quality control instrument during brief session observations repeatedly throughout the week during their RBT’s sessions. After the instrument has been validated in-house, this process could be implemented as part of routine clinical quality control practices and a QD-KPI assigned. For example, the manager could collect baseline data on the weekly average score per week (i.e., the QD-KPI) based on data collected across all RBTs by clinical BCBA supervisors, then set an average weekly target of “35 points” and arrange OGC in a behavioral process improvement project focused on increasing the quality of discrete trial training. No single employee can engage in behaviors that result in the average weekly score, thus the data representing the QD-KPI is the AP of this quality control process. Thus, the target AP (i.e., a weekly score of 35 or higher) could be produced by (1) each clinical BCBA supervisor using an independent group contingency to reinforce one or more specific RBT behaviors aligned with the guidelines; (2) the clinical manager using a dependent or interdependent group contingency targeting RBT behaviors aligned with the guidelines; or (3) the clinical manager applying a metacontingency in the form of delivering a reinforcer to all clinical supervisors and RBTs contingent on the average weekly score meeting the QD-KPI target. Figure 3 is a diagram illustrating such a metacontingency-based intervention.

Fig. 3.

Theoretical metacontingency implemented by clinical or operations manager to help a clinical team meet a target key performance indicator for discrete trial training clinical quality

Standard Pretreatment Assessment Process Example

When a child with autism is referred to insurance funded ABA autism therapy, the agency verifies their insurance, obtains authorization from the insurance company to complete an initial assessment, and submits a proposed treatment plan based on the results of the assessment. ABA autism therapy providers typically conduct an in-house standard battery of behavioral and developmental, and other assessments during the assessment process. The complexity of this process suggests considerable variability in how it is carried out across ABA autism therapy organizations.

For example, ABA autism therapy providers may aim to complete the assessment battery within a specified span of time, such as 5 cumulative hours (not including treatment plan writing), if it benefits the receiving system (e.g., a treatment plan review process within the clinical quality department). Thus, in this example having a battery of valid and appropriate assessments is an operational standard, the KPI is the time it takes to complete the assessment battery, and the target KPI value is set at 5 cumulative hours.

The assessment battery usually involves a clinical team comprised of at least one supervising BCBA, and others such as an RBT or perhaps a scheduler and/or clinical director (as well as caregivers and the child). Within that context, no single team member (i.e., employee) can complete the assessment battery. Two or more team members coordinate their behavior to get it done. Thus, managers have a variety of contingencies they could apply differentially to select KPI values at 5 hr or less. For example, an operant contingency between a monetary bonus and the supervising clinical BCBA supervisor completing the assessment battery within 5 hr. A potential disadvantage of this contingency is the clinical BCBA supervisor may not be able to meet it consistently if other members of the clinical team fall behind. Other clinical team members might even resent the idea of the clinical BCBA supervisor getting a bonus for their performance when no other team members do. As an alternative, group-oriented contingencies could be applied. Independent and interdependent OGCs are probably irrelevant because team members engage in different behaviors that ultimately lead to the completion of the assessment battery, and in OGCs the reinforcer is contingent on a single target behavior. If one specific team member tends to slow the process down, a dependent OGC could be applied in which every member of the clinical team receives a monetary bonus if the team member (i.e., variation of the dependent OGC) exhibits the target behavior needed for the team to complete the assessment battery on-time. However, under this arrangement the clinical team could complete the assessment battery longer than 5 hr and still obtain the reinforcer so long as the target team member exhibits the target behavior. Finally, a metacontingency could be applied in which a CC is delivered to the whole team for submitting a completed test battery to the manager with a duration of 5 or fewer hr. In this contingency, the time to complete assessment batteries is free to vary, the bonus is contingent on an AP (i.e., assessment battery completed within 5 hr) rather than any specific behavior of any particular clinical team members, and team members are free to vary IBCs as needed to get the job done. Clinical managers could experimentally evaluate the effects of the metacontingency alternating between conditions in a withdrawal design in which the monetary bonus is delivered contingent on different KPI value thresholds. Figure 4 is a diagram illustrating such a metacontingency-based intervention.

Fig. 4.

Theoretical metacontingency implemented by clinical or operations manager to help an assessment team meet a target key performance indicator

Conclusion

ABA autism therapy organizations principally exist to produce large, lasting, meaningful improvements in the quality of life for individuals with autism and their families through behavior change accomplished with ABA technology. It is reasonable to assume that better outcomes are associated with higher quality services. What exactly constitutes “high-quality services” remains elusive in the behavior analytic literature due to a lack of conceptual work on the concept of service delivery quality in the industry. Silbaugh and El Fattal (2021) took an initial step to expose that gap in the literature and start to address it by defining service quality at the organizational level and suggesting ABA autism therapy organizations could use an ASDQ framework to empirically determine the quality of services. The ASDQ framework emphasizes a focus on understanding and arranging contingencies within organizations that control clinical and operational practices and processes that affect ASDQ. Recent trends in the ABA industry serving individuals with autism suggest an increasingly urgent need for advances in understanding how ABA autism therapy organizations can use contingencies to improve behavioral systems and processes underlying ASDQ.

The current article extends emerging research on ASDQ by describing how a variety of contingencies at the operant and cultural level of selection could be used in an experimental approach to evaluating the effects of organizational change initiatives (i.e., attempts at process improvement) on KPIs to promote ASDQ in ABA autism therapy organizations using empirically supported tools and concepts from organizational behavior management. This approach to change management is compatible, and therefore could be integrated, with a culturo-behavioral systems model to promote the success of such efforts (Malott, 2022).

The current article also extends the conceptual analysis of OGC and metacontingencies by arguing that they should be classified as group-oriented contingencies operating at the level of operant selection (i.e., OGC) and cultural selection (i.e., metacontingencies). This approach to classifying group-level interventions may draw increased attention to metacontingencies as “group contingencies” and the fact that research is needed to understand the relative practical utility of group-level interventions in organizations based on operant selection, cultural selection, or both. For many years, the field of ABA has used the term “group contingencies” only to refer to operant group contingencies. The term “group-oriented contingencies” enables one to talk about both operant-level and cultural-level interventions on the behavior of group members while preserving the differences in the level of selection on which each type of intervention is based.

The current conceptual analysis identified plausible differences between OGC and metacontingencies (see Table 1) that have not been discussed or addressed explicitly in recent, broad scope reviews of the experimental metacontingency literature (Albuquerque et al., 2021; Cihon et al., 2020; Zilio, 2019). In doing so, the article reveals the need for more research to understand how to apply group-oriented contingencies in organizational settings, and what their different effects may be with respect to individual employee performances and the products of behavioral processes. Research is needed to develop assessments managers can use systematically to identify potential existing group contingencies, or the absence thereof, and based on the context select an appropriate operant contingency, OGC, or metacontingencies. Based on the current conceptualization of group-oriented contingencies, such an assessment could provide a foundation for new research lines aimed at understanding the contingencies controlling service quality at the operant and cultural level and developing contingency-based culturo-behavioral interventions that promote cultural practices conducive to high ASDQ.

Whether experiments on group contingencies have controlled for the variables listed in Table 1 is unknown, suggesting that it is possible some group contingencies in published studies represent methodological outliers, which could perhaps be reconceptualized as metacontingencies. For example, to the extent that consequences were dependent on an AP, or group members had to interact with each other to produce the AP, new alternative interpretations of findings in group contingency studies may be possible. As such, future research could improve our understanding of differences between effects of OGC and metacontingencies by screening studies to examine whether different results were obtained by studies with high relative to low conformance to characteristics of OGCs and metacontingencies. Likewise, future research could adopt the current conceptualization of group-oriented contingencies to experimentally compare the effects of OGCs and metacontingencies on KPIs in applied settings where processes are found to adversely affect ASDQ. Future research could also experimentally evaluate the relative effects of the Good Behavior Game with and without IBC requirements on disruptive student classroom behavior.

A limitation of the current article is that examples of the application of operant, OGC, and metacontingencies to behavioral processes were hypothetical. Research in ABA autism therapy organizations is needed to test some of the assumptions and experimentally evaluate the efficacy of such interventions. Research is needed to better understand methods compatible with organizational settings that leaders and managers can use to experimentally evaluate the effects group-oriented contingencies for behavioral process improvement on QD-KPIs. In addition, this article has only addressed the role of contingency-based interventions on behavioral process improvement to increase ASDQ. However, future research could also consider the utility of antecedent manipulations such as “nudging” (Tagliabue, 2022) for process improvements that increase ASDQ.

Data Availability

None to report.

Declarations

Informed Consent

Not applicable.

Conflicts of Interest

None to report. No businesses have provided monetary support, and the authors are not affiliated with the organizations identified in this paper.

Ethics Approval

Not applicable.

Consent to Participate

Not applicable.

Consent for Publication

The author consents to publication of the current manuscript.