The detrimental effects of extrinsic reinforcement on "Intrinsic motivation"

Extrinsic consequences have been criticized on the grounds that they decrease intrinsic motivation or internally initiated behavior. Two popular rationales for this criticism, Lepper's overjustification hypothesis (1981) and Deci's motivational theory (Deci & Ryan, 1985), are reviewed and the criticism is then redefined behaviorally. "Intrinsically controlled" behavior is defined as behavior maintained by response-produced reinforcers, and the question concerning extrinsic consequences is thus restated as follows: When behavior is maintained by response-produced stimuli, does extrinsic reinforcement decrease the reinforcing value of those stimuli? The empirical support for this detrimental effect is summarized briefly, and several possible explanations for the phenomenon are offered. Research results that reflect on the effect's generality and social significance are discussed next, with the conclusion that the effect is transient and not likely to occur at all if extrinsic rewards are reinforcing, noncompetitive, based on reasonable performance standards, and delivered repetitively.     

Evaluations of the Overjustification Effect

The utility of reinforcement-based procedures has been well established in the behavior analysis literature and is commonly used in educational settings. However, the overjustification effect is one commonly cited criticism of programs that use tangible items as reinforcers. In the current studies, we evaluated the effects of tangible rewards contingent on engagement with leisure activities. We further evaluated the amount of engagement after the reward condition was discontinued. In Studies 1 and 2, tangible rewards were delivered for engaging with a preferred activity when only one leisure activity or several activities were available, respectively. Study 1 results did not support the overjustification hypothesis. Study 2 participants showed a decrease in activity engagement following the reward condition. It was not clear whether this was a result of extended exposure or overjustification. Thus, in Study 3, we extended exposure to preferred activities in one context without rewards. Study 3 results suggested that extended exposure to a preferred activity had effects that look similar to the overjustification effect but are probably best described as satiation.     

A quantitative review of overjustification effects in persons with intellectual and developmental disabilities

The overjustification hypothesis suggests that extrinsic rewards undermine intrinsic motivation. Extrinsic rewards are common in strengthening behavior in persons with intellectual and developmental disabilities; we examined overjustification effects in this context. A literature search yielded 65 data sets permitting comparison of responding during an initial no‐reinforcement phase to a subsequent no‐reinforcement phase, separated by a reinforcement phase. We used effect sizes to compare response levels in these two no‐reinforcement phases. Overall, the mean effect size did not differ from zero; levels in the second no‐reinforcement phase were equally likely to be higher or lower than in the first. However, in contrast to the overjustification hypothesis, levels were higher in the second no‐reinforcement phase when comparing the single no‐reinforcement sessions immediately before and after reinforcement. Outcomes consistent with the overjustification hypothesis were somewhat more likely when the target behavior occurred at relatively higher levels prior to reinforcement.     

Establishing praise as a conditioned reinforcer: Pairing with one versus multiple reinforcers

Praise and other forms of attention may not function as reinforcers for the behavior of children with autism. Previous research demonstrated that contingently pairing praise with reinforcers (response stimulus) can establish praise as a conditioned reinforcer. We evaluated a procedure for establishing praise as a generalized conditioned reinforcer (pair praise with four reinforcers) and compared it with a procedure to establish praise as a conditioned reinforcer (pair praise with one reinforcer). We compared the two conditions in a reversal design with “praise” and “no programmed consequence” conditions rapidly alternated after pairing phases. With three out of five participants with autism, responding in the “pair‐with‐four” condition showed higher levels of responding or more increasing trends compared with that in “pair‐with‐one” condition. We also replicated previous contingent pairing research with two participants in a multiple baseline across participants design. A limitation was the absence of manipulating motivating operations in assessing the efficacy of praise as a generalized conditioned reinforcer. Nevertheless, these data contribute to the research on establishing praise as a reinforcer, and we recommend directions for future research.     

Extrinsic rewards undermine altruistic tendencies in 20-month-olds

The current study investigated the influence of rewards on very young children's helping behavior. After 20-month-old infants received a material reward during a treatment phase, they subsequently were less likely to engage in further helping during a test phase as compared with infants who had previously received social praise or no reward at all. This so-called overjustification effect suggests that even the earliest helping behaviors of young children are intrinsically motivated and that socialization practices involving extrinsic rewards can undermine this tendency.     

An evaluation of the properties of attention as reinforcement for destructive and appropriate behavior

The analogue functional analysis described by Iwata, Dorsey, Slifer, Bauman, and Richman (1982/1994) identifies broad classes of variables (e.g., positive reinforcement) that maintain destructive behavior (Fisher, Ninness, Piazza, & Owen-DeSchryver, 1996). However, it is likely that some types of stimuli may be more effective reinforcers than others. In the current investigation, we identified 2 participants whose destructive behavior was maintained by attention. We used concurrent schedules of reinforcement to evaluate how different types of attention affected both destructive and appropriate behavior. We showed that for 1 participant praise was not an effective reinforcer when verbal reprimands were available; however, praise was an effective reinforcer when verbal reprimands were unavailable. For the 2nd participant, we identified a type of attention that effectively competed with verbal reprimands as reinforcement. We then used the information obtained from the assessments to develop effective treatments to reduce destructive behavior and increase an alternative communicative response.     

The differential effects of tangible rewards and praise on intrinsic motivation: A comparison of cognitive evaluation theory and operant theory

Substantial research indicates that tangible rewards, such as money, prizes, and tokens, decrease response rates by undermining intrinsic motivation. In contrast, praise appears to increase response rates by enhancing intrinsic motivation. Based on their interpretation of available evidence, many social-cognitive researchers warn not to use tangible rewards in applied settings and to use praise instead. Furthermore, they suggest that the differential effects of the two types of rewards on intrinsic motivation cannot be explained using principles of operant psychology. Cognitive evaluation theory provides one of the most recent and widely cited social-cognitive explanations for the different effects of the two types of rewards on intrinsic motivation (Deci & Ryan, 1985). However, a review of existing research found little support for the explanations based on this theory and revealed three potential confounding effects: (a) temporal contiguity, (b) the number of reward administrations, and (c) discriminative stimuli associated with reward availability. These three confounding factors provide explanations for the effects of tangible rewards and praise on intrinsic motivation that are consistent with principles of operant psychology.     

Examining stimuli paired with alternative reinforcement to mitigate resurgence in children diagnosed with autism spectrum disorder and pigeons

In two laboratory experiments, we examined whether stimuli paired with alternative reinforcers could mitigate resurgence of a previously reinforced target response with pigeons (Experiment 1) and children diagnosed with Autism Spectrum Disorder (Experiment 2). In Phase 1, we arranged food reinforcement according to a variable-ratio schedule for engaging in a target response. In Phase 2, we arranged extinction for target responding and differentially reinforced alternative responding according to a fixed-ratio schedule, with every alternative-reinforcer delivery paired with a change in keylight color (Experiment 1) or automated verbal (praise) statement (Experiment 2). In Phase 3, we assessed resurgence during extinction of target and alternative responding in the presence versus absence of continued presentation of the paired stimulus. Despite variation across sessions, resurgence on average was lower when continuing to present the paired stimuli in all pigeons and children while maintenance of alternative responding did not differ between assessments. These findings indicate that stimuli paired with alternative reinforcement can modestly decrease resurgence, but further examination of their efficacy and a better understanding of the underlying processes are necessary before they can be recommended for clinical use in reducing resurgence of clinically relevant problem behavior.     

Control by reinforcers across time and space: A review of recent choice research

Reinforcers affect behavior. A fundamental assumption has been that reinforcers strengthen the behavior they follow, and that this strengthening may be context‐specific (stimulus control). Less frequently discussed, but just as evident, is the observation that reinforcers have discriminative properties that also guide behavior. We review findings from recent research that approaches choice using nontraditional procedures, with a particular focus on how choice is affected by reinforcers, by time since reinforcers, and by recent sequences of reinforcers. We also discuss how conclusions about these results are impacted by the choice of measurement level and display. Clearly, reinforcers as traditionally considered are conditionally phylogenetically important to animals. However, their effects on behavior may be solely discriminative, and contingent reinforcers may not strengthen behavior. Rather, phylogenetically important stimuli constitute a part of a correlated compound stimulus context consisting of stimuli arising from the organism, from behavior, and from physiologically detected environmental stimuli. Thus, the three‐term contingency may be seen, along with organismic state, as a correlation of stimuli. We suggest that organisms may be seen as natural stimulus‐correlation detectors so that behavioral change affects the overall correlation and directs the organism toward currently appetitive goals and away from potential aversive goals. As a general conclusion, both historical and recent choice research supports the idea that stimulus control, not reinforcer control, may be fundamental.     

Competence and the maintenance of behavior

The overjustification effect is manifested by decreased interest in an attractive activity following removal of a reward made contingent upon participation in the activity. Evidence is reviewed which suggests that rewards conveying information that the person who is performing the activity is competent cause the behavior to become insulated from the overjustification effect. This maintenance of behavior resulting from competence information is also shown to be a powerful factor in increasing the transfer and posttherapy persistence of behavior in both behavioral and traditional therapies.I wish to thank Pamela Sheridan, Tom Eddy, Larrie Hutton, and Jon Aronoff for their help during the course of my research on this topic. I am also grateful to Sylvia Snavely for sharing her helpful ideas with me.     

The analysis and treatment of vocal stereotypy in a child with autism

This study examined procedures for the assessment and treatment of automatically reinforced vocal stereotypy of a 6‐year‐old girl with autism. Stimulus assessments were conducted to identify toys that were correlated with higher rates of vocal stereotypy and toys that were not. A concurrent operants assessment identified preferred stimuli (toys that produced auditory stimulation), which were then used as reinforcers for the non‐occurrence of vocal stereotypy. A reversal design was used to compare the effects of a fixed time schedule of reinforcement (FT 1‐min) to differential reinforcement for the non‐occurrence of behavior (DRO) to reduce vocal stereotypy. Implementation of the FT schedule revealed no effect, whereas the DRO schedule led to a reduction in the target behavior during treatment sessions and across the school day. This study adds to the body of literature supporting the identification of matched stimuli to reduce non‐socially mediated problem behavior. Copyright © 2005 John Wiley & Sons, Ltd.     

The effects of verbal reinforcement of intrinsic motivation for sex-linked tasks

Recent studies have demonstrated that positive verbal feedback enhances males' intrinsic motivation for a task, while decreasing that of females. This result has been attributed to differential socialization forces which make the controlling aspects of rewards more salient for females as opposed to males, for whom the informational aspects are prepotent. A weakness in this conclusion stems from the use of a masculine sex-linked task in prior experiments, producing a result which may be due to gender congruence/incongruence rather than to socialization differences. Two laboratory experiments were performed to examine the effects of verbal praise on intrinsic motivation, controlling for sex-typing of the experimental activity. Experiment I found that verbal praise increased females' and males' intrinsic motivation on both a feminine and masculine task, and intrinsic motivation was higher for sex-appropriate than sex-inappropriate tasks. The sexes did not show a differential pattern of response to praise. In Experiment 2, verbal praise also enhanced sex-role traditional females' intrinsic motivation on the same activity used in prior research. These results suggest that socialization content has changed sufficiently to make concerns about competence and self-determination equally salient for females and males.     

Behavior Analysts' Use of and Beliefs in Treatments for People with Autism: A 5‐Year Follow‐up

As autism prevalence rises, parents of children with autism rely upon professionals to recommend the most effective treatment for their children. Historically, these professionals have chosen a range of treatments from empirically supported to nonempirically supported treatments. This study replicated and extended that work by surveying Board Certified Behavior Analysts to determine what treatments they promoted and used. Results showed that although most respondents used evidence‐based, behavioral procedures, many did not. Influential stimuli most likely to contribute to behavior analysts' choices of treatments were beliefs in effectiveness, ease of implementation, and persuasion by others. Self‐reported reinforcers for treatment use included perceived observed success, praise from others, and financial gain. Behavior analysts continue to use a variety of nonscientifically supported treatments, including those that have been deemed ineffective and harmful to people with autism. Future training of Board Certified Behavior Analysts must stress adherence to the ethical code for certified behavior analysts, adherence to data‐based decision‐making, and strict fealty to the scientific process in which all behavior analysts have been trained. Copyright © 2016 John Wiley & Sons, Ltd.     

Operant discrimination training to establish praise as a reinforcer

Social deficits are a hallmark feature of autism spectrum disorder (ASD). Consequently, socially mediated consequences (i.e., praise) may not function as reinforcers for individuals with ASD. Given the frequency in which praise is used as a programmed consequence in empirical research and in clinical practice, it is necessary to explore procedures that could condition praise to function as reinforcers. Operant discrimination training has emerged as a potential procedure to condition stimuli to acquire reinforcing properties. This study aimed to utilize the operant discrimination training procedure to condition a praise statement to function as a reinforcer. Four children diagnosed with ASD participated in the study. A nonconcurrent multiple baseline design across participants was used to evaluate this emerging pairing procedure. Findings suggest that immediate effects were observed across all participants, but responding did not maintain, suggesting that reinforcing value of the praise statement diminished.     

Extending the revelation effect to faces: Haven't we met before?

The revelation effect is an episodic memory phenomenon where participants are more likely to report that they recognise an item when it is judged after an interpolated task than when it is not. Although this effect is very robust, nearly all of the extant research has used verbal or readily verbalisable stimuli. The present two experiments examined whether a revelation effect could be produced with non‐verbal stimuli such as faces. A revelation effect was found in both experiments, for both targets and lures, using faces as stimuli. The findings are integrated into the prevailing empirical frameworks for the revelation effect and face recognition memory.     

Increasing Teachers' Use of a 1:1 Praise‐to‐Behavior Correction Ratio to Decrease Student Disruption in General Education Classrooms

Contingent, behavior‐specific praise is universally recommended as an effective tool to increase students’ academic achievement and proscocial behavior. Despite this recommendation, little research has examined the effects of training teachers to increase their praise‐to‐behavior correction ratio in classroom settings. This study evaluated the effects of training teachers to provide a 1:1 ratio of praise‐to‐behavior correction to decrease student disruption in three general‐education classrooms. Three urban middle‐school general‐teachers who exhibited very low rates of praise participated in the study. Teachers received training, including modeling and performance feedback, to achieve a 1:1 ratio of praise‐to‐behavior correction administered within a multiple‐baseline across participants design. Results demonstrated that (a) the teachers were able to achieve and maintain a 1:1 praise‐to‐behavior correction ratio; (b) a reduction in student disruption coincided with teachers’ use of a 1:1 praise‐to‐behavior correction ration; and (c) two teachers evidenced moderate levels of generalization to classrooms where no training took place. Implications for practitioners and future research are discussed. Copyright © 2011 John Wiley & Sons, Ltd.     

Delivering alternative reinforcement in a distinct context reduces its counter‐therapeutic effects on relapse

Delivery of alternative reinforcers in the presence of stimuli previously associated with reinforcement for target behavior increases the susceptibility of target behavior to relapse. To explore contingencies that might mitigate this counter‐therapeutic effect, we trained pigeons on a procedure that entailed extinction of previously reinforced target‐key pecking, access to a distinct stimulus context contingently on refraining from target behavior (differential‐reinforcement‐of‐other‐behavior; DRO), and reinforcement of alternative‐key pecks (differential‐reinforcement of alternative behavior; DRA) in that context. This DRO‐DRA treatment was compared with standard DRA in successive conditions, counterbalanced across pigeons. Target behavior extinguished more rapidly in the Standard‐DRA condition. When alternative reinforcement was discontinued, however, there was less resurgence after DRO‐DRA than after Standard DRA. In a third condition, the DRO contingency was suspended so that the former DRA stimuli were not presented (DRO‐NAC), and resurgence was greater than in the Standard‐DRA and DRO‐DRA conditions. Reinstatement produced by response‐independent reinforcers was small and similar across conditions. Subsequent reacquisition of target‐key pecking under baseline reinforcement conditions was faster following DRO‐NAC than Standard‐DRA or DRO‐DRA. These findings suggest that DRO‐DRA might serve as a useful method in clinical settings for reducing problem behavior while minimizing the threat of posttreatment relapse.     

Training intraverbal naming to establish equivalence class performances

The purpose of this three‐experiment study was to evaluate whether performance consistent with the formation of equivalence classes could be established after training adults to tact and intraverbally relate the names of visual stimuli. Fourteen participants were exposed to tact training, listener testing, and intraverbal training (A'B’ and B'C’) prior to matching‐to‐sample (MTS) and intraverbal posttests presented in different sequences across experiments. All participants demonstrated emergent MTS and intraverbal relations consistent with equivalence class formation. More importantly, all participants emitted experimentally defined or self‐generated tacts or intraverbally named the correct sample‐comparison pairs at some point during posttests. These results are consistent with the intraverbal naming account (Horne & Lowe, 1996) in that participants who passed novel relations MTS tests also demonstrated emergence of corresponding intraverbal relations. However, verbal reports and latency data suggest that participants did not necessarily have to use intraverbal naming as a problem solving strategy continuously throughout MTS posttests. These results extended previous research by showing that verbal behavior training of baseline relations (A'B’ and B'C’) is sufficient to establish novel conditional relations consistent with equivalence class formation.     

Further evaluation of the use of preference categories to identify novel reinforcers: A systematic replication

Preference assessments are frequently used to identify reinforcers for individuals with autism spectrum disorder and intellectual and developmental disabilities. Although well‐established procedures have been shown to identify preferred stimuli for individual items that can be used in skill acquisition and behavior reduction programs, little research has been conducted on identifying categories of preferred items. In this study, paired‐stimulus preference assessments were conducted with 4 individuals with autism spectrum disorder. Researchers classified the edible stimuli as belonging to 1 of 4 categories: chocolate (e.g., chocolate chips and M&M's^®), salty/crunchy (e.g., chips and crackers), gummy (e.g., Swedish Fish^® and Starburst^®), or fruit/vegetable (e.g., grape and apple). Preference hierarchies were identified for individual stimuli and for categories of stimuli. For all participants, at least 3 of the 4 most preferred items came from the same category. Novel items (i.e., items not included in the preference assessments) identified by the researchers as belonging to the high‐preference category functioned as effective reinforcers during subsequent reinforcer assessments for all participants. This finding suggests that clinicians could identify likely effective edible reinforcers based on an individual's categorical preference without explicit testing.