Token reinforcement: Translational research and ... · for analysis and application. Unlike standard translational research, based on a unidirectional model in which the analysis

TOKEN REINFORCEMENT: TRANSLATIONAL RESEARCH ANDAPPLICATION

TIMOTHY D. HACKENBERG

REED COLLEGE

The present paper provides an integrative review of research on token reinforcement systems,organized in relation to basic behavioral functions and economic variables. This type of func-tional taxonomy provides a useful way to organize the literature, bringing order to a wide rangeof findings across species and settings, and revealing gaps in the research and areas especially ripefor analysis and application. Unlike standard translational research, based on a unidirectionalmodel in which the analysis moves from laboratory to the applied realm, work in the area oftoken systems is best served by a bidirectional interplay between laboratory and applied research,where applied questions inspire research on basic mechanisms. When based on and contributingto an analysis, applied research on token economies can be on the leading edge of theoreticaladvances, helping set the scientific research agenda.Key words: behavioral economics, comparative analysis, review, token reinforcement, token

economy, translational research

Token reinforcement systems are among theoldest and most widely used procedures inapplied behavior analysis. Beginning withgroundbreaking studies by Ayllon and Azrin(1965) with adult patients in a psychiatric wardand by Phillips (1968) with delinquent youthin a group home, numerous studies have docu-mented the therapeutic and educational benefitsof token procedures across a wide range ofsettings and populations (for reviews, see Ayllon& Azrin, 1968; Doll, McLaughlin, & Barretto,2013; Glynn, 1990; Kazdin, 1982; Kazdin &Bootzin, 1972; O’Leary & Drabman, 1971).Unlike other successful technologies in behavioranalysis, however, there has been little substan-tive contact between applied and basic researchwith token reinforcement over the years.

Despite laboratory research on token systemsdating to the 1930s, little is known about howthis body of knowledge relates to token rein-forcement procedures in the applied realm. Anddespite some 50 years of applied work on tokeneconomies, surprisingly little is known aboutthe behavioral mechanisms responsible for theireffectiveness.Laboratory research on token systems has

centered mainly on basic mechanisms(e.g., acquisition and maintenance of behavior,conditioned reinforcement, temporal organiza-tion, behavioral units), whereas applied researchon token systems has centered mainly on prac-tical issues (e.g., generalization and mainte-nance, treatment integrity, staff training).Although token reinforcement systems havegenerally been quite successful in the appliedrealm, it is often unclear what variables areresponsible for their success; the token systemsare rarely based on an understanding of thebasic principles involved. As a result, laboratoryand applied work on token systems have devel-oped largely in parallel, with little cross-fertilization of ideas and concepts.A major aim of this paper is to bring

together what is known about token

Manuscript preparation was aided by grant R01DA02617 from the National Institute on Drug Abuse. Iam indebted to John Borrero, Jason Bourret, Mary-KateCarey, and Francesca degli Espinosa for helpful commentson earlier versions of the paper.Address correspondence to: Timothy D. Hackenberg,

Psychology Department, Reed College, 3203 SE Wood-stock Blvd., Portland OR 97202-8199. E-mail: [email protected]: 10.1002/jaba.439

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2018, 51, 393–435 NUMBER 2 (SPRING)

© 2018 Society for the Experimental Analysis of Behavior

393

reinforcement systems from laboratory andapplied perspectives. From this, we should gaina better understanding of some of the criticalvariables, how they operate in both settings andin relation to broader economic contexts. I willsuggest promising new lines of research thatemerge from a synthesis of laboratory andapplied work. Although I will endeavor topoint out some of the practical implications ofan integrated approach to basic and appliedwork, the wide-ranging therapeutic and educa-tional benefits of token reinforcement proce-dures are beyond the scope of the paper.Rather, my emphasis is on conceptual integra-tion of basic and applied work based on a func-tional analysis of behavior across species andsettings. To that end, the review of the litera-ture is selective, emphasizing studies and find-ings that illustrate general principles ofbehavior.The paper is organized into five sections. I

begin with an historical overview of token rein-forcement research, tracing the parallel develop-ments in laboratory and applied work that ledto the current disjointed state of affairs. I thenreview the literature on token reinforcementeffects in applied settings, comparing and con-trasting what is known about such effects fromlaboratory research, using a functional taxon-omy similar to that used in a review of labora-tory research on token systems (Hackenberg,2009). I then consider an economic perspectiveon token reinforcement systems, building onthe emerging synthesis of behavioral economicsand behavior analysis. Finally, I consider somepractical implications of the findings, offeringsome recommendations for implementingtoken-based procedures in applied settings inlight of the literature reviewed in the earlier sec-tions. I conclude by encouraging an integratedtranslational approach to research and applica-tion, borrowing concepts and methods fromboth behavior analysis and behavioraleconomics.

HISTORICAL BACKGROUND

Token systems have been in existence in oneform or another since antiquity (O’Leary,1978), but it was not until the 1930s thattoken systems were studied systematically inthe laboratory, and not until the 1960s thattoken systems were implemented in applied set-tings. As the 1960-70s was a period of rapidgrowth in applied behavior analysis, fueled bylaboratory-based research, one might assumethat the application of token systems in theapplied realm grew out of laboratory researchon token reinforcement. In actuality, however,work in the two areas has never been well con-nected. Despite occasional calls for greater inte-gration (Kagel & Winkler, 1972), laboratoryand applied work on token systems have devel-oped largely in parallel.Laboratory research on token reinforcement

dates back to a series of groundbreaking studiesby Wolfe (1936) and Cowles (1937) withchimpanzees as subjects. Each study consistedof a series of interconnected experiments on awide range of topics, including discriminationof tokens with and without exchange value, acomparison of tokens and food reinforcers inthe acquisition and maintenance of behavior,response persistence under immediate anddelayed reinforcement, preference betweentokens associated with different reinforcer mag-nitudes and qualities (e.g., food vs. water, playvs. escape), and social behavior engendered bytoken reinforcement procedures, to nameonly a few.The primary focus of research was on condi-

tioned reinforcement—the circumstances underwhich neutral stimuli come to acquire reinfor-cing functions. Many of the experiments weredesigned to assess whether or to what extenttokens as acquired (conditioned) reinforcerswere functionally equivalent to food (uncondi-tioned) reinforcers. At the same time, theseauthors were clearly aware of some of the dis-tinctive characteristics of token reinforcement

TIMOTHY D. HACKENBERG394

systems, including their durability in the face ofdifferent motivational states—what Skinner(1953) would later term generalized reinforce-ment—and their relevance to economic systemsmore generally. An impressive number ofexperiments from these studies prefigured laterdevelopments in the field of conditioned rein-forcement (Hackenberg, 2009).Despite this promising early research, the

study of token reinforcement lay dormant fortwo decades, until Kelleher’s work in the mid1950s. Also using chimpanzees and poker-chiptokens, Kelleher conducted a number of inter-connected experiments (Kelleher, 1956, 1958),which formed part of the emerging field ofreinforcement schedules (Ferster & Skinner,1957) and was on the leading edge of researchand theory at the time. These schedules permit-ted a broader analysis of conditioned reinforce-ment and of the conditions responsible forestablishing and maintaining tokens as acquiredreinforcers. This emphasis on basic principlesof conditioned reinforcement and temporallyorganized behavior has persisted in more recentlaboratory research, conducted mainly with ratsand pigeons as subjects. And although themethods used to study conditioned reinforce-ment have changed over the years, the labora-tory study of token reinforcement has beenfocused primarily on the elucidation of basicprinciples.Applied research on token reinforcement has

developed along different lines. Within adecade of Kelleher’s groundbreaking work,token systems had been established in hospitals,schools, and group homes. The main source ofinspiration for these studies was not Kelleher’sresearch with chimpanzees, however, butrather, a pioneering study by Ayllon and Azrin(1965), conducted with patients with chronicpsychosis in an institutional setting. The studyused a token reinforcement program to modifya variety of vocational and self-care activities ofthe patients. Tokens were small metal coinsthat could be exchanged at regularly scheduled

exchange periods three times per day for a vari-ety of functionally defined terminal/backupreinforcers (e.g., interaction with staff, preferreddorm rooms, off-ward passes, snacks, cigarettesetc.). On the whole, the program wasimmensely successful, generating and maintain-ing a wide range of positive behavior in mostpatients.While the Ayllon and Azrin (1965) study

was analytic, emphasizing experimental controland isolation of effective variables, the clinicalsetting and the use of socially adaptiveresponses made its applied significance readilyapparent. The Ayllon and Azrin study was fol-lowed by dozens of studies on token systemsapplied to a wide range of settings and subjectpopulations, including delinquent youth in agroup home setting (Phillips, 1968; Phillips,Phillips, Fixsen, & Wolf, 1971), postwar sol-diers in a psychiatric ward (Boren & Colman,1970), and children in classroom settings(Kaufman & O’Leary, 1972; McLaughlin &Malaby, 1972, 1976). While some of theseearly studies provided important insights intobehavioral functions (described in more detailbelow), the rapid proliferation of token systemsin applied settings led to calls for more researchon issues pertaining to practical exigencies ofprogram implementation, staff training, clientresistance, generalization, and so on (Kazdin &Bootzin, 1972). A decade later, Kazdin (1982)reviewed progress in these areas and identifiedseveral additional areas in need of research,including treatment integrity, administrative/organization context, and dissemination. Allwere concerned with large-scale adoption andmaintenance of token systems.Applied research on token systems thus came

to be increasingly governed by practical mattersof implementation and program evaluation(often in conjunction with other treatment pro-grams) rather than by the illumination of basicbehavioral processes. And although one findspassing references to some of the seminal labo-ratory studies in the major reviews of token

395TOKEN REINFORCEMENT

economies, substantive contact between labora-tory and applied work on token systems hasbeen virtually nonexistent.In addition to drifting from its laboratory

roots, the frequency of applied research ontoken economies has decreased markedly overtime. If one were to conduct a keyword searchwith the terms token economy and token rein-forcement, one would find that the annual rateof published papers rose sharply in the late1960s, peaked in the mid-1970s, and has sincedeclined. To be sure, citation analyses are likelyto underestimate the prevalence of token rein-forcement procedures in applied settings. Alongwith timeout procedures, token reinforcementprocedures are used routinely in applied set-tings (e.g., classroom settings, group homes,institutions) as part of the background behaviormanagement system, but are rarely the subjectof research per se. In some ways, token rein-forcement may be a victim of its own successes:The effectiveness of token systems across such arange of settings and subject populations mayhave created a sense that further research isunnecessary.Whatever the reasons, there has been a rela-

tive dearth of published applied research ontoken reinforcement in the past few decades.Despite the decreasing publication trends,token economies hold great promise as a trans-lational research environment, permitting rapidadvancement of theory and practice. As a start-ing point, I describe the characteristics of com-mon token systems, considering the variouspotential functions of tokens in appliedsettings.

FUNCTIONAL TAXONOMY OF TOKENSYSTEMS

Tokens are stimuli paired with other rein-forcers that can be earned, accumulated, andexchanged for other reinforcers (sometimescalled terminal or backup reinforcers) duringscheduled exchange opportunities. Tokens are

often manipulable objects (e.g., poker chips,coins, marbles), but can be nonmanipulable aswell (e.g., stickers, points, checkmarks on alist). Terminal reinforcers can be unconditionedreinforcers (e.g., food, water) or conditionedreinforcers (e.g., money, privileges, play oppor-tunities). Because clear distinctions betweenunconditioned and conditioned are often diffi-cult, I will make no further divisions betweenthem, but instead, adopt a procedural defini-tion: a terminal reinforcer is the final reinforcerin the chain for which the token is exchanged.A token system, or token economy, specifies theinterrelations between token-earning behavior,exchange opportunities, and terminalreinforcers.There are a variety of ways to arrange a

token economy. Contingencies can be arrangedfor individual or aggregate behavior and forhighly specific responses or broad classes ofbehavior; they might involve gain only, lossonly, or some combination; and they might beembedded within larger sets of contingenciesthat govern transition to other levels. Thesecontingencies may themselves be embedded instill-larger sets of social and institutional con-tingencies. Clearly, token systems may exist inmany forms, and it is important to catalogueand study those differences. My emphasis here,however, will not be on the detailed analysis ofparticular token systems, but rather on themore general principles that govern the func-tional operation of all token systems.In the following section, I will outline some

elementary functions of tokens, organizedaround a behavioral taxonomy similar to thatused in a review of laboratory research on tokenreinforcement (Hackenberg, 2009). Thisshould help identify features common to bothsettings and the degree to which laboratory andapplied research on token systems share func-tional (as opposed to merely superficial) charac-teristics. This organizational structure will alsohelp reveal gaps in our understanding of tokensystems and therefore prime topics of further


research. In later sections, I will consider someof the ways in which these basic functions com-bine and interact in a broader token economy,as well as the applied significance of tokenresearch, including the types of token econo-mies in which particular variables are mostlikely to be important.

Reinforcing FunctionsIt is generally assumed that tokens serve as

conditioned reinforcers (i.e., reinforcers whoseeffectiveness has been established through expe-rience with other events already effective asreinforcers). In early experiments by Wolfe(1936), for example, food-paired tokens wereused to establish new behavior. Similarly, Ayl-lon and Azrin (1965) showed that tokens couldbe used to increase the frequency of workaround the hospital ward (e.g., cleaning).Figure 1 shows the number of hours per dayon work-related tasks across conditions, aggre-gated across 44 patients, in which tokens wereor were not contingent on performance(i.e., tokens were delivered independent ofresponding). As mentioned above, the tokenswere metal coins exchangeable at regular timesduring the day for a variety of tangibles(e.g., cigarettes) and activities (e.g., off-wardpasses). Job performance deteriorated whentokens were not contingent on performance,dropping to less than half of the baseline levelin 3 days, and to near-zero levels by the end ofthe 20-day condition, but performancerebounded quickly when the token reinforce-ment contingency was reinstated.These findings were subsequently replicated

and extended by Winkler (1970), also in a psy-chiatric institutional setting. Various patterns ofself-care and independent living were reinforcedwith tokens, the functional control of whichwas demonstrated through control conditionsin which the tokens were delivered independentof behavior. These types of yoked-control pro-cedures in which tokens are presented in a

noncontingent fashion provide an alternativeway to arrange an extinction procedure whileholding roughly constant the overall reinforce-ment rate. That is, to the extent that theresponse-independent token rates approximatethe response-dependent rates, any differencescan be attributed to the token reinforcementcontingency per se, rather than to changes inreinforcement rate that would accompany amore traditional extinction procedure, in whichreinforcers are discontinued (Thompson &Iwata, 2005). These remain important controlprocedures, and their use in applied token sys-tems research would augment the surprisinglysparse literature on reinforcing functions oftokens.Pairing with the terminal reinforcer. Tokens

are thought to derive their value as conditionedreinforcers from repeated pairings with the ter-minal reinforcers for which they are exchanged.

Figure 1. Total hours of daily work by 44 patients ina closed psychiatric ward as a function of token reinforce-ment contingencies (from Ayllon & Azrin, 1965). See textfor additional details. Reprinted with permission.


In experiments by Wolfe (1936) and Cowles(1937), for example, poker chip tokens wereestablished as reinforcers by allowing them tobe exchanged for food. Metal tokens, on theother hand, were not exchangeable for food,and therefore did not acquire reinforcing func-tions. The chimpanzees came to readily workfor and exchange the poker chips, while quicklylosing interest in the metal tokens.In one of the few studies with humans on the

topic, Smith (1972) explored the role of token–reinforcer pairings with kindergarten childrenby comparing the effectiveness of tokens thathad and had not been paired with other rein-forcers during exchange periods. Tokens werefirst presented on a fixed-time (FT) schedule(i.e., response-independently) every 10 s. Twotoken types were presented in strict alternation:One type could be exchanged for a toy trinketeach time it was earned (paired token), whereasthe other type was not exchangeable (unpairedtoken). This condition lasted until 50 tokenshad been delivered, 25 of each type. The twotoken types were then made contingent onpressing one of two buttons in a choice task.Pressing one button produced tokens (of onetype or the other), whereas pressing of the otherdid not. The effects of the previous pairing his-tory were studied across two groups of children.For one group, the tokens were those previouslypaired with toys, whereas for the other groupthe tokens were those previously unpaired.There was a strong preference for the token-producing button for both groups (i.e., priorhistory of pairing with toys did not matter), andno difference in the response rates for the pairedand unpaired groups. In a subsequent 60-sextinction condition, however, clear differencesbetween paired and unpaired tokens emerged:Nearly three times as many responses occurredon the token alternative than on the no-tokenalternative in the paired group, whereas no dif-ferences were seen in the unpaired group.The Smith (1972) study is important in

demonstrating the importance of pairing, using

proper comparisons to unpaired tokens. It alsounderscores the need to study token-reinforcedbehavior under various conditions, as differ-ences not apparent in the reinforcement condi-tion appeared in extinction. The study also hadnotable shortcomings, including brief contactwith a limited range of contingencies and abetween-subject design not well suited to anexperimental analysis. Another limitation was alimited range of schedule values with minimalresponse requirements, which significantlyrestricts the range of response rates over whichan effect might be seen.Motivational operations. Another common

assumption is that token reinforcers (like allconditioned reinforcers) maintain their reinfor-cing functions only insofar as that function ismaintained for the terminal reinforcer withwhich it has been paired. For example, Wolfe(1936) showed that chimpanzees’ choicesbetween working for yellow tokens exchange-able for water and black tokens exchangeablefor peanuts depended on the motivational con-text: the chimpanzees favored the yellow/watertokens when water deprived and the black/foodtokens when food deprived.Similarly, in a classroom setting with chil-

dren and adolescents with intellectual disabil-ities, Moher, Gould, Hegg, and Mahoney(2008) showed that the reinforcing efficacy ofspecific tokens depended on the motivationalcontext. Following a training procedure inwhich one token type was paired with a high-preference edible reinforcer and a second tokentype was paired with a low-preference ediblereinforcer, response rates were compared underthe following four conditions (high-preferenceedible, high-preference token, low-preferenceedible, low-preference token) as a function ofdeprivation and satiation conditions. Responserates were clearly differentiated in the depriva-tion conditions, with higher rates maintainedby the high-preference edibles and tokens thanthe low-preference edibles and tokens.Responding (both for edibles and tokens)


quickly extinguished under satiation conditionswith presession free access to the reinforcers,showing that the efficacy of the tokensdepended on a given deprivation.For four participants in the study, the pair-

ings were established through direct contactwith the contingencies; for a fifth, the pairingswere established through verbal instructions.The practical import of these latter findings isthat with verbal participants, tokens can beestablished as reinforcers via instructions ratherthan via direct experience with the pairing oper-ation. That generally similar results wereobtained across participants in this study sug-gests that verbal instructions may substitute fordirect contact with the contingencies. In theseconditions, instructions serve as a reasonableproxy for direct contact; to the extent that theyhasten acquisition of token-reinforced behavior,instructions may be pragmatically useful. At thesame time, such strong substitutability ofinstructions and contingencies cannot always beassumed. Whereas some research shows anadaptive role of instructions, other findingsshow instructions can impair sensitivity to othercontingencies, resulting in markedly suboptimalbehavior (Fox & Pietras, 2013; Galizio, 1979;Hackenberg & Joker, 1994). Given how little isknown about the role of direct pairings in estab-lishing reinforcing functions of tokens, instruc-tions should be used with caution and in fullrecognition of their complex and varied effects.Token-directed behavior. Tokens are thought

to be initially neutral stimuli that acquire rein-forcing functions by virtue of their repeatedpairings with other reinforcers. Although thismay be true of some tokens (e.g., poker chips,checkmarks), tokens can themselves engenderbehavior directed toward them. In some cases,this behavior takes the form of consummatoryresponses presumably conditioned by thetoken–food pairings; for example, rats(Malagodi, 1967) and chimpanzees (Kelleher,1958) have been observed mouthing marblesand poker chips, respectively.

The origins of other token-directed behaviorare less tied to food reinforcers. For example,the self-stimulatory behavior often observed inpeople with autism can occur in relation totokens. Charlop-Christy and Haymes (1998)showed that the success of a token reinforce-ment program with children with autism in anacademic setting depended, in part, on thedegree to which the tokens evoked self-stimulatory behavior. Standard arbitrary tokens(e.g., stars on a sheet of paper) were comparedwith idiosyncratic tokens, determined individu-ally for each student. These tokens were the so-called “objects of obsession” for each student,stimuli in the presence of which self-stimulatorybehavior was allowed to occur. In general,the tokens that occasioned self-stimulatorybehavior produced higher levels of correctresponding than standard tokens. They alsoresulted in lower levels of problem (off-task)behavior. Moreover, the use of self-stimulatoryopportunities as terminal reinforcers did notlead to a general increase in self-stimulatorybehavior.These findings were extended more recently

by Carnett et al. (2014) with a 7-year-old childwith autism. The basic design and results canbe seen in Figure 2. Following a baseline periodwithout a token economy, tokens that engen-dered what the authors termed perseverativeinterest (PI tokens) were compared against stan-dard arbitrary tokens, such as pennies, in analternating treatments design in which the twoprocedures were presented in succession overtime. Both token types were exchangeableaccording to the same schedule and for thesame type of terminal reinforcers (food). Bothtoken procedures were effective, but the proce-dure with the embedded PI tokens was moreeffective than the standard token procedure,increasing on-task behavior and decreasingchallenging behavior. The effects of PI tokensthen generalized to a classroom setting.Despite the similarities between this type of

token-directed behavior and that seen in other


animals, there is an important difference.Unlike the animal studies, in which the token-directed behavior derives from pairings with theterminal reinforcer (food), the PI tokens usedin the studies with children with autism derivefrom preexisting reinforcers, established outsidethe study. These tokens then likely served pri-marily discriminative functions, occasioningself-stimulation, perhaps adding to the reinfor-cing value of that procedure. In other words,the relatively greater efficacy of the PI

procedures over the standard procedures maythus derive from the combined effects of thetoken procedure and the opportunity to engagein self-stimulatory behavior (a high-probabilityresponse).Although such self-stimulatory behavior is

normally considered maladaptive, the rationalewas based on the Premackian notion thatopportunities to engage in high-probabilitybehavior (in this case, self-stimulatory behavior)can reinforce less probable behavior (in this

Figure 2. Percentage of intervals on-task (top panel) and challenging behavior (bottom panel) for one child withautism across token economy conditions and token types: open symbols (standard tokens) and closed symbols (tokensengendering perseverative interest [PI]) (from Carnett et al., 2014). See text for additional details. Reprinted withpermission.


case, correct responses); stimuli that occasionsuch behavior should therefore acquire impor-tant stimulus functions. At first glance, suchprocedures may seem counterproductive in thatthey explicitly arrange for self-stimulatorybehavior to occur—behavior that may in turninterfere with the task being trained. This wasnot the case in either of the studies describedabove but is something to carefully consider indesigning a token economy.

Antecedent FunctionsMuch of the evidence reviewed above is con-

sistent with a conditioned reinforcement func-tion, based on repeated pairing of tokens andterminal reinforcers during exchange periods. Intypical token economies, however, tokens alsooccasion other behavior (namely, token produc-tion and exchange behavior), and thus likely alsoserve important discriminative as well as reinfor-cing functions. For example, in the Smith (1972)study, described above, the poker chips were notonly paired with trinkets during exchangeperiods, but also provided occasions upon whichexchange responses (deposits) were reinforced. Inrobust token economies, these two functions(reinforcing and discriminative) are complemen-tary and mutually supportive: Stronger condi-tioned reinforcers tend to go hand in hand withstronger discriminative control. And for mostpractical purposes, there may be little need toseparate the multiple functions of tokens. But foranalytic purposes, because reinforcing and dis-criminative functions of tokens are perfectly con-founded in typical procedures, little is currentlyknown about the separate contributions of thetwo functions in token economies.Future research should explore the effects of

different types of pairing operations andschedules—variables known to exert substantialimpact on conditioned reinforcers generally.For example, applied research has shown thatresponse contingency plays an important rolein establishing conditioned reinforcing

functions of stimuli, such that stimuli firstestablished as discriminative become effectiveconditioned reinforcers (Isaksen & Holth,2009; Taylor-Santa, Sidener, Carr, & Reeve,2014). When compared with stimulus-pairingmethods in which neutral stimuli are temporallypaired with reinforcers, such discriminative pro-cedures tend to be superior (Holth, Vandbakk,Finstad, Gronnerud, & Sorenson, 2009). Todate, however, the findings have been limited tothe new-response method, in which the effectsof the different procedures (discriminativevs. stimulus pairing) are tested to determinewhether they are capable of supporting a newresponse in extinction (i.e., the absence of fur-ther pairings with reinforcement). Suchmethods have two main limitations. First,because the stimulus-pairing and discriminativeoperant procedures are compared in extinction,while responding is in decline, the effects aretypically brief and transient. Second, suchmethods, in which the conditioned reinforcersare presented in the signaled absence of uncon-ditioned reinforcement, are quite unlike theconditions in which they will be used in treat-ment. For both of these reasons, comparisons ofthe two procedures for establishing conditionedreinforcers would be greatly enhanced if con-ducted under steady-state conditions, such aswith extended chained and concurrent chainedschedules, in which added stimuli continue tobe paired with terminal reinforcers. Such proce-dures generate more robust responding, andhave proven useful in the analysis of conditionedreinforcement more broadly (Fantino, 1977;Gollub, 1977; Shahan, 2010; Williams, 1994).Directing these types of procedures to token-reinforced behavior would provide importantinsights into the relative contributions made bystimulus-pairing, discriminative, and condi-tioned reinforcing functions.

Generalized Reinforcing FunctionsTokens are frequently exchangeable for more

than one terminal reinforcer (Ayllon & Azrin,


1965; McLaughlin & Malaby, 1972; Phillips,1968). Such tokens are said to be generalizedconditioned reinforcers, in the sense of notbeing tied to a particular deprivation. This con-ceptualization of generalized reinforcementoriginates with Skinner (1953). According toSkinner, and to subsequent interpretations(including virtually every behavioral textbookthat covers the topic), generalized reinforcersare assumed to retain reinforcing functionsacross a wider range of conditions than nonge-neralized, specific reinforcers (those associatedwith a single terminal reinforcer). This isthought to be due to the wider range of termi-nal reinforcers and motivational conditions towhich generalized reinforcers are related—anyone or more of which is likely to prevail at thetime the response is made.Generalized reinforcers, ranging from mone-

tary to social reinforcers, abound in everydayhuman contexts, including those in whichmuch of applied behavior analysis occurs. Tobegin with, monetary reinforcers (the prototyp-ical generalized reinforcer) have been shown toplay a powerful role in contingency manage-ment and other incentive-based interventionprograms (Higgins, Heil, & Lussier, 2004).Skinner also implicated generalized reinforce-ment in social behavior. Signs of attention andapproval are closely related to a wide range ofother reinforcers, and thereby acquire general-ized functions. Such social reinforcers play acrucial role in functional assessment and treat-ment programs as well. Yet despite the practicaland theoretical importance of the topic, little iscurrently known about the provenance of gen-eralized reinforcement—how generalized rein-forcers are established and maintained, andhow they compare to nongeneralized, specificreinforcers.In applied research, Sran and Borrero (2010)

found that three typically developing childrenpreferred generalized tokens (paired with multi-ple edible reinforcers) over specific tokens(paired with a single edible reinforcer). The

children worked on a tracing task in whichevery five completed units produced a token(a fixed ratio [FR] 5 token production sched-ule; see below). At the end of the 3-min ses-sions, any earned tokens could be exchangedfor preferred edibles (a fixed-interval [FI] 3-minexchange production schedule; see below).Response rates were slightly but consistentlyhigher in sessions in which the generalizedtokens were earned. Moreover, a modest prefer-ence for the generalized over the specific tokenswas observed, suggesting that the generalizedtokens were of somewhat higher efficacy thanthe specific tokens.These findings are broadly consistent with

laboratory research with pigeons in token econ-omies (DeFulio, Yankelevitz, Bullock, & Hack-enberg, 2014). Under various experimentalarrangements, pigeons could produce eitherspecific tokens or generalized tokens. In oneexperiment, food components (in which greentokens could be earned and exchanged forfood) alternated with water components(in which red tokens could be earned andexchanged for water) in a multiple schedule for-mat (i.e., two schedule components presentedin alternation, each correlated with distinctivestimuli). In both components, white (general-ized) tokens could be earned and exchanged foreither food or water. The cost of producing thespecific or generalized tokens was always thesame, yet the pigeons produced disproportion-ately more generalized than specific tokens.Together with the results of a more recentstudy with the same subjects and apparatus(Andrade & Hackenberg, 2017), pigeons pre-ferred generalized to specific tokens in 32 of40 conditions in which both options wereavailable at equal costs.Preference for generalized over more specific

reinforcers may be due to the expanded rangeof motivational conditions over which the gen-eralized reinforcers are effective. One set ofconditions in the Moher et al. (2008) study isrelevant to this issue. The design and main


findings for one participant are portrayed inFigure 3. The first three conditions were similarto an earlier set of conditions in the study,described briefly above, in which the reinfor-cing efficacy of the token was shown to dependon deprivation from the edible with which ithad been paired. Following the return to depri-vation (third phase), the token was paired witha second highly preferred edible and subse-quently tested under satiation conditions, inwhich presession free access to one or the otherreinforcer (but not both) was provided. If thetoken had acquired generalized functions,responding should be less sensitive to satiationconditions because deprivation remained ineffect for one of the terminal reinforcers. Con-sistent with this prediction, response ratesremained high across satiation conditions.Thus, only two terminal reinforcers were suffi-cient to produce generalized functions of thetokens. These effects were replicated with thetwo other participants, although for one, threeterminal reinforcers were needed to producegeneralized functions (see also Becraft &

Rolider, 2015, for replication and extension ofthe basic findings).In sum, the studies reviewed in this

section suggest a few promising directions forresearch on generalized functions of tokens,and many important questions remain. Tobegin with, we know very little about the his-tory of experiences needed to establish tokensas generalized reinforcers. It is common intoken economies to include a store in whichtokens can be exchanged for an array of backupreinforcers (preferred items and activities). It isassumed—but rarely demonstrated—that suchexperiences are sufficient to establish the tokensas generalized reinforcers.And although one finds occasional recom-

mendations to use a healthy variety of terminalreinforcers (Kazdin, 1982), little research hasbeen directed to this topic per se—what mightbe called the generalizability of the reinforcer. Itmay be most useful to view reinforcers on acontinuum of generalizability, ranging fromhighly specific (dependent on a single backupreinforcer and its deprivation) to highly

Figure 3. Response rates for one participant across different pairing and motivational conditions in Experiment 3 ofMoher et al. (2008). See text for additional details. Reprinted with permission.


generalized (associated with multiple backupreinforcers and less dependent on specific depri-vation conditions, such as money). Laboratorymethods with nonhumans are well suited toexploring parts of this continuum, but thenumber of terminal reinforcers is usually quitelimited (typically food, water, or drug). Appliedtoken research, on the other hand, can greatlyexpand the number and range of terminal rein-forcers, sharpening the analysis and permittinggreater quantification. When expanded toinclude a wider range of terminal reinforcers,procedures such as those reviewed here on pref-erence and schedule assessments (Becraft &Rolider, 2015; Moher et al., 2008; Sran &Borrero, 2010) are well suited to an analysis ofthis continuum of generalizability.

Schedules of Token ReinforcementBeginning with Kelleher’s work in the

1950s, early token reinforcement research over-lapped quite substantially with work onextended chained and second-order schedules.Boren and Colman (1970), for example, usedinsights gained from laboratory research onchained and higher-order schedules of rein-forcement to conceptualize and interpret theresults of token-based experiments conductedon a psychiatric ward with soldiers. One experi-ment was concerned with assessing the effec-tiveness of a token program on attendance at amorning meeting when organizational details ofthe token system were discussed. Althoughtokens were available for attendance, many sol-diers skipped the meetings. It was only whenattendance at the meeting (initial link of thechain) was required to access the exchangeperiod (the terminal link of the chain) thatattendance increased.Conceptualized as extended chained or

second-order schedules, a token reinforcementschedule can be analyzed with respect to threeprincipal components: (a) the token-productionschedule, by which tokens are earned, (b) the

exchange-production schedule, by which exchangeperiods are produced, and (c) the token-exchangeschedule, by which tokens are exchanged for ter-minal reinforcers. Research related to these func-tional components of a token economy will bereviewed below, along with recommendationsfor further research.Token-production schedules. The token-

production schedule specifies the contingenciesfor earning tokens. There are two main lines ofevidence of token-production schedule effectsin laboratory research with nonhumans. In one,Kelleher (1958) and Malagodi (1967) showedthat token-production schedules produce localpatterns with chimpanzees and rats, respec-tively, that resemble the patterns typically seenwith simple schedules. That is to say, FRschedules of token production generate typical“break-run” patterns observed under simple FRschedules of reinforcement, characterized by atwo-state pattern of pausing shortly after rein-forcement with an abrupt transition to a higherrate of responding until reinforcement.The second line of evidence comes from the

functional relationship between response ratesand FR schedule value. In simple FR schedules,this function is bitonic—it first increases andthen decreases (Mazur, 1983). Bullock andHackenberg (2006) showed that a similarbitonic function characterized the relationshipbetween response rates and FR token-production schedule value with pigeons. Takentogether, these patterns of findings show thattokens resemble food reinforcers in functionalcontrol over behavior, suggesting that thetokens indeed serve important conditionedreinforcing functions.There is little in the way of a direct analogue

in applied research, though a few studies haveexamined schedule-related effects in token pro-duction schedules, taking as a starting pointwhat is known about schedule-related differ-ences from laboratory research. For example,De Luca and Holborn (1990) compared therelative efficacy of FR and FI token-production


schedules on exercise and weight loss in sixth-grade boys. The target responses were wheelturns of a stationary bike in four boys, two ofwhom were obese and two of whom were non-obese, according to standard criteria. Followinga baseline phase during which wheel turns hadno programmed consequences, wheel turnsproduced tokens (points signaled by light andbell) across three successive phases. In the first,tokens were produced according to an FI 30-sschedule, in which the first wheel turn after30 s produced a token. In a second phase,tokens were produced according to an FRschedule, in which a fixed number of responseswas required to produce a token (with theresponse requirement matched to the responsesin the preceding FI condition). This was fol-lowed by a reversal to the FI 30-s schedule.Tokens earned during the 30-min sessionscould be later exchanged for functionally deter-mined terminal reinforcers (priced according toeach participant’s value ratings). Response rateswere consistently higher for the nonobese thanfor the obese boys, but a clear reinforcementeffect was seen for all four participants. Tobegin with, response rates for all participantsincreased in the transition from baseline to FIschedule. Rates increased further in the subse-quent FR condition, consistent with the higherresponse rates under FR than comparable FInontoken reinforcement schedules (Ferster &Skinner, 1957). Responding declined in thereturn to the FI condition, providing furtherevidence of schedule control.De Luca and Holborn (1992) replicated and

extended these findings, using variable ratio(VR) rather than FR schedules of token pro-duction, in which a variable number ofresponses was required to produce a token. Thetoken procedures were used to establish andmaintain exercise in 11-year-old obese andnonobese boys over a 12-week period. Therationale for using VR schedules was derivedfrom basic schedule research showing higherresponse rates under VR than comparable FR

schedules (Mazur, 1983). Following a baselineperiod in the absence of tokens with little activ-ity, wheel turns increased immediately uponintroduction of the VR token-reinforcementcontingency, and then increased across condi-tions in which the mean VR requirementsincreased by 15%. A subsequent 3-day returnto baseline produced rapid declines in respond-ing, showing functional control by the contin-gency. This was followed by a return to thehighest VR in the study, ranging from 110 to130 wheel turns per token. This generatedresponse rates in excess of 100 per min in allsix boys. These findings are important in dem-onstrating clear functional control by a high-rate contingency over a clinically importantactivity.Also taking a lead from basic research on

schedule-related differences in response rates,Repp and Deitz (1975) compared FR 60 andVR 60 schedules of token production with twoboys, aged 10-12. Button presses produced amarble token immediately exchangeable for apenny. In Phase 1 (20 sessions), the FR andVR token-production schedules were arrangedin a multiple schedule, with FR and VR com-ponents strictly alternating after each tokendelivery and exchange period. Response rateswere approximately equal in both components.In Phase 2, the FR and VR token-productionschedules were arranged concurrently in aFindley-type changeover procedure (Findley,1958). In this procedure, only one of the twoschedules was active; a response on a second(changeover) button changed the schedule fromone to the other. Both boys made a majority ofchangeover responses from the FR to the VR,indicating a strong preference for the VRschedule. This preference for VR over FR(owing perhaps to the disproportionate influ-ence of the occasional shorter reinforcer delaysin VR distributions) parallels results with otherschedules and species (Mazur, 2004). The par-allel between these effects obtained with tokensand with other reinforcers suggests that the


tokens were serving as conditioned reinforcers.That concurrent schedules can reveal schedule-related differences not observed with the sched-ule in isolation is also consistent with priorresearch with pigeons and food reinforcers(Neuringer, 1967).Together with the results of De Luca and

Holborn (1990, 1992), the Repp and Deitz(1975) results show clear functional controlover behavior by token-production schedules.These studies are conceptually significant inderiving their rationale from an understandingof basic schedule effects, using what is knownabout rate differences between FI, FR, and VRschedules to develop maximally effective tokenschedules. One limitation of these studies,however, is that because the tokens could beexchanged as soon as they were earned, thedelays to token delivery were confounded withdelays to the exchange period (when the tokenscould be exchanged for other reinforcers). Andit is this variable—the delay to the exchangeperiod—that has emerged as a powerful vari-able in token reinforcement research, both inlaboratory and applied settings.Exchange-production schedules. In addition to

token-production schedules, several studieshave shown that rates and patterns of token-reinforced behavior are systematically related toexchange-production schedules. In laboratorystudies, for example, when token-productionschedules are held constant, pausing increasesand response rate decreases with increases inFR exchange-production ratio (Bullock &Hackenberg, 2006, 2015; Foster, Hackenberg,& Vaidya, 2001), similar to that seen undersimple FR schedules (Felton & Lyon, 1966;Ferster & Skinner, 1957; Powell, 1968).Similar exchange-schedule effects have been

reported in applied studies. For example, anearly study by Staats, Staats, Schutz, and Wolf(1962) showed that tokens were effective ingenerating and maintaining behavior in a read-ing program with children, even withexchange-production ratios as high as

24 (i.e., tokens could be exchanged after24 had been earned). Staats et al. also reportedevidence of break-run patterns similar to thoseobtained with FR exchange-production sched-ules with chimpanzees (Kelleher, 1956), rats(Webbe & Malagodi, 1978), and pigeons(Bullock & Hackenberg, 2006; Foster et al.,2001). The similar schedule-related effects seenacross settings and species suggest that token-reinforced behavior in the laboratory and inapplied settings may be governed by similarprinciples.In one of the experiments in the Phillips

et al. (1971) study with delinquent youth in agroup home, an attempt was made to encour-age savings by making tokens contingent ondeposits into a bank account. Figure 4a (leftcolumn) shows cumulative deposits, averagedacross five subjects, in baseline (when depositswere recorded but did not produce any addi-tional consequences) and contingent points(when deposits produced points). Depositsoccurred infrequently in baseline but increasedsubstantially when tokens (points) were contin-gent on deposits. And although deposits couldoccur at any time in the week, the frequency ofdeposits increased with proximity to theexchange period, scheduled each Friday. Mostof the deposits occurred in close proximity toexchange, showing that token-reinforced behav-ior was under temporal control of exchangeopportunities.Figure 4b shows data from an analogous

experiment with rats (Waddell, Leander,Webbe, & Malagodi, 1972), in which leverpresses produced marble tokens according to anFR 20 schedule and exchange periods accordingto FI schedules. The FI schedules varied para-metrically across conditions (1.5, 4.5, and9 min), with a representative cumulative recordfrom each shown in Figure 4b. The rate oftoken-reinforced behavior tended to increaseacross the exchange-production interval, inproximity to the exchange periods. That simi-larities in the temporal patterning of token


production occurred in the Waddellet al. (1972) and Phillips et al. (1971) studies,despite differences in responses (lever pressing,money deposits), tokens (marbles, points), spe-cies (rats, humans), and settings (laboratory,group home), suggest that exchange-scheduleeffects have wide generality.Further evidence of exchange-schedule

effects in token economies comes from studiesin which exchange-production schedules aredirectly manipulated. For example, in spite ofgenerally similar temporal distribution of tokenproduction across the different FI curves inFigure 4b, increasing the FI value tended todecrease the overall response rates, mainlythrough increasing pausing early in the interval.This type of inverse relationship between

response rates and FI exchange-schedule dura-tion has also been seen with FR exchangeschedules (Bullock & Hackenberg, 2006;Foster et al., 2001). Applied research in whichexchange-schedule variables were manipulatedin this fashion would make a significant contri-bution to basic understanding of exchange-schedule effects, expanding the generality ofthis functional relationship across species andsettings. It would also hold important practicalsignificance, as a method for identifying opti-mal schedule values to use in applied tokeneconomies.Another way to manipulate exchange-

schedule variables is to compare behavior underfixed and variable schedules. In general,response rates tend to be higher under variable

Figure 4. (a, left panel): cumulative deposits in savings account across days of week in a residential token economywith delinquent youth. Different panels indicate conditions in which savings were reinforced (Points) or not (Baseline);arrows indicate exchange periods (from Phillips et al., 1971). (b, right panel): cumulative token-production lever pressesper minute across time in a laboratory token economy with rats. The different panels include records from threeexchange-production schedules: 1.5, 4.5, and 9 min FI schedules; circles indicate exchange periods (from Waddell et al.,1972). Both figures reprinted with permission.


than under otherwise comparable fixed sched-ules. This has been seen in laboratory researchwith both simple fixed and variable schedules(Mazur, 1983), as well as with fixed and vari-able exchange-production schedules in tokenreinforcement procedures (Foster et al., 2001;Webbe & Malagodi, 1978). It has also beenshown in a classroom token economy withsixth graders (McLaughlin & Malaby, 1972).Points could be earned for assignment comple-tion and lost for inappropriate behavior. Pointscould be exchanged for functionally definedterminal reinforcers (mainly classroom privi-leges) scheduled after either a fixed or variablenumber of days. Assignment completions inboth token conditions were superior to controlconditions in which the privileges were notcontingent on assignment completion, showingthat the mere presence of the token economywas not sufficient to maintain academic gains;the tokens had to be contingent on the targetbehavior. Although the variable schedule (tech-nically, a variable-time, or VT) seemed to pro-duce even stronger effects than the fixed-time(FT) schedule, this was evident in only a subsetof students with inconsistent academic perfor-mance; the others were already performing athigh levels under the FT exchange schedule.The effects of the VT schedule were thusobscured by a ceiling effect.Because the VT schedule had a slightly lower

mean requirement than the FT schedule (4.5vs. 5.0 days), however, the effects of the vari-able scheduling could not be completely disen-tangled from the relatively shorter exchangedelay for the VT exchange-production sched-ule. To remedy this, the authors conducted afollow-up study that equated the FT and VTexchange schedules (McLaughlin & Malaby,1976). Each was equal to 5 days, but the VTexchange was scheduled according to a variabledistribution with an equal likelihood of 3, 5,7, and 9 days between exchange periods. Usinga combined reversal and multiple-baselinedesign across responses (math, language, and

spelling), both FT and VT exchange schedulesgenerated high rates of assignment completion.Similar to the earlier study, the VT schedulereduced variability, producing near perfectlevels of completion (though the effects wereagain obscured by ceiling effects). Followingthe second VT 5 was a higher-valued VT(mean = 13.2 days, range = 8-21 days) with aminimum delay exceeding the mean of theother conditions. Assignment completionincreased from 88%-100% during FT exchangeto nearly 100% during VT exchange (includingthe VT 13.2).Tarbox, Ghezzi, and Wilson (2006) also

demonstrated the importance of the exchange-production schedule in the context of a thor-ough within-subject experimental analysis oftoken-reinforcement effects with a 5-year-oldboy with autism. The response of interest,attending to the therapist, was examined sys-tematically as a function of token reinforce-ment: Each response produced a sticker laterexchangeable for the terminal reinforcer—periods away from the session and access topreferred items. Attending occurred on nearly100% of opportunities under token reinforce-ment conditions, and at low levels under base-line conditions when the token reinforcementschedule was inoperative. Responding was sus-tained across substantial increases in theexchange-production ratio, from an initial FR10 up to FR 50 (i.e., 50 tokens required toproduce the exchange period).Token-exchange schedules. The Tarbox

et al. (2006) study went on to demonstrate theimportance of two additional variables: pairingand token-exchange schedule. Results fromthese conditions are summarized in Figure 5,which shows the percentage of trials withattending behavior across a series of reversalscomparing tokens paired or not with the termi-nal reinforcer (designated TR, for Token Rein-forcement and NB for No Backup, in thefigure). In both conditions, the exchange-production schedule was fixed at FR 20. In the


NB conditions, social praise followed theexchange but no terminal reinforcer occurred.Removing the terminal reinforcer led tomarked reductions in attending, but attendingquickly recovered when a token reinforcementschedule was reinstated, demonstrating func-tional control by this variable.Then, with the exchange-production sched-

ule held constant at FR 20, exchange opportu-nities were delayed according to a fixed-time(FT) schedule, which increased in 5-s incre-ments across blocks of sessions, up to delays of190 s. In these FT token-exchange schedules,tokens could be produced but could not beexchanged until after the delay. Attendingdecreased but was not eliminated as exchangeopportunities were delayed; even at FT dura-tions of 190 s, attending occurred at approxi-mately 40% of the FT 0-s delay levels.Attending recovered to high levels when the

token reinforcement contingencies werereinstated.A more extensive parametric analysis of

delayed reinforcement effects was conducted byLeon, Borrero, and DeLeon (2016) in a tokenreinforcement context with three individualswith intellectual disabilities. Three conditionswere compared: (a) delayed access to preferredfood (no token, delayed food); (b) delayedaccess to tokens exchangeable for preferred foodat the end of the session (delayed token,delayed exchange/food); and (c) immediateaccess to tokens exchangeable for preferred foodat the end of each trial (immediate token,delayed exchange/food). The delays were sys-tematically increased across successive sessionsuntil responding dropped to low levels. For thetwo participants who completed all three condi-tions, responding weakened most quickly underdelayed token delivery, reaching low levels with

Figure 5. Percentage of trials with a target (attending) response for one child with autism across token reinforcementconditions: baseline (BL), token reinforcement (TR), schedule thinning (ST), no terminal/backup reinforcer (NB), anddelayed exchange (DR) (from Tarbox et al., 2006). See text for additional details. Reprinted with permission.


6-s delays, and remained relatively strong underthe other two conditions until longer delayswere reached. Delay tolerance (delay values atwhich responding dropped to low levels) wasslightly higher under delayed food than underdelayed exchange for one participant, whereasfor the other, delay tolerance was slightly higherfor delayed exchange than for delayed food.The findings from the delayed exchange con-

dition are consistent with what is known abouteffects of delayed exchanges under laboratoryconditions. In one early experiment with chim-panzees by Wolfe (1936), response mainte-nance under delayed exchange conditions wascompared across four conditions, in which aresponse produced (a) delayed food (no token,delayed food); (b) immediate access to a poker-chip token exchangeable for food after a delay(immediate token, delayed exchange/food);(c) immediate access to a brass token that wasnot exchangeable for food (immediate token/unpaired); and (d) immediate access to a poker-chip token and deposit but a delay to food.Within each condition, the delays increasedprogressively with each earned reinforcer, aprogressive-interval (PI) schedule. Only condi-tion (b) sustained responding, reaching break-points (or last completed ratios) of 20 and60 min for two subjects and 60+ for the othertwo (that continued to respond at the maxi-mum delay studied). The other three condi-tions generated weak and inconsistentresponding, with breakpoints rarely in excessof 3 min.It is worth noting that the conditions that

generated the strongest behavior (immediatetokens, delayed exchange) were also those inplace in the Tarbox et al. (2006) study and insome conditions in the Leon et al. (2016)study, in which the student produced tokensbut had to tolerate increasingly longer delays tothe exchange period. These procedures couldreadily be adapted to include conditions likethose used by Wolfe (1936) to systematicallyidentify the optimal arrangement of token-

production and exchange contingencies. Theprocedures might also be expanded to includegeneralized tokens rather than the specifictokens used by Leon et al. (2016). Identifyingways to enhance exchange delay tolerance hasobvious applied significance; it has theoreticalimplications as well, bearing on fundamentalissues such as the necessity of token–food pair-ings in maintaining token-reinforced behavior.Antecedent functions revisited. In the research

reviewed in these prior sections, various ante-cedent functions are difficult to disentangle. Aspart of an extended chained schedule, a tokenis both paired with a terminal reinforcer andoccasions responses that occur in its presence.As such, tokens will normally have multiplefunctions, the relative influence of which islikely to vary across conditions.A laboratory study by Bullock and Hacken-

berg (2015) attempted to disentangle thesemultiple stimulus functions of tokens. Thetokens were lights arranged in a horizontal rowabove the response keys and the subjects werepigeons, well experienced with token econo-mies. Two-component multiple schedules wereused to compare responding in token schedulesto responding in otherwise identical (tandem)schedules but without tokens present. In thetoken components, tokens were earned accord-ing to a FR 50 token-production schedule andexchanged according to a FR x exchange-production schedule (in which x varied acrossconditions: 2, 4, and 8). In the tandem compo-nents, an equivalent number of responses wererequired to produce food reinforcers, but in theabsence of tokens. So, for example, in the com-parison with the FR 4 exchange-productionschedule, 200 responses were required to pro-duce four food reinforcers in the tandem com-ponent. These tandem components served as acontrol for the added stimuli in token proce-dures. If the tokens are serving primarily rein-forcing functions, one might expect higherresponding in the token components than inthe tandem components, owing to the presence


of four additional conditioned reinforcers.Overall response rates were in fact consistentlylower in token than in tandem components, afinding that is consistent with prior resultscomparing behavior in chained and tandemschedules of food reinforcement (Jwaideh,1973). Such differences strongly suggest dis-criminative effects, namely, weak responding inthe presence of stimuli early in the chain(i.e., temporally distant from the terminalreinforcer).It is worth noting that the rate differences

between token and tandem were not merely aproduct of weak responding in the initial linksin the absence of tokens, but rather, to lowerrates across the entire link. Figure 6 shows run

rates (responses with preratio pauses subtractedout) under token and tandem componentsacross links within the chain at all threeexchange-production ratios. The functions aregraded in proximity to the exchange period,reflecting temporal control, but are displaceddownward for the token component, reflectingdiscriminative control by the tokens. A laterexperiment showed that similar such patternswere evident even in the signaled absence of acontingency between responding and tokenpresentation, albeit at somewhat lower levels.Thus, tokens can potentially serve one of threefunctions—reinforcing, discriminative, oreliciting—depending on the particular arrange-ment of contingencies.

Figure 6. Token-production response rates across token-production segments for three pigeons in token (filled sym-bols) and tandem (open symbols) schedules (from Bullock & Hackenberg, 2015). The different columns represent dif-ferent exchange-production schedules: FR2, FR4, and FR8. See text for additional details. Reprinted with permission.


Although no comparable studies from theapplied realm exist, including tandem controlconditions alongside token reinforcement pro-cedures can be beneficial in disentangling con-trol by conditioned reinforcement from controlby discriminative stimuli. Such multiple controlis not limited to token reinforcement, but istrue of conditioned reinforcement more gener-ally, both in the laboratory (Shahan, 2010;Williams, 1994) and in applied settings, as dis-cussed earlier (Holth et al., 2009). Token rein-forcement procedures that separate thesedifferent stimulus functions may thereby aid ina more general understanding of conditionedreinforcement and discriminative control.

Aversive FunctionsIf token gains serve reinforcing functions, to

what extent do token losses serve aversive func-tions? The two main ways to assess the aversivefunctions of a stimulus event are via punish-ment (response-contingent loss) and negativereinforcement (response-contingent reductionin losses). In a seminal early laboratory investi-gation of token-loss negative reinforcement,Weiner (1963) studied adult human subjectson avoidance and escape procedures under lab-oratory conditions. The aversive events werepoint-loss periods (PLPs), or periods of signaledtoken loss. Responding was established andmaintained across three components in a multi-ple schedule: avoidance, in which responsespostponed the upcoming PLP; escape, in whichresponses terminated the PLP; and escape/avoid-ance, in which both termination and postpone-ment were possible. Responding wasestablished and well maintained across all threecomponents, and adjusted to the local contin-gencies: just after PLP onset in escape, andprior to PLP onset in avoidance. On the whole,the response patterns are impressively consis-tent with those seen with other species andother aversive stimuli, suggesting that tokenlosses are functionally equivalent with other

aversive stimuli (Azrin & Holz, 1966; Pazuli-nec, Meyerrose, & Sajwaj, 1983).Additional evidence for the aversive func-

tions of token loss comes from the aforemen-tioned study by Phillips et al. (1971). In oneexperiment, the target was promptness for din-ner; tokens (points tallied on index cards) wereearned for appropriate behavior and lost forinappropriate behavior. Tokens could beexchanged for privileges (e.g., hobbies andgames, allowance). A punishment contingencywas instituted, such that 100 tokens were lostfor each minute late. The contingency waseffective in reducing lateness from about10 min to 0-1 min in an ABAB design.Threats of losing points had an initial but tran-sient effect; only when lateness cost tokens wasit reduced.A second experiment involved negative rein-

forcement contingencies: If 80% of the avail-able points were earned, points were given forthe day; if less than 80% of the points wereearned, points were subtracted. Corrective feed-back was also given. This contingency waseffective in producing and sustaining a highlevel of behavior relative to a second conditionwith no points. Reinstating the point contin-gency brought behavior back to high levels.The contingency was then faded, first byremoving corrective feedback, then by decreas-ing the proportion of days on which the con-tingency was in effect, while holdingreinforcement magnitude constant. High levelsof behavior were maintained across the fadingand 6-month follow-up conditions.Comparing token-loss with other aversive con-

trol procedures. An important question sur-rounding the aversive functions of token loss ishow token loss compares to other aversive stim-uli. In an early systematic analysis of this ques-tion, Burchard and Barrera (1972) comparedpunishment via token loss to comparable time-out (TO) procedures on antisocial behavior inadolescents diagnosed with intellectual disabil-ities in an institutional setting. The magnitude


of the punisher was manipulated across groupsof participants: smaller magnitude (five tokensor 5-min TO) versus larger magnitude(30 tokens or 30-min TO). The results can beseen in Figure 7, which shows the overall fre-quency with which each type of consequencewas delivered, averaged across participants ineach group. (Although the y-axis label reads“total timeouts,” a timeout referred to either aresponse-cost fine or to the more traditionaldefinition of removal from the setting.) Bothtypes of punishment produced magnitude-related reductions in the rate of each conse-quence, reflecting the rate of responding.Greatest suppression was seen under the largermagnitude conditions, consistent with punish-ment magnitude effects seen in laboratory

research with humans (Weiner, 1962). It is alsobroadly consistent with shock-punishmenteffects seen in a wide range of nonhuman spe-cies (Azrin, 1960). The consistency in the func-tional relationship between responsesuppression and punishment magnitudestrongly shows that token losses are effectivepunishing events.Along with TO procedures, token loss proce-

dures are negative punishment procedures, inthat the aversive functions arise from theremoval of positive reinforcers. Thus, unlikepositive punishers (e.g., shock, social disap-proval), token-loss contingencies depend on thereinforcing value of token gain. That is,because losses, by definition, reduce the overallrate of token reinforcement, the response-suppressive effects of punishment are difficultto tease apart from the response-weakeningeffects of (sometimes marked) reductions inreinforcement rate or quality. In the Burchardand Barrera (1972) study, for example, thepoint-loss (and TO) contingencies also reducedoverall reinforcement, making it difficult to iso-late controlling variables. Disentangling thesedifferent sources of control requires controlprocedures in which token losses are separablefrom concomitant changes in overall reinforce-ment density.No applied research on token loss has been

directed to isolating punishment effects fromunderlying changes in reinforcement rate, butsome laboratory research may provide a startingpoint. In a laboratory-based token economywith pigeons, Raiff, Bullock, and Hackenberg(2008) assessed token-loss punishment acrosstwo components of a multiple schedule. A stan-dard token reinforcement schedule operated inboth components in baseline no-loss conditions.Under loss conditions, a token-loss punishmentcontingency operated conjointly with thetoken-gain schedule in one component, andonly the gain contingency operated in the othercomponent. That is, in loss conditions,responses both produced and subtracted

Figure 7. Frequency of timeouts (TO) across timeand procedure type: token loss (circles) and timeout (tri-angles); small magnitude (five tokens, 5-min TO) andlarge magnitude (30 tokens, 30-min TO) punishers arerepresented by open and closed symbols, respectively(from Burchard and Barrera, 1972). Reprinted withpermission.


tokens, whereas in no-loss conditions, responsesonly produced tokens. Responding was substan-tially reduced in the loss component, relative tothe no-loss component, suggesting a punishingfunction. The conditions also resulted in sub-stantial reductions in the overall number oftokens exchanged for food, however. To teaseapart the influence of these variables, a subse-quent control condition provided noncontin-gent food, yoked to the obtained rate in the no-loss component. Responding was somewhatlower in the yoked-food than in the baseline no-loss conditions, reflecting some effects of reducedreinforcement rate, but was even lower in theloss condition, reflecting the additional suppres-sive effects of punishment. Thus, even with pos-itive reinforcement variables held constant,token loss appears to function as an effectivepunisher (see also Pietras & Hackenberg, 2005,for analogous findings using different yoked-control methods). Similar effects have beenreported with TO as an aversive event: respond-ing can be maintained by TO postponementeven in the absence of increases (and sometimesin spite of decreases) in overall food reinforce-ment rate (Pietras & Hackenberg, 2000).Together with the findings discussed earlier,

these findings support a view of tokens as con-ditioned reinforcers and token losses as condi-tioned aversive stimuli. These laboratoryfindings also point to some critical control con-ditions needed to separate important behavioralfunctions. Such procedures could easily beadapted to applied research. Indeed, presentingtokens irrespective of behavior (such as theyoked-control methods of Raiff et al., 2008) isa variant of a commonly used procedure inapplied research—noncontingent reinforcement(NCR; Phillips, Iannaccone, Rooker, & Hago-pian, 2017). Incorporating NCR-like proce-dures as control procedures into token-losspunishment programs will help reveal keybehavioral functions of tokens.Comparing token-loss with token-gain proce-

dures. Another important question surrounding

token-loss procedures is how they compare totoken-gain procedures. Although many tokeneconomies involve combinations of loss andgain, a few studies have directly compared therelative efficacy of loss and gain procedures. Inan early study along these lines, Kaufman andO’Leary (1972) examined token reinforcementand punishment effects in a classroom tokeneconomy. Tokens could be earned for appro-priate behavior or lost for disruptive behavior.One group of students could earn up to10 points per class period, whereas a secondgroup could avoid losing up to 10 points. Bothsets of contingencies (loss and gain) producedclear and systematic reductions in disruptivebehavior; one was not appreciably superior tothe other.Similarly, Iwata and Bailey (1974) compared

token reinforcement and punishment effects onacademic performance in a classroom tokeneconomy. Unlike the Kaufman and O’Leary(1972) study, in which gain and loss conditionswere varied across groups, the procedures herealternated on a within-subject basis, with ordercounterbalanced across groups of students.Thus, students in each group received both lossand gain conditions, but in a differentsequence, along with interspersed baseline con-ditions with neither contingency in effect. Sim-ilar to Kaufman and O’Leary, both procedureswere about equally effective in reducing disrup-tive behavior.More recently, Donaldson, DeLeon, Fisher,

and Kahng (2014) compared token-loss andtoken-gain contingencies on disruptive behaviorin a classroom setting. Loss and gain conditionswere varied in a multielement design, with theoverall earning potential in each condition heldconstant at 10 tokens. A time-sampling methodwas used whereby students either received atoken (check mark) for appropriate behavior, orhad one removed for inappropriate behavior,under gain and loss conditions, respectively.Tokens were later exchangeable for tangibleand edible reinforcers. Similar to prior research,


both procedures reduced disruptive behavior tocomparably low levels, and no adverse collateralbehavior was reported. On average, participantsearned slightly more tokens in, and displayed amodest preference for, the token-loss than thetoken-gain conditions. The preferences for lossover gain procedures parallels previous findings(Hanley, Piazza, Fisher, & Maglieri, 2005),raising at least some doubts about negative sideeffects of the token-loss procedures.Collateral effects of token loss. One consider-

ation surrounding the use of aversive controlprocedures, more generally, is the incidence ofcollateral effects (e.g., emotional behavior,escape, social disruption) observed with nonhu-man animals and noxious stimuli such as elec-tric shock (Azrin & Holz, 1966; Hutchinson,1977). How prevalent are such effects withtoken loss, and how do they compare withother aversive stimuli?One common collateral effect of some aver-

sive stimuli is an increase in physiologicalarousal (Brady, Kelly, & Plumlee, 1969). Doesloss produce similar emotional arousal?Although there are few studies along theselines, an early experiment by Schmauck (1970)compared autonomic arousal just after shock,token loss, and social disapproval. Autonomicresponses were elicited by shock but not by theother two aversive stimuli; all were judgedequal in subjective reports of anxiety. Thus,despite the equivalent subjective effects, thephysiological response profile differed, withtoken loss producing no change in physiologi-cal responding. Although further study isclearly needed, the limited available evidencesuggests that token losses do not necessarilygive rise to the negative emotional responsesseen with electric shock. Thus, emotionalresponding does not seem to be an inevitableor perhaps even likely accompaniment oftoken-loss procedures.Another commonly reported side effect of

aversive events is escape from the situation inwhich the aversive events are occurring

(Azrin & Holz, 1966). The study by Borenand Coleman (1970), mentioned earlier,reported negative side effects associated with atoken-loss procedure with soldiers in a psychiat-ric ward. When tokens were contingent uponattendance at a morning meeting, most but notall soldiers attended. When penalties, in theform of token losses, were introduced forabsences while holding the reinforcement con-stant, however, attendance decreasedmarkedly—the opposite of a punishment effect.In addition to poor attendance, the authorsreported clear changes in other aspects of behav-ior. The subjects acted more rebelliously(e.g., encouraging others to miss the meeting)and generally more aggressively with each otherand with staff. When the point-loss punishmentprocedure was discontinued in a subsequentcondition, attendance increased to previouslevels and the punishment-induced aggressivebehavior quickly extinguished. In this case, atleast, the indirect effects of the token-loss con-tingency were large and generally detrimental inthat they opposed the target behavior.Aversive control has also been known to pro-

duce social disruption. An illustrative case isprovided by a study conducted in a residentiallaboratory setting (Emurian, Emurian, &Brady, 1985). Participant triads (three-persongroups) lived together for up to 3 weeks, com-pleting work units for points later exchangeablefor money. Points contributed to a groupaccount, with earnings divided evenly amongthe three participants at the end of the study.Positive and negative reinforcement contingen-cies were compared across blocks of days on awithin-triad basis. In positive reinforcementconditions, each completed work unit added atoken to the group account; in negative rein-forcement conditions, satisfying a fixed require-ment (yoked to earnings from the prior positivereinforcement condition) prevented reductionsin accumulated earnings.Four triads were studied with repeated rever-

sals across blocks of daily sessions to


demonstrate functional control. Despiteroughly similar levels of overall productivityunder the two sets of contingencies, major dif-ferences in collateral behavior were observed. Inpositive reinforcement conditions, the partici-pants worked together well and were sociallyamicable, earning and exchanging a high rate oftokens for other reinforcers. In negative rein-forcement conditions, on the other hand, therewere marked increases in negative verbalresponses about the experiment. Figure 8 showsmean daily ratings on a 4-point scale (rangingfrom 1, not bothered by the experiment, to4, extremely bothered by the experiment) onnegative reinforcement (avoidance) and positivereinforcement (appetitive) days. Ratings wereconsistently more negative under negative rein-forcement conditions. Similar patterns were evi-dent in analogous scales rating theexperimenters and other participants. In onegroup (G-4, in Figure 8), these changes in ver-bal behavior coincided with decrements in pro-ductivity. One participant stopped workingaltogether, and another reduced productivityunder negative reinforcement conditions, both

in response to a third group member whoseproductivity rate was insufficient to meet theperformance criterion. The authors interpretedsuch negative collateral behavior as a side effectof aversive control, analogous to those seenwith nonhuman animals and noxious stimulisuch as electric shock.It is worth noting, however, that collateral

behavior is not always produced by token-lossprocedures. Neither Kaufman and O’Leary(1972) or Iwata and Bailey (1974) found evi-dence of negative collateral behavior; indeed,both studies reported positive indirect effects ofthe token-loss procedures: increases in academicoutput (not specifically targeted by the token-loss contingency). Clearly, the effects of point-loss procedures, including the incidence of col-lateral byproducts of such procedures, are acomplex function of many variables. In somecases, token-loss procedures produce effectscomparable to token-gain procedures and arefree of unwanted collateral effects. In othercases, token-loss procedures produce similardirect effects but negative collateral effects. Instill other cases, token-loss procedures produceweaker direct effects along with negative collat-eral effects.The variability in findings may be due in

part to different subsets of contingencies oper-ating within the token economy. One differ-ence between the studies was in how thetokens were presented. In the studies reportingmainly positive effects of point-loss contingen-cies, the tokens were presented noncontingentlyat the beginning of a class period; they werethen subject to loss for specified responses. Inthe studies reporting weaker and negativeeffects, on the other hand, the tokens removedaccording to the token-loss procedure had pre-viously been earned. Why should this make adifference?Loss aversion. One possibility is that this dif-

ference in how tokens are presented (whetherthey are given freely or have to be earned) inthe token economy may alter the value of the

Figure 8. Mean daily verbal ratings on a 4-point scale(ranging from 1, not bothered by the experiment, to4, extremely bothered by the experiment) on negativereinforcement (avoidance) and positive reinforcement(appetitive) days for four social groups of three subjectsapiece (from Emurian et al., 1985). Reprinted withpermission.


tokens. A token earned, as opposed to onefreely given, may be more highly valued. Astudy by Miller, DeLeon, Toole, Lieving, andAllman (2016), for example, found that partici-pants who earned their own tokens were morerisk averse (less likely to gamble those tokens)than participants given noncontingent (free)tokens. These findings, along with those from asimilar study (Brandt, Sztykiel, & Pietras, 2013),suggest that the contingent delivery of tokensmay enhance their value relative to those freelydelivered. Perhaps at least some of the negativecollateral effects reported by Boren and Colman(1970) and Emurian et al. (1985), when point-loss conditions were instituted, reflected a similarkind of value adjustment, in which tokens in aloss context are weighted more heavily than thesame tokens in a gain context.This tendency to weigh losses more than

gains may be related to the literature on the so-called endowment effect—the tendency to valuemore a commodity in one’s possession(to which one is endowed) to the same com-modity at a price. In a prototypical experiment(e.g., Kahneman, Knetsch, & Thaler, 1990),participants engage in a market exchange fortwo items of equivalent value: candy bars andcoffee mugs. Despite being randomly assigned,the people tend to place greater value on itemsin their possession, charging more to sell theitem than they were willing to pay for the sameitem. The endowment effect is normally con-ceptualized as reflecting loss aversion—the ten-dency to value losses more heavily than gains.Loss aversion has been empirically demon-strated across a wide range of studies in variousdecision-making domains, and is consideredtheoretical bedrock in some prominentapproaches to behavioral economics, such asProspect Theory (Kahneman & Tversky, 1979).Yet the studies to date are based almost entirelyon verbal-hypothetical scenarios with adulthumans (but see Levitt, List, Neckermann, &Sadoff, 2012, for an interesting exception),which raises questions about their applicability

both to laboratory-based and everyday situa-tions with repeated choices and real outcomes.Token systems permit a more general

approach to assessing the relative value of gainsand losses, appropriate to either verbal or non-verbal tasks. That roughly comparable resultsare produced in gain and loss contingencies(Donaldson et al., 2014; Iwata & Bailey, 1974;Kaufman & O’Leary, 1972) suggests losses andgains are equally weighted—a result broadlyinconsistent with loss aversion. On the otherhand, the modest preferences for token lossover token gain reported by Donaldsonet al. (2014) may be viewed as consistent withloss aversion. Similarly, in a laboratory-basedtest of loss aversion, Rasmussen and Newland(2008) found evidence supporting loss aversion.Adult humans made repeated choices betweentwo schedules of reinforcement (token gain),one of which sometimes also included punish-ment (token loss). Punishment conditions alter-nated with no-punishment conditions acrossblocks of sessions. If losses loom larger thanotherwise equivalent gains, then strong bias forthe unpunished alternative would be predicted.This is indeed what happened. Such dispropor-tionate effects of token loss are consistent witha loss aversion account.An earlier laboratory study by Critchfield,

Paletz, MacAleese, and Newland (2003), how-ever, did not find evidence of loss aversion.Like Rasmussen and Newland (2008), theygave humans repeated choices between sched-ules of token gain and loss. The conditionswere arranged in such a way as to evaluate thepredictive accuracy of a punishment model thatviews losses and gains as symmetrical but oppo-site in their effects. Across a broad parametricrange of conditions, preferences were generallywell aligned with the predictions of the model,supporting a version of the symmetrical law ofeffect dating back to at least Thorndike (1911),and counter to loss aversion.Although the few studies designed to assess

the symmetry between losses and gains have


yielded mixed results, token economies areespecially well suited to functionally scalinglosses and gains along a common dimension, aslosses and gains can be defined in conceptuallyanalogous terms, as common currency units ofloss or gain. And with currency units more likethose of monetary currencies, token proceduresmay provide a bridge between laboratory-basedchoice procedures used in behavior analysis andthose used in other behavioral economicapproaches. Such token-based procedures withhumans would also connect to token-basedprocedures with other animals, facilitatingcross-species comparisons more generally (seeHackenberg, 2009).

APPLIED BEHAVIORAL ECONOMICS

In a series of papers published between 1970and 1980, Kagel, Winkler, and colleaguesencouraged a view of token systems as experi-mental economies (Battalio et al., 1974; Kagel,1972; Winkler, 1973, 1980). Kagel andWinkler (1972), who were credited with havingcoined the term behavioral economics (Madden,2000), argued that token systems were ripe fora behavioral economic analysis. Tokens areearned for completion of some task andexchanged for access to commodities. In eco-nomic terms, token production is analogous toa wage, exchange opportunities to procurementcosts, exchange rate to the price of a good,accumulation to savings, and terminal rein-forcers to expenditures.Using this type of analysis, several early stud-

ies, conducted mainly in institutional settings,found broad support for economic demand the-ory—a family of models and concepts con-cerned with how consumption of a commoditychanges with price and the availability of othercommodities (Winkler, 1971, 1972). Overall,the results were consistent with those obtainedin other token economic research (Battalioet al., 1974; Schroeder & Barrera, 1976). Theyare also in general agreement with econometric

data on consumption patterns in the nationaleconomy, supporting the general view of tokensystems as analogues of real economic systems.Despite the promise of this work, however,

an economic approach to token systems didnot gain a strong foothold in applied behavioranalysis at the time. And while behavioral eco-nomic methods and concepts have been usedproductively in applied behavior analysisresearch over the past two decades (Bickel &Vuchinich, 2000; DeLeon, Iwata, Loh, &Worsdell, 1997; Francesco, Madden, & Bor-rero, 2009; Madden, 2000; Roane, Call, & Fal-comata, 2005; Tustin, 1994), little of this workhas focused on token economies perse. Combining an economic approach withtoken systems has great translational appeal,poised to yield rapid advances in the analysisand application of token reinforcement effects.In this section, I review some findings relevantto an economic account, and consider somepromising areas of research and application sug-gested by an applied behavioral economics.

Different Price MeasuresIn a token economy, wages correspond to

the token earnings—the number of tokens perunit of work (labor). A straightforward way tomanipulate wages in a token economy is viachanges in reinforcement magnitude. This hasemerged as a powerful variable in contingencymanagement programs, in which vouchers(tokens) are earned for drug abstinence. Pro-grams with higher wages (monetary incentives)tend to produce better treatment outcomes(Dallery, Silverman, Chutuape, Bigelow, &Stitzer, 2001; Levitt et al., 2012). Similar wageeffects have been reported in token economiesin institutional settings. Fisher (1979), forexample, altered the wage rate in a token econ-omy with 16 adults in an institutional mentalhealth ward. The target response, brushingteeth, was compared across baseline (no token)and two different magnitudes of token


reinforcement (one vs. five tokens) in an A-B-A-C-A within-subject design. Relative to base-line (no-token) conditions, brushing teethincreased under both token reinforcement con-ditions, and was sensitive to wage rates, increas-ing to a greater extent in the high-wage(5 token) than in the low-wage (1 token)conditions.In addition to labor, costs can be defined in

terms of the work to produce an exchangeopportunity (procurement costs) and the numberof tokens to exchange for a terminal reinforcer(prices). These distinctions are critical in under-standing reinforcement effects in token systems,as there is clear evidence from schedule research(summarized above) that token-reinforcedbehavior is jointly determined by all three typesof costs. When economic analyses are appliedto behavior, costs are generally labor costs—thenumber of responses to produce a reinforcer—but with a token system, it is possible to definecosts in terms of common currency units(tokens)—the price (number of tokens) one iswilling to pay for it. This more closely resem-bles real-world economic systems, in whichmoney is used to purchase different reinforcers,and one can infer relative reinforcer value fromexpenditures. Of perhaps greater importance isthat token systems provide the basis for a com-mon scale for comparing qualitatively differentreinforcers in a quantitatively precise way—amajor aim of applied behavioral economics.

Quantifying Reinforcer ValueAmong the main strengths of an economic

approach to behavior are tools for identifyingand quantifying different reinforcers. The prin-cipal method for quantifying reinforcer valuefrom an economic standpoint is the demandanalysis: consumption as a function of price.Demand for commodities highly sensitive toprice changes (i.e., greater than proportionalchanges in price) are said to be elastic, whereasthose less sensitive to price changes (i.e., less

than proportional changes) are inelastic. Rein-forcers are typically characterized by mixedelasticity—they are less elastic at low prices andmore elastic at higher prices. For example,demand for a highly valued good such as coffeemight be inelastic at relatively low prices, butelastic if the price rises enough. The resultingresponse output (work) functions are bitonic—increasing over the inelastic portion of thecurve before eventually declining at higherprices.Considering the mathematical characteristics

of demand functions for different reinforcerscan yield important information about optimalschedule values for generating peak responseoutput. For example, in early investigationalong these lines, Tustin (1994) analyzed thereinforcing value of visual, auditory, and socialreinforcers with young adults with moderate tosevere intellectual disabilities. Social reinforcersconsisted of brief bouts of teacher attentionand sensory reinforcers consisted of briefperiods of computer-generated auditory, visual,or combined auditory and visual stimuli. Forone participant, demand functions wereobtained separately for social and sensory rein-forcers. Consumption (reinforcers earned) andwork output (response rate) was similar forboth reinforcer types at the lowest price (FR 1).As the price (FR size) was increased, both con-sumption and work output declined for bothsocial and sensory reinforcers, but the slope ofthe function was shallower for sensory than forsocial reinforcers. In economic terms, demandfor sensory reinforcers was less elastic than forsocial reinforcers, suggesting higher relativereinforcer efficacy. Such differences in rein-forcer efficacy have important practical implica-tions, as typical reinforcer assessments use lowcost (typically FR 1) procedures and could thusmask potentially significant differences in rein-forcer value that emerge at higher prices. Forexample, in this case, probing only the FR1 price would provide a misleading picture ofthe relative reinforcing efficacy of the two


reinforcers. Roane and colleagues (Roane,2008; Roane, Lerman, & Vorndran, 2001)made a similar point regarding the use of pro-gressive ratio (PR) schedules as assessmenttools.Similar demand methods have been applied

to token reinforcement contingencies. Tan andHackenberg (2015), for example, generateddemand functions for specific and generalizedreinforcers in a token economy with pigeons.The pigeons could earn specific tokens(exchangeable for either food or water) or gen-eralized tokens (exchangeable for both food andwater). Production of all three token types(food-specific, water-specific, and generalized)declined systematically with price, consistentwith a demand analysis. Demand curves for allthree token types were well described by Hurshand Silberberg’s (2008) essential value model,which has proven successful in a range of exper-iments with a variety of species and reinforcers(Barrett & Bevins, 2012; Bentzley, Fender, &Aston-Jones, 2013; Cassidy & Dallery, 2012;Christensen, Silberberg, Hursh, Huntsberry, &Riley, 2008).In addition to the similarities, there were also

some differences between the token types. Pro-duction was less elastic for food-specific tokensthan the other two token types, but more gen-eralized than specific tokens were produced atthe lowest price. Thus, as in the Tustin (1994)study, demand elasticity was not predictable onthe basis of production at the lowest price;these two aspects of reinforcer efficacy (con-sumption at low price and sensitivity tochanges in price) must therefore be consideredseparately.In subsequent phases of the study, Tan and

Hackenberg (2015) included additional assess-ments of token reinforcer efficacy—PR break-point and preference—to assess the correlationbetween the different measures. Elasticity fromthe demand analysis was highly correlated withpreference in pairwise choices between the dif-ferent token types. That is, the less elastic the

demand for the token, the more it was pre-ferred, and vice versa. The correlation betweenpreference and PR breakpoints was weaker thanthat for demand elasticity, however. Theseresults, along with other findings with rats(Madden, Smethells, Ewan, & Hursh, 2007),suggest limits on a unitary conception of rein-forcer efficacy, based on convergence betweenthe measures. This lack of perfect correspon-dence between different measures of reinforcervalue has important implications for appliedresearch, as it is commonly assumed that a rein-forcer assessment conducted under one sched-ule arrangement will generalize to others. Theseresults, however, urge a more cautious, empiri-cal approach to determining when, and underwhat conditions, reinforcer efficacy in one situ-ation predicts efficacy in a different situation.Demand analyses can also be applied to con-

currently available reinforcers. In the finalphase of the Tan and Hackenberg (2015)study, pigeons were given three sets of pairwisechoices between the three token types: foodversus water, food versus generalized, and waterversus generalized. When food tokens werepitted against water tokens, pigeons producedand exchanged approximately 80% of thetokens for food and 20% for water. In the con-ditions with the generalized tokens, preferencesdepended on whether the alternative was foodor water tokens. When the alternative was foodtokens, pigeons produced about 80% foodtokens and about 20% generalized tokens—which were subsequently exchanged for water.When the alternative was water tokens, pigeonsproduced about 20% water tokens and about80% generalized tokens—which were subse-quently exchanged for food. In other words,the generalized tokens were spent on (function-ally substituted for) water in one context andon food in another, enabling roughly equiva-lent levels of food and water consumptionacross the different choice contexts.This type of functional substitution is key to

understanding the higher relative reinforcing


efficacy of generalized token reinforcers. Gener-alized reinforcers can be exchanged for multipleterminal reinforcers and, in this sense, serve asfunctional substitutes for those reinforcers. Theeconomic concept of substitution is simply aname given to this type of relationship. Morethan a mere synonym, however, substitution—when linked to a demand analysis—puts asharper quantitative point on the relationship,permitting one to determine not only whetherone reinforcer substitutes for another but alsothe degree to which it does so. In the Tan andHackenberg (2015) study, generalized tokenssubstituted to a higher degree for food (nearlyfull substitutes) than for water (partial substi-tutes). The type of quantification permitted byan economic analysis is a valuable tool in map-ping the continuum of generalizability dis-cussed above, and for reinforcer assessments inapplied work.

Collateral Effects of Token ReinforcementAn economic perspective calls attention to the

broader context in which specific reinforcementcontingencies are embedded, and suggests theneed to measure more global changes in the distri-bution of behavior produced by contingencychanges. In the Winkler (1970) study, describedabove, noise and aggression were monitored butnot specifically targeted for reinforcement. Reduc-tions in these two patterns of problem behavioraccompanied the changes in appropriate behaviorproduced by the token reinforcement procedure.In other words, noise and aggression were part ofa functional class of responses affected by thetoken reinforcement contingency—what aresometimes called collateral effects of reinforce-ment contingencies. Such response covariation isa critical but often overlooked dimension ofbehavior-change procedures (see Parrish, Cataldo,Kolko, Neef, & Egel, 1986), sometimes as impor-tant as the main effects of contingencies.Token economies are also ideal contexts in

which to explore collateral effects of

contingencies, including potential adverseeffects of reinforcement procedures. Fisher(1979), for example, designed an experiment toassess overjustification effects, defined as reducedresponse output following periods of extrinsicreinforcement (Greene & Lepper, 1974).According to standard interpretations of over-justificaton effects, the low responding stemsfrom reductions in intrinsic motivation pro-duced by extrinsic reinforcement, reducingbehavior to below baseline levels. Another wayto think about overjustification effects is interms of behavioral contrast: a greater-than-expected reduction in responding in transitionfrom a rich to a lean reinforcement context(Reynolds, 1961; Williams, 1983). Howeverthey are best conceptualized, such effects, ifvalid, would pose major challenges to tokenreinforcement procedures, as they usually entailexplicit tangible reinforcement for sociallydesirable behavior.In a study discussed briefly above, Fisher

(1979) compared brushing teeth across baseline(no token) and two magnitudes of reinforce-ment (one and five tokens). In the comparisonsmost critical to the overjustification hypothesis(no-token conditions before and after tokenconditions), levels of brushing teeth in no-token conditions following token reinforcementconditions did not differ from baseline levels.This finding runs counter to the overjustifica-tion hypothesis, which predicts that brushingteeth would drop below baseline levels, owingto the reduced interest in the activity. That nei-ther this study nor others (Feingold & Maho-ney, 1975; McGinnis, Friman, & Carlyon,1999; Vasta & Stirpe, 1979) have found evi-dence for detrimental effects of token reinforce-ment suggests that token economies mayinsulate behavior from such effects. Unlike thetypical laboratory-based study, in which expo-sure to conditions is extremely brief, the partic-ipants in Fisher’s study were long-timemembers of the token economy, with variablessystematically manipulated over time.


Moreover, tokens were contingent on job ortask performance—variables shown to attenuatethe so-called adverse effects of contingent rein-forcement (Eisenberger & Cameron, 1996).These results thus join with the results reportedfor reinforcement procedures more generally(Eisenberger & Cameron, 1996; Levy et al.,2016), in showing no detrimental effects ofreinforcement contingencies.

Economic ContextA token economy is a system of interlocking

contingencies, and to properly understand onepart of it (e.g., responding on a specific proce-dure), it is useful to know more about thebroader context surrounding it. One part ofthis context is the price and availability of otherreinforcing events and activities, shown in aprevious section to be powerful determinants ofdemand and preference. In this section, I willtouch on other aspects of the economic con-text. In considering the impact of these back-ground economic variables, Winkler andBurkhard (1990) distinguished between inter-nal interdependencies and external interdepen-dencies; the former are factors external to aspecific procedure but internal to the tokeneconomy (e.g., income), whereas the latter arefactors external to the token economy (e.g., thebroader national economy).Internal interdependencies. An example of an

internal interdependency is income—the num-ber of tokens available for spending. This is afactor that is external to a specific contingencybut internal to the token economy as a whole.For example, Fisher et al. (1978) showedincome to be a powerful determinant of tokenearning and spending in token economies. Forlow-income earners, the higher wages producedincreases in income, and did not adverselyaffect other token-earning behavior; theyenjoyed a higher standard of living, and workedat about the same rate. For high-incomeearners, on the other hand, the higher wages

led to reductions in other (lower-wage) token-earning behavior; they maintained a roughlysimilar standard of living, and worked less onnontarget tasks.The therapeutic implications are clear. The

success of any specific intervention (e.g., self-help skills) will depend not only on the specificprocedure (e.g., token reinforcement), but alsoon how output and consumption on that jobrelate to broader patterns of income and savingsof the individual within the token economy. Ifthe wage increases for brushing teeth result inless behavior directed toward other activities ofapplied significance (e.g., academic work), thentradeoffs will need to be made between differ-ent responses, as they relate to the overall goalsof the program. The utility of an economicanalysis is its explicit focus on the broaderchanges surrounding specific reinforcementcontingencies, bringing the tradeoffs into focus.Winkler (1972) discussed how income com-

bines with savings in determining the marginalrate of consumption—the ratio of changes inspending to changes in income. In both tokenand natural economies, the marginal rate ofconsumption is linear but with slopes typicallyless than one, indicating expenditures do notcompletely keep pace with income. Winklershowed how the marginal rate of consumptionis affected both by income and savings (tokensaccumulated for later exchange). This has clearapplied significance. Given that tokens are usu-ally earned for socially desirable behavior, waysto increase token earning (marginal consump-tion) translates directly into more effective clin-ical practices. Winkler showed how the efficacyof token procedures varied with savings (tokensin the bank). The procedures were highly effec-tive for individuals with little savings; withgreater levels of savings, however, behavior wasless well maintained by the procedures.To account for these differential effects,

Winkler (1971) introduced the concept of criti-cal range of savings (CRS)—a savings thresholdbelow which response output and consumption


will increase and above which will decrease.Manipulations that decrease savings shouldtherefore increase consumption, and this isindeed what happens (Winkler, 1971). Whenthe price of tea was doubled, more was spenton tea. This in turn reduced savings, and pro-duced increments in token-earning behavior.Savings may also be decreased in other ways(e.g., improving the quality of the backup rein-forcers). By pointing to variables that alterCRS, and thereby enhance token-earningbehavior, this approach has important appliedimplications (see below).External interdependencies. In addition to the

internal interdependencies described above,there are external interdependencies—structuralfactors external to the token economy that mayinteract with the behavior within the economy.One example of an external factor is thebroader economy in which the token system issituated. Winkler (1980), for example, com-pared two different token economies in termsof their interaction with economic factors out-side the setting. In one setting, an institutionwith long-term patients, the economy wasclosed, in the sense that tokens could only beexchanged within the setting. Here the econ-omy thrived, with high levels of token earningand spending, little savings, and high levels ofmarginal consumption. In a second setting, a98-day experimental program with volunteermarijuana smokers paid to work on a variety oftasks, the token economy was open, in that itinteracted with broader economic circum-stances. Tokens could be exchanged for goods(including marijuana) within the program, orcould be saved and exchanged for money at theend of the program. This essentially opened theeconomy to broader influences, namely, themoney available at the end of the program.Indeed, an increasing portion of earnings wentinto savings, with corresponding changes inconsumption and income.The two settings correspond to different

points on a continuum ranging from relatively

open (the experimental setting) to relativelyclosed economies (the ward setting). Closedeconomies are increasingly common in basicbehavioral economic research, including muchof the research on token reinforcementreviewed above. With a few notable exceptions(Roane et al., 2005), closed economies aresomewhat less common in applied behavioraleconomics. The study by McLaughlin andMalaby (1972), discussed earlier, provides auseful illustrative example of a relatively closedtoken economy. In this study, sixth-grade stu-dents earned tokens for assignment completionand other appropriate behavior, and lost tokensfor inappropriate behavior, with exchangeopportunities scheduled every 5-6 days over thecourse of an entire academic year. With onlynegligible investment by the teacher, the pro-gram was highly successful in establishing andmaintaining high levels of academic perfor-mance over sustained time periods. A distinc-tive feature of the token system was the use offunctionally defined activities (e.g., caring forthe class pet) that occurred only within theclassroom setting. This helped maintain a rela-tively closed and self-contained economy inwhich tokens were earned and spent onlywithin the economy. A practical benefit ofusing only reinforcers intrinsic to the settingwas that no additional financial expenditureswere needed. The token economy simplyrestructured the activities in such a way thatpreferred objects and activities were made moreusefully contingent on academic behavior.The applied implications are substantial.

First, the study utilized functionally definedreinforcers. Second, by limiting reinforcers tothose available within the classroom setting, thestudy essentially closed the economy, eliminat-ing interactions with extraclassroom reinforcers.Both of these factors enhance the value of thereinforcers and the efficacy of the token econ-omy as a whole. In holding constant factorsexternal to the classroom economy, such class-room token systems provide a behaviorally rich


and analytically tractable environment in whichto explore basic mechanisms of token-economicbehavior: token earning, saving, and spendingin a controlled and relatively closed economy.These types of settings provide applied labora-tories for studying basic processes.

TRANSLATIONAL IMPLICATIONS

The focus of the paper to this point has beenon reviewing what is known about token rein-forcement systems from laboratory and appliedperspectives. Although some applied implica-tions have been mentioned along the way, theprimary focus has been on identifying gaps inthe literature and promising areas for futureresearch. As an overall translational program ofresearch, consisting of (a) basic research,(b) applied research, and (c) practice, theemphasis thus far has been on (a) and (b)—thescience of token reinforcement. In this section,I emphasize part (c), the practical implementa-tion of token reinforcement procedures in thelight of the science.

AcquisitionResearch reviewed above shows how much

we still have to learn about the specific condi-tions responsible for establishing the variousfunctions of tokens. Nevertheless, a few specificsuggestions can be made. As with other condi-tioned reinforcers, tokens should occur in a richreinforcement context, closely coupled withother already-effective reinforcers. Simply pair-ing the tokens with terminal reinforcers may besufficient in some cases to establish conditionedreinforcement value, but there are reasons tofavor response-contingent pairings of tokensand terminal reinforcers, in which an exchangeresponse is required to access terminal rein-forcers. First, response-contingent pairings(i.e., training a discriminative operant) are moreeffective than simply pairing the tokens withthe terminal reinforcers. Second, there are prac-tical benefits as well: Training an exchange

response and bringing it under stimulus controlof a token is the initial step in a more generaltoken reinforcement program. It is advisable inthe early stages to use FR 1 schedules, when-ever possible, adding links via backward chain-ing until the entire unit of token productionand exchange is established.

MaintenanceOnce token-reinforced behavior has been

established, an important applied concern iswith the maintenance of behavior—as exchangeperiods are often temporally removed from theperiods during which tokens are earned.Indeed, increasing this delay between tokendelivery and token exchange is frequently a goalof token reinforcement programs. Examiningways to sustain behavior in the face of suchchanges thus holds important practical value.What variables should be taken into account inthe maintenance of such behavior?When viewed as an extended chained sched-

ule, token reinforcement procedures consist ofa sequence of interconnected schedules. Forexample, suppose one were to arrange a tokenprogram with the following three componentschedules: (a) FR 5 token production (five aca-demic problems completed produce a star),(b) FR 10 exchange production schedule(10 stars produce an exchange period), and(c) FR 1 exchange schedule (handing over thefull token board to a teacher produces 30-saccess to a preferred toy train). Suppose onewished to maintain responding while thinningthe schedule; which schedule (token productionor exchange production) should be changed?Research on schedule effects suggests it will bemost effective to raise exchange-productionrequirements and keep token-productionrequirements low, rather than the reverse. Thiswas indeed the strategy employed by Tarboxet al. (2006), and perhaps the reason they sawrelatively small decrements in responding in theface of large increases in FR exchange price.


This ability to sustain responding in the faceof substantial increases in the FR value may bedue, in part, to a unique feature of token rein-forcement schedules: The number of tokensproduced is correlated with the magnitude ofthe terminal reinforcer. Each increment in theFR exchange-production price means that moretokens will be available to exchange for the ter-minal reinforcer. Hence, increases in theexchange-production price produce correspond-ing increases in terminal reinforcer magnitude.In the above example, assuming each tokencould be cashed in for 30-s access to the train,increasing from FR 10 to FR 20 would alsoincrease the toy access time from 300 to 600 s.As a result, despite a two-fold increase in FR,the overall unit price (responses per terminalreinforcer duration) remains constant. A two-fold increase in the token-production FR, onthe other hand, would double the unit price,because each increment in the FR size does notproduce corresponding increases in terminalreinforcer magnitude. Thus, incrementally rais-ing the exchange-production schedule is likelyto sustain higher levels of responding thancomparable increases in the token-productionschedule.Not all applied token programs are arranged

in such a way that terminal reinforcer amountis correlated with number of tokens earned,however. In some, tokens signal proximity tothe exchange but not amount of reinforcementin exchange. This is the case, for example,when the terminal reinforcer magnitude is con-stant, irrespective of the numbers of tokens pre-sent at exchange. In the above example,suppose the terminal reinforcer is 30-s toyaccess whether the exchange-production sched-ule is FR 10 or FR 20. Unlike standard tokenschedules with correlated reinforcer magnitude,this type of token schedule is more akin to atypical extended chained schedule, with a singleterminal reinforcer.Absent the correlation between token num-

ber and terminal reinforcer amount, the

methodological advantages of manipulatingthe exchange-production schedule diminishsomewhat. Even so, however, there are addi-tional reasons to recommend changing theexchange-production over token-productionschedule. For one thing, exchange-productionmanipulations involve relatively more tokensthan otherwise comparable token-productionmanipulations. That is, target responses con-tinue to produce tokens when the exchange-production ratio is increased, but not whenthe token-production ratio is increased. Asconditioned reinforcers, the added tokensalone should facilitate responding, even withcomparable unit prices and reinforcer magni-tude. Thus, if an applied goal is to sustainresponse output as the schedule is thinned, itwill be more efficient to increase the workrequirements for exchange production(e.g., number of tokens needed to access theexchange) than for the token productionschedule (e.g., number of responses needed toproduce a token).

Discriminative ControlOne exception to this general rule is if the

exchange-production FR is too large, and thetokens in the early links exert strong discrimi-native effects. For example, research reviewedearlier showed that weak early-link respondingis due to the discriminative functions of tokenstemporally removed from exchange periods andterminal reinforcers. As additional tokens areearned, exchanges and terminal reinforcersgrow nearer. As a result, the number of earnedtokens is correlated with delays to the exchangeand terminal reinforcers. In the above example,with the FR 10 exchange-production schedule,relatively large numbers of tokens (e.g., 8,9, 10) signal relatively short exchange delays,whereas relatively small numbers of tokens(e.g., 0, 1, 2) signal relatively long exchangedelays. These discriminative functions effec-tively weaken early-link responding.


One method to strengthen such early-linkbehavior is to reverse the usual stimulus-controlrelations, such that late-link stimuli becomeearly-link stimuli, and vice versa (e.g., begin thetoken-production link in the presence of10 tokens). If the tokens are serving discrimina-tive functions, this manipulation might work inbringing weak early-link responding under thecontrol of stimuli that normally accompanystronger later-link behavior. Such effects arelikely to be temporary, however, as new stimu-lus functions based on temporal proximity toterminal reinforcers would eventually beestablished.Another method for increasing weak behav-

ior in the early links of token reinforcementschedules is to use variable schedules, whichhave been shown to produce relatively higherresponse rates than otherwise comparable fixedschedules (as reviewed above). In terms of theabove example, using a VR 10 exchange-production schedule (e.g., where exchangeperiods might occur after 2, 6, 10, 14, or18 tokens) should sustain higher response out-put than an FR 10 schedule, owing to thestronger early-link behavior in the VR schedule.Thus, putting this together with the above,when arranging token reinforcement schedulesfor maintenance, one should, whenever possi-ble, (a) manipulate exchange-production ratherthan token-production schedules, and (b) usevariable rather than fixed schedules. By point-ing to variables that strengthen respondingwhen it is at its weakest (in periods remotefrom exchange), these recommendations sug-gest ways to increase rates of token-reinforcedbehavior—an important applied goal of mosttoken programs.

Aversive ControlJust as tokens can be used to increase behav-

ior, token losses can be used to decrease behav-ior. How might such research be translatedinto practical recommendations? Returning to

the above example, suppose that, in addition tothe token-gain schedule (FR 5 token produc-tion, FR 10 exchange production) arranged foracademic behavior, a token-loss contingency isarranged for challenging behavior. What typeof token-loss procedure should be used?In light of the punishment magnitude effects

described above, it might seem appropriate torecommend losses as large as possible. Suchconsiderations must also take into accounttoken-gain contingencies, however. With ratio-based token-gain schedules, as in our example,removing a star from the token board is essen-tially a return to an earlier link in the chain.This results in longer delays to the exchangeand toy access, and thus may weaken behavioron its own. For this reason, ratio-based sched-ules of token gain may not be the best choicein sustaining behavior, especially given thatsuch token-earning behavior is usually desirable(in this case, academic work). Suppose insteadthat a time-based schedule (e.g., FI, VI) is used,in which an exchange is produced following thefirst token after a period of time. These sched-ules are less sensitive to punishment effects,and may therefore provide a better way to sus-tain the target academic responding via tokengain while punishing challenging behavior viatoken loss. On the other hand, all else equal,ratio schedules generally produce higher ratesthan interval schedules. The particular arrange-ment of token-gain and token-loss contingen-cies may depend on tradeoffs between thesedifferent goals, including the relative impor-tance of the academic and challenging behavior.It may also depend on practical considerations(e.g., time-based schedules may be more diffi-cult to implement than response-based sched-ules). However these tradeoffs are resolved, theneed for balancing losses against gains is espe-cially crucial in negative punishment proce-dures (including both token loss and timeout),in which the reinforcing value of token gainsand the aversiveness of token losses areinterdependent.


Another variable contributing to the aver-siveness of the loss is the manner in which thetokens are presented. Research reviewed abovesuggests that earned tokens may be valuedmore than otherwise equivalent freely deliveredtokens. And just as access to earned tokens maybe more reinforcing, removal of earned tokensmay be more aversive. In our example, it mightbe more effective to punish the challengingbehavior by removing earned stars from thetoken board than by removing freely providedstars.A final consideration in the decision about

whether to use token-loss procedures is theirpotential for negative collateral effects of aver-sive control (e.g., escape, social disruption),seen with other types of aversive stimuli.Although some such effects have occasionallybeen reported, their overall incidence is low.The majority of studies have reported no nega-tive side effects, some have reported positiveside effects (e.g., increasing academic behavior),whereas others have reported preference for lossover gain procedures.On the whole, token-loss procedures are

both effective and safe; when designed andimplemented in light of what is known aboutpunishment effects more generally, token-lossprocedures can be valuable behavior-changetools. When such procedures are used, considera few general points. First, with respect to theloss schedule, losses should be immediate, fre-quent, and of moderate to high magnitude. Inour example, all stars might be removed fromthe board (return to the initial link) each timethe challenging behavior occurs. Second, withrespect to the token-gain schedules, whethertime-based or response-based schedules are useddepends on how the token-reinforced behavior(academic work, in our example) interacts withthe token-loss schedule. If the loss schedule isweakening token-earning behavior due to thedelays to the exchange and terminal reinforcer(toy access) brought on by the loss schedule,then time-based schedules should be used.

Otherwise, ratio-based schedules would befavored, as they tend to produce higherresponse rates than comparable time-basedschedules. And third, with respect to bothschedules, both gains and losses should beresponse-dependent. This will enhance boththe reinforcing value of the tokens and theaversiveness of token loss.

Economic VariablesSeveral of the issues raised in the applied

behavioral economics section have importanttranslational implications. To begin with, thedemand analysis provides a valuable supple-ment to existing reinforcer assessment methods,suggesting quantitatively precise ways to assessrelative reinforcer efficacy in ways most usefulto treatment. It is common, for example, inapplied studies to identify a reinforcer based onthe results of a preference assessment under FR1 schedules. Work described above, however,shows how relative reinforcer efficacy can andoften does change with schedule variables(e.g., FR price). If limited to FR 1 schedules,these differences in reinforcer efficacy wouldnot be apparent. The typical FR 1-type ofassessment may be sufficient if the interventionschedules are similar (i.e., FR 1 or small FRrequirements). If, on the other hand, onewishes to use leaner and more intermittentschedules for behavior maintenance, it may beprudent to conduct a demand analysis, asses-sing preference under a wider range of schedulevalues that map better onto maintenance con-ditions. A demand analysis expands the rangeof schedule values used in assessment, provid-ing a more comprehensive and nuanced pictureof reinforcer efficacy, and for these reasons maybe more predictive of treatment success.Should one choose to conduct a demand

analysis, what type of procedure should beused? The typical demand analysis would entailmeasuring response output and terminal rein-forcers earned as the (token or exchange-


production) price was increased incrementallyacross sessions (e.g., FR 1, FR 5, FR 10, FR20). An alternative method, gaining currency inrecent years, is a PR schedule, in which theprice changes within a session, usually aftereach reinforcer (Roane, 2008). The same spanof ratios can be programmed, but on a com-pressed time scale. This has clear practicaladvantages over typical methods, not the leastof which is the time needed to generate thedemand function (the main practical disadvan-tage of the standard method). Unfortunately,research has shown that reinforcing efficacy, asmeasured by demand and PR methods, are notalways highly correlated, and thus may be mea-suring different aspects of reinforcer value.More specifically, price sensitivity from ademand analysis better predicts preferencebetween reinforcers than does PR breakpoint(Madden et al., 2007; Tan & Hackenberg,2015). The choice of measure will depend,then, on the treatment goals. If the goal is topredict preferences between different rein-forcers, then the demand analysis may be thebest choice; if, on the other hand, the goal is topredict broad differences in response output ina single-response situation, then the quicker PRmethod may suffice.A major insight from economic analyses is

that the value of one reinforcer depends notonly on its own price, but also on the price andavailability of other reinforcers. This is espe-cially pertinent to applied settings, in which amyriad of reinforcers are generally available,and needs to be considered in designing treat-ment programs. Returning to our prior exam-ple, suppose that the token-based academicprogram is embedded in a larger token econ-omy, in which tokens are produced and lost foradaptive and challenging behavior, respectively.The success of the program will depend, inpart, on the degree to which the academicbehavior and these other behavior patterns pro-duce functionally similar reinforcers; or in eco-nomic terms, the degree to which they serve as

substitutes, one for the other. Most token pro-grams have this type of functional substitutionbuilt into them, in that the reinforcer for allresponses is the same: functionally equivalentgeneralized tokens exchangeable for a widearray of reinforcers.Considering the broader context surrounding

specific token programs can call attention tolarger classes of variables to manipulate inapplied settings. One such variable is income—the number of tokens available to spend.Income is a joint function of the price andavailability of tokens for adaptive behavior(e.g., academic work) and the number oftokens already accumulated in the bank(i.e., savings, in economic terms). Suppose thatin our example, stars can be saved acrossexchange periods, with unspent stars remainingin a token bank account. Because the stars inthe bank and the stars earnable on the task arefunctionally interchangeable, a large savingsaccount is essentially a supply of substitutablereinforcers, which may then weaken token-earning behavior. A robust token economy isone in which savings are low or nonexistent;this promotes the high rates of token earningand spending critical to the overall success ofthe program.One way to encourage spending is to

improve the quality of the terminal reinforcers(e.g., including a wider range of preferred rein-forcers). Another is to arrange mandatoryexchange periods, either after each token andexchange-production cycle (as commonly done)or periodically (e.g., at the end of each sessionor at the end of each day), thereby emptyingthe token bank account and maintaining amore robust economy. Either way, methodsthat shift the income toward the present taskand away from the savings account will benefitthe economy as a whole.A final variable to consider in designing

effective token programs is the overall eco-nomic context; more specifically, the availabil-ity of substitutable reinforcers outside the


token-earning context. Suppose that in ourexample, toy trains (the preferred terminal rein-forcer) were abundantly available before andafter the token-based academic sessions(in economic terms, an open economy). Thiswould likely weaken academic behavior, as thiswould be only one of many ways to access toytrains. Instead, restricting access to trains out-side the academic sessions (in economic terms,a closed economy), would likely enhance thevalue of the tokens and the academic respond-ing on which they depend. In general, thegreater the restriction placed on access to termi-nal reinforcers in the setting as a whole(i.e., the more closed the economy), the moreeffective the token economy.

SUMMARY AND CONCLUSIONS

For a variety of reasons, laboratory andapplied research on token reinforcement havedrifted apart over the years. This split betweenthe laboratory and applied work on token sys-tems has had unfortunate consequences for thefield. Although token reinforcement systemshave generally been quite successful in theapplied realm, it is often unclear what variablesare responsible for their effectiveness; they arerarely based on an understanding of the basicprinciples involved. We have known for overfive decades that token reinforcement proce-dures work; we need to know more about whythey work. We need a better analysis of thecritical variables operating in token economies.A key premise of the paper is that a better

understanding of the basic principles operatingin token systems will enhance application. Totake just one example from a prior section,understanding that token-reinforced behavior isbetter maintained by variable than by fixedschedules (e.g., Ferster & Skinner, 1957) leadsto more effective ways of scheduling exchangeperiods in classroom token economies(e.g., McLaughlin & Malaby, 1972). Knowl-edge of basic reinforcement schedules, derived

largely from laboratory research with nonhu-man animals, translates into more effectiveapplications with students in a classroomenvironment.This is but one of numerous examples pro-

vided in the present paper where appliedresearch serves translational research in thisway—mainly as a vehicle for extrapolation oflaboratory-based principles to an applied prob-lem. This is the prototypical model of transla-tional research, in which principles uncovered inthe laboratory move outward to application. Inthe realm of token reinforcement, however,applied research can play a more active role intranslational work; it can drive the analysis for-ward, opening new research vistas. For example,as discussed earlier, exploring the continuum ofgeneralizability requires a healthy range of termi-nal reinforcers. These conditions are common-place in applied token environments (e.g., in themenu of terminal reinforcers for which tokenscan be exchanged), and therefore provide anideal context for an experimental analysis of acritical function of token reinforcers—theirdegree of generalizability—an analysis that isbest suited to an applied setting.I have suggested the utility of a theoretical

framework that combines basic behavior-analytic and behavioral-economic principles.This illustrates another aspect of translationalresearch—the intertranslation of concepts andmethods from related disciplines. Although thedirect translation between certain concepts(e.g., FR size in behavior analysis and price inbehavioral economics) may lead some to ques-tion the utility of consulting two different the-oretical approaches, my aim here was to showways in which the disciplines complement oneanother by asking similar questions butemphasizing different experimental andexplanatory variables. For instance, behavior-analytic methods have traditionally focused onqualitatively similar reinforcers, whereasbehavioral-economic methods have focused onqualitatively different reinforcers. In situations


with a variety of reinforcers, as in the case ofgeneralized token reinforcers, well-establishedbehavioral-economic methods (e.g., substitu-tion effects) may therefore serve as a valuablesupplement to traditional behavior-analyticmethods.Combining these behavioral economic

methods with applied token research—what Icall applied behavioral economics—would simul-taneously advance the science and the applica-tion of token reinforcement. In the case ofgeneralized reinforcement, elasticity measureswould advance the science by providingdetailed quantitative information on interac-tions between qualitatively different reinforcers;such information would be extremely valuablein designing more effective applied contingen-cies. And because the analyses would take placein the applied context itself, the results wouldbe maximally generalizable.In this and other domains, there are impor-

tant questions applied research is actually bettersuited than laboratory research to address.Indeed, in this domain, clear distinctionsbetween laboratory and applied research are dif-ficult to make, as the analysis takes place in thevery context in which it is applied. This type ofbidirectional interplay between analysis andapplication is one dimension of use-inspiredresearch (Critchfield, 2011; after Stokes, 1997),in which applied problems provide the theoreti-cal context for basic research.A major aim of this paper has been to stimu-

late such use-inspired research, in which tokeneconomies serve as applied laboratories foraddressing fundamental questions, and theoreti-cal questions about basic principles go hand inhand with practical questions about how bestto implement them. Figuring out how thingswork in the course of solving important socialproblems is a defining feature of applied behav-ior analysis (Baer, Wolf, & Risley, 1968), andwhen combined with the tools of behavioraleconomics, offers fresh new insights into thescience and application of token reinforcement.

REFERENCES

Andrade, L. F., & Hackenberg, T. D. (2017). Substitu-tion effects in a generalized token economy withpigeons. Journal of the Experimental Analysis ofBehavior, 107, 123-135. https://doi.org/10.1002/jeab.231

Ayllon, T., & Azrin, N. H. (1965). The measurementand reinforcement of behavior of psychotics. Journalof the Experimental Analysis of Behavior, 8, 357-383.https://doi.org/10.1901/jeab.1965.8-357

Ayllon, T., & Azrin, N. H. (1968). The token economy: Amotivational system for therapy and rehabilitation.New York, NY: Appleton-Century-Crofts.

Azrin, N. H. (1960). Effects of punishment intensityunder variable-interval reinforcement. Journal of theExperimental Analysis of Behavior, 3, 123-142.https://doi.org/10.1901/jeab.1960.3-123

Azrin, N. H., & Holz, W. C. (1966). Punishment. InW. K. Honig (Ed.), Operant behavior: Areas ofresearch and application (pp. 380-437). New York,NY: Appleton-Century-Crofts.

Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Somecurrent dimensions of applied behavior analysis. Jour-nal of Applied Behavior Analysis, 1, 91-97. https://doi.org/10.1901/jaba.1968.1-91

Barrett, S. T., & Bevins, R. A. (2012). A quantitativeanalysis of the reward-enhancing effects of nicotineusing reinforcer demand. Behavioural Pharmacology,23, 781-789. https://doi.org/10.1097/fbp.0b013e32835a38d9

Battalio, R. C., Kagel, J. H., Winkler, R. C.,Fisher, E. G., Jr., Basman, R. L., & Krasner, L.(1974). An experimental investigation of consumerbehavior in a controlled environment. Journal of Con-sumer Research, 1, 52-60. https://doi.org/10.1086/208591

Becraft, J. L., & Rolider, N. U. (2015). Reinforcer varia-tion in a token economy. Behavioral Interventions,30, 157-165. https://doi.org/10.1002/bin.1401

Bentzley, B. S., Fender, K. M., & Aston-Jones, G.(2013). The behavioral economics of drug self-administration: A review and new analytical approachfor within-session procedures. Psychopharmacology,226, 113-125. https://doi.org/10.1007/s00213-012-2899-2

Bickel, W. K., & Vuchinich, R. (2000). Reframing healthbehavior change with behavioral economics. Mahwah,NJ: Lawrence Erlbaum & Associates.

Boren, J. J., & Colman, A. D. (1970). Some experimentson reinforcement principles within a psychiatric wardfor delinquent soldiers. Journal of Applied BehaviorAnalysis, 3, 29-37. https://doi.org/10.1901/jaba.1970.3-29

Brady, J. V., Kelly, D., & Plumlee, L. (1969). Autonomicand behavioral responses of the rhesus monkey tobehavioral conditioning. Annals of the New York


https://doi.org/10.1002/jeab.231


https://doi.org/10.1901/jeab.1965.8-357


https://doi.org/10.1901/jaba.1968.1-91


https://doi.org/10.1097/fbp.0b013e32835a38d9

https://doi.org/10.1097/fbp.0b013e32835a38d9

https://doi.org/10.1086/208591

https://doi.org/10.1086/208591

https://doi.org/10.1002/bin.1401

https://doi.org/10.1007/s00213-012-2899-2

https://doi.org/10.1007/s00213-012-2899-2



Academy of Sciences, 15, 959-975. https://doi.org/10.1111/j.1749-6632.1969.tb12991.x

Brandt, A. E., Sztykiel, H., & Pietras, C. J. (2013). Labo-ratory simulated gambling: Risk varies acrossparticipant-stake procedure. The Journal of GeneralPsychology, 140, 130-143. https://doi.org/10.1080/00221309.2013.776509

Bullock, C. E., & Hackenberg, T. D. (2006). Second-order schedules of token reinforcement with pigeons:Implications for unit price. Journal of the Experimen-tal Analysis of Behavior, 85, 95-106. https://doi.org/10.1901/jeab.2006.116-04

Bullock, C. E., and Hackenberg, T. D. (2015). The sev-eral roles of stimuli in token reinforcement. Journal ofthe Experimental Analysis of Behavior, 103, 269-287.https://doi.org/10.1037/e625362007-001

Burchard, J. D., & Barrera, F. (1972). An analysis oftimeout and response cost in a programmed environ-ment. Journal of Applied Behavior Analysis, 5, 271-282. https://doi.org/10.1901/jaba.1972.5-271

Carnett, A., Raulston, T., Lang, R., Tostanoski, A.,Lee, A., Sigafoos, J., & Machalicek, W. (2014).Effects of a perseverative interest-based token econ-omy on challenging and on-task behavior in a childwith autism. Journal of Behavioral Education, 23,368-377. https://doi.org/10.1007/s10864-014-9195-7

Cassidy, R. N., & Dallery, J. (2012). Effects of economytype and nicotine on the essential value of food inrats. Journal of the Experimental Analysis of Behavior,97, 183-202. https://doi.org/10.1901/jeab.2012.97-183

Charlop-Christy, M.H., & Haymes, L. K. (1998). Usingobjects of obsession as token reinforcers for childrenwith autism. Journal of Autism and DevelopmentalDisorders, 28, 189-198. https://doi.org/10.1023/a:1026061220171

Christensen, C. J., Silberberg, A., Hursh, S. R.,Huntsberry, M. E., & Riley, A. L. (2008). Essentialvalue of cocaine and food in rats: tests of the expo-nential model of demand. Psychopharmacology, 198,221-229. https://doi.org/10.1007/s00213-008-1120-0

Cowles, J. T. (1937). Food-tokens as incentives for learn-ing by chimpanzees. Comparative PsychologicalMonographs, 12, 1-96. https://doi.org/10.1037/14268-000

Critchfield, T. S. (2011). Translational contributions ofthe experimental analysis of behavior. The BehaviorAnalyst, 34, 3-17. https://doi.org/10.1007/bf03392227

Critchfield, T. S., Paletz, E. M., MacAleese, K. R., &Newland, M. C. (2003). Punishment in humanchoice: Direct or competitive suppression? Journal ofthe Experimental Analysis of Behavior, 80, 1-27.https://doi.org/10.1901/jeab.2003.80-1

Dallery, J., Silverman, K., Chutuape, M. A.,Bigelow, G. E., & Stitzer, M. L. (2001). Voucher-

based reinforcement of opiate plus cocaine abstinencein treatment-resistant methadone patients: effects ofreinforcer magnitude. Experimental and ClinicalPsychopharmacology, 9, 317-325. https://doi.org/10.1037//1064-1297.9.3.317

DeFulio, A., Yankelevitz, R. L., Bullock, C. E., &Hackenberg, T. D. (2014). Generalized conditionedreinforcement with pigeons in a token economy.Journal of the Experimental Analysis of Behavior, 102,26-46. https://doi.org/10.1002/jeab.94

DeLeon, I. G., Iwata, B. A., Loh, H. L., &Worsdell, A. S. (1997). Emergence of reinforcer pref-erence as a function of schedule requirements andstimulus similarity. Journal of Applied BehaviorAnalysis, 30, 439-449. https://doi.org/10.1901/jaba.1997.30-439

De Luca, R. V., & Holborn, S. W. (1990). Effects offixed-interval and fixed-ratio schedules of token rein-forcement on exercise in obese and non-obese boys.Psychological Record, 40, 67-82.

De Luca, R. V., & Holborn, S. W. (1992). Effects of avariable-ratio reinforcement schedule with changingcriteria on exercise in obese and non-obese boys.Journal of Applied Behavior Analysis, 25, 671-679.https://doi.org/10.1901/jaba.1992.25-671

Doll, C., McLaughlin, T. F., & Barretto, A. (2013). Thetoken economy: A recent review and evaluation.International Journal of Basic and Applied Science, 2,131-149.

Donaldson, J. M., DeLeon, I. G., Fisher, A. B., &Kahng, S. W. (2014). Effects of and preference forconditions of token earn versus token loss. Journal ofApplied Behavior Analysis, 47, 1-12. https://doi.org/10.1002/jaba.135

Eisenberger, R., & Cameron, J. (1996). Detrimentaleffects of reward: Reality or myth? AmericanPsychologist, 51, 1153-1166. https://doi.org/10.1037//0003-066x.51.11.1153

Emurian, H. H., Emurian, C. S., & Brady, J. V. (1985).Positive and negative reinforcement effects on behav-ior in a three-person microsociety. Journal of theExperimental Analysis of Behavior, 44, 157-174.https://doi.org/10.1901/jeab.1985.44-157

Fantino, E. (1977). Conditioned reinforcement: Choiceand information. In W. K. Honig &J. E. R. Staddon (Eds), Handbook of operant behavior(pp. 313-339). Englewood Cliffs, NJ: Prentice Hall.

Feingold, B. D., & Mahoney, M. J. (1975). Reinforce-ment effects on intrinsic interest: Undermining theoverjustification hypothesis. Behavior Therapy, 6,367-377. https://doi.org/10.1016/s0005-7894(75)80111-0

Felton, M., & Lyon, D. O. (1966). The post-reinforcement pause. Journal of the ExperimentalAnalysis of Behavior, 9, 131-134. https://doi.org/10.1901/jeab.1966.9-131

Ferster, C. B., & Skinner, B. F. (1957). Schedules of rein-forcement. New York, NY: Appleton-Century-Crofts.


https://doi.org/10.1111/j.1749-6632.1969.tb12991.x

https://doi.org/10.1111/j.1749-6632.1969.tb12991.x

https://doi.org/10.1080/00221309.2013.776509

https://doi.org/10.1080/00221309.2013.776509



https://doi.org/10.1037/e625362007-001


https://doi.org/10.1007/s10864-014-9195-7

https://doi.org/10.1007/s10864-014-9195-7



https://doi.org/10.1023/a:1026061220171

https://doi.org/10.1023/a:1026061220171

https://doi.org/10.1007/s00213-008-1120-0

https://doi.org/10.1007/s00213-008-1120-0

https://doi.org/10.1037/14268-000

https://doi.org/10.1037/14268-000

https://doi.org/10.1007/bf03392227

https://doi.org/10.1007/bf03392227


https://doi.org/10.1037//1064-1297.9.3.317

https://doi.org/10.1037//1064-1297.9.3.317





https://doi.org/10.1002/jaba.135


https://doi.org/10.1037//0003-066x.51.11.1153

https://doi.org/10.1037//0003-066x.51.11.1153


https://doi.org/10.1016/s0005-7894(75)80111-0

https://doi.org/10.1016/s0005-7894(75)80111-0



Findley, J. D. (1958). Preference and switching underconcurrent scheduling. Journal of the ExperimentalAnalysis of Behavior, 1, 123-144. https://doi.org/10.1901/jeab.1958.1-123

Fisher, E. B., Jr. (1979). Overjustification effects in tokeneconomies. Journal of Applied Behavior Analysis, 12,407-415. https://doi.org/10.1901/jaba.1979.12-407

Fisher, E. B., Jr., Winkler, R. C., Krasner, L., Kagel, J.,Battalio, R. C., & Basmann, R. L. (1978). Economicperspectives in behavior therapy: Complex interde-pendencies in token economies. Behavior Therapy, 9,391-403. https://doi.org/10.1016/s0005-7894(78)80083-5

Foster, T. A., Hackenberg, T. D., & Vaidya, M. (2001).Second-order schedules of token reinforcement withpigeons: Effects of fixed- and variable-ratio exchangeschedules. Journal of the Experimental Analysis ofBehavior, 76, 159-178. https://doi.org/10.1901/jeab.2001.76-159

Fox, A. E., & Pietras, C. J. (2013). The effects ofresponse-cost punishment on instructional control ina choice task. Journal of the Experimental Analysis ofBehavior, 99, 346-361. https://doi.org/10.1002/jeab.20

Francesco, M. T., Madden, G. J., & Borrero, J. (2009).Behavioral economics: Principles, procedures, andutility for applied behavior analysis. The BehaviorAnalyst Today, 10, 277-293. https://doi.org/10.1037/h0100671

Galizio. M. (1979). Contingency-shaped and rule-governed behavior: Instructional control of humanloss-avoidance. Journal of the Experimental Analysis ofBehavior, 31, 53-70. https://doi.org/10.1901/jeab.1979.31-53

Glynn, S. M. (1990). Token economy approaches for psy-chiatric patients: progress and pitfalls over 25 years.Behavior Modification, 14, 383-407. https://doi.org/10.1177/01454455900144002

Gollub, L. (1977). Conditioned reinforcement: Scheduleeffects. In W. K. Honig & J. E. R. Staddon (Eds),Handbook of operant behavior (pp. 288-312). Engle-wood Cliffs, NJ: Prentice Hall.

Greene, D., & Lepper, M. R. (1974). Effects of extrinsicrewards on children’s subsequent intrinsic interest.Child Development, 45, 1141-1145. https://doi.org/10.2307/1128110

Hackenberg, T. D. (2009). Token reinforcement: Areview and analysis. Journal of the Experimental Analy-sis of Behavior, 91, 257-286. https://doi.org/10.1901/jeab.2009.91-257

Hackenberg, T. D., & Joker, V. R. (1994). Instructionalversus schedule control of humans’ choices in situa-tions of diminishing returns. Journal of the Experi-mental Analysis of Behavior, 62, 367-383. https://doi.org/10.1901/jeab.1994.62-367

Hanley, G. P., Piazza, C. C., Fisher, W. W., &Maglieri, K. A. (2005). On the effectiveness of andpreference for punishment and extinction

components of function-based interventions. Journalof Applied Behavior Analysis, 38, 51-65. https://doi.org/10.1901/jaba.2005.6-04

Higgins, S. T., Heil, S. H., & Lussier, J. P. (2004). Clini-cal implications of reinforcement as a determinant ofsubstance abuse disorders. Annual Review ofPsychology, 55, 431-461. https://doi.org/10.1146/annurev.psych.55.090902.142033

Holth, P., Vandbakk, M., Finstad, J., Gronnerud, E. M., &Sorenson, J. M. A. (2009). An operant analysis of jointattention and the establishment of conditioned socialreinforcers. European Journal of Behavior Analysis,10, 143-158. https://doi.org/10.1080/15021149.2009.11434315

Hursh, S. R., & Silberberg, A. (2008). Economic demandand essential value. Psychological Review, 115,186-198.

Hutchinson, R. R. (1977). By-products of aversive con-trol. In W. K. Honig & J. E. R. Staddon (Eds),Handbook of operant behavior (pp. 415-431). Engle-wood Cliffs, NJ: Prentice Hall.

Isaksen, J., & Holth, P. (2009). An operant approach toteaching joint attention skills to children with autism.Behavioral Interventions, 24, 215-236. https://doi.org/10.1002/bin.292

Iwata, B. A., & Bailey, J. S. (1974). Reward versus costtoken systems: An analysis of the effects on studentand teacher. Journal of Applied Behavior Analysis, 7,567-576. https://doi.org/10.1901/jaba.1974.7-567

Jwaideh, A. R. (1973). Responding under chained andtandem fixed-ratio schedules. Journal of the Experi-mental Analysis of Behavior, 19, 259-267. https://doi.org/10.1901/jeab.1973.19-259

Kagel, J. H. (1972). Token economies and experimentaleconomics. Journal of Political Economy, 80, 779-785.https://doi.org/10.1086/259926

Kagel, J. H., & Winkler, R. C. (1972). Behavioral eco-nomics: Areas of cooperative research between eco-nomics and applied behavior analysis. Journal ofApplied Behavior Analysis, 5, 335-342. https://doi.org/10.1901/jaba.1972.5-335

Kahneman, D., Knetsch, J. L, & Thaler, R. H. (1990).Experimental studies of the endowment effect andthe Coase theorem. Journal of Political Economy, 98,1325-1348. https://doi.org/10.1086/261737

Kahneman, D., & Tversky, A. (1979). Prospect theory:An analysis of decisions under risk. Econometrica, 47,263-291. https://doi.org/10.2307/1914185

Kaufman, K. F., & O’Leary, K. D. (1972). Reward, cost,and self-evaluation procedures for disruptive adoles-cents in a psychiatric hospital school. Journal ofApplied Behavior Analysis, 5, 293-309. https://doi.org/10.1901/jaba.1972.5-293

Kazdin, A. E. (1982). The token economy: A decadelater. Journal of Applied Behavior Analysis, 15, 431-445. https://doi.org/10.1901/jaba.1982.15-431

Kazdin, A. E., & Bootzin, R. R. (1972). The token econ-omy: An evaluative review. Journal of Applied





https://doi.org/10.1016/s0005-7894(78)80083-5

https://doi.org/10.1016/s0005-7894(78)80083-5





https://doi.org/10.1037/h0100671

https://doi.org/10.1037/h0100671



https://doi.org/10.1177/01454455900144002

https://doi.org/10.1177/01454455900144002

https://doi.org/10.2307/1128110

https://doi.org/10.2307/1128110







https://doi.org/10.1146/annurev.psych.55.090902.142033

https://doi.org/10.1146/annurev.psych.55.090902.142033

https://doi.org/10.1080/15021149.2009.11434315

https://doi.org/10.1080/15021149.2009.11434315






https://doi.org/10.1086/259926



https://doi.org/10.1086/261737

https://doi.org/10.2307/1914185




Behavior Analysis, 5, 343-372. https://doi.org/10.1901/jaba.1972.5-343

Kelleher, R. T. (1956). Intermittent conditioned rein-forcement in chimpanzees. Science, 124, 679-680.https://doi.org/10.1126/science.124.3224.679

Kelleher, R. T. (1958). Fixed-ratio schedules of condi-tioned reinforcement with chimpanzees. Journal ofthe Experimental Analysis of Behavior, 1, 281-289.https://doi.org/10.1901/jeab.1958.1-281

Leon, Y., Borrero, J. C., & DeLeon, I. G. (2016). Para-metric analysis of delayed primary and conditionedreinforcers. Journal of Applied Behavior Analysis, 49,1-17. https://doi.org/10.1002/jaba.311

Levitt, S. D., List, J. A., Neckermann, S., & Sadoff, S.(2012). The behavioralist goes to school: Leveragingbehavioral economics to improve educational perfor-mance. ZEW Discussion Papers, No. 12-038, http://nbn-resolving.de/urn:nbn:de:bsz:180-madoc-323358

Levy, A., DeLeon, I. G., Martinez, C. K., Fernandez, N.,Gage, N. A., Sigurdsson, S. Ó., & Frank-Crawford, M. A. (2016). A quantitative review ofoverjustification effects in persons with intellectualand developmental disabilities. Journal of AppliedBehavior Analysis, 50, 206-221. https://doi.org/10.1002/jaba.359

Madden, G. J. (2000). A behavioral economics primer. InW. K. Bickel & R. Vuchinich (Eds), Reframing healthbehavior change with behavioral economics (pp. 3-26).Mahwah, NJ: Lawrence Erlbaum & Associates.

Madden, G. J., Smethells, J. R., Ewan, E. E., &Hursh, S. R. (2007). Tests of behavioral-economicassessments of relative reinforcer efficacy II: Eco-nomic complements. Journal of the ExperimentalAnalysis of Behavior, 88, 355-367. https://doi.org/10.1901/jeab.2007.88-355

Malagodi, E. F. (1967). Acquisition of the token-rewardhabit in the rat. Psychological Reports, 20, 1335-1342.https://doi.org/10.2466/pr0.1967.20.3c.1335

Mazur, J. E. (1983). Steady-state performance on fixed-,mixed-, and random-ratio schedules. Journal of theExperimental Analysis of Behavior, 39, 293-307.https://doi.org/10.1901/jeab.1983.39-293

Mazur, J. E. (2004). Risky choice: Selecting between cer-tain and uncertain outcomes. Behavior Analyst Today,5, 190-203. https://doi.org/10.1037/h0100031

McGinnis, J. C., Friman, P. C., & Carlyon, W. D.(1999). The effect of token rewards on “intrinsic”motivation for doing math. Journal of Applied Behav-ior Analysis, 32, 375-379. https://doi.org/10.1901/jaba.1999.32-375

McLaughlin, T. F., & Malaby, J. E. (1972). Intrinsicreinforcers in a classroom token economy. Journal ofApplied Behavior Analysis, 5, 263-270. https://doi.org/10.1901/jaba.1972.5-263

McLaughlin, T. F., & Malaby, J. E. (1976). An analysisof assignment completion and accuracy across timeand under fixed, variable, and extended tokenexchange periods in a classroom token economy.

Contemporary Educational Psychology, 1, 346-355.https://doi.org/10.1016/0361-476x(76)90054-0

Miller, J. R., DeLeon, I. G., Toole, L. M.,Lieving, G. A., & Allman, M. J. (2016). Contin-gency enhances sensitivity to loss in a gambling taskwith diminishing returns. The Psychological Record,66, 301-308. https://doi.org/10.1007/s40732-016-0172-5

Moher, C. A., Gould, D. D., Hegg, E., &Mahoney, A. M. (2008). Non-generalized and gener-alized conditioned reinforcers: Establishment and val-idation. Behavioral Interventions, 23, 13-38. https://doi.org/10.1002/bin.253

Neuringer, A. J. (1967). Effects of reinforcement magni-tude on choice and rate of responding. Journal of theExperimental Analysis of Behavior, 10, 417-424.https://doi.org/10.1901/jeab.1967.10-417

O’Leary, K. D. (1978). The operant and social psychol-ogy of token systems. In A. C. Catania &T. A. Brigham (Eds.), Handbook of applied behavioranalysis: Social and instructional processes. New York,NY: Irvington.

O’Leary, K. D., & Drabman, R. (1971). Token reinforce-ment programs in the classroom: A review. Psychologi-cal Bulletin, 75, 379-398. https://doi.org/10.1037/h0031311

Parrish, J. M., Cataldo, M. F., Kolko, D. J.,Neef, N. A., & Egel, A. L. (1986). Experimentalanalysis of covariation among compliant and inappro-priate behaviors. Journal of Applied Behavior Analysis,19, 241-254. https://doi.org/10.1901/jaba.1986.19-241

Pazulinec R., Meyerrose. M., & Sajwaj T. (1983). Pun-ishment via response cost. In S. Axelrod (Ed.), Theeffects of punishment on human behavior (pp. 71-86).New York, NY: Academic Press.

Phillips, C. L., Iannaccone, J. A., Rooker, G. W., &Hagopian, L. P. (2017). Noncontingent reinforce-ment for the treatment of severe problem behavior:An analysis of 27 consecutive applications. Journal ofApplied Behavior Analysis, 50, 357-376. https://doi.org/10.1002/jaba.376

Phillips, E. L. (1968). Achievement Place: Token rein-forcement procedures in a home-style rehabilitationsetting for pre-delinquent boys. Journal of AppliedBehavior Analysis, 1, 213-223. https://doi.org/10.1901/jaba.1968.1-213

Phillips, E. L., Phillips, E. A., Fixsen, D. L., &Wolf, M. W. (1971). Modification of the behaviorsof pre-delinquent boys within a token economy. Jour-nal of Applied Behavior Analysis, 4, 45-59. https://doi.org/10.1901/jaba.1971.4-45

Pietras, C. J., & Hackenberg, T. D. (2000). Timeoutpostponement without increased reinforcement fre-quency. Journal of the Experimental Analysis ofBehavior, 74, 147-164. https://doi.org/10.1901/jeab.2000.74-147


https://doi.org/10.1901/jaba.1972.5-343%20

https://doi.org/10.1901/jaba.1972.5-343%20

https://doi.org/10.1126/science.124.3224.679



http://nbn-resolving.de/urn:nbn:de:bsz:180-madoc-323358

http://nbn-resolving.de/urn:nbn:de:bsz:180-madoc-323358





https://doi.org/10.2466/pr0.1967.20.3c.1335


https://doi.org/10.1037/h0100031





https://doi.org/10.1016/0361-476x(76)90054-0

https://doi.org/10.1007/s40732-016-0172-5

https://doi.org/10.1007/s40732-016-0172-5




https://doi.org/10.1037/h0031311

https://doi.org/10.1037/h0031311











Pietras, C. J., & Hackenberg, T. D. (2005). Response-cost punishment via token-loss with pigeons. Beha-vioural Processes, 69, 343-356. https://doi.org/10.1016/j.beproc.2005.02.026

Powell, R. W. (1968). The effects of small sequentialchanges in fixed-ratio size upon the post-reinforcement pause. Journal of the ExperimentalAnalysis of Behavior, 11, 589-593. https://doi.org/10.1901/jeab.1968.11-589

Raiff, B. R., Bullock, C. E., & Hackenberg, T. D.(2008). Response-cost punishment with pigeons: Fur-ther evidence of response suppression via token loss.Learning & Behavior, 36, 29-41. https://doi.org/10.3758/lb.36.1.29

Rasmussen, E. B., & Newland, M. C. (2008). Asymmetryof reinforcement and punishment in human choice.Journal of the Experimental Analysis of Behavior, 89,157-167. https://doi.org/10.1901/jeab.2008.89-157

Repp, A. C., & Deitz, S. M. (1975). A comparison offixed-ratio and variable-ratio token-production sched-ules with human subjects. The Psychological Record,25, 131-137. https://doi.org/10.1007/bf03394296

Reynolds, G. S. (1961). Behavioral contrast. Journal of theExperimental Analysis of Behavior, 4, 57-71. https://doi.org/10.1901/jeab.1961.4-57

Roane, H. S. (2008). On the applied use of progressive-ratio schedules of reinforcement. Journal of AppliedBehavior Analysis, 41, 155-161. https://doi.org/10.1901/jaba.2008.41-155

Roane, H. S., Call, N. A., & Falcomata, T. S. (2005). Apreliminary analysis of adaptive responding underopen and closed economies. Journal of Applied Behav-ior Analysis, 38, 335-348. https://doi.org/10.1901/jaba.2005.85-04

Roane, H. S., Lerman, D. C., & Vorndran, C. M.(2001). Assessing reinforcers under progressive sched-ule requirements. Journal of Applied Behavior Analysis,34, 145-167. https://doi.org/10.1901/jaba.2001.34-145

Schmauck, F. J. (1970). Punishment, arousal, and avoid-ance learning in sociopaths. Journal of AbnormalPsychology, 76, 325-335. https://doi.org/10.1037/h0030398

Schroeder, S. R., & Barrera, F. J. (1976). Effects of pricemanipulations on consumer behavior in a shelteredworkshop token economy. American Journal of Men-tal Deficiency, 8, 172-180.

Shahan, T. A. (2010). Conditioned reinforcement andresponse strength. Journal of the Experimental Analysisof Behavior, 93, 269-289. https://doi.org/10.1901/jeab.2010.93-269

Skinner, B. F. (1953). Science and human behavior.New York: Macmillan.

Smith, T. D. (1972). Development of tokens as secondaryreinforcers. Journal of Experimental Child Psychology,14, 133-138. https://doi.org/10.1016/0022-0965(72)90038-0

Sran, S.K, & Borrero, J.C. (2010). Assessing the value ofchoice in a token system. Journal of Applied BehaviorAnalysis, 43, 553-557. https://doi.org/10.1901/jaba.2010.43-553

Staats, A. W., Staats, C. K., Schutz, R. E., & Wolf, M.(1962). The conditioning of textual responses using“extrinsic” reinforcers. Journal of the ExperimentalAnalysis of Behavior, 5, 33-40. https://doi.org/10.1901/jeab.1962.5-33

Stokes, D. E. (1997). Pasteur’s quadrant: Basic science andtechnological innovation. Washington, DC: BrookingsInstitution Press.

Tan, L., & Hackenberg, T. D. (2015). Pigeons’ demandand preference for specific and generalized condi-tioned reinforcers in a token economy. Journal of theExperimental Analysis of Behavior, 104, 296-314.https://doi.org/10.1002/jeab.181

Tarbox, R. S. F., Ghezzi, P. M., & Wilson, G. (2006).The effects of token reinforcement on attending in ayoung child with autism. Behavioral Interventions, 21,155-164. https://doi.org/10.1002/bin.213

Taylor-Santa, C., Sidener, T. M., Carr, J. E., &Reeve, K. F. (2014). A discrimination training proce-dure to establish conditioned reinforcers for childrenwith autism. Behavioral Interventions, 29, 157-176.https://doi.org/10.1002/bin.1384

Thompson, R. H., & Iwata, B. A. (2005). A review ofreinforcement control procedures. Journal of AppliedBehavior Analysis, 38, 257-278. http://doi.org/10.1901/jaba.2005.176-03

Thorndike, E. L. (1911). Animal intelligence. New York,NY: Macmillan.

Tustin, R. D. (1994). Preference for reinforcers undervarying schedule arrangements: A behavioral eco-nomic analysis. Journal of Applied Behavior Analysis,27, 597-606. https://doi.org/10.1901/jaba.1994.27-597

Vasta, R., & Stirpe, L. A. (1979). Reinforcement effectson three measures of children’s interest in math.Behavior Modification, 3, 223-244. https://doi.org/10.1177/014544557932006

Waddell, T. R., Leander, J. D., Webbe, F. M., &Malagodi, E. F. (1972). Schedule interactions insecond-order fixed-interval (fixed-ratio) schedules oftoken reinforcement. Learning and Motivation, 3, 91-100. https://doi.org/10.1016/0023-9690(72)90050-1

Webbe, F. M., & Malagodi, E. F. (1978). Second-orderschedules of token reinforcement: Comparisons of per-formance under fixed-ratio and variable-ratio exchangeschedules. Journal of the Experimental Analysis of Behavior,30, 219-224. https://doi.org/10.1901/jeab.1978.30-219

Weiner, H. (1962). Some effects of response cost uponhuman operant behavior. Journal of the ExperimentalAnalysis of Behavior, 5, 201-208. https://doi.org/10.1901/jeab.1962.5-201

Weiner, H. (1963). Some effects of response cost andaversive control of human operant behavior. Journal


https://doi.org/10.1016/j.beproc.2005.02.026%C2%A0

https://doi.org/10.1016/j.beproc.2005.02.026%C2%A0



https://doi.org/10.3758/lb.36.1.29

https://doi.org/10.3758/lb.36.1.29


https://doi.org/10.1007/bf03394296









https://doi.org/10.1037/h0030398

https://doi.org/10.1037/h0030398



https://doi.org/10.1016/0022-0965(72)90038-0

https://doi.org/10.1016/0022-0965(72)90038-0








http://doi.org/10.1901/jaba.2005.176-03

http://doi.org/10.1901/jaba.2005.176-03



https://doi.org/10.1177/014544557932006

https://doi.org/10.1177/014544557932006

https://doi.org/10.1016/0023-9690(72)90050-1




of the Experimental Analysis of Behavior, 6, 415-421.https://doi.org/10.1901/jeab.1963.6-415

Williams, B. A. (1983). Another look at behavioral con-trast in multiple schedules. Journal of the ExperimentalAnalysis of Behavior, 39, 345-384. https://doi.org/10.1901/jeab.1983.39-345

Williams, B. A. (1994). Conditioned reinforcement:Neglected or outmoded explanatory construct? Psy-chonomic Bulletin & Review, 1, 457-475. https://doi.org/10.3758/bf03210950

Winkler, R. C. (1970). Management of chronic psychoticpatients by a token reinforcement system. Journal ofApplied Behavior Analysis, 3, 47-55. https://doi.org/10.1901/jaba.1970.3-47

Winkler, R. C. (1971). The relevance of economic theoryand technology to token reinforcement systems.Behavior Research and Therapy, 9, 81-88. https://doi.org/10.1016/0005-7967(71)90064-7

Winkler, R. C. (1972). A theory of equilibrium in tokeneconomies. Journal of Abnormal Psychology, 2, 169-173. https://doi.org/10.1037/h0032722

Winkler, R. C. (1973). An experimental analysis of eco-nomic balance, savings, and wages in a token econ-omy. Behavior Therapy, 4, 22-40. https://doi.org/10.1016/s0005-7894(73)80072-3

Winkler, R. C. (1980). Behavioral economics, tokeneconomies, and applied behavior analysis. InJ. E. R. Staddon (Ed.). Limits to action (pp. 269-297). New York, NY: Academic Press. https://doi.org/10.1016/b978-0-12-662650-6.50015-5

Winkler, R. C., & Burkhard, B. (1990). A systemsapproach to behavior modification through behavioraleconomics. In L. Green and J. H. Kagel (Eds.),Advances in behavioral economics, Vol. 2 (pp. 288-315). Norwood, NJ: Ablex.

Wolfe, J. B. (1936). Effectiveness of token rewards forchimpanzees. Comparative Psychological Monographs,12, 1-72.

Received June 6, 2016Final acceptance October 2, 2017Action Editor, Iser DeLeon





https://doi.org/10.3758/bf03210950

https://doi.org/10.3758/bf03210950



https://doi.org/10.1016/0005-7967(71)90064-7

https://doi.org/10.1016/0005-7967(71)90064-7

https://doi.org/10.1037/h0032722

https://doi.org/10.1016/s0005-7894(73)80072-3

https://doi.org/10.1016/s0005-7894(73)80072-3

https://doi.org/10.1016/b978-0-12-662650-6.50015-5

https://doi.org/10.1016/b978-0-12-662650-6.50015-5

Documents

Token reinforcement: Translational research and ... · for analysis and application. Unlike standard translational research, based on a unidirectional model in which the analysis