FAILURE TO REPLICATE THE ‘‘WORK ETHIC’’ EFFECT IN PIGEONS

FAILURE TO REPLICATE THE ‘‘WORK ETHIC’’ EFFECT IN PIGEONS

MARCO VASCONCELOS AND PETER J. URCUIOLI

PURDUE UNIVERSITY

KAREN M. LIONELLO-DENOLF

UNIVERSITY OF MASSACHUSETTS MEDICALSCHOOL/SHRIVER CENTER

We report six unsuccessful attempts to replicate the ‘‘work ethic’’ phenomenon reported by Clement,Feltus, Kaiser, and Zentall (2000). In Experiments 1–5, pigeons learned two simultaneousdiscriminations in which the S+ and S2 stimuli were obtained by pecking an initial stimulus once ormultiple (20 or 40) times. Subsequent preference tests between the S+ stimuli and between the S2stimuli mostly revealed indifference, on average, between the S+ from the multiple-peck (high-effort)trials and the S+ from the one-peck (low-effort) trials, and likewise between the two respective S2stimuli. Using a slightly different procedure that permitted assessment of the relative aversiveness of lowversus high effort, Experiment 6 again revealed a pattern of indifference despite showing that pigeonstook considerably longer to begin pecking on high- than on low-effort trials. Our findings call intoquestion the reliability of the original findings and the sufficiency of the hypothesized within-trialcontrast mechanism to produce them.

Key words: work ethic, stimulus value, simultaneous discrimination, within-trial contrast, delay-reduction hypothesis, key peck, pigeons

_______________________________________________________________________________

This paper reports a series of experimentsconducted over the course of a few years toinvestigate the interesting and provocative‘‘work ethic’’ effect reported by Clement,Feltus, Kaiser, and Zentall (2000). Our initialfollow-up experiments were designed to testa particular hypothesis about the origins of theeffect. Unfortunately, we were unable toconduct an adequate test of that hypothesisbecause we were unable to replicate thefindings of Clement et al. This failure led toadditional replication attempts involving a va-riety of procedural variations, but all returnedthe same pattern of null findings. We alsoconducted an experiment resembling onedescribed by Clement and Zentall (2002) tosee if we could obtain results similar to theirs

and evidence for the mechanism proposed byZentall and his associates to explain theirfindings. Once again, however, we did notobtain the preference effects they reported.Considering what we view as an extensive andthorough effort at replication, we are report-ing our null findings to underscore theelusiveness of the work ethic effect and thepossibility that the original findings may havebeen a Type I error (Davison & Nevin, 2005).

Clement et al. (2000) trained pigeons ontwo simultaneous discriminations in which thediscriminative stimuli (S+FR1 and S2FR1) forone task were obtained by pecking once toa center-key stimulus and those for the othertask (S+FR20 and S2FR20) were obtained bypecking 20 times to that same initial stimulus.These low-effort and high-effort trials, respec-tively, were randomly intermixed in eachtraining session. After acquiring both discri-minations to high and stable levels of accuracy,pigeons then received periodic, nondifferen-tially reinforced probe trials on which theychose between the two S+ stimuli and, onseparate probe trials, between the two S2stimuli. Clement et al. found that pigeonspreferred the high-effort S+ over the low-effortS+, and the high-effort S2 over the low-effortS2, even though the two S+ stimuli had

Preparation of this article was made possible by NationalInstitute of Mental Health Grant MH 66195 to Peter J.Urcuioli. The authors thank Andrea Friedrich, ChristineHoeing, Benny Kartono, Natasha Killen, Nicole Madou,Sarah Michalek, Maya Romero, John Schmeling, SamanthaSteen, Steven Suska, Kristine Trent, Matt Watts, DawnWhite and Holly Youngs for assistance in running subjects.

Correspondence concerning this article should beaddressed to Marco Vasconcelos or Peter J. Urcuioli,Department of Psychological Sciences, Purdue University,703 Third Street, West Lafayette, IN 47907-2081. E-mails:[email protected] and [email protected].

doi: 10.1901/jeab.2007.68-06

AND

JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR 2007, 87, 383–399 NUMBER 3 (MAY)

383

virtually identical reinforcement histories (asdid the two S2 stimuli). Averaged across testsessions and three probe-presentation formats,pigeons selected the high-effort S+ 69% of thetime and the high-effort S2 84% of the time.Thus, pigeons clearly preferred the discrimi-native stimuli that they had to work relativelyhard to obtain in training.

Clement et al.’s (2000) explanation for thisfinding was that stimuli following high efforthave greater conditioned reinforcing value.This greater value purportedly arises becausehigh effort is aversive (Eisenberger, 1992)which, in turn, produces a particularly largewithin-trial contrast between this aversive stateand the positive state generated when choos-ing S+ is reinforced (Clement et al., 2000;Clement & Zentall, 2002; Friedrich & Zentall,2004). We hypothesized, however, that differ-ences in the conditioned reinforcing values ofthe high- versus low-effort stimuli were expli-cable in terms of the delay-reduction hypoth-esis of conditioned reinforcement (Fantino,1977). Our reasoning was as follows. It takeslonger to complete 20 pecks than one peck, sothe overall wait (or delay) to reinforcementtimed from the onset of the initial stimulus islonger on high- than on low-effort trials, TH .TL. Consequently, even though the delay toreinforcement timed from the appearance ofthe discriminative stimuli is the same for bothsets of stimuli (tH 5 tL), the reduction in theoverall delay to reinforcement associated withthe appearance of the high-effort stimuli isgreater than it is with the low-effort stimuli:(TH 2 tH) . (TL 2 tL). This should make thehigh-effort discriminative stimuli more potentconditioned reinforcers.

To distinguish between a purely effort-basedaccount and a time-based account, we imple-mented a design that involved a low- versushigh-effort difference (e.g., 1 vs. 20 pecks tothe initial stimulus) but an equal overall delayto reinforcement. In other words, our pro-cedure insured that (TH 2 tH) 5 (TL 2 tL). Byour hypothesis, this should eliminate thepreference for the high-effort stimuli in theClement et al. (2000) procedure.

Unfortunately, there was no preference toeliminate: Pigeons trained under conditionsidentical to those of Clement et al. (2000)were indifferent, on average, to the two S+stimuli and, for the most part, to the two S2stimuli on the probe-trial preference tests. The

first five experiments detail the null findingswe obtained using the Clement et al. method-ology. Experiment 6 was modeled after Clem-ent and Zentall (2002, Experiment 1) in orderto see if a different procedure might yielda stimulus preference and, at the same time, ifprechoice patterns of performance were con-sistent with greater aversiveness of high effort.Once again, preferences failed to emergedespite independent evidence that high effortis indeed aversive relative to low effort.

EXPERIMENTS 1–5

Experiments 1–5 were structurally verysimilar. They differed in terms of the stimuliused for simultaneous discrimination training,the size of the large fixed-ratio (FR) compo-nent, whether or not the initial-link FR re-quirement was signaled, and whether or notthey included a separate group for which trialdurations on small- and large-FR trials wereequated.

METHOD

Subjects

White Carneau retired breeders obtainedfrom the Palmetto Pigeon Plant (Sumter, SC),and maintained at 80% of their free-feedingbody weights by restricted feeding, were used.Each pigeon was housed individually in a stain-less steel, wire-mesh cage in a colony room ona 14 hr:10 hr day–night cycle with lights on atapproximately 07:00. Grit and water werealways available in the home cage. Daily foodrations were obtained in the experimentalapparatus except on the 1 day per weekpigeons were not run. Four experimentallynaıve pigeons were used in Experiment 4.Eight pigeons, all with prior experience in two-alternative delayed matching and randomlydivided into two groups of four, participated inExperiments 1, 2, 3, and 5. Different pigeonswere used in each experiment.

Apparatus

The experiments were run using either oneor two pigeon chambers, consisting of BRS/LVE (Laurel, MD) Model PIP-016 panelsinside BRS/LVE Model SEC-002 enclosures.The pigeon’s compartment in each chamber(36.8 cm high 3 30.5 cm wide 3 34.3 cmdeep) was equipped with an aluminum grid

384 MARCO VASCONCELOS et al.

floor. Three clear plastic pecking keys, 2.5 cmin diameter and spaced 8.3 cm center-to-center, were about 25 cm above the floor onthe panel. Each had a BRS/LVE Model IC-901-IDD in-line projector mounted behind it fordisplaying visual stimuli. The center key pro-jector could display a solid white dot and threewhite vertical or horizontal lines on blackbackgrounds (BRS/LVE Pattern No. 692).The side-key projectors could display a whitevertical or a white horizontal line on a blackbackground, a white triangle and a whiteannulus also on black backgrounds (BRS/LVE Pattern No. 696) and various homoge-neous colored fields (red, green, blue, andyellow) via filters provided with the projectors(Experiments 1 and 2) or via Wratten filters#26, #60, #38a, and #9, respectively (Experi-ments 3–5). A partially covered GE #1829bulb 5.7 cm above the center key served asa houselight. Its light was directed toward theceiling. Purina ProGrains was delivered ina magazine which, when raised, was accompa-nied by lighting a small miniature bulb (ESB-28) in a metal housing that covered it. A 5.8-cm-square opening centered approximately13 cm below the center key allowed access tothe back-mounted magazine. Ventilation andmasking noise was provided by continuouslyrunning blower fans attached to the chambers.IBM-compatible 386 computers located in anadjacent room controlled and recorded allexperimental events.

Procedure

Each experiment involved preliminary train-ing, choice training on two concurrent simul-taneous discriminations, and 5 days of prefer-ence testing. The general aspects of theseexperiments are described first, followed bythe particulars for each experiment.

Preliminary training. Experimentally naıvepigeons (Experiment 4) initially were taughtto eat quickly and reliably from the lit foodmagazine, after which their key peck responseto the center-key dot was shaped via themethod of successive approximations. Next,they learned to peck at red and green, or blueand yellow, stimuli on the left and right keys. Asingle peck to the color appearing on eachtrial turned it off and produced 3-s access tofood. Successive trials in these two 60-trialsessions were separated by a 10-s intertrialinterval (ITI). Each color appeared equally

often on both keys, and the houselightremained on throughout each session.

Next, the pigeons in Experiments 1–4learned to obtain food by pecking the center-key dot multiple times. There were 60 stimuluspresentations per session, and the number ofpecks required for reinforcement was in-creased from 2 to 20 across seven to eightsessions. Completion of the FR requirementimmediately turned off the dot and producedfood. The houselight was off during each trialand was on throughout the ensuing 10-s ITI.After two sessions with the FR 20, pigeonsreceived two additional sessions in whichpecking the dot was reinforced on a mixedFR 1 FR 20 schedule. On half of the 60 trials inthese sessions, food was contingent on a singlepeck to the dot; on the other half of the trials,it was contingent on completing an FR 20.Pigeons in Experiment 5 received eithermixed- or multiple-schedule training fromthe outset (see below), with the large FRgradually raised from 3 to 40 over the courseof 15 to 16 sessions. The duration of foodreinforcement was constant within a sessionbut varied from 2 to 6 s across sessions so as tomaintain body weights as close to the 80% free-feeding values as possible.

Choice training. During choice training,pigeons learned two simultaneous discrimina-tions, one preceded by a single peck to thecenter-key stimulus (FR 1 trials) and the otherby pecking the center-key stimulus multipletimes (FR 20 in Experiments 1–4 or FR 40 inExperiment 5). Completion of the scheduledFR on each trial turned off the center-keystimulus and immediately produced (or even-tually produced—see Experiment 1 procedurebelow) an S+ and an S2 stimulus on the twoside keys. One set of discriminative stimulialways appeared on FR 1 trials, and the otherset always appeared on FR 20 (or FR 40) trials,with each stimulus in a set appearing equallyoften on the left and right keys. The first keypeck to either discriminative stimulus turnedoff the other side-key stimulus and initiated a 6-s interval that ended response-independentlywith the offset of the remaining stimulus andpresentation of food (if the S+ had beenchosen) or a timeout of equivalent duration (ifthe S2 had been chosen).

The two simultaneous discriminations oc-curred in one of two random orders with theconstraints that each occur 48 times per 96-

ELUSIVE ‘‘WORK ETHIC’’ EFFECT 385

trial session and that none of the four possibletrial types—FR 1 versus FR 20 (or FR 40) initiallink 3 the two left–right positions of theterminal-link discriminative stimuli—occurmore than three times in a row. The house-light was off during a trial and was onthroughout the 10-s ITI. The duration of foodreinforcement was again varied across sessionsas needed to maintain each pigeon’s 80%body weight.

Pigeons were trained to a criterion of onesession in which they chose the S+ stimulus90% or more of the time on both the FR 1 andFR 20 (or FR 40) simultaneous discrimina-tions. Twenty overtraining sessions were thengiven.

Preference testing. Choice training was fol-lowed by five preference-test sessions. Eachcontained 60 baseline trials, 30 of eachsimultaneous discrimination, and 36 test trials.Half of the test trials involved a choice betweenthe S+ from the FR 1 simultaneous discrimi-nation and the S+ from the FR 20 (or FR 40)discrimination. The other half involveda choice between the two respective S2stimuli. A single peck to either of the two S+or the two S2 stimuli immediately turned offthe other stimulus and initiated a 6-s intervalthat ended with the remaining stimulus turnedoff and with food with a .5 probability. In otherwords, choices on the test trials were partiallyand nondifferentially reinforced.

Twelve test trials (six with the two S+ stimuliand six with the two S2 stimuli) were pre-ceded by a single peck (FR 1) to the center-keystimulus. Another twelve were preceded bycompleting an FR 20 (or 40) to the center-keystimulus. On the remaining twelve test trials,the two S+ or the two S2 stimuli simplyappeared on the side keys at the end of the ITI(viz., they were not preceded by a center-keystimulus). The 36 test trials were randomlyintermixed among the 60 training trials withthe constraints that each of the four possibletrial types in training occur at least oncebefore the first test trial and that no morethan two test trials occur in a row.

Experiment 1. Two groups of 4 pigeons wererun, one under conditions identical to thosedescribed by Clement et al. (2000) and theother for which FR 1 and FR 20 trial durations(timed from the onset of a center-key dot tothe onset of the choice stimuli) were equated.This was accomplished by recording FR 20 trial

durations on a trial-by-trial basis and, on an FR1 trial, inserting a blank interval between theoffset of the center-key dot and onset of the FR1 choice stimuli that equaled a previous FR 20duration. The FR 20 durations were usedsequentially. A ‘‘seed’’ duration was used ifthe first trial of a session was an FR 1 trial.1

For half of the pigeons in each group, redand yellow were the choice stimuli for the FR 1trials, and blue and green were the choicestimuli for the FR 20 trials. For the other half,these assignments were reversed. Also, red andblue served as the S+ stimuli, and green andyellow as the S2 stimuli, for half of the pigeonsin each group and vice versa for the other half.During the preference test, then, pigeonschose between red versus blue and betweengreen versus yellow.

Experiment 2. Two groups like those de-scribed above were run, except that red andgreen served as the S+ stimuli (and yellow andblue as the S2 stimuli) for half of the pigeons,and vice versa for the other half. Test trials,then, involved choices between red versusgreen and blue versus yellow. This changekept the color pairings identical to those usedby Clement et al. (2000).

Experiment 3. This was identical to Experi-ment 2 except that Wratten filters #26, #60,#38a, and #9, respectively, were used toproduce the red, green, blue, and yellowdiscriminative stimuli. These were the samefilters used by Clement et al. (2000) and wereprovided to us by the Zentall lab.

Experiment 4. Experimentally naıve pigeonswere used, and all were run under conditionslike those described by Clement et al. (2000).The Wratten filters again were used to producethe color stimuli. After the five color prefer-ence tests between red and green and betweenblue and yellow, pigeons were retrained usingvertical and horizontal lines, and the triangleand annulus forms, as the choice stimuli.These were counterbalanced with respect to

1 The seed duration for each pigeon in the equalduration condition was computed by taking the averagedifference in trial duration between FR 20 and FR 1 trialsfor the two sessions of mixed FR 1 FR 20 trainingpreceding choice training and adding 6 to 7 s to thedifference. The latter compensated for the 6-s interval oftime between the pigeon’s choice of one of the discrim-inative stimuli and food presentation (or timeout) duringchoice training. The seed duration for each pigeon wasconstant throughout its training, but differed acrosspigeons.


the FR initial-link requirements and their rolesas S+ and S2. Test trials involved choicesbetween vertical and triangle, and betweenhorizontal and annulus.

Experiment 5. Two groups of pigeons weretrained with FR 40 as the large requirementand with the same choice stimuli as inExperiment 3. For Group Dot, a mixed FR 1FR 40 schedule was in effect for pecking thecenter-key dot. For Group Line, vertical andhorizontal center-key lines signaled the tworesponse requirements (a multiple FR 1 FR 40schedule), with the line-FR relations counter-balanced across pigeons.

RESULTS

Choice training. In all five experiments,pigeons rapidly acquired the two simultaneousdiscriminations in training. With hues as thediscriminative stimuli, the average number ofsessions to criterion (including the criterionsession) across all five experiments was 2.7,with most pigeons needing only 2 or 3 sessionsto reach 90% or better accuracy on bothdiscriminations. With the line orientations andforms as stimuli (Experiment 4), pigeonsrequired an average of 4.8 sessions to meetcriterion (range: 4 to 6 sessions). Terminallevels of accuracy were virtually indistinguish-able for the two discriminations. Over the lastfive overtraining sessions, the mean across-experiment accuracy on the discriminationthat followed the FR 1 initial link was 99.5%(range: 98.3%–99.9%) versus 99.8% on thediscrimination that followed the FR 20 (or FR40) initial link (range: 98.8%–100.0%). Therewere no significant between-discriminationdifferences within any experiment, largestF(1, 3) 5 2.01.

Preference testing. The data from the three‘‘standard’’ groups (cf. Clement et al., 2000)in Experiments 1–3 were combined, as werethe data from the three equal-duration groups,because preliminary analyses showed no sig-nificant across-experiment differences in over-all S+ preferences or in overall S2 preferencesfor either training condition, largest F(2, 9) 52.35, and no significant interactions betweentraining condition and the response require-ment (none, FR 1, or FR 20) preceding the S+or S2 preference test, largest F(4, 18) 5 1.16.The combination was also justified on thegrounds that all pigeons were experienced,hues were used as discriminative stimuli in

each experiment, and the same contingenciesand procedures were in effect across experi-ments.

Figure 1 shows mean preferences averagedover the five test sessions for choosing the S+stimulus (solid bars) and the S2 stimulus(hatched bars) that followed the FR 20 peckrequirement in training. The results from thestandard and equal-duration groups appear inthe top and bottom panels, respectively, andare broken down according to the initial-linkwork requirement that preceded the two S+stimuli or the two S2 stimuli on test trials.

Fig. 1. Average percentage choices (+ SEM) of the FR 20stimuli from training on the probe-trial tests in Experi-ments 1, 2, and 3 combined as a function of initial-linkwork requirement. Solid bars show the results from the S+probe trials; hatched bars show the results from the S2probe trials. Top panel: Data from groups run in the samefashion as that reported in Clement et al. (2000). Bottompanel: Data from groups for which trial durations timedfrom the onset of the initial stimulus to onset of thediscriminative stimuli were the same on lowversus high-effort trials.


(‘‘None’’ indicates that there was no initiallink prior to the appearance of the teststimuli—i.e., those stimuli were presentedimmediately after the preceding intertrialinterval). Appendix A shows the preferencedata for individual subjects.

None of the mean preferences in thestandard group deviated significantly fromchance, all ts (11) , 1.60, although there wasa clear tendency toward choosing the low-effort(FR 1) S+ and S2 on test trials preceded by anFR 20 initial link. For example, 7 of the 12pigeons chose the FR 1 S2 from training 83%of the time or more when given a choicebetween that stimulus and the FR 20 S2training stimulus. In the equal-duration group,there was a significant preference for the high-effort (FR 20) S2 on the S2 test trialspreceded by an FR 1 initial link, t(11) 52.41. No other preferences were significant,ts(11) , 1.80, although again there wasa tendency toward choosing the low-effort S2on FR 20 initial-link test trials.

Figure 2 depicts the corresponding meanpreferences for the experimentally naıve pi-geons in Experiment 4 averaged across theirfive hue and five form test sessions. There wereno significant overall preferences for the high-effort S+ (solid bars) on either test, ts(3) ,2.06, although there was a tendency in thatdirection on the form S+ test following an FR20 initial link. On the S2 test trials (hatchedbars), the preference for the high-effort S2approached significance on the form test trialspreceded by an FR 1 initial link, t(3) 5 2.86, p5 .06, two-tailed. All other S2 preferenceswere not significant despite appearances, ts(3), 1.61, due to considerable pigeon-to-pigeonvariability. For example, on the hue S2 testtrials preceded by an FR 1 initial link, 2pigeons chose the high-effort S2 more than90% of the time but the other 2 pigeons chosethe high-effort S2 less than 50% of the time(see Appendix A).

Figure 3 shows the high-effort stimuluspreferences in Experiment 5 where the largeFR was 40 (rather than 20). Data from thegroup trained on the mixed FR 1 FR 40schedule with the center-key dot (Group Dot)are shown in the top panel; data from thegroup trained on the mult FR 1 FR 40 schedulewith different center-key line orientations(Group Line) are shown in the bottom panel.There were no significant preferences on the

S+ test trials (solid bars) in either group.Likewise, there were no significant preferencesin Group Line on the S2 test trials (hatchedbars), although there was a tendency towardchoosing the low-effort S2 following an FR 40initial link, t(3) 5 22.17, p , .12, two-tailed.Despite appearances, there were no significantoverall preferences on the S2 test trials inGroup Dot, ts(3) , 1.21, because 1 pigeonin this group showed a very strong preferencefor the low-effort (FR 1) S2 even thoughthe other 3 pigeons showed strong preferencesfor the high-effort (FR 40) S2 (see AppendixA).

Fig. 2. Average percentage choices (+SEM) of the FR 20stimuli from training on the probe-trial tests in Experi-ment 4 as a function of initial-link work requirement. Solidbars show the results from the S+ probe trials; hatched barsshow the results from the S2 probe trials. Top panel:Preference data following simultaneous discriminationtraining with hue stimuli. Bottom panel: Preference datafollowing simultaneous discrimination training withform stimuli.


DISCUSSION

None of our five experiments replicated thework-ethic effect reported by Clement et al.(2000) despite procedures virtually identical totheirs and, in three experiments, using huestimuli generated from the same Wrattenfilters. One reasonable concern is that ourgroups contained fewer subjects than inClement et al., so we had less power ofdetecting a true preference effect. But thisconcern is muted by the fact that pooling thedata from Experiments 1 to 3, thus creatinglarger overall Ns than used by Clement et al.,did not change the outcome, except for

a significant preference for the high-effortS2 on one type of test trial in the equal-duration group. Also, if the subject-to-subjectvariability observed with our small sample sizesis indicative of the variability we would haveseen had we run another 4 subjects percondition or group, the results would likelyhave been the same. The fact that thisvariability was frequently reproduced acrossexperiments gives us confidence in this assess-ment.

But what about the obvious trends towarda preference for the high-effort S2 in Experi-ments 4 and 5 and the significant preferencefor this stimulus on the FR 1 test trials for theequal-duration groups of Experiments 1–3?Assuming there are true preferences here, weare left with the problem of explaining howthey could arise despite no preference for thehigh-effort S+. After all, the latter would seemto be necessary for the former (Clement et al.,2000) because reinforcing correct choice ofthe high-effort S+ during training purportedlyendows that S+ with a higher overall value thanthe value imparted to the low-effort S+ when itis correctly chosen. This value difference isthought to arise from a larger within-trialcontrast between the positive attributes ofreinforcement and the substantially negativeattributes of a large work requirement vis-a-visthe less negative attributes of a small workrequirement (e.g., Friedrich & Zentall, 2004).The larger positive value of the high-effort S+in turn creates a preference for the high-effortS2 through the process of value transfer (e.g.,Urcuioli, 2006; Zentall & Sherburne, 1994): Ahigh-effort S2 has more positive value avail-able to be transferred to it from the high-effortS+ than does the low-effort S2 from the low-effort S+. But this analysis must certainlypredict that the hypothesized greater positivevalue of the high-effort S+ would be reflectedin a preference for that stimulus over the low-effort S+. None of our experiments, however,showed such a preference.

An alternative explanation for the trendtoward preference for the high-effort S2 issuggested by the fact that this preference wasapparent only (or mostly) when choices werepreceded by an FR 1 initial link or by no initiallink. The bias toward pecking the high-effortS2 in testing could have occurred if instead oflearning to peck the S+ on FR 1 trials duringchoice training, the pigeons learned not to

Fig. 3. Average percentage choices (+ SEM) of the FR 40stimuli from training on the probe-trial tests in Experi-ment 5 as a function of initial-link work requirement. Solidbars show the results from the S+ probe trials; hatched barsshow the results from the S2 probe trials. Top panel:Preference data from pigeons trained with a dot initialstimulus. Bottom panel: Preference data from pigeonstrained with vertical and horizontal lines as initial stimulisignaling lowversus high-effort requirements.


peck the S2 on these trials (Clement et al.,2000). Avoidance of the low-effort S2 on theFR 1 test trials would result in the selection ofthe alternative, high-effort S2. Also, if notpecking is more similar to one peck than it isto 20 (or 40 pecks), generalization wouldproduce the same bias on the S2 test trialsconducted without an initial link. This sameargument, of course, should apply to test trialsin which the initial link was an FR 20 (or FR40): Pigeons should be more likely to selectthe low-effort S2 if they had learned to avoidthe high-effort S2 during training. This over-all trend can also be seen in most of the data.

EXPERIMENT 6

Clement and Zentall (2002) proposed thatthe within-trial contrast purportedly responsi-ble for the work-ethic effect can also occurfrom the anticipation of differential effort. Intheir Experiment 1, trials began with thepresentation of either vertical or horizontallines. Pecking the vertical lines produceda simultaneous discrimination on half of thetrials and a low-effort (FR 1 reinforced)stimulus on the other trials. Pecking thehorizontal lines produced a different simulta-neous discrimination on half of the trials anda high-effort (FR 30 reinforced) stimulus onthe other trials. During subsequent testing,pigeons periodically chose between the S+stimuli from the simultaneous discriminationsand between the corresponding S2 stimuli.Consistent with their hypothesis, Clement andZentall found a preference for the high-effortS+ and for the high-effort S2. Because thehigh- versus low-FR stimuli and the simulta-neous discriminations were presented onseparate trials (i.e., differential effort did notdirectly precede the discriminative stimuli),the authors suggested that their pigeonsjudged the value of each S+ and S2 relativeto their expectation of differential effort assignaled by the initial line stimuli, a second-order contrast effect.

To follow up on these more recent findingsand to evaluate the sufficiency of the mechan-ism(s) proposed by Clement and Zentall(2002; see also Friedrich & Zentall, 2004), wearranged that both the anticipation of differ-ent effort and differences in actual effortwould precede presentation of different setsof discriminative stimuli for one group (Group

Signaled). This was accomplished by usingdifferent initial stimuli to signal low versushigh forthcoming effort. For a second group(Group Unsignaled), the initial stimulus wasalways the same and, hence, uninformativeregarding forthcoming effort. In short, thediscriminative stimuli for Group Signaled werepreceded by the anticipation of effort andactual effort, whereas for Group Unsignaledthey were preceded only by actual effort.

For both groups, the high-effort FR re-quirement was set to 80 in an attempt toincrease the chances of seeing a work ethiceffect which, to this point, had failed tomaterialize. According to Clement et al.(2000; see also Clement & Zentall, 2002;DiGian, Friedrich, & Zentall, 2004; Friedrich& Zentall, 2004), preference for the discrim-inative stimuli preceded by this especially largeeffort should be apparent in both groups.Furthermore, if the anticipation of effortcombines with actual effort, those preferencesshould be more pronounced in Group Sig-naled than in Group Unsignaled or, ata minimum, should appear in the formergroup even if they remain absent in the lattergroup.

METHOD

Subjects and Apparatus

Sixteen White Carneau retired breedersparticipated in Experiment 6. All were experi-enced with a variety of procedures orthogonalto those of the present experiment. They werehoused and maintained as described previous-ly. Two operant chambers similar to those usedin Experiments 1 to 5 were used.

Procedure

Pigeons were randomly assigned to GroupSignaled or Group Unsignaled and to eachexperimental chamber, with 8 pigeons pergroup and chamber.

Preliminary training. Subjects initially weretrained to obtain food by pecking a dot on thecenter key, with the response requirementprogressively increased to 80 pecks (FR 80)over 19 to 27 sessions. Next, pigeons weretrained to peck vertical and horizontal lines onthe center key, with these trials randomlyintermixed within each session. The FR re-quirement was progressively increased to FR 5over three sessions. Completion of the FR


requirements produced 2 to 6-s access to food,adjusted individually for each pigeon in eachsession so as to maintain its 80% body weight.Each preliminary training session had 60 trialswith successive trials separated by a 10-s ITI.The houselight remained on during the ITIand off during the presentation of the stimuliand food reinforcement.

Choice training. As shown in Figure 4, eachdiscrimination training trial for Group Signaledbegan with the offset of the houselight and theillumination of either vertical or horizontallines on the center key. Completing an FR 5requirement turned off the center-key linestimulus and turned on a center-key dot, towhich 1 or 80 pecks (FR 1 or FR 80) wererequired, depending on which initial linestimulus had been presented. For 4 pigeons,one peck to the dot was required when theinitial stimulus was vertical and 80 pecks to thedot were required when the initial stimulus washorizontal. For the remaining 4 pigeons, thesecontingencies were reversed. Completion ofthe FR 1 or FR 80 requirement terminated thedot and produced two different hues on theside keys. On trials with the FR 80 workrequirement, one set of hues (high-effort S+and high-effort S2) was presented, whereas ontrials with the FR 1 work requirement, a differ-ent set of hues (low-effort S+ and low-effort S2)was presented. Once the discriminative stimuliappeared, a single peck to either hue turned off

the other side-key stimulus and initiated a 6-speriod that ended in food reinforcement if theS+ was pecked first or in nonreinforcement ifthe S2 was pecked first. A 10-s ITI with thehouselight on then commenced.

For Group Unsignaled, the order of the lineand dot stimuli was reversed. The initialcenter-key stimulus was the dot to which anFR 1 or FR 80 was required, and the secondstimulus was either vertical or horizontal towhich an FR 5 was then required. For 4pigeons, the vertical lines appeared aftercompleting an FR 1 to the dot and thehorizontal lines appeared after completingan FR 80 requirement. For the remainingpigeons, these contingencies were reversed.After subsequently completing the FR 5 re-quirement to the lines, the discrimination trialproceeded as described for the signaled group(i.e., one set of hues appeared on FR 1 trialsand a different set appeared on FR 80 trials).

For both groups, red and yellow, and greenand blue always appeared together as discrim-inative stimuli. The colors within a set ap-peared equally often as high-effort S+ andhigh-effort S2, or as low-effort S+ and low-effort S2, across the pigeons in each group.Each training session contained 96 trials, 48involving high effort and 48 involving loweffort, with left-right locations of the S+ andS2 stimuli counterbalanced across trials. Trialswere presented in a pseudorandom order, with

Fig. 4. High- and low-effort training trials for Groups Signaled and Unsignaled in Experiment 6. Location of the S+ andS2 (left or right) was counterbalanced over trials, and the S+ and S2 hues were counterbalanced within each group. V 5vertical lines; H 5 horizontal lines; D 5 dot ; R 5 red; Y 5 yellow; G 5 green; B 5 blue; + 5 reinforced choice; 2 5nonreinforced choice.


the constraint that no trial type (high- vs. low-effort) occur more than three times insuccession. During choice training (and sub-sequent preference testing), the shortest pos-sible reinforcement duration was 1.0 s ratherthan 2.0 s.

Pigeons were trained to a criterion of at least90% correct on both discriminations within thesame session. After meeting this criterion, eachpigeon received an additional 20 overtrainingsessions. During overtraining, 3 pigeons fromGroup Signaled and 2 from Group Unsignaledstopped pecking during high-effort trials. For 3of these pigeons, the FR 80 requirement wasreduced to FR 60; for the remaining 2 pigeons,the FR requirement was further reduced to FR50 and FR 40. One pigeon from GroupSignaled was eventually dropped from theexperiment due to an injury.

Preference testing. Five test sessions immedi-ately followed discrimination training. Eachtest session included 36 test trials randomlyintermixed among 60 training trials. Half ofthe test trials involved a choice between thetwo S+ stimuli (high-effort S+ vs. low-effort S+)and the other half a choice between the twoS2 stimuli (high-effort S2 vs. low-effort S2).All test trials commenced with the presenta-tion of the two S+ or the two S2 stimuli, withleft–right locations counterbalanced acrosstest trials. As on training trials, a single peckto either side key turned off the other side keyand initiated a 6-s period. After the 6-s intervalelapsed, however, food reinforcement ran-domly occurred 50% of the time independent-ly of which stimulus was initially pecked.

RESULTS

Choice training. Group Signaled and GroupUnsignaled learned the two hue discrimina-tions in an average of 2.7 and 3.0 sessions,respectively. A two-way, mixed ANOVA withgroup and trial type as factors yielded nosignificant effect of group, trial type, or theirinteraction, all Fs(1,13) , 2.67. During over-training, both groups continued to performaccurately on each trial type: 99.8% and 99.7%correct on high- and low-effort trials, respec-tively, for Group Signaled; 99.6% and 99.8%correct, respectively, for Group Unsignaled.There were no significant effects of group, trialtype, or their interaction, all Fs(1,13) , 1.00.

The top panel of Figure 5 shows the averagelatency to peck the initial stimulus (the dot for

Group Unsignaled, and vertical or horizontallines for Group Signaled) over the last fivetraining sessions. On the small FR (low-effort)trials, the groups did not differ in theirlatencies: 1.4 s for both groups. On the largeFR (high-effort) trials, however, Group Sig-naled showed much longer latencies thanGroup Unsignaled (52.5 s vs. 2.1 s, respective-ly). A two-way, mixed ANOVA performed onthese data yielded a significant effect of group[F(1,13) 5 10.60], trial type [F(1,13) 5 11.33],and their interaction [F(1,13) 5 10.73]. Thesignificant interaction confirms that GroupSignaled could anticipate low versus higheffort upon seeing the initial stimulus whereasGroup Unsignaled could not.

Fig. 5. Top panel: Average latency (+ SEM) to peck theinitial stimulus on low-effort (Small FR) and high-effort(Large FR) trials over the last five training sessions for eachgroup in Experiment 6. Bottom panel: Trial duration (+SEM) measured from the onset of the initial center-keystimulus to the onset of the choice stimuli averaged overthe last five training sessions for each group in Experiment6 on low-effort (Small FR) and high-effort (Large FR) trials.


The bottom panel of Figure 5 plots theduration of each trial type measured from theonset of the initial center-key stimulus to theappearance of the hue choice stimuli averagedover the last five training sessions. As expected,the duration of the large-FR (high-effort) trialswas significantly greater than the duration ofthe small-FR (low-effort) trials for both groups.ANOVA showed an overall effect of trial type,F(1,13) 5 32.92, but no effect of group[F(1,13) , 1.00] or Group 3 Trial-type inter-action [F(1,13) , 1.00].

The absence of a Group 3 Trial typeinteraction seems, at first glance, at odds withthe significant interaction found for latencies.This apparent inconsistency is explained bythe behavior of the pigeons in Group Un-signaled on large-FR trials. If the first peck tothe initial, center-key dot did not remove thedot, then these pigeons typically would pausebefore continuing to peck. Thus, the high-effort trials for Group Unsignaled were char-acterized by an initial short latency followed bya long run time whereas, for Group Signaled,these trials were characterized by a longlatency followed by a somewhat shorter runtime. Stated otherwise, the long run time ofGroup Unsignaled compensated for the longlatency of Group Signaled.

Preference testing. During testing, the twogroups continued to perform accurately onthe baseline discriminations. Accuracy onhigh- and low-effort trials averaged 99.9%and 99.0%, respectively, in Group Signaledand 99.7% and 99.9%, respectively, in GroupUnsignaled. ANOVA revealed no significanteffects, all Fs(1,13) , 4.38.

As shown in the top panel of Figure 6, onthe first test session the pigeons in GroupSignaled chose the high-effort S+ over the low-effort S+ 38.1% of the time (filled bars) andchose the high-effort S2 over the low-effortS2 61.9% of the time (hatched bars). Thecorresponding percentages for Group Un-signaled were 39.6% and 54.2%, respectively.t-tests showed that there were no significantdeviations from chance in Group Signaled[t(6) 5 1.99 for the high-effort S+ and t(6) 521.26 for the high-effort S2] or in GroupUnsignaled [t(7) 5 21.36 for the high-effortS+ and t(7) 5 1.0 for the high-effort S2].

ANOVA using test-trial type and group asfactors showed a significant overall effect oftest-trial type, F(1,13) 5 7.45, but no signifi-

cant group effect or Group 3 Trial typeinteraction, Fs(1,13) , 1.00. In short, thepreference for the high-effort S2 averagedacross groups was higher than the overallpreference for the high-effort S+.

The bottom panel of Figure 6 shows thepreferences averaged over all five test sessions.The pattern of results is similar to the first testsession. Group Signaled chose the high-effortS+ 43.3% of the time (filled bars) and thehigh-effort S2 62.4% of the time (hatchedbars). The corresponding percentages forGroup Unsignaled were 42.9% and 50.8%,respectively. t-tests again showed no significantdeviation from chance either in Group Sig-naled [t(6) 5 20.89 and t(6) 5 1.64 for thehigh-effort S+ and the high-effort S2, re-

Fig. 6. Average percentage choices (+SEM) of the large-FR S+ stimulus (solid bars) and large-FR S2 stimulus(hatched bars) stimuli from training on the probe-trialtests for each group in Experiment 6. Top panel:Preference data from the first test session. Bottom panel:Preference data averaged across the five test sessions.


spectively] or in Group Unsignaled [t(7 ) 521.18 for the high-effort S+ and t(7) 5 0.26for the high-effort S2]. Individual-subject dataaveraged over the five test sessions are shownin Appendix A.

As for the first test session, ANOVA yieldeda significant overall effect of test-trial type,F(1,13) 5 4.73, but no significant group effector Group 3 Trial-type interaction, Fs(1,13) ,1.00. Thus, across the five test sessions, theoverall preference for the high-effort S2 wassignificantly higher than the overall prefer-ence for the high-effort S+.

DISCUSSION

According to Clement and Zentall (2002),within-trial contrast can operate on the basis ofactual differential effort or the anticipation ofdifferential effort. In this experiment, bothgroups experienced differential effort and, forGroup Signaled, conditions were manipulatedto produce the anticipation of differentialeffort as well. Additionally, the high-effort FRrequirement was increased to FR 80 in order toenhance any within-trial contrast that mightoperate.

Despite our efforts, neither group showeda preference for the high-effort S+ or for thehigh-effort S2. Overall, results from GroupUnsignaled followed the pattern of the pre-vious five experiments in the sense that nowork-ethic effect was observed. Likewise,Group Signaled revealed no significant devia-tions from chance during testing, although thelatency data strongly suggest that these pi-geons were able to anticipate the FR require-ment involved in each trial. Thus, the work-ethic effect did not materialize even whenactual differential effort was combined withthe anticipation of effort.

Another interesting feature of the data isthat during training, the two simultaneousdiscriminations were preceded by differentdelays (i.e., different times required to com-plete all response requirements prior to theappearance of the S+ and S2; Figure 5,bottom panel). In principle, this fact should,by itself, produce within-trial contrast giventhat a long delay is considered relatively moreaversive than a short one (DiGian et al., 2004).In DiGian et al., completion of an FR 20requirement to an initial stimulus immediatelyproduced one set of discriminative stimuli onsome trials (i.e., after a 0-s delay) and

a different set of discriminative stimuli onother trials after a 6-s delay. For some pigeons,delay was unsignaled (the initial stimulus waswhite for both delays) whereas for others, itwas signaled (the initial stimulus was eithervertical or horizontal depending on the delayit would follow). In testing, DiGian et al. founda significant preference for the long-delay S+in the signaled group but not in the un-signaled group. Although it is not entirelyclear why different delays must be signaled toproduce the effect, one would nonethelesspredict given their findings that a preferencefor the high-effort S+ should be found inGroup Signaled in the present experiment.Clearly, that was not the case.

A possible objection is that effort and timewere confounded in the present experiment.After all, we observed different delays onlybecause different work requirements wereimposed. But it seems implausible that thecombination of long delay to reinforcementand high work would detract from thehypothesized influence of within-trial contrast.Besides, in the original work-ethic demonstra-tion (Clement et al., 2000), work and delaywere confounded but preferences nonethelessmaterialized. In fact, Experiments 1–3 re-ported here were run precisely to separatethe effect, if any, of each confounded variable.Theoretically, contrast should be enhanced byany factor capable of producing differentiallypreferred events (for details, see DiGian et al.,2004).

In summary, despite several ingredients thatshould reasonably be expected to increase theprobability of observing a work-ethic effect—differential effort, an especially large, high-effort FR, and (in one case) the anticipation ofhigh effort—no significant preferences wereobserved. These findings are, however, consis-tent with the results of Experiments 1 to 5.

GENERAL DISCUSSION

Studies of the effect of effort on thereinforcing value of discriminative stimulipreviously associated with, or produced bythat effort date back at least to the 1960s. Forexample, Lewis (1965) trained rats to obtainfood by pulling an 80-g weight across a cham-ber compartment painted with stripes in oneorientation (vertical or horizontal) and bypulling a 5-g weight across another compart-


ment painted with stripes in the oppositeorientation. Later, rats were tested for theirpreference for vertical versus horizontal stripesin a T maze that had differently striped sidearms. There was no significant choice prefer-ence over the first 10 test trials, but preferencefor the high-effort stripes became apparentduring 10 subsequent test trials. The reasonfor the difference in results across test blocks isunclear given that lighting conditions variedbetween the first and second block as did thespacing of the trials within blocks. In anyevent, Lewis concluded that the choice dataprovided only ‘‘…limited support…’’ for theinfluence of effort on the secondary reinfor-cing value of the striped stimuli (see alsoLewis, 1964a, b).

Subsequent instrumental conditioning stud-ies by Armus (1999, 2001) and Jellison (2003)demonstrate how limited this support is.Armus (1999), for example, trained rats torun on a 26u incline to a baited black (orwhite) goal box and on a horizontal runway toa baited goal box of the opposite brightness.Preference for black versus white was latercarried out in a Y maze with stems angledupward 13u. No significant preference wasfound for the goal-box brightness that hadfollowed greater running effort (i.e., the 26uincline) during training. Similar null resultswere reported by Armus (2001) and by Jellison(2003) when differently flavored food pelletsfollowed high- and low-effort lever pressingduring training. In those studies, the forcerequired to operate one lever was low (5 or8 g) relative to the force required to operatethe other (50 or 75 g). When later givena choice between the differently flavoredpellets in a T or Y maze, rats were indifferent.

On the other side of the ledger are theeffects of effort and other apparently aversiveconditions on choice preferences in avians.Pigeons have been shown to prefer discrimi-native stimuli that follow high effort (Clementet al., 2000), signaled long delays to reinforce-ment (DiGian et al., 2004), or the absence offood (Friedrich, Clement, & Zentall, 2005).This preference also extends to discriminativestimuli that do not directly follow high effortor the absence of food but, instead, followstimuli which signal high effort or no food onother trials (Clement & Zentall, 2002). Kacel-nik and his associates (Kacelnik & Marsh,2002; Marsh, Schuck-Palm, & Kacelnik, 2004;

Pompilio & Kacelnik, 2005) have reportedsimilar findings with European starlings. Forexample, in Kacelnik and Marsh, starlingslearned to obtain food during training bypecking a key lit by one color after 16 1-mflights (high-effort trials) and by peckinganother key lit by a different color after 4 1-m flights (low-effort trials). When later givena choice between these two colors duringtesting, 10 out of 12 starlings showed a signif-icant preference for the color associated withhigh effort.

But the reported manipulations do notalways yield comparable effects. For instance,Kacelnik and Marsh (2002) did not requireexplicit choices during training (i.e., only anS+ stimulus appeared following low effort andfollowing high effort) but they nonethelessobtained the work-ethic effect in their starlingsduring a subsequent choice test. By contrast,Clement et al. (2000) reported that in a pre-liminary experiment, pigeons did not exhibitpreference for a high-effort stimulus aftersimilar training during which just a single(S+) stimulus followed low-effort (FR 1) andhigh-effort (FR 20) responding.

Likewise, Marsh et al. (2004) showed thata starling’s energetic state at the time itencounters certain stimuli influences the valueof those stimuli. In that study, starlings wereeither food deprived or prefed before eachtraining session during which they learned torespond to different discriminative stimuli.Later, they preferred the stimulus to whichthey responded when food-deprived over theone to which they responded when prefed (seealso Pompilio, Kacelnik, & Behmer, 2006). Butthese results contrast with previous findingsobtained with rats (although see Revusky,1967). For example, in a series of experiments,Capaldi, Myers, Campbell, and Sheffer (1983)found that rats were either indifferent to tastesconsumed under high versus low deprivationor preferred the taste consumed under lowdeprivation. Although these sorts of compar-isons are confounded by across-species differ-ences and all of the methodological differ-ences they typically entail, the differencesshould nonetheless be a relatively minorconcern if preferences are primarily drivenby a common mechanism (e.g., within-trialcontrast).

More important, methodological differ-ences between our training and testing proto-


cols in Experiments 1 to 5 and those ofClement et al. (2000) were essentially absent;in some cases, we used exactly the same choicestimuli. Nevertheless, preference for high-effort discriminative stimuli obtained by Clem-ent et al. did not materialize here.

Why, then, were we unable to obtain a work-ethic effect in pigeons whereas Clement et al.(2000) obtained a clear effect? Could inconspic-uous albeit functionally important details beresponsible for the discrepancy? For instance,our stimuli appeared on three 2.5-cm-diametercircular pecking keys spaced 8.3 cm apart centerto center. Clement et al. reported using threerectangular response keys (3.0 cm wide, 2.5 cmhigh) that were closer together (0.8 cm apart or3.8 cm apart center-to-center). Comparing pre-test performances (e.g., number of sessions tocriterion, terminal levels of accuracy) in ourexperiments versus theirs gives no indicationthat these differences were important.

Our experiments and Clement et al. (2000)also differed in the reinforcement durationsused throughout choice training and testing.Clement et al. used a fixed 1.5-s durationwhich usually required postsession feedings inorder to maintain pigeons at 80–85% bodyweights (T. Zentall, personal communication,November 27, 2006). Our reinforcementdurations were adjusted individually and dailyso that pigeons would obtain sufficient foodduring their experimental sessions to maintain80% body weights. But these adjustments wereuncommon and unnecessary after pigeonsreached asymptotic performance on thechoice training task. Furthermore, 73% ofthe 275 preference tests across our six experi-ments were run using a 2.0-s reinforcementduration. Still, our 2.0-s duration versus the1.5-s duration used by Clement et al. was largeenough that their pigeons needed postsessionfeedings whereas our pigeons did not. Mightthe ‘‘open economy’’ conditions of Clement etal. versus our ‘‘closed economy’’ conditions(e.g., Hursh, 1980) be important? Perhaps, butit seems to us that if pigeons do not receive anyextra food outside of the experimental ses-sions, there should be a larger positive effect offood within a session (i.e., larger within-trialcontrast effects). Our conditions, then, oughtto enhance rather than diminish the chancesof observing the work-ethic effect.

Another way in which a work-ethic effectmight be obscured is if at least some pigeons

choose between the stimuli on the preferencetests conditional upon the events or re-sponses—of the sort suggested in the formtest of Experiment 4 and the S+ preferencetest for Group Dot in Experiment 5—thatimmediately precede their appearance (seealso Capaldi & Myers, 1982, Experiment 1 fora similar effect in rats). Although not an issuein Clement et al. (2000), other data reportedby Zentall and his associates indicate thatpigeons may indeed learn which choice tomake and which choice not to make intraining, conditional upon the particularevents preceding them. DiGian et al. (2004),for example, found that pigeons were morelikely to choose the S+ that was preceded bya delay during training when the test trialswere initiated by a delay than when they wereinitiated by no delay. Although not statisticallysignificant by their analyses, the trend seemsvery clear. Furthermore, an effect of theinitiating event on choices made in testingwas obtained in Friedrich et al. (2005):Pigeons were significantly more likely tochoose the S+ that had been preceded by nofood during training when the test trials beganwith no food than when they began with food(81.2% vs. 53.1%, respectively). The authorspointed out, however, that a significant overallpreference for the S+ preceded by no food stillremained when the results were pooled overthe three possible initiating events for the testtrials (food, no food, and no stimulus).However, it seems clear that if a high pro-portion of pigeons learn to make their choicesin training as if those choices were conditionalupon a prior event (even though they werenot), the chances of detecting an overallpreference will diminish.

This sort of unnecessary learning, however,does not appear to contribute in any sub-stantive way to our null findings. Consequent-ly, at least in terms of the specific paradigmused here, our systematic failed attempts toshow preference for discriminative stimulifollowing high effort, over discriminativestimuli following low effort, lead us to con-clude that the seminal demonstration ofa work-ethic effect in pigeons (Clement et al.,2000) is not a reliable finding.

Although failed replications such as ours areoften regarded (at least implicitly) as lessnewsworthy than demonstrations of noveleffects, Davison and Nevin (2005) argue that


this sort of bias, and the reinforcementcontingencies from which they arise, can fosterinaccurate representations of psychologicalprocesses and phenomena. At a minimum,then, our findings should leave open (or re-open) the question of whether a pigeon’sprior effort enhances the conditioned reinfor-cing properties of stimuli that follow sucheffort.

REFERENCES

Armus, H. L. (1999). Effect of response effort onsecondary reward value. Psychological Reports, 84,323–328.

Armus, H. L. (2001). Effect of response effort on thereward value of distinctively flavored food pellets.Psychological Reports, 88, 1031–1034.

Capaldi, E. D., & Myers, D. E. (1982). Taste preferences asa function of food deprivation during original tasteexposure. Animal Learning & Behavior, 10, 211–219.

Capaldi, E. D., Myers, D. E., Campbell, D. H., & Sheffer, J.D. (1983). Conditioned flavor preferences based onhunger level during original flavor exposure. AnimalLearning & Behavior, 11, 107–115.

Clement, T. S., Feltus, J. R., Kaiser, D. H., & Zentall, T. R.(2000). ‘‘Work ethic’’ in pigeons: Reward value isdirectly related to the effort or time required to obtainthe reward. Psychonomic Bulletin & Review, 7, 100–106.

Clement, T. S., & Zentall, T. R. (2002). Second-ordercontrast based on the expectation of effort andreinforcement. Journal of Experimental Psychology: Ani-mal Behavior Processes, 28, 64–74.

Davison, M., & Nevin, J. A. (2005). On science and thediscriminative law of effect. Journal of the ExperimentalAnalysis of Behavior, 83, 85–92.

DiGian, K. A., Friedrich, A. M., & Zentall, T. R. (2004).Discriminative stimuli that follow a delay have addedvalue for pigeons. Psychonomic Bulletin & Review, 11,889–895.

Eisenberger, R. (1992). Learned industriousness. Psycho-logical Review, 99, 248–267.

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.

Friedrich, A. M., Clement, T. S., & Zentall, T. R. (2005).Discriminative stimuli that follow the absence ofreinforcement are preferred by pigeons over thosethat follow reinforcement. Learning & Behavior, 33,337–342.

Friedrich, A. M., & Zentall, T. R. (2004). Pigeons shift theirpreference toward locations of food that take moreeffort to obtain. Behavioural Processes, 67, 405–415.

Hursh, S. R. (1980). Economic concepts for the analysis ofbehavior. Journal of the Experimental Analysis of Behavior,34, 219–238.

Jellison, J. L. (2003). ‘‘Justification of effort’’ in rats:Effects of physical and discriminative effort on rewardvalue. Psychological Reports, 93, 1095–1100.

Kacelnik, A., & Marsh, B. (2002). Cost can increasepreference in starlings. Animal Behaviour, 63, 245–250.

Lewis, M. (1964a). Effect of effort on value: An exploratorystudy of children. Child Development, 35, 1337–1342.

Lewis, M. (1964b). Some nondecremental effects of effort.Journal of Comparative and Physiological Psychology, 57,367–372.

Lewis, M. (1965). Effect of effort on choice: Value ofa secondary reinforcer. Psychological Reports, 16,557–560.

Marsh, B., Schuck-Palm, C., & Kacelnik, A. (2004). State-dependent learning affects foraging choices in star-lings. Behavioral Ecology, 15, 396–399.

Pompilio, L., & Kacelnik, A. (2005). State-dependentlearning and suboptimal choice: When starlingsprefer long over short delays to food. AnimalBehaviour, 70, 571–578.

Pompilio, L., Kacelnik, A., & Behmer, S. (2006, March 17).State-dependent learned valuation drives choice in aninvertebrate. Science, 311, 1613–1615.

Revusky, S. H. (1967). Hunger level during food con-sumption: Effects on subsequent preference. Psycho-nomic Science, 7, 109–110.

Urcuioli, P. J. (2006). When discrimination fails (or at leastfalters). Journal of Experimental Psychology: AnimalBehavior Processes, 32, 359–370.

Zentall, T. R., & Sherburne, L. M. (1994). Transfer of valuefrom S+ to S– in a simultaneous discrimination.Journal of Experimental Psychology: Animal Behavior Pro-cesses, 20, 176–183.

Received: October 10, 2006Final acceptance: December 18, 2006


Appendix AIndividual subject preferences for the high-effort S+ and the high-effort S2 averaged over the five test sessions in

Experiments 1–6. Preference data are divided according to whether the choice stimuli were obtained by a single initial-link peck (FR 1), 20 or more initial-link pecks (FR 20+), or no initial-link peck (i.e., no initial-link stimulus; None).

Experiment-Group/Condition Subject

High-effort S+ Preference High-effort S2 Preference

None FR1 FR20+ None FR1 FR20+

Exp. 1 -Standard Group ST1 46.67 36.67 3.33 6.67 13.33 0.00ST2 46.67 33.33 46.67 36.67 43.33 16.67ST3 56.67 56.67 93.33 16.67 20.00 40.00ST4 66.67 63.33 26.67 50.00 46.67 50.00

Exp. 1 - Equal-durationGroup

EQ1 43.33 40.00 53.33 46.67 50.00 3.33EQ2 53.33 56.67 50.00 46.67 70.00 56.67EQ3 60.00 60.00 60.00 80.00 80.00 76.67EQ4 56.67 73.33 50.00 63.33 76.67 50.00

Exp. 2 - Standard Group ST5 16.67 23.33 6.67 60.00 53.33 13.33ST6 53.33 86.67 56.67 6.67 0.00 0.00ST7 60.00 46.67 20.00 23.33 16.67 10.00ST8 43.33 63.33 36.67 56.67 56.67 66.67


EQ5 36.67 23.33 36.67 36.67 36.67 3.33EQ6 66.67 63.33 76.67 73.33 96.67 46.67EQ7 33.33 46.67 33.33 16.67 30.00 13.33EQ8 56.67 46.67 40.00 63.33 66.67 13.33

Exp. 3 - Standard Group ST9 40.00 16.67 43.33 96.67 90.00 93.33ST10 60.00 70.00 26.67 83.33 96.67 13.33ST11 30.00 13.33 43.33 26.67 10.00 6.67ST12 46.67 26.67 60.00 93.33 96.67 93.33


EQ9 23.33 36.67 10.00 40.00 90.00 6.67EQ10 46.67 60.00 46.67 66.67 93.33 13.33EQ11 60.00 86.67 80.00 90.00 93.33 86.67EQ12 66.67 56.67 86.67 43.33 33.33 40.00

Exp. 4 - Hue Test ST13 26.67 30.00 26.67 36.67 46.67 20.00ST14 66.67 70.00 80.00 93.33 96.67 86.67ST15 53.33 30.00 40.00 36.67 43.33 0.00ST16 53.33 53.33 43.33 76.67 93.33 56.67

Exp. 4 - Form Test ST13 66.67 86.67 76.67 100.00 100.00 90.00ST14 40.00 26.67 63.33 43.33 73.33 46.67ST15 66.67 50.00 56.67 60.00 53.33 26.67ST16 50.00 53.33 50.00 66.67 90.00 10.00

Exp. 5 - Group Dot D1 33.33 36.67 53.33 10.00 13.33 0.00D2 46.67 50.00 50.00 100.00 96.67 90.00D3 20.00 16.67 33.33 93.33 96.67 93.33D4 50.00 60.00 76.67 66.67 93.33 83.33

Exp. 5 - Group Line L1 40.00 56.67 36.67 0.00 0.00 0.00L2 50.00 70.00 13.33 23.33 43.33 0.00L3 53.33 40.00 76.67 86.67 93.33 60.00L4 56.67 53.33 50.00 53.33 56.67 16.67


Subject

High-effort S+ Preference High-effort S2 Preference

None None

Exp 6 - Group Unsignaled U1 46.67 50.00U2 50.00 33.33U3 50.00 43.33U4 70.00 56.67U5 36.67 63.33U6 10.00 50.00U7 36.67 56.67U8 43.33 53.33

Exp. 6 - Group Signaled S1 13.33 43.33S2 43.33 63.33S3 40.00 93.33S4 30.00 46.67S5 70.00 50.00S6 66.67 53.33S7 40.00 86.67

Appendix A(Continued)


Documents

FAILURE TO REPLICATE THE ‘‘WORK ETHIC’’ EFFECT IN PIGEONS