SENSITIVITY AND STRENGTH EFFECTS OF INSTRUCTIONS O

ThePsychologicalRecord,2006,56,303-320SENSITIVITYANDSTRENGTH:EFFECTSOFINSTRUCTIONSONRESISTANCETOCHANGECHRISTOPHERA.PODLESNIKandPHILIPN.CHASEWestVirginiaUniversitySeveralresearchlaboratorieshavefoundthatinstructedbehaviorcanbelesssensitivetochangesincontingenciesthanshapedbehavior.Thecurrentexperimentexaminedwhetherthesedifferencesinsensitivitycouldberelatedtoresistancetochange.Twogroupsofsubjects,whowerematchedonthebasisofaninitialdisruptionassessment,wereexposedtoavariable-interval30-sscheduleofreinforcementwithandwithoutadisrupter.Thedisrupterwasavideopresentationofapopulartelevisionsituationcomedy.Onegroupreceivedminimalinstructions(MI)thattoldthemonlythattheycouldearnpointsexchangeableformoney.Eachmemberofthesecondgroupreceivedacompleteinstruction(CI)thatdescribedthetopographyofthetargetresponsethatwasyokedtoaMIsubject'sstablebaselineresponserate.TheresponseratesunderthedisruptionconditionfortheCIsubjectsweremoreresistanttochangethantheMIsubjectsin74outof15disruptionsessions.Thesefindingsarediscussedintermsofresistancetochangebeingincreasedbyinstructionalconditionslikethosemanipulatedandthattheproceduresusedtotestdisruptionprovideanadditionalmethodtoevaluatedifferencesbetweeninstructedandcontingency-governedbehavior.Rules,instructions,policies,anddirectionsareassumedtoinfluencebehaviorandareusedtoteach,inform,orprescribebehaviorinmanykindsofsituations.Howrulesaffectbehavior,however,isstillnotclearlyunderstood.Forthisreason,researchontheeffectofrulesonbehaviorhasbeenamajorfocusforanumberofresearchers.Skinner(1969)definedrulesascontingency-specifyingdiscriminativestimuli(SD)thatdescribebehaviorandthecontrollingenvironment.Indefiningrules,Skinneremphasizedthedifferencesbetweenbehaviorthatisinfluencedbydirectcontactwithcontingenciesofreinforcementandbehaviorthatisinfluencedbydescriptionsofthosecontingencies.AccordingtoSkinner,contingency-governedbehaviorisselectedbydirectcontactwiththeconsequencesforthebehaviorinquestionwhereasrule-governedCorrespondencemaybesenttoPhilipN.Chase.P.O.Box6040,DepartmentofPsychology,WestVirginiaUniversity,Morgantown,WV26506-6040.(E-mail:[e-mail]).304PODLESNIKANDCHASEbehaviorismaintainedbyrulesandthecontingenciessupportingthefollowingofrules.Skinnersuggestedthatthecontrollingvariablesaredifferentforcontingency-governedandrule-governedbehaviorevenwhenthebehaviorsaretopographicallysimilar.SensitivityOneoftheprimaryissuesconcerningthesedifferencesbetweenrule-governedbehaviorandotherbehavioristhatmanystudieshavefoundthatinstructedbehaviorisinsensitivetochangesincontingencies.Wheninstructedononeschedule(e.g.,fixed-ratio),subjectswilloftencontinuetorespondinaccordancewithinstructionseventhoughthecontingencychanges(e.g.,toavariable-intervalschedule)(Kaufman,Baron,&Kopp,1966).Forexample,Shimoff,Catania,andMatthews(1981)taughtsubjectstorespondoneitheratandemrandom-interval(RI)differential-reinforcement-of-low-rates(DRL)oratandemrandom-ratio(RR)DRLwitheitherinstructionsorbyexposingthesubjectstothecontingencies.Aftertrainingwascomplete,uninstructedsubjects'behaviorwassensitivetotheremovaloftheDRLintervalsin10outof13cases,whereasinstructedsubjects'behaviorwassensitiveinonly8outof20cases.Shimoffetal.concludedthatinsensitivitytochangesincontingencieswasduetotheinstructions.Althoughinstructedbehaviorcanbemadesensitivetothechangesinlocalcontingenciesofreinforcement(Galizio,1979;Joyce&Chase,1990;LeFrancois,Chase,&Joyce,1988),theinsensitivityofsomekindsofinstructedbehaviorisimportantforapplied,methodological,andtheoreticalreasons.Ifthegoalofteachingortrainingistoproducebehaviorthatadaptstochangesinconsequences,thenoneshouldnotassumethatinstructionswillproduceadaptivebehavior.Infactsomehaveimplicatedtherelationbetweeninstructionalcontrolandrigidity(Wulfert,Greenway,Farkas,Hayes,&Dougher,1994).Ifinstructionsareusedaspartofthemethodsofanexperiment,thenoneshouldnotassumethatthebehaviorachievedisthesameasthebehaviorachievedwithoutinstructions(Galizio,1979).Atthetheoreticallevel,thesedifferencesbetweeninstructedandnoninstructedbehaviorsuggestquestionsrelatedtothecontingenciesoperatingonrule-governedbehavior.Onesuchquestioniswhetherrule-governedbehavioristypicallyatgreaterstrengththanthespecificclassesofrespondingtowhichitiscompared.Forexample,onemayaccountfortheinsensitivityofrule-governedbehaviorfoundinthetypicalexperimentbyreferringtothestrengthofrule-governedbehaviorasaresultoftheextensiveandgeneralizedhistorythehumansubjectshavehadforfollowingrules.Inmostcases,rulefollowinghasbeenreinforcedwitharangeofdifferenttopographiesofbehaviorandspecificconsequences.Givensuchtypicalhistories,oneshouldnotbesurprisedtofindthatinstructedbehavioroftentendstobelesssensitivetochangesincontingenciesbecauseitisstrongerthanthecontingency-governedbehaviortowhichitiscompared.Incurrentbehavioralterms,differencesinstrength,maybedescribedintermsofresistancetochange.EFFECTSOFINSTRUCTIONSONRESISTANCETOCHANGE305ResistancetoChangeNevin,Mandell,andAtak(1983)suggestedthatresponsestrengthisanalogoustomomentuminphysicsandcanb