behavioral learning theory: its relevance …special... · greatest value may be in the development...

MICHAEL L ROTHSCHILD & WILLIAM C. GAIDIS

Behavioral learning theory has been generally overlookedinthe development of marketing thought. The central conceptstates that behavior that is positively reinforced is more likelyto recur than nonreinforced behavior. This runs parallel tothe marketing concept and may be a sufficient mode! fordealing with most low involvement purchase situations. Itsgreatest value may be in the development of promotionalstrategies. This paper extends some of the ideas presentedin an earlier paper in this journal.

BEHAVIORAL LEARNINGTHEORY: ITS RELEVANCE TO

MARKETING AND PROMOTIONS

IN his 1977 presidential address to the Associationfor Consumer Research, Kassarjian (1978) called

for a return to simpler models of consumer behavior.He felt that most consumer decisions were "un-important, uninvolved, insignificant (and) minor,"and for these "we do not need a grand theory ofbehavior." Kassarjian felt that a theory that couldcontribute some degree of parsimony might bebehaviorism, also known as behavioral learningtheory, instrumental conditioning, behavior modi-fication, or operant conditioning deriving from thework of Skinner (1953) and Thorndike (1911).

Nord and Peter (1980) have recently presentedan explication of several such behavioral conceptsunder the general rubric of a "Behavior Modifica-tion Perspective on Marketing." They are to becommended for introducing this paradigm into themarketing hterature, and for covering a broad rangeof materials in a basic exposition. This paper willtake the opposite approach and will examine one

Michael L. Rothschild is Associate Professor of Businessand William C. Gaidis is a doctoral student, both at theUniversity of Wisconsin, Madison.

aspect of behaviorism (and the Nord and Peterpaper) in greater depth. The topic dealt with hereinis behavioral learning theory. It has been selectedbecause of its philosophical similarity to the mar-keting concept and because of its strong potentialas a contributor to marketing thought.

An Expanded View of BehavioralLearning TheoryBehavioral learning theory is the paradigm generallyreferred to when a layperson speaks of "behaviormodification." A review of some of its basic con-cepts can be found in Nord and Peter and willnot be reviewed here. Nord and Peter correctlystate that behavioral learning occurs when a re-sponse behavior precipitates the appearance of astimulus. This paradigm is not new to marketers;the marketing concept is an example of its principlesin that a transaction occurs when purchase behavior(response) takes place and a product (stimulus) isreceived by the consumer. If the product is pleasing(e.g., meets needs), the probability of repeat behav-ior will increase. Additionally, one can enhance the

70 / Journal of Marketing, Spring 1981Journal of Marketing

pleasingness of the product through appropriatemanipulation of price, distribution, and promotionalvariables. Since the key to successful marketingis closely tied to repeat purchase behavior, thenotion of providing positive reinforcement for de-sired behavior is crucial; therefore, positive rein-forcement must be the ultimate goal of the marketer.A more rigorous examination of the principles ofbehavioral learning theory will allow marketers totake advantage of what behaviorists have learned.Such an examination follows.

Figure 1 shows an overview of the relationshipsbetween various common stimuli and responses inmarketing. S, (advertising) is felt to lead to theresponses of awareness and knowledge (R,). Thisparadigm is based on the verbal learning model andthe power of a repetitive S,. The concepts ofvicarious learning and modeling discussed by Nordand Peter are relevant here.

S (product attribute) is felt to reinforce Rj(purchase behavior). If Sj is viewed positively, thenthe probabihty of future purchase behavior (R3 ^)is increased. If the goal of marketing is to eUcitcertain long run behaviors, the reinforcement ofa good product will be much more powerful thanthe preceding commercial stimulus. Sj can promisea benefit and establish a manufacturer's implicitcontract with consumers; Sj must deliver the per-ceived benefit. Behavioral learning is primarilyconcemed with the relationship between S2, Rj andR3 „.

The present discussion will focus upon fivecomponents of the behavioral learning paradigm thatare relevant to marketing—shaping, extinction, re-inforcement schedules, immediate versus delayedreinforcement, primary versus secondary rein-

FIGURE 1Learning Theory in a Marketing Context

Advertising

TrialPurchaseBehavior

Product{Bundle ofPositive andNegativeAttributes)

IncreasedProbabilityof RepeatBehavior

forcers. Of these, shaping and reinforcement sched-ules were introduced by Nord and Peter and areexpanded here. The others represent concepts thathave proven to be useful in other fields, wouldseem to have potential for marketing, and werenot discussed by Nord and Peter.

Shaping Procedures

Shaping may be the single most potentially usefulconcept for marketers. Nord and Peter, though,gave a relatively weak example of shaping; theyhave reduced what is generally regarded as a multi-step process to a one-step process in their illustra-tion.

Shaping is an essential process in deriving newand complex behavior because a behavior cannotbe rewarded unless it first occurs; a stimulus canonly reinforce acts that already occur. New, com-plex behaviors rarely occur by chance in nature.If the only behavior to be rewarded were the finalcomplex sought behavior, one would probably haveto wait a long time for this to occur by chance.Instead, one can reward simpler existing behaviors;over time, more complex patterns evolve and theseare rewarded. Thus, the shaping process occursby a method of successive approximations.

Shaping is important to marketers since the initialpurchase of any new product involves a complexset of behaviors. To elicit repeat purchase behavioris even more complex. One way to reach this finalbehavior is through a series of successive approx-imations. Such a series might begin with the useof a free sample (for a frequently purchased lowpriced product). A coupon would be included inthis sample for a large discount on the first purchase,and in the first purchase the consumer would finda coupon for a smaller discount on later purchases.As these incentives are reduced, the behavior ap-proximates repeat purchase of the product at itsfull retail price. Soon no artificial reinforcers maybe necessary. After the approximate responses havebeen evoked and firmly estabhshed, the arbitrarystimulus supports are "faded" or gradually with-drawn as control is transferred to stimuU likely tofunction as the major elicitors under naturalisticconditions.

In the above example the sample was given toallow perusal and trial of the product. The trialwas reinforced by a good product and a coupontoward the next purchase. The coupon reinforcedconsideration of product purchase. Each ensuingpurchase was reinforced by the product and theenclosed coupon. Each coupon had a lesser value;ultimately, the product was sufficient rein-forcement, repeat behavior was achieved, and the

Behavioral Learning Theory / 71

FIGURE 2Application of Shaping Procedures toMarketing

Teimlnal Goal: Repe

Induce pi^ehaae withlittle financialobligation

Induce purchase ulChmoderate financlal

Shaping Procedure Reinforcement Applied

Free samples dis- Product Performance; Coupotrlbuted, large

enclosed

Product Performance; CoupoDiscount couponprompts purchasewith little cost.

small discounton next purchaseenclosed.

i

iSmall d i

purchase withmoderat

Product Performance

Induce pufchafull flnanciaobligation

ate cast.

aTe occursPurchaTe occurs

coupons (contingent reinforcers) were faded out.In a series of successive approximations, behaviorhas gone from nothing, to trial without financialobligation, to usage with slight obligation, to repeatusage with financial obhgation. Figure 2 summarizesthis process. Not only has the desired behaviorbeen accomplished, but the desired behavior isultimately reinforced by the product itself and notby promotional incentives. A common error in theuse of promotional tools for shaping purposes maybe the improper fading out of the ancillary incen-tives. When the incentives are eventually droppedwithout gradual shaping of approximate behaviorand gradual fading out of incentives, sales may dropas consumers revert to the brand used before theincentives were employed.

A second common error may be the tendencyto overuse these aids; as a result, purchase maybecome contingent upon the presence of a promo-tional tool. Removal of the promotion may leadto the extinction of purchase behavior. If long-termbehavior toward the product is desired, promotionaltools should not overshadow the product. In amarketing situation, it is paramount that rein-forcement for purchase be derived primarily fromthe product, lest purchase become contingent upona never ending succession of consumer deals.

As seen from the above, shaping can be avaluable concept in developing strategies concerningpromotional tools. Given that expenditures on thesetools is now greater than expenditures on advertisingin the U.S. marketplace (Strang 1976), a set of

guidelines for the use of promotional tools is cer-tainly called for. Shaping procedures may form thebasis for such guidelines.

ExtinctionExtinction is the removal of a correlation betweena response and a reward. This is generally doneby removal of the reward or by introduction ofrewards not correlated with the response. Sincemost products attempt to reinforce continuously,lack of reinforcement (poor product performance)will lead to rapid extinction of behavior. This wouldbe especially pronounced in a competitive situation.

In the case of promotional incentives, behaviortowards a product may be based primarily onreinforcement caused by the incentive. If this isthe case, then removal of the incentive will leadto extinction of behavior. It is important, then, tobuild behavior towards the product and not towardsthe promotional incentive. Conversely, the role ofthe incentive should be to shape appropriate behav-ior toward the product. Extinction, then, may bethe result of improper shaping techniques and anoverreliance on incentives that are later removed.

Reinforcement Schedules and Locus of PowerNord and Peter note that appropriate behavior canbe reinforced on a continuous basis or on a varietyof intermittent schedules; most of these scheduhngnotions, though, have limited value for marketing.The concepts of intermittent scheduhng work verywell in situations where there is an imbalance ofpower and/or lack of competition. For example,when pigeons are starved to 80% of their normalbody weight, they work very hard on an intermittentschedule; factory piece workers work hard on anintermittent schedule in part because they do notperceive themselves as having much power in theiremployer relationships. (Strong unions do not allowtheir members to be caught up in a piece worksituation.)

In a simultaneous choice situation (such asgenerally exists in marketing), it is incumbent upona marketer to maintain continuous reinforcement.If reinforcement becomes intermittent, consumerswill shift behavior to the purchase of a competitiveproduct providing continuous reinforcement. Inter-mittent reinforcement in such a situation may beseen as punishment by the consumer.

Not acknowledging this situational differencemay lead, in turn, to the development of an inappro-priate reinforcement strategy. For behavioral learn-ing to work most effectively, one party must controlthe situation. This implies a lack of competition,an imbalance of power, and a closed system. Private

72 / Journal of Marketing, Spring 1981

sector marketing in competitive situations, there-fore, may require reinforcement procedures verydifferent from the noncompetitive examples com-monly used in discussions of reinforcement sched-ules. WhUe there may be few opportunities to usenoncontinuous schedules in relation to the productitself, there are many opportunities to use thenotions of scheduhng with regard to promotionaltools.

Immediate Versus Delayed Reinforcers

In almost all cases, a delayed reinforcement is worthless than immediate reinforcement during acquisi-tion of a behavior; delayed reinforcement inhibitslearning and will lead to a lower probabihty of afuture occurrence. If the reinforcement is delayed,then irrelevant behaviors will occur between thedesired behavior and the reinforcement. As a result,the most recent behavior will be more stronglyreinforced than will the desired behavior.

For example, if a consumer collects proof-of-purchase coupons and then mails them to themanufacturer for a premium, the typical time toreinforcement is four to six weeks. When thepremium arrives it may strongly reinforce the be-havior of opening one's mailbox rather than thebehavior of making multiple purchases of the prod-uct. If the premium has the quahties of a goodreinforcer, it will hkely lead to the pursuit of othermail premiums. The goal of the manufacturer,though, is reinforcement of the purchase rather thanthe pursuit of premiums. Mail premiums are theo-retically weak reinforcers of purchase behavior dueto this delay in reinforcement.

If purchase behavior is to be reinforced (otherthan through a quality product), the reinforcingpremium should be in or on the package. If repeatpurchase behavior is to be reinforced, a multipartpremium may need to be included in or on thepackage.

Primary Versus Secondary Reinforcers

The notion of delayed reinforcement can be extend-ed by considering primary versus secondary rein-forcers. Primary reinforcers, loosely defined, haveintrinsic utility (the product) while secondary rein-forcers (tokens, coupons, trading stamps) have nosuch utility and must be converted. When, forexample, trading stamps are given as reinforcers,delayed gratification results; when they must becollected before redemption (delayed secondaryreinforcers), the potential for success of the promo-tion may be further eroded. Secondary reinforcershave taken on value over time because individualshave learned that they can be converted for primary

reinforcers; they are still, theoretically, less power-ful than primary reinforcers. Given this relationship,perhaps marketers should concentrate on dealsgiving more product per unit of price. This focuseson the primary reinforcer (product) and gives imme-diate reinforcement.

A Review of Some Promotions Literature

Given the above discussion of behavioral learningconcepts, it becomes clear that much of the valueof these concepts lies in their relationship to promo-tional tools. A review of relevant literature showsvery little work done concerning promotions andnone relating promotions to operant conditioning.'

Two recent papers considered promotions fromthe point of view of self perception theory. Scott(1976) found weak repeat behavior in response toa number of different one-time incentives, althoughsome (a discounted trial) did better than others (freetrial, free trial plus premium). From a behaviorallearning perspective the results are also weak, butserve as a case to support the notion presentedearlier that shaping procedures are generally poorlyused. In Scott's experiment the treatments wereall single incentives where consumers received sav-ings and/or gifts ranging in value from 25C to $1on a 50C item.

Since the item (a community newspaper) wasa low involvement item, there was low priorawareness, low subscriber level after introduction,and low cost of product; there would be no behav-ioral reason to suspect a one-time incentive toalter long run behavior. Given a low involvementproduct, behavior would need to be shaped slowlyover time, otherwise any behavior would be duemerely to the more highly involving financial incen-tive. When the high involvement incentive wasremoved, behavior would be expected to extinguish.This was, indeed, the case.

In Scott's case, the incentive, and not the prod-uct, was the primary reward for purchase behavior.When it was withdrawn,' the behavior was extin-guished. In contrast, if initial purchase were ac-companied by a small premium, the product andnot the premium would be the primary reward forpurchase behavior. When the premium is then

'There is a large body of work outside the traditional marketingliterature that essentially examines promotions and operant condi-tioning for issues such as transit usage, energy conservation, andcurtailing littering. This work, while relevant, was felt to beinappropriate here, since it comprises the body of mainstreambehavioral learning research and generally uses noncompetitivesituations of limited relevance to marketers.

^For a general discussion of involvement, see Houston andRothschild (1978). among others.


withdrawn, purchase behavior may be unaffected.Scott, quite rightly, suggests that incentives mayexist upon a continuum and that researchers needto identify the size/impact of various incentivelevels. Behavioral learning theory offers marketersa framework to study the impacts of incentivesupon purchase behavior. This framework includesconsideration of various types of reinforcers (pri-mary, secondary), various types of reinforcementschedules (continuous, intermittent), and the impactof any delay that might occur between a behaviorand its reinforcement (immediate or delayed rein-forcement). It seems essential, however, that anystudy of the incentive/purchase behavior relation-ship employ a shaping paradigm to provide thegreatest insight into the impact of consumer dealsupon long run behavior.

Dodson, Tybout, and Stemthal (1978) also ex-amined incentives from a self-perception frame-work, examined one-time incentives (mediacoupons, cents-off deals, and in-package coupons)for low involvement products (margarine and flour),and found weak but sometimes significant results.In this study there was strong brand switchingbehavior when dealing occurred, but little follow-upbehavior when deals were retracted.

Behavioral learning theory would suggest thatdeals cause brand switching because the deal ismore likely to be reinforcing than the product. Infact the authors found that the greatest switchingoccurred for the greatest financial incentive; nextmost switching occurred where there was less fi-nancial incentive but great ease of use (the cents-off deal required no prior purchase or coupon clip-ping). It is reinforcing to save money if little effortneeds to be expended.

When they considered repeat purchase theimpact of the deals was reversed. Now the in-pack-age coupon outperformed the media coupon andcents off deal. Operant conditioning would predictpoor results from the latter incentives because noshaping had gone on over time. While the in-packagecoupon also provided poor shaping, it was betterthan the others in that it provided a two step processleading to changed behavior. Again self-perceptiontheory may be inappropriate due to the low involve-ment nature of the products; repeat behavior maynot be attained because the one-time incentive wasnot powerful enough to overcome apathy and iner-tia.

A consideration of behavioral learning theorywould suggest that in both the case of a mediacoupon and a cents-off deal, the primary rein-forcement for purchase behavior was derived fromthe deal, not the product itself. Hence when the

deal was withdrawn, purchase behavior was extin-guished. The authors provide some support for thiswhen they suggest a media coupon may be advisableto induce switching when the dealt product isperceived to be superior on some relevant attribute.In this case one presumes that primary rein-forcement would be derived from the product itself;the deal would play a secondary role.

It is interesting to note that in the two reviewedpapers, self-perception and behavioral learningtheories make similar predictions as to the outcomeof the one-time incentive, but for different reasons.The two theories, though, suggest different strate-gies to alleviate the ineffectual one-time incentive.Both theories predict a low level of success, butself-perception predicts this because the most ef-fectual change will come when people attribute theirbehavior to irmer causes rather than extemal incen-tives; therefore, incentives should be minimized.Behavioral learning predicts greater behaviorchange through a shaping process and thereforemore incentives would be necessary.

Each theory would seem to have its place. Inhigh involvement situations where complex cogni-tive activity would seem to take place, self-percep-tion based strategy may be more appropriate. Inlow involvement cases where httle cognitive activityis necessary for adequate decision making, behav-ioral learning based strategy may be more appro-priate. The two cited studies were low involvementcases; and, therefore, the lack of success in eachshould be more easily remedied by behavioral learn-ing strategies.

In low involvement cases, Kassarjian has calledfor a return to simple models of behavior explana-tion. Self-perception and other cognitive theoriesassume a high level of involvement, since individualsare felt to go through a complex mental activity.If such activity does not take place, then a one-timeincentive will not be successful; shaping proceduresare needed to avoid early extinction.

Another set of contributing data is from a pro-prietary study reported by Ogilvy and Mather (n.d),who found that samples that include a coupon foran initial purchase have a 20% higher initial purchaserate than do samples without coupons. This is atest of a two step versus a one step shapingprocedure that clearly favors the two step proce-dure, and is consistent with the theory.

A final contribution to this discussion comesfrom Prentice (1975), who divided the commonpromotional tools into classes; those that contributeto the Consumer Franchise Building effort (referredto as CFB) and those referred to as non-CFB. Thiswork is a report on current industry practice and


is not based on any theoretical model. Conclusionsinclude that most promotional deals are too shorttemporally, and that for a promotion to work, abrand must provide value of its own. Behaviorallearning theory would support the following conclu-sions: Shaping takes time; promotions of shortduration cannot do the tasks asked of them; behaviortoward a brand of little value will rapidly extinguish,since there is no primary reinforcement.

In separating tools into CFB and non-CFBclasses, the author suggests that non-CFB tools donot emphasize product/brand value but rather keyon other issues. For example, price-off deals andrefund offers stress financial issues, premiumsstress other objects, and contests / sweepstakesstress games. CFB tools stress the product. Exam-ples here are samples and coupons. This classifica-tion is also intuitively consistent with behaviorallearning, which suggests that secondary reinforcers(money, objects, games) are less hkely to lead toappropriate behavior than will good product. TheCFB /non-CFB model has not been rigorously testedbut is generally consistent with behavioral learningtheory. Figure 3 summarizes the relationships be-tween the three models discussed above.

DiscussionIt can be seen that modem marketing thought isat least implicitly related to behavioral learning.Since repeat purchase is necessary for survival inthe marketplace, behavior must be positively rein-forced. Since new products are continuously enter-ing the market, shaping procedures can be usedso that trial is more than a random process. Inbehavioral learning terms, purchase is a behavior,and the product is a positive (or negative) reinforcer.This discussion will consider the theoretical issuesemerging from behavioral learning theory, researchquestions which must be answered before wholesaleadoption of these principles can be made, and thepotential implications of behavioral learning formarketing strategy.

Theoretical Issues Emerging from BehavioralLearning Theory

As Nord and Peter (1980) point out, there are manyareas where benefits to marketing can be derived,and one should consider the differences betweencognitive learning and behavioral leaming. Cogni-tive leaming (especially as it applies to marketing)

FIGURE 3Promotional

PromotionalTool

Samples

In-PackageCoupon

MediaCoupon

Price-OffDeal

Premiums

Contests/Sweepstakes

Refund Offer

Effects Predictions Made by the

BehavioralLearningTheory

Good, Positively Reinforcing

Good since it requiresapproximate repeatbehavior in order toredeem

Depends on method—need to use as shapingmechanism

Depends on method—need to use as shapingmechanism

Poor, does not reinforceappropriate behavior

Poor, does not reinforceappropriate behavior

Poor, reinforcement is toodelayed to be effective

Three Models

Self PerceptionTheory

Poor (Scott 1976)—Free trialdid not help

Good (Dodson, Tybout,Sternthal 1978), Led torepeat purchase

Poor (Dodson, Tybout,Sternthal 1978), Did notlead to repeat purchase

Poor—Did not lead torepeat purchase (Dodson,Tybout, Sternthal 1978)

Good—Led to repeatsubscription (Scott 1976)

No effect of premium plusfree trial (Scott 1976)

Consumer FranchiseBuilding Model (CFB)

(Prentice 1975)

Good, focuses on productvalue

Good, won't be usedwithout product value

Good

Poor, too much financialemphasis

Poor, nonproduct emphasis

Poor, nonproduct emphasis

Poor, too much financialemphasis


takes place without reinforcement, and can, indeed,be forced upon consumers through the judicioususe of repetition in advertising (Ray et al. 1973)and can occur as a result of mere exposure.

Behavioral learning occurs through the use ofreinforcers. While Ray shows that trial purchasebehavior may also occur as a function of advertisingrepetition, such behavior is often stimulated by theuse of promotional devices. Repeat behavior, onthe other hand, is certainly based on reinforcementof the trial. Kmgman (1965) posits that this rein-forcement leads to a positive attitude after the trial.(A behaviorist would not add the hypothetical con-struct of attitude onto a discussion of these events.)

Two issues emerge;

• Reinforcement is necessary to behavioralleaming, especially beyond the trial stage.Advertising (modeling, vicarious learning)can help in behavioral leaming, but product,price, and place must be favorable in thelong run.

• Cognitive leaming focuses on the internal-ization of messages rather than the learningof behavior patterns. In behavioral leamingthe message cue announces the upcomingreinforcement opportunity.

Returning to the concept of attitude, the behav-iorist feels that since he/she carmot observe anystate of consciousness, he/she should merely beconcemed with that which is observable (i.e., behav-ior). This is a very different concept from thatheld by consumer researchers who use a socialpsychology model. It is not, though, inconsistentwith notions concerning low involvement (Kassar-jian 1978). As Kassarjian suggests, perhaps themiddle range social psychology theories are bestreserved for high involvement cases, and behaviorallearning theory should be used in low involvementcases.

In recent years the notion of involvement hasbecome popular in consumer behavior (e.g., Hous-ton and Rothschild 1978). Part of its popularity isdue to its intuitive appeal as a simpler model ofbehavior development for unimportant decisions.It is in these cases of decision making when con-sumers have low involvement and attempt to satis-fice, that behavioral leaming theory has its greatestpotential. As involvement increases, cognitiveprocesses become more complex; here the middlerange theories of cognitive dissonance, attribution,personahty, perceived risk, and so on have theirvalue. In low involvement cases, behavioral learningtheory may be most relevant for consumer behavior.

While economic man may be an appropriate modelfor high involvement cases, marketing man (Bagozzi1975) is more the prototype for low involvementcases.

Research Issues Emerging from BehavioralLearning Theory

There are some questions that have not been for-mally raised by marketers but should be exploredbefore there can be wholesale adoption of behavioralprinciples. Each of these has been explored in thebehavioral learning hterature but none has beenresolved in a marketing context:

• Behavioral learning works well in the labora-tory because the enviroimient can be con-trolled by the experimenter. In the competi-tive marketplace the marketer has much lesscontrol; hence, research is needed to assesstbe impact of the low degree of control themarketer can exercise on the predictions ofbehavioral leaming theory. For example, dosimultaneous choice situations necessitate theuse of continuous reinforcement schedules?

• Is shaping through successive approximationan efficient/effective method for inducingnew product trial and long-term purchasebehavior? Would the employment of shapingprocedures have decreased the attrition ratesfound by Dodson, Tybout, and Stemthal(1978) and Scott (1976) when deals were usedon a one-time basis?

• Are premiums and other promotional devicesmore effective when based upon primaryversus secondary reinforcers? Are premiumsand other promotional devices more effectivewhen based upon immediate versus delayedreinforcement? Is there an interactive effectbetween type of reinforcement (primary, sec-ondary) and the delay between behavior andreinforcement (immediate, delayed)?

Figure 4 dichotomizes both types of rein-forcement and delay between behavior and rein-forcement. Behavioral leaming theory suggestsgreater long run rates of behavior change will beexperienced when respondents are moved througha n A ^ B - ^ D - ^ E o r A - ^ C ^ D ^ E sequencethan when any one treatment is presented in isola-tion. This test of the greater value of shapingprocedures versus a one-time consumer deal wasnot performed in the studies of Scott or Dodson,Tybout, and Sternthal.


FIGURE 4The Impact of Reinforcement Type andReinforcement Delay Within the Context ofBehavioral Learning Theory

Crpiter 1

crater .»e.i.tB

Delayed

Primary

PriBsry (A)

DclavedPfinjry (C)

Secondary

Secondary IB)

CBlivedSetoTidory (D)

(E> No

There are also situational types of researchquestions that can be raised;

• Are there differences in the above questionsthat are a function of high or low involvementsituations?

• Can the issues of behavioral leaming theorygive insight to increasing survey responserates through the use of behaviorally basedincentives for respondents?

• Can sales force compensation schedules bemade more effective through the use ofbehavioral leaming theory?

Application and Strategy Issues Emerging fromBehavioral Learning TheoryTurning to a more apphed view of behavioralleaming, there are a number of concepts discussedabove which may lend themselves to the moreorderly and efficient development of marketingstrategy. The following points are suggested by theliterature reviewed above and by Nord and Peter:

Motivation and meeting needs. Both the market-ing concept and behavioral learning stress meetingneeds. Appropriate long run behavior only takesplace when the reinforcer meets some need. Areinforcer can't be positive if it does not meet needs.

Controlling the environment. Behavioral leamingworks well in the laboratory because the environ-ment can be controlled. In the marketer's worldthere is much less control; therefore, the 4 P's mustbe organized. Due to the lack of total control, thoseareas where control is possible must be utilizedto their fullest.

Shaping through successive approximation. Itmay be possible to aid consumers in their acquisitionof new behaviors through the use of shaping proce-dures. In cognitive terms, risk is reduced; in behav-

ioral terms, any behavior is a minor change fromthe preceding one.

Immediate reinforcement. In addition to creatingtime utility, in itself reinforcing, immediate rein-forcement is strong because it is linked most directlywith the appropriate behavior.

The need for systematic and consistent applica-tion. Behavioral learning has worked best in labora-tory enviroimients where the world of the subjectis orderly and predictable (to the subject). In trans-ferring these principles to the marketplace, behav-ioral leaming theory suggests that reinforcersshould be immediate, should be as described inthe discriminative stimulus, and should be consis-tent.

The use of promotional tools. These tools canbe used in both the shaping of new behavior andthe reinforcement of existing behavior. There isa tendency to overuse these aids; as a result, adependency develops so that removal of the promo-tion leads to extinction of behavior. Promotionaltools should not overshadow the products they areaiding (if long-term behavior toward the productis desired).

In addition to the similarities between behavioralleaming theory and marketing, and the insights thatthe former can provide to the latter, there are alsoa number of problem areas which need to beconsidered. For example, in the controlled environ-ment where behavioral research is generally con-ducted (e.g., a Skinner box), the experimenter hastotal control of the situation and absolute power;there are no competitive choices available to thesubject. Such a situation rarely exists in the privatesector marketplace; as a result, some behavioralnotions cannot be transferred.

A Concluding CommentBehavioral leaming theories offer a frameworkwithin which to organize and structure marketingand promotional activities, and a simple but elegantmodel of the generic concept of marketing. Inmarketing, the desired end is appropriate behaviormanipulation and control to further the goals ofthe organization. The currently recognized mostefficient means to these ends is through the useof the marketing concept. By developing reinforcerswhich meet needs, marketing would seem to havealready embraced behavioral learning theory. Be-havior is a function of its consequences and en-vironment; by assessing needs, marketers are bestable to meet their own needs and control portionsof the consumer's environment.

This model may be appropriate for many of the


". . . unimportant, uninvolved, insignificant, minordecisions that are made in the marketplace everyday. . ." (Kassarjian 1978, p. xiv). In these situations.

consumers have weak cognitive processes, and asa result, behavioral leaming theory can offer in-sightful direction to the marketer.

REFERENCESBagozzi, R. P. (1975), "Marketing as Exchange," Journal

of Marketing, 15 (October), 32-39.Dodson, J. A., M. A. Tybout, and B. Stemthal (1978),

"Impact of Deals and Deal Retraction on Brand Switch-ing," Journal of Marketing Research, 15 (February),72-78.

Houston, M. J. and M. L. Rothschild (1978), "A Paradigmfor Research on Consumer Involvement,'' working paper.University of Wisconsin, Xl-ll-Ad.

Kassarjian, H. H. (1978), "Presidential Address, 1977:Anthroposorphism and Parsimony" in Advances in Con-sumer Research, Volume V., H. K. Hunt, ed., Ann Arbor:Association for Consumer Research.

Krugman, H. (1965), "The Impact of Television Advertising:Leaming without Involvement," Public Opinion Quarter-ly, 29 (Fall), 349-356.

Nord, W. R. andJ. P. Peter(1980) "A Behavior ModificationPerspective on Marketing," Journal of Marketing, 44(Spring), 36^7.

Ogilvy and Mather (n. d.), How to Create Advertising ThatSells, New York: Ogilvy and Mather.

Prentice. R. M., (1975). "The CFB Approach to Advertis-ing/Promotion Spending," in The Relationship BetweenAdvertising and Promotion In Brand Strategy, R. A.Strang, ed., Cambridge: Marketing Science Institute,75-90.

Ray, M. L., A. G. Sawyer, M. L. Rothschild, R. M. Heeler,E. C. Strong, and J. B. Reed (1973), "Marketing Com-munications and the Hierarchy of Effects," in NewModels for Mass Communications Research, Volume II,Sage Annual Reviews of Communications Research, P.Clarke, ed., Beverly Hills, CA: Sage.

Scott, C. A. (1976), "The Effects of Trial and Incentiveson Repeat Purchase Behavior," Journal of MarketingResearch, 13 (August), 263-269.

Skinner, B. F. (1953), Science and Human Behavior, NewYork: The Free Press.

Strang, R. A. (1976), "Sales Promotion—Fast Growth,Faulty Management," Harvard Business Review, 54(July-August), pp. 115-124.

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

ADVERTISERS' INDEX

Addison-Wesley PubUshing Company Page 168American Warehousemen's Association Page 165Burke Marketing Services, Inc Cover 4The Ehrhart-Babic Group Page 1Elsevier North-Holland, Inc ; Page 3Harper & Row, Publishers Page 167Richard D. Irwin, Inc Cover 2, 3, Page 166Martinus Nijhoff Publishing Page 4John Wiley & Sons, Inc Page 2


behavioral learning theory: its relevance …special... · greatest value may be in the development...

Documents