cognitive response

8/7/2019 cognitive response

1/12

REPRINTED PROMPSYCHOLOGKAL FOUNDATIONS OF ATTITUDES01968ACADEMIC PRESS INC., NEW YORK

Cognitive Learning, CognitiveResponse to Persuasion, and Attitude Change1

ANTHONY G. GREENWALDDEPARTMENT OF PSYCHOLOGY

OHIO STATE UNIVERSITYCOLUMBUS, OHIO

It is a common assumption that the effectiveness of a persuasivecommunication is, at least in part, a function of the extent to whichits content is learned and retained by its audience. This assumedlearning-persuasion relation is based on a reasonable analogy be-tween the persuasive communication and an informational com-munication such as a classroom lecture. In th e lec ture, it is by defi-nition of the educational situation that retention of content is taken asa measure of effectiveness. In the persuasion situation, however, theessential criterion of effectiveness is acceptance of content . It remainsan empirical question to determi ne whether acceptance of a persua-sive communication is related to retention of its content.

The hypothesis that acceptance of a communication is, in some part,a function of learning or retentio n of its content has received explic itendorsement by a number of attitude researchers and theorists(e.g., Hovland, Janis, & Kelley, 1953; McGuire, this volume; Miller

'Preparation of this report and the research reported h ere wer e supported in largepart by grants from the National Science Foundati on (GS-1601) and the Mershon SocialSciences Program at Ohio State University. The author is particularly indebted toRosita Daskal Albert, Dallas Cullen, Robert Love, and Joseph Sakumura who haveparticipated actively in various phases of the research reported here. A condensedversion of this chapter was pre sented at the American Psychological Association sympo-sium, "Alternatives to consistency theory in the study of attitude change," Washington,D.C., September, 1967.


2/12

148 6. COGNITIVE LEARNING AND COGNITIVE RESPONSE ANTHONY G. GREENWALD 149& Campbell, 1959; Watts & McGuire, 1964) and has aroused n opublished opposition. Indeed, this cognitive learning model ofpersuasion is most reasonable. I t is widely accepted that cognitionsbearing on th e object of an attitude form a major component of thestructure of the attitude toward that object (see, for example, Camp-bell, 1947; Katz & Stotland, 1959; Krech, Crutchfield, & Ballachey,1962; Rokeach, 1960; Smith, Bruner, & White, 1956). Since the indi-vidual is not born with his cognitions, but acquires them, there seemsto be no reasonable alternative to the assumption that cognitionsbearing on attitude objects are learned. Further, the most obvioussource of such cognitions is the wealth of persuasive messages towhich o ne is exposed via the public communications media as wellas through face-to-face communications.

In light of the overpowering reasonableness of the persuasion-as-a-function-of-retention hypothesis, it is rather surprising how un-supporting the research evide nce is. A few studies have directlyexamined the relation between the learning and attitudinal effectsof persuasive communications (Insko, 1964; Miller & Campbell,1959; Watts & McGuire, 1964). These studies have generally foundthat both communication retention and persuasion diminish withincreasing time between communication and posttest, consistentwith the hypothesis that retention is necessary for persuasion. Onthe othe r hand, these same studies have found only weak and variablecorrelations be tween communication retention and persuasion amongsubjects tested at the same posttest interval, suggesting that therelation betwe en retention an d persuasion is not a necessary one. Ina conceptually relevant study on impression formation, Anderson andHubert (1963) have concluded that there are separate memory sys-tems for retention of a set of person-des criptive adjectives and of theperson-impression derived from them; their results thus also suggestlittle or no necessary relation betwe en a communication's retentionand its effectiveness. Additionally, in a numbe r of studies scatteredthroughout the attitude literature (and reviewed in Ho vland et al.,1953), variables sh own to affect opinion demons trably - uch as credi-bility, fear arousal, and organization of argum ents- have not be enfound to have corresponding effects on retention of communicatedarguments. Such negative findings on retention measures have typi-cally been used to counter any possible interpretation for obtainedattitude change differences in terms of unintentionally induceddifferences in attention to or retention of communication content.In sum, the research evidence must be interpreted as uncongenialto the hypothesis that persuasion is a function of retention of per-

suasive arguments. (In partial reviews of the relevant literature,Insko, 1967, and McGuire, 1968, reach similar conclusions.)

It must be concluded that either (a) learning of attitude-relevantcognitions (i.e., persuasive argu ments) is unre lated to attitude forma-

t tion an d chan ge; or (b) persuasive communications can indu ce atti-tude change without necessarily providing the cognitive content onwhich the attitude is based. Th e first conclusion carries the implica-tion that learned cognitions are not fundamental to the structure ofattitudes. Rather than accept this conclusion, which run s counter tomost conceptions of attitud e, it seem ed worth some effort to exploresources other than persuasive communications as possible originsof learned attitudinal cognitions.

C O G N I T I V E R E S P O N S E S T O P E R SU A S IO NThere is, of course, an important extracommunication source of

cognitive content in th e persuasion situation: the cognitive reactionsof the communication recipient to incoming persuasive information.When a person receives a communication an d is faced with the deci-sion of accepting or rejecting the persuasion, he may be expectedto attempt to relate the new information to his existing attitudes,knowledge, feelings, etc. In the course of doing this, he likely re-hearses substantial cognitive content beyond that of the persuasivemessage itself. The present hypothesis is, then, that rehearsal andlearning of cognitive responses to persuasion may provide a basisfor explaining persisting effects of communications in terms of cogni-tive learning. The learning of cognitive response content may,indeed , be more fundamental to persuasion than is the learning ofcommunication content.

This hypothesis is not a new one. The following passages fromprevious works indicate views bearing the essence of the presenthypothesis.

. . . there is reason to expect that those audience members who are alreadyopposed to the point of view being presented may be distracted [from thecontent of a communication] by "rehearsing" their own arguments while thetopic is being presented and will be antagonized by the omission of the argu-ments on their side [Hovland, Lumsdaine,& Sheffield, 1949,p. 2011.

When exposed to [a persuasive communication], a member of the audienceis assumed to react with at least two distinct responses. He thinks of his own[opinion], and also of the [opinion] suggested by the communicator. . . .Merely thinking about the new opinion along with the old would not, in itself,


3/12

ANTHONY G. GREENWALD 151lead to opinion change. The individual could m em or i z e the content of the [newopinion] while his opinion remained unchanged. Practice, which is so im-portant for memorizing verbal material in educational or training situations,is not sufficient for bringing about the accep t ance of a new opinion [Hovlande t a l . , 1953, p. 111.

It was hypothesized that conformity in the communication situation willincrease attitude change to the extent to which implicit s uppor t i ng responsesare produced, and decrease attitude change to the extent to which implicitinterfer ing responses are produ ced. . .By supporting response is meant any implicit response made by the indi-vidual (usually a self-verbalization), which prov ides argum ents in favor ofthe overt response he makes; which produces further motivations in thedirection of the overt response; or which relates the overt response to otherstimulus situations. By interfering response is meant any implicit responsemade by the individual which provides motivation against the overt responsehe makes; which limits the stimulus situations to which the overt responseis applicable; or which is generally irrelevant (such as aggressive or dis-tracting responses) [Kelman, 1953, p. 187, 2111.

Despite a number of speculations similar to the ones just cited,there has been no direct experimental exploration of the role ofcognitive responses in persuasion, and, in fact, there is not muchresearch that is even relevant. The research on active participationin the communicat ion process (e.g., Hovland et al., 1953, Ch. 7)comesclosest to being relevant. Elsewhere, isolated experiments (e.g.,Brock, 1967; Janis & Terwilliger, 1962; Kelman, 1953) have explo reddependent variables approximating the present conception of cogni-tive responses to persuasion. Research in which cognitive responsesto persuasion are employed as independent variables in experi-mental persuasion situations is particularly needed. Before pro-ceedi ng to a consideration of evide nce collected in the author'slaboratory, it will be useful to state the present hypothesis withsome precision.

It is proposed that the persuasion situation is usefully regardedas a complex stimulus that evokes in the recipient a complex cogni-tive response. The essentia l dimensions of the recipient's cognitiveresponse are, at t he least, (a) response content, i.e., deg ree of ac-ceptance versus rejection of the position advocated in th e communi-cation, and (b ) intensity, or vigor, of response. The latt er dimension,as well as other possible dimensions of cognitive response, will notbe considered further in this chapter. The essential components ofthe persuasion situation as a stimulus- hat is, as determinant of thecognitive response content- re setting, source, and communicationcontent. An additional major set of determinants of the cognitiveresponse content is the set of characteristics brought by the recipient

to the persuasion situation, including his existing repertory of atti-tude-relevant cognitions as well as personality traits and groupmemberships.

As in many othe r treatments of persuasion, the cognit ive responseanalysis assumes that attitude change can be a chieved by the modifi-cation, through learning, of the recipient's repertory of attit ude-relevant cognitions. Such modification might include strengtheningof existing cognitions as well as introduction of new ones. Thepresent emphasis on the mediating role of the recipient's owncognitive responses to persuasion may be formulated as an assertionthat cognitive modification of attitu des requires active (no t neces-sarily overt) rehearsal of attitude-relevant cognitions at a time whenthe attitude object or opinion issue is salient. Thus the effects ofpersuasive communications might range from persuasion -whe n therecipient rehearses content supporting the advocated position-toboomerang-when the recipient rehearses content opposing theadvocated position.

As a consequence of the present emphasis on the recipient'srehearsal of his own responses to persuasion, it is assumed thatlearn ing of communication content does not play an essential rolein mediating the effects of persuasive communications. The presentformulation, therefore, is capable of maintaining an analysis ofpersuasion effects in terms of cognitive learning while being com-patible with findings indicating no necessary relation betweencommunication retention and persuasion.

It is possible to formulate the cognitive response analysis in termsof an analogy to the classical conditioning paradigm. In this analogy,the persuasion situation corresponds to the unconditioned stimulusin that it has a response evocation capacity; that is, it influences thecontent of the recipient's cognitive response. As an analog of theunconditioned response, the cognitive response becomes transferredto the attitude object, which is analogous to the conditioned stimulusof the classical paradigm. While this analogy may be decidedlyuseful, especially in relating t he present analysis to other treatmentsthat have invoked the classical conditioning model (see the chapte rsby Staats and Weiss in this volume), it would be inappropriate cur-

I rently to regard the m odel as more than a possibly suggestive analogy.A point of difficulty that would a rise if the model is taken li terally,for example, would concern the nature of the conditioned responsein persuasion; it would be unnecessarily cumbersome, at this stage,to incorporate in the present analysis an analogy to the conditioningmodel's assumption that conditioned responses are ei ther fractional


4/12

ANTHONY G. GREENWALD 153components of unconditioned responses or preparatory adjustmentsto unconditioned stimuli.STUDIES OF COGNITIVE LEARNING AND ATTITUDECHANGE

The cognitive response analysis of persuasion has guided a seriesof experiments cond ucted at Ohio State University. While the pri-mary focus of this research has concerned the properties and func-tions of recipients' cognitive responses in persuasion situations, asecondary focus has been on the role of communication content inpersuasion. The present section is devoted largely to the findings ofcompleted portions of this research program. Th e first two subsectionsdeal with experiments in which communication content learning andcognitive response rehearsal have been employed as manipulatedindependent variables, enabling conclusions about their causalinvolvement in persuasion. The next two subsections are concernedwith experiments in which retention of communicated argumentsand occurrence and retention of cognitive responses have beenobserved as dependent variables; from these experiments it ispossible to draw conclusions about the determi nants and correlatesof cognitive response content and of retention of cognitive responsesand communication content. A final subsection provides a briefreview of literature on variables in fluenc ing the reten tion of com-municated persuasive arguments.

It is to be emphasized that the research program described hereis very much in progress. At approp riate points below, the natureof planned and in-progress research will be indicated; these com-ments will serve, perhaps, more to indicate the author's awarenessof questions that remain to be answered than to enlighten the reader.

Previous studies of the relationship between communication re-tention and opinion change have tested this relationship correla-tionally- ubjects received a communication and were subsequentlytested for both retention of content and acceptance of the viewpointof the communica tion. With the assistance of Rosita Albert and DallasCullen, th e author conducted a study (unpublish ed) in which com-munication retention was a manipulated indepe ndent variable, withopinion as the dependent variable. Such a design, it may be noted,is more appropriate to drawing a conclusion about the causal role of

communication learning in attitude change than is a correlationaldesign.

The experiment employed three communications concerningUnited States foreign aid policy. One of these con sisted of 6 briefarguments favoring foreign aid, 1 of 6 arguments opposing foreignaid, while the third was composed of all 12 of these arguments anddid not draw either a favorable or unfavorable conclusion. For eachcommunica tion, a group of subjects was asked to rehearse th e indi-vidual arguments contained in it, in preparation for a subsequentretention test; in addition to these three groups, two other groupsreceived the 12-argument communication, and, after briefly readingit once, in entirety, were asked to rehearse only 6 arguments that hadbeen underlined. For one of these groups, the 6 underlin ed argumentswere those favoring foreign aid, while for the other group the 6opposing foreign aid were underlined. Subjects were tested foropinion on th e issue and then for retention of all arguments in theircommunications (not just rehearsed ones) immediately followingthe learning task and again, unexpectedly, after a 1-week interval.

For the group receiving the one-sided pro-foreign aid message(N = 16), an average of 3.7 (out of 6) arguments wer e retaine d for theimmediate retention test and 3.1 for the unexpected retention testone week later. The corresponding means for the group receiving theone-sided anti-message (N=20) were 3.6 arguments retained imme-diately and 2.4 after a week. Since these two groups differed signifi-cantly from each other in opinion in th e directions advocated in theircommunications, both immediately (F=9.50, df =1, 34, p < .01) andafter a week (F=7.03, p < .02), these data were consistent with thehypothesis that communication learning and persuasion are related.Among subjects receiving the two-sided communications, those as-signed to learn pro arguments (N = 18) retained substantially morepro than con arguments (p< .001, for each testing); th e reverse wastrue, as expected, for those (N=19) asked to rehearse con arguments(p < 001, for each testing); and those asked to learn both sets (N = 18)retained approximately equal numbers of each.

The effectiveness of the partial learning assignments for thetwo-sided communications can be indicated by th e fact that subjectsretained an average of 3. 0 assigned arguments compared to 0.3 non-assigned ones on the immediate retention test; for the delayed re-tention test, the corresponding means were 1.8 assigned and 0.3nonassigned arguments. If there is, indeed, a causal relation betweenargument learning and persuasion, these manipulated differences inretention s hould have produced corresponding differences in opinion.


5/12

154 6. COGNITIVE LEARNING AN D COGNITIVE RESPONSE ANTHONY G. GREENWALD 155However, there we re no opinion differences among the three groupsreceiving two-sided communications, either immediately (F< 1,df= 2, 52) or after a week ( F< 1).

It must be concluded that learning of communication content isnot a sufficient condition, and perhaps not eve n a necessary condition,for persuasion. That is, learned arguments supporting one side of theforeign aid issue were ineffective when, at the time of learning,subjects were made aware that credible opposing arguments wereavailable- ven though the opposing arguments were poorly retained.This finding stands in contrast with those of studies demonstratingacross-cell con-elations between learning and persuasion (e.g., Insko,1964; Miller & Campbell, 1959; Watts & McGuire, 1964). The lack ofopinion differences between conditions with decided retention dif-ferences in the present study suggests that such previously obtainedacross-cell correlations should not be interpreted in terms of a causalrelation between communication retention and persuasion.ACCEPTANCEND RECALL F IMPROVISEDRGUMENTS

If, as is presently supposed, attitudes change in the direction ofcognitive content rehearsed during a persuasion situation, then pro-cedures that manipulate the content of the recipient's cognitiveresponses should have persuasive effect. A traditional persuasionprocedure that may be viewed as a manipulation of cognitive responsecontent is t he improvised role-playing procedure in which a subjectis asked to deliver a persuasive message supporting a positioninitially unacceptable to him. The majority of research evidenceindicates that role playing produces greater persuasion toward theunacceptable position than does passive receipt of a persuasivecommunication (cf. Insko, 1967, p. 222). Such results are quitecompatible with the present point that a communication recipient'srehearsal of his own arguments may be more important in per-suasion than is his rehearsal of arguments contained in a communica-tion to which h e is exposed. However, some commentators feel thatthe currently available evidence on role playing is equivocal (seeMcGuire, 1966, p. 498). Because of this empirical uncertainty, thepresent research program included an experiment intended toassess the effect of improvised role playing on both acceptance andretention of arguments.

In this experiment (Greenwald & Albert, 1968), each subjectimprovised five arguments in response to an assignment to advocateeither specialized (career preparatory) or general (liberal arts) under-

graduate education. Improvisation was obtained in response to fiveneutrally worded questions that could be answered with an argumentsupporting either viewpoint. Assignment to positions was randomwith approximately 90 subjects being assigned to each side. In addi-tion to being exposed to their own improvised arguments supportingone side of the issue, subjects carefully read and studied-for aboutthe same amount of time they had spent improvising-a set of argu-ments supporting the opposite side that was actually written byanother subject in the study. Since each subject's improvisationsserved once as an improvised set of arguments and once as an ex-ternally originated set (for a different subject), this procedure servedto equa te quality of arguments for the two sets over the sam ple ofsubjects, although not necessarily for each subject. After a 20-minuteirrelevant task, the subjects were tested for opinion on the general-specialized education issue and were then asked to recall as manyarguments as they could of those to which they had been exposed-both their own and the ones that had been improvised by anothersubject.

Figure 1 gives the opinion and retention results for this study.It may be seen that subjects arrived at opinions consistent, on theaverage, with the position to which they had been assigned. Theopinion difference between the two groups was statistically significant(F = 7.42, df=1, 177, p < .01). A more powerful effect was obtainedwith t he retention measure, with subjects tending to recall much moreof their own improvised arguments than they did of those improvisedby another subject (F= 168.10,df= 1,777, p < .001).A supplementaryfinding was that subjects tended to rate their own arguments as moreoriginal than the other's ( p< .001), indicating that, other things beingequal, one tends to evaluate his own thoughts more favorably thanothers'. In summary, these results demonstrated that the subject'srehearsal of his own cognitions and externally originated cognitionsof comparable quality tended to result in attitude change in the direc-tion of the content of the subject's own cognitions. These results sup-port the gene ral trend of findings in the role-playing liter ature . Moreimportantly for present purposes, they add force to the hypothesisthat the recipient's rehearsal of his own cognitive responses plays animportant role in persuasion.

It should be noted that t he Greenwald-Albert experiment does notprovide an explanation of the special efficacy of the recipient 's ownimprovised arguments. It could be that subjects' increa sed retentionof their own arguments was responsible for the observed attitudeeffect; however, alternative explanations are possible in terms of


6/12

156 6. COGNITIVE LEARNING AND COGNITIVE RESPONSE ANTHONY G. GREENWALD 157

Opinion data

Specialized

General

Retent~ondata

I SpecializedGeneral I

SD units for mean of 90 ubjectsFIG.1. Opinion and retention as a function of position assigned for improvisation.

(Bars are labeled with the assigned position. Positive scores on the opinion measurerepresent favorableness to specialized education; positive scores on the retention mea-sure represent greater recall of arguments favoring specialized education than of onesfavoring general education.) (Greenwald & Albert, 1968.)perception of the self as the source of arguments or in terms ofsubjects' ability to genera te argument s particularly effective in regardto their own attitude structure (cf. Hovland e t al . , 1953, pp. 233-237).Aseries of experiments designed to select among these explanationswill be conducted in the near future as part of the present researchprogram.CONTENT ND CORRELATESF COGNITIVE ESPONSES TOPERSUASION

A series of experiments conducted by Dallas Cullen, under theauthor's supervis ion, explored the content of subjects' thoughts

relevant to th e topic of a persuasive communication that they had justreceived. T he communication, on the issue of general versus special-ized undergraduate education, advocated the view that college educa-tion should be directed specifically at career preparation (that is,should be specialized). It was assumed that cognitive effects of thecommunication could be assessed by comparing the issue-relevantthoughts of subjects who received this communication with thoughtson the same issue expressed by control subjects who had receiveda communication on a different topic.

The subjects in this series of experiments were Ohio State Univer-sity introductory psychology students who participated in classroomsize groups. Four separate experiments were conducted with onlyminor variations in procedure. The basic procedure for Experiment Iincluded an introduction to the issue of general versus specializededucation consisting mainly of a definition of the alternative posi-tions. The n a communicatio n of about 250 words contai ning 12 dis-tinct arguments supporting specialized undergraduate education wasadministered in printed form. Control subjects, at this time, read acommunication concerning instructional television. Next, all subjects(including controls) were asked to "collect their thoughts" on theissue of general versus specialized undergraduate education bylisting thoughts that were:

. . . pertinent to forming and expressing an opinion on the issue of generalversus specialized education. These thoughts may consist of (a) informationfavorable to on e or the other viewpoint; ( b) personal values of yours that arefavorable to on e or the other viewpoint; (c) features of either viewpoint thatyou perceive as good; (d) features of either viewpoint that you perceive as bador harmful; and (e) any other thoughts you feel to b e pertinent .

In writing down these thoughts, please separate your thoughts into individ-ual ideas to be written do wn separately. An "individual idea" is one that, tothe best of your judgment, expresses only a single fact, value, good or badfeature, or thought.

Following this "thought-listing" procedure, subjects completed abrief questionnaire of four Likert-type items measuring opinion onthe general-specialized education issue. Finally, subjects wereinstructed to look back at the thoughts they had listed concerninggeneral and specialized education and to judge, for each individualthought, whether it was favorable to general or specialize d educationand how favorable it was, on a three-p oint sca le of slightly favorable(I) , moderately favorable (2), and very favorable (3) .

Experiment I1 added an opinion pretest to the basic procedure; it


7/12

158 6. COGNITIVE LEARNING AND COGNITIVE RESPONSE ANTHONY G. GREENWALDconsisted of th e four Likert-type items also used as the opinion1,osttest.

For Experiment 111, subjects were given an additional judgmenttask, at the end of the experiment, in regard to their listed thoughts.They were to assign each thought to one of three categories: (a) thosehaving their source in the experimental materials (the introductorydefinitions and th e persuasive communication); (b) modifications ofthe experimental materials (such as illustrations of, qualifications of,and reactions to communicated arguments); and (c) deas not traceableto the experimental materials. These thr ee categories are to be identi-fied here, respectively, as (a)externally originated, (b) recipient-modi-fied, and (c) recipient-generated cognitions. Th e aim of Experiment111 was to compare subject co ding of listed-thought responses intothese t hree categories with inde pende nt coding by judges; no controlgroup was employed.

In t he last experiment in this series, two different communicationconditions were employed. One (group IVA) employed the same com-munication used in the previous experiments while the other (groupIVB) was modified by addi ng material acknowledging opposing argu-ments (ones favoring general education) and refuting them wherepossible (cf. Hovland e t a l . , 1949, Ch. 8). Subjects in Experiment IValso performed the additional categorization of their listed thoughtsthat had been requested of subjects in Experiment 111.

For Experiments I, 11, and I11 the classification of thoughts intoexternally originated, recipient-modified, and recipient-generatedcategories was performed by judges who were able to agree on theseclassifications for 859% of their judgments. This extent of agreementwas not considered entirely satisfactory, especially when it wasfound, in Ex perime nt 111, that judges' classifications agreed withsubjects' classifications for only 62% of judgments (chance agreementwould be 33%). Since subjects were able to perform t he judging taskwith relative ease, it was decided that it would be most satisfactoryto use subject self-scoring for subsequent groups.

The data for the four experiments we re first examined in terms ofthe quantitie s of thoughts as distributed among the three categories-externally originated, recipient-modified, and recipient-generated.These data are given in Fig. 2. Judges' categorizations were usedfor Experiments I and 11; for Experimen ts I11 and IV, subjects' owncategorizations were used. The most significant feature of the datasummarized in Fig. 2 is that the recipient-generated category ac-counted for the majority of thoughts listed by communication-receiving subjects -despite the fact that their thoughts were tapped

Externally originated thoughts7

Reci~ ientmodified thoughts

8 Recipient generated thoughts

I I1 IU I T A X B 1 1 1 mCommunication receiving groups Control groups

FIG. . Mean quantities of listed thoughts in t hree categories. (Numb ers of subjectsare as follows: communication-receiving groups: I (48), I1 (48), 111 (68), IVA (26), IVB(28);control groups: I (16), I1 (16), IV (13).) (Greenwa ld & Cullen, unpublished data.)

immediately after reading a communication containing a dozen dis-tinct relevant thoughts that could have been listed.*

Subjects' judgments of the position supported by each of theirlisted thoughts were used to calculate an index summarizing thethoughts' content on the general-specialized education issue. It willbe recalled that each thought was weighted according to its degreeof support for the position it supported. T he index was calculated bysubtracting the sum of weights for thoughts favoring general educa tion

2Thequantities of listed thoughts in the externally originated and recipient-modifiedcategories for control groups were greater than zero bec ause control subje cts could anddid list or react to material contained in introductory definitions of the concepts ofgeneral and specialized education.


8/12

160 6. COGNITIVE LEARNING AND COGNITIVE RESPONSE ANTHONY G. GREENWALD 161Externally originoted thoughts

.40

.200

Recipient modified thoughts

Rec~pientgenerated thoughts

Posttest attitude responses

I I1 III D A I Y B I 11 ECornmunicotion receiving groups Control groupsFIG.3. Mean directional content of listed thoughts in three categories and mean

posttest attitude responses. (The extreme scores on the three category indexes are+ 1.00 and - .00; for the posttest attitude responses, the extremes are +2.50 and- .50.) (Greenwald & Cullen, unpublished data.)from the sum for those favoring specialized education, then divid-ing this difference by t he sum of weights for all thoughts. This calcu-lation was do ne separately for thoughts in each of the three categories- xternally originated, recipient-modified, and recipient-generated.Figure 3 gives the mean values of these directional content indexesfor the various groups in the present series of experiments. Forcomparison, the mean posttest opinion questionnaire responses aregiven at the bottom of Fig. 3. The effects of th e communications canbe seen in the generally more positive index values and posttestopinion scores for communication groups than for controls. The

significance levels for these comparisons, combining over Experi-ments I , 11, and IV (no control group was used in Experim ent 111)were: externally originated category, p < 02; recipient-modifiedcategory, p < .O2; recipient-generated category, p < 07; and posttestopinion, p < OO1 (all one-tailed).

It may be noted that the communication received by subjects ingroup IVB -which differed from the others in that it acknowledgedopposing arguments -prod uced the most favorable posttest opinionquestionnaire responses and the most positive responses in therecipient-generated category, while producing the least positiveresponses in the externally originated category. This pattern offindings is suggestive of the importance of recipient-generatedcognitions, relative to externally originated ones, in persuasion.

Figure 4 gives for each group th e correlations with posttest opinionfor the three category directional content indexes as well as for onebased on all thoughts combined. It may be noted that the correlationsinvolving the recipient-generated category index were quite highrelative to those for the externally originated category. This suggests,once more, the importance of recipient-generated cognitions in therecipient's attitude str~cture.~The correlations with posttest opinion for the recipient-generatedcategory index and for the index based on all thoughts combinedwere sufficiently high to suggest that the thought-listing procedureused in the present experiments might be very useful as a measureof opinion. A very desirable aspect of the thought-listing procedureis that it is applicable to virtually any attitude issu e without necessityfor time consuming scaling and item selection procedures. Researchcurrently underway, being conducted by Cullen, is exploring thereliability, validity, and sensitivity-to-change properties of thethought-listing procedure in comparison with more traditional atti-tude scaling procedures. (See experiment reported on pp. 163- 165for an illustration of the use of the thought-listing procedure as anopinion measure.)

Table 1 gives the additional correlational data that were obtained%plit-half reliability of the externally originated category index (.74)was actually

slightly higher than that for the recipient-generated category (.68),which indicates thatthe difference between correlation magnitudes could not be attributed to relative un-reliability of the externally originated category index. However, it must be noted thatthe externally originated material was nearly uniformly favorable to the specializededucation position; thus, it is possible that range restriction for the externally origi-nated category index was responsible, in part, for the low-magnitude correlations ob-tained between externally originated thought content and posttest opinion.


9/12

ANTHONY G. GREENWALD 163

Recip~entmodified though

Recipient generated thoughts80.60.40.2000

All thoughts combined

Communication receiving groups Control groupsFIG.4. Correlations with posttest attitude for four listed thought directional content

indexes. (Greenwald & Cullen, unpublished data.)

TABLE 1CORRELATIONSF LISTEDTHOUGHTONTENT

INDEXES ITH PRETEST ND POSTTEST TTITUDE MEASURES"

Correlations withContent index Pretest attitude Posttest attitudeb

Externally originated thoughts .20 .06Recipient-modified thoughts .56' -.02Recipient-generated thoughts .52' .3gCAll thoughts combined . 7 lC 65 "

"Experiment I1 Data; N =48.bPretest attitude has be en partialed o ut of correlations with posttest attitude.c p< .01.

from Experiment I1 by virtue of the use of a pretest opinion measure.On the left side of the table, it may be s een that pretes t opinion waspositively correlated with all of the directional content indexes,indicating that pretest opinion was an important determinant of thecontent of cognitions rehearsed i n response to t he persuasion situa-tion. On th e right side of the table, correlations between the categorydirectional content indexes and posttest opinion are given, withpretest opinion partialed out. Significant positive correlations wereobtained only for the recipient-generated category index and for theone based on all thoughts combined-again suggesting the impor-tance of recipient-generated cognitions in persuasion, relative toexternally originated ones.In summary, Cullen's experiments provided a variety of evidenceindicating important involvement of cognitive responses to persua-sion in attitude structure. These findings, especially those for recipi-ent-generated cognitive responses, were consistent with the presenthypothesis that rehearsal of cognitive responses to persuasive com-munications is an essential mediator of cognitive attitude change.

PERSUASIONND RETENTION F COGNITIVE ESPONSESThe preceding series of experiments examined attitude-relevant

cognitions that were recorded shortly after the receipt of a persuasivecommunica tion. Robert Love's master's thesis research (in prepara--tion), conducted unde r the author's direction, examined recipient'scognitions during a persuasive communication situation. Th e aim ofthis experiment was chiefly to assess correlations of persuasion effec-tiveness with retention of communication content and retention ofcognitive response content, observed imme diately following and oneweek after the initial communication situation.

Love's sample was divided into two groups, one receiving a per-suasive communication on admitting Puerto Rico as the fifty-firststate (N=33), the other receiving a communication advocating popu-lar election of the Secretary of State (N = 35); both communicationswere adapted with minor modifications from ones used by Watts andMcGuire (1964). Subjects first received an opinion pretest (Likert-type), following which the a ppropriate communication was adm inis-tered in printed form. Each communication presented three mainsupporting arguments in separate paragraphs: following each para-graph, four blank lines were provided for the subject to write a one-sentence reaction to the main point of the paragraph ( the main point


10/12

164 6. COGNITIVE LEARNING AND COGNITIVE RESPONSE ANTHONY G. GREENWALD 165was underlined in the printed communication). This procedureserved to obtain a sample of cognitive responses rehearsed duringthe persuasion sit~ation.~fter the communication, subjects' opinionswere again assessed, this time using the thought-listing procedure(see p. 1571, following which subjects were given an unexpected testfor recall of (a) h e three main points of th e communication and (b) hereactions that had been written following each. The same subjectswere recruited for an ostensibly different experiment, one week later,at which time the opinion posttest and the recall test were unex-pectedly readministered.

The data were analyzed for partial correlations (pretest partialedout) of posttest opinion with (a ) the con tent of cognitive reactions tothe communication (number of favorable minus number of unfavor-able reactions), (b) retention of th e main arguments of the communi-cation (nu mber of arguments recalle d; maximum = 3), and (c) reten-tion of cognitive reactions (number of favorable minus number ofunfavorable reactions recalled). Thes e correlational results are givenin Table 2. It is apparent that the best predictor of the effect of thecommunications was t he measure of content of the cognitive reactionsthat were written during the communication exposure (average of4 T'S = 52); the next best predictor was the retention index for thecognitive reactions (average of 4 r's = 30); decidedly the poorestpredictor was the retention measure for the persuasive communica-tion itself (average of 4 TS = 03). These results strongly support thepresent contention that cognitive responses to persuasion are impor-tant in mediating persuasion effects while retention of communica-tion content is not.

Supplementary data were obtained from subjects' classificationsof the thoughts they had listed for the opinion posttests into the cate-gories of (a) having originated in the persuasive message, ( b) havingoriginated in written reactions to the persuasive message, and (c)traceable to neither of these sources. Subjects' written reactions weremost often represented in their listed thoughts (mean number=2.26),

4Note that this proced ure explicitly encoura ged rehearsal of recipients ' cognitiveresponses in that subjects were instructed to produce thoughts and to spend timewriting their thoughts. This removes the p resent experimental situation a bit from thetype of communication situation invo lving only a one-way transmission from source torecipient. However, this was unavoidable in the interests of obtaining usable infor-mation about the content of recipients' cognitions during persuasion. Moreover, thesituation was analogous to another important type of persuasion situation in whichsource and recipient are in face-to-face confrontation, the source typically beinginterrupte d by the recipient's reactions to the communication.

TABLE 2PARTIAL ORRELATIONSF COGNITIVE EACTION AND RETENTION MEASURES

WITH IMMEDIATEND DELAYED OSTTESTOPINIONCommunication topic

Puerto Rico Secretary of State( N= 33) ( N=35)

I Immediate Delayed Immediate Delayedopinion opinion opinion opinion

Predictors of posttest opinion posttest posttest posttest posttest1. Content of cognitive reactions

during communication .47" .52O .59" .48"2. Retention of communication content

Immediate -. 12 - .19 -Delayed - .15 - -.09

3. Retention of cognitive reactioncontent

Immediate .22 - .23 -Delayed - .45" - .30b

Note: Thes e data are from a master's thesis by Ro bert E. Love (in preparation). Pretestopinion was partialed out of the correlations reported in this table."p < .01, ne-tailed.

. *p< .05, one-tailed.

while the communication content was least often represented (meannum ber= 1.21). (An average of 1.90 thoughts was traceable to nei thersource.) These findings lend further support to the conclusion thatcognitive responses to the communications were more significant inproviding content for cognitive attitude change than were the com-munications themselves.

The chief theoretical aim of this chapter has been to establish thebasis for useful analysis of attitude change as a cognitive learningprocess. To do this, it was necessary to focus on cognitive responses

4 rehearsed during persuasion situations. Specific learning-theoreticaltopics, such as the roles of incentives, reinforcers, and conditions ofpractice, have been ignored for the moment.

I At present, knowledge concerning the determinants of learning ofattitude-relevant cognitions is quite limited. Certainly, much isknown theoretically about verbal learning, including learning of


11/12

166 6. COGNITIVE LEARNING AND COGNITIVE RESPONSE ANTHONY G. GREENWALD 167meaningful material (see McGeoch & Irion, 1952; Woodworth &Schlosberg, 1954). However, th e particular variables involved inattitude-relevant learning- for example, covert rehearsal, preexistingattitudes, prior familiarity with information, and comprehension ofpersuasive messages -are not well understood in learning-theoret icalterms. Number of presentations of a persuasive argument is aboutthe only variable that is unequivocally established as a determinantof a rgument retention (see, for example, Greenwald & Sakumura,1968; Jones & Kohler, 1958; Levine & Murphy, 1943; Waly & Cook,1966). Until quite recently, another widely accepted principle oflearning of persuasive arguments was that audiences would selec-tively attend to and remember information consonant with theirpre-existing attitudes. Several studies had demonstrated effects ofthis nature (e.g., Jones & Kohler, 1958; Levine & Murphy, 1943).Recent attempts to replicate this phenomenon (Greenwald & Saku-mura, 1968; Waly & Cook, 1966) have met with absolutely no suc-cess, so that the phenomenon of selective learning of attitude-consonant information must currently be regarded as of dubiousvalidity.

Other determinants of attitude-relevant learning that have beenimplicated by empirical research are information utility (Jones &Aneshansel, 1956) and novelty (Greenwald & Sakumura, 1968).Since utility and novelty are variables known to increase attentionto persuasive information: it appears likely that their effects onlearning of persuasive arguments may be mediated by these atten-tional effects rather than by any direct role in the learning process.

The findings of the studies just mentione d, and of other relevantstudies not cited here, do not require the supposition that any vari-able other than duration of exposure (i.e., atten tion) to persuasiveinformation is a determinant of information learning (cf. Cooper &Pantle, 1967). It s eems likely also that comprehension of informationis a determinant of retention (see Fitzgerald & Ausubel, 1963); how-ever, minimal evidence is available concerning this relationship.Additionally, conditions of practice, particularly distribution of prac-tice over time, should be expected to affect learning of persuasiveinformation in much the same manner that they affect learning ofother verbal material; again, little pertinent evidence is available,although the literature concerned with primacy and recency effectsin persuasion ( see Rosnow's chapter in this volume, and references;

5Brock,T. C. , Albert, S . M. , & Becker, L. A. Familiarity, utility and supportivenessas determinants of information receptivity. (in preparation)

also Anderson & Hubert, 1963) may be interpreted in terms of theconditions-of-practice variable.

The effects of rewards and punishments occurring in th e persuasionsituation are certainly relevant to theoretical interpretations of cog-nitive learning. However, although effects of rewards and punish-ments on attit ude measures have frequently bee n demonstrated, the

, processes underlying such effects are poorly understood. Competingexplanations in terms of classical conditioning, instrumental learning,dissonance reduction (Brehm & Cohen, 1962), and attention mecha-nisms (Janis & Gilmore, 1965), all can be justified by appeal to por-tions of the relevant literature. (The reader will find extensive dis-cussion of reward a nd punis hment effects, interprete d in terms ofconditioning processes, in the chapters by Lott and Lott, Rosnow,Staats, and Weiss in this volume.)The procedure of having subjects actively rehearse their ownpersuasive arguments was found by Greenwa ld and Albert (1968; alsosummarized earlier in this chapter) to produce substantial enhance-ment of argument retention as well as noticeable self-persuasion. Atthe moment, it is unknown whether these effects were due to en-hanced original attention to the improvised arguments or to otherfactors. Nonetheless, the focus on learning of persuasive argumentsactively rehearsed in a persuasion situation has provided t he basis forpresently reasserting th e importance of cognitive learning in persua-sion. Thus, the study of determinants of persuasive-argument learn-ing-a problem area in which, to summarize the present brief litera-ture review, current ignorance is considerable- an be justified notonly as an interesting exercise in learning theory, but in terms of itspractical value in interpreting the basis for effective and durablepersuasion.

CONCLUSION-COGNITIVE PROCESS, COGNITIVELEARNING, AND ATTITUDE CHANGEThe present program of research set out to establish the legitimacy

of a conception of attitude change through persuasive communicationas, at least in part, a cognitive learning process. In t he course of doingthis, the obtained experimental evidence repeatedly indicated thatthe effects of persuasive communications are strongly mediated by

I the content of attitude-relevant cognitions elicited (and thus re-hearsed and l earned) during the persuasion situation. The analysis ofdeterminants of cognitive response content may very well require


12/12

ANTHONY G . GREENWALD 16 9explanatory principles outside the scope of the learning principleswith which the present research was concerned. There is, however,no lack of theories in the cognitive integration area that might beapplied to the analysis of determinants of cognitive response con tent.Cognitive consistency theories, for example, can be used to predictthat cognitive responses to persuasion will be consistent with pre-existing attitude-relevant cognitions. The assimilation-contrast ap-proach (Sherif & Hovland, 1961) predicts that the individual reactswith favorable cognitions to persuasive statements within his latitudeof acceptance and with unfavorable cognitions to statements outsidehis latitude of acceptance. Reactance theory (see Brehm's chapter inthis volume) predicts that unfavorable cognitive reactions will occurin persuasive situations of a coercive nature. Brock's commodityanalysis (se e his chap ter in this volum e) predicts favorability ofcognitive reaction to persuasive information to be a decreasing func-tion of the perceived availability of the information. Functional anal-yses of attitude change (Katz, 1960; Sarnoff, 1960; see also Baron'schapter in this volume) provide more complex principles that mightbe u sed to predict cognitive reactions to persuasion given know ledgeabout the motivational basis of existing attitude structures.In light of these observations, it would appear fruitful to approachthe study of persuasive comm unication effectiveness with a combina-tion of cognitive process theory and learning theory. The effects ofindep enden t variable manipulations in persuasion situations, such ascredibility, organization of argum ents, communication mediu m, etc.,could be studied simultaneously in terms of their effects on the con-tent of cognitive responses to persuasion and on learning of per-suasive information. Existing data, noted at the outset of this chapter,suggest that most of the traditional independent variables of persua-sion do not significantly affect retention of persuasive information.(Their effects on attitude presumably are mediated strictly throughtheir effects on cognitive responses to persuasion.) The absence ofreten tion effects of "traditional" persuasion va riables may, howev er,only reflect the fact that attitude change researchers h ave been moreinterested in manipulating variables that affect acceptance of persua-sive information than ones that affect attention to and retention ofpersuasive information. In the combined application of cognitiveintegration theory and learn ing theory to persuasion, cognitive inte-gration theory should offer an account of the processes involved inacceptan ce of persuasio n while lea rning theory should seek to ac-count for persistence of induced changes through learning and reten-tion processes. Neither of these areas of theory, when considered

alone, can currently b e expe cted to provide a complete account of theprocesses by which attitudes are formed or lastingly changed inresponse to persuasive com munications.

ReferencesI

Anderson, N. H., & Hubert, S. Effects of concomitant verbal recall on order effects inpersonality impression formation.Journal o f Verbal Learning and V erbal Behavior,1963, 2, 379-391.

Brehm, J. W., & Cohen, A. R. Explorations in cognitive d issonance. New York: Wiley,1962.

Brock, T. C. Communication discrepancy and inten t to persuade as determinants ofcounterargument production. Journal o f Experimental Social Psychology, 1967,3, 296-309.

Campbell, D. T. The generality of a social attitude. Unpublished doctoral disserta-tion, University of California, Berkeley, 1947.

Cooper, Elaine H., & Pantle, A. J. The total-time h ypothesis in verbal learning. Psy-chological Bulleti n , 1967, 68, 221-234.Fitzgerald, D., & Ausubel, D. P. Cognitive versus affective factors in the learning and

retention of controversial material. Journal o f Educational Psychology, 1963,54, 73-84.Greenwald, A. G. An amended learning model of persuasion. Paper read at AmericanPsychological Association, Washington, D. C., September 1967.

\ Greenwald, A. G., & Albert, Rosita D. Acceptance and recall of improvised arguments.Journal of Personality and Social Psychology, 1968, 8, 31-34.Greenwald, A. G., Albert, Rosita D., & Cullen, Dallas M. Persuasion as a function of

communication content learning. Unpub lished manuscript, Ohio State Univ., 1968.4 Greenwald, A. G., & Sakumura, J. S. Attitude and selective learning: Where are the

phenomena of yesteryear? Journal o f P ersonality and Social Psychology, 1967, 7,387-397.

Hovland, C. I., Janis, I. L., & Kelley, H. H. Comm unica tion and persuasion. NewHaven: Yale University Press, 1953.

Hovland, C. I. , Lumsdaine, A. A., & Sheffield, F. D. Experiments on m a ss co m m u n i-cation . Princeton, N . J.: Princeton University Press, 1949.Insko, C. A. Primacy versus recency in pers uasion as a function of the timin g of argu-

ments and measures.Journal o f Abnormal and Social Psychology, 1964,69,381-391.Insko, C. A. Theories o f a ttitude change. New York: Appleton, 1967.Janis, I. L. , & Gilmore, J. B. The influence of incentive conditions on the success of

role playing in modifying attitudes. Journal of Personality and Social Psychology,1965, 1, 17-27.

t Janis, I. L. & Terwilliger, R. F. An experimental study of psychological resistances tofear-arousing communications. Journal o f Abnormal and Social Psychology,1962, 65, 403-410.

Jones, E. E., & Aneshansel, J. The learning and utilization of contravaluant material.I Journal o f Abnormal and Social Psychology, 1956, 53, 27-33.Jones, E. E., & Kohler, R. The effec ts of plausibility on t he learn ing of controversial

statements.Journa1of Abnormal and Social Psychology, 1958,57,315-320.

Documents

cognitive response