Advances in Consumer Research Volume 25, 1998 Pages 532-537
DYNAMIC CHARACTERISTICS OF MOTIVATION, ABILITY AND OPPORTUNITY TO PROCESS COMMERCIAL INFORMATION
Johan de Heer, Tilburg University
Theo B.C. Poiesz, Tilburg University
The present paper focuses upon several complexities regarding the three main information processing antecedents motivation, ability and opportunity. One puzzling complexity involves their dynamic character: they may influence each other over time. To examine the impact of exposure duration on motivation, ability and opportunity to process commercial information, a study was set up in which each of twenty subjects viewed one hundred WWW-advertising pages under different conditions of exposure duration: 1, 2, 4, 6, and 8 seconds. It is surprising to note that if exposure duration is shorter than 4 seconds, ability and motivation differ significantly, while the difference is not significant at an exposure duration of longer than 4 seconds. A related finding was that intercorrelations between the three information processing antecedents reached a maximum at 4 seconds. The assmed conceptual independence of the antecedents increases over time at an exposure duration longer than 4 seconds.
In the last two decades much insight has been acquired concerning the theoretical information processing antecedents of message elaboration and eventual persuasion. Three main generic separate antecedents have been proposed, viz. individuals motivation, ability and opportunity to process information (Andrews, 1988; Batra and Ray, 1986; MacInnis and Jaworski, 1989; Petty and Cacioppo, 1983; Poiesz, 1989). Moreover, several integrative framework models in this area, in particular, the Elaboration Likelihood Model (Petty and Cacioppo, 1986), and the Integrative Attitude Formation Model (MacInnis and Jaworski, 1989) have explored relationships among combinations of these three information processing antecedents to message elaboration. Likelihood of elaboration can be viewed as a function of motivation and ability and opportunity (Andrews, 1988; Petty and Cacioppo, 1986; Poiesz, 1989). In general, it is assumed that increasing the level of the information processing antecedents deepens information processing (e.g., more attention and processing capacity is allocated to the advertisement) resulting in advertising effectiveness responses that range from mood-generated affect (minimal consequence) to self-generated persuasion (maximum consequence) (see, MacInnis and Jaworski, 1989).
Even though it can be argued that the information processing antecedents motivation, ability, and opportunity are conceptually distinct and should be treated as such in experimental settings, several operational complexities may occur (Andrews, 1988). The first complexity relates to mutual relationships. As indicated by several authors, measures of motivation and ability have been found to be significantly correlated (Batra and Ray, 1986; Bettman and Park, 1980; Brucks, 1985; Kuusela, 1992; Lutz, MacKenzie, and Belch, 1983; Maheswaran and Sternthal, 1990; Petty, Cacioppo, and Goldman, 1981; Zaichkowsky, 1985). More recently, Pieters, Rosbergen and Hartog (1996), in an eye-movement recording study, found that the motivation manipulation for the third repeated exposure for the same stimulus was no longer significant. This indirectly suggests that motivation to process possibly decreases by increasing opportunity to process the same stimuli. Thus, the information processing antecedents motivation and opportunity may have a complex relationship as well (other examples can be found in Batra and Ray, 1986; Petty and Caciopppo, 1986; Rethans, Swasy, and Marks, 1986; Vuokko, 1992). A second type of complexity is presented by so- called reciprocal or trade-off effects: executional cues that positively affect one information processing antecedent and negatively affect another (MacInnis, Moorman and Jaworski, 1991). A third possible complexity is that motivation, ability, and opportunity to process may all interact in some situation (Wright, 1981). Although, there are possible explanations for (some of) these complexities (Andrews, 1988; Poiesz, 1989, 1994; Robben and Poiesz, 1993), this nature is not fully understood and no comprehensive theory does exist as yet.
Andrews (1988) suggests three possible explanations, however: (1) motivation to process can, over time, develop into an antecedent condition influencing ones ability to process (and vice versa; our emphasis); (2) self-report measures may be inflated measures (for an excellent review see also MacInnis, Moorman and Jaworski, 1991); and, (3) opertionalizations differ from experimental manipulations. Although the term 'over time is not specified and could refer to both unspecified time and/or exposure time, Andrews first explanation implies that the antecedents have dynamic characteristics, however time is defined. Poiesz (1989, 1994) and Robben and Poiesz (1993) extended Andrews first explanation by suggesting that the likelihood that the three antecedents (as subjectively perceived) change and influence one another may increase with increasing exposure duration. This would be expressed in higher intercorrelations among the three information processing antecedents captured on successive times.
The influence of exposure duration on the three main antecedents of information processing is not clear, however. Should the three antecedents be taken as 'dynamic information processing antecedents? For theoretical, methodological and practical reasons, the factor exposure duration is a critical factor in need for careful consideration. Yet, the choice for a particular level of exposure duration in advertising research often seems determined by subjective criteria which are not communicated by the respective researchers. Instead, there seems to be a rather broad range within which the manipulated or allowed exposure duration may fall, and from which a seemingly random and ad hoc choice seems to be made. By consequence, exposure time is not systematically varied, nor are comparisons allowed between studies. To complicate matters even more, some authors use self-selected exposure duration by subjects or viewing time to infer the level of elaboration, thereby confusing cause and effect. In all, the role of a critical factor in advertising effects is obscure. Findings in one study with a particular exposure level X may be very different from findings in a very similar study with a particular exposure level Y. This may possibly bias results and may limit the generalizability of the conclusions.
In advertising research more emphasis is given to the relationship between exposure duration and recall and/or recognition. The general finding is that the ability to recall / recognize advertisements increases with increasing exposure duration. For example, using magazine illustrations, Potter and Levy (1969) found only 50 percent could be recognized at a 1/2 second exposure duration, whereas 93 percent were recognized at a 2 second exposure duration. In Shepards (1967) experiment an accuracy of 97 percent was obtained for immediate recognition. Shepards experiment, in which subjects were exposed to 612 pictures including advertisements, reported that the average viewing time chosen by subjects was about 5 seconds. More recently, Pieters and Bijmolt (1994) found that exposure duration has a positive effect on aided and unaided recall of television advertising. It has to be noted that recall and recognition are memory-based processing outcomes, and are likely to be the result of the combination of motivation and ability and (perceived) opportunity to process the information and not by exposure duration alone.
Based on Andrews (see his first explanation, 1988) and Poiesz (1989, 1994) and Robben and Poiesz (1993) suggestions we have three expectations: (1) The three information processing antecedents change over time. (Here, the direction of change is not specified, as it may depend upon the initial perceived levels of motivation, ability, and opportunity. For example, a high initial level of motivation may reduce as a consequence of a lower level of perceived ability and/or opportunity; conversely, a low level of initial motivation may not be reduced further, no matter what the levels of ability and/or opportunity). Expectation (2) is that the intercorrelations between the three information processing antecedents will increase over time. Finally, expectation (3) is that the information processing antecedents (alone and their interaction component) contribute to the explained variance of the third antecedent, and that the contributions change over time as well. The latter two expectations are related and explicitly address the assumed conceptual independence of the information rocessing antecedents. Note that, even though these expectations are not formulated as explicit hypotheses, the literature generally fails to recognize the likelihood of dynamic relationships and influences among these generic variables altogether.
In order to assess dynamic characteristics of the three information processing antecedents the following experiment is conducted. Subjects were asked to assess and report their subjective motivation, ability and opportunity to process commercial information. The following single generic operationalizations are used: 'To what extent do you think the information on this homepage is interesting? (Motivation); 'To what extent do you think the information about the advertisers brand on this homepage is understandable? (Ability); and, 'To what extent do you have enough time to absorb the information on this homepage (that is, if you would want to do so)? (Opportunity). (See for discussion on similar single generic operationalizations Robben and Poiesz ,1993). Motivation, ability and opportunity were measured using a 9-point semantic differential scale with end-points 'To a minor extent and 'To a major extent and no semantic labels for points 2 through 8 (the non-extremes).
Subjects and Material
Twenty-one female first year undergraduate Psychology students participated in the present study. Subjects volunteered and received credit points for their participation. All subjects were naive regarding the purpose of the study. The stimulus material was not previously encountered by the subjects. In fact, most of the subjects were novice to the World Wide Web. An Olivetti P90 with a NEC MultiSync XV17 monitor was used to present the stimuli and the operationalized questions of the generic information processing antecedents.
Stimuli and Procedure
One hundred commercial World Wide Web-pages were randomly sampled from a Dutch tv advertising site and served as the stimulus material. The advertisements varied widely over product and service categories (audio & video, cars, banks & insurance, business & communication, food & drinks, computers, movie-companies, traveling & leisure). Four presentations were created, each consisting of 25 WWW-pages. Exposure durations were set at 1, 2, 4, 6, and 8 seconds and randomly assigned to the stimulus material. This range was selected on the basis of the average viewing time of 5 seconds found by Shepard (1967). Each individual subject saw 100 WWW-pages: within each of the 4 presentations 5 for 1 second,5 for 2 seconds, 5 for 4 seconds, 5 for 6 seconds, and 5 for 8 seconds. The four presentations were shown to each subject in a different order. Further, the initial order of the stimulus material within each presentation was used for the first 5 subjects. For the next 5 subjects the stimuli were counterbalanced within each presentation. For subjects 11 to 15 the initial order was split for each presentation and subjects saw stimuli 13-25 first, followed by stimuli 1-12. For the last 5 subjects this order was counterbalanced again. In sum, WWW-pages were presented in a different order for all exposure durations. Each stimulus was followed by the operationalizations of motivation, ability and opportunity to process the stimulus. Each next stimulus was selected by the subject by pressing the space bar.
Data from one subject were not used for data analyses because she did not follow the instructions. For each of the 5 levels of exposure duration scores were averaged over stimuli (100) and over subjects (20), resulting in 400 data points for each exposure duration. Therefore, a repeated measures design was used with exposure duration as the within subject variable. A repeated measures analysis indicates that our first expectation is supported; i.e., motivation, ability and opportunity to process all differ significantly between exposure durations. The effect of exposure duration on ability F (4, 1596)=63.56, p<0.05; on motivation F (4, 1596)=16.61, p<0.05; on opportunity F (4, 1596)=259.99, p<0.05. Figure 1 shows that all three information processing antecedents vary over time. Although, we did not specify the direction of change, Figure 1 shows that motivation, ability and opportunity all increase by increasing exposure duration. Figure 1 also shows that motivation has the smallest increase over time as compared to ability and opportunity.
Though our repeated measures analysis shows that all three information processing antecedents vary over time, it is noteworthy to examine if they vary across all levels of exposure duration. Contrast analyses show that between the exposure duration levels 1 and 2 seconds, and between 2 and 4 seconds, ability and motivation differ significantly from one another (F (1, 399)=220.32, p< 0.05 (ability between 1-2 seconds) and F (1, 399)=16.25 , p<0.05 (ability between 2-4 seconds) and F (1, 399)=69.62, p<0.05 (motivation between 1-2 seconds) and F (1, 399)=4.81, p<0.05 (motivation between 2-4 seconds). In other words, there is a significant difference between 1 and 4 seconds exposure duration both for ability and motivation. Not unexpectedly, between all exposure durations a significant difference was found for opportunity (F (1, 399)=1124.40, p<0.05, F (1, 399)=11.35, p<0.05, F (1, 399)=2.94, p<0.10, and F (1, 399)=6.72, p<0.05 between 1-2, 2-4, 4-6, and 6-8 seconds respectively). The differences between the 4, 6 and 8 seconds exposure duration for ability and motivation were not significant. In sum, regarding the first expectation, results showed that motivation, ability and opportunity significantly differed from one another under different conditions of exposure duration. However, the reported levels of motivation and ability increased up to the 4 seconds exposure duration. At longer exposure durations than 4 seconds, these antecedents show no difference. Not surprisingly, opportunity increased by increasing exposure duration. Thus, the first expectation is partly supported.
Figure 2 presents the Pearson correlation coefficients between the information processing antecedents for each level of exosure duration. We expected to find that the intercorrelations increase over time. Conventional statistical analyses do not allow us to report whether the observed increases and decreases are significant. Note, however, that the intercorrelations are obtained by averaging over subjects and stimuli for each different exposure duration. It then is interesting to observe from the pattern of intercorrelations between each possible combination (motivation & ability, ability & opportunity, motivation & opportunity) that the maximum intercorrelations are found at 4 seconds. Thus, the second expectation is partly supported as well: intercorrelations between the information processing antecedents increased over time up to 4 seconds after which they decreased.
In case of single item operationalizations it is not possible to assess whether the mutual relationships between variables are determined by their conceptual overlap, operational overlap, or a combination of both. Of course, if the variables prove to be unrelated, there is both conceptual and operational independence. For the present study we can assess how the mutual relationships between the three processing antecedents develop over time. Changes in the strength of a relationship can be attributed to conceptual or operational convergence/divergence. Even though no specific hypotheses are presented to this effect, we want to assess how the mutual dependence between the antecedents develops over time. In order to so, regression analyses were performed, each with one antecedent as the criterion variable, and the other two factors and their interaction component as the predictor variables. Table 1 lists the results. Note that these regressions address the contribution of the information processing antecedents to the explained variance of the third antecedent, and not the contribution to the explanation of variance of a dependent variable. This explains why the explained variances are relatively low. It is interesting to find that motivation and opportunity both contribute to the explained variance of ability for all exposure durations. However, the interaction between motivation and opportunity contributes negatively to the explained variance of ability for the 4 seconds exposure duration. Ability contributes to the explained variance of motivation for all exposure durations. For exposure duration 1 second also the interaction between ability and opportunity is significant. Ability contributes for the explained variance of opportunity for all exposure durations. However, for exposure durations 1 and 6 seconds the interaction between ability and motivation also contributes significantly. Further, for the 6 second exposure duration a significant, however, negative contribution, of motivation is found as well. In general, the value of the regression coefficients increased over time up to 4 seconds after which they decreased. This means that dependence between the three processing antecedents increases over time up to 4 seconds and independence increase at an exposure duration longer than 4 seconds.
The intercorrelation coefficients and the results of the regression analyses are taken to indicate to what extent in terms of subjective assessments, subjects judged the antecedents to be independent from one another. We expected to find that, using two information processing antecedents (alone and their interaction component) as the predictor variables, contribute to the explained variance of the criterion variable and that these contributions change over time as well. The combined results partly supported our third expectation: the value of the regression coefficients increased over time up to 4 seconds after which they decreased.
DISCUSSION AND IMPLICATIONS
The purpose of the present study was to investigate the dynamic characteristics of motivation, ability and opportunity to process commercial information. We measured motivation, ability and opportunity to process commercial information under different conditions of exposure duration (1, 2, 4, 6, and 8 seconds). We expected to find (1) that the three information processing antecedents change over time (the direction of change was not specified), and (2) that the intercorrelations between the three information processing antecedents will increase over time, and (3) that the information processing antecedents (alone and their interaction component) contribute to the explained variance of the third antecedent, and that the contributions change over time as well. All expectations were partly supported.
The present results are interesting because Andrews (1988) suggested that motivation to process can, over time, develop into an antecedent condition of ability to process. Our results may support this suggestion, although the direction of causality could not be assessed. Moreover, ability to process can, over time, develop into an antecedent condition of motivation to process as well. Petty, Cacioppo, and Goldman (1981) also argued that motivation to engage in information processing is, in part, based on the preexisting information about the issue (ability to process). However, in our study this seems to hold for very short exposure durations only. One reviewer pointed to the possibility that the responses collected are not actual changes in MOA but changes in Ss (in)abilities to respond to the measures, in particular, for short exposure durations. For example, at shorter exposure times, Ss are less certain about their motivation to process, and hence, respond close to the mid-scale point. This means that, if the reviewers suggestion is valid, we expect to find that standard deviations and variances, in particular, for motivation and ability are different (smaller) as compared to standard deviations and variances at longer exposure durations. Our data do not support that, however. This means that Ss were able to respond to the measures even at short exposure durations.
Robben and Poiesz (1993) reported intercorrelations of 0.18 (motivation and ability), 0.15 (motivation and opportunity) and 0.26 (ability and opportunity) over experimental stimuli in a study in which subjects were exposed to stimuli for 10 seconds. Together with the results found in this study it seems that independence increases over time at an exposure duration longer than 4 seconds. The present study does not address the question what takes place if exposure duration is longer than 10 seconds. A closer examination is needed in future studies.
Obviously, the objective variable exposure duration has a positive effect on the subjective variable opportunity to process. Note, however, that the contribution of ability, motivation and the interaction between ability and motivation to the explanation of variance of (perceived) opportunity. Although single item operationalizations were used in this study for generality and practical reasons subjects may not have perceived them as referring to distinct and separate aspects of the total exposure situation. This may call for multi-item assessments.
The results of the present study showed, among other things, that the intercorrelations between the information processing antecedents reached a maximum at 4 seconds. If 4 seconds is a meaningful exposure duration level than this may be expressed in the actual viewing time. To check the actual viewing time of our stimulus material we set up a small study similar to the one but with different subjects (20). We let the subjects watch each stimulus "as long as they would like to do so". An average viewing time of 5.56 seconds was found, however, the median viewing time was 4.01 seconds (results obtained from 20 (stimuli) * 100 (subjects)=2000 recordings).
RESULTS OF THE REGRESSION ANALYSES
In sum, the results may uggest that within 4 seconds all relevant knowledge structures are activated. Note that ability does not change after 4 seconds. Further, the motivation to process does not differ significantly over time when all relevant knowledge structures are activated. A rather interesting observation is that the independence of the information processing antecedents increases at an exposure duration longer than 4 seconds. The combined results call for a closer examination of the role of particular exposure duration levels. Note that these results were obtained over different types of 'static ads. Further research on the dynamic antecedent characteristics for dynamic ads (e.g., television advertising) is needed as well but seems premature at this point, considering the complexities of research with static ads only.
In conclusion, given the complexities (e.g., changing of the antecedents, mutual influences of the antecedents, reciprocal or trade-off effects, and interactions between the antecedents) advertising researchers are recommended to take all three information processing antecedents into consideration simultaneously and consistently in order to stimulate meaningful comparisons between different studies in the future.
Andrews, J. Craig (1988), "Motivation, Ability, and Opportunity to Process Information: Conceptual and Experimental Manipulation Issues," in Advances in Consumer Research, 15, Michael J. Houston, ed., Association for Consumer Research, 219-225.
Batra, Rajeev and Michael L. Ray (1986), "Situational Effects of Advertising Repetition: The Moderating Influence of Motivation, Ability, and Opportunity to Respond," Journal of Consumer Research, 12, 432-445.
Bettman, James R. and C.W. Park (1980), "Effects of Prior Knowledge and Experience and Phase of the Choice Process on Consumer Decision Processes: A Protocol Analysis," Journal of Consumer Research, 7, 234-248.
Brucks, Merrie (1985), "The Effects of Product Class Knowledge on Information Search Behavior," Journal of Consumer Research, 12, 1-16.
Kuusela, Hannu (1992), "The Effects of Actual and Self-perceived Knowledge on the Use of Elementary Information Processes (EIPs) in a Choice Task," Acta Universitatis Tamperensis, ser A vol 333, Academic dissertation, University of Tampere.
Lutz, Richard J., Scott B. MacKenzie and George E. Belch (1983), "Attitude Toward the Ad as a Mediator of Advertising Effectiveness: Determinants and Consequences," in Advances in Consumer Research, Vol. 10, Richard P. Bagozzi and Alice M. Tybout, eds., Ann Harbor, MI: Association for Consumer Research, 532-539.
MacInnis, Deborah J. and Bernard J. Jaworski (1989), "Information Processing from Advertisements: Toward an Integrative Framework," Journal of Marketing, 53, 1-23.
MacInnis, Deborah J., Christine Moorman and Bernard J. Jaworski (1991), "Enhancing and Measuring Consumers Motivation, Opportunity, and Ability to Process Brand Information from Ads," Journal of Marketing, 55, 32-53.
Maheswaran, Durairaj and Brian Sternthal (1990), "The Effects of Knowledge, Motivation, and Type of Message on Ad Processing and Product Judgements," Journal of Consumer Research, 17, 66-73.
Petty, Richard E. and John T. Cacioppo (1983), "Central and Peripheral Routes to Persuasion: Application to Adverising," in Advertising and Consumer Psychology, L. Percy & A. Woodside, eds., Lexington, MA: Lexington Books, 3-24.
Petty, Richard E. and John T. Cacioppo (1986), Communication and Persuasion: Central and Peripheral Routes to Attitude Change, New York: Springer., 10, 135-146.
Petty, Richard E., John T. Cacioppo and Rachel Goldman (1981), "Personal Involvement as a Determinant of Argument-based Persuasion," Journal of Personality and Social Psychology, 41(5), 847-855.
Pieters, Rik M.G. and Tammo H.A. Bijmolt (1994), "The Effect of Duration, Serial-position and Competitive Clutter on Consumer Memory for Television Advertising: An Extension and a Test of Generalizability," Unpublished research report. Tilburg University.
Pieters, Rik M.G., Edward Rosbergen and Michel Hartog (1996), "Visual Attention: The Impact of Motivation and Repetition," In Advances in Consumer Research, 23, Corfman, Kim P. And John G. Lynch Jr., eds., Association for Consumer Research, 242-248.
Poiesz, Theo B.C. (1989), De Transformatie van een Karikatuur [The Transformation of a Caricature]. Inaugural lecture: Tilburg University.
Poiesz, Theo B.C. (1994), The Assessment of Communication Quality-Theory and Application of the Triad model. In Proceedings 46th Marketing research Congress: 'The many faces of quality now and in the future. European Society for opinion and Marketing Research, Copenhagen, 137-159.
Potter, M.C. and E.I. Levy (1969), "Recognition memory for rapid sequence of pictures," Journal of Experimental Psychology, 81, 10-15.
Rethans, Arno, John L. Swasy and Lawrence J. Marks (1986), "The Effects of Television Commercial Repetition, Receiver Knowledge and Commercial Length," Journal of Marketing Research, 23, 50-61.
Robben, Henry S.J. and Theo B.C. Poiesz (1993), "The Operationalization of Motivation, Capacity, and Opportunity to Process an Advertising Message," In: W. Fred van Raaij and Gary J. Bamossy (eds.). European Advances in Consumer Research, 1, 160-167.
Shepard, Roger N. (1967), "Recognition Memory for Words, Sentences, and Pictures," Journal of Verbal Learning and Verbal Behavior, 6, 156-163.
Vuokko, Pirjo (1992), "Advertising Repetition Effects-Conceptual Framework and Field Study in Four Product Categories," Publications of the Turku School of Economics and Business Administration, Series A-1.
Wright, Peter (1981), "Cognitive Responses to Mass Media Advocacy," In Cognitive Responses to Persuasion, Richard E. Petty, Thomas M. Ostrom, and Timothy C. Brock, eds., Hillsdale, NJ: Lawrence Erlbaum Associates, 263-282.
Zaichkowsky, Judith Lynne (1985), "Measuring the Involvement Construct," Journal of Consumer Research, 12, 341-352.