When Does Consumer Information Processing Research Actually Have Anything to Do With Consumer Information Processing?

Michael L. Ray, Stanford University
ABSTRACT - Consumer information processing (CIP) research relates to consumer information processing if the situation in the research is the same as the situation in the processing. This paper concentrates on what may be the most common CIP or persuasion situation, that of advertising. The advertising situation differs from the typical CIP research one in audience involvement, attention, noise, exposure time and audience control of exposure. In general, motivation and opportunity to make cognitive responses and use attitude structures is lacking in advertising persuasion and heightened in CIP-persuasion research. Several examples are given of the results of this dramatic difference, and researchers are cautioned to take care in applying results from highly artificial to natural situations. In the future CIP researchers should extend their theories and initial findings to settings like the advertising one.
[ to cite ]:
Michael L. Ray (1977) ,"When Does Consumer Information Processing Research Actually Have Anything to Do With Consumer Information Processing?", in NA - Advances in Consumer Research Volume 04, eds. William D. Perreault, Jr., Atlanta, GA : Association for Consumer Research, Pages: 372-375.

Advances in Consumer Research Volume 4, 1977   Pages 372-375

WHEN DOES CONSUMER INFORMATION PROCESSING RESEARCH ACTUALLY HAVE ANYTHING TO DO WITH CONSUMER INFORMATION PROCESSING?

Michael L. Ray, Stanford University

ABSTRACT -

Consumer information processing (CIP) research relates to consumer information processing if the situation in the research is the same as the situation in the processing. This paper concentrates on what may be the most common CIP or persuasion situation, that of advertising. The advertising situation differs from the typical CIP research one in audience involvement, attention, noise, exposure time and audience control of exposure. In general, motivation and opportunity to make cognitive responses and use attitude structures is lacking in advertising persuasion and heightened in CIP-persuasion research. Several examples are given of the results of this dramatic difference, and researchers are cautioned to take care in applying results from highly artificial to natural situations. In the future CIP researchers should extend their theories and initial findings to settings like the advertising one.

THE ANSWER

My answer to the question "When does consumer information processing research actually have anything to do with consumer information processing?" is, in general, "Sometimes." More specifically, I believe that the majority of persuasion situations involve mass media advertising which is almost always different, in at least six ways, from the artificial conditions of CIP research.

1. Advertising is a low involvement situation while the CIP research setting is a high involvement one. People don't care about the ads. They are far from as motivated to process information as are the experimental "subjects" in CIP labs.

2. In advertising attention is difficult to achieve while in CIP research attention is forced in virtually all cases.

3. In advertising there is, in an information theory sense (Shannon and Weaver, 1949), much noise (consisting of competing messages, etc.) while CIP re- search is usually conducted in a sterile environment in order to create "experimental realism" (Aronson and Carlsmith, 1968).

4. Advertising, particularly broadcast, involves short exposure time while CIP research averages much longer exposure time and time to consider messages.

5. For advertising the audience has little or no control of exposure time. In CIP research the audience often has control over exposure time.

6. So in the advertising situation there is minimal motivation or opportunity to develop cognitive responses or use attitude structures. But in consumer information processing or persuasion research the situation is engineered to produce cognitive responses and use of attitude structures.

In short, the most typical persuasion situation in the world is one that is ignored by the bulk of the social psychological research on persuasion.

At the same time, however, consumer information processing research--including the cognitive response analysis and cognitive attitude structure analysis approaches we are concentrating on today as well as the perceptual mapping-conjoint measurement approaches and hierarchy of effects analysis research--seems to me to hold the key to understanding this most common persuasion and, therefore, consumer information processing situation. This is because consumer information processing researchers are truly concerned with people and the way they handle the very difficult task of using tremendous amounts of information to make very complex decisions under very difficult conditions. No longer are we content to consider audiences en masse almost mindlessly stepping through the stages of cognition, affect, and conation. So despite the fact that most of the initial research in this area has concentrated on rather unrealistic conditions, it has been and will lead to rather clear implications for campaign planners attempting to deal with the consumer information processing situation represented by advertising.

The point I'm making is not really too different from Hovland's (1959) in his classic paper published seventeen years ago and entitled "Reconciling Conflicting Results Derived from Experimental and Survey Studies of Attitude Change." The importance of situational characteristics was established well throughout all of the Hovland communication and attitude change studies. This tradition is represented recently by our discussant Timothy Brock's (Keating and Brock, 1971; Osterhouse and Brock, 1970) cautions that the distraction technique will only work when people are involved and antagonistic to the message position and when distraction does not impede learning. And this session's chairman has made a very valuable contribution by pointing out how situational constraints can affect the choice or aggregation model a consumer might use (Wright, 1974).

But today I would like to go a bit further than just cautioning about the effects of the situation. I would like to suggest that consumer information processing researchers attempting to deal with persuasion processes must begin to realize that there is more than a quantitative difference between persuasion in their laboratory studies and persuasion in the real world of advertising. In fact the difference is qualitative. Today I would like to give a number of examples of the unusual results one can get when there is an attempt to bring "mundane realism" (Aronson and Carlsmith, 1968) into persuasion research. My main point is that by concentrating on the differences between information processing and persuasion situations we can begin to make "translations" from CIP research to advertising applications. And, in the future we can begin to bring a wider variety of persuasion situations into CIP research.

THREE HIERARCHIES

The results in this paper come primarily from a program of research at Stanford in which we have tried to replicate natural exposure situations in controlled laboratory experimentation so that the findings of communication research could be examined in a more realistic way. In virtually all the studies repetition is an independent variable so that the procedure used resembles the free recall paradigm of verbal learning research, with the exception that the stimuli are advertisements rather than nonsense syllables.

When we began the research we assumed a hierarchy relationship between learning, attitude and behavior that would be predicted from the rtl-component notion of attitude or most learning or information processing formulations. We are surprised to find, however, that for most of our results a different pattern emerged. This was the pattern predicted by Krugman's (1965) notions about the process of television advertising effects. In this "low involvement" hierarchy, repetitive messages worked to affect gross brand and advertising awareness and subsequently had a slight effect on purchase intention but little effect on attitude.

Later we realized that a third hierarchy was possible under other conditions. In this one, which we called the "dissonance-attribution hierarchy," a choice had been made before advertising exposures and combined with the exposures to affect attitude first and finally complex cognitive learning.

These three hierarchies--learning (cognitive to affective to conative), low involvement (cognitive to conative to affective), and dissonance-attribution (conative to affective to cognitive)--were more fully described in a 1973 paper which was a collaborative effort of the Stanford researchers (Ray et al., (1973). Two aspects of the findings are critical for the present purposes however. First, each of the three hierarchies represents a large body of social psychological research on particular situations which are well defined. Knowing about the hierarchies and what causes them allows a very functional analysis of situations to determine how consumers will process information in them.

The second conclusion that might be made from the Stanford repetition and hierarchy research is that the majority of situations in advertising, particularly that for the broadcast media, tend to be low involvement ones. That is, the decisions being made are not salient or important, there are almost no differences between alternative brands and products, advertising is the main source of information, and consumers are very familiar with the product category in the sense that it is in the maturity stage of its life cycle.

If this is an accurate description of most advertising persuasion situations, we have great reason to question the applicability of most information processing research. Certainly the likelihood of detailed cognitive response in such a situation is Low. At best it may be that cognitive response operates in some very different way. Also, since there are few product differences and the decisions are not salient to people, it isn't likely that multi-attribute attitude structures hold or that the more textured choice models apply to the use of advertising information.

LIMITED INFORMATION PROCESSING

At the most basic level, the low involvement situation means a limit in information processing. Even with the tremendous repetition potential of advertising, researchers have found that there is a definite limit to what people get out of messages. For instance, Krugman (1972) expanded his low involvement idea by surmising that there are really only three relevant exposures in a number of repetitions of television advertising. The first one produces a "what is it?" response. The second one produces whatever information processing is going to occur, and the third exposure and all subsequent ones are equivalent to reminders of whatever was learned in the second exposure.

Once again, there is much in the Stanford repetition research data that supports Krugman's idea. Although cognitive learning curves did not stop automatically at the three exposure level, there did seem to be some very low level of exposure at which point nothing further was taken out of a message exposed in the television manner. Further repetitions just reinforced the basic message. Even when there might be some involvement in the situation, consumers definitely have developed particular ways of using advertising. For instance, in one study comparing the repetition effect of advertising for convenience goods and shopping goods, it was found that the repetition response function for ad recall went up steadily for convenience goods but stopped after four exposures for the shopping goods advertisements (Ray and Sawyer, 1971). Our interpretation for that finding was that consumers did not use ads for shopping goods (e.g., washing machines, television sets, and low price cars) to gather any more than the most basic information. The important information gathering phase for shopping goods was in the store actually examining the product and discussing it with salespeople. Thus repetitions of the ads had an effect only up to a certain point.

The same kind of limit to what consumers will process was found in a study of color versus B&W advertisements (Ray, Sawyer and Strong, 1971). In this study we examined the repetition results for two series of ads which were identical except that one series had one color added to the pictures of fruits or vegetables depicted. The color ads were superior in getting the message recalled, but the black and white ads, if recalled, were superior in generating learning of the main message. In a follow-up field experiment using the two ad series, the black and white print ads were more effective than the color in producing ad awareness. It seemed that, in this real advertising situation, the black and white ads generated more information processing while the color ads elicited just a surface awareness after a certain point, no matter what the repetition level.

All of these examples--evidence of the low involvement hierarchy, the limit to information processing in advertising situations and the specific cases of that limit--show what can happen to cherished CIP folklore in the real advertising situation.

DEALING WITH DIFFERENTIAL ATTENTION

As I have already pointed out, one of the key differences between most lab persuasion studies and the typical advertising situation is that attention is a given in the labs but a goal hard to attain in advertising. Our original assessment of this bias was that it was a main effect, that is, if an ad or message did well in the lab it would do relatively well in the field--pro-vided it had the appropriate schedule.

Now we are beginning to believe that there is not only a greater main effect difference than we had originally presumed but also there are interactions. That is, ads that do well in a forced exposure lab situation may be the very ones that do poorly in the field.

My evidence for this comes from a series of lab studies on television clutter carried out by Peter Webb (Ray and Webb, 1976, Webb and Ray, 1977). Peter first determined that his lab approach was producing attention patterns similar to that found in field research (e.g., Burke, 1972). Once this was established he was able to manipulate the commercial break pattern in several television program types.

His main finding was that consumers don't just randomly ignore commercials. They seem to actually avoid them--and with considerable skill at that.

Consumers were amazingly well-schooled as to the pattern of commercial breaks on their favorite programs. This was particularly true for soap operas. When Peter tried different commercial schedules during those programs than the schedules to which consumers were accustomed, they very often asked why this unusual pattern was being used. During debriefing respondents told us more about how carefully they watched for the pattern of commercial interruptions. For instance, one working woman actually made an audio tape of soap operas so that she could plan her housework during the commercial breaks while she was home having lunch and watching her program.

The data supported the patterning beyond such anecdotal evidence, however. Even those respondents who made no visual or verbal indication of noting an unusual pattern were able to avoid being affected by the most common interruption schedule for the program. In other words, virtually all response measures were lower for the most common type of interruption schedule as opposed to the more unusual ones.

Before this research we had assumed that the typical laboratory information processing of communication research studies erred in forcing people to read or view the messages while in a more natural situation consumers would not be biased toward or away from viewing the messages. NOW this research indicates that the error of typical laboratory communication research studies may be even greater than we had previously imagined. While the lab studies bias people toward viewing the material, the audience/in the natural situation may have learned to be biased completely in the other direction, that is, away from viewing and processing the messages. In advertising we have to be concerned not just with getting attention. We have to gain attention from people who are actively avoiding us.

INVOLVEMENT AS A HEDGE AGAINST CLUTTER

Given this bias, however, is it possible that certain commercials--which may not do well in typical persuasion research--might be less susceptible to the vagaries of clutter and consumer avoidance? Peter Webb examined the extent to which involvement would enhance message effectiveness. He used the "connections" measure developed by Krugman (1966-67). As hypothesized, those commercials shown to be high in involvement in a prestudy seemed unaffected by changes in clutter (4 versus 8 commercials) or position in a string of commercials, while low involvement commercials were dramatically effected.

WHEN COUNTERARGUING INCREASES PERSUASION

Even further we found that commercials which would not do well in typical persuasion research did well in a normal exposure situation, possibly because they generated involvement and attention.

More specifically we showed, at least in one case, that counterarguing can enhance persuasion.

In CIP and lab persuasion studies in which the respondents are forced to be exposed to the messages, counter-arguing is a negative to be avoided at all cost. But when exposure conditions are more realistic and attention is an important goal, counterarguing can sometimes enhance persuasion by increasing attention and making the message more memorable.

In an NIMH-sponsored study on the effects of anti-drug abuse messages, we found exactly that effect of counter-arguing (Ray, Ward and Reed, 1976). We studied, among others, two commercials which had quite opposite patterns of cognitive response. "Walkout" was a memorable commercial which produced nearly violent counterarguing from the parents audience. "Big Brother" was not quite as memorable but produced positive cognitive responses such as "connections." In natural lab experiments and in a split cable field experiment the Walkout commercial was substantially more effective than the Big Brother commercial in producing, for instance, positive attitude change, booklet interest and reported numbers of times that respondents thought, read about, and talked about drug abuse during the time the commercials were running.

In this case, as in many situations in the real world, some response was better than no response at all. In such situations the counterarguing so worrisome to information processing researchers, can be a benefit.

Thus, with regard to attention, the advertising situation is more difficult than we expected and involving commercials surmounted the difficulty somewhat, even when they were the opposite of those that do well in the typical lab study.

NEED FOR CARE IN CIP RESEARCH APPLICATION TO OTHER SITUATIONS

The results I've discussed thus far indicate that we must take great care in translating CIP and persuasion lab research to the field in terms of advertising. In many cases lab and field results will be markedly different. Not only should we apply lab results carefully but we should start to alter our research procedures to take account of the nature of the advertising situation.

But there is another approach that might be taken and is illustrated by my final example. Instead of trying to translate to advertising, we can ignore it, and--like the drunk looking for his lost keys under the street light--we cam look for situations in the real world that are similar to those in the lab.

This was and is the approach used by many applied persuasion researchers in the late 60's and in the 70's. For instance, on our campus, Philip Zimbardo trained hundreds of Stanford students--on one occasion in a mass rally--to use social psychological principles in canvassing for peace candidates and issues in 1970. This led to a little booklet entitled Canvassing for Peace which Zimbardo wrote with Robert Abelson (1970). A more general statement of this approach is in the Zimbardo and Ebbesen paperback Influencing Attitudes and Changing Behavior (1969) and in Varela (1971). In consumer research and application it is represented by those who see the most direct applications in personal selling or at the very least in print advertising, usually in an industrial or public service setting--where there might be involvement and the type of processing found in the lab.

Even in such set-up situations, however, I would argue that we need to take great care in application.

For our example here, I'll use the "foot-in-the door" technique. The seminal foot-in-the-door study was conducted by Freedman and Fraser (1966), who asked residents to place a small card in their window. Two weeks later these residents were asked to display a large, clumsily lettered sign. Subjects who had been approached to comply with the earlier request were much more likely to agree with the second one.

Although the foot-in-the-door or small request approach has been found to be quite effective in the social psychological literature, one field experiment done by William Swinyard demonstrated a situation in which adding a small request to a sales message actually decreased the combined effect of the sales message and later advertising exposures (Swinyard and Ray 1976, 1977).

The key response was intention to volunteer for Red Cross work and the small requests were to put a little sign in the prospect's window and wear a Red Cross pin for at least two shopping trips. Both in terms of main effect and interaction with advertising exposures, the small request manipulation was less effective than a persuasive message without the small request. Again, the explanation appears to be related quite directly to the situation. The Red Cross is an issue about which respondents may feel a good deal of guilt. They "know" that they should give blood or volunteer, yet they are fearful or don't want to take the time. They look for simple ways to relieve the guilt and still avoid the anxiety brought on by the needle and the inconvenience of time demands. Compliance with the small request may have provided an easy way for respondents to relieve this guilt, bringing with it a feeling that, "I have made my 'contribution' already...I don't need to do any more.".

Whether you agree with this explanation or not, you should know that ours is not the only research that found the small request manipulation lacking in social marketing situations. Alice Tybout (1976) found that a small request manipulation worked only when the commitment was made salient to respondents. This obviously fits the concern in this paper with the differential involvement found in persuasion research and in the field.

CONCLUSION

In conclusion, I believe I have an answer to the question that started this paper. Consumer information processing research relates to consumer information processing if the situation in the research is the same as the situation in the processing. Practitioners can use our findings if they are very careful to determine whether there is a match between the situation in the research and the situation in which real world processing takes place. At this point I perceive a difference between the situation in most research and that in the most common persuasion category, advertising Most advertising, especially the broadcast kind, represents a low involvement situation. But it is possible to take even present research findings and analyze their implications for advertising. And fortunately the trend in information processing research seems to be toward more natural exposure and response conditions.

REFERENCES

Robert P. Abelson and Philip Zimbardo, Canvassing for Peace (Ann Arbor, Michigan: Society for the Psychological Study of Social Issues, 1970).

Burke Marketing Research, Inc. "Viewer Attitudes Toward Commercial Clutter on Television and Media Buying Implications,'' Presentation to 18th ARF Conference (November 14, 1972).

Jonathan L. Freedman and S. C. Fraser, "Compliance Without Pressure: The Foot-in-the-Door Technique," Journal of Personality and Social Psychology, 4 (March 1966), 195-202.

Carl I. Hovland, "Reconciling Conflicting Results Derived from Experimental and Survey Studies of Attitude Change," American Psychologist, 14 (January 1959), 8-17.

J. P. Keating and Timothy C. Brock, "A Myth About Distraction," American Scientist, 59 (December 1971), 416-9.

Robert A. Osterhouse and Timothy C. Brock, "Distraction Increases Yielding to Propaganda by Inhibition Counterarguing," Journal of Personality and Social Psychology, 15 (August 1970), 344-58.

Michael L. Ray and Alan G. Sawyer, "Repetition in Media Models: A Laboratory Technique," Journal of Marketing Research, 8 (February 1971), 20-30.

Michael L. Ray in collaboration with Alan G. Sawyer, Michael L. Rothschild, Roger M. Heeler, Edward C. Strong, and Jerome B. Reed, "Marketing Communication and the Hierarchy-of-Effects," in Peter Clarke (Ed.), New Models for Mass Communication Research, Beverly Hills, California: Sage Publications, 1973).

Michael L. Ray, Alan G. Sawyer and Edward C. Strong, "Frequency Effects Revisited," Journal of Advertising Research, 11 (February 1971), 14-20.

Michael L. Ray, Scott Ward and Jerome B. Reed, "Pretesting of Anti-Drug Abuse Education and Information Campaigns: Summary Report of a Marketing Science Institute Special Project," in Ronald E. Ostman (Ed.) Communication Research and Drug Education (Beverly Hills, California: Sage Publications, 1976).

Michael L. Ray and Peter Webb, "Experimental Research on the Effects of TV Clutter: Dealing With the Difficult Media Environment," Report No. 76-102, Marketing Science Institute, Cambridge, Massachusetts (April 1976).

Claude E. Shannon and W. Weaver, The Mathematical Theory of Communication (Urbana: University of Illinois Press, 1949).

William R. Swinyard and Michael L. Ray, "The Impact of Psychological Labeling and Small Requests on Consumer Behavior: New Evidence," Submitted to Journal of Consumer Research, 1976

William R. Swinyard, "Advertising-Selling Interactions," Journal of Marketing Research, 14 (1977).

Alice M. Tybout and Bobby J. Calder, "Threats to Internal and External Validity in the Field Setting," in William D. Perreault, Jr. (Ed.) Advances in Consumer Research, Vol. IV, (Chapel Hill, North Carolina: Association for Consumer Research, 1977).

Jacobo A. Varela, Psychological Solutions to Social Problems: An Introduction to Social Technology (New York: Academic Press, 1971).

Peter Webb and Michael L. Ray, "Experimental Research on the Effects of TV Clutter," Advertising Quarterly (1977).

Peter L. Wright, "The Harassed Decision Maker: Time Pressures, Distractions, and the Use of Evidence," Journal of Applied Psychology, 59 (1974).

Philip Zimbardo and Ebbe Ebbesen, Influencing Attitudes and Changing Behavior (Reading, Massachusetts: Addison-Wesley Publishing, 1969).

----------------------------------------