The Dimensions of Advertising Involvement

Andrew A. Mitchell, Carnegie-Mellon University
ABSTRACT - A conceptual definition of involvement and a conceptual model of how involvement affects the information acquisition process are presented. This definition views involvement as an internal state of an individual with both intensity and directional properties. As such, it is one of many different variables that may affect the information acquisition process, This definition and model is then used to discuss alternative definitions of involvement and previous research examining the effect of involvement on communication processes.
[ to cite ]:
Andrew A. Mitchell (1981) ,"The Dimensions of Advertising Involvement", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 25-30.

Advances in Consumer Research Volume 8, 1981      Pages 25-30

THE DIMENSIONS OF ADVERTISING INVOLVEMENT

Andrew A. Mitchell, Carnegie-Mellon University

ABSTRACT -

A conceptual definition of involvement and a conceptual model of how involvement affects the information acquisition process are presented. This definition views involvement as an internal state of an individual with both intensity and directional properties. As such, it is one of many different variables that may affect the information acquisition process, This definition and model is then used to discuss alternative definitions of involvement and previous research examining the effect of involvement on communication processes.

INTRODUCTION

Interest in understanding the effect of involvement on communication processes has been increasing in recent years. There seems to be a general understanding of communication effects under conditions of "high involvement" and a general agreement that communication effects under conditions of "low involvement" are different from those under "high involvement". However, these effects under "low involvement'' conditions are not well understood. Recently, there have been a number of attempts to examine these differences experimentally (e.g. Petty and Cacioppo 1979, Mitchell 1980, Mitchell, Russo and Gardner 1980), however, the "involvement" manipulations differ in these studies so it is difficult to understand the results.

Much of the confusion surrounding involvement and its effect on communication processes seems to center on our failure to understand conceptually what involvement is and how it may affect the information acquisition process. In this paper, I will attempt to clarify this misunderstanding. In the next section I will present a definition of involvement and a conceptual model of how involvement affects the information acquisition process. Next, I will discuss our research which has been based on this model and then use the model to discuss previous research examining the effect of involvement on communication processes and alternative conceptual definitions of involvement.

CONCEPTUAL MODEL

In a previous paper, I defined involvement as an individual level, internal state variable whose motivational properties are evoked by a particular stimulus or situation (Mitchell 1979). As such, it has two dimensions, intensity and direction. Consequently, in discussing involvement in a particular context, we must define both the level of involvement (e.g., high vs. low) and also the direction of involvement (e.g., directed at the advertised brand or at an element of the advertisement).

A conceptual model of the effect of involvement on the information acquisition process is presented in Figure 1. The double arrows indicate the critical stages of the information acquisition process that affect attitude formation and the retrieval of processed information from long term memory. These critical stages are attention and processing. Attention is conceptualized as having capacity limitations (e.g., Kahneman 1973) which causes individuals to focus their cognitive resources on a limited number of stimuli in the environment. In processing information, then, one decision individuals must make is which stimuli to attend to and the amount of attention to devote to each stimuli.

FIGURE l

CONCEPTUAL MODEL

The second stage is processing. Here individuals interpret the information from the environment, make inferences and evaluate the information. Numerous theories from cognitive psychology indicate that how individuals process information affects their ability to retrieve the information at a later point in time (Mitchell 1980). For instance, the depth of processing approach suggests that the use of semantic processing and the amount of elaboration that occurs during processing enhances the ability to retrieve the processed information (e.g., Craik and Tulving 1975, Craik 1979). Other theories hypothesize that other dimensions of processing affect recall (e.g., Paivio 1971, Nelson 1979). Although there is general agreement that the type of processing that occurs during exposure to a stimulus affects the recall of that stimulus, there is currently disagreement as to exactly what dimensions of processing affect recall.

There is also considerable research that indicates that the type of evaluative processing or the type of cognitive responses generated during exposure affects attitude formation and change (e.g., Greenwald 1968, Wright 1973, 1980). For instance, counterarguments reduce the favorableness of the attitudes that are formed while support arguments enhance their favorableness. In summary, then, the two critical stages in the information acquisition process are attention and processing. These two stages will affect what information can be retrieved from long term memory and the formation and change of attitudes.

The factors which affect attention and processing are the stimulus (e.g., advertisement) and the goals of the individual while exposed to the stimulus. These factors, in turn, determine the level and direction of involvement and the particular memory schema that is used to process the information. Here we define memory scheme as an organized set of information about a concept in long term memory (e.g., Rumelhart and Ortony 1978).

To illustrate the model, assume that a number of individuals are exposed to an advertisement for a new fuel-efficient automobile. In the advertisement, the automobile is shown on a dock surrounded by a number of sailboats.  If an individual, for instance, was planning to purchase a small fuel efficient automobile in the near future and had not decided which brand to purchase, he or she would probably be very interested in this product and may be actively searching for information about the various alternatives on the market. Consequently, this individual would probably devote full attention to the advertisement and activate a memory schema that contains information about fuel-efficient automobiles so that he or she can critically evaluate the information in the advertisement. As a consequence of this processing, the information about the new fuel-efficient automobile would be organized into a tightly integrated knowledge structure and, in most cases, an evaluation of the new automobile would be formed.

Another individual, however, may not be interested in purchasing an automobile or even obtaining information about new brands of fuel-efficient automobiles. When this individual is exposed to the same advertisement, he or she may be attracted to the picture of the sailboat in the advertisement due to an interest in sailing. This individual might also devote full attention to the advertisement and may skim part of the copy; however, he or she would probably activate their memory schema for sailboats in processing information from the advertisement. This would result in a different type of processing than in the first case. Since the individual did not activate the memory schema that contains information about fuel efficient automobiles, he or she would not have comprehended the message (Rumelhart and Ortony 1978) and would generate few, if any, counterarguments or support arguments. As a result, he or she may have acquired some information about the advertised car, but it would be much less information than in the previous example and it probably would not be tightly organized into a knowledge structure about the advertised automobile. In addition, the individual probably would not have formed an evaluation of the automobile during exposure to the advertisement.

Finally, a third individual may have little interest in automobiles or sailboats and, therefore, this individual would probably devote little attention to the advertisement and would probably do little active processing of the information in the advertisement. In this case, the individual may be unable to retrieve any information about the new automobile and may not be aware of its existence. In summary, then, the content of the stimulus and the goals of the individual determine the amount and direction of involvement during exposure to the advertisement. The intensity of the involvement determines how such attention is devoted to the advertisement. The direction of the involvement determines which memory schema is activated which, in turn, determines the type of processing that occurs during exposure.

OUR RESEARCH

In our research (Mitchell 1980, Mitchell, Russo and Gardner 1980), we have directly manipulated processing strategy and attention levels to examine the resulting effects on attitudes, intentions and the retrieval of brand related information from long term memory. In this research, we differentiate between a brand and a nonbrand processing strategy. A brand processing strategy involves the active processing of brand information from an advertisement to either form an evaluation of the brand or acquire information about the brand. A nonbrand processing strategy involves the processing of information from the advertisement to achieve some other goal. This may be caused, for instance, by the use of an attention getting device such as an attractive model in the advertisement.

To manipulate processing strategy, we instruct subjects to either examine the advertisement to form an evaluation of the advertised brand (brand processing strategy) or to evaluate the effectiveness of the advertisement (nonbrand processing strategy). For the latter manipulation, we provide subjects with a set of criteria to use in evaluating the effectiveness of the advertisement. These two different instructions presumably result in the activation of different memory schema for use in processing the information.

In our initial study, (Gardner, Mitchell and Russo 1978, Mitchell, Russo and Gardner 1980), we examined the effect of the two different processing strategies with high attention levels on attitudes and the retrieval of information from long term memory. The subjects in each condition were shown advertisements for four new brands in four different product categories. The results indicated that subjects executing brand processing strategy were able to retrieve more product related information faster and at a greater level of accuracy than subjects executing nonbrand processing strategy (Mitchell, Russo and Gardner 1980). In addition, subjects using a brand processing strategy formed less favorable attitudes toward the advertised brand. Although the subjects using s brand processing strategy generated almost twenty times as many counterarguments, support argument and support derogation statements as subjects using a nonbrand processing strategy, these differences only partially explained the differences in attitudes. More specifically, the use of the number of counterarguments, support arguments and source derogations as covariates did not eliminate all the reliable difference in attitudes between the two groups. These results suggest the possibility that other mediators may affect attitude formation in these conditions.

In a second study, these two different processing manipulations were used along with two different manipulations of attention levels (Mitchell 1980). For one attention level manipulation, subjects were shown advertisements for new products and were instructed to execute a brand processing strategy while watching a red and a blue light that randomly flashed at the rate of approximately 25 flashes per second. Subjects were instructed to raise their left hand whenever one of the two lights flashed. The second attention level manipulation required the subjects to count backward by three's from a three digit number while examining the advertisements. The subjects in each condition saw the same eight advertisements for new brands in eight different product categories. The results indicated differences in attitudes and the amount of product related information retrieved from long term memory between all conditions except the nonbrand processing condition and the second attention level manipulation where subjects had to count backward by three's.

These results suggest that differences in involvement may result in three different types of information acquisition processes. One process is caused by a "high involvement" condition and two are caused by low involvement" conditions (Figure 2). Here "high involvement'' refers to high interest levels in the advertised brand. Under the "high involvement" condition, individuals devote all their attention to the advertisement and execute a brand processing strategy. Consequently, they will critically evaluate the brand information in the advertisement and will, generally, form an overall evaluation of the advertised brand during exposure to the advertisement. This means that their verbal thought processes during exposure to the advertisement will contain a large number of counterarguments and support arguments.

Under the first "low involvement" condition, individuals will also execute a brand processing strategy; however, this strategy will be executed with reduced attention level. This will generally result in a reduction in the number of counterarguments and support arguments generated which, in turn, will frequently result in a difference between this condition and the "high involvement" condition in the attitudes that are formed. Under this condition, then, individuals will comprehend the message since the proper schema is activated, but they will not be as critical of the information as they were under conditions of "high involvement".

FIGURE 2

ALTERNATIVE MODELS OF INFORMATION ACQUISITION UNDER DIFFERENT INVOLVEMENT CONDITIONS

Finally, the second "low involvement" condition occurs when individuals etcher execute a nonbrand evaluation strategy or there are severe attention deficits. There will be few, if any, counterarguments or support arguments generated and the individuals will not totally comprehend the message. They will, however, acquire some information about the advertised brand.

These three different information acquisition processes should cause differences in the amount, content and organization of information about the advertised brand in memory. In the previously mentioned study (Mitchell 1980) subjects in the nonbrand brand processing condition and the second attention level condition could recall significantly less information about the advertised brands than subjects in the brand processing condition and the first attention level condition. It also took them longer to verify statements about the attributes of the advertised brands and their evaluation of them. No differences were found on these two measures between the brand strategy condition and the first attention level condition or the nonbrand strategy condition and the second attention level condition. Since attitudinal differences were found between the first two conditions, however, we might expect differences in the content of the information that would be recalled.

In summary, our conceptualization of involvement and our research suggest that differences in involvement may result in three different types of information acquisition processes -- one caused by a "high involvement" condition and two by "low involvement" condition. In addition, we believe that the latter two processes may be caused by factors other than involvement. One of these may be a lack of knowledge about the topic of the communication.

ALTERNATIVE APPROACHES TO INVOLVEMENT

There have been a number of articles over the last ten years either discussing the effect of involvement on the reception of communications or experimental studies which manipulate involvement and examine the resulting effects. As mentioned previously, these conceptualizations and manipulations have generally differed. In this section, will discuss these alternative manipulations and conceptualizations using the model presented in the previous section.

Conceptual Approaches

Krugman:  In his original article, Krugman (1965) suggested that the cognitive processes that occur during exposure to television advertisements were similar to those that occur, during the learning nonsense syllables. As evidence of this similarity, Krugman mentioned that the recall of three consecutive television commercials displays the same U shaped relationship as the recall of a series of nonsense syllables -- strong primacy and recency effects. He further hypothesized that television advertising produces subtle shifts in our perceptions of brands which result in changes in the saliency of the different attributes of the product. In this article, and in a second article (Krugman 1967), he defines involvement as the number of "bridging experiences, connections or personnel references per minute that the viewer makes between his own life and stimulus", not as the "amount of attention, interest or excitement".

This definition of involvement differs from the one presented here in that Krugman defines involvement as one of the dimensions of the type of processing that occurs during exposure to the advertisement. In contrast, I define involvement as a particular state of the individual at a point in time. Our state variable conceptualization of involvement affects the type of processing that occurs during exposure, however, it is not the only variable that may affect it. The amount and type of knowledge that an individual has and the memory schema that is activated during exposure also affects the resulting processes. Therefore, it is entirely possible that according to our definition an individual may be highly involved, however, if this person has little knowledge about the advertised product, the resulting process might be a low involvement one according to Krugman's definition.

In general, we believe that information acquisition under the second low involvement condition is similar to the one Krugman describes in his original paper. Under our second low involvement condition individuals acquire information about the advertised brand but do not actively organize this information into a conceptual understanding of the product. At some later point in time, something may trigger the individual to organize this information about the advertised brand, and at this point, an attitude may be formed.

In an experiment which I am currently conducting, we find that subjects who executed a nonbrand processing strategy during exposure to a set of advertisements lose information more rapidly during a two week interval after exposure than subjects that executed a brand processing strategy during exposure to the same advertisements.

This result also seems to be consistent with Krugman's conceptualization of what should occur under low involvement conditions.

Greenwald, Leavitt and Obermiller:  Greenwald, et al (1980) also define involvement as a process and suggest that there may be a "high involvement" process and two "low involvement" processes. Their "high involvement" process agrees with my conceptualization of information acquisition under conditions of "high involvement", however, we differ on the other two processes. The first "lay involvement" process proposed by Greenwald, et al (1980) seems to contain both of our information acquisition processes under conditions of "low involvement". We, however, believe that the distinction we make is an important one since, as discussed previously, these two processes result in differences in the amount, content and organization of the processed information in long term memory and may result in attitudinal differences.

The second "low involvement" process suggested by Greenwald, et al (1980) is based on the "mere exposure" literature (Zajonc 1980). This literature indicates that attitudes may be formed toward a stimulus even though the individual cannot determine, at better than chance levels, whether he or she has seen the stimuli (Wilson 1979). Greenwald, et al (1980) refer to this as a pre-attentive process.

Although these ideas are interesting, they have yet to be demonstrated with complex as opposed to simple stimuli. Part of the problem with examining information acquisition under these conditions is defining exactly what non-attention means in terms of experimental manipulations. Kellogg (1980), for instance, has recently suggested that individuals can recognize unattended stimuli. This conclusion is based on the results from a series of experiments where subjects looked at pictures of faces while performing a complex multiplication task. Kellogg found that under these conditions his subjects could recognize the pictures at greater than chance levels, however, there was a decrement in recognition performance compared to when the subjects looked at the faces under conditions of full attention.

Problems arise in these types of studies, however, since individuals can acquire information through different channels. Research, for instance, has indicated that individuals have both a verbal and visual channel for acquiring information (e.g., Brooks 1968) and a single distractor task may not use up the capacity of both channels. Kellogg (1980), presents evidence that his complex multiplication task used up the capacity of the verbal channel, however, it is not clear whether his instructions to produce a mental image of the task used up the capacity of the visual channel.

In the experiment discussed previously, for instance, we found that when the advertisements contained visual information, our subjects formed different attitudes under the second attention level condition than a control group That did not see the advertisements (Mitchell 1980). This did not occur for advertisements that contained primarily verbal information. Consequently, in order to examine whether learning can occur under these conditions, we need to define more precisely what we mean by nonattention during exposure.

Issue Involvement

Concern with the effects of issue involvement on persuasion goes back at least as far as the research by Hovland and Janis on communication effects (Hovland, Janis and Kelley 1953). This early research led to the development of social judgment theory which provides an explanation of the effect of issue involvement on persuasion (Sherif, Sherif and Nebergall 1965). According to this theory, individuals that are highly involved with an issue will have large "latitudes of rejection" and will, therefore, be more likely to reject a persuasive message on this issue.

Recently, Petty and Cacioppo (1979) examined the effect of issue involvement on the number of counterarguments and support arguments generated during exposure to a persuasive communication. In two different experiments, Petty and Cacioppo (1979) manipulated issue involvement by presenting subjects with a persuasive communication advocating a change in coed visitation hours. In the high involvement condition, these changes were advocated for the student subjects' own university, while in the low involvement condition the change was advocated for another university. The other manipulations included the direction of change advocated and the strength of the arguments supporting the change. As might be expected, increased involvement enhanced persuasion for the proattitudinal message (i.e., increase visitation hours) and reduced persuasion for the counterattitudinal message (i.e., decrease visitation hours). Increasing involvement also enhanced persuasion for the message with the strong arguments even though the message was counterattitudinal, but reduced persuasion for the message with weak arguments. Perhaps one of the most interesting findings was that under lay issue involvement conditions, the correlations between cognitive response measures (e.g., counterarguments) and attitudes were considerably lower then under conditions of high involvement (average absolute values of .29 vs. .69).

In terms of the model that was presented here, the topic of the message affected the subjects' involvement with the message. This, in turn, resulted in differences in attention levels and processing strategies. In each case, the authors found that low issue involvement resulted in a general reduction in the amount of evaluative processing (i.e., number of counterarguments and source arguments) that occurred. This indicates that the involvement manipulation used in these experiments resulted in information acquisition under our first low involvement condition.

As mentioned previously, they also found that the magnitude of the relationships between the number of counterarguments and support arguments and the resulting attitudes were greater under the high involvement manipulation than under the low involvement manipulation. It is interesting to note that in our study (Mitchell, Ruses and Gardner 1980), we found that the magnitude of the relationship between a measure of attitude and attitudes predicted from the Fishbein model was greater with the brand processing strategy than with the nonbrand processing strategy. In other studies, we have generally found that the number of counterarguments and support arguments and predicted attitudes from the Fishbein model explained the same variance in independent measures of attitudes (Mitchell 1981). Therefore, it would seem that the type of semantic processing that occurs during exposure to a communication and the semantic information in memory after exposure explains more of the variance in attitudes when information is acquired under conditions of high involvement as opposed to low involvement. This suggests that either other mediators of attitude formation and change may be operating under these latter conditions or that the resulting attitudes may be less reliable.

Media Involvement

Krugman (1965, 1966) has suggested that, based on his definition of involvement, the media used in advertising a product determines the resulting level of involvement during exposure. According to Krugman, television advertising results in low involvement conditions while print results in high involvement conditions.

Wright (1974) tested this hypothesis by examining the effect of providing the same information both visually (print) and auditorily (radio) and found that his subjects generated fewer counterarguments and support arguments with the audio message as opposed to the visual message. Wright argued that these differences occurred because individuals do not have the opportunity to verbally respond to audio messages mince they cannot control the rate at which they receive information. In other words, capacity constraints limit the amount of verbal processing that occurs during an audio message. It should be noted that if Krugman is correct in his hypothesis concerning the similarity in the type of processing that occurs during exposure to a television commercial and the processing that occurs during the learning of nonsense syllables, then Wright's explanation is consistent with the currently accepted explanation for the finding of primacy and recency effects in verbal learning experiments.

Much of this argument is based on how involvement is conceptualized. As mentioned previously, Krugman views involvement as a process -- essentially the type of processing that occurs during exposure to a communication.  If, however, involvement is viewed as a state variable and only one of many variables that may affect the type of processing that occurs during exposure to an advertisement, it would be possible for someone to be highly involved with a television advertisement. However, because less time is available to process the information, he or she may execute a different type of processing than would occur if the same information is presented in print.

Brain Wave Activity and Involvement

In a provocative article, Krugman (1977) suggested that the processing of information from television advertisements may be primarily a right brain activity while the processing of information from print advertisements may be primarily a left brain activity. Based on this hypothesis, he also suggested that the primary effect of most television advertising may be the storing of visual images in the right hemisphere of the brain. Under these conditions, individuals would not necessarily be able to recall specific television commercials, however, individuals would be able to recognize whether or not they had seen a television commercial. Supposedly, according to Krugman, their recognition of a specific commercial may have an effect at the point of purchase.

To test this hypothesis, Appel, Weinstein and Weinstein {1979) measured the brain waves of subjects exposed to high recall and low recall television advertisements that were presented three times to each subject. The results indicated that advertisements with higher recall scores generated higher levels of brain wave activity (e.g., more seconds of alpha waves), however, there was no difference in the amount of brain wave activity between the left and right hemispheres. Krugman (1980), however, after reexamining the data, found brain wave activity in the left hemisphere declined with repetition, but brain wave activity in the right hemisphere remained constant with repetition. In a later study, Weinstein, Appel and Weinstein (1980) reported no difference in the amount of brain wave activity in the left and right hemispheres for subjects exposed to print and television advertisements. Magazine advertisements, however, generated a higher level of brain wave activity in both hemispheres as opposed to television advertisements.

These results seem to indicate that the amount of brain wave activity in each hemisphere is generally the same during exposure to advertisements in both television and print. There may, however, be differences after a number of repetitions. In addition, magazine advertisements seem to generate more brain wave activity. If there is a strong relationship between brain wave activity and the amount of cognitive response generated, and I suspect that there is, these letter results would tend to confirm the findings of Wright (1974). Cacioppo and Petty (1979), for instance, found a positive relationship between a number of physiological measures (e.g., oral muscle, cardiac and respiratory activity) and the number of counterarguments and support arguments generated.

DISCUSSION

In this paper, I have presented a definition of involvement and a conceptual model of how involvement affects the information acquisition process. This definition, which defines involvement as an internal state variable whose motivational properties are evoked by a particular stimulus, differs from other definitions which define involvement in terms of the type of processing that occurs during exposure to an advertisement. Under the conceptualization of involvement presented here, involve-mint is only one of many possible causes of these processes. Others are the amount of knowledge that an individual has about the topic of the communication and the opportunity to cognitively respond to the communication. In general, much of the confusion in the literature concerning involvement seems to center on different conceptualizations of involvement.

We have also suggested that different levels of involvement may cause three different types of information acquisition processes. One is caused by high involvement and two by low involvement. The issue involvement manipulation used by Petty and Cacioppo (1979) seems to have resulted in information acquisition under the first Low involvement condition, while we suggest that Krugman (1965) is essentially discussing information acquisition under our second low involvement condition.

One interesting effect of involvement on the information acquisition process that has been found in two different experiments is a reduced relationship between semantic information and attitudes. Petty and Cacioppo (1979), found smaller correlations between counterarguments and attitudes and support arguments and attitudes under conditions of low issue involvement as opposed to high issue involvement. Mitchell, Russo and Gardner (1980) found smaller correlations between predictions from the Fishbein model and attitudes under a nonbrand processing strategy as opposed to s brand processing strategy. These lower correlations may occur for two different reasons. First, attitudes formed under conditions of low involvement may have lower reliability. This would result in lower observed correlations. Second, other mediators of attitude formation may have a greater effect under these conditions. Clearly, additional research is required to both verify this effect and, if valid, to determine its cause.

Finally, it should be noted that the examination of factors affecting the information acquisition process can occur at different levels. Petty and Cacioppo (1979) and Wright (1974), for instance, examined the effects of message involvement and modality while Mitchell (1980) and Mitchell, Russo and Gardner (1980) examined the effects of different attention levels and processing strategies. Mitchell (1981) has referred to the latter manipulations as Level One variables affecting the information acquisition process while the former are called Level Two variables. Obviously, research at both levels is required to understand communication effects, however, it may be very difficult to obtain Model III information acquisition processes with Level Two variables in the laboratory. This may occur because knowledge that they are participating in a laboratory experiment may automatically raise the involvement level of subjects so that only Model I or Model II processes may occur.

REFERENCES

Appel, V., S. Weinstein and C. Weinstein (1979), "Brain Activity and Recall of TV Advertising," Journal of Advertising Research, 19, 7-15.

Brooks, L. R. (1968), "Spatial and Verbal Components of the Act of Recall," Canadian Journal of Psychology, 22, 349-350.

Cacioppo, J. T. and R. E. Petty (1979), "Attitudes and Cognitive Response: An Electrophysiological Approach," Journal of Personality and Social Psychology, 37, 2181-2199.

Craik, F. I. M. (1979), "Levels of Processing: Overview and Closing Consents," in L. S. Cermak and F. I. M. Craik, eds., Levels of Processing in Human Memory, Hillsdale, NJ: Lawrence Erlbaum Associates, 447-462.

Craik, F. I. M. and R. S. Lockhart (1972), "Levels of Processing: A Framework for Memory Research," Journal of Verbal Learning and Verbal Behavior, 11, 671-684.

Craik, F. I. M. and E. Tulving (1975), "Depth of Processing and the Retention of Words in Episodic Memory," Journal of Experimental Psychology: General, 104, 268-294.

Gardner, M., A. A. Mitchell and J. E. guess (1978), "Chronometric Analysis: An Introduction and an Application to Low Involvement Perception of Advertisements," in H. K. Hunt, ed., Advances in Consumer Research Vol. 5, Ann Arbor: Association for Consumer Research, 581-589.

Greenwald, A. G. (1968), "Cognitive Learning, Cognitive Response to Persuasion and Attitude Change," in A. G. Greenwald, et al. eds., Psychological Foundations of Attitudes, New York: Academic Press, l47-170.

Greenwald, A. G., C. Leavitt and C. Obermiller (1980), "What is Low Consumer Involvement?" Paper presented at the American Psychological Association Meeting in Montreal.

Hovland, C., I. L. Janis and H. H. Kelley (1953), Communication and Persuasion, New Haven, Conn.: Yale University Press.

Kahneman, D. (1973), Attention and Effort. Englewood Cliffs, NJ: Prentice-Hall, Inc.

Kellogg, R. T. (1980), "Is Conscious Attention Necessary for Long Term Storage?" Journal of Experimental Psychology: Human Learning and Memory, 6, 379-390.

Krugman, H. E. (1980), "Sustained Viewing of Television," Journal of Advertising Research, 20, 65-68.

Krugman, H. E. (1977), "Memory Without Recall, Exposure without Perception," Journal of Advertising Research, 17, 7-12.

Krugman, H. E. (1971), "Brain Wave Measures of Media Involvement," Journal of Advertising Research, 11, 3-10.

Krugman, H. E. (1967), "The Measurement of Advertising Involvement," Public Opinion Quarterly, 30, 583-596.

Krugman, H. E. (1965), "The Impact of Television Advertising: Learning without Involvement," Public Opinion Quarterly, 29, 349-356.

Mitchell, A. A. (1981), "Cognitive Processes Initiated by Exposure to Advertising," In R. Harris, ed., Information

Processing Research in Advertising, Hillsdale, NJ: Lawrence Erlbaum Associates.

Mitchell, A. A. (1980), 'Using an Information Processing Approach to Understand Advertising Effects," in J. Olson, ed., Advances in Consumer Research, Vol. 7, Ann Arbor: Association for Consumer Research, 171-177.

Mitchell, A. A. (1980), "The Relationship Between Cognitive Processes and Co8nitive Structures," in. I. Ross, ed., Proceedings. Division 23 Program. 87th Annual Convention, American Psychological Association, Vol. IX.

Mitchell, A. A. (1979), "Involvement: A Potentially Important Mediator of Consumer Behavior," in W. Wilkie ed., Advances in Consumer Research, Vol. 6, Ann Arbor: Association for Consumer Research, 191-196.

Mitchell, A. A., J. E. Ruses and M. Gardner (1980), "Strategy Induced Low Involvement Processing of Advertising Messages," Working Paper, Graduate School of Industrial Administration, Carnegie-Mellon University, Pittsburgh, PA. 15213.

Nelson, D. L. (1979), "Remembering Pictures and Words: Appearance, Significance and Name," in L. S. Cermak and F. I. M. Craik, eds., Levels of Processing In Human Memory, Hillsdale, NJ: Lawrence Erlbaum Associates, 45-76.

Paivio, A. (1971), Imagery and Verbal Processes, New York: Holt, Rinehart and Winston.

Petty, R. E. and J. T. Cacioppo (1979), "Issue Involvement Can Increase or Decrease Persuasion by Enhancing Message -Relevant Cognitive Responses," Journal of Personality and Social Psychology, 37, 1915-1926.

Rumelhart, D. E. and A. Ortony (1978), "The Representation of Knowledge in Memory," in R. C. Anderson, R. J. Spire and W. E. Montague, eds., Schooling and the Acquisition of Knowledge, Hillsdale, NJ: Lawrence Erlbaum Associates, 37-54.

Sherif, C. W., M. Sherif and R. E. Nebergall (1965), Attitude and Attitude Change, Philadelphia, PA., Sanders.

Weinstein, S., V. Appel and C. Weinstein (1980), "Brain-Activity Responses to Magazine and Television Advertising," Journal of Advertising Research, 20, 57-68.

Wilson, W. R. (1979), "Feeling More Than We Know: Exposure Effects without Learning," Journal of Personality and Social Psychology, 37, 811-821.

Wright, P. L. (1980), "Message-Evoked Thoughts: Persuasion Research Using Thought Verbalizations," Journal of Consumer Research, 7, 151-175.

Wright, P. L. (197&), "Analyzing Media Effects on Advertising Responses," Public Opinion Quarterly, 38, 192-205.

Wright, P. L. (1973), "The Cognitive Processes Mediating Acceptance of Advertising," Journal of Marketing Research, 10, 53-62.

Zajonc, R. B. (1980), "Feeling and Thinker: Preferences Need No Inferences," American Psychologist, 35, 151-175.

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