A Framework For Relating Consumer Involvement to Lateral Brain Functioning

Banwari Mittal, State University of New York at Buffalo
ABSTRACT - The past literature on involvement and hemispherical specialization has advanced two hypotheses: i) high (low) involvement induces left (right) brain processes, and ii) print (TV) is dominantly processed in the left (right) brain. Both these assumptions are questioned here. We base our argument on a recent view of involvement, namely that involvement can be cognitive or it can be affective. We develop this conception further, and then link it to left/right brain dominance via a chain of intermediate processes. Our conceptual framework: a) assigns high-involvement to both hemispheres, and b) predicates left/ right brain engagement upon the type of involvement a stimulus engenders regardless stimulus modality.
[ to cite ]:
Banwari Mittal (1987) ,"A Framework For Relating Consumer Involvement to Lateral Brain Functioning", in NA - Advances in Consumer Research Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT : Association for Consumer Research, Pages: 41-45.

Advances in Consumer Research Volume 14, 1987      Pages 41-45


Banwari Mittal, State University of New York at Buffalo


The past literature on involvement and hemispherical specialization has advanced two hypotheses: i) high (low) involvement induces left (right) brain processes, and ii) print (TV) is dominantly processed in the left (right) brain. Both these assumptions are questioned here. We base our argument on a recent view of involvement, namely that involvement can be cognitive or it can be affective. We develop this conception further, and then link it to left/right brain dominance via a chain of intermediate processes. Our conceptual framework: a) assigns high-involvement to both hemispheres, and b) predicates left/ right brain engagement upon the type of involvement a stimulus engenders regardless stimulus modality.


Brain-wave measures of advertising processing have in recent consumer research received increased attention (Olson and Ray 1983; Alwitt 1983; Rothschild and Thorson 1983). A well-established theory in neuropsychology is that the left and right halves of the human brain process information differently (see Moscovitch 1979 for a review). Some advertising researchers have subsequently proposed correspondence between differential brain activity and a) exposure to print versus TV; b) low versus high consumer involvement. Since involvement has been demonstrated to be an important mediator of consumer behavior, linking involvement to differential brain activity would provide researchers new capability to unobtrusively measure advertising effects. This potentially fruitful line of theorizing has not been adequately pursued, however. The purpose of the present paper is to develop a conceptual framework of the relationship between involvement and hemispherical lateralization.

Existing Theory

Krugman (1977) was the first marketing researcher to propose that the "medium of print is a left-brain function and TV largely or relatively a right-brain function." Krugman went on to suggest, "I would add to that my own views that high involvement is more a left-brain, and low involvement a right-brain activity ." Krugman s was a heuristic, general essay on the cultural impact of mass media. He could not have offered empirical support for his hypothesis, for none was available then.

In a comprehensive survey of marketing applications of the lateralization theory, Hansen (1981) followed Krugman and stated that "in terms of hemispherical specialization, this should imply that in low involvement situations, right brain processes dominate, whereas higher degrees of involvement give rise to left-brain processes." As support, Hansen cited the theory that left-brain specializes in verbal, cognitive, attributional information processing whereas right-brain specializes in nonverbal, pictorial, holistic perception. This theory about brain-lateralization is well supported, of course (see Olson and Bay 1983, Rothschild and Thorson 1983, and Tucker 1981). Critical to Hansen's equation of high/low involvement with left/ right brain dominance is this implicit assumption: low involvement generates noncognitive, nonverbal, pictorial, holistic processing, whereas high involvement generates verbal, cognitive, attributional processing of information. As Stephens (1985) has pointed out, this assumption seems oversimplistic and is examined below in view of some recent developments in involvement theory.


Although there are diverse conceptions of involvement in the literature (see Antil 1984, and Muncy and Hunt 1984, for recent reviews), Hansen's own definition is quite appropriate. Hansen (1981) defines involvement as "variations in the extent to which the individual is more or less motivated toward a specific piece of information, product, or the like." He goes on to say that with high involvement, "more psychic energy is released for handling incoming information, sorting it out, and making choices." This view of involvement is appealing for several reasons. First, it is quite congruent with some major, recent definitions advanced by involvement researchers (Cohen 1983; Mitchell 1981, Park and Mittal 1985). Second, it is general enough to apply to the viewing of an advertisement, product use, or purchase task. Third, it defines involvement in terms of motivational energy, and not in terms of information processing.

Because involvement is motivational energy brought to bear upon whatever task is at hand, such energy can be characterized along a "level" (i.e., low/high) dimension and also along a dimension that recognizes differences in the underlying motives. Regarding the latter, although there are various classification schemes of motives, Park and Mittal (1985) have found McGuire's (1974) broad grouping of motives into cognitive" and "affective" categories to be particularly useful. Accordingly, they have put-forth the idea of cognitive and affective types of involvement. Elsewhere (Mittal 1982, 1983; Park and Mittal 1985) we have specified that affective-involvement obtains when psycho-social interpretation of the stimulus is pertinent to one's goals. Such psycho-social appraisal proceeds with the consideration of overall social image of the stimulus, or experience of hedonic satisfaction or emotional experience. This contrasts with the cognitive, information processing paradigm where information about performance levels on some functional criteria is assumed to be considered. AT&T's commercial which depicts an emotionally moved mother ("He said, 'I called you, Mom, because I love you'") is a case in point. That commercial is purported, it would seem, to induce affective involvement in the viewer. If and when the viewer is moved by the commercial, he/she would be in a heightened state of arousal. In approaching/avoiding the long-distance-call decision in that state of mind, he/she may find himself/ herself incapacitated for logical thinking, as high arousal interferes with problem solving task performance (Berlyne 1978). Moreover, the "cold cognitions" about the choice alternatives (i.e., the cost of the call, the past month's phone bill, the current month's budget, availability of things to say, etc.) would be overshadowed or made less salient by the experience of an overwhelming, global, intense emotional reaction.

The above effect was vividly demonstrated by Park and Young (1986) who induced in their subjects low, high-cognitive, and high-affective involvement. They fount that the high-affective involvement group generated substantially less cognitive responses than tit the high-cognitive involvement group. This result is explained by positing that in the affective-involvement condition as opposed to the cognitive-involvement condition, products, brands, advertisements, or other stimuli are processed holistically without much attention to the specific, disparate features. Hansen assumes that holistic, gestalt-like processes (which no doubt predominantly occur in the right-half brain) represent low-involvement conditions. It now seems logical that these holistic processes could be high-affective involvement processes.

One obvious contest where affective rather than cognitive involvement would be expected is that of experiential or hedonic consumption. Dwelling on experiential/hedonic consumption, Hirschman and Holbrook (1982) advance propositions which would link affective involvement to right-brain processes quite readily. These authors view hedonic consumption as being tied to the imaginative construction of reality. The imaginative construction entails multi-sensory imageries and fantasizing. In the selection of brands of symbolic/hedonic products, therefore, a consumer appraises brand, holistically via multisensory imageries. The consumer is affectively involved, and his involvement is high, not low. Involvement is high because the consumer is emotionally aroused, ant, to borrow another of Hirschman-Holbrook propositions, there is generated "substantial mental activity on the part of the consumer." This mental activity would occur predominantly in the right-half brain because most normal-brained, right-handed humans have been fount to possess a right-hemisphere advantage for emotional and imagery experience (Moscovitch 1979, Nevid 1984).

A Framework of Correspondence between Involvement and Hemispherical Lateralization

Table I presents a framework that relates the types and levels of involvement to hemispherical specialization via a chain of intermediate processes. First, 3 categories of involvement are distinguished: high-cognitive, high-affective, and low involvement (i.e., low in cognitive and low in affective involvement). See Column 1.

In column 2 are described aspects of the stimuli that are relevant to the 3 categories of involvement. As is logical, when the involvement is of the high-cognitive type, the product's performance dimensions or information about product-features are of utmost relevance. When the involvement is of the high-affective type, the product's image-dimensions and the emotional or image situation contained in an advertisement are relevant. If neither performance nor image dimensions are significantly important. then low involvement occurs.

The third column describes as to what aspects of the stimuli are processed. By "processed" is meant "attended to," "encoded," or "utilized." Entries in this column follow closely those in column 2 for the high-affective and high-cognitive involvement cases. The entry for the low-involvement case needs some clarification. A consumer may attend to a single feature or so if the information about it is readily available (Rothschild 1979). Or, he/she may form a quick, vague and shallow impression of the product. In many cases the information about the single feature attended to may constitute the vague image. In other cases, the vague image may be tied to no discernible feature. A brand of salt may be processed, for example, by promptly noting that it is iodized and bence desirable; or it may be chosen simply because an avoidance-disposition was not experienced. Now, the brand selection of salt can be viewed as a low-cognitive involvement task due to a utilitarian motive underlying its purchase. A greeting cart may, on the other band, be 8 low-affective involvement product for many consumers. (For some consumers in some situations, it may be a high-affective involvement product.) The greeting card may be chosen, then, by checking that its message is relevant or that it is not too costly; alternatively it may be chosen because it looked like a "fairly attractive" card. In other words, either an objective feature or two may be processed or a quick, rough impression might be formed as prelude to the card selection. It may be tempting to link low-cognitive involvement with the processing of a single feature and low affective involvement with the vague, shallow image formation. But this view would seem to be oversimplistic, and as the salt and card examples cited above show, either type of processing may occur for low-cognitive, and likewise for low-affective involvement. Presently, we simply to not know enough to make categorical statements one way or another, and there ore leave the low-involvement category as undifferentiated.

Moving further on in Table 1, column 4 describes the extent of mental activity. Suffice here to note that mental activity is expected to be high for high-affective involvement as well as for high-cognitive involvement. (There is no need to equate mentalism with cognitivism!)

Now, some clarification of terminology is in order. Following Zajonc (1980), affect is viewed as an approach/avoidance tendency. Therefore, the attitude that is typically modeled in multiattribute models (c.f. Wilkie and Pessemier 1973) is a kind of affect. To distinguish it from emotion, for example, emotion is labeled as "hot" affect, and multi-attribute attitudes as "cold-affect." Pot-affect is an attitude which cannot be decomposed into its componential attributes. It may be proper, too, to attach an intensity dimension to affect so that affect may be not merely cold or hot but also weak or strong. In the social judgment paradigm (Sherif and Hovland 1961), such intensity is operationalized by latitude of acceptance/ rejection.

In column 5, we hypothesize high-cognitive involvement to engender feature cognitions, i.e., brand beliefs; and through a synthesis of these brand beliefs, strong, cold affect is produced. In the high affective-involvement condition, rich imagery is produced, encoding is global and sensory, and this imagery evokes strong, hot affect. Finally, in the low-involvement condition (low-cognitive and low-affective), shallow cognitive processing (i.e., the processing of the most easily accessible feature information) may occur or a surface level general image may be formed. Consequently, very weak colt affect or very weak hot affect will be expected. Once again, not enough is known to make more definitive statements about this condition.

Finally, Column 6 specifies the engagement of the left/right brains. In the high-cognitive involvement condition, the left-brain is engaged dominantly and the right-brain moderately. In the high-affective involvement condition, the right brain is engaged strongly, and the left-brain moderately. In the low-involvement condition, both the hemispheres are engaged minimally. In the high-cognitive involvement condition, we posit the engagement of the right-brain to accommodate the strong colt affect generated in this condition. Because this affect is colt, we assume that the right brain is engaged only at a moderate level. In the high-affective involvement condition, the right brain is assumed to be engaged deeply tue to the high hot affect which this condition engender. But we also assume that the left brain is also engaged, although moderately. This is because any object or stimulus which causes an intense hot affect to be experienced will also cause at least a few cognitions or beliefs. If you immensely like someone, you might allow yourself to believe that he is intelligent, for example. Such a cognition might even be an afterthought. Although it may not necessarily be produced, it seems intuitively plausible, and therefore one must allow for it. The simultaneous engagement of the two brain-halves (although to differing degrees) is compatible with and may indeed be required to accommodate the evidence that the two hemispheres constantly communicate with each other in "normal" humans (see Katz 1980).

Now, to capture the dynamic nature of brain processes, a time dimension is incorporated and hypothesized brain-waive patterns are shown in Figure l. A pattern market by early onset and longer persistence of a more intense (i.e., of greater amplitude) brain-wave will be termed "primary," and a less marked pal tern m "secondary.

Panel 1 of Figure 1 shows the progression of the left and right brain processes as a stimulus is processed in the high-cognitive involvement condition. Here, the left-brain process is hypothesized to be the primary process whereas the right-brain shows a secondary process. In column 5 of the table a time sequence is implied between the mental activities specified there. Congruent with the multi-attribute models of attitude, the strong, colt affect is assumed to follow upon the formation of brand-beliefs which precede such affect. Therefore, the left-brain processes must precede in time the right-brain processes. Also, they begin, at a fairly high level. At some point during the processing, the left brain processes are assumed to peak, presumably when the associative belief network (Mitchell 1983) has grown tense. The engagement of the left-brain is assumed to taper off as it completes its task of synthesizing the multiple beliefs and transfers the resulting overall affect to the right hemisphere. The right-brain process is assumed to peak at this point, although even at its peak, it is only moderate in magnitude.

We allow some engagement of the right-brain from the very beginning in the high-cognitive condition, and, likewise, some engagement of the left-brain from the very beginning in the high-affective condition, See Figure l. This is at least for two reasons. First, the modality of the stimuli (i.e., verbal or visual; print or TV) will influence the initial engagement of the left or right brain. Such engagement, necessary for initially attending to the stimulus, need not be at a high level, however. Nor need it be permanent or on-going. Beyond the initial, quite a low level of engagement, which hemisphere is engaged will depend upon the type of involvement condition rather than modality. Thus, a verbally presented message can be highly affect-laden and can engender strong hot affect. Likewise, a pictorially presented message may dwell entirely upon feature specific objective information, and thus engender brand-beliefs first and colt affect thereafter. The second reason is that it is not reasonable to entirely-separate the cognitive and affective effects of communication. Various hierarchy-of-effects models (Smith and Swinyard 1982, Park and Mittal 1985) prescribe a sequence of these effects only because they assume a "macro" time-frame. In contrast, discussions of brain-processes require a "micro" time frame; indeed, brain wave measures are taken continuously and ca be analyzed by time units of fractional seconds if desired. In this micro time-frame, the two effects (namely, the experience of affect, and the formation of a belief) must be posited to occur in an intertwined manner.



To illustrate, as one attends to 8 highly informative advertisement, for example, it seems inevitable that one will have formed some "impression" of the stimulus at the outset. Affect is an approach/avoidance tendency, and in order for someone to continue attending to the information in the 'ad', an "approach" disposition must have already been formed. This "approach" disposition (or "affect," that is) might be tentative, and one may hold in abeyance one's "final" attitudinal judgment until all the brand beliefs are formed and processed. But affect, low level as it is, is experienced none the less, and, correspondingly, the right-brain is engaged from the very beginning

For the same reason, in the high-affective involvement condition, (see Panel 2), the left-brain is assumed to be engaged (at modest levels) from the very beginning. Even when a communication begins with an affect-laden scene (e.g., a sexy model), the viewer may generate a cognition (e.g., "that model is tall," or "this must be an 'ad' for a brand of shampoo"). From as much as we know about the complex interrelationship between cognition and affect (see Zajonc and Marcus 1982), the safest statement we can make about the occurrence of cognition and affect is that on a micro time frame typically used for brain-wave measures, the two processes are intertwined. Therefore, both hemispheres are always engaged.

For the high-affective involvement condition (see Panel 2), the right-brain process may show peaks and troughs depending upon the affect experienced at a specific moment. Unlike the left-brain process in Panel l, however, the right-brain process in Panel 2 does not have to begin at a relatively lover level and rise progressively. This is because the left-brain processes information sequentially, whereas the right-brain attends to the stimulus holistically (Moscovitch 1979, Tucker 1981).

In the low-involvement condition, recall that we were unwilling to delineate as to whether cognitive processing or affective processing will occur. Accordingly, it is not possible to specify (see Panel 3 of Figure l) whether the right-brain or the left-brain processes will dominate Both the processes will be at low levels, so the question of which process is primary and which secondary may be indeterminate or even immaterial.



Suggested Operationalizations

Stimuli. The high-cognitive, high-affective and low-involvement processing situations can be relatively easily created by a manipulation used by Park and Young (1986). Alternatively, advertisements differing in their potential to engender involvement levels and type (cognitive versus affective) can be prepared and u led as stimuli delivered via both print and TV media.

Measures of Effects. Brain-wave measures using EEG procedures must be taken at multiple sites on each hemisphere. Suppression of alpha waves and (correspondingly but not identically) increase in beta waves have in the past research been used as indicators of brain-activation (see Olson and Ray, 1983). "Percentage of time each wave occurred," a measure used in pas: research, seems inappropriate or in any case inadequate because our framework requires dynamic plot of brain waves against the time axis. Amplitude of the alpha and beta waves (i.e., voltage) as used by Alwitt (1983) and Rust et al (1985) will therefore be more appropriate. In view of individual designs must be employed, and comparisons of EEG amplitudes across the two hemispheres be made only after the initial, base rate (i.e., activity levels during nontask, idle time) amplitudes have been partialled out within each hemisphere. Additionally, the use of new biomedical techniques that provide computer-simulated topographical displays of brain activation patterns (for description see Nevid 1984) may enable more direct observations of differential "primary" and "secondary" processes hypothesized in Figure 1.

The validity of EEG measures remains inconclusive (Stewart 1984). But theory testing and measure validation are interdependent (Cacioppo and Petty 1985), and both require a theoretical framework such as the one this paper furnishes.


Three categories of involvement (high-cognitive, high-affective, and low-cognitive/low-affective) were delineated, and they were in turn related to differential engagements of the hemispheres. Our reasoning let us to postulate that both hemispheres are always engaged, but that their engagement differs in being either primary or secondary. We grant that holistic processes occur in the right brain but we believe that when these occur with high intensity they represent high-affective involvement rather than low involvement conditions. So in our framework, the left brain is primarily engaged in the high-cognitive involvement condition whereas the right brain is primarily engaged in the high-affective involvement condition. In the low involvement condition, both the brains are engaged but at a low level. Moreover, either modality (print or TV, verbal or pictorial) can engender either type of involvement. Then, it is the type of involvement, not modality, which will influence the relative engagement of the hemispheres.

Our derivations are consistent with the few empirical findings that are available. Appel, Weinstein and Weinstein (1979) failed to support their hypotheses that (a) TV commercials will generate more right-brain- than left-brain activity; and (b) that higher recall commercials will generate left-brain dominance. Weinstein, Appel, and Weinstein (1980) also found inconclusive evidence for left-brain dominance for magazine advertisements and for right-brain dominance for TV advertisements. According to our framework, one must classify the content and effects (not modality) of the advertisement in order to specify in a theoretically correct way the relative engagement of the hemispheres.


Alwitt, L. F., (1983), "EEG Activity Reflects the Content of Commercials," in L. Percy and A. G. Woodside (eds.), Advertising and Consumer Psychology, Lexington: D. C. Heath, 239-252.

Antil, J. R., (1984), "Conceptualization and Operationalization of Involvement," in T.C. Kinnear (ed.), Advances ln Consumer Research, XI, 203-209.

Appel, V., S. Weinstein, and C. Weinstein (1979), "Letters to the Editor," Journal of Advertising Research, 19, 70.

Berlyne, D. E. (1978), "Chapter 3: 'Arousal' a posthumously published chapter from Berlyne's unfinished book entitled Curiosity and Emotion'," Journal of Motivation and Emotion, 2, 146-175.

Cacioppo, J. T. and R. E. Petty, (1985), "Psychological Responses and Advertising Effects," Psychology and Marketing, 5, 115-126.

Cohen, J. B. (1983), "Involvement and You: 1,000 Great Ideas," in R. Bagozzi and A. Tybout (eds.), Advances in Consumer Research, IX, 325-328.

Hansen, F. (1981), "Hemispheral Lateralization: Implication for Understanding Consumer Behavior," Journal of Consumer Research, 8, 23-36.

Hirschman, E. C. and M. B. Holbrook (1982), "Hedonic Consumption: Emerging Concepts, Methods and Propositions," Journal of Marketing, 46 (Summer), 92-101.

Holbrook, M. B. and E. C. Hirschman (1982), "The Experiential Aspects of Consumption: Consumer Fantasies, Feelings, and Fun," Journal of Consumer Research, 132-40.

Katz, W. (1980), "The Split Brain Syndrome," Journal of Professional Marketing Research Society, 1, 14-18.

Krugman, B. E. (1977), "Low Involvement Theory in Light of Brain Research," in J. C. Maloney and B. Silverman (eds.), Attitude Research Plays for High Stakes, Chicago: AMA.

McGuire, R. J. (1974), "Psychological Motives and Communication Gratification," in J. G. Blumler and E. Katz (eds.), The Uses of Mass Communications: Current Perspectives and Gratifications Research, Sage.

Mitchell, A. A. (1981), "The Dimensions of Advertising Involvement," in t. B. Monroe (et.), Advances in Consumer Research, VII, 25-30.

Mittal, B. (1982), "Understanding the Bases and Effects of Involvement in the Consumer Choice Process," Ph.D. Dissertation. University of Pittsburgh.

Mittal, B. (1983), "Consumer's Cognitive Journey Through the Product Forest," in Bagozzi and Tybout (eds.), Advances in Consumer Research. X, 464-469.

Moscovitch, M. (1979), "Information Processing and the Cerebral Hemispheres," in M. S. Gazzaniga (et.), Handbook of Behavioral Neurobiology, 2 (Nev York: Plenum).

Muncy, J. A. and S. D. Runt (1984), "Consumer Involvement: Definitional Issues and Research Directions," in T. C. Kinnear (ed.). Advances in Consumer Research, VI, 193-196.

Nevid, J. S. (1984), "Methodological Considerations in the Use of Electroencephalographic Techniques in Advertising Research," Psychology and Marketing, 1, 5-19.

Olson, J. C. and W. J. Ray (1983), "Using Brain-Wave Measures to Assess Advertising Effects," MSI Report No. 83-108, Cambridge, MA: Marketing Science Institute.

Park, C. W. and B. Mittal (1985), "A Theory of Involvement in Consumer Behavior", in J. N. Sheth (ed.), Research in Consumer Behavior. Greenwich: JAI Press, 201-232.

Park, C. W. and S. M. Young (1986), "Consumer Response to TV Commercials: The Impact of Involvement and Background Music on Brand Attitude Formation," Journal of Marketing Research. (February), 11-24.

Rothschild, M. L. (1979), "Advertising Strategies for High and Low Involvement Situations," in J. C. Maloney and B. Silverman (eds.), Attitude Research Plays for High Stakes, Chicago: American Marketing Association, 74-93.

Rothschild, M. L. and E. Thorson (1983), "Electroencephalographic Activity as a Response to Complex Stimuli," in L. Percy and A. G. Woodside (eds.), Advertising and Consumer Psychology, Lexington: D. C. Heath, 239-252.

Rust, T. R., L. L. Price and V. Kumar (1985), "EEG Response to Advertisements in Print and Broadcast Media," Report No. 85-111, Marketing Science Institute.

Sherif, M. and C. W. Hovland (1961), Social Judgment, Nev Haven: Yale University Press.

Smith, R. E. and W. R. Swinyard (1982), "Information Response Models: An Integrated Approach," Journal of Marketing, 46, 81-93.

Stephens, D. (1985), "Hemispheric Function and Involvement," in E. C. Hirschman and M. B. Holbrook (eds.), Advances in Consumer Research, XII, 285-289.

Stewart, D. W. (1984), "Physiological Measurement of Advertising Effects," Psychology and Marketing, 1, 43-48.

Tucker, D. (1981), "Lateral Brain Function, Emotion, and Conceptualization," Psychological Bulletin, 89, 19-46.

Weinstein, S., V. Appel, and C. Weinstein (1980), "Brain-Activity Responses to Magazine and TV Advertising, Journal of Advertising Research, (June), 57-63.

Wilkie, W. L. and E. A. Pessemier (1973), "Issues in Marketing's Use of Multi-Attribute Models," Journal of Marketing Research, 10, 428-41.

Zajonc, R. B. (1980), "Feeling and Thinking: Preferences Need No Inferences " American Psychologist, 35, 151-75.

Zajonc, R. B. and R. Marcus (1982), "Affective and Cognitive Factors in Preferences," Journal of Consumer Research, 9 (September), 123-131.