Hemispheric Function and Involvement

Debra Stephens, The University of Chicago
ABSTRACT - A recent hypothesis relates high and low involvement to information processing in the left and right cerebral hemispheres, respectively. This paper examines current research findings in the areas of involvement and hemispheric function, in an attempt to determine whether that hypothesis has merit.
[ to cite ]:
Debra Stephens (1985) ,"Hemispheric Function and Involvement", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 285-289.

Advances in Consumer Research Volume 12, 1985      Pages 285-289


Debra Stephens, The University of Chicago


A recent hypothesis relates high and low involvement to information processing in the left and right cerebral hemispheres, respectively. This paper examines current research findings in the areas of involvement and hemispheric function, in an attempt to determine whether that hypothesis has merit.


The way in which degree of involvement with an issue affects attitude has been a topic of concern to social psychologists for a number of years (Chaiken 1980; Sherif and Cantril 1947; Sherif, Sherif, and Nebergall 1965). Experimentally, an individual's involvement is classified as high or low depending on whether an issue has personal relevance for him (Apsler and Sears 1968; Petty and Cacioppo 1979 Rhine and Severance 1970). Consumer behavior researchers (e.g., Gardner, Mitchell, and Russo, forthcoming) as well as social psychologists (Petty, Cacioppo, and Heesacker 1981; Petty, Cacioppo, and Schumann 1983) have found that level of involvement helps determine how persuasive a message is, and aspects of it have greater impact on attitude.

It has been suggested that the left half of the cerebral cortex guides information processing under conditions of high involvement, while the right hemisphere predominates when involvement is low (Hansen 1981 Krugman 197?). The primary purpose of this paper is to evaluate and reformulate that hypothesis in light of what is known in the respective areas of involvement and hemispheric function. Let us take a brief look at each field separately, so that we may determine how the hypothesis might be clarified.


Investigators of involvement have operationalized the term in a number of ways. Krugman (1967) suggested that television induces a passive, low involvement state while print evokes higher involvement. It is interesting that the opposite has also been posited (Worchel, Andreoli, and Eason 1975). Others have placed individuals in high or low involvement groups on the basis of the personal importance of an issue or product category (Hovland, Harvey, and Sherif 1957; Newman and Dolich 1979). Finally, some researchers have classified certain products and issues as high- or low-involvement (Lastovicka and Gardner 1979; Rhine and Severance 1970).

As Russo (personal communication) points out, in all of these cases involvement is conceptualized in terms of its possible antecedents. Doubtless certain product categories and methods of message presentation tend to evoke higher involvement than do others. However, one cannot assume that an individual's internal state will necessarily be altered by a given antecedent. Moreover, in each of the above cases, presumed involvement level is confounded with variations in characteristics of the medium, subject, or product.

An accurate picture of information processing under different levels of involvement is more likely to emerge when all factors (product, medium, subject type) except level of involvement are kept constant (see Petty, Cacioppo, and Schumann 1983 for a discussion of these issues). Several investigators have taken this approach, and have experimentally induced states of high and low involvement in subjects (Gardner et al., forthcoming; Park and Young 1984; Petty et al. 1983). For these studies the question of external validity arises (Cook, personal communication). Probably the proper procedure is to develop a theory of involvement in a well-controlled laboratory environment, and then to test and refine it in field settings. Let us examine the attempts at theory development.

Petty, Cacioppo, and Schumann (1983) informed both high and low involvement groups that they would be required to evaluate a series of advertisements. The high involvement subjects were told that the product in the experimental at would soon be available in their area, and that following the experiment they would be allowed to select a free gift from the experimental product category. The low involvement individuals were informed that the product in the experimental ad would not soon be available in their local area, and that they could choose a free gift from another product category.

Four experimental advertisements were composed for Edge razors. Each contained either strong or weak arguments, and a photograph of a famous or non-famous endorser. Each version was viewed by a high involvement group and a low involvement one. The experimental advertisement was Presented as one in a series of advertisements.

Subjects rated their overall impression of the product on the scales good-bad, satisfactory-unsatisfactory, and favorable-unfavorable. To check that the manipulations of argument quality and endorser identity were effective, several additional questions were asked.

Analyses of the attitudinal measures revealed that involvement level interacted significantly with endorser identity as well as with argument quality. Endorser identity had a greater impact on the low involvement group than on the high involvement one. Argument quality, on the other hand, had more influence on attitudes of the latter group than on those of the former.

To account for these results, Petty et al. present the notions of central versus peripheral routes to persuasion. When an individual is highly involved in deciding about an issue or product, he processes the contents of a communication which are directly relevant to a thoughtful evaluation of the issue or product. As Petty et al. point out, the most important product-related information might be an attractive endorser. The defining characteristic of central processing is thus not whether the information attended to is in pictorial or verbal form. It is the effortful evaluation of any message contents that are directly related to product attributes or aspects of an issue.

The peripheral route, utilized under low involvement conditions, is distinguished by a less effortful, more associationistic style of processing information. Positive or negative cues associated with the object under evaluation are fed into a simple decision rule. In this study a pictorial cue, endorser type, was employed. Petty and Cacioppo (1984) found that a decision rule may be activated by number of message arguments (the more the better). Chaiken (1980) observed a peripheral style of processing in response to a verbal description of a message source as likeable or dislikeable. Thus either the central or the peripheral route to persuasion may involve a consideration of verbal and/or pictorial information, and source as well as message cues.

Gardner and her colleagues view involvement as the internal motivational state of an individual at a given moment. Intensity of involvement affects attention level, and the individual's goal in processing information will determine the type of strategy he employs to do so. With respect to the effects of advertisements, the authors note that two kinds of strategies are relevant: a brand processing and a nonbrand one. The authors postulate that a brand processing strategy will be used by individuals with a recognized interest in the product. During exposure to the advertise it they retrieve the knowledge about the brand or product class, and are likely to integrate new information from the advertisement with their pre-existing knowledge. In the course of this integration process, arguments in favor of or against the brand are frequently generated.

People utilizing a nonbrand strategy have another set of goals in viewing the advertisement, such as enjoyment, censorship, or stylistic evaluation. They do not fully retrieve their knowledge of the brand or product category, and so make few inferences about product characteristics and performance. Therefore counterarguments and support arguments are not as probable.

When attention level is high, Gardner et al. define high involvement in terms of the use of a brand processing strategy, and low involvement in terms of a nonbrand strategy. (The authors observe that a low attention level always leads to low involvement.) In this investigation, attention was held at a high level, and subjects were instructed in the use of a brand or nonbrand processing strategy.

As compared to the nonbrand processors, the high involvement group displayed superior memory for brand information, and formed less favorable attitudes toward the brands advertised.

The third investigation to be discussed here is interesting because the authors distinguish two different types of high involvement (Park and Young 1984). Like Gardner et al., they assert that the individual's goal in attending to an advertisement is one major determinant of how the information content will be processed. The two kinds of high involvement which Park and Young distinguish are cognitive and affective. Cognitive involvement occurs when the individual's goal is to learn about the functional performance of a brand. Affective involvement results from the association of emotional or aesthetic components of an advertisement with one's desire to express a specific type of self-image to the world. Park and Young define low involvement in terms of lowered arousal toward an advertisement, either because of distractors or a lack of interest.

The authors' description of attitude formation in a cognitive involvement situation resembles the explanation of the central route to persuasion presented by Petty et al., as well as the outline of a brand processing strategy in the Gardner et al. study. Park and Young describe attitude formation under conditions of affective involvement as follows. When watching an advertisement, the individual identifies an image that he matches to an actual or desired image of himself. Attitude toward the brand advertised derives from the affect which is stored with the image of self. The image-matching is accomplished through an analogical process in which stimulus and internal representation are compared in a global fashion. Information processing in an affective involvement situation is less effortful than under cognitive involvement conditions, because it is more automated.

The investigators characterize attitude formation in a Low involvement situation as "ephemeral in nature, weak in intensity, and transient in stability, depending on the frequency of exposure" (p. 10). Visual aspects of a commercial should have a greater influence on attitude than should verba' aspects, because the former are easier to process. Park and Young assert that the peripheral route of Petty et al. corresponds to this situation. The nonbrand processing strategy described by Gardner et al. does not appear to have a clear analog in the Petty et al. classification scheme or in the Park and Young taxonomy. Likewise, the affective involvement condition of Park and Young does not appear to correspond to any classification level of the other two groups of investigators.

In this study, as in the other two, involvement was varied with instruction, and the stimulus was an advertisement (commercial). The investigation revealed that brand attitude varies not only as a function of involvement type, but also with the presence or absence of music, and with attitude toward the advertisement. Specifically, attitude toward the advertisement was not significantly associated with brand attitude in the cognitive involvement condition, but was an important determinant of brand attitude in the affective and low involvement situations. The presence of music in the commercial appeared to reduce the amount of brand information processing by the cognitively involved subjects, and resulted in less favorable brand attitudes. The authors suggest that this result is explicable by the assumption that music interferes with the generation of support arguments. Music had a positive effect on the brand attitudes of the low involvement group. Its effect on the affectively involved subjects was not clear.

In sum, recent investigations of involvement indicate that level or type of involvement helps determine the extent to which brand-related information will be processed and affect brand attitude. Moreover, in producing an effect on attitude, the involvement variable apparently interacts with several other mediators of brand attitude formation, such as music, attitude toward the advertisement, and endorser type. It is not clear that one taxonomy of involvement is superior to the other two. A combination which distinguishes processing strategy from attentional level, and cognitive from affective brand processing might be most productive at this time.


The most compelling evidence that the two cerebral hemispheres subserve different functions has come from studies of so-called "split-brain" patients. Split-brain surgery, a technique for controlling epileptic seizures, involves the cutting of some nerve fibers which connect the two hemispheres. People who have undergone this surgery have virtually intact but independent hemispheres. The two hemispheres may be studied separately in the same person, by presentation of information to one hemisphere alone, via connections from the opposite visual field, ear, or hand.

Roger Sperry and his colleagues found that the two halves of the brain showed differences in ability as well as style, in the performance of a variety of tasks. The left hemisphere is the center for speech, writing, and calculation. The right hemisphere cannot respond in speech or writing, and can only calculate very simple additions. It excels at functions which Sperry (1974) characterizes as "non-linguistic, non-mathematical . . . holistic and unitary rather than analytic and fragmentary . . . involving concrete perceptual insight rather than abstract sequential reasoning" (?. 11). For example, the left hand (right hemisphere) of split-brain patients is superior to the right hand (left hemisphere) at constructing block designs, and at copying and drawing two-dimensional representations of three-dimensional objects such as Necker cubes and houses. The right hemisphere is also superior at detecting direction of motion and discriminating the orientation of lines and two-dimensional objects, and has better binocular depth perception than the left (see Springer and Deutsch 1981 for a review).

Studies of normal adults reveal that virtually all right-handers display the pattern of left and right hemisphere asymmetries outlined above. In the intact brain, speech dominance can only be directly determined by observing whether an individual can speak when each hemisphere in turn is inactivated by a barbiturate such as sodium amytal (Wada and Rasmussen 1960). However, several non-intrusive tests have been developed, from which specialization for language and other tasks may be inferred. In these tests, stimuli are usually presented either tachistoscoPicallY or dichotically.

In a tachistoscopic task, the subject must respond to each of a series of stimulus items briefly presented to the right or left visual field, or to both. Accuracy and average response time is calculated for each visual field, and the hemisphere opposite the one which receives a better score is considered to be dominant for the task. Right-handers show a right visual field superiority, indicating left hemisphere dominance, on word and syllable identification (e.g., MacKavey, Curcio, and Rosen 1975; Levy and Reid 1978), and left visual field (right hemisphere) dominance on visuo-spatial tasks such as dot location (Levy and Reid 1978), face recognition (Leehey and Cahn 1979), and recognition of braille letters (Hermelin and O'Connor 1971).

On dichotic tasks, two auditory stimuli are presented simultaneously, one to each ear, and the subject is to report what he or she heard in one or both ears. Most right-handers display a right ear (left hemisphere) advantage for identifying words and syllables (Kimura 1967; Shankweiler and Studdert-Kennedy 1967), and a left ear (right hemisphere) superiority for recognizing melodic contours (Kimura 1964) and nonverbal emotional sounds such as crying and laughing (King and Kimura 1972).

A right hemisphere advantage has also been found for identifying emotions conveyed by intonation (Safer and Leventhal 1977). Moreover, studies of brain-damaged populations indicate that an intact right hemisphere is essential for the accurate interpretation of emotional aspects of stimuli, such as the emotional content of stories (Wechsler 1973), intonational cues (Heilman, Scholes, and Watson 1975), and facial expression (Cicone, Wapner, and Gardner 1979).

Superimposed on these patterns of functional specialization are stable, task-independent differences in hemispheric arousal. Individuals who show higher left than right hemisphere activation, as indexed by EEG and cerebral blood flow, perform better on verbal tasks and worse on spatial ones than do persons displaying higher right hemisphere arousal (e.g., Gur and Reivich 1980). Because asymmetrical_y greater arousal of ore hemisphere leads to motoric and attentional orientations toward the side of space opposite the more highly aroused hemisphere, it is possible to infer which hemisphere is more aroused by administering various tachistoscopic tests. In such tests, stimuli which both hemispheres are equally skilled at processing are flashed to one or both visual fields. The hemisphere opposite the field for which performance is better is inferred to be the more highly aroused (Levy, Heller, Banich, and Burton 1983a). In addition, Jerre Levy and her colleagues at the University of Chicago have developed a free vision test which is sensitive to individual differences in arousal asymmetry (Levy, Heller, Banich, and Burton 1983b).

Evidence indicates that differences in personality characteristics are associated with variations in the direction of arousal asymmetry. For example, Dawson, Tucker, and Swenson (1981) observed that individuals who show a predominance of leftward lateral eye movements (higher right hemisphere arousal) tend to be extroverted and give positive self-descriptions. Right-movers are inclined toward introversion and negative self-descriptions.


At first glance, the sequential, analytic processing style of the left hemisphere appears well suited for handling information under conditions of cognitive involvement. The holistic strategy employed by the right hemisphere, on the other hand, seems ideal for processing information in a situation of affective involvement. Unfortunately, a second and longer look reveals a very important unresolved issue.

The processing of information in an advertisement is clearly a much more complicated task than are the laterality tasks commonly administered to normals. The relative simplicity of the latter is of course unavoidable under conditions of tachistoscopic presentation. This is because a guarantee of initial reception by one hemisphere necessitates brevity of stimulus presentation, or severe competition between simultaneously presented stimuli.

Studies of brain-damaged populations indicate that optimal performance on complex linguistic as well as pictorial tasks is impossible without two intact hemispheres (Huber and Gleber 1982; Wapner, Hamby, and Gardner 1981; Winter and Gardner 1977). However, right- and left-hemisphere damaged groups do exhibit different sorts of deficits on such tasks.

Huber and Gleber presented scrambled stories, some in pictorial and others in sentence form, to left- and right-damaged patients, as well as to normals. All brain-injured patients made more errors than did the normals. ,lost of the left-hemisphere damaged subjects scored higher on the pictorially presented stories than on those in sentence form. The right-damaged group obtained fewer errors on the verbal than on the pictorial presentations.

Wapner, Hamby, and Gardner required neurologically impaired patients to recount stories presented auditorily or on film, and to answer various sorts of questions. Right-hemisphere injured subjects used correct phonology and syntax, and accurately recalled basic facts and many details. However, their stories often contained embellishments and confabulations. They also had difficulty explaining the moral of a story, and could only describe characters' motivations if these were straightforward. Left-hemisphere damaged patients were hardly eloquent, but showed none of the deficits exhibited by the right-damaged subjects.

Advertisements (and everyday communications in general) frequently contain figurative or metaphorical use of language to convey abstract concepts in concrete form. The study by Winter and Gardner (1977) strongly suggests that the right and left hemispheres play different but equally essential roles in the use of metaphor. Two tests of metaphor comprehension--a pictorial and a verbal one-- were administered to right- and left-damaged patients. The right-hemisphere injured subjects were able to explain the meaning of a given metaphor, but could not select a picture demonstrating the standard interpretation. Left-damaged patients could perform the pictorial task, but were not able to explain the meaning.

What these investigators suggest is that the two hemispheres play different but equally crucial roles in certain complex tasks. Where does this leave us with respect to involvement? Clearly, we cannot fall prey to the tempting analogy between differences in hemispheric processing style and variations in the handling of information under the two (or more) involvement levels. However, the strong associations between certain personality characteristics and side of greater arousal to indicate that the two hemispheres may indeed play differentially important roles in some higher order mental activities.

For this reason, it is suggested that the hypothesis be reformulated to take into account interactions between involvement type and stimulus properties, and then tested properly. A productive reformulation would be stated something like the following: if an advertisement contains only a straightforward verbal enumeration of functional characteristics of a brand, and if an individual is cognitively involved, then the left hemisphere will predominate in processing. If, on the other hand, the advertisement contains no verbal elements, but is instead pictorial and attempts to match the brand to some cultural metaphor, then the right hemisphere will be the more active Processor.

Finally, proper testing with normals is likely to require meticulous and expensive physiological measures, of cerebral blood flow for example. It is crucial that the population be administered standard laterality tests, and that hemispheric arousal be assessed as well. Without such rigorous testing, the debate about the importance of brain function in involvement will remain unresolved.


Apsler, Robert and David O. Sears (1968), "Warning, Personal Involvement, and Attitude Change," Journal of Personality and Social Psychology, 9 (June), 162-166.

Chaiken, Shelly (1980), "Heuristic Versus Systematic Information Processing and the Use of Source Versus Message Cues in Persuasion," Journal of Personality and Social Psychology, 39 (November), 752-766.

Cicone, Michael, Wendy Wapner, Nancy Foldi, Etgar Zurif, and Howard Gardner (1979), "The Relation between Gesture and Language in Aphasic Communication," Brain and Language, 8, 324-349.

Cook, Victor J., Jr. (1984), Personal communication, New Orleans, Louisiana.

Gardner, Meryl P., Andrew A. Mitchell, and J. Edward Russo (forthcoming), "Strategy-Induced Low Involvement With Advertising," Journal of Advertising Research.

Gur, Ruben C. and Martin Reivich (1980), "Cognitive Task Effects on Hemispheric Blood Flow in Humans: Evidence for Individual Differences in Hemispheric Activation," Brain and Language, 9, 78-92.

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

Heilman, Kenneth, R. Scholes, and R. Watson (1975), "Auditory Affective Agnosia: Disturbed Comprehension of Affective Speech," Journal of Neurology, Neurosurgery, and Psychiatry, 38, 69-72.

Hermelin, B. and N. O'Connor (1971), "Functional Asymmetry in Reading of Braille," NeuroPsychologia, 9, 431-435.

Hovland, Carl I., O. J. Harvey, and Muzifer Sherif (1957), "Assimilation and Contrast Effects in Reactions to Communication and Attitude Change," Journal of Abnormal and Social Psychology, 55 (September), 244-252.

Huber, Walter and Jochen Gleber (1982), "Linguistic and Nonlinguistic Processing of Narratives in Aphasia," Brain and Language, 16, 1-18.

Kimura, Doreen (1964), "Left-Right Differences in the Perception of Melodies," Quarterly Journal of Experimental Psychology, 16, 355-358.

Kimura, Doreen (1967), "Functional Asymmetry of the Brain in Dichotic Listening," Cortex, 3, 163-178.

King, F. and Doreen Kimura (1972), "Left-Ear Superiority in Dichotic Perception of Vocal Nonverbal Sounds," Canadian Journal of Psychology, 26, 111-116.

Krugman, Herbert E. (1967), "The Measurement of Advertising Involvement," Public Opinion Quarterly, 30 (Winter), 583-396.

Krugman, Herbert E. (1979), "Low Involvement Theory in the Light of New Brain Research," in Attitude Research Plays for High Stakes, eds. John C. Maloney and Bernard Silverman. Chicago: American Marketing Association, 16-22.

Lastovicka, John L. and David M. Gardner (1979), "Components of Involvement," in Attitude Research Plays for High Stakes, eds. John C. Maloney and Bernard Silverman. Chicago: American Marketing Association, 53-73.

Leehey, Susan C. and Andrew Cahn (1979), "Late-al Asymmetries in the Recognition of Words, Familiar Faces and Unfamiliar Faces," Neuropsychologia, 19, 619-629.

Levy, Jerre, Wendy Heller, Marie Banich, and Leslie Burton (1983a), "Are Variations Among Right-Handed Individuals in Perceptual Asymmetries Caused by Characteristic Arousal Differences between Hemispheres?" Journal of Experimental Psychology, 9, 329-359.

Levy, Jerre, Wendy Heller, Marie Banich, and Leslie Burton (1983b), "Asymmetry of Perception in Free Viewing of Chimeric Faces," Brain and Cognition, 2, 404-419.

Levy, Jerre and Mary Lou Reid (1978), "Variations in Cerebral Asymmetries as a Function of Handedness, Hand Posture in Writing, and Sex," Journal of Experimental Psychology: General, 107, 119-144.

MacKavey, W., F. Curcio, and J. Rosen (1975), "Tachistoscopic Word Recognition Performance under Conditions of Simultaneous Bilateral Presentation," Neuropsychologia, 13, 27-33.

Newman, Larry M. and Ira J. Dolich (1979), "An Examination of Ego-Involvement as a Modifier of Attitude Changes Caused from Product Testing," Advances in Consumer Research, Vol. 6, ed. William L. Wilkie. Ann Arbor, MI: University of Michigan, 41-57.

Park, C. Whan and S. Mark Young (1984), "The Effects of Involvement and Executional Factors of a Television Commercial on Brand Attitude Formation," working paper no. 84-100, Cambridge, MA: Marketing Science Institute.

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

Petty, Richard E. and John T. Cacioppo (1984), "The Effects of Involvement on Responses to Argument Quantity and Quality: Central and Peripheral Routes to Persuasion," Journal of Personality and Social Psychology, 46 (January), 69-81.

Petty, Richard E., John T. Cacioppo, and Martin Heesacker (1981), "The Use of Rhetorical Questions in Persuasion: A Cognitive Response Analysis," Journal of Personality and Social Psychology, 40 (March), 432-440.

Petty, Richard E., John T. Cacioppo, and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.

Rhine, Ramon and Laurence J. Severance (1970), "Ego-Involvement, Discrepancy, Source Credibility, and Attitude Change," Journal of Personality and Social Psychology, 16 (October), 175-190.

Russo, J. Edward (1984), Personal communication, Chicago, Illinois.

Safer, Martin and Howard Leventhal (1977), "Ear Differences in Evaluating Emotional Tone of Voice and Verbal Content, " Journal of Experimental Psychology: Human Perception and Performance, 3, 75-82.

Shankweiler, Donald and Michael Studdert-Kennedy (1967), "Identification of Consonants and Vowels Presented to Left and Right Ears," Quarterly Journal of Experimental Psychology, 19, 59-63.

Sherif, Muzifer and M. Cantril (1947), The Psychology of Ego Involvement, New York: John Wiley.

Sherif, Carolyn W., Muzifer Sherif, and Roger E. Nebergall (1965), Attitude and Attitude Change, Philadelphia: Saunders.

Sperry, Roger (1974), "Lateral Specialization in the Surgically Separated Hemispheres," in The Neurosciences Third Study Program, eds. F. Schmitt and F. Worden. Cambridge, M.A: M.I.T. Press, 5-19.

Springer, Sally P. and George Deutsch (1981), Left Brain, Right Brain. San Francisco: W. H. Freeman and Company.

Wada, John K. and T. Rasmussen (1960), "Intracarotid Injection of Sodium Amytal for the Lateralization of Cerebral Speech Dominance: Experimental and Clinical Observations," Journal of Neurosurgery, 17, 266-282

Wapner, Wendy, Suzanne Hamby, and Howard Gardner (1981), "The Role of the Right Hemisphere in the Apprehension of Complex Linguistic Materials," Brain and Language, 14, 15-33.

Wechsler, A. (1973), "The Effect of Organic Brain Disease on Recall of Emotionally Charged Versus Neutral Narrative Text," Neurology, 23, 130-135.

Worchel, Stephen, Virginia Andreoli, and Joe Eason (1975), "Is the Medium the Message? A Study of the Effects of Media, Communicator, and Message Characteristics on Attitude Change," Journal of Applied Social Psychology, S (April-June), 157-172.