Consumers' Interpretations of Advertising Imagery: a Visit to the Hell of Connotation

CONSUMERS' INTERPRETATIONS OF ADVERTISING IMAGERY: A VISIT TO THE HELL OF CONNOTATION

David Glen Mick, University of Florida

Laura G. Politi, University of Florida

[The authors are indebted to the following individuals who provided valuable comments on an earlier draft of this article: Joseph Alba, Morris Holbrook, Richard Lutz, Dean McCannell, Grant McCracken, Winfried Noth, Srinivasan Rathneshwer, and Richard Zakia.]

ABSTRACT -

The semanticity of advertising imagery is investigated through the use of a protocol interpretive method. It is shown that, beyond contributions to theory building, interpretive methods are also valuable for testing propositions. In addition, a close inspection of the protocols reveals several dimensions of advertising comprehension formerly ignored. We conclude that the concept of denotation in advertising illustrations is misguided and that an alternative meaning model of advertising consumption is needed to more fully appreciate the complexities and nuances of consumers' interpretations of advertising imagery.

INTRODUCTION

Consumers' interpretations of advertising messages have long been an important and controversial topic in advertising research (Jacoby and Hoyer 1982a, 1982b, 1987; Mick 1988b; Russo, Metcalf, and Stephens 1981; Shimp and Preston 1981). [The terms comprehension and interpretation are used interchangeably here. Readers who question the synonymy of those terms are also likely to believe that denotation and connotation are separate constructs- -likely because they associate comprehension with denotation and interpretation with connotation. This article seeks to defy the denotation/connotation distinction with respect to advertising illustrations. Comprehension itself is conceptualized generally in this article from a levels of processing perspective, specifically as a cue-induced spread of activation of semantic concepts (knowledge structures). Inferencing is the basic mechanism of this activation and also accounts for the meanings constructed as a function of bridging two already - activated concepts.] Historically, most empirical work has centered on consumers' processing of linguistic information; recently, researchers have payed increased attention to nonlinguistic features (e.g., Childers and Houston 1984; Edell and Staelin 1983; Johnson, Zimmer, and Golden 1987; Richards and Zakia 1981; Rossiter and Percy 1980; Zakia 1986). Lengthy theoretical writings on visual communication in advertising have also appeared (Rossiter and Percy 1983). Despite this trend, few have actually studied the semiotic substance of consumers' interpretations of advertising illustrations. For instance, Thematic Apperception Tests remain widespread in the advertising industry for Pretesting visual content. Yet, advertising scholars have exerted little effort to employ such semantic -unveiling methods to inform theory and assess propositions about consumers' interpretations of nonverbal ad information.

In several linguistic domains, there has been a surge of interest in protocol (think -out-loud) techniques: textlinguistics (Ballstaedt and Mandl 1984); reading research (Graesser and Clark 1985); and literary analysis (Kintgen 1983). These researchers eschew the traditional, circumscribed, post hoc measures of comprehension, e.g., recall and recognition tests. Instead, they favor concurrent protocols for the fertile insights provided about the dynamic, iterative nature of meaning processes, including spontaneous inferences, tentative hypotheses, and reconstructed meanings. According to Olson, Duffy, and Mack (1984), p. 257), protocol data can reveal processing strategies used, knowledge structures employed, and types of representations (meanings) constructed. Only a handful of advertising researchers have utilized open-ended, on-line measures of interpretive processes, and they have concentrated strictly on verbal information (e.g., Levy 1986; Mick 1987). In a recent expose' on psychological meaning in advertising, Friedmann and Zimmer (1988) called for wider usage of these methods. [Although there is a long tradition of collecting cognitive responses in advertising studies, the coding schemes have remained wedded to attitude theory in social psychology (support arguments, counterarguments, etc.). The semantic content of the responses are treated superficially, at least in terms of what they can reveal about comprehension. Two recent exceptions to this trend include Chattopadhyay and Alba (1988) and Mick (1987).] Data produced by unstructured techniques can be analyzed both quantitatively and qualitatively, the latter emphasis being closer in spirit to the interpretive methods and raison detre of this volume (see also Belk, Sherry, and Wallendorf 1988; Hirschman 1987; Holbrook and Grayson 1986; Hudson and Ozanne 1988; Mick 1988a). Most researchers, regardless of their methodological predilictions, are willing to accept that interpretive methods can serve as a logic of discovery, i.e., to develop theory and empirically testable propositions (see Hunt 1983). However, there is far less agreement on whether interpretive methods can and should be used as a logic of justification, i.e., to support or falsify theory, including the testing of propositions as commonly done with positivist techniques such as surveys and experimentation (see Calder and Tybout 1987; Hudson and Ozanne 1988). [It must be noted that some researchers who are sympathetic to interpretive methods reject altogether the distinction between a logic of discovery and a logic of justification.]

The purpose of our study is twofold. First, we adopt a protocol technique to capture consumers' interpretations of a pictorial print ad. Through our interpretation of the protocol data, we seek to contribute to both theory and method aimed at understanding consumers' interpretations of visual advertising messages. Second, as a preliminary demonstration work, we show how these interpretive data can be used to test certain propositions about advertising.

ADVERTISING ILLUSTRATIONS: OPEN OR CLOSED TEXTS?

The classic model of communication employed by advertisers involves a sender who encodes a message which is then transmitted through a specific medium to a receiver who, in turn, decodes the message. Ideally, the receiver is a member of a targeted market segment whose linguistic and nonlinguistic codes have been sufficiently researched, the result being a code-sharing between sender and receiver that maximizes efficient, effective communication. A major goal is to have targeted consumers interpret brand messages as the advertiser intends (denotates). When this occurs, comprehension of the message is said to have taken place. According to a number of longstanding advertising theories (e.g., McGuire 1969), comprehension is a necessary step on the route to a positive brand attitude and purchase behavior.

In the terminology of semiotics, it seems clear that advertisers construct their messages to be "open" texts, i.e., understandable as intended by a select group of individuals. According to Eco (1979, p.3),

An 'open' text cannot be described as a communicative strategy if the role of its addressee (the reader, in the case of verbal texts) has not been envisaged at the moment of its generation qua text. An open text is a paramount instance of a syntactic-semantico-pragmatic device whose foreseen interpretation is a part of its generative process.

Related specifically to advertising illustrations, the proposition that ad messages are open texts in turn reflects Bryson's (1983) notion of the essentialness of denotation in painting, i.e., interpretation of visual content is guided by iconographic codes which lead the presupposed observer to its intended, fixed meaning. Hence, a carefully crafted advertising photograph should communicate the same message to individuals from a relevant market segment.

In his incisive semiotic article, however, Baker (1985) charges that the concept of denotation is worthless in the analysis of visual imagery, because it is impossible. He argues that there is rarely a shared, descriminable code but rather "a sloppy, blurry thing, firmest at the center where its stereotypes are most stupid and its prejudices most entrenched, but always slipping into abjection at its edges" (Baker 1985, p. 173; see also Holbrook 1983, pp.68-69). According to Baker's position, then, advertising pictorials are more likely to be "closed" texts, as semioticians; would call them. As Eco (1979, p.8) explains,

Those texts that obsessively aim at arousing a precise response on the part of more or less precise empirical readers .... are in fact open to any possible 'aberrant' decoding. A text which is so immoderately 'open' to every possible interpretation [is] a closed one.

Thus, a closed text attempts to overdetermine receiver response and always fails.

It must be stressed that the open/closed text distinction represents poles on a continuum, not an all-or-none dichotomy. Nevertheless, Baker's (1985) discussion suggests that, irrespective of apparent success or effectiveness, an advertising illustration is inherently more like a closed text than advertising theory and practice assumes. From that proposition Baker goes on to pose some troubling thoughts for advertisers, quoting the French serniologists Baudrillard (1981, p. 196) and Barthes (1964, p. 43):

If the possibility of denoting is removed, what is left? Is it, as Jean Baudrillard suggests the Bauhaus designers feared, that beyond the desperate illusion of a nucleus of intended meanings or functions, 'all the rest is coating, the hell of connotation: residue, superfluity, excrescence, eccentricity, ornamentation, uselessness'? .... What Barthes called 'the terror of uncertain signs' is far more likely to be felt in cases where direct and unambiguous communication is attempted, where it is thought necessary to get a specific message from sender to receiver.

In a recent empirical Work, Zakia (1986) partly addressed the advertising implications surrounding Baker's (1985) claims. Zakia had a small sample of consumers respond rapidly with a single word upon exposure to a highly visual liquor ad. The reactions were multifarious, including "absorbing," "cold," "embraceable," "evil," "primitive," "seductive," and .'vulnerable." Zakia (1986) concluded that without multiple potential meanings, an extensive and diverse audience could not be attracted to the product. In other words, from a managerial perspective, a closed text in pictorial advertising may be superior to an open text. Like Zakia, the textlinguist Noth (1987, 1988) has noted that advertisements are usually unambiguous with respect to their purpose, but vary so dramatically in surface structure that the range of possible semantic interpretations is virtually limitless.

In essence then, the standard theories and models of advertising communication are diametrically opposite to the Baker-Zakia-Noth viewpoint, i.e., whether advertising visuals are generally open or closed texts for their destined market segments. If the prevailing perspective is wrong, then the notions of comprehension and denotative meaning will require serious reconsideration with respect to advertising pictorials and advertising theory.

Although revealing, Zakia's work requires extension beyond single word responses in order to more thoroughly investigate the semanticity of advertising illustrations. Such an effort would also permit a more detailed examination of the open/closed text issue. We reasoned that by having consumers from a given market segment describe a segment-appropriate pictorial ad and indicate what they think the advertiser intended with the visual content, we could determine the extent of interpretive agreement (disagreement) as an assessment of ad-text openness (closedness).

DOES SEX SELL?

As a message strategy, sexuality is replete across advertising illustrations. Given the premise that the primary goal of an advertisement is to produce consumer desire for the brand, MacCannell (1987, p.528) contends that sexual innuendo is effective in advertising pictorials precisely because it models desire. This hypothesis derives from MacCannell's scanning of contemporary advertising, based on sociocultural insights about myths and a semiotic appreciation for the role of kinesics and proxemics in nonverbal communication. According to MacCannell (1987, p.528),

a close reading reveals that it is not actual bio-reproductive sex that is used for selling. Rather, it is a kind of sexual suggestion which covers over the censure of actual sex.... Specifically, what is represented is not sex but the blockage, censure, or inhibition of the sex act which is simultaneously suppressed from the image and is also the only possible motive for it.

By MacCannell's account, a person desires only what he or she does not have already; uninterrupted desire is maximally maintained when the object of desire is nearby, but just beyond reach. Hence, pervasive sexual suggestion in mass advertising serves as a model for culturally-engrained product desire, coalescing the desire for commodification with the commodification of desire. MacCannell notes that most sexual imagery is masculine oriented and, as such, "anything that simultaneously suggests and denies, or seems to invite while barring male sexual involvement, reproduces the essential structure of the cultural production of desire under a male sign" (p. 530).

In saying that sex in advertising models desire, we take MacCannell to mean--at the least--that there is an implicit association between depicted sexual desire and brand desire. Whether this means that the two desires are in some sense parallel but separate, or perhaps parallel and Superimposed, is not wholly clear from MacCannell's comments, though that distinction does not appear critical to his thesis. It is the association which is tantamount. The association could be instrumentalist (the picture displays brand desire in which product acquisition may serve to fulfill sexual desire); the association could be metaphorical (the picture suggests that brand desire has intrinsic similarities to sexual desire); or the association could be metonymic (the picture simply co-presents brand desire and sexual desire contiguously). Additional associations may be possible and certainly more than one type could be involved in a given illustration. [The mechanisms by which this proposed correlation has developed between product/brand desire and sexual desire are likely to be complicated and subtle on both biological and cultural levels. One explanation derived from classical conditioning would suggest that a positive brand attitude might emerge through the association of the brand with other message stimuli (sexuality) that are reacted to positively (see Gorn 1982).]

The feasibility and the power of MacCannell's proposition depend on consumers interpretively associating brand desire with depicted sexual desire. Those who interpret such an association must represent more than an inconsequential percentage of sample consumers, otherwise the proposition bears no import.

METHOD

Consumers for this study consisted of a convenience sample of junior and senior undergraduate students from a major southeastern university. Twenty subjects were scheduled; two failed to keep their appointments and one was subsequently dropped from the study because English was not her native language. This left 17 subjects, nine females and eight males. Subjects were compensated with monetary payment ($3) or course credit, whichever they preferred.

Subjects were processed individually and their spoken responses were taped with a small, unobstrusive recorder. Introductory instructions informed subjects that the study focused on people's interpretations of magazine advertisements. They were told that we were interested in their natural thoughts and feelings in answering some questions about each ad they were about to be shown. Subjects were then exposed to four consecutive full-page magazine advertisements (not facsimiles), each matted inside a separate manilla folder. After opening each folder, subjects were given 15 seconds to view the ad and then, with the ad still available, subjects were handed four consecutive 3x5 cards on which a different response request was typed. Subjects read the cards silently and then verbalized. At the end of each initial response, the researcher probed once more with "Anything else?" Then the next card was handed to them. After the fourth card for the fourth ad, subjects completed a one-page information sheet about their prior use of the products and brands shown in the ads. They were then debriefed, paid, and dismissed.

The four cards for each of the four ads were identical in terms of content and order of presentation. The questions were:

1) In your own words, please describe what's going on in the picture.

2) Ignore for now what the advertiser may have intended and tell us your opinions and feelings about what's going on in this picture.

3) What do you think the advertiser was trying to communicate with the picture in this particular ad?

4) What kinds of evidence can you identify in the picture to support your ideas about what the advertiser tried to communicate?

The four ads in this study were taken from general circulation magazines which college students occasionally view. Each ad was predominantly nonlinguistic, with minimal verbal information. It is worth emphasizing that ads which have no verbal content are practically nonexistent and, for this initial Study, we chose not to alter the actual ads. The advertised brands represented product classes for which 20-22 year old adults demonstrate strong interest: sunglasses, cigarettes, liquor, and perfume. The ads themselves each included young adult models.

The target ad for this study was a Gordon's gin ad (see the Figure) and it was always the third ad shown to the subjects. This particular ad has been used by the firm for several years; obviously, company executives have judged the ad to be an effective communication about their brand to young adults.

The ad portrays a man and a woman on a beach. She is painting, seated on a stool in front of an easel; he is sitting on the sand close behind her, pulling on her dresscoat. He is looking up at her, whereas she is looking downward at her painter's palette. The brand is placed in the lower-right comer, accompanied in the middle-bottom with the quote "I could go for something Gordon's" and the slogan "The possibilities are endless." The picture itself is black and white, while the colors yellow, orange, and green appear on the bottle or in the lime inside the glass.

As already noted, this ad was selected because young adults can easily relate to both the product and the situational content of the ad. In addition, for assessing the propositions previously discussed, the picture is vivid and plainly harbors sexual innuendo. The standard open-text proposition would be called into question if more than one interpretation theme was produced by the subjects. MacCannell's proposition would be undermined if few subjects associated the sexual desire with desire for the brand.

ANALYSIS AND RESULTS

After the protocols were transcribed, the first author reviewed them and discerned three themes specifically related to subjects' thoughts on what the advertiser was trying to communicate with the picture. A theme represented the central, dominating idea expressed by the subject. A fourth category was used for those subjects who were unable to associate the picture and the brand. The four categories were:

1) A peacefulness theme expressed through associations between the physical setting (beach context) and the brand (e.g., its relaxing effects);

2) A desire theme expressed through associations between the sexual innuendo and desire for the brand;

3) A status theme expressed through associations between the type of people shown in the ad (e.g., young, healthy, stylish, etc.) and the brand (its ability to transform the viewer into a similar sort of individual); and

4) No association apparent between the picture and the brand.

The two authors then independently classified each subject's protocol according to the four categories above, with extra attention given to subjects' comments about the intended message and related evidence in the illustration (i.e., the third and fourth response tasks). Agreement between the two authors was 15 out of 17 classifications; the remaining two were resolved through review and discussion of the subjects' protocols.

The results indicated that six subjects generated the peacefulness theme, five subjects produced the desire theme (three males, two females), one subject derived the status theme, and five subjects perceived no relationship between the ad illustration and the brand. No subject expressed more than one theme. Examples of six subjects' partial protocols appear in the Table, along with their respective classifications.

DISCUSSION

Propositions Tested

ne proposition that ad pictorials typically represent open texts to their respective market segments is not supported by our data. If the gin ad was an open text, then more agreement and certainty across the total seventeen subjects should have emerged with respect to their picture-brand interpretations. Consequently, while many pictorial ads are meticulously designed for a target audience, intended to be relatively constrained to key denotative meanings, it may be that the majority of pictorial ads are semantically closed. Denotatively they are sterile; connotatively they border on anarchy. Myers'(1983, p. 214-16) warning is prophetic for ad comprehension researchers:

It seems that pictorial ads for highly symbolic products such as liquor, jewelry, automobiles, cologne and perfume, fashion clothing and accessories, and home furnishings would be most likely to spur elaborate inferential interpretations. Whether these same findings would hold for less symbolic products (e.g., kitchen appliances, house and-garden tools) remains to be shown.

MacCannell's proposition appears marginally supported by our data. Five out of the seventeen total subjects (29%) produced interpretations congruent with the proposition that sexual desire in advertising imagery can serve as a model for product desire. However, in looking only at the twelve subjects who constructed interpretations associating the picture and the brand, 42% produced the "desire" theme (5/12). Several of our subjects definitely linked the ad's sexual innuendo with brand desire.

Limitations

Some limitations of this study must be acknowledged before moving to further discussion topics. Foremost is the fact that in this study we utilized only one ad, for a single product class (liquor). [Data from the other ads are being analyzed as part of a larger study.] Therefore, our results are strictly suggestive, not definitive. They should only be taken as a springboard to further theory and research on the substantive and methodological issues addressed here.

Secondly, a set of limitations revolves around our use of a protocol technique to address the semantic content of consumers' interpretations during visual processing. By requesting verbalized reports, we may have encouraged some subjects to construct meanings which would not have emerged under normal processing conditions. Moreover, Lutz and Lutz (1977) and Rossiter (1982) have pointed out that pictures are quite effective in eliciting mental imagery, some of which probably formed the semantic substance of our subjects' interpretations. According to MacInnis and Price (1987, p.485), verbal reports of mental imagery require translations between imagery and discursive (verbal) processing modes, potentially confounding them. Since we were not focused on separating the modes per se, this is only a serious problem in our study to the degree that the translations altered the semantic contents of subjects' basic thematic interpretations. Though our data do not permit us to deflect this concern completely, we have no evidence to suggest that it happened. Prior research has shown that adults spontaneously assign verbal labels to most pictorial stimuli (Pezdek and Evans 1979) and, as a whole, the subjects who generated picture-brand associations in our study responded effortlessly.

Another limitation concerns articulate individuals who may have performed the imagery-to-discursive translation easier than less verbose individuals. If so, the less verbose subjects may have been more inclined to say there was no association between the illustration and the brand when, in fact, they may have mentally imaged an association.

Other limitations (and criticisms) surrounding the general Use Of Protocol methods to uncover mental processes are well known and will not be repeated here (Ericsson and Simon 1984; Russo, Johnson, and Stephens 1986). We make no claims as to capturing the whole or the majority of meanings our subjects abscribed to the Gordon's gin ad. Our intent was to glimpse the pictorial interpretation process sufficiently enough through verbalization in order to identify significant semantic concepts and to characterize overall themes. The protocol method used in this study was partly retrospective and partly concurrent, given that our subjects were allowed 15 seconds to view the ad before verbalizing. Future studies may seek to reduce retrospectiveness by requesting verbalizations from the moment of stimulus presentation.

Further Insights and Future Directions

Baker's (1985) provocative article sensitizes us to the ambiguous and arguable division between denotation and connotation in graphic art. As he points out, for years semioticians and other communication theorists have recognized the arbitrariness of the relation between linguistic signs (words) and their referents (concepts)--a condition which makes the meaning of words anything but intransigent. As a result, the boundary line between denotation and connotation quickly fades, bluffing those comfortable and seemingly self-evident distinctions between intended versus unintended meaning, primary versus secondary meaning, and literal versus inferential meaning. This is especially true when viewed from the message receiver's perspective.

FIGURE 1

TABLE

SIX PARTIAL PROTOCOLS AND THEIR CLASSIFICATIONS

Nonetheless, arbitrariness between sign and referent has not been Seriously considered in graphic art. The received view, which Baker strives to dislodge, is that graphics --compared to verbal language--more readily incorporate a demarcation between denotative and connotative meaning. In marketing and consumer research this belief is variously propagated (e.g., Rossiter and Percy 1983). Also, in their thorough review on mental imagery processing, MacInnis and Price (1987, p. 475) state that "Compared with symbolic or language-like processing, imagery processing bears a non-arbitrary correspondence to the thing being represented." In fact, there is broad support in psychology for the contention that an object and its mental image are structurally isomorphic (see Shepard 1978). However, it would be erroneous to conclude that that structural isomorphism is related to denotated meaning when consumers have mental images elicited by advertising illustrations. If this were true, the denotative meaning of Gordon's pictorial ad should have been straightforward and widely shared across our subjects. It was not.

The immediate implications of this argument form a momentous hurdle to advertisers who seek lucid communication through pictures and to researchers who strive to understand the nature of pictorial interpretation. Perhaps the only path from the horror of connotative anarchy is to return to the marketing nucleus: like the proverbial customer, the interpreter is always right. For, as Baker (1985, p.173) reminds us, "it is really of very little consequence to the viewers if their reading is not the intended one." A possible reaction to this viewpoint is to say that if the reading is not the intended one, then perhaps it is false, in the sense of advertising miscomprehension (see Jacoby and Hoyer 1987). We are skeptical of that rationale and find ourselves in broad agreement with Freundlieb (1984, p.82):

Interpretive statements, although undoubtedly the result of complex cognitive processes, are not empirically true or false of some independently existing object in the external world .... For empirical statements to be true or false, there must be external objects or states of affairs of which those statements are true or false. In the case of interpretations concerning higher level meanings, no such independent truth conditions exist.

Borrowing from Schmidt's (1984) philosophical argument, we submit that the interpretation of advertising is a subject-dependent, internal monologue through which consumers generate, change, and maintain their individual reality.

McCracken (1987, p.122) has also articulated a meaning model of advertising consumption, characterized by consumers "looking [at ads] for symbolic resources, new ideas, and better concrete versions of old ideas with which to advance their project," i.e., the construction of a unified life system of self, family, community, nation, and world. This is not an altogether novel idea, since a small band of researchers has long argued that consumers interpret advertising for the cultural and psychological meanings it symbolizes (Friedman and Zimmer 1988; Levitt 1970; Levy 1959; McCracken 1986; Sherry 1987; White 1959). However, the systematic application of this meaning model in advertising theory and research has been thwarted by the combined forces of the techno-rational information processing model and a reluctance to adopt open-ended, interpretive methodologies. On the other hand, our data substantiate the value of this revisionist perspective on advertising consumption, including the barrenness of denotation in traditional models of advertising communication. Beyond our study, future research must look at a wider array Of product categories and different market segments. In addition, selecting or manipulating ads with varying degrees of pictorial clarity and verbal-visual message -interactions should also prove insightful.

Other than attempting to show the efficacy of an interpretive protocol in assessing Propositions about advertising, a number of other issues naturally arose which are quite germane to a meaning model of advertising consumption. We turn now to some of those related insights.

Same Ad? Different Worldviews. Research has shown that sex and gender are strong determinants of visual processes (Johnson, Zimmer, and Golden 1987). In our study, males and females provided some conspicuously dissimilar descriptions of the ad illustration (first response task). Male subjects commented that the ad-male was attracted to the ad-female:

S#2: "He seems to be very attracted to her and is trying to seduce her."

S#4: "The way he's looking at her, you'd assume that he cares for her, that they might even be lovers."

S#6: "Guy is lying on the ground, tugging on her, seems to be only thinking of her."

Female subjects expressed a more profound interpretation:

S#5: "She's very artistic. He looks like he adores her."

S#10: "He is admiring her because she's really pretty and her work on the canvas."

S#17: "'The guy is kind of in awe."

For female subjects, the picture suggests both romance and the ad-female's artistic talents. Females acknowledge the ad-female as something more than a here-and-now sexual target; she is lovable and has skills and value that transcend the ad setting. At least, that is what the female subjects want the ad-male to recognize. If the diffused feminist's disposition of the 1980s is truly operative, our female subjects' interpretations may well reflect what they want the men in their own lives to recognize about them.

For male subjects, the picture reads quite differently: sexual gameship (though not all the males thematically associated the sexual desire with brand desire). For them, the image is erotic and hints at behaviors prefatory to consummation. As one male subject put it, "He's in an enviable position, beside her and alone."

While these interpretational discrepancies between males and females are hardly surprising, they do underscore that the meaning of the ad image, even on a surface descriptive level, is not literal or inherent to the ad, i.e., it is not denotational. The ad we used in this study depicts an asymmetry of gender quite different from those in Goffman's (1979) earlier work. Unlike his ads, ours reveals a woman in a comparatively more powerful, less dependent, and less attentive Posture. While several of our female subjects interpreted the ad-female to be in control of the pictorial scene, Most Of OUT male subjects were oblivious to such a possibility.

Our data also reveal that pictorial advertising which evokes human sexuality invites readings commensurate with acculturated sexual orientations. Whether such ads are equally capable of garnering positive brand attitudes, ad attitudes, and brand recall across the sexes remains unknown. Apparently, the pictorial interpretation routes taken by males and females can be quite different, and deserve more detailed research.

Idiosyncracies in the Mindspace. Although our study used neither a Thematic Apperception Test nor any similar request for fabricated elaborations, several of our subjects automatically constructed ministories around the ad illustration, including unique inferences. One female subject hypothesized that the ad- male "might have come up from out of nowhere ... and he just decided to sit down next to her and watch her and admire her." One male subject conjured up the unseen private destination the ad-male had in mind, i.e., trying to "get her inside the beach house." Another male subject speculated that "they might have some alcohol in them, they might be a little tipsy." Instantly and unabashed, one female had a positive hedonic response to the beach setting. This subject went on to say,

S#5: The way she's painting the picture, it's like a changing thing, and being on the beach, the beach is a changing thing itself, the waves and the sand shift .... She could paint anything, you know her colors, different mixtures would create a whole different mood with the painting .... the possibilities arc endless and also with their relationship.

Personalized meanings have been highlighted by some advertising researchers (e.g., Leavitt, Waddell, and Wells 1970; Shavitt and Brock 1986), though neither idiosyncratic nor personalized meanings have been accorded any status in advertising comprehension theory or research. Recently, however, in a study by Morris et al. (1985), recall results suggested that illogical conclusions drawn from an OTC drug ad were precipitated by consumers' inclinations to elaborate idiosyncratically while processing the message. Our data also suggest that idiosyncratic inferences can play an important part in the interpretation of pictorial advertising.

Interpretive Knots. Subjects' interpretations also varied along a continuum of complexity not commonly considered or uncovered in ad comprehension studies. With the gin ad, intricate interpretation was often dependent on a keen eye for the proxemic and kinesic cues of flirtation. Laing (1970, p.48) has poignantly noted the spiral of courtship cognition in this way:

She wants him to want her

He wants her to want him

To get him to want her

she pretends she wants him

To get her to want him

he pretends he wants her

Jack wants                   Jill wants

Jill's want of Jack        Jack's want of Jill

so                                so

Jack tells Jill                 Jill tells Jack

Jack wants Jill             Jill wants Jack

                a perfect contract

Some of our subjects echoed Laing in interpreting the ad image:

S#12: She looks like she's totally ignoring him and he thinks it's kind of funny too, that she's ignoring him, like maybe he knows that she's purposefully ignoring him.

S#17: She seems to be in her own little world maybe like, not even ignoring him, but strategically ignoring him, kind of playing hard to get because he's tugging on her shirt and she has like a little smile on her face, so she's kind of like teasing or something. ... Ignoring him, but not really ignoring him, kind of also acknowledging at the same time.

However, other subjects were insensitive to these subtle nonverbal signs. They developed comparatively simple interpretations:

S#4: They're out on a beach somewhere. She's painting a seascape, intimate, there's nobody around. Just a time with the two of them. Not dressed for swimming. They're dressed, just in their regular street clothes, casual, comfortable. Obviously she came down there to paint. She's got everything she needs.

Overall, female subjects tended to be more attuned to the range of body language signs in the illustration, especially those produced by the ad-female.

Besides the complexity dimension, a different interpretive knot of paradox and contradiction also appeared. In particular, some subjects casually constructed inconsistent meanings:

S#1: His expression, it looks like he's almost pleading with her to go. Her expression, she's almost ignoring him. It seems like there could be a conflict of interest here. And again, the sand, the water, the ocean, it looks like they're out in the open and not really into anything serious. It looks like they're both in peace of mind, just by where they are and what they're doing. [our italics]

Consistent, deductive cognition is no more commonplace in advertising interpretation than it is in consumer choice behavior. Yet, past studies have focused on ad comprehension using measures that penalize the Creative interpreter. This is shortsighted because, at a metaphorical level, life is a struggle between opposing forces which consumers arbitrate or admix in the daily marketplace: indulgence versus discipline, the profane versus the sacred, commonality versus uniqueness, etc. Hence, paradoxical interpretations of advertising pictorials are not necessarily degenerate. Their nature and role in advertising consumption is completely undetermined.

Envisionment. Meaning is molded through a formative interpretation process. Language researchers such as Miyake (1986) refer to this aspect of interpretation as iteration, Rumelhart (1984) likens it to hypothesis testing, and Langer (1986) calls it envisionment. As Langer (1986) puts it,

Meaning derived from any given portion of the text is shaped by how earlier segments were interpreted and continues to develop and change in light of later segments. These changing "envisionments" are records of the "text internal" world that is constructed by the reader while processing the text. These envisionments are the primary "dynamics" through which the reader experiences the message.

Several of our subjects demonstrated that envisionment is quite relevant to pictorial interpretation as well:

S#11: She's trying to paint and he's bothering her and she doesn't want him to bother her. She may be a bit, well, she doesn't look like she's irritated. No, she has a slight smile on her face.

S#15: He's having a decent time, just watching, although maybe not, he's pulling on her, maybe looking for attention for himself.

Focusing on the tentative, groping, and even playful character of interpretation should provide insights about other significant dimensions of ad-picture comprehension which have gone unresearched. These include foci of attention (Chafe 1979, 1980), the activation of schemas (Langer 1986), and metacomprehension strategies (Flavel 1976; Raphael et al. 1981).

CONCLUSION

Though it is often remarked that ads are intricate, multidimensional stimuli and that consumers perceive them through a cocktail of contexts and epistemic structures (individual, social, and cultural), most advertising researchers cling to the belief that meaning can be affixed denotatively to advertising messages. Moreover, this denotated meaning is considered both primary and necessary to the derivation of secondary, connotative meanings. This article has challenged those premises.

Research in advertising comprehension requires a fundamental reprioritization of its goals, most dramatically in the case of pictorial advertising. The central question is not "How much of advertising illustrations do consumers understand or misunderstand?" (see Watkins 1984) Instead, the question should be "What characterizes the process of advertising interpretation, such that consumers pass through these consumption experiences reinforced or changed with respect to the meaning of their lives?" Lannon and Cooper (1986), Levy (1986), McCracken (1987), Mick (1986), and Sherry (1987) have each argued for a cultural, subjective-meaning model of advertising consumption. Empirical research within this framework is greatly needed.

This study was conducted as a demonstration work, illustrating that open-ended, interpretive methods are not strictly for theory development; they can also be utilized to assess propositions in much the same manner experiments and surveys are employed. In addition, when coupled with an imagistic ad, the interpretive protocol method revealed a number of advertising consumption issues which have been undervalued or overlooked in past comprehension research. Adopting a cultural, subjective-meaning model, interpretive protocol methods appear especially useful for examining the semanticity of advertising illustrations, notably their connotations.

The mind is its own place, and in itself Can make a heav'n of hell, a hell of heav'n.

    Milton, Paradise Lost, Book 1, line 253

REFERENCES

Baker, Steve (1985), 'The Hell of Connotation," Word and Image, 1 (2), 164-175.

Ballstaedt, Steffen-Peter and Heinz Mandl (1984), "Elaborations: Assessment and Analysis," in Learning and Comprehending of Text, eds. Heinz Mandl, Nancy Stein, and Tom Trabasso, Hillsdale, NJ: Erlbaum, 331-352.

Barthes, Roland (1964), "Rhetoric of the Image," in Working Papers in Cultural History, (Spring 1971), translated by Brian Trench, 37-50.

Baudrillard, Jean (1981), For a Critique of the Political Economy of the Sign, translated by Charles Levin, St. Louis, MO: Telos Press.

Belk, Russell W., John F. Sherry, Jr., and Melanie Waldorf (1986), "A Naturalistic Inquiry into Buyer and Seller Behavior at a Swap Meet," Journal of Consumer Research, 14 (4), 449-470.

Bryson, Norman (1983), Vision and Painting: The Logic of Gaze, London: MacMillan.

Calder, Bobby J. and Alice M. Tybout (1987), "What Consumer Research Is..." Journal of Consumer Research, 14 (1), 136-140.

Chafe, Wallace L. (1980), "The Development of Consciousness in the Production of a Narrative," in The Pear Stories: Cognitive, Cultural, and Linguistic Aspects of Narrative Production, ed. Wallace L. Chafe, (Volume III, in Roy 0. Freedle, ed., Advances in Discourse Processing), Norwood, NJ: Ablex.

Chafe, Wallace L. (1979), "The Flow of Thought and the Flow of Language," in Syntax and Semantics, ed. T. Givon, New York: Academic Press.

Chattopadhyay, Amitava and Joseph W. Alba (1988), 'The Situational Importance of Recall and Inference in Consumer Decision Making," Journal of Consumer Research, 15 (1), 1-12.

Childers, Terry L. and Michael Houston (1984), "Conditions for a Picture Superiority Effect on Consumer Memory," Journal of Consumer Research, 11 (2), 643-655.

Eco, Umberto (1979), The Role of the Reader: Explorations in the Semiotics of Texts, Bloomington, IN: Indiana University Press.

Edell, Julie A. and Richard Staelin (1983), "The Information Processing of Pictures in Print Advertising, Journal of Consumer Research, 10 (1), 45-61.

Ericsson, Anders K. and Herbert A. Simon (1984), Protocol Analysis: Verbal Reports as Data, Cambridge, MA: MIT Press.

Flavell, J. H. (1976), "Metacognitive Aspects of Problem Solving," in The Nature of Intelligence, ed. B. Resnick, Hillsdale, NJ: Erlbaum.

Freundlieb, Dieter (1984), "Explaining Interpretation: The Case of Henry James's The Turn of the Screw," Poetics Today, 5 (1), 79- 95.

Friedmann, Roberto and Mary R. Zimmer (1988), "The Role of Psychological Meaning in Advertising, Journal of Advertising, 17 (1), 41-48.

Goffman, Erving (1979), Gender Advertisements, New York: Harper and Row.

Gorn, Gerald J. (1982), "ne Effects of Music in Advertising on Choice Behavior: A Classical Conditioning Approach," Journal of Marketing, Vol. 46 (Winter), 94-101.

Graesser, Arthur C. and Leslie F. Clark (1985), 'The Generation of Knowledge-based Inferences during Narrative Comprehension," in Inferences in Text Processing, eds. G. Rickheit and H. Strohner, Amsterdam: North Holland Publishing, 53-90.

Hirschman, Elizabeth C. (1987), "Movies as Myths: An Interpretation of Motion Picture Mythology," in Marketing Signs: New Directions in the Study of Signs for Sale, ed. Jean Umiker-Sebeok, Berlin: Mouton de Gruyter, 335-373.

Holbrook, Morris B. (1983), "Product Imagery and the Illusion of Reality: Some Insights from Consumer Esthetics," in Advances in Consumer Research, Vol. 10, eds. Richard P. Bagozzi and Alice M. Tybout, Provo, UT: Association for Consumer Research, 65-71.

Holbrook, Morris B. and Mark W. Grayson (1986), "The Semiology of Cinematic Consumption: Symbolic Consumer Behavior in Out of Africa," Journal of Consumer Research, 13 (3), 374-381.

Hunt, Shelby D. (1983), Marketing Theory: The Philosophy of Marketing Science, Homewood, IL: Richard D, Irwin.

Hudson, Laurel Anderson and Julie L. Ozanne (1988), "Alternative Ways of Seeking Knowledge," Journal o Consumer Research, 14 (4), 508-521.

Jacoby, Jacob and Wayne D. Hoyer (1982a), "Viewer Miscomprehension of Televised Communication: Selected Findings," Journal of Marketing, 46 (Fall), 12-26.

Jacoby, Jacob and Wayne D. Hoyer (1982b), "On Miscomprehending Televised Communication: A Rejoinder, " Journal of Marketing, 46 (Fall), 35-43.

Jacoby, Jacob and Wayne D. Hoyer (1987), The Comprehension and Miscomprehension of Print Communications: An Investigation of Mass Media Magazines, New York: Advertising Educational Foundation.

Johnson, Keren A., Mary R. Zimmer, and Linda L. Golden (1987), "Object Relations Theory: Mate and Female Differences in Visual Information Processing," in Advances in Consumer Research, Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT: Association for Consumer Research, 83-87.

Kintgen, Eugene R. (1983), The Perception of Poetry, Bloomington, IN: Indiana University Press.

Laing, R. D. (1970), Knots, New York: Vintage Books.

Langer, Judith A. (1986), "The Construction of Meaning and the Assessment of Comprehension: An Analysis of Reader Performance on Standardized Test Items," in Cognitive and Linguistic Analyses of Test Performance, ed. Roy 0. Freddie, New York: Ablex.

Lannon, Judie and Peter Cooper (1983), "Humanistic Advertising: A Holistic Cultural Perspective," International Journal of Advertising, 2, 195-213.

Leavitt, Clark, C. Waddell, and William Wells (1970), "Improving Day-After Recall Techniques," Journal of Advertising Research, 10, 13-17.

Levitt, Theodore (1970), "The Morality(?) of Advertising," Harvard Business Review, 48 (July-August), 84-92.

Levy, Sydney (1986), "Meanings in Advertising Stimuli," in Advertising and Consumer Psychology, eds. Jerry C. Olson and Keith Sentis, New York; Praeger, 214-226.

Levy, Sydney (1959), "Symbols for Sale," Harvard Business Review, (July-August), 117-124.

Lutz, Kathy A. and Richard J. Lutz (1978), "Imagery-Eliciting Strategies: Review and Implications of Research," in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Ann Arbor, MI: Association for Consumer Research, 611-620.

MacCannell, Dean (1987), "'Sex Sells': Comment on Gender Images and Myth in Advertising," in Marketing Signs: New Directions in he Study of Signs for Sale, ed. Jean Umiker-Sebeok, Berlin: Mouton de Gruyter, 521-531.

MacInnis, Deborah J. and Linda L. Price (1987), "The Role of Imagery in Information Processing: Review and Extensions," Journal of Consumer Research, 13 (4), 473-491.

McCracken, Grant (1987), "Advertising: Meaning or Information," in Advances in Consumer Research, Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT: Association for Consumer Research, 121-124.

McCracken, Grant (1986), "Culture and Consumption: A Theoretical Account of the Structure and Movement of the Cultural Meaning of Consumer Goods," Journal of Consumer Research, 13 (1), 71-84.

McGuire, William J. (1969), 'The Nature of Attitudes and Attitude Change," in Handbook of Social Psychology, eds. G. Lindzey and E. Aronson, Reading, MA: Addison-Wesley.

Mick, David Glen (1988a), "Contributions to the Semiotics of Marketing and Consumer Behavior 1985-1988," in The Semiolic Web: A Yearbook of Semiotics, eds. Thomas A. Sebeok and Jean Umiker-Sebeok, Berlin: Mouton de Gruyter, 535,584.

Mick, David Glen (1988b), "A Critical Review of the Comprehension Construct in Marketing Communications Research," Working Paper, Marketing Department, College of Business Administration, University of Florida.

Mick, David Glen (1987), "Levels of Comprehension in Consumers' Processing of Print Advertising Language," Unpublished Ph.D. Dissertation, Indiana University.

Mick, David Glen (1986), "Consumer Research and Semiotics: Exploring the Morphology of Signs, Symbols, and Significance," Journal of Consumer Research, 13 (2), 196-213.

Miyake, Naomi (1986), "Constructive Interaction and the Iterative Process of Understanding," Cognitive Science, 10, 151-177.

Morris, Louis A., David Brinberg, Ron Klimberg, Carole Rivera, and Lloyd G. Millstein (1986), "Miscomprehension Rates for Prescription Drug Advertisements," Current Issues and Research in Advertising, 9, 93-117.

Myers, Kathy (1983), "Understanding Advertisers," in Language, Image, Media, eds. Howard Davis and Paul Walton, Oxford: Basil Blackwell, 205-223.

Noth, Winfried (1988), personal communication.

Noth, Winfried (1987), "Advertising: The Frame Message," in Marketing Signs: New Directions in the Study of Signs for Sale, ed. Jean Umiker-Sebeok, Berlin: Mouton de Gruyter, 279-294.

Olson, Gary M., Susan A. Duffy, and Robert L. Mack (1984), "Thinking -Out- Loud as a Method for Studying Real-Time Comprehension Processes," in New Methods in Reading Comprehension Research, eds. David E. Kieras and Marcel A. Just, Hillsdale, NJ: Erlbaum, 253-286.

Pezdek, K. and G. W. Evans (1979), "Visual and Verbal Memory for Objects and their Spatial Locations," Journal of Experimental Psychology: Human Learning and Memory, 5, 360-373.

Raphael, Taffy E., Ann C. Myers, William C. Tirre, Mary Fritz, and Peter Freebody (1981), "The Effects of Some Known Sources of Reading Difficulty on Metacomprehension," Journal of Reading Behavior, 13, 325-334.

Richards, Jeff and Richard Zakia (1981), "Pictures: An Advertiser's Expressway through FTC Regulations," Georgia Law Review, 16 (1), 77134.

Rossiter, John R. (1982, "Visual Imagery: Applications to Advertising," in Advances in Consumer Research, Vol. 9, ed. Andrew A. Mitchell, Ann Arbor, MI: Association for Consumer Research, 101-106.

Rossiter, John R. and Larry Percy (1983), "Visual Communication in Advertising," in Information Processing in Advertising, ed. Richard Jackson Harris, Hillsdale, NJ: Erlbaum, 83-125.

Rossiter, John R. and Larry Percy (1980), "Attitude Change Through Visual Imagery in Advertising, Journal of Advertising, 9 (2), 10-16.

Rumelhart, David E. (1984), "Understanding Understanding," in Understanding Reading Comprehension: Cognition, Language, and the Structure of Prose, ed. J. Flood, Newark, DE: International Reading Association, 1-20.

Russo, J. Edward, Barbara L. Metcalf, and Debra L. Stephens (1981), "Identifying Misleading Advertising," Journal of Consumer Research, 8 (2), 119-131.

Russo, J. Edward, Eric J. Johnson, and Debra L. Stephens (1986), "The Validity of Verbal Protocols," Working Paper, Marketing Department, Cornell University.

Schmidt, S. J. (1984), "The Fiction is that Reality Exists: A Constructivist Model of Reality, Fiction, and Literature," Poetics Today, 5 (2), 253274.

Shavitt, Sharon and Timothy C. Brock (1986), "Self-Relevant Responses in Commercial Persuasion: Field and Experimental Tests," in Advertising and Consumer Psychology, eds. Jerry C. Olson and Keith Sentis, New York: Praeger, 149-171.

Shepard, Roger N. (1978), "The Mental Image," American Psychologist, 33 (February), 125-137.

Sherry, John F., Jr. (1987), "Advertising as a Cultural System," in Marketing Signs: New Directions in the Study of Signs for Sale, ed. Jean Umiker-Sebeok, Berlin: Mouton de Gruyter, 441-461.

Shimp, Terence A. and Ivan L. Preston (1981), "Deceptive and Nondeceptive Consequences of Evaluative Advertising," Journal of Marketing, 45 (Winter), 22-32.

Watkins, Thomas (1984), "Misunderstanding Pictures in Print Ads," Viewpoint, Chicago: Ogilvy and Mather, Inc., Spring, 31-37.

White, Irving (1959), "The Functions of Advertising in Our Culture," Journal of Marketing, 24 (July), 8-14.

Zakia, Richard (1986), "Adverteasement," Semiotica, 59 (1/2), 1-11.