The Dynamic Component in Attitudes Towards the Stimulus

ABSTRACT - The effects of emotions in consumer behavior are mainly discussed as attitudes-towards-the-ad/or stimulus (Aad) This study is to show that Aad contains a dynamic affective component Thus Aad will be a mixture of emotional state at the beginning of the exposure and modulated instantaneous emotions during exposure In addition, a direct effect of Aad on intention to buy could be found.

Citation:

Bruno Neibecker (1987) ,"The Dynamic Component in Attitudes Towards the Stimulus", in NA - Advances in Consumer Research Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT : Association for Consumer Research, Pages: 482-486.

Advances in Consumer Research Volume 14, 1987      Pages 482-486

THE DYNAMIC COMPONENT IN ATTITUDES TOWARDS THE STIMULUS

Bruno Neibecker, University of the Saarland (Fed. Rep. of Germany)

[The author wishes to express his gratitude to W Kroeber-Riel, J E Russo and R. P Bagozzi for ideas on the subjects and helpful editorial comments. Authors address Institute fur Konsum- u. Verhaltensforschung, 6600 Saarbrucken 11, FRG.]

ABSTRACT -

The effects of emotions in consumer behavior are mainly discussed as attitudes-towards-the-ad/or stimulus (Aad) This study is to show that Aad contains a dynamic affective component Thus Aad will be a mixture of emotional state at the beginning of the exposure and modulated instantaneous emotions during exposure In addition, a direct effect of Aad on intention to buy could be found.

INTRODUCTION

Much of the previous research in advertising and communication effects has concentrated on "controlled" verbal processes. The cognitive structure approach, which emphasizes the beliefs about a stimulus that are formed or changed after exposure, explains between 15% to 30% of the variance in attitudes (e.g., Ryan and Bonfield, 1980). By controlling for measurement error up to 50% can be achieved (Bagozzi, 1982).

In addition, much of the cognitive response approach has concentrated on "verbal" processing that occurs during exposure to a stimulus. In this approach, protocols are collected at least immediately after exposure (e.g. Wright, 1981; Batra and Ray, 1983). This kind of process measurement more adequately accesses the dynamic component of consumers' processing of a message. By collecting data during stimulus exposure. the on-going attitude and intention-formation process becomes transparent.

It is estimated, that the analysis of verbal processes can explain up to 70% of the variance in attitudes or (purchase) intention (Mitchell, 1983, p 28) The remaining 30% of the variance may be accounted by emotional behavior and worth a closer look which is the goal of this paper.

Attitude-towards-the-stimulus or more specific attitude towards-the-at (Aad) is seen as a predominant construct to analyze consumers' emotion. Aad is considered as a mediator of advertising's effects on brand attitudes and purchase intentions (e.g. Mitchell and Olson 1981; Lutz, 1985)

In general, three central ways of studying consumers' emotions can be distinguished:

- The category approach which postulates a set of basic or primary emotion categories like: joy, anger, sadness, fear, acceptance (Izard, 1977; Plutchik, 1980).

- Dimension theorists who formulate a small set of (independent) dimensions that best describe the differences in emotional behavior (Mehrabian and Russel, 1974). This approach provides verbal measures of pleasure, activation (arousal), and, with less consistency, the amount of dominance elicited by a stimulus The pleasantness dimension has been shown to be highly correlated with the evaluation factor of the semantic differential (Ward and Russell, 1981; see also Neibecker 1984; 1986)

- Also recent cognitive response research has concentrated on the emotional component in verbal protocols (Batra and Ray. 1985). Coding of affective elements in cognitive responses seems to be another promising way to analyze consumers' emotions.

Whereas, there are signs to believe, that strongly verbal oriented approaches in detecting emotions remains captured in the information processing tradition (Derbaix and Abeele, 1985; Kroeber-Riel 1984) Emotion psychology as well as research on hemispheric specialization gained theoretical insights, which draw our attention to the specific processing schema of our visual and emotional channel (Zajonc, 1980; Tucker, 1981; Lazarus, 1982; Bryden and Ley, 1983).

Of course, the crucial and specific question of the debate, whether the original affective reactions themselves must also contain cognitive components cannot be answered empirically given the current state of knowledge and research technology (ZaJonc and Markus, 1982) However, significant proportions of emotional reactions, even if cognitive mediation should be necessary, rely on nonverbal channels and are difficult to verbalize (i.e. Ekman and Friesen, 1969; Buck, Baron and Barrette, 1982) And a posteriori questions about the reasons for an affective evaluation may be often yield justifications rather than substantial explanations.

As part of the distinction between automatic and con trolled processes, Derbaix and Abeele (1985, p. 160) pointed out, not to equate controlled process with cognition only and automatic processes with emotion or action habits only. There can also be cognitive automatism, as in schemata or controlled emotions, as for example in the voluntary facial expression As a consequence, the autonomic-controlled distinction should not be considered as absolute. So nowadays, there is a clear tendency to integrate both viewpoints and theorists try to consider both. cognitive and emotional processes, including global behavior such as motor components of affect (Bower, 1981; Clynes, 1980; Leventhal, 1982)

Further indications for the relevance of the Aad-construct are found in persuasion research where two basic routes are posited: one, called the central route, says that attitude change results from a person's careful consideration of information Here message content is the primary influence. However, people are not always motivated to think about the information to which they are exposed, yet attitudes may change nonetheless. Attitude change that occurs via this second or peripheral route does not occur because the person has diligently considered the pros and cons of the issue. In this case, the persons are affected more by the source of the message or contextual factors than by actual message content (Petty and Cacioppo. 1984).

ATTITUDES TOWARDS THE STIMULUS: A DYNAMIC PERSPECTIVE

At the present stage, Aad is conceptualized as a rather complex construct. Credibility, attitude towards the advertiser, attitude towards advertising and finally emotion (moot) are considered as relevant components of Aad (Lutz, 1985). The study focuses upon the emotional effects, perhaps the most relevant factor in Aad

Related terms for emotion found in Aad-literature are "warmth" and "mood or feeling state" (e.g. Holbrook and O'Shaughnessy 1984). Warmth is defined as "a positive, milt, volatile emotion involving physiological arousal and precipitated by experiencing directly or vicariously a love, family or friendship relationship" (Aaker, Stayman and Hagerty, 1986). Here a single, but complex emotional state is described, which may be characterized by 8 high degree of pleasantness combined with a medium arousal level Moot on the other side, has been defined as the consumer's affective state at the time of exposure to the stimulus and some kind of affect transfer from mood on Aad is assumed (Lutz, 1985). Other researchers describe the purpose of the advertisement as to create a favorable Aad by leaving the viewer in a positive emotional state after processing the ad (Hill and Mazis, 1986)

Most of the traditional definitions share a static view to the emotional component in Aad. On the other side, Izard (1977, p 5) defines emotional state as a particular emotional process of limited duration, which may last from seconds to hours. For TV-spots, pop-music and other dynamic stimuli of about 15 seconds or longer, the emotional state at the beginning of exposure may be modulated during exposure and even different emotions may be induced by the stimulus

In analogy to the recently outlined conceptualization of low involvement as a situation-specific state variable, a dynamic conceptualization for Aad is suggested (Gardner et al , 1985). This process-oriented view supplements the conventional Aad-construct Thus Aad will be a mixture of emotional state at the beginning of the exposure. modulated instantaneous emotions during exposure, individual differences etc.

As a further support for the "dynamic view" of Aad, it has been fount that short-run affective states can be induced by emotionally laden slides, films or everyday success conditions (e.g. Isen and Shalker 1982; Kroeber-Riel 1979; Buck 1982; Zuckerman et al. 1981) Therefore, in this study Aad is not considered to depend on the actual long-lasting emotional state only, but also is expected to be modulated in the short-run by the emotional content of the stimulus itself

Up to now lt remains to be shown that the dynamic component in Aad is able to explain a significant portion of the variance. Thus, the present study attempts to demonstrate that techniques designed to measure momentary emotions can add an additional amount of explained variance to the Aad-construct.

METHOD

Program Analyzer

To operationalize the construct "instantaneous emotion." like/dislike responses (affective pleasantness) were gathered throughout stimulus presentation, using the program-analyzer Related methods are discussed as "warmth monitor" (Aaker, Stayman and Hagerty, 1986) and "dialing procedure" (Friestad and Thorson. 1986). The program analyzer (PA), allows one to record reactions to an on-going program and was first devised by Paul F. Lazarsfeld and Frank Stanton in 1937 (Peterman. 1940; Brockhaus and Irwin, 1958; Levy, 1982; Neibecker, 1985). The subJect operates it with two buttons, in our study a ret button in the right hand and a grey one in the left hand. The instructions for the subjects were as follows:

We should like you to indicate whether you like or dislike these songs while you are listening to them. If, during any part of the program, you feel that you like what you are listening to, please press the button in your right hand. If you dislike what you are listening to, press the button in your left hand. If any parts leave you indifferent, just do not press either button.

In addition to this instruction, two "mnemonic" sheets were placed in front of the subjects showing two figures: on the left site with the label "dislike" and a sad face, and on the right side with the label "like" and a smiling face. Since the appearance of the original Lazarsfeld-Stanton program analyzer, other audience-response recorders have been devised For our study, a computer-assisted facility is used with on-line data recording.

Study Specification

The experimental data reported herein were collected in the Laboratory of the Institute for Consumer and Behavioral Research (University of the Saarland, FRG). First the subJects were asked to evaluate two German popmusic songs using program analyzer. In addition other songs were presented but are not considered in this paper. After passing this step, a paper-and-pencil session followed with affective (attitude) measures and a constant sum scale used as a purchase intention measure (for details, see Figure l). The semantic differential attitude scales were bipolar sevenCpoint rating scales and the Aad-scales were constructed with unipolar itemized rating scales with seven points. Instead of advertisements, popmusic songs were used in this study. Analogous to the Aad literature, attitudes towards the songs were measured Consequently it would be more correct to talk about attitude-towards-the-stimulus - but to avoid the introduction of a new term, we talk about Aad throughout this paper.

A total of 130 subjects was chosen from the main target population for these kind of songs and was placed at our disposal by one of the performers. Therefore, all subjects were females from 12 to 64 years old. They were recruited by an advertisement in a local newspaper. A gift was given to every respondent for participating in the study.

Stimuli were two new German pop songs interpreted by well-known performers comparable to Barry Manilow or Neil Diamond, and not yet released when the experiment was started.

FIGURE 1

DESCRIPTION OF MEASURES

Analysis Method

A structural equation modeling method (LISREL) is used to represent the constructs and test the hypothesis that instantaneous emotions are a relevant subcomponent of the Aad-construct To model the dynamic component, Aad is postulated BS a molecular structural construct. This follows the distinctive view of Bagozzi (1985) to differentiate between molar (global) and molecular (micro representation) level of abstraction The proposed model (see Fig. 2) employs higher-order latent variables within the contest of simultaneous equation systems. Thus Aad is modeled as a latent variable which has no direct measures but rather achieve meaning through their connection to lower-order latent variables which do have direct measures A static and a dynamic pleasantness subdimension represent the complex Aad-construct.

Furthermore the effect of Aad on performers' attitudes and purchase intention are considered In addition, the effects of familiarity on attitudes and intention are modeled (cf. Batra and Ray, 1985; Moore and Hutchinson, 1985).

For analysis performed at the level of individual respondents, responses on the program analyzer were calculated as a like/dislike index for every part of the songs representing 8 subtracted linear combination of the reaction scores for like, neutral, and dislike (e 8. Nunnally, 1978, p 152). After correcting for extreme-response tendency, the "normalized values" do not deviate much from normality, a notable fact for further calculations (typical skewness about -0.3 to 0.4, typical kurtosis in the range of -0.9 to 0.1).

FIGURE 2

DYNAMIC COMPONENT MODEL

FIGURE 3

REACTION PROFILES FOR SONG I

RESULTS

In early research as well as in actual commercial and noncommercial research studies, the usual way to analyze data of the program analyzer is restricted to descriptive statistical procedures (Peterman, 1940; Brockhaus and Irwin, 1958; Mielke and Chen, 1980) It is common practice to give the so-called "reaction profile" an aggregated score over all subjects for every second. In market research practice. the aggregated scores of reaction profiles are compared with the results from past experience and conclusions as above or below average are drawn.

TABLE 1

PARAMETER ESTIMATES

In this research, we are going one step further, looking for correspondence between variables on the individual person level of analysis. To achieve representative scores. the different parts of the songs are divided into refrain, verse, and instrumental segments In the more tender songs, as used here, instrumental segments are rare and verse and refrain dominate. This procedure differs from earlier research with musical stimuli, in which electronically generated sound sequences ant/or music such as Waltzes, Tangoes, Jazz. etc have been mainly used (Gabrielsson. 1973; Crozier. 1974).

For our songs. the over-all reactions to single segments indicate that the greatest acceptance for pop-songs is achieved if primarily the retrains and secondly the verses are liked For SI. 68 subjects (54%) liked the refrains Compared with SII. the songs with the lower purchase intention. especially the refrains and verses are noticeably disliked. Only the instrumental part of SII. a segment without performer specific potential and no singing etc . found agreement (Figure 3)

However, our central question remains open: can these like/dislike scores explain a significant portion of the variance in Aad? To discuss this question first let us consider the results for the model estimation (Table 1)

At first, the model as a whole is not disconfirmed by the data and can be "accepted." Chi values as well as the goodness of fit index (GFI) confirm the adequate fit of the model for both songs with pCvalues greater 0 10 or GFI values of at least 0.94. The model thus shows that Aad can be represented as a multidimensional molecular construct. In this respect the two subdimensions "static component" and "dynamic component" - are of particular interest. By "static" is meant that the traditional measurement procedures are circumscribed i.e. they are measured about ten minutes after hearing the songs. The "dynamic" subdimension is operationalized by instantaneous like/dislike reactions during exposure to the program. The resulting parameter estimates for the static component are quite high and significant (see rL) with value ranging from 0.83 and 0.92. for SI and SII, respectively.

Second, as hypothesized, the parameter estimates for the dynamic component are relatively high and significant, with values of 0.35 and 0.34. To test if this contribution to the variance in Aad is of relevance, an alternative model is assumed which postulates a zero influence These two restricted models yield in Chi2 - differences of 14.9 for SI (1 df) and 15.0 for SII (1 df). The 1088 in explanatory power for these restricted models is significant at the 0.0005 level. Thus, our hypothesis is strongly supported, and the dynamic component adds a significant portion to the variance in Aad

As a further result, a performers' familiarity and Aad seem not to be strongly interrelated for popmusic songs. Only for SI does a marginal intercorrelation exist between Aad and performers' familiarity. On the other side, familiarity (knowing the performer) has a positive effect on purchase intention for SI and a negative effect for SII. The former reflects high and the latter low popularity for the respective performers

These findings can be interpreted as a result of the situation of the performer of SII who changed his style in a time period of low popularity. A 3-way classification for Aad. attitudes towards performer and familiarity clarifies the situation. If Aad is negative. the familiarity with the performer has no significant effect on attitudes toward performer. But, if Aad is positive, familiarity with the performer becomes a central variable. Surprisingly 85% of the subjects with positive Aad and without knowing the performer also have a positive attitude towards the performer. On the other side, 54% of the respondents indicating a positive Aad but knowing the performer (which means high familiarity) expressed a weak attitude towards the performer They liked the new song spontaneously, but recalled the name of the performer during the questionnaire procedure and become aware of the actual low popularity of this performer - resulting in a subsequent bad evaluation.

Further, Aad has a highly significant direct effect on performers' attitudes and on purchase intention for both songs. And surprisingly enough, the direct effect of attitudes toward performers on purchase intention is Low for both songs and lower than the direct effect of Aad on intention to buy

CONCLUSIONS

Aad is considered as a mediating construct to explain stimulus effects on attitudes, intentions, and preferences Past research was concentrated on analyzing final Aad-states measures some minutes or longer after stimulus exposure. Analogous to persuasion research. where process measures like cognitive responses are used to analyze verbal processes that occur during stimulus exposure, a related process measure is used to analyze emotional Processes that occur during exposure - without the necessity of a verbal translation.

The typical like/dislike reactions using the program analyzer are interpreted as (dynamic) pleasantness responses. The possibility to analyze homogeneous subparts of the stimuli enables a detailed control which emotional-state is induced by which part of the stimulus Thus, by elimination of irritating and disliked scenes. the overall acceptance of a stimulus can be improved

The dynamic component in the Aad-construct, as outlined in this paper. could be shown to explain a significant portion of the variance in Aad. Thus, affective processes should gain our attention, when analyzing consumer's behavior.

REFERENCES

Aaker, D A , D M Stayman and M R. Hagerty (1986), "Warmth in Advertising Measurement, Impact and Sequence Effects," Journal of Consumer Research 13, (in press)

Bagozzi, R P (1982), " A Field Investigation of Causal Relations Among Cognitions, Affect, Intentions, and Behavior," Journal of Marketing Research 19, 562-584.

Bagozzi, R P (1985), " Expectancy-Value Attitude Models An Analysis of Critical Theoretical Issues," International Journal of Research in Marketing 2, 43-60.

Batra, R and M L Ray (1983), "Operationalizing Involvement as Depth and Quality of Cognitive Response," Advances in Consumer Research 10, R P Bagozzi and A M Tybout, Eds, Ann Arbor Association for Consumer Research 309-313.

Batra, R and M L Ray (1985), "How Advertising Works at Contact," In Psychological Processes and Advertising Effects Theory, Research, and Application L F Alwitt and A A Mitchell, Eds, Hillsdale N.J.: Lawrence Erlbaum 13-43.

Bower, G H (1981), " Mood and Memory," American Psychologist 36, 129-148.

Brockhaus, H H and J V Irwin (1958), "The Wisconsin Sequential Sampling Audience Analyzer," Speech Mono- graphs 25, 1-13.

Bryden, M P and R G Ley ( 1983), "Right Hemispheric Involvement in Imagery and Affect," In Cognitive Processing in the Right Hemisphere, E Perecman, Ed , New York et al Academic Press 111-123.

Buck, R (1982), " Spontaneous and Symbolic Nonverbal Behavior and the Ontogeny of Communication," In Development of Nonverbal Behavior in Children, S Feldman, Ed New York Springer 29-62.

Buck, R , R Baron and D Barrette ( 1982), "Temporal Organization of Spontaneous Emotional Expression A Segmentation Analysis," Journal of Personality and Social Psychology 42 506-517.

Clynes , M (1980), " The Communication of Emotion Theory of Sentics," In Emotion Volume 1 Theories of Emotion, R Plutchik and H Kellerman, Eds, New York et al Academic Press 271-301.

Crozier, J B (1974), " Verbal and Exploratory Responses to Sound Sequences Varying in Uncertainty Level," In Studies in the New Experimental Aesthetics, D E Berlyne, Ed, Washington D C Hemisphere Publishing 27-90.

Derbaix, C and P V Abeele (1985), "Consumer Inferences and Consumer Preferences The Status of Cognition and Consciousness in Consumer Behavior Theory", International Journal of Research in Marketing 2, 157-174.

Ekman, P and W V Friesen (1969), "The Repertoire of Nonverbal Behavior Categories, Origins, Usage and Coding," Semiotica 1 49-98.

Friestad, M and E Thorson (1986), "Emotion-Eliciting Advertising Effects on Long Term Memory and Judgement," Advances in Consumer Research 13, R Lutz, Ed, Provo UT Association for Consumer Research, 111-116.

Gabrielsson, A (1973), "Adjective Ratings and Dimension Analyses of Auditory Rhythm Patterns," Scandinavian Journal of Psychology 14, 244-260.

Gardner, M P , A A Mitchell and J E Russo (1985) "Low Involvement Strategies for Processing Advertisement", Journal of Advertising 14, No 2, 4-12.

Hill, R P and M B Mazis (1986), "Measuring Emotional Responses to Advertising," Advances in Consumer Research 13, R Lutz, Ed , Provo UT Association for Consumer Research, 164-169.

Holbrook, M B and S O'Shaughnessy (1984), "The Role of Emotion in Advertising", Journal of Psychology and Marketing 1, 45-54.

Isen, A M and T E Shalker (1982), "The Effect of Feeling State on Evaluation of Positive, Neutral, and Negative Stimuli When You "Accentuate the Positive," Do You "Eliminate the Negative"?," Social Psychology Quarterly 45, 58-63.

Izard, C E (1977), " Human Emotions,"New York et al Plenum Press.

Joreskog, K G and D Sorbom (1982), "Recent Developments in Structural Equation Modeling," Journal of Marketing Research 19, 404-416.

Kroeber-Riel, W (19797, "Activation Research Psychobiological Approaches in Consumer Research", Journal of Consumer Research 5, 240-250.

Kroeber-Riel, W (1984), "Konsumentenverhalten" (Consumer behavior), Munchen Vahlen (3rd ed).

Lazarus, R S (1982), " Thoughts on the Relations Between Emotion and Cognition," American Psychologist 37, 1019-024.

Leventhal, H (1982), " The Integration of Emotion and Cognition A View from the Perceptual-Motor Theory of Emotion," In Affect and Cognition, M S Clark and S T Fiske, Eds , Hillsdale Lawrence Erlbaum 121-156.

Levy, M R (1982), " The Lazarsfeld-Stanton Program Analyzer An Historical Note," Journal of Communi- cation, 30-38.

Lutz, R J (1985), " Affective and Cognitive Antecedents of Attitude Toward the Ad A Conceptual Framework," In Psychological Processes and Advertising Effects Theory, Research, and Application, L. F. Alwitt and A A Mitchell, Eds , Hillsdale N J Lawrence Erlbaum 45-63.

Mielke, X W and M Chen (1980), "Making Contact Formative Research in Touch with Children " Children's Television Workshop International Research Note Ga New York 9-14.

Mitchell, A A (1983), " Cognitive Processes Initiated by Exposure to Advertising," In Information Processing Research in Advertising, R J Harris; Ed , Hillsdale, N J Lawrence Erlbaum 13-42.

Mitchell, A A and J C Olson (1981), "Are Product Attribute Beliefs the Only Mediator of Advertising Effects on Brand Attitude?," Journal of Marketing Research 18, 318-332.

Moore, D L and J W Hutchinson (1985), "The Influence of Affective Reactions to Advertising Direct and Indirect Mechanisms of Attitude Change," In Psycho- logical Processes and Advertising Effects, L F Alwitt and A A Mitchell, Eds , Hillsdale N J Lawrence Erlbaum 65-87.

Neibecker, B (1984), "The Validity of Computer-Controlled Magnitude Scaling to Measure Emotional Impact of Stimuli", Journal of Marketing Research 21, 325-331.

Neibecker, B (1985), " Konsumentenemotionen Messung durch computergestutzte Verfahren", (Consumers' Emotion), Wurzburg-Wien Physica.

Neibecker, B (1986), "Processing Complexity in Magnitude versus Category Scaling", Advances in Consumer Research 13, R Lutz, Ed , Ann Arbor Association for Consumer Research, 286-290.

Nunnally, J C (1978), " Psychometric Theory," 2 Ed , New York et al McGraw-Hill.

Peterman, J N (1940), " The "Program Analyzer"," Journal of Applied Psychology 24, 728-741.

Remaining References Available from The Editors.

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