The Effect of Background and Ambient Color on Product Attitudes and Beliefs

ABSTRACT - The process by which attitude change occurs is a key issue in consumer research. This study addresses this theoretical question by examining the effect of background color on product attitudes and beliefs. Background color was used as the independent variable because, like music, it appears to be an aspect of the communication and exposure situation which carries little explicit information about the product. Hence, many would argue that differences in product attitude produced by background color occur via and are evidence for nonbelief-based change processes. In this study, respondents presented with a slide of a pen against a blue background exhibited a more positive attitude toward buying the product than those shown the same pen against a red background. However, beliefs as well as attitudes were measured and it was possible to demonstrate that those in the blue condition also held different beliefs about buying the product. Thus, although background color would seem, on the surface, to carry no product information, the background color affected the underlying belief structure as well as the attitude. It is argued that belief-based processes may be responsible for attitude change previously assumed to be due to affective processes.

Citation:

Susan E. Middlestadt (1990) ,"The Effect of Background and Ambient Color on Product Attitudes and Beliefs", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 244-249.

Advances in Consumer Research Volume 17, 1990      Pages 244-249

THE EFFECT OF BACKGROUND AND AMBIENT COLOR ON PRODUCT ATTITUDES AND BELIEFS

Susan E. Middlestadt, University of Illinois, Urbana-Champaign

ABSTRACT -

The process by which attitude change occurs is a key issue in consumer research. This study addresses this theoretical question by examining the effect of background color on product attitudes and beliefs. Background color was used as the independent variable because, like music, it appears to be an aspect of the communication and exposure situation which carries little explicit information about the product. Hence, many would argue that differences in product attitude produced by background color occur via and are evidence for nonbelief-based change processes. In this study, respondents presented with a slide of a pen against a blue background exhibited a more positive attitude toward buying the product than those shown the same pen against a red background. However, beliefs as well as attitudes were measured and it was possible to demonstrate that those in the blue condition also held different beliefs about buying the product. Thus, although background color would seem, on the surface, to carry no product information, the background color affected the underlying belief structure as well as the attitude. It is argued that belief-based processes may be responsible for attitude change previously assumed to be due to affective processes.

One of the ongoing questions within the behavioral sciences concerns the nature of the processes by which attitudes, intentions and behaviors change. On the one hand, some theorists argue that attitudes, intentions and behaviors are determined by underlying beliefs and that change ultimately occurs by changing this cognitive structure. This belief-based approach has been most clearly articulated by Fishbein and Ajzen (1975). As applied to the attitude toward a behavior, these theorists propose the following compensatory expectancy-value model:

A(behavior) = fn (Sum(Bi x Ei)), for i from l to n salient consequences of performing the behavior, where A(behavior) is the attitude toward the behavior, Bi is the strength of the belief that performing the behavior will lead to the ith consequence, Ei is the evaluation of this outcome and a belief times evaluation cross-product is formed for each of n salient consequences of performing the behavior.

Notice that attitude is determined by the totality of the evaluative implications of an underlying cognitive structure, here expressed as a sum of belief times evaluation cross-products. Furthermore, according to this model, the underlying beliefs function not merely as predictor variables, but as causal ones. Thus, a change in attitude must be mediated by a change in this sum of cross-products which could occur by changing the belief strengths linking the behavior to its consequences, by changing the evaluation of the outcomes, and/or by changing the set of salient outcomes.

In contrast to this highly cognitive approach, many theorists argue that it is necessary to recognize an additional process of attitude and behavior change, one which is not mediated by belief change, one which is more affective and less cognitive. This "other" process takes many forms from theory to theory. For example, Petty and Cacioppo (1981) contrast peripheral from central routes to persuasion. Zajonc and Markus (1982) argue that, under certain circumstances, affective reactions occur which precede and are independent of cognition. And, a number of researchers (Mitchell and Olson, 1981; Gresham and Shimp, 1985; Lutz, MacKenzie and Belch, 1982) suggest that the affective response to the ad might be directly transferred to the product without changing the underlying beliefs.

Evidence for an "other than belief-based" process of attitude change has come from a wide range of research. Here, it will be argued that the possible role of belief-based change as a mechanism for attitude change has frequently been neglected in this work. For example, in his study of the effect of music on choice behavior, Gorn (1982) presented a product with liked or disliked music. He found that the product associated with the liked music showed a significantly higher preference in comparison to the product associated with the disliked music. Since "minimal product information" was presented in the exposure situation and since few beliefs about the product were elicited to a free response question about the reason for choosing the product, Gorn argued that the change in behavior was accomplished via some process other than a belief-based one and that the positive affect toward the music directly transferred to the product. Gorn suggested that the effect of music on choice occurred by way of classical conditioning.

This study stimulated much debate primarily about whether, when and how classical conditioning processes occur in the advertising and consumer behavior domain (e.g., Allen and Madden, 1985; McSweeney and Bierley, 1984; Stuart, Shimp and Engle, 1987). Here, instead, the very existence of direct transfer of affect will be questioned. That is, it will be argued that music may have affected choice by changing some aspect of the underlying cognitive structure. Although the presentation of the product with background music contained no explicit verbal information about the product, it is possible that the liked music caused the respondents to form or change beliefs about the product which led to a more positive attitude in comparison to the disliked music.

Can background music during a product presentation affect the consumer's beliefs about the product? More generally, do aspects of the exposure situation which provide no explicit verbal information about the product have an effect on the cognitive structure underlying product attitudes? An answer of yes implies that belief-based change processes may be responsible for some of the changes previously assumed to be affect-based. Evidence for differences in cognitive structure as a function of these presumably "informationless" manipulations would suggest that it is premature to rule out a belief-based explanation for change. The purpose of this study is to demonstrate that aspects of the communication and exposure situation which, on the surface, do not appear to have the capacity of changing underlying cognitive structure may in fact do so. In another paper, the effect of music on beliefs has been demonstrated (Middlestadt, 1988). Here, the color associated with the product is shown to have an effect on both the attitude toward buying the product and the cognitive structure underlying this attitude.

Color is an important variable in a wide range of applied and theoretical disciplines and has been argued to produce a variety of sensory, perceptual, cognitive and affective effects. Briefly, from an applied perspective, scholars are concerned with how color affects the aesthetic experience (Ball, 1965) and with color as an element in design (Sharpe, 1974) and photocommunication. In the consumer domain, the focus is on the use of color in packaging, retail design and display (Bellizzi, Crowley, and Hasty, 1983) its impact on advertising (Warner and Franzen, 1947; Dunlap, 1950; Starch, 1959). Color is also the subject of theoretical research. For example, at the physiological level, color has been shown to affect arousal (Wilson, 1966), leg strength (Pellegrini, Schauss and Birk, 1980), performance of simple psychomotor skills (Nakshian, 1964) and perception of time (Smets, 1969). Within the learning theory tradition, there is evidence that color connotations can be conditioned to terms with which they are associated (Harbin and Williams, 1966) and that color preferences can be changed with simple learning trials (Peters, 1943). There is a long history of assumed association between color and feeling, emotionality or affectivity (Pressey, 1921; Norman and Scott, 1952). Within the psychodynamic tradition, Schachtel (1943) argues that responses to color are similar to responses to affect in that they are immediate, direct and evoked rather than deliberated or mediated. There is evidence that color affects subjects' reports of moods (Rosenshine, 1985; Levy, 1980). Furthermore, although there are cultural, developmental and individual differences, colors and color words have been shown to be consistently associated with mood-tones (Hevner, 1935; Odbert, Karwoski, and Ekerson, 1942; Wexner, 1954; Murray and Deabler, 1957; Schaie, 1961) and affective meanings (Adams and Osgood, 1973)

Two points from this varied work are relevant to the task of examining the processes by which attitude change occurs. First, like music, color would be expected to affect how people respond to products. And, like background music, background color does not convey explicit product information but seems to represent affectivity and emotion. Thus, like a music manipulation, a color manipulation would qualify as a relatively noncognitive one. Second, while the specific findings about color effects are complex and vary from study to study, red is consistently found to be different from the cool colors of blue and green. Thus, contrasting red with blue would appear to be the most effective experimental manipulation.

In the present study, three products were presented against blue and red backgrounds. It was expected that a difference in attitude toward buying the product would be found between these two color conditions. More importantly, in view of the comparison of affective and belief-based models of attitude change, it was expected that if a difference in attitude occurred, it would be accompanied by a difference in cognitive structure. Note that focus of the study is not on the effects of color per se but rather on the process by which color produces attitude change.

METHOD

The subjects in the study were 84 female undergraduate students from an introductory advertising course at a major midwestern university. These students were majors from a wide variety of departments and colleges within the university and participated in partial fulfillment of a course research requirement.

The experimental manipulation. Subjects were randomly assigned to one of two experimental conditions, red or blue. Those in the red condition viewed slides containing one of three products (a bottle of perfume, a gold and silver pen and a bottle of mineral water) against a red background and those in the blue condition- viewed slides of the same products against a blue background. The light from the slide was not sufficient to read the questionnaire. Therefore, lights in a color matching the background of the slide were projected onto the ceiling at the front of the room. Thus the color manipulation was of both the background color on the slide as well as ambient color, that is, the surrounding color in the room. It should be noted that the room was kept relatively dark and that no attempt was made to control the luminance of the projected light.

The questionnaire. For each product, attitude toward buying the product was assessed directly with six bipolar semantic differential items. For example,

        My buying the pen pictured in the slide is

good   _:        _:      _:      _:        _:        _:       _   bad

      extreme quite slight. neith. slight. quite extrem.

Underlying cognitive structure was measured as recommended by Ajzen and Fishbein (1980). More

TABLE 1

EFFECT OF COLOR ON ATTITUDE TOWARD BUYING PRODUCT: MEAN ON ATTITUDE ITEMS BY CONDITION

specifically, belief and evaluation items were written for eight modal salient outcomes elicited in a pretest study. For the belief item, each respondent indicated on a seven-point bipolar scale, how likely or unlikely the outcome would be if they bought the product. For the evaluation item, they indicated on a seven-point bipolar scale, how good or bad the outcome was.

The procedure. The subjects arrived at the experimental room in groups of about 6 to 10. Under regular light, they read the instructions which described how to fill out the scales and indicated that the purpose of the study was to determine the effect of color on sensory processes in the eye. Then the regular room light was turned off and blank colored slides were projected on the screen and the ceiling. To allow the subjects to adjust to the color, they completed three pages of filler questions. Then the slide with the perfume on the colored background was projected on the screen and subjects indicated their attitude toward buying the perfume and completed the questions on the underlying cognitive structure. The slide was changed to the pen and subjects completed attitude and cognitive measures, this time with respect to buying the pen. Finally, the slide of the mineral water was projected and respondents answered attitude and cognitive questions with respect to buying the mineral water. Throughout the session, the color on the background of the slide and in the room was red for those in the red condition and blue for the blue condition. The session took approximately 25 minutes.

RESULTS

There were no significant effects of the color manipulation on attitudes or beliefs with respect to two of the products, the perfume and the mineral water. Therefore, this report focuses on the effects of color on the respondents' ratings of the pen. The results for this product are presented in three sections: (1) attitude toward buying the pen; (2) overall evaluative implications of underlying cognitive structure; and (3) individual belief and evaluation aspects of the salient outcomes comprising the cognitive structure.

Effect of color on attitude toward buying the pen. The top of Table 1 presents the means within experimental condition for the six semantic differential items used as direct measures of attitude. Over the six items, the multivariate F (Approx. F (6,77) = 3.21) was statistically significant at beyond the .01 level. For each of the items, respondents who were exposed to the pen with the blue background and ambient color were more positive than those exposed to the pen with the red background and ambient color. The mean difference was statistically significant at beyond the .01 level for two of these items, pleasant-unpleasant and enjoy able-unenjoyable.

Thus, color had a significant effect on attitude toward buying the pen pictured in the slide. Compared to the red condition, those exposed to a pen against a blue background with blue ambient color held a more positive attitude toward buying the product. It is next reasonable to ask whether this effect could have been mediated by an effect on the underlying cognitive structure.

Effect of color on overall evaluative implications of cognitive structure. It will be recalled that cognitive structure underlying the attitude toward buying the product was assessed according to Fishbein and Ajzen (1975). That is, for each of eight salient consequences of performing the behavior, a belief item assessed the likelihood

TABLE 2

EFFECT OF COLOR ON COGNITIVE STRUCTURE UNDERLYING BUYING PRODUCT: MEAN ON BELIEFS AND EVALUATIONS BY CONDITION

that this outcome would result if the pen were bought and an evaluation item assessed the evaluation of the outcome. For each outcome, the belief and evaluation aspects were scored from +3 to -3 and multiplied to create a cross-product. The sum of these eight belief times evaluation cross-products provides an indirect assessment of attitude and represents the evaluative implications of the underlying cognitive structure.

First, as support for the assumption that the sum of the cross-products represents an indirect attitude measure consider the correlations of this measure with the individual semantic differential items. These correlations range from a high of .59 for the good-bad, pleasant-unpleasant and like-dislike scales to a low of .45 for wise-foolish. These correlations are all statistically significant and about the size typically found using these types of measures.

Second, did the experimental manipulation of background color affect the overall evaluative implications of the cognitive structure underlying the attitude toward buying the product? The mean sum of cross-products for each color group is presented at the bottom of Table 1. Again, those who were exposed to the pen in the blue condition were more positive toward buying the pen than those exposed to the pen in the red condition. And again, this difference was significant at beyond the .01 level (F (1,82) = 63.57). This difference implies that the cognitive structure of those in the blue condition had significantly more positive evaluative implications than that of those in the red condition. In other words, compared to those exposed to the pen against the red background, those who viewed the pen against the blue background perceived that buying the pen was likely to lead to positive outcomes and unlikely to lead to negative ones. At this point, it is important to determine which outcomes were affected by the color manipulation.

Effect of color on belief and evaluation of individual outcomes. Multivariate analyses of variance on the individual outcomes revealed a significant effect of color over the eight beliefs (Approx. F (8,75) = 3.14) at beyond the .01 level, but no significant effect over the eight evaluations (Approx. F (8,75) = 1.32). Table 2 presents the means on these items for each color group, the beliefs at the top of the table and the evaluations at the bottom.

The background and ambient color had the strongest effect on two beliefs, buying the pen means-buying something which is elegant and which is unique. Compared to those in the red condition, those in the blue condition believed it more likely that the pen was elegant as well as unique. More specifically, those who were exposed to the pen with the blue background believed that buying the pen would quite likely lead to buying an elegant pen; whereas, those who were shown a pen again a red background perceived this outcome as neither likely nor unlikely. And, those who viewed the pen on a blue background indicated that it would be slightly likely that buying this pen would mean buying a unique pen; whereas those exposed to it against the red background believed that this outcome was slightly unlikely to occur.

Marginal effects of color are evident on two beliefs, buying the pen means buying something which is good quality and which is a pleasant color. Again, those exposed to the pen against the blue background believed that these two positive outcomes were be more likely to occur than those exposed to the pen against the red background. Although not statistically significant, it is interesting to note that the pen presented in the blue condition was perceived to be more likely to be expensive (overall, a slightly negative consequence) than the pen presented in the red condition. Thus, exposure to the pen in the blue condition did not invariably lead to more positive evaluative implications.

The effect of the color manipulation on the individual evaluations is presented at the bottom of the table. Consistent with the multivariate F reported above, for none of the eight consequences did the background and ambient color significantly affect the evaluation of the outcome. That is, those exposed to the pen in the blue condition did not evaluate the outcomes significantly more or less positively than those in the red condition. The lack of significant effects on the evaluations is not surprising. One might expect that a person's favorability toward buying products with attributes, such as uniqueness, elegance and durability would be more stable and difficult to change than the beliefs about whether buying a specific product will lead to obtaining these outcomes.

CONCLUSION

In this study, it was shown that an aspect of the exposure situation, the background color, had an effect on attitude toward buying a product. Subjects exposed to a pen presented against a blue background with blue ambient color evaluated buying the pen more positively than those exposed to the same pen with a red background and ambient color.

From the perspective of understanding attitude change processes, the key issue is determining how color produced this effect on product attitudes. Since the background color did not explicitly convey any product information, it might be argued that the affect associated with the color was directly transferred to the product and thus created a positive attitude through some other than belief-based process. However, in this study, the underlying cognitive structure was measured in detail and it was possible to show that the color affected the respondents' beliefs about buying the product. Compared to those in the red condition, those who were exposed to the pen in the blue condition believed more strongly that buying the pen would lead to two positive outcomes, buying a pen which was elegant and which was unique. This difference in beliefs resulted in a cognitive structure which was more positive for those exposed to the pen against a blue background. Thus, it could be argued that the change in attitude toward buying the pen found in this study was caused by a belief-based change process.

These data do not prove that a cognitive difference mediated the attitude and behavior effects found here or in other research. As argued by Gorn (1982), it is possible that the change in the beliefs occurred after and in justification of the change in attitude. And, disentangling the causal relations among beliefs and attitudes is unlikely to be an easy task. However, the data presented here do demonstrate that-subtle differences in how a product is presented, differences which do not appear to involve product information,-do produce differences in beliefs about the product. Thus, belief-based change remains a viable explanation for the observed changes in product attitude.

REFERENCES

Adams, Francis, M. and Charles E. Osgood (1973). "A cross-cultural study of the affective meanings of color," Journal of Cross-Cultural Psychology, 4, 135-156.

Ajzen, Icek and Martin Fishbein (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall.

Allen, Chris T. and Thomas J. Madden (1985). "A closer look at classical conditioning," Journal of Consumer Research, 12, 301-315.

Ball, Victoria (1965). "The aesthetics of color: A review of fifty years of experimentation," Journal of Aesthetics and Art Criticism," 23, 441452.

Bellizzi, Joseph, A., Ayn E. Crowley and Ronald W. Hasty (1983). "Color effects in store design," Journal of Retailing, 59, 1, 21-45.

Dunlap, J. William (1950). "The effect of colo in direct mail advertising," Journal of Applied Psychology, 3, 280-281.

Fishbein, Martin and Icek Ajzen (1975). Belief, attitude, intention and behavior. Reading, Mass: Addison-Wesley.

Gorn, Gerald J. (1982). "The effects of music in advertising on choice behavior: A classical conditioning approach," Journal of Marketing, 46 (Winter), 94-101.

Gresham, Larry G. and Terence A. Shimp (1985). "Attitude toward the advertisement and brand attitudes: A classical conditioning perspective," Journal of Advertising, 14, 10-17.

Harbin, Susan P. and John E. Williams (1966). "Conditioning of color connotations," Perceptual and Motor Skills, 22, 217-218.

Hevner, Kate (1935). "Experimental studies of the affective value of color and lines," Journal of Applied Psychology, 19, 385-398.

Levy, Bernard, I. (1980). "Research into the psychology meaning of color," American Journal of Art Therapy, 19, 87-91.

Lutz, Richard J., Scott B. MacKenzie and George E. Belch (1983). "Attitudes toward the ad as a mediator of advertising effectiveness: Determinants and consequences," in Advances in Consumer Research, Richard P. Bagozzi and Alice M. Tybout (eds.), 10, 532-539.

McSweeney, Frances K. and Calvin Bierley (1984). "Recent developments in classical conditioning," Journal of Consumer Research, 11, 619-631.

Middlestadt, Susan E. (1988). "The effect of music on brand attitudes: Affect or belief-based change." Working paper. University of Illinois.

Mitchell, Andrew A. and Jerry C. Olson (1981). "Are product attribute beliefs the only mediator of advertising effects on brand attitude?" Journal of Marketing Research, 18, 318-332.

Murray, David C. and Herdis L. Deabler (1957). "Color and mood-tones," Journal of Applied Psychology, 41, 279-283.

Nakshian, Jacob, S. (1964). "The effects of red and green surroundings on behavior," Journal of General Psychology, 70, 143-161.

Norman, Ralph D. and William A. Scott (1952). "Color and affect: A review and semantic evaluation," Journal of General Psychology, 46, 185-223.

Odbert, H. S., F. Karwoski and A. B. Ekersen (1942). "Studies in synesthetic thinking: I. Musical and verbal associations of color and mood," Journal of General Psychology, 26, 153173.

Pellegrini, Robert J., Alexander G. Schauss and Thomas J. Birk (1980). Leg strength as a function of exposure to visual stimuli of different hues," Bulletin of the Psychonomic Society, 16, 111-112.

Peters, Henry, N. (1943). "Experimental studies of the judgmental theory of feeling: V. The influence of set upon the affective value of color," Journal of Experimental Psychology, 33, 285 -298.

Petty, Richard E. and John T. Cacioppo (1981). Attitudes and persuasion: Classical and contemporary approaches. Dubuque, Iowa: Wm. C. Brown.

Pressey, S. L. (1921). "The influence of color upon mental and motor efficiency," American Journal of Psychology, 32, 326-356.

Rosenstein, L. D. (198' ). "Effect of color of the environment on task performance and mood of males and females with high and low scores on the Scholastic Aptitude Test," Perceptual and Motor Skills, 60, 550.

Schachtel, Ernest G. (1943). "On color and affect: Contributions to an understanding of Rorschach test," Psychiatry, 6, 393-409.

Schaie, K. Warner (1961). "Scaling the association between color and mood tones," American Journal of Psychology, 74, 266-273.

Sharpe, Deborah T. (1974). The psychology of color and design. Chicago: Nelson-Hall Company.

Smets, Gerda (1969). 'Time expression of red and blue," Perceptual and Motor Skills, 29, 511-514.

Starch, Daniel (1959). "How size, color, position and location affect inquiries," Mediascope, 3, 23-27.

Stuart, Elnora W., Terence A. Shimp and Randall W. Engle (1987). Classical conditioning of consumer attitudes: Four experiments in an advertising context," Journal of Consumer Research, 14, 334-349.

Warner, Lucian and Raymond Franzen (1947). "Value of color in advertising," Journal of Applied Psychology, 31, 260-270.

Wexner, Lois B. (1954). 'The degree to which colors (hues) are associated with mood-tones," Journal of Applied Psychology, 38, 432-435.

Wilson, Glenn D. (1966): "Arousal properties of red versus green," Perceptual and Motor Skills, 23, 947-949.

Zajonc, Robert B. and Hazel Markus (1982). "Affective and cognitive factors in preferences," Journal of Consumer Research, 9, 123-131.

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