Priming Effects by Visual Image Information in On-Line Shopping Malls

Priming Effects by Visual Image Information in On-Line Shopping Malls Eun-young Kim*, Si-cheon You**, Jin-ryeol Lee*** *Chosun University Division of Design 375 Seosukdong Dong-gu Gwangju Korea, key1018@hanmail.net ** Chosun University Division of Design 375 Seosukdong Dong-gu Gwangju Korea, scyou@chosun.ac.kr *** Chosun University Division of Design 375 Seosukdong Dong-gu Gwangju Korea, bayhunt@chosun.ac.kr Abstract: "Priming Effect", as part of "Contextual Effect" is the phenomenon that pre-searched product information influences consecutive product evaluation. "Priming Effect" is different from other types of "Contextual Effects" in that pre-searched information doesn't have a direct effect on product evaluation, but an indirect effect which means that pre-searched information influences interpretation of consecutive product evaluation. On the previous theoretical background, this study investigated how visual images can influence consumer preference and product choice, introducing concept of priming effects into the design of on-line shopping malls. This experiment adopted the digital cameras as the experimental stimulus target products and designed the individual web pages by priming the attributes of price and size. In result of analysis, the difference of product preference by priming attributes was proved to be statistically significant. In this reason, suggesting the background of commercial web page can be the significant factor in choosing products, the priming effects introduced into the design can be positively employed in online advertising. Key words: Priming Effect, On-line Shopping malls, Visual Image 1. Introduction Internet users are growing alongside the spread of Internet use and the development of high-speed networks in recent years. The structuring of information infrastructure has led to a distinction between online and offline divisions in people s lives. The development of the Internet has changed people s lifestyles and brought many changes in the way they purchase products and do business, as represented by the high-growth posted by e-commerce every year. As on-line e-commerce continues to increase and more people use the Internet, much research has been conducted on consumers behaviors on on-line networks. This study is an example, focusing on how consumers preference and choice of products in on-line shopping malls are influenced by the priming effects of visual images. As part of contextual effect, priming effect is the phenomenon wherein the proceeding information affects the interpretation of how products are evaluated. It differs from contextual effect since it is not the actual information about the product but the interpretation of the information that directly influences its evaluation [1]. This research aims to introduce the concept of priming effects to design, determine how visual images presented in commercial websites influence the preference and choice of products, and how these can be

applied to designing on-line shopping malls. 2. Theoretical Background 2.1 Definition of Priming Effect The term priming used in psychology refers to various phenomena occurring under an identical mechanism. If one is exposed to a certain stimulus, this increases the access to certain information already existing in one s memory [2]. Prime effect, in case of an ambiguous interpretation of the information, is the phenomenon wherein a particular concept provoked by the preceding information affects the interpretation of the ensuing information. Previously presented information is referred to as priming stimulus while the ensuing information of an ambiguous interpretation is referred to as target information. According to studies on priming effect, the evaluator s interpretation of the target information is affected by what kind of priming stimulus he or she is exposed to before evaluating the ambiguous target information [3]. In the research by Yi (1990), each evaluator was shown articles on either safe flight or oil before he or she was shown an ad containing information that a car is big. Evaluators who read the flight article got the safety concept and consequently interpreted the target car as a safer car, while those who read the articles on oil interpreted the target car as consuming much oil [4]. The differences in interpreting information affected the evaluation and purchase of products. Readers of the safe flight article compared with those of the oil article showed more favorable attitude and high purchase intention. Thus, contextual elements enable certain products to be accessed by consumers exposed to relevant ads and lead consumers to perceive information that affect their evaluation of them. Likewise, overall evaluation of products is improved by priming positive attributes of the products. As Yi s research showed, different effects can follow depending on what attributes are primed, even with the same ad. This priming effect is grouped into two categories: assimilation effect and contrast effect [5]. Assimilation effect occurs when the priming results in an interpretation of new information that matches the preconceived concept. Contrast effect is the result of interpreting information based on a new concept used as the basis for judgment [6]. 2.2 Mechanism of Priming Effect Priming effect occurs when categorization, i.e., the cognitive course unconsciously occurring in interpreting information, is affected by the succeeding information. Categorization also refers to automatically grouping and storing similar information unconsciously stored in our head as we interpret information. As such, if the target information presented as a priming effect is categorized as a specific group in consumers heads, several pieces of stored information related to such categorization are activated. The evaluator uses knowledge within the category in evaluating the target information. A schema is a network-type knowledge structure related to descriptive knowledge such as eventual knowledge and semantic knowledge after consumers store information related to specific products in a type of network association. This is a cognitive structure of a stimulus concept that includes stimulus attributes and relations with attributes. For instance, the trademark knowledge of a specific product is a schema on the trademark. Consumers have diverse consumption environmental schemas such as product categories, trademarks, and stores. Thus, an activation of these cognitive schemas greatly influences consumers purchase and consumption activities, with the prime effect enhancing their accessibility and affecting consumers judgment.

2.3 Priming Effect in Web Pages The research by Johnson and Mandel (1999) manipulated background screens in the first few pages of online shopping malls using money and cloud and primed prices and quality. People exposed to price priming stimulus, compared with those exposed to quality priming stimulus, preferred and chose low-quality yet low-priced products[7]. They also preferred exploring priming stimulus-related attributes than the other attributes. This research verified the benefit of creating priming effects even on online channels. It contested the general belief that in the online environment, due to the characteristics of the Internet s interactive media, consumers enhance their decision-making ability and control their exposure to surrounding stimuli; thus minimizing the priming effect. Kwak Jun-sik, in his research (2001), indicated the possibility that the experiment conducted by Johnson and Mandel (1999) directly exposed priming stimuli; hence allowing participants to know the experimenter s intention. He used a subliminal method wherein evaluators were unable to know which stimuli they were exposed to; thus adopting the priming stimuli [8]. The abovementioned studies were experiments that forecast the assimilation effect using the priming stimuli. They also examined how the priming effect influences product preference in online shopping malls, presupposing the creation of assimilation effect. Unlike these studies by Johnson and Mandel and Kwak Jun-sik, however, this research seeks to use visual images in the background of products to induce priming. 3. Experiments 3.1 Stimulus Design This experiment research has the purpose of confirming the priming effects by visual image information in on-line shopping malls using experiment stimulus. For this research, we adopted 3-4 million pixel digital cameras as the experimental stimulus target products. These are also new models of a specific brand: camera A is a high-priced product yet very portable with a compact design and camera B is a low-priced product but less portable. The experiment sought to find out whether the two products characteristics of size very portable and compact design and price high or low price have a priming effect. It first designed a general page, shown as Figure 1, not primed by any elements. Another page was then designed that used images and presented the prices and sizes of the two products as primed attributes. On the page with the primed price, shown as Figure 2, images of paper money and coins associated with price were included in the design, with the prices determined in a pre-survey [Appendix A]. The other page, shown as Figure 3, was a prime- sized page that used an image of a hand-held camera, with its actual size to be determined by the reader. Table1. Stimulus Type Prototype Stimulus 1 Stimulus 2 Camera A (Type A') Type A' Price Priming (Type A'-1) Size Priming (Type A'-2) Camera B (Type B') Type B' Price Priming (Type B'-1) Size Priming (Type B'-2)

Type A' Fig.1 Prototype (Non Priming) Type B' Type A'-1 Fig.2 Stimulus 1 (Price Priming) Type B'-1 Type A'-2 Fig.3 Stimulus 2 (Size Priming) Type B'-2

3.2 Surveys Method The experiment focused on the non-primed page (Type A' and Type B'), the price-primed page (Type A'-1 and Type B'-1) and the size-primed page (Type A'-2 and Type B'-2). Six groups of 30 people or a total of 180 people participated as evaluators. These were actually college students who belong to the generation with a high interest in digital cameras. Each group was shown a designed web page and allowed to choose to answer five questions. These five questions were directly borrowed from the five questions used by Peracchio and Meyer-Levy (1997) aimed at measuring responses to advertising. Measurements were classified as: very low quality/very high quality, low product value/high product value, not worth purchasing/worth purchasing, not favorite at all/very favorite, and not a good product at all/very good product. They were based on the seven-point Likert scale (1= not at all, 7= very much so). [9] In advance, a validity test was required since measurements were based on multi-items. On the five questions, an internal consistency test was given by conducting Cronbach's α test and a validity test; thus resulting in Cronbach's α value as Type A' (α=0.9019), Type A'-1 (α=0.9031), and Type A'-2 (α=0.8746) in the camera A questionnaire. Likewise, the value for the B camera questionnaire was Type B' (α=0.9031), Type B'-1 (α =0.8729), and Type B'-2 (α=0.8271). α value, if above 0.6, is generally valid in social sciences. The six experimental stimuli all had above 0.8 values, which is considered as highly valid measurements. A one-way ANOVA was conducted on the average of the five items. 3.3 Results With the results shown in Figure 4, it showed the high-priced yet very portable and compact-designed camera A had its price-primed page (Type A'-1, M=4.3667) earning a low preference compared with the not-primed page (Type A', M=4.84) and the size-primed page (Type A'-2, M=5.0533). Likewise, the size-primed page (Type A'-2, M=5.0533) earned a significantly higher preference compared with the not-primed page (Type A', M=4.8400) and the price-primed page (Type A'-1, M=4.3667). (F(2,87)=3.164, p<.05) Shown in Table 2 is the ANOVA of the analysis of evaluation values of the three groups surveyed regarding camera A. 5.2 5.2 5 4.8 4.84 5.0533 5 4.8 4.9467 4.6 4.4 4.3667 4.6 4.4 4.5133 4.38 4.2 4.2 4 TypeA' TypeA'-1 TypeA'-2 (Non-Priming) (Price Priming) (Size Priming) 4 TypeB' TypeB'-1 TypeB'-2 (Non-Priming) (Price Priming) (Size Priming) Fig.4 Camera A s Preference changes by Priming factors Fig.5 Camera B s Preference changes by Priming factors

Table2. Analysis Results of Three Groups on Camera A s Web Page (ANOVA Table) Between Groups Within Group Total Sum of Squares df Mean Square F Sig 7.411 2 3.705 3.164. 047 101.873 87 1.171 109.284 89 Table3. Analysis Results of Three Groups on Camera B s Web Page (ANOVA Table) Between Groups Within Group Total Sum of Squares df Mean Square F Sig 5.267 2 2.633 2.783. 067 82.317 87.946 87.584 89 Figure 5 presented the results for the low-priced yet less portable camera B. As shown in the graph, the price-primed page (Type B'-1, M=4.3667) earned a higher preference compared with the not-primed page (Type B', M=4.5133) and the size-primed page (Type B'-2, M=4.38). The size-primed page (Type B'-2, M=4.38) earned a lower preference compared with the not-primed page (Type B', M=4.5133) and the price-primed page (Type B'-1, M=4.9467). The significance in the ANOVA, presented in Table 3, registered 0.067; thus showing significance within the validity level of 90 percent. (F(2,87)=2.783, p<.1) Shown in the Table 3 is the ANOVA table analyzing the evaluation values of the three groups that responded to camera B s questionnaire. 3.4 Discussion Based on the results of the experiment, camera A earned a low preference on the price-primed web page (Type A'-1) since it was perceived to be expensive. On the other hand, it rated a high preference on the size-primed web page (Type A'-2). In contrast, camera B registered a high preference on the price-primed web page (Type B'-1) but had a low preference on the size-primed web page (Type B'-2) since it was perceived to have an inconvenient size. As presupposed before the experiment, the two types demonstrated assimilation effect and contrast effect due to primed attributes; hence revealing a significant preference difference. As seen in the above analysis of the results, web page advertising can be effective when using positive attributes of a product to prime positive stimulus for consumers; thus enhancing overall preference for the product. Likewise, the evaluation of target information changes depending on what attributes the priming stimulus activates. 4. Conclusions This research verified the priming effect of information on the visual image of the background screen of a web page in choosing products in online shopping malls, with the result proving to be statistically significant. Research on the priming effect on consumers behavior in choosing products continues in the fields of marketing and social psychology, and in recent years, in online channels as well. This particular research is significant since it positively introduced priming effects to the field of design to activate information inherent in the memory of consumers when they visit commercial web pages; hence influencing their evaluation of products and their intention to buy them. Priming effects can be positively employed in online advertising. Future research needs to pay keener attention to whether evaluators in the experiment have varying degrees of

involvement. This was not clarified in this research. Including evaluators accessibility to information on products as a component will make the findings more accurate. This entails evaluators participating as either a low involvement or high involvement group to ensure validity. References 1. Yoon Sung-A. The Study of Priming Effect on Consumers evaluation: Verification of alternative Mechanism. Thesis of Masters Degree, Korea University. (1998) 2. Naomi Mandel, Eric J. Johnson. When Web Page Influence Choice: Effects of Visual Primes on Experts and Novices. Journal of Consumer Research, vol.29 p235-245 (2002) 3. Yoon Sung-A. op. cit 4. You-jae Yi. Cognitive and affective priming effects of the context for print advertisements. Journal of Advertising, Spring v19 n2 p40(9) (1990) 5. The terms Assimilation and Contrast Effects in priming effect were firstly mentioned in the study of Paul, Steven & Russell (1983) - Paul M. Herr, Steven J. Sherman, and Russell H. Fazio. On the Consequences of Priming: Assimilation and Contrast Effects. Journal of Experimental Social Psychology, (July) v19, p323-340(1983) 6. Choi Se-Jun. When Priming Influence Consumers Choice: Recognizable and Important Role of Priming. Thesis of Masters degree, Korea University. (1997) 7. Naomi Mandel, Eric J. Johnson. op. cit 8. Gwak Joon-Sik. The Influence of Priming Effect on On-line Product Choice. Thesis of Masters degree, Korea University. (2001) 9. Meyers-Levy, Joan and Laura A. Peracchio. Understanding the Effects of Color : How the Correspondence between Available and Required Resources Affects Attitudes. Journal of Consumer Research, 22 (September), 121-138. (1995) Appendix A The pre-survey was conducted to set the price of the products. It involved 20 college and graduate school students and their perception on how much cameras A and B cost. The questions were: What price would be suitable for this camera?, What price do you think would be considered expensive for this camera? and What price do you think would be considered cheap for this camera? The prices suggested ranged from under KRW200,000 to over KRW2 million. The finding showed that an average price of KRW890,000 was considered expensive for camera A and KRW395,000 for camera B. Accordingly, these prices were set for cameras A and B. Product description for the price determination is as Figure 6

Survey (Camera A) This Survey is designed to examine consumer s reaction prior to its brand new release. Your sincere cooperation is immensely appreciated. The description of new model is as follows; Survey (Camera B) This Survey is designed to examine consumer s reaction prior to its brand new release. Your sincere cooperation is immensely appreciated. The description of new model is as follows; Effective pixels : 4.13 million Weight/Size : 230g / 95x69x52mm(pocket size) Function : Movie, Multi-Shot 16, 16M(CF)Card Produced by NICON / Made in Japan Effective pixels : 33.4 million Weight/Size : 360g / 130x73x50mm(Normal size) Function : Movie, Multi-Shot 16, 16M(CF)Card Produced by NICON / Made in Japan Fig.6 Product Description on Pre-Survey