Testing the Effects of Incidental Advertising Exposure in Online Gaming Environments

Adam Acar

University of Connecticut


This study addresses incidental advertising exposure effects in online games by manipulating both the location (low proximity versus high proximity) and the message content (visual versus verbal) of an incidental cue while participants’ primary attention is diverted by playing an online game. The results reveal that incidental exposure can be effective if the secondary stimulus appears very close to the focal attention area. The study also finds no incidental effects of text inside bars or images in address bars.


In 2005, advertisers spent $80 million to reach video game players; this spending is expected to top $400 by 2009 (Park Associates 2006). Advertising through digital games appears widely in popular print media and industry magazines, even though few empirical studies attempt to explain the effects of the ads that target game players. Yet because 75% of people who have Internet access spend more than an hour per month playing online games (Park Associates 2006), and as online gaming continues as the fastest growing area of the entertainment industry (IGDA Online Games SIG Steering Committee 2005), the analysis of online game advertising and incidental ad processing via the Internet can offer enormous help to practitioners and academicians who are interested in the impact of new communication channels on end users.

Incidental advertising processing, states of unconscious learning, and preattentive exposure effects remain hotly debated topics in advertising and marketing literature (Janiszewski 1993; McQuarrie and Mick, 2003; Shapiro, MacInnis, and Heckler 1997; Zajonc 1980). People often do not actively involve or process the majority of the ads in their immediate environments (Bauer and Greyser 1968; Webb and Ray 1979), but those ads can cause changes in people’s subconscious minds (Hawkins and Hoch 1992; Krugman 1981), because most learning, emotional reactions, and thought generation take place beyond awareness (Coulter 2007). Previous studies indicate that consumers might unknowingly process advertisements that are not their main focus and make future judgments about brands accordingly (Janiszewski 1993; Shapiro, MacInnis, and Heckler 1997). In this context, researchers find that the relevancy of the ad (Janiszewski 1988), information quantity, ad placement (Janiszewski 1993), and visual versus verbal cues (Janiszewski 1993; McQuarrie and Mick 2003) represent the main factors that can interfere with incidental advertising effects and alter levels of brand familiarity and brand liking, as well as the formation of consideration sets. Yet no empirical study has explored incidental advertising effects on computer game players, nor has research investigated the placement of the ads designed to be subconsciously processed.

This study attempts to replicate incidental advertising exposure effects in an underexplored realm of online games by manipulating message type (visual versus verbal) and ad placement (low versus high proximity). As one of the very few articles pertaining to game advertising, this study provides valuable insights into the development and placement of ads in online games, as well as theoretical advances about unconscious learning processes. In addition, the results of this study hold specific interest for online marketing practitioners, because the methodology allows for the analysis of the effects of favicon images and sidebar text, which previously have not been researched.

Literature Review


In the 20 years since Sega Games decided to put Marlboro billboards in its racing games (Chambers 2006), which may mark the start of the advergaming industry, applications of in-game advertising have grown dramatically. By 2000, 50% of the top 25 video games featured branded products (Nelson 2002).Although advertising practitioners, who were trying to find new ways to communicate with the hard-to-reach cohort of 18-35-year-old men, welcomed this development, the approach also raised concerns about controlling the brand’s role in games and testing the effectiveness of the placements (Nelson 2002). Consumers also developed ambivalent feelings toward product placements in games, supporting the increased realism and affordable game prices that resulted from advertising on the one hand but complaining about the clutter and irrelevant brand appearances on the other (Molesworth 2006). Recently, especially online, product placements appear to have been replaced largely by game sponsorships (Chambers 2006), interactive competitions (Wang 2006), pregame advertisements, customized arcade games, and banner advertisements that integrate brands within virtual games. Following the shift in attention from off- to online advertising, this study investigates the effectiveness of online game advertisements without product placement.

According to the Organization for Economic and Cooperative Development’s (OECD 2004, p. 9) digital games report, online games include “any computer-based game played over the internet including PC, console and wireless games including extensions of stand-alone games so that small groups of players (six to 16) can play together, to Massively Multiplayer Online Role Playing Games, with more than 10 000 players playing at the same time.” Although the online gaming industry owns only 6.4% of the digital games market, it continues to grow rapidly, in parallel with the broadband user base, because of greater cost effectiveness and the significant interaction among different players during the game (OECD 2004). According to the Pew Research Center (Jones 2003), 57% of female and 75% of male college students regularly play online games; among the men, gamers spend up to 15 hours per week playing online. The popularity of online games also seems to have increased demand for game portals; for example, Sony’s The Station (2007) reports more than 13 million registered users. This huge online demand and new tracking technologies have heightened advertisers’ interest in advergaming, which offers a targeted, measurable, adjustable form of advertising (OECD 2006). A preliminary scan of the major online gaming portals (e.g., Yahoo Games, MSN Gaming Zone, Freeonlinegames) reveals that it has become almost impossible to play an online game without exposure to a banner ad surrounding the game screen.

Although banner ad effectiveness commonly appears in advertising studies, no particular study relates banner ads in digital games. Research findings of general Internet advertising cannot be applied directly to the online gaming platform because of its unique form (e.g., sound effects, motion images, time perception, involvement and engagement levels). As Chou and Ting (2003) note, online gaming involves a “flow state” experience, in which people develop distorted perceptions of time and display unusual exploratory or playful behavior that can lead to a form of addiction. Such flow, defined as “a state of total involvement where one moment flows holistically in to the next without conscious intervention” (Csikszentmihalyi, qtd. in Celsi, Rose, and Leigh 1993, p. 11), also relates to “clear objective and immediate feedback, challenge encounter and adequate skill, combination of action and consciousness, concentration, sense of control, curiosity, and loss of self-consciousness,” which are commonly observed during online game playing (Wan and Chiou 2006, p. 318).Thus, participants likely are more engaged with their primary task (playing the game) and therefore have less attention capacity for the surrounding secondary information compared with their level of attention when they are engaging in usual online activity (e.g., reading online newspaper, paying bills online). In this sense, the findings of this study can expand Internet advertising literature by clarifying consumer behavior in online gaming environments and pinpointing the effects (if any) of text that appears in sidebars or images in address bars.

This study purposefully uses very small text and images to ensure participants do not process secondary cues actively during the experiment (e.g., before or after the game, while loading weapons). Therefore, in the low proximity/verbal cue condition, the text stimulus appears in the side bar, whereas in low proximity/visual cue condition, the address bar includes an embedded visual stimulus. The sidebar text consists of 11-point, white, Ariel font letters on blue backgrounds that are approximately .25 inches in height and 11 inches in length (depending on screen size). In most cases, the sidebar text consists of Web site names, which help users to click on the correct boxes if they minimize the page, but it also could include text longer than 100 characters (see Figure 1).

FIGURE 1. Examples of Sidebar Texts

Examples of Sidebar Texts

Address bar icons (a.k.a., favicons) are 16 × 16 pixel images that appear on the left-hand side of the address bar, next to the URL, and can be uploaded as a .favicon.ico format on the Web server. Most popular Web browsers (e.g., Internet Explorer, Mozilla, Firefox, Safari) display favicons as a default, but older versions of Internet Explorer require users to list specific Web sites in a favorites folder before they can view the favicon image. Favicons can exist in any shape, color, or texture and appear as either static or animated; Figure 2 provides some examples.

FIGURE 2. Examples of Address Bar Icons

Examples of Address Bar Icons

Theoretical Framework

Mere Exposure and Attitude Change. Robert Zajonc (1968, p. 1) defines mere exposure as “a condition which just makes the given stimulus accessible to the individual’s perception” and empirically demonstrates that people’s attitudes toward previously unknown objects become enhance simply through exposure to them. Although Zajonc acknowledges the strong impact of sociopsychological factors on cognitive and affective processing, he also bases his proposition on the premise that frequency breeds familiarity, which positively affects attitudes. He further indicates that the positive relationship between simple exposure and liking is consistent across cultures and species, regardless of the stimulus type and consciousness level (Zajonc 2001). Other researchers also offer alternative explanations to illustrate why mere exposure may be sufficient for a positive attitude change toward a message.

Perceptual Fluency. People process messages or objects they have already encountered more easily because a representation of those objects already exists in people’s brains (Monin 2002). Once a person forms a mental representation of a specific item, any subsequent processing occurs faster and, in many cases, makes the messages seem more likeable and factual than they would be if they were difficult or complicated to process (Reber and Schwarz 1999; Reber, Winkielman, and Schwarz 1998). Furthermore, Bornstein and D’Agostino (1994) explain the likeability of more readily available information according to a cognitive perceptual fluency/misattribution model, whereas Winkielman and Cacioppo (2001) propose a hedonistic fluency model that assumes fluent processing creates positive affective responses toward easily processed stimuli. According to the hedonistic fluency model, mild and positive affective responses occur because (1) familiar stimuli usually mean a harmless situation, (2) successful recognition creates good feelings, and (3) coherent interpretation leads to positive moods (Winkielman and Cacioppo 2001). Another potential reason people develop positive attitudes toward fluently processed messages, as offered by Monin (2002), may be that smooth processing makes people perceive the stimuli as more attractive, which provides a precursor to likeability.

Liking Through Familiarity. Human beings inherently tend to generate favorable affective responses toward those things with which they are familiar (Bornstein 1989; Rindfleisch and Inman 1998). Even if initial attitudes toward an unfamiliar object are negative, people experience feelings of intimacy and comfort when they gain a sense of familiarity, which enhances their attitudes toward the familiar object (Monin 2002; Zajonc 1968). The same process occurs in interpersonal relationships; as many empirical studies demonstrate, viewers like familiar faces more (Monin 2002; Peskin and Newell 2004) and perceive familiar people as more trustworthy (Kollock 1994). In addition, marketing research dedicated to familiarity liking indicates that familiarity increases brand loyalty (Hoyer and Brown 1990), decreases negative competitive advertising effects (Kent and Allen 1994), and positively affects brand preferences (Carpenter and Nakamoto, 1994).

Mere Exposure and Subconscious Processing: In 1989, Borstein conducted a meta-analysis and reviewed 134 studies that had tested various forms of mere exposure effects on cognitive and affective responses. He thus concluded that stimulus type, stimulus complexity, presentation sequence, exposure duration, stimulus recognition, age of the subject, and delay between exposure and ratings all affect the strength of mere exposure effects. Bornstein also determined that mere exposure effects are stronger when subjects are not aware of the exposure. Subsequently, Bornstein and D’Agostino (1992) revealed that respondents like subliminally processed (5 ms) stimuli better than consciously processed (500 ms) stimuli. Although this phenomenon can be explained away as a result of counterargumentation or satiation (Bornstein 1989), which occur during conscious processing, Zajonc (2001) notes natural differences among people in their cognitive processing abilities and therefore proposes subliminal exposure effects may be stronger than supraliminal exposure, because they create mostly affective responses.

Whereas early mere exposure studies mainly investigate the consequences of short-term repeat exposures, Janiszewski (1993) establishes the relationship between mere exposure and incidental exposure effects by empirically demonstrating that preattentively processed constant information can create subsequent familiarity and ease of processing, similar to Zajonc’s (1968) proposal. If receivers do not recognize the existence of secondary information when they focus on a major task, “like repetition, duration of exposure [becomes] an effective agent for promoting affective responses, provided that increased duration does not result in the recognition of the stimulus” (Janiszewski 1993, p. 377). His findings thus reveal that incidental advertising exposure (i.e., availability of attentive resources for the target ad is very low) is more effective when nonfocal verbal stimuli appear on the right side and nonfocal visual stimuli are on the left. Shapiro, MacInnis. and Heckler (1997) seek further support for the incidental advertising exposure hypothesis by asking subjects to read an article for 5 minutes on a computer screen while being exposed to ads placed outside their attention area. Those subjects who were incidentally exposed to ads were more likely to include the target products in their consideration sets in both familiar and unfamiliar buying situations. These findings remain consistent for both stimulus- and memory-based consideration formation phases, even though the subjects do not recognize the ads to which they had been exposed. In a similar study, testing consideration set formation under the influence of incidentally processed print ads, Shapiro (1999) successfully demonstrates that feature and semantic analyses occur during subconscious advertising processing and find empirical support for greater preference for advertised products processed with minimum attention. His findings also indicate incidental exposure is more effective when subsequent cues appear in a familiar form and within a relevant context.

Because of such unconscious processing, people may not realize that they see ads outside of their main focus area, so when asked, they cannot recall or recognize the advertising messages. As Shapiro (1999, p. 17) explains, “when consumers are preoccupied with a primary task, attentional resources available for processing secondary information are limited, and, therefore, memory traces for this information are assumed not to be strong enough to be retrievable during a directed search of a memory.” Referring to active memory, Krugman (1986) claims that if the audience does not remember its ad exposure, advertising effectiveness declines, but “minimally attended exposure” can still create positive advertising effects, even if it cannot be detected through recall measures. In parallel with this proposition, past studies clearly demonstrate that subconsciously processed messages can affect conscious decision making by creating positive affective responses toward the stimuli (Bornstein 1989; MacInnis and Jaworski 1989; Zajonc 1980). Janiszewski (1990) further explains that positive affective responses that result from incidental exposure increase preferences and favorable attitudes toward the stimulus, thus making it more appealing among the set of alternatives. In response to existing findings, which indicate positive subconscious affective reactions to incidentally processed messages that result in higher preferences and tendencies to behave toward the stimulus,

H1A: Participants in the experimental conditions will reveal higher preferences for the stimulus than will participants in the control group.

H1B: Participants in the experimental conditions will be more likely to choose the stimulus than will participants in the control group when asked to make a choice.

Proximity. Preattentive filtering may depend mainly on the location of the stimulus that surrounds the primary focus area (Ramstrom 2004). When the Euclidean distance from the foveal perception area increases, attention to the peripheral cues decreases correspondingly (Duncan, 1984; Egly, Driver, and Rafal 1994). Eriksen and St. James (1986) use their spotlight theory of attention to suggest that people inherently focus on behaviorally relevant information within their environments and that the focal area always takes the shape of a circle or donut, similar to a spotlight. However, this spotlight lacks definite borders, and perceived subsequent information gradually diminishes until no more object-specific information gets recorded on the visual cortex (Chun and Wolfe 2001). Eriksen and St. James also use a zoom lens model and thereby empirically demonstrate that a wider spotlight area decreases processing efficiency, whereas greater stimuli complexity makes the spotlight narrower.

Two studies in the consumer behavior domain analyze the location effects of subsequently processed ads and their consequences. Clark and Brock (1994) place warning messages about ice cream and tobacco products in either the peripheral or the central perception area (i.e., the bottom versus the center) of advertisements and ask participants to focus on the overall advertisement message. After the initial exposure, participants in the central perception condition displayed more negative attitudes toward the brands than did those in the peripheral perception condition. Raymond (2002) also tests upper versus lower field effects in unattended banner ads and finds that when incidentally processed banner ads appear above the primary focus field, recognition is higher, mainly because of upwardly driven saccades.

Despite these findings, which confirm that the placement of the incidental cue determines subsequent processing abilities, most incidental ad exposure studies still fail to recognize the relationship between spatial proximity and capacity allocation and regard any secondary information as a perceivable information source, regardless of its distance from the primary task region. Therefore, this study considers the proximity effects for implicit learning and investigates if these effects remain consistent in online environments. To manipulate the proximity of incidental cues, the study design places the stimulus either near or above participants’ focus area. Because digital games are complicated tasks that require higher levels of attention, players likely will concentrate on the game window, which should dramatically reduce the chance that they unconsciously perceive incidental cues in the sidebar or address bar.

H2A: Participants in high proximity conditions will have higher preferences for the stimulus than participants in low proximity conditions.

H2B:Participants in high proximity conditions will be more likely to choose the stimulus than will participants in low proximity conditions when asked to make a choice.

Visual versus Verbal Cues. Preliminary work pertaining to the cognitive processing of verbal and nonverbal (imagery) information, undertaken by Paivio (1971), suggests the superiority of pictures for associative learning. According to dual coding theory (DCT), processing and storing incoming information can occur in either the visual or the verbal subsystems in the brain, which are independent but also can work together. Furthermore, DCT asserts that pictures, which are processed by both visual and verbal subsystems, may be better recalled and recognized than texts, which are processed only by verbal subsystems. Although DCT seems dominant in advertising effectiveness literature, some alternatives exist, including the conceptual hypothesis (Potter et al. 1986), which presumes pictures are not processed, stored, and retrieved with their features but rather with regard to their meanings only.

Existing studies of the effectiveness of visual versus verbal cues in actively processed ads offer mixed results and suggest that effectiveness may depend on audience characteristics (Childers, Heckler, and Houston 1986). However, in general, people remember visual messages more (Gardner and Houston 1986), perceive them as more familiar (Hirschman 1986), and use them as a means to enhance their product attribute perceptions (Holbrook 1997; Smith 1991), though they do not affect brand preferences (Costley and Bruck 1992) or ad evaluations (Stafford 1996). Meyers-Levy (1989, p. 76) draws special attention to the right versus left brain processing of advertising messages, noting that “visual spatial or visual information seems to activate the undifferentiated, holistic processing style associated with the right hemisphere, while linguistic or verbal information appears to activate the detail-sensitive, differentiated processing style associated with the left hemisphere.” In another study, Hansen (1981) links the right brain with low involvement situations and asserts that visual messages create positive mere exposure effects better than verbal messages.

Janiszewski (1993) further extends the hemispheric approach to the realm of incidental advertising exposure and discovers that unattended verbal messages are more effective when they appear on the right side of the main focus area, but they become inferior to visual messages if they are placed on the left. According to Janiszewski, interpretations of secondary cues are driven mainly by feature analysis, which entails the recognition and processing of the “perceptual features of the stimulus” in the subconscious mind (Shapiro 1999, p. 17). Therefore, when both text and visual stimuli appear to the left of the focal point, viewers’ brains naturally allocate more capacity to process the visual cues, which results in greater familiarity and likeability for the visual messages during future encounters. However, McQuarrie and Mick (2003) find that incidentally exposed ads with figures produce more favorable attitudes and improved memory, whereas Clark and Brock (1994) report no significant effects of images in subsequently processed ad warnings, along with greater attitude changes after the exposure to peripheral verbal warnings. Because both these studies involve a consistent location but still achieve conflicting findings about the effects of message form (image versus text), this research investigates the following questions:

RQ1: Does the content of a message (visual versus verbal) that a player processes incidentally processed while playing an online game influence that player’s product preferences and product choice?

RQ2: Is there any interaction effect of the location (low  versus high proximity) and the message content (visual versus verbal) of the stimulus?



In exchange for class credit, 190 undergraduate students (96 men, 94 women) from a large northeastern U.S. university participated in the study. All students were enrolled in a general education course and were aged between 17 and 33 years (m = 18.3).The data collection took place during the last two weeks of February 2007, and students received debriefings after the conclusion of the data collection.

Design and Procedure

This study uses a 2 (proximity) × 2 (message content) between-subjects factorial design, along with a control group. Participants were assigned to each of the conditions according the day they arrived to participate in the experiment. Therefore, in the five experimental conditions run on five different days, 35 participants joined condition 1 (low proximity/visual stimulus), 35 participated in condition 2 (low proximity/verbal stimulus), 58 in condition 3 (high proximity/verbal stimulus), 30 in condition 4 (high proximity/visual stimulus), and 31 people participated in the control condition, which meant they played the online game but were not exposed to any stimulus. Approximately 5 subjects participated in each experimental session, and the entire data collection process took two weeks, which minimizes the potential for confounding factors such as seasonal effects or participant influences on one another.

Primary Task

During the recruitment stage, subjects were told they would play an online game and answer a brief survey about their online gaming experience. Upon arrival for the study, they received instructions about how to play a flash arcade game (The Professionals), in which the player shoots moving targets and gets points. Subjects then played the game on the lab computers continuously for five minutes (i.e., if they finished early, they restarted the game) in an attempt to score as many points as they could. Immediately after the playing time, subjects clicked on a link on the computer screen that took them to the online survey.


The verbal stimulus for this experiment consists of text that reads “Drink Lemonade.” In both the low and high proximity conditions, the white, 11-point Arial text appears on a blue background. The visual stimulus is a 16 × 16 pixel illustration of a lemon (see the appendix).

Manipulation Check

In the high proximity conditions, the verbal and visual stimuli are adjacent (distance = .01 inch) to the game window, whereas in the low proximity conditions, they appear approximately 2.5 inches above the game window. To ensure the manipulation of the stimulus location is effective, 17 different people from the same convenient sample pool participated in a manipulation check. After being told that a lemon image or text appears on the computer screen next to the game window, these participants searched for the visual or verbal cue; the time they take to figure out the location of the image/text was measured. It took significantly longer for participants to spot the stimuli in the low proximity conditions, which indicates the low and high proximity manipulations work well.

TABLE 1. Comparison of High and Low Proximity for Spotting the Stimulus

Comparison of High and Low Proximity for Spotting the Stimulus

Additional Measures

Shapiro, MacInnis, and Heckler (1997) identify three main issues in preattentive advertising exposure studies: controlling attention, measuring attention, and measuring preattentive processing. In validating the methodology, this study takes extra precautions with regard to these issues.

Controlling Attention. At the end of the survey, each participant indicated if he or she noticed any text, image, or ad on the computer screen outside the game window. Participants who answered “yes” and indicated anything related to a lemon or lemonade are eliminated from the data analysis (four subjects in high proximity/verbal condition, one subject in low proximity/verbal condition). Thus, all subjects in the analysis have been only incidentally exposed to the cues, and no subject has processed the target messages actively.

Measuring Attention. Participants also indicated the overall attention they paid to the primary task, which indicates their engagement with the primary task (i.e.,  “What was the overall attention you paid while playing the game?”). Those who indicated an overall attention level lower than the midpoint (3.5 on 7-point scale) do not appear in the data analysis (one subject in high proximity/verbal condition, two subjects in the high proximity/visual condition). This precaution ensures that all participants pay some attention to the primary task while passively processing the experiment stimuli.

Measuring Preattentive Processing.  In general, affective measures are more effective in subliminal processing studies (Zajonc 2001), because after subliminal processing, participants cannot recall ads and might not recognize those to which they previously were exposed. Therefore, in line with Janiszewski’s (1993) measure of attitudes toward incidentally processed unfamiliar text, this study asks participants to indicate their preferences, on a scale of 1-7, for two different beverages (lemonade and cranberry juice), with the assumption that incidental exposure should will generate positive attitudes toward lemonade and thus increase preferences for it. In addition, to measure the behavioral intention to choose lemonade over cranberry juice, the study requests their immediate preference (i.e., “If you were offered a drink right now, would you want to drink lemonade or cranberry juice?”) while also controlling for participants’ weekly lemonade and juice consumption. Although this technique appears different than Shapiro, MacInnes, and Heckler’s (1997) consideration set formation methodology, it employs the same logic and simply limits the consideration set to one.

Finally, the juice drinks are unbranded in the study to control for any preexisting brand attitudes. The beverage preference questions appear at the top of the questionnaire, which acknowledges the possibility that the effects of subliminal exposure might be short term (Bornstein 1989).


The questionnaire includes four items, measured on seven-point, Likert type, task involvement scale inspired by Zaichowski (1985). It also features questions about demographics, past beverage consumption, intentions to drink lemonade, and the incidental exposure measures.

Data Anlaysis

A  2 (proximity) × 2(content) plus control condition between-subjects factorial ANOVA compares the preferences for lemonade among all conditions, while controlling for past beverage consumption and involvement level. No significant difference emerges among the groups (F(4, 176) = 1.09, p > .05). As expected, prior lemonade consumption represents a significant predictor of preference for lemonade (F(1,176) = 25,31, p < .05). Thus, H1A and H2A must be rejected.

TABLE 2 Analysis of Variance for Lemonade Preference

Analysis of Variance for Lemonade Preference

According to a binary logistic regression model that predicts intentions to choose lemonade on the basis of the experimental condition (with the control group as a base), involvement level, and past beverage consumption, the model is statistically significant (Beta = 1.2, standard error = .18, p < .05). Both low proximity conditions and the high proximity/visual condition do not differ from the control condition in terms of choosing lemonade over cranberry juice, whereas the high proximity/verbal condition differs significantly (Beta = 1.9, standard error = .77, p < .05). Thus, H1B and H2B are partially supported; incidental exposure has more impact when the stimulus is closer to the target activity and when it appears in the form of text.

TABLE 3 Summary of Logistic Regression Analysis for Choice of Lemonade (N = 184)

Summary of Logistic Regression Analysis for Choice of Lemonade (N = 184)


The results of this study demonstrate that incidental exposure to ads during digital games influences consumer decisions. However, this effect may be small or nonexistent in certain circumstances, such as when the secondary stimulus appears far from the focus of attention or in a form that is hard to detect (e.g., small size, blended background). Exposure to text (e.g., “Drink Lemonade”) during online game playing systematically increases the choice of lemonade over an alternative drink, even though none of the participants recall processing this text actively. However, when the text resides three inches above the game window, it has no effect on beverage choice. This finding suggests that incidental exposure can be effective only if the secondary stimulus is close to the focal attention area.

The lack of an observed impact of the text message in the sidebar or the visual message in address bar indicates participants neither actively see these messages nor passively process the cues. Even though the cues appear only a few inches above the game screen at which participants looked for at least five minutes, they may have been unable to perceive them, because they never focused their attention on that part of the screen. Participants instead likely perceived the image in the address bar as a yellow dot and the text in the sidebar as a white line, because detail perception correlates negatively with distance from the focal area. These results also weaken the proposition that subliminal advertising in static images (e.g., erotic figures in ice cubes) can create attitudinal or behavioral changes. Just because something is within a person’s sight, it will not necessarily be processed as it were. If the decoding of images excludes their definitional characteristics (e.g., seeing a line instead of text), no meaningful perception process takes place.

A key findings of this study pertains to the difference between preferences and behavioral intentions as a result of incidental exposure. When participants indicated their preferences for lemonade on a Likert-type scale, no difference appears among the control and the other four conditions. However, when subjects had to choose between lemonade and cranberry juice, those in the high proximity/verbal condition chose lemonade significantly more than did those in the control condition. This finding may indicate that incidental exposure does not make great changes in people’s overall attitudes toward advertised products; instead, it may just increase familiarity, which then increases the probability that consumers will choose the target product among the various alternatives. Therefore, a person who expresses the same level of preference for both lemonade and cranberry juice will likely use a choice heuristic, because he or she weighs both alternatives the same. When asked to choose, this person probably heuristically selects the more familiar option, even without being aware of his or her previous exposure to the product messages. This finding implies that advertising in games may be more beneficial for brands that need to increase their familiarity or have low involvement level products (e.g., soft drinks, gum).


In contrast with previous research, this study indicates that visual messages have less impact than verbal messages. Although this phenomenon might be a natural occurrence that demands further research, the size of the text (i.e., the height of both the verbal and the visual message is the same, but the text is 2 inches longer) and the background color (i.e., verbal white on blue background, visual yellow on white background) in this study also might have caused this difference.

Despite attempts to control for many possible confounding factors (e.g., room temperature, season of the year, computer settings, social desirability effects, demand artifacts), limitations still mark this study. The study takes place in a computer lab, and the setting forces participants to pay full attention to the online game and disregard everything in the surrounding area. Although this design ensures incidental processing, it also might prevent players from diverting their attention to other things involuntarily, such as they may do when they play games in a natural settings. In addition, all participants are college students, who likely have advanced-level computer and game-playing skills, unlike the rest of society.

This study does not include every experimental condition that could test the hemispheric processing of incidental cues. Instead, in all conditions, the stimuli appear above the game window and lean toward the left-hand side. Therefore, additional studies should manipulate the location of the stimulus fully (i.e., left and right side, as well as below and above the focal task region).


Bauer, Raymond and Stephen A. Greyser (1968), Advertising in America: The Consumer’s View, Boston: Harvard University Press.

Bornstein, Robert F. (1989), “Exposure and Affect: Overview and Meta-analysis of Research, 1968-1987,” Psychological Bulletin, 106 (September), 265-289.

— and Paul R. D’Agostino (1992), “”Stimulus Recognition and the Mere Exposure Effect,” Journal of Personality and Social Psychology, 63 (4), 545-552.

Carpenter, Gregory S. and Kent Nakamoto (1994), “Reflections on Consumer Preference Formation and Pioneering Advantage,” Journal of Marketing Research, 31 (November), 570-573.

Celsi, Richard L., Randall L. Rose, and Thomas W. Leigh (1993), “An Exploration of High-Risk Leisure Consumption through Skydiving,” Journal of Consumer Research, 20 (June), 1-23.

Chambers, Jason (2006), “The Sponsored Avatar: Examining the Present Reality and Future Possibilities of Advertising in Digital Games,” available at http://ir.lib.sfu.ca/retrieve/1630/8878e0c3d9c0a0bc67670b8d9a0f.doc (accessed 11/1/2006).

Childers, Terry L., Susan E. Heckler, and Michael J. Houston (1986), “Memory for the Visual and Verbal Components of Print Advertisements,” Psychology & Marketing, 3, 137-150.

Chou, Ting-Jui and Chih-Chen Ting (2003), “The Role of Flow Experience in Cyber-Game Addiction,” Cyber Psychology & Behavior, 6 (December), 663-675.

Chun, Marvin M. and Jeremy M. Wolfe (2001), “Visual Attention,” in Blackwell’s Handbook of Perception, B. Goldstein, ed. Malden, MA: Blackwell, 272-310.

Clark, Eddie M. and Timothy C. Brock (1994), “Warning Label Location, Advertising and Cognitive Responding,” In Attention, Attitude, and Affect in Response to Advertising,  Timothy C. Brock, Eddie M. Clark, and David W. Stewart, eds. Hillsdale, NJ: Lawrence Erlbaum Associates, 287-300.

Costley, Carolyn L. and Merrie Brucks (1992), “Selective Recall and Information Use in Consumer Preferences,” Journal of Consumer Research, 18 (March), 464-473.

Coulter, Robin A. (2007), “Consumption Experiences as Escape: An Application of the Zaltman Metaphor Elicitation Technique,” in Handbook of Qualitative Research Methods in Marketing, Russell W. Belk, ed. Northampton: Edward Elgar, 400-418.

Duncan, J. (1984), “Selective Attention and the Organization of Visual Information,” Journal of Experimental Psychology, 113 (4), 501-517.

Egly, R., R. Driver, and R. Rafal (1994), “Shifting Visual Attention between Objects and Locations: Evidence from Normal and Parietal Lesion Subjects,” Journal of Experimental Psychology: General, 123, 161-177.

Eriksen, C. and J. St. James (1986), “Visual Attention Within and Around the Field of Focal Attention: A Zoom Lens Model,” Perception & Psychophysics 40 (4), 225-240.

Gardner, Meryl P. and Michael J. Houston, (1986), ‘The Effects of Verbal and Visual Components of Retail Communications,” Journal of Retailing, 62 (Spring), 64-78.

Greenwald, Anthony G. and Clark Leavitt (1984), “Audience Involvement in Advertising: Four Levels,” Journal of Consumer Research, 11 (June), 581-592.

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

Hawkins, Scott A. and Stephen J. Hoch (1992), “Low-Involvement Learning: Memory without Evaluation,” Journal of Consumer Research, 19 (September), 212-225.

Hirschman, Elizabeth C. (1986), “The Effect of Verbal and Visual Advertising Stimuli on Aesthetic, Utilitarian and Familiarity Perceptions,” Journal of Advertising, 15 (2), 27-34.

Holbrook, Morris B. (1997), “Stereographic Visual Displays and the Three-Dimensional Communication of Findings in Marketing Research,” Journal of Marketing Research, 34 (November), 526-536.

Hoyer, Wayne D. and Steven P. Brown (1990), “Effects of Brand Awareness on Choice for a Common, Repeat-Purchase Product,” Journal of Consumer Research, 17 (September), 141-148.

IGDA Online Games SIG Steering Committee (2005), “Casual Games White paper,” available at http://www.igda.org/casual/IGDA_CasualGames_Whitepaper_2005.pdf (accessed 11/1/2006).

Janiszewski, Chris (1988), “Preconscious Processing Effects: The Independence of Attitude Formation and Conscious Thought,” Journal of Consumer Research, 14 (September), 199-209.

— (1990), “The Influence of Print Advertisement Organization on Affect Toward a Brand Name,” Journal of Consumer Research, 17 (June), 53-65.

— (1993), “Preattentive Mere-Exposure Effects,” Journal of Consumer Research, 20 (December), 376-392.

Jones, Steve (2003), “Let the Games Begin,” available at http://www.pewinternet.org/pdfs/PIP_College_Gaming_Reporta.pdf (accessed 11/1/2006).

Kent, Robert J. and Chris T. Allen (1994), “Competitive Interference Effects in Consumer Memory for Advertising: The Role of Brand Familiarity,” Journal of Marketing, 58 (July), 97-105.

Kollock, Peter (1994), “The Emergence of Exchange Structures: An Experimental Study of Uncertainty, Commitment, and Trust,” American Journal of Sociology, 100 (September), 313-345.

Krugman, Herbert E. (1986), “Low Recall and High Recognition of Advertising,” Journal of Advertising Research, 26 (February/March), 79-86.

MacInnis, Deborah J. and Bernard J. Jaworski (1989), “Information Processing from Advertisements: Toward an Integrative Framework,” Journal of Marketing, 53 (October), 1-23.

McQuarrie, Edward F. and David Glen Mick (2003), “Visual and Verbal Rhetorical Figures under Directed Processing Versus Incidental Exposure to Advertising,” Journal of Consumer Research, 29 (March), 579-587.

Meyers-Levy, Joan (1989), “Priming Effects On Product Judgments: A Hemispheric Interpretation,” Journal of Consumer Research, 16 (June), 76-87.

Molesworth, Mike (2006), “Real Brands in Imaginary Worlds: Investigating Players’ Experiences of Brand Placement in Digital Games,” Journal of Consumer Behavior, 5 (July/August), 355-367.

Monin, Benoit (2002), “The Warm Glow Heuristic: When Liking Leads to Familiarity,” Dissertation Abstracts International: Section B: The Sciences and Engineering, 62 (June), 5430.

Nelson, Michelle R. (2002), “Recall of Brand Placement in Computer/Video games,” Journal of Advertising Research, 42 (2), 80-92.

OECD (2006), “Digital Broadband Content: The Online Computer and Video Game Industry,” available at http://www.lessig.org/blog/archives/OECD_Games.pdf (accessed 11/1/2006).

Paivio, A. (1971), Imagery and Verbal Processes, New York: Holt, Rinehart & Winston.

Park Associates (2006), “PC In-Game Advertising Revenue to Top $400 Million by 2009,” http://newsroom.parksassociates.com/article_display.cfm?article_id=313 (accessed 11/1/2006).

Peskin, Melissa and Fiona N. Newell (2004), “Familiarity Breeds Attraction: Effects of Exposure on the Attractiveness of Typical and Distinctive Faces,” Perception, 33 (2), 147-157.

Potter, Mary C., Judith F. Kroll, Betsy Yachzel, Elizabeth Carpenter, and Janet Sherman (1986),”Pictures in Sentences: Understanding Without Words,” Journal of Experimental Psychology: General, 115 (3), 281-94.

Ramstrom, Ola (2004), “Visual Attention using Game Theory,” available at http://www.nada.kth.se/utbildning/forsk.utb/avhandlingar/lic/RamstromOla.pdf (accessed 11/1/2006).

Raymond, Jane E.(2002), “Upper-Lower Visual Field Effects on the Visual Memory for Incidentally Viewed Branded Images,” Advances in Consumer Research, 29(1), 223-224.

Reber, Rolf and Norbert Schwarz (1999), “Effects of Perceptual Fluency on Judgments of Truth,” Consciousness and Cognition, 8 (September), 338-342.

—, Piotr Winkielman, and Norbert Schwarz (1998), “Effects of Perceptual Fluency on Affective Judgments,” Psychological Science, 9 (January), 45-48.

Rindfleisch, Aric and Jeffrey Inman (1998), “Explaining the Familiarity-Liking Relationship: Mere Exposure, Information Accessibility, or Social Desirability,” Marketing Letters, 9 (February), 5-19.

Shapiro, Stewart (1999), “When an Ad’s Influence is Beyond our Conscious Control: Perceptual and Conceptual Fluency Effects Caused by Incidental Ad Exposure,” Journal of Consumer Research, 26 (June), 16-36.

—, Deborah J. MacInnis, and Susan E. Heckler (1997), “The Effects of Incidental Ad Exposure on the Formation of Consideration Sets,” Journal of Consumer Research, 24 (June), 94-104.

Smith, Ruth A. (1991), “The Effects of Visual and Verbal Advertising Information on Consumers’ Inferences,” Journal of Advertising, 20 (December), 13-25.

Stafford, Marla R. (1996), “Tangibility in Services Advertising: An Investigation of Verbal versus Visual Cues,” Journal of Advertising, 25 (Fall), 13-29.

The Station (2007), www.thestation.com (accessed 11/2/2007).

Wan, Chin-Sheng and Wen-Bin Chiou (2006), “Psychological Motives and Online Games Addiction: A Test of Flow Theory and Humanistic Needs Theory for Taiwanese Adolescents,” CyberPsychology & Behavior, 9 (June), 317-324.

Wang, Alex (2006), “Ad Engagement: A Driver of Message Involvement on Message Effects,” Journal of Advertising Research, 46 (Fall), 355-368.

Webb, Peter and Michael Ray (1979), “Effects of TV Clutter,” Journal of Advertising Research, 19 (June), 7-12.

Winkielman, Piotr and John T. Cacioppo (2001), “Mind at Ease Puts a Smile on the Face: Psychophysiological Evidence that Processing Facilitation Elicits Positive Affect,” Journal of Personality and Social Psychology, 81 (December), 989-1000.

Zaichkowsky, Judith L. (1985), “Measuring the Involvement Construct,” Journal of Consumer Research,12 (December), 341-52.

Zajonc, Robert B. (1968), “Attitudinal Effects of Mere Exposure,” Journal of Personality and Social Psychology, 9 (June), 1-29.

— (1980), “Peeling and Thinking: Preference Needs no Inference,” American Psychologist, 35 (February), 151-175.

— (2001), “Mere Exposure: A Gateway to the Subliminal,” Current Directions in Psychological Science, 10 (December), 224-228.


CONDITION 1: Low Proximity/Verbal Cue

Low Proximity/Verbal Cue

CONDITION 2: Low Proximity/Visual Cue

Low Proximity/Visual Cue

CONDITION 3: High Proximity/Verbal Cue

High Proximity/Verbal Cue

CONDITION 4: High Proximity/Visual Cue

High Proximity/Visual Cue

CONDITION 5: Control Condition

Control Condition

About the Author

Adam Acar (Ph.D Candidate, University of Connecticut) has been majoring in marketing communications and currently is a Ph.D. candidate at the University of Connecticut. His research interests include advertising effectiveness, social network marketing and market mixed modeling.

Note: The author thanks Dr. Robin Coulter and Dr. Alex Wang for their valuable contributions to this research.