In-Game Advertising Effects: Examining Player Perceptions of Advertising Schema Congruity in a Massively Multiplayer Online Role-Playing Game

Ben Lewis and Lance Porter

Louisiana State University

Abstract

This quasi-experimental survey involves 100 participants aged 18-24 years, conducted to measure effects of advertising schema congruity in a massively multiplayer online role-playing game (MMORPG) and other aspects of gameplay experience. Results from a posttest questionnaire suggest that though moderately incongruent advertising in an MMORPG can trigger high awareness rates, extremely incongruent in-game advertising can reduce a game's perceived sense of realism and annoy players if not coordinated appropriately with the game environment. Whereas previous research suggests players usually accept in-game advertising when relevant, this study shows that the opposite occurs when advertisers fail to contextualize ads within the game world.

Keywords: in-game advertising, schema theory, massively multiplayer online role-playing games (MMORPGs).

Introduction

Recent research has documented the increased migration of men, aged 18-34 years, from watching television to playing console and computer games during their leisure time (Activision 2005; Chaney, Lin, and Chaney 2004; di Cesare 2005; Donaton 2003; Reynolds 2004; Stanley 2004; Takahashi 2000). Nielsen Media Research has found a gradual decline in television viewing habits among this same audience, who make up approximately 12% of the total television audience and account for around $4.3 billion in targeted network and cable advertising (Reynolds 2004). Although some may express difficulty accepting the notion that "video games have spawned a technology of substantial importance" (Castronova 2005, p. 264), advertisers have begun to pay increasing attention to the game industry in their attempts to reach this elusive demographic. As gaming audiences widen over the coming years, advertisers that cater primarily or exclusively to this key demographic would be remiss to ignore the immense appeal video games hold with the U.S. population.

The purpose of this study is to gauge player perceptions of advertising in an online multiplayer computer game environment by drawing comparisons between reactions to advertising from an incongruent group playing an ad-supported online role-playing game and a congruent group playing the same game without real-world brands present. We seek to observe the role of schema congruency in gameplay and its impact on advertising awareness, unaided recall, message recognition, brand interest, and intent to purchase. Results may help researchers learn the effects of interactive media and video games on consumers to develop a better understanding of how games can be used effectively for communication purposes, as well as provide game publishers, developers, and advertisers with a firsthand look at the factors involved in effective in-game advertising.

The video game industry has recently eclipsed U.S. box office revenue figures and captivates audiences of all demographics worldwide. The market research company NPD Group estimates there are 174 million gamers in the United States (Caoili 2008), and IGA Partners claims that computer and console games represented a $28 billion worldwide entertainment medium in 2005 (BBC News 2005), having generated 30 billion eyeball hours in 2002 alone (IGA Partners 2004). Some analysts had anticipated that the international video game market would be worth up to $55.6 billion by 2008 (Nicovich 2005), though DFC Intelligence pegged the industry at a more conservative $44 billion by 2011 (Graft 2006b).

Advertisers, agencies, game developers, and media publishers are constantly seeking new developments in research on advertising in video games, with projected in-game ad spending expected to rise significantly in the near future. As additional revenue from advertisers lowers the overhead costs of producing games, individual unit prices could drop as well (Campbell 2006; Nelson, Keum, and Yaros 2004), prompting higher unit sales and potentially higher numbers of eyeballs for advertisers. Consumers stand to benefit from research on the effects of in-game advertising as well: Consistent revenue streams to publishers from advertisers could induce a drop in product prices, and more effective placement of advertising in games could lead to more immersive consumer experiences.

Literature Review

In-Game Advertising Spending Increasing

Just as advertisers began incorporating the Internet into their marketing mix in the late 1990s (Bush, Bush, and Harris 1998), many advertisers and agencies have turned to in-game marketing as a crucial facet of their modern communications objectives. In January 2006, San Francisco media agency Engage In-Game Advertising announced it would be working with video game advertising networks to deliver in-game promotion opportunities to advertisers. Engage's first campaign involved placing Subway restaurant ads into the first-person shooter game Counter-Strike, resulting in 31,000 unique impressions in key markets over three weeks (Gibson 2006). In May 2006, Engage joined with San Francisco's Greystripe mobile in-game ad network to produce a branded game for SpeeDee Oil Change & Tune-Up, which offered consumers free play and a $5 discount on their next oil change (Sanders 2006).

Although spending on in-game advertising is currently modest, several consultancies and researchers anticipate spending to skyrocket in the near future. According to a 2005 U.S. Association of National Advertisers survey, 9% of companies surveyed had advertised in video games over the past year, and 13% said they planned to do so in the near future (Sennott 2005). In-game ad spending jumped from $10 million in 2003 (Stanley 2004) to nearly $200 million in 2004 (Hershman 2005) and continues to grow. The Yankee Group estimated that the market would reach between $800 million (Oser 2005a) and $900 million by 2009 (Oser 2005b), and the entertainment industry research firm DFC Intelligence predicted a jump to $1 billion by 2008 (Hershman 2005).

Traditional Advergaming Practices Defined

Two prominent forms of advertising through games exist today. Traditional "advergaming" involves designing an entire game around a brand or product promotion (usually PC-based and played online), whereas in-game advertising involves the placement of real-world marketing into preexisting console and computer games, commonly in the form of billboards, posters, and sponsor signage in sports and racing games (Schwarz 2005). For example, an automotive company might engage in advergaming by commissioning a developer to create an online game in which players compete for high scores online by racing computerized versions of that company's cars on its Web site. In contrast, the same company could pay to insert its digitized billboards into a virtual racetrack environment in a NASCAR game available for purchase in retail stores. Although some sources refer to in-game ads and advergaming interchangeably (Graft 2006a), this study establishes a clear distinction between the two formats and focuses primarily on effects of in-game advertising alone; advergaming traditionally offers only one unique marketing message at a time.

Factors in Using In-Game Advertising Effectively

Recent studies on the effects of advertising in online games have been encouraging yet sparse. Chaney, Lin, and Chaney (2004) strategically placed billboards featuring fictional products in a first-person shooter computer game and invited 42 participants to compete in "deathmatches" for 15 minutes at a time while being exposed to the ads. The researchers recorded each session to measure individual exposure to ads and observe conversations among players during the matches. Participants recalled passing billboards, but many could not recall the names of the products or brands after the session was over. However, players recalled the in-game billboards at a rate higher than that produced during real-life sporting events, perhaps because "gamers view only one billboard at a time as they pass through various ‘rooms' and the information for each product has some chance of being assimilated" (Chaney, Lin, and Chaney 2004, p. 66). In real-life sports events, ads appear presented simultaneously and therefore compete for viewers' attention. These researchers also attribute this moderate effect to the game's high concentration requirements and the advertisements' minimal impact on enhancing the game experience, which supports the argument that effective in-game advertising must contribute to, not detract from, a game's level of immersion and realism.

Several studies also focus on how willing players are to accept in-game advertising. Nielsen Entertainment and Activision conducted a study among 1,350 male gamers aged 13-44 years and found "67 percent of gamers believe in-game advertising makes games more realistic and 40 percent of male gamers say in-game ads influence their purchasing decisions" (Wegert 2005, p. 8). Furthermore, most gamers who recalled a product advertised in a game believed that the advertisement "fit the game they were playing" (Activision 2005, p. 4). Although the study's findings suggest an encouraging acceptance of in-game advertising, all four games featured in these experiments primarily involve sports and racing categories-game genres in which advertising makes the most sense, considering their settings' resemblance to real-world environments (di Cesare 2005; Nelson 2002). These results may support the relevance of in-game marketing to advertisers; however, research involving advertising in MMORPGs remains relatively scarce.

Several studies (Activision 2005; Book 2004; Chaney, Lin, and Chaney 2004; Nelson 2002; Nelson, Keum, and Yaros 2004) find that players generally accept in-game advertising when placed relevantly in games, especially when a game's sense of realism is undisturbed or improved. For example, advertising signage in sports and racing games often translates well to players, whereas advertising in fantasy games may appear unnatural and intrusive (Nelson 2002).

Seamless integration is a crucial factor in successful in-game advertising. If a marketing message interrupts immersion or makes the experience appear less realistic, the advertiser has failed (di Cesare 2005; Nelson, Keum, and Yaros 2004). Integrating a product directly into a game's main storyline runs the risk of player backlash if the placement seems artificial in any way, yet research shows this approach may be the most effective method of in-game advertising when contextually positioned (Activision 2005; Nelson 2002). Nelson (2002, p. 80) also finds superior brand recall when brands were "a major part of game-play" and considered "relevant to the consumer."

The purpose of this study is to measure player acceptance of varying degrees of congruent and incongruent advertising in an MMORPG environment to understand the effects of advertising context in games more effectively. To support this approach, we employed schema congruity theory as an underlying framework.

Schema Congruity as Framework

Perhaps a better way to examine brands within video games is through schemata. Schemata represent people's expectations or cognitive models about other people, content, objects, or messages. People use schemata to help them make sense of their environment and then behave accordingly (Fiske and Linville 1980; Schank and Abelson 1977). Speck, Schumann, and Thompson (1988, p. 70) indicate that schemata "organize perception by organizing expectations." Bartlett's (1932) original experimental work measured long-term memory by asking participants to recall folk tales over the course of a year. His results demonstrate that participants altered their memories constantly to fit individual schemata.

Brewer and Treyens (1981) develop schema theory further by showing how people's expectations can create memories of things that never occurred. After placing participants in a room identified as a study room, these researchers asked those participants to recall that room's contents later. Numerous participants recalled seeing books in the room even though no books were present. These results show that participants' schemata of an ideal study room colored their recollections of the room's contents.

Similarly, schemata influence overall information encoding. Depending on how congruent incoming messages are with their existing objectives, people either process or ignore messages (Wyer et al. 1982). They then attempt to reconcile the information with their existing schema (Mandler 1982).

This reconciliation often determines the efficacy of advertisements in delivering messages. In terms of advertising, audiences use schemata developed over time to decide which messages to elaborate on and which to discard from among the hundreds of messages that bombard them each day. Many studies have found that moderately incongruent events inspire extensive processing and elaboration and evoke more positive evaluations (Campbell and Goodstein 2001; Mandler 1982; Meyers-Levy and Tybout 1989).

Mandler's (1982) seminal work shows that though incongruent advertising tends to encourage more processing, consumers usually dismiss it. Mandler initially proposed that schema congruity would lead to more positive responses because people favor information that meets their expectations. Yet though congruent messages may lead to a slight positive evaluation, they do not stimulate much arousal. Conversely, moderately incongruent advertising, which includes incongruities that can be successfully resolved, leads to more elaboration, because audiences can reconcile them with their existing schemata. Finally, extreme incongruity, which can be resolved only by changing existing cognitive structures, often leads to frustration and avoidance. Mandler's work has received empirical support from various studies (Meyers-Levy and Tybout 1989; Peracchio and Tybout 1996; Schmidt and Hitchon 1999; Stayman, Alden, and Smith 1992).

However, Meyers-Levy and Tybout (1989) find slightly different results. They examine product preference and cognitive processing and reveal that schema incongruity leads to greater elaboration and product recall.

Sujan, Bettman, and Sujan (1986) find more nuance; atypical salespeople elicit more systematic or analytical processing of product attributes, whereas typical salespeople encourage people to engage in simpler, heuristic processing. Furthermore, when salespeople are congruent with participants' expectations, the participants recall more about the salesperson than about the product.

In defining Web schema congruity, Bellman and Rossiter (2004) also find that the more congruent a Web site is with users' expectations, the easier users find the site to navigate, and the better their attitudes toward brands featured or advertised on the site, as well as the higher is the quality of consumers' brand choices.

However, Lee and Shen (2006) find that processing load might moderate the schema congruity effect in evaluations of advertisements. In their experiment, which involves advertisements featuring multiple products and different cognitive loads, they find that in conditions with high cognitive processing loads, participants process congruent ads more easily. Because incongruent ads require more elaboration, participants tend to dismiss incongruent ads in high cognitive load situations. Video game environments entail high cognitive load conditions, which implies advertising in these environments may provoke similar reactions. Therefore, we suggest:

This study also explores the following research questions, related to topics left unresolved by prior studies:

Method

Quasi-Experimental Survey

We conducted a quasi-experimental survey using a convenience sample of 100 undergraduate students aged 18-24 years (48 men, 52 women), recruited from a mass communication introductory course at a southeastern U.S. university. This sample provided a wide scope of responses, from people with little or no gaming experience to those who play video games daily. Each participant played the MMORPG Anarchy Online on a supplied PC gaming setup for 30 minutes. We placed participants in one of two groups: The incongruent group played online with other study participants in the same room as teammates in a version of the game featuring real-world advertising, whereas the congruent group played a version of the game featuring default fictional billboards but identical other conditions. The gaming sessions took place over a period of seven days.

Anarchy Online is an online-only, third-person role-playing game set about 30,000 years in the future. Funcom, the developer of Anarchy Online, currently offers a free client program supported by in-game advertisements, delivered and maintained by the Massive Incorporated advertising network. These ads are displayed as billboards and video screens in the game and cycle frequently as players move through each area to guarantee several impressions for each brand placed in the game. This game was ideal for our study because it contained more frequent and varied advertising than other games supported by the Massive Incorporated network at the time of the study, such as Splinter Cell: Chaos Theory, S.W.A.T. 4, and Rainbow Six: Lockdown.

During the week of the quasi-experimental sessions, ads for seven organizations and brands appeared in the game: static and video ads for the U.S. Army, Air Force, and Navy and static ads for Subway restaurants, Castrol Syntec motor oil, and Under Armour athletic apparel; later (after 39 participants had completed the survey), Garnier Fructis hair styling product ads also appeared. Late additions of questions to the post-quasi-experimental survey attempted to measure the effectiveness of the Garnier Fructis ads among the remaining 11 participants in the incongruent group (see Figure 1) and test message recognition among the congruent group. Of the video-based ads, the U.S. Army clips were shown much more frequently than those for the Air Force or Navy; the Navy ads appeared most sparingly.

All billboards in the congruent group sessions were displayed as static ads for Dreadloch Arms & Gear, a fictional weapons manufacturer in the Anarchy Online world (see Figure 2). Many of the thousands of custom items and weapons in the game are identified as manufactured by Dreadloch and available for purchase in locations throughout the game. These ads therefore are relevant to the typical Anarchy Online player experience and represent the default billboards that players with paid accounts would normally see in the game. No advertisements featuring video or audio were present in the congruent group's version of the game.

Upon arrival at the test site, participants were randomly assigned to one of four computers and briefly instructed on how to play the game. Participants wore headphones for a more immersive experience but were allowed to communicate vocally with teammates if they had become separated from their group in the game. For the purposes of this study, we placed players into teams of three or four and instructed each team to make their way through different parts of virtual cities using teleport stations. We invited each team to explore particular areas for five to ten minutes at a time. This approach adequately mimicked usual gameplay in these areas; players must travel through cities to buy items, meet up with other teammates, rent in-game apartments, and reach their destinations, where they carry out their required quests. No in-game combat occurred during the game sessions in this study. Participants were exposed to numerous ads along the paths from city to city and were free to play the game as they saw fit during exploration periods.

Figure 1. Incongruent Group Treatment

Incongruent Group Treatment

Figure 2. Congruent Group Treatment 

Congruent Group Treatment

Posttest Questionnaire

After the 30-minute game session, participants answered a computer-based questionnaire online to measure advertising awareness, unaided ad recall, acceptance or rejection of in-game advertisements, potential contribution of in-game ads to the game's sense of realism, suitability of advertising in specific game genres, common demographic information, prior gaming experience information, and more. All participants were debriefed after completing their survey and were free to leave afterward.

Following models in previous studies (McMillan and Hwang 2002; Nelson 2002; Nelson, Keum, and Yaros 2004; Sundar and Kim 2005), the questionnaire used Likert scales and open-ended questions to measure unaided recall, advertising awareness, and message recognition. After the questions about participants' perceptions of the game and their previous gaming experience, we asked unaided recall questions about each product category presented in the game, such as "When you think of athletic apparel, what brand name comes to mind?" Next, the claimed advertising awareness items included, "I have recently seen advertising in video games for the following types of athletic apparel: (Click all that apply)." We assessed message recognition by asking, "Which of these offers the following message in its advertising?" Sometimes the list contained messages from competing brands, not present in the game, to measure the effectiveness of in-game messages. We measured interest and intent to purchase in the subsequent questions. To assess attitudes toward advertising across various media, we asked participants to rate their level of agreement with statements such as "Advertising/product placement in TV shows is annoying/obtrusive to me" on a five-point scale, with 1 = "Strongly Agree" and 5 = "Strongly Disagree."

Similar to Nelson's (2002) questioning to test for acceptance of in-game advertising ("Do you think product placement impaired or helped the game experience?"), the survey measured different levels of approval or rejection of in-game advertising by asking participants to rate their level of agreement with statements such as "Advertising in video games is annoying/obtrusive to me," "Advertising should never be present in video games," "Advertising in video games makes the games seem more realistic and immersive," and "Advertising in video games is acceptable when the products/brands shown are real" (see Table 1). We reverse-coded some items in the results because of their phrasing in the survey; in these cases, higher mean scores reflect more unfavorable attitudes.

Table 1. Example of Survey Format

Part III: Advertising Position

The following will ask about your personal position on advertising. Please choose the best answer that reflects your viewpoint.
44. Advertising in general is annoying/obtrusive to me.

       ☐ Strongly agree

       ☐ Agree

       ☐ Neutral

       ☐ Disagree

       ☐ Strongly disagree
45. Advertising/product placement in movies is annoying/obtrusive to me.

       ☐ Strongly agree

       ☐ Agree

       ☐ Neutral

       ☐ Disagree

       ☐ Strongly disagree
46. Advertising/product placement in TV shows is annoying/obtrusive to me.

       ☐ Strongly agree

       ☐ Agree

       ☐ Neutral

       ☐ Disagree

       ☐ Strongly disagree
47. Advertising/product placement in video games is annoying/obtrusive to me.

       ☐ Strongly agree

       ☐ Agree

       ☐ Neutral

       ☐ Disagree

       ☐ Strongly disagree

We measured players' potential for in-game advertising to alter participants' purchasing habits in the near future by their level of agreement with statements such as "I would welcome advertising in video games if the retail price dropped by $20 because of advertising included in the game" and "I would pay more for an advertising-free version of a video game that I was interested in." Finally, we collected standard demographic information and thanked participants for their contribution to the study. After debriefing them on the purpose of the study, we asked that they not discuss the study with others.

Results

Our hypothesis received support: Participants in the congruent group who played games with advertisements that featured fictitious and congruent brands exhibited significantly higher levels of acceptance of in-game advertising as unobtrusive additions to the game environment than did those in the incongruent group. Furthermore, the results of a one-way analysis of variance (ANOVA) revealed that the incongruent group found in-game advertising significantly more annoying and obtrusive than the congruent group (see Table 2).

The incongruent group also agreed significantly more with the statement that advertising should never be present in games and agreed significantly less with the statement that advertising in games makes the games seem more realistic and immersive compared with the congruent group. The incongruent group agreed significantly less with the statement that advertising in video games is acceptable when the products/brands shown are real and that they would welcome advertising in video games if the retail price dropped by $20 because of advertising, again compared with the congruent group. Finally, the incongruent group agreed significantly more that they would pay more for an advertising-free version of a video game than did the congruent group (see Table 2).

Table 2. Differences between Quasi-Experimental Groups in Reaction to Advertising

  Incongruent Group Congruent Group    
    Mean (SD) Mean (SD) F-value (d.f.) Significance
Ads in TV annoying 3.26 (1.14) 3.76 (.87) 6.08 (1, 98) .015
Ads in games annoying 2.92 (1.24) 3.56 (.95) 8.363 (1, 98) .005
Ads never in games 
 3.1 (1.07) 3.72 (.76) 11.135 (1, 98) .001
Ads in games realistic 2.84 (1.24) 2.34 (.772) 5.893 (1, 98) .017
Ads acceptable if real 2.92 (1.12) 2.48 (.68) 5.636 (1, 98) .020
Welcome ads if $20 off 2.1 (.86) 1.72 (.9) 4.62 (1, 98) .034
Pay more for no ads 3.42 (1.33) 4.1 (.79) 9.709 (1, 98) .002
Notes: N = 100, 1 = "Strongly Agree," 5 = "Strongly Disagree."

Results of Research Question Testing

Among those in the incongruent group who responded to the statement, "Advertising is realistic or appropriate when placed in the following game genre(s): (Click all that apply)," 42 (85.7%) chose sports, 41 (83.7%) chose racing/driving, 11 (22.4%) chose fighting, 9 (18.4%) chose role-playing/adventure, 8 (16.3%) chose action/first-person shooter, 1 (2%) chose strategy/puzzle, and 1 (2%) replied "all." Among those who responded in the congruent group, 41 (85.4%) chose sports, 36 (75%) chose racing/driving, 15 (31.3%) chose role-playing/adventure, 13 (27.1%) chose fighting, 13 (27.1%) chose action/first-person shooter, 3 (6.2%) chose strategy/puzzle, and 1 (2.1%) replied with "none."

Regarding the second research question, a one-way ANOVA yielded no significant results among different ages, ethnicities, or income levels; however, significant differences emerged between male and female participants in their reactions to advertising (see Table 3). More women disagreed that advertising in games is annoying or obtrusive and believed that in-game advertising makes video games seem more realistic than did men. More men believed that in-game advertising is acceptable when the products advertised are fictional; conversely, more women believed that in-game advertising is acceptable when the products actually exist.

Table 3. Differences between Genders in Reactions to Advertising

  Females Males    
    Mean (SD) Mean (SD) F-value (d.f.) Significance
Ads in movies annoying 3.88 (.758) 3.33 (1.06) 9.069 (1, 98) .003
Ads in TV annoying 3.83 (.857) 3.17 (1.18) 11.095 (1, 98) .001
Ads in games annoying 3.48 (1.0) 2.98 (1.25) 4.965 (1, 98) .028
Ads in games realistic 2.27 (.87) 2.94 (1.14) 11.029 (1, 98) .001
Ads acceptable if fake 3.42 (.83) 3.02 (1.02) 4.730 (1, 98) .032
Ads acceptable if real 2.46 (.78) 2.96 (1.05) 7.288 (1, 98) .008
Notes: N = 100, 1 = "Strongly Agree," 5 = "Strongly Disagree."

A one-way ANOVA also revealed significant differences in reactions to advertising between nongamers (who reported playing video games for 0 hours per week) and avid gamers (those who reported playing video games for 11 hours or more per week) (see Table 4). Nongamers found advertising in general less annoying or obtrusive than did avid gamers. Nongamers also found advertising in movies, television, and video games less annoying or obtrusive. However, the gamers agreed less with the statement that ads should never be present in video games than did avid gamers and agreed more that advertising in video games makes the games seem realistic. Avid gamers expressed more agreement that in-game advertising is acceptable when the products advertised are fictional; conversely, nongamers considered in-game advertising more acceptable when the products advertised actually exist.

Table 4. Differences between Levels of Previous Gaming Experience in Reactions to Advertising

  Nongamers Avid Gamers    
    Mean (SD) Mean (SD) F-value (d.f.) Significance
Ads in general annoying 3.18 (.83) 2.44 (1.03) 8.204 (1,59) .006
Ads in movies annoying 3.91 (.63) 2.75 (.93) 30.637 (1,59) .000
Ads in TV annoying 3.75 (.86) 2.81 (1.05) 12.705 (1,59) .001
Ads in games annoying 3.44 (.99) 2.38 (1.26) 11.912 (1,59) .001
Ads never in games 3.6 (.84) 2.81 (1.11) 8.77 (1,59) .004
Ads in games realistic 2.4 (.89) 3.5 (1.12) 15.377 (1,59) .000
Ads acceptable if fake 2.63 (1.15) 3.42 (.81) 9.08 (1,59) .004
Ads acceptable if real 2.51 (.76) 3.56 (1.09) 17.827 (1,59) .000
Notes: N = 100, 1 = "Strongly Agree," 5 = "Strongly Disagree."

The third research question pertains to whether in-game advertising influences a player's perception of navigation through the game. Results of a one-way ANOVA indicated that participants who saw congruent in-game ads rated the ease of navigation much more positively (M = 2.2, SD = .95) than did those who saw incongruent in-game ads (M = 2.94, SD = 1.27, F(1, 98) = 10.923, p = .001).

Regarding the influences of congruency in in-game advertising, the results of a one-way ANOVA revealed no significant differences in measured product/brand interest, advertising message recognition, or intent to purchase between those who played the game with incongruent versus ads present. However, there were some notable differences in unaided recall and advertising awareness testing. Specifically, the open responses to unaided recall questions were coded with numbers to represent the most popular answers for each item, and a one-way ANOVA indicated significant differences in advertising awareness levels, though only for three brands: Under Armour, Marines, and Subway. More participants in the incongruent group recalled seeing in-game Under Armour ads (M = .62, SD = .49) than did those in the congruent group (M = .26, SD = .44, F(1, 98) = 14.84, p < .001). Fewer participants in the incongruent group recalled seeing in-game Marines ads (M = .1, SD = .3) compared with the congruent group (M = .34, SD = .48, F(1, 98) = 8.98, p = .003). More players in the incongruent group recalled seeing in-game Subway ads (M = .6, SD = .5) than did those in the congruent group (M = .16, SD = .37, F(1, 98) = 25.34, p < .001).

Finally, do players perceive incongruent but real in-game advertising as contextual-that is, adding realism and immersion to the game environment? The ANOVA results showed that the incongruent group agreed significantly less with the statement that in-game advertising makes the game seem more realistic and immersive (M = 2.84, SD = 1.24) than did those in the congruent group (M = 2.34, SD = .772, F(1, 98) = 5.89, p = .017). Thus, players in the incongruent group perceived in-game advertising as less contextual than did players in the congruent group.

Discussion

This study indicates that though incongruent advertising placement in online video games can trigger high advertising awareness, it also can reduce a game's perceived sense of realism if not coordinated with the game environment. Whereas previous research (Activision 2005; Book 2004; Chaney, Lin, and Chaney 2004; Nelson 2002; Nelson, Keum, and Yaros 2004) shows that players can accept in-game advertising when relevant, we suggest the opposite can occur when advertisers make little or no effort to contextualize their ads within the game world by placing incongruent ads.

Why, then, did the advertising in this study seem to provoke ambivalent responses in so many players? Simply put, most of the ads did not match their environment. A significant number of players in the incongruent group perceived in-game ads as less contextual than did those in the congruent group, with reasonable cause. Current-day ads placed in a sci-fi game based 30,000 years in the future cannot possibly boost the game's sense of realism. Furthermore, the advertisements are extremely incongruent with the advertising environment. Anarchy Online offers a virtual adventure grounded entirely in the distant future yet inundated with marketing efforts of the 21st century-unless players pay a monthly fee for the ad-free version.

If the gameplay mechanics in Anarchy Online were focused primarily on player combat, perhaps real-life military advertisements would seem congruent, but we do not find support for this claim. Anarchy Online is a multifaceted virtual environment in which players can experience the game as they see fit, complete with in-game social areas, guild houses, apartments, bars, and several layers of player customization. The game is arguably no more focused on combat than it is on socializing or exploration, and this study's participants were never exposed to combat during the game sessions. Moreover, the game's blatant futuristic setting is a poor venue for present-day imagery as congruent objects; abundant teleport stations, floating video screens, and ultra-sleek attire all contribute to the idea of a world set in the faraway future.

Only two brands (Under Armour and Subway) enjoyed substantial awareness rates in this study. These two ads contained no period-specific material in their presentation; only Under Armour's logo was shown, and, in Subway's case, two abstract characters and a brief message ("Fresh Toasted") were displayed. Both ads used simple, eye-catching designs and likely appeared more convincing than other billboards in the game's futuristic setting, assuming that players could believe that both brands would still exist in 30,000 years. As such, these ads could be considered moderately incongruent, which enabled players to resolve it. Other ads, such as those for the Navy and Air Force, depicted examples of present-day people and technology, compromising their credibility in a futuristic setting. These observations offer a further indication that acceptable in-game advertising can be moderately incongruent and still enhance a game's sense of realism. Moderately congruent advertising though may offer the best condition for player acceptance and maintenance of immersion.

Contextual and congruent in-game advertising begins by matching appropriate brands and products to the game worlds. Just as a McDonald's logo has no place in World of Warcraft, certain products should not be advertised where they do not normally belong. Rather than trying to monetize gameplay by placing irrelevant ads in several games, in-game ad firms should work more closely with agencies and demographics services to deliver what players are truly interested in seeing. For example, in an MMORPG such as Anarchy Online, tech-savvy gamers may be far more interested in moderately congruent ads offering player-exclusive discounts on electronics at Best Buy or NewEgg.com than incongruent hairstyling products offering the same message as they do in television campaigns. Moreover, by making such advertisements truly interactive to players, such as by adding a clickable interface to offer more information or open the product's site in a browser, advertisers could build further engagement with their audience, and players may perceive a substantial benefit from engaging with the ads.

Conclusions

Corporate advertisers, media publishers, game developers, and ad agencies still have much to learn about the relatively nascent world of in-game advertising. Advertisers have found their captive audience of men aged 18-34 years, and they are reachable in a medium arguably more engrossing than television, radio, or movies; however, matching relevant brand messages with contextual game environments remains a challenge. Constructing moderately incongruent ads is even more difficult. In this study, we sought to examine the effects of advertising in an online role-playing game, in the hopes that all parties involved with in-game advertising may be better informed of the advantages and downfalls of placing ads in games.

Our results also suggest that when employing in-game advertising, advertisers should strive to produce ads that are moderately congruent with the game environment. Faced with the high cognitive load of a virtual world gaming environment, players do not seem to have the capacity to process fully incongruent advertising. A substantial percentage of participants in the congruent group (31.3%) believed that advertising in role-playing games could be considered appropriate and realistic. These results, coupled with Nicovich's (2005) findings, suggest that advertising can be executed effectively within the role-playing game genre if the material "makes sense" in the game environment-for example, in games based in modern settings rather than pure fantasy worlds or period-specific environments (e.g., medieval settings, futuristic worlds, outer space).

The results that show more players from the incongruent group reported in-game advertising as annoying or obtrusive imply that incongruent ads negatively affected players' schemata for the game. If players in the study were already familiar with popular MMORPGs, which typically contain no advertising, their schemata may have instantly changed on seeing billboards in Anarchy Online. In contrast, players who may have been accustomed to seeing advertising in other types of games may not have had their schemata similarly affected.

Differences in advertising acceptance between the genders are interesting to note. Female players were significantly more accepting of in-game advertising overall and believed that in-game advertising actually contributed to immersion and realism. Perhaps female players are accustomed to advertising in Web-based casual games space, a section of the market that enjoys a 76% female player population (Dobson 2006) and is commonly monetized by ads. The same results suggest that certain advertisers may discover significantly more positive reactions to marketing in female-oriented virtual worlds, such as Second Life, as long as the advertised material corresponds with the needs of the product's target audience and is congruent with the environment.

Nongamers showed similar responses; they found advertising in movies, television, and games less annoying or obtrusive than did avid gamers and agreed more that in-game advertising makes games seem more realistic. Again, this finding suggests that players without preexisting schemata about MMORPGs will be far less affected in their schemata by advertising in the game environment than will those who actively play similar games and have developed their own systems of expectations.

Players in the incongruent group found advertising in television shows significantly more annoying and obtrusive than did those in the congruent group. These results suggest that players' negative attitudes toward the advertising in Anarchy Online may have influenced their opinions, albeit possibly temporarily, about advertising in other media. Whereas Nelson, Keum, and Yaros (2004) find that those players who respond negatively to in-game advertising are often opposed to advertising in general, the results of our study revealed no significant difference in reactions to advertising in general or advertising in movies between the two groups. This exception may suggest that avid gamers mentally associate product placement in television shows more closely with advertising in games than with product placement in movies.

We were somewhat surprised to find participants in the incongruent group had rated the game's ease of navigation significantly lower than those in the congruent group, though these results support Bellman and Rossiter's (2004) findings that users consider congruent Web sites easier to navigate than those less congruent with their expectations. Players in the incongruent group may have experienced confusion when encountering incongruent ads in an unfamiliar space; perhaps others spent time taking in the advertising and, in retrospect after the game session, realized they could have better spent that time navigating the environment with other players. When players perceive a game environment as unusually troubling to navigate due to the presence of overbearing advertising, the immersive experience inarguably has been compromised. To deliver a marketing message effectively, advertisers must be sure not to interrupt a player's sense of immersion in the game. Developers and advertisers must work together to avoid consequences such as these, through inoffensive presentation and cautious ad placement.

In summary, this study suggests that out-of-context advertising placed in games can weaken the immersive experience for players on a variety of levels. Players might not be able to justify their complaints if the games are free to play, though this is not an ideal model for attracting paying customers. Fully incongruent advertisements will not have the desired effect anyway. Although this study reveals that participants who played the game with ads present were most likely to pay more than regular retail prices for ad-free versions of games in which they were interested, the companies behind in-game advertising would be remiss if they simply annoyed their customers into spending more. In-game ad networks must strive to match relevant advertising with appropriate game settings to ensure positive results on all sides.

Limitations and Strengths

Because participants were recruited from a large undergraduate class without random sampling, the results of this study cannot generalize to all video game players. Yet the research should offer an informative basis for further studies on advertising in video games, because in-game advertisers frequently spend to reach this very demographic. Massive Incorporated could not divulge specific in-game advertising cycle dates and times, so it was impossible for us to predict when particular ads would leave the current cycle or new ads would appear. Fortunately, all brands represented in advertising at the start of the study remained until the incongruent group's final game session; however, the Garnier Fructis ads entered the cycle after 39 participants had already completed their sessions, rendering the item virtually incomparable to the other brands in-game.

The sample size (N = 100) compares favorably with previous game research, such as Nelson's (2002) study (N = 16) and Chaney, Lin, and Chaney's (2004) work (N = 42). A predetermined play time of 30 minutes also presumably allowed for more conclusive results than previous studies. Chaney, Lin, and Chaney (2004) and Nelson (2002) both implemented play times of 15 minutes; Activision (2005) allowed a minimum of 20 minutes for play time; and Nelson, Keum, and Nelson (2004) allowed exposure to racing games for only three minutes.

Further Research

The processes and technologies behind in-game advertising have advanced at a remarkable rate. Since the inception of this study, brick-and-mortar clothing retailers and ad agencies have set up shops in virtual worlds, in-game ad networks have introduced interactive billboards to multiplayer game environments, and acquisitions worth hundreds of millions of dollars have been inked with framework providers. With the advent of next-generation consoles and a steadily increasing worldwide gaming population, the opportunities for research in this space are endless.

Aarseth (2001, p. 9) describes the video game business as "a billion dollar industry with almost no basic research"; it is the responsibility of researchers from virtually all fields to continue to fill this void. In the interest of examining the comparative effectiveness of in-game advertising, research should expose participants to advertising in video games and traditional advertising vehicles (e.g., magazines, television, online) in the same study and compare recall rates and response attitudes across the various media.

Activision and Nielsen Entertainment took a step in this direction by including three cells focused on television advertising (Activision 2005); however, further experiments should be conducted to compare recall and attitudes toward advertising in all media with those toward in-game ads to help pinpoint potential areas of intrinsic opportunity in games that other media cannot provide. It may also be worthwhile to investigate the effect of in-game advertising on players' schemata of brand recognition and other perceptions within a game environment.

Advertising in virtual worlds should receive increased academic attention, particularly regarding perceived and intended interaction between users and ads and the effects of social trends on virtual purchasing habits. With subscription rates and virtual economies skyrocketing as word spreads of the creative and lucrative possibilities available in virtual worlds, researchers should actively observe and participate in these worlds as they grow. The residents of these worlds have already devised internal advertising networks; it will be fascinating to watch for the further involvement of real-world advertisers and the impact of their presence in a fantasy-based setting.

References

Aarseth, Espen (2001), "Computer Game Studies, Year One," Game Studies: The International Journal of Computer Game Research, 1 (1), <http://www.gamestudies.org/0101/editorial.html> (accessed November 28, 2005).

Activision (2005), "Activision and Nielsen Entertainment Release Results of Pioneering Research on In-Game Advertising," Press Release (December 5, 2005), <http://investor.activision.com/ReleaseDetail.cfm?ReleaseID=181109> (accessed February 4, 2006).

Bartlett, Frederic C. (1932), Remembering: A Study in Experimental and Social Psychology. Cambridge: Cambridge University Press.

BBC News (2005), "Global Ad Network Eyeballs Gamers," <http://news.bbc.co.uk/1/hi/technology/4485959.stm> (accessed November 28, 2005).

Bellman, Steven and John R. Rossiter (2004). "The Website Schema," Journal of Interactive Advertising, 4 (Spring), <http://www.jiad.org/article48> (accessed June 1, 2009).

Book, Betsy (2004), "These Bodies are FREE, so Get One NOW! Advertising and Branding in Social Virtual Worlds," <http://ssrn.com/abstract=536422> (accessed November 8, 2005).

Brewer, William T. and Treyens, James C. (1981), "Role of Schemata in Memory for Places," Cognitive Psychology, 13 (2), 207-230.

Bush, Alan J., Victoria Bush, and Sharon Harris (1998), "Advertiser Perceptions of the Internet as a Marketing Communications Tool," Journal of Advertising Research, 38 (2), 17-28.

Campbell, Colin (2006), "Jay Cohen," Next Generation, <http://www.nextgen.biz/index.php?option=com_content&task=view&id=2331> (accessed February 25, 2006).

Campbell, Margaret C. and Goodstein, Ronald C. (2001), "The Moderating Effect of Perceived Risk on Consumers' Evaluations of Product Incongruity: Preference for the Norm," Journal of Consumer Research,28 (3), 439-449.

Caoili, Eric (2008), "NPD: Only 3 Percent of Gamers Are ‘Extreme,' Most Are ‘Young and Heavy,'" Gamasutra, <http://www.gamasutra.com/php-bin/news_index.php?story=19802> (accessed September 9, 2008).

Castronova, Edward (2005), Synthetic Worlds: The Business and Culture of Online Games. Chicago: University of Chicago Press.

Chaney, Isabella M., Ku-Ho Lin, and James Chaney (2004), "The Effect of Billboards Within the Gaming Environment," Journal of Interactive Advertising, 5 (1), 54-69.

di Cesare, Christopher (2005), "The Name of the Advergame," Brand Strategy, 195, 5.

Dobson, Jason (2006, September 13). "Survey: PopCap Releases Casual Game Findings," Gamasutra, <http://www.gamasutra.com/php-bin/news_index.php?story=10861> (accessed September 15, 2006).

Donaton, Scott (2003), "We Unravel the Mystery of Where the Boys (18-34) Are," Advertising Age, 74 (44), 23.

Fiske, Susan T. and Patricia W. Linville (1980), "What Does the Schema Concept Buy Us?" Personality and Social Psychology Bulletin, 6, 543-557.

Gibson, Ellie (2006), "New In-Game Advertising Agency Launches," GamesIndustry.biz, <http://www.gamesindustry.biz/content_page.php?aid=13991> (accessed May 31, 2006).

Graft, Kris (2006a), "TWST: ‘Advergaming' to Hit $4 Bln in '08," Next Generation, <http://next-gen.biz/index.php?option=com_content&task=view&id=2109> (accessed February 4, 2006).

--- (2006b), "DFC: Games Industry Worth $44 Bln in 2001," Next Generation, <http://www.next-gen.biz/index.php?option=com_content&task=view&id=3885&Itemid=2> (accessed September 27, 2006).

Hershman, Tania (2005), "Advertisers: Game On," Technology Review, <http://www.technologyreview.com/articles/05/05/issue/forward_advertisers.asp> (accessed November 8, 2005).

IGA Partners (2004), "Key Facts about In-Game Advertising," <http://www.ingameadvertising.com/docs/IGA_factsheet.pdf> (accessed November 28, 2005).

Lee, Sang Yeal and Fuyuan Shen (2006), "Joint Advertising and Brand Congruity: Effects on Memory and Attitudes," in American Academy of Advertising Conference Proceedings, 73-74.

Mandler, George (1982), "The Structure of Value: Accounting for Taste," in Affect and Cognition: The 17th Annual Carnegie Symposium, M.S. Clark and S.T. Fiske, eds. Hillsdale, NJ: Lawrence Erlbaum Associates, 3-36.

McMillan, Sally J. and Jang-Sun Hwang (2002), "Measures of Perceived Interactivity: An Exploration of the Role of Direction of Communication, User Control, and Time in Shaping Perceptions of Interactivity,"Journal of Advertising, 31 (3), 29-42.

Meyers-Levy, Joan and Alice M. Tybout (1989), "Schema Congruity as a Basis for Product Evaluation," Journal of Consumer Research, 16 (1), 39-54.

Nelson, Michelle R. (2002), "Recall of Brand Placements in Computer/Video Games," Journal of Advertising Research, 42 (2), 80-93.

---, Heejo Keum, and Ronald A. Yaros (2004), "Advertainment or Adcreep? Game Players' Attitudes toward Advertising and Product Placements in Computer Games," Journal of Interactive Advertising, 5 (1), 3-30.

Nicovich, Stefan G. (2005), "The Effect of Involvement on Ad Judgment in a Video Game Environment: The Mediating Role of Presence," Journal of Interactive Advertising, 6 (1), 38-51.

Oser, Kris (2005a), "WildTangent, Massive, IGN Stake Claims on Frontier of $900M Market," Advertising Age, 76 (16), 6-7.

--- (2005b), "Grand Theft Rating Flap Could Spook Marketers," Advertising Age, 76 (30), 3.

Peracchio, Laura A. and Alice M. Tybout (1996), "The Moderating Role of Prior Knowledge in Schema-Based Product Evaluation," Journal of Consumer Research, 23 (3), 177-192.

Reynolds, Christopher (2004), "Game Over," American Demographics, 26 (1), 34-38.

Sanders, Adrienne (2006), "S.F. is Ground Zero for Emerging Video Game Advertising Industry," San Francisco Business Times, <http://sanfrancisco.bizjournals.com/sanfrancisco/stories/2006/05/29/newscolumn1.html> (accessed May 31, 2006).

Schank, Roger C. and Robert P. Abelson (1977), Scripts, Plans, Goals, and Understanding: An Inquiry into Human Knowledge Structures. Hillsdale, NJ: Lawrence Erlbaum Associates.

Schmidt, Toni L. and Jacqueline C. Hitchon (1999), "When Advertising and Public Relations Converge: An Application of Schema Theory to the Persuasive Impact of Alignment Ads," Journalism & MassCommunication Quarterly, 76 (3), 433-455.

Schwarz, Joyce (2005), "R-U-GAME?" iMedia Connection, <http://www.imediaconnection.com/content/5354.asp> (accessed November 18, 2005).

Sennott, Sarah (2005), "Gaming the Ad," Newsweek, 145 (5), E2.

Speck, Paul Surgi, David W. Schumann, and Craig Thompson (1988), "Celebrity Endorsements-Scripts, Schema and Roles: Theoretical Framework and Preliminary Tests," in Advances in Consumer Research, Vol. 15, Michael J. Houston, ed. Provo, UT: Association for Consumer Research, 69-76.

Stanley, T.L. (2004), "Massive Launches Ad Server for Games," Advertising Age, 75 (42), 36-38.

Stayman, Douglas M., Dana L. Alden, and Karen H. Smith (1992), "Some Effects of Schematic Processing on Consumer Expectations and Disconfirmation Judgments," Journal of Consumer Research, 19 (2), 240-255.

Sundar, Shyam S. and Jinhee Kim (2005), "Interactivity and Persuasion: Influencing Attitudes with Information and Involvement," Journal of Interactive Advertising, 5 (2), 6-29.

Sujan, Mita, James R. Bettman, and Harish Sujan (1986), "Effects of Consumer Expectations on Information Processing in Selling Encounters," Journal of Marketing Research, 23 (November), 346-353.

Takahashi, Dean (2000), "Games Get Serious," Red Herring, <http://www.redherring.com/Article.aspx?a=2405> (accessed October 21, 2005).

Wegert, Tessa (2005), "Gaming 101," ClickZ Network, <http://www.clickz.com/experts/media/media_buy/article.php/3550216> (accessed November 28, 2005).

Wyer, Robert S., Thomas K. Srull, Sallie E. Gordon, and Jon Hartwick (1982), "Effects of Processing Objectives on the Recall of Prose Material," Journal of Personality and Social Psychology, 43 (4), 674-688.

About the Authors

Ben Lewis (M.M.C., Louisiana State University) is a marketing consultant and community manager in the game industry in Austin, Texas. His research interests include in-game advertising, virtual economies, online communities, and emerging trends in game industry business models. E-mail: benblewis@gmail.com.

Lance Porter (Ph.D., University of Georgia) is the Advertising Area Head in the Manship School of Mass Communication at Louisiana State University. He has focused on digital media since 1995, when he built his first commercial Web site. His research focuses on how digital media affect communication and culture. He holds a joint appointment with the Center for Computation and Technology (CCT). E-mail: lporter@lsu.edu.