Examining the Role of Exposure to Incongruent Messages on the Effect of Message Framing in an Internet Health Search

Abstract

This study examines the effects of exposure to messages incongruent with one’s motivational orientation. In a factorial design with regulatory orientation and message frame as independent variables, participants (N = 106) conducted an information search on a web program. Participants selected online information that was congruent with their activated motivational orientation. Compatibility effects resulted in promotion orientation/gain frame and prevention orientation/loss frame participants reporting more favorable attitudes than promotion orientation/loss frame and prevention orientation/gain frame participants irrespective of exposure to messages incongruent with the activated motivational orientation. A similar pattern of results occurred with message recall. For behavioral intention, significant differences occurred for only the promotion orientation condition where gain-framed messages elicited greater behavioral intentions than the loss-framed message.

Keywords

Internet search health messages motivational orientation goals regulatory focus

Although health communication campaign messages are increasingly delivered via the Internet (Noar, 2006) and through other communication technologies that are often bundled with Internet access (Kreps & Neuhauser, 2010; Rice & Katz, 2001), studies on message design rarely take into account the role of the Internet search in the processing of such messages. Health messages commonly emphasize the positive or negative consequences of a health behavior. That a message framed to accentuate the consequences of a health behavior can lead to persuasive effects is but one of the reasons for the steady stream of research on message-framing effects in the literature. While there has been research on framing effects over the past three decades, the pattern of framing effects has not been consistent and a parsimonious explanation for such framing effects has been elusive (Lee & Aaker, 2004; O’Keefe & Jensen, 2009). Most of these studies have been based on health issues (Levin, Schneider, & Gaeth, 1998) and, in particular, the persuasive effects of exposure to a single health message. What has not been considered is that the exposure to one health message often leads to exposure to other related health messages. Such exposure to multiple messages is particularly relevant to the Internet where 80% of Internet users have searched online for health information (Fox, 2011).

The current study relies on regulatory focus theory (Higgins, 1997), a goal pursuit theory, to predict the effectiveness of message goal frames. Regulatory focus theory posits that individuals tend to construe goals in distinct ways and pursue goals by utilizing different strategies (herein referred to as regulatory or motivational orientation). While individuals tend to have one motivational orientation (Higgins, 1997), it is possible to activate temporarily a motivational orientation through exposure to a health message (Lee & Aaker, 2004; Spiegel, Grant-Pillow, & Higgins, 2004). The current study utilizes two types of health messages—one activates a motivational orientation (or the strategies utilized to pursue goals) and one pertinent to the outcome of the goals (i.e., gain- or loss-framed messages). The current study examines, first, the informational choices made by information seekers when exposed to these two types of messages and, second, whether previously documented persuasive effects (Lee & Aaker, 2004; Spiegel et al., 2004) remain salient when information seekers are exposed to messages that are incongruent with an activated motivational orientation.

Health Messages and Goal Outcomes

The current study utilizes the message framing concept that originated from prospect theory (Kahneman & Tversky, 1979) and the subsequent research in the persuasion literature (Dillard & Marshall, 2003; O’Keefe & Jensen, 2009; Rothman, Bartels, Wlaschin, & Salovey, 2006). This conceptualization of message frames in the persuasion literature is in contrast to the message framing literature in mass communication where a frame pertains to a central idea that emerges through selection, emphasis, exclusion, and elaboration of a topic (Gamson, 1992; Scheufele, 1999). In the mass communication literature, for example, whereas individuals exposed to newscasts with an episodic frame (e.g., depiction of specific events) tend to attribute responsibility of issues to individuals rather than society, individuals exposed to newscasts with a thematic frame (e.g., depiction of general evidence) tend to attribute responsibility of issues to society (Iyengar, 1991). While framing in both the mass communication and persuasion literature suggests that subtle cues in the presentation of a message can influence a person’s decision making, a distinction is that mass communication frames do not stipulate a specific goal outcome. For example, one study spoke of news frames on urban growth as being positive or negative but not whether they stipulated a specific policy recommendation (Shah, Kwak, Schmierbach, & Zubric, 2004). To contrast, in the persuasion literature, a distinguishing characteristic of goal framing, which is the basis of the current study, is that both frames promote the same behavior (i.e., goal), but one frame emphasizes the positive consequences of a behavior whereas the other frame emphasizes the negative consequences of not performing a behavior (Dillard & Marshall, 2003; O’Keefe & Jensen, 2006; Rothman et al., 2006). In addition, message frames in the persuasion literature depict an outcome as one that is attained or avoided (O’Keefe & Jensen, 2006). The latter characteristic of message goal frames in the persuasion literature is in contrast to the general use of frames in the mass media literature where description of attaining or avoiding an outcome is generally not made. Gain/loss frames are also sometimes referred to as advantage/disadvantage and positive/negative frames (Shen & Dillard, 2007). The term gain/loss frame is used in the current study.

Among the theoretical work explicating the framing effects mechanism, perhaps the most prevalent is the line of research that posits that the function of a health behavior moderates framing effects (Rothman & Salovey, 1997). Rothman and Salovey’s taxonomy of health behaviors framework posits that loss-framed messages are best suited for detection behavior whereas gain-framed messages are best for prevention behavior (Rothman & Salovey, 1997; Salovey, Schneider, & Apanovich, 2002). This taxonomy is based on prospect theory (Kahneman & Tversky, 1979; Tversky & Kahneman, 1981), which contends individuals avoid risk (i.e., are risk adverse) when they consider potential gains and seek risk (i.e., are risk seeking) when they consider potential losses associated with their decision. Thus, more uncertainty is associated with detection behavior (e.g., mammogram) because individuals risk finding out that they have a health condition that they were previously unaware of. In that sense, because individuals are risk seeking when they engage in health detection behavior, a loss-framed message would be most persuasive (Rothman & Salovey, 1997). In contrast, for messages that pertain to health prevention behaviors (e.g., wearing sunscreen), the promoted behavior is often construed as safe (i.e., risk adverse), such that gain-framed messages are most persuasive.

However, such a pattern of effects has not been documented by some research that, instead, has found no significant advantage of either loss- or gain-framed messages (Lerman et al., 1992; O’Connor, Pennie, & Dales, 1996; Wilson, Wallston, & King, 1990). A meta-analytic review found that, while gain frames were more persuasive than loss frames for disease prevention topics, loss frames were not more persuasive for disease detection behavior (O’Keefe & Jensen, 2006). In another meta-analysis that distinguished between specific types of detection behavior, loss frames had a small, but significant, advantage for detection behavior, but breast cancer detection accounted for most of the effect (O’Keefe & Jensen, 2009). This ambiguity in findings undermines the assumption that prevention behaviors are perceived as low risk whereas detection behaviors are perceived as high risk (O’Keefe & Jensen, 2006). For example, a gain-framed message elicited a higher HIV testing rate but only when participants were certain that they were HIV negative (Apanovitch, McCarthy, & Salovey, 2003). That is, while HIV is a health detection behavior, those who were confident that they were not HIV positive likely construed the test to be risk adverse and, thus, were persuaded by the gain-framed message.

Such sensitivity to positive or negative consequences is the basis of regulatory focus theory (Higgins, 1997, 1998), which suggests that differences in the representation of goals (i.e., promotion vs. prevention orientation) moderate the effects of framing appeals (Lee & Aaker, 2004; Spiegel et al., 2004). Specifically, individuals with a promotion orientation are more persuaded by gain-framed messages whereas individuals with a prevention orientation respond better to loss-framed messages (Lee & Aaker, 2004). This pattern of effects is similar to the aforementioned taxonomy of health behaviors framework, as well as research on regulatory focus (Rothman et al., 2006), which is elaborated upon in the next section. That is, individuals considering a health detection behavior may be induced to a prevention orientation whereas consideration of a health prevention behavior may induce a promotion orientation (Rothman et al., 2006).

Motivation and Regulatory Focus Theory

Individual predispositions drive information-seeking behavior (Johnson, 1997). In particular, motivational factors drive information seeking in various contexts, including interpersonal exchanges (Afifi & Weiner, 2004), the mass media (David, 2009), and the Internet (Kim, 2007). For example, different dimensions of motivation (e.g., surveillance, social utility) can lead to differential exposure and attention to news information seeking (David, 2009). The literature suggests that intrinsic motivation can influence information selection behavior. For example, personal issue importance (i.e., a form of intrinsic motivation) influences online selection of political content (Kim, 2007). Health consciousness, yet another form of intrinsic motivation, predicts whether individuals will seek additional health information beyond the physician (Dutta-Bergman, 2005). These studies highlight how individual differences, and in particular traits and dispositions, moderate media effects (Oliver, 2002; Oliver & Krakowiak, 2009).

Regulatory focus theory (Higgins, 1997, 1998) posits that there are two fundamental self-regulatory systems (i.e., promotion orientation and prevention orientation), each of which has a preferred means of pursing goals. Regulatory orientation is a motivation orientation in which both promotion- and prevention-oriented individuals desire achieving goals but differ in how they construe goals and go about achieving those goals. Promotion-oriented individuals view goals as aspirations and prefer to utilize eagerness (approach) strategic means to accomplish goals (Higgins, 2000). Thus, promotion-oriented individuals are concerned with advancements. In contrast, prevention-oriented individuals view goals as obligations and prefer to utilize vigilance (avoidance) strategic means to accomplish goals (Higgins, 2000). Thus, prevention-oriented individuals are concerned with safety. The eagerness (approach) and vigilance (avoidance) strategies for goal pursuits are conceptually similar to goal framing. Specifically, gain frames emphasize the favorable outcomes of performing a behavior (approach), and loss frames emphasize the unfavorable outcomes of not performing a behavior (avoidance) (Levin et al., 1998). Furthermore, approach and avoidance behaviors are commonly utilized in the management of information in health contexts (Brasher, Goldsmith, & Hsieh, 2002).

This conceptual similarity was supported by research on the interactive effect of regulatory focus and message frames (Lee & Aaker, 2004; Spiegel et al., 2004). Related research on individual sensitivity to positive or negative outcomes has found similar framing effects (Shen & Dillard, 2007). Specifically, the behavioral inhibition/behavioral approach systems (BIS/BAS) is based on the premise that two systems regulate behavior—approach and avoidance systems. BIS (i.e., avoidance system) controls aversive motivation while BAS (i.e., approach system) controls appetitive motivation. Whereas BIS is associated with cues of nonrewards, BAS is sensitive to cues of rewards. That BIS/BAS is conceptually similar to regulatory orientation (Rothman et al., 2006) would suggest a similar pattern of framing effects. Indeed, advantage frames (i.e., gain frames) are associated with BAS whereas disadvantage frames (i.e., loss frames) are associated with BIS (Shen & Dillard, 2007). Although these two concepts are analogous, regulatory orientation can be momentarily activated through message exposure (Lee & Aaker, 2004; Spiegel et al., 2004) whereas BIS/BAS is a chronic orientation that requires a self-reported measure scale for assessment (Shen & Dillard, 2007).

It has been demonstrated that individuals pursue goals that complement their motivation orientation (Elliot & McGregor, 2001). When individuals are in situations that support their preferred goal pursuit strategies (eagerness vs. vigilance), their orientation (promotion vs. prevention) is said to align (or fit) with their preferred goal-pursuit strategies (Higgins, 2000). Such alignment results in “regulatory fit” or compatibility effects. Thus, regulatory fit effects occur when (1) motivation orientation (promotion vs. prevention) aligns with (2) the preferred strategies of goal pursuits (eagerness vs. vigilance).

One mechanism for the regulatory fit principle is the enhanced strength of engagement (Higgins, 2006). Specifically, this process mechanism for regulatory compatibility leads to people’s ease of processing a message (Lee & Aaker, 2004). Lee and Aaker found that, under regulatory fit, participants reported greater fluency in processing messages. The persuasiveness associated with the pairing of promotion orientation/gain frame and prevention orientation/loss frame messages is consistent with the larger body of research on the effectiveness of message matching. For example, prevention programs that incorporate culturally tailored content elicited more drug prevention behaviors among Mexican Americans (Hecht et al., 2003). In addition, the persuasiveness of health messages has been demonstrated when messages are matched to specific individual characteristics such as ethnic identity (Herek et al., 1998) and need for cognition (Williams-Piehota, Schneider, Pizarro, Mowad, & Salovey, 2003). When messages match an enduring characteristic of an individual (e.g., one’s motivational orientation), they are more likely to attend to the message (Brinol & Petty, 2006). Such closer attention to messages can be attributed to the higher relevance of the message per elaboration likelihood model (Petty & Cacioppo, 1986). Importantly, while individuals may exhibit one dominant regulatory orientation, either of the two orientations can be momentarily activated through messages that highlight concerns associated with promotion or prevention orientation (Cesario, Grant, & Higgins, 2004; Lee & Aaker, 2004; Spiegel et al., 2004).

Congruent and Incongruent Informational Selections in Internet Health Search

Although studies have demonstrated that message compliance results when motivation orientation (i.e., regulatory orientation) is compatible with framing appeals (Lee & Aaker, 2004; Spiegel et al., 2004), theoretical issues remain unaddressed. First, a single health message framed to elicit an intended persuasive effect is likely to lead to exposure to additional health messages on the Internet. However, the literature has not addressed the informational choices made upon exposure of such additional persuasive messages under regulatory fit. While a typical Internet information seeker may be exposed to hundreds of websites, only a few are selected in the search process. For example, while the majority of health information searchers peruse two or more websites during each session (Fox, 2006), with each website containing multiple webpages, Internet users view on average between five and eight webpages during a typical web session (Liu, Gwizdka, Liu, & Belkin, 2011).

Information seeking behavior on the web is characterized to encompass discrete stages. The information seek cycle (ISC) (David, Song, Hayes, & Fredin, 2007) is an iterative model by which the outcome of each stage influences a subsequent stage. ISC contains the following three discrete stages of the Internet search process: (1) preparation, (2) exploration, and (3) consolidation (David et al., 2007; Fredin & David, 1998). Preparation entails choosing from a menu of hyperlinks. Exploration entails information processing of the results of the aforementioned choices. Consolidation entails evaluating the results based on the goals set in the preparation stage. The information seeking process cycles again beginning with the preparation stage. The three-stage iterative model of ISC is similar to the three stages in the information management process in the theory of motivated information management for interpersonal exchanges (Afifi & Weiner, 2004). In the current study, two discrete stages of the Internet search process were examined: (1) preparation, and (2) exploration (David et al., 2007). Participants were given a menu choice of six messages, three of which had taglines for promotion concerns and three of which had taglines for prevention concerns. Participants were instructed to pick only three of the six choices (i.e., preparation stage of information seeking).

From the literature on selective exposure, individuals attend to and expose themselves to information that is congruent with their preexisting attitudes (Oliver, 2002; Zillman & Bryant, 1985). The limited effects model, which posits that the media reinforce preexisting attitudes, is based on the premise that individuals avoid counterattitudinal messages and, instead, prefer attitude-consistent messages (Klapper, 1960). Because of the abundance of information and ease of access, the Internet is a medium particularly ripe for selective exposure effects (Oliver & Krakowiak, 2009). This observation is reminiscent of a similar observation made over two decades ago when new communication technologies (e.g., cable and remote control) made television fertile ground for research on selective exposure (Zillman & Bryant, 1985). Specific to the information seeking process (Jonas, Schulz-Hardt, Frey, & Thelen, 2001) and information seeking on the Web (Knobloch-Westerwick & Meng, 2009), individuals seek information that is congruent with their preexisting attitudes. By selecting information that is complementary to preexisting attitudes, information is easier to process because individuals avoid a negative arousal state of cognitive dissonance (Festinger, 1957). It has also been found that selective exposure to congruent information occurs because such information fits with preexisting attitudinal structure (Frey, 1986). In addition, similar to the current study, selective exposure effects have been found when participants were given just article titles (or one-line text descriptions of information) (Brannon, Tagler, & Eagly, 2007; Jonas et al., 2001), news lead paragraphs (Knobloch-Westerwick & Meng, 2009), and when a limited amount of information is presented (Fisher, Jonas, Frey, & Schulz-Hardt, 2005). Thus, in the context of the current study, we would expect that individuals would select information that is congruent with their motivational orientation (i.e., promotion vs. prevention orientation) when presented with information that highlights both promotion- and prevention-oriented concerns.

However, while individuals may select more informational messages that are congruent with their motivational orientation, a minority of the selected messages may be incongruent with their motivational orientation. For example, across communication channels (e.g., web, TV, newspapers), while individuals attend to information that is consonant with their preexisting political attitudes, they are also exposed to information dissonant from their preexisting attitudes (Chaffee, Saphir, Graf, Sandvig, & Hanh, 2001). However, the literature also suggests that selective exposure affects subsequent information processing in that individuals attend to and recall information that is congruent with their initial attitudes (Oliver, 2002; Smith, Fabrigar, Powell, & Estrada, 2007). That is, attitude-incongruent information is not as centrally processed as attitude-congruent information and is not processed and recalled to the degree of attitude-congruent information. In summary, individuals tend to select, perceive, and recall information that is consonant with their preexisting attitudes and beliefs (Oliver, 2002). Thus, the effect of message framing and motivational orientation predicted by regulatory fit should occur irrespective of the amount of exposure to information that is incongruent with one’s motivational orientation. These observations are articulated in the following hypotheses:

Hypothesis 1a (H1a): Individuals with promotion orientation will select more information congruent with their respective motivational orientation.

Hypothesis 1b (H1B): Individuals with prevention orientation will select more information congruent with their respective motivational orientation.

Hypothesis 2a (H2a): Compatibility between message frame and motivational orientation results in more favorable attitudes toward recommended behavior irrespective of exposure to information incongruent to motivational orientation.

Hypothesis 2b (H2b): Compatibility between message frame and motivational orientation results in greater behavioral intention irrespective of exposure to information incongruent with motivational orientation.

That messages are easier to process when gain-framed and loss-framed messages are paired with promotion and prevention orientation, respectively (Lee & Aaker, 2004), suggests that enhanced message recall should occur when the motivational orientation is aligned with the corresponding message frame. Such processing fluency is the result of the enhanced strength of engagement, which is one effect that results when goal pursuits align with motivational orientation (Higgins, 2006). While previous research found self-reported processing fluency for the promotion orientation/gain frame and prevention orientation/loss frame participants (Lee & Aaker, 2004), such research was limited to self-reported ease of processing. The extent of message recall through standard techniques (e.g., thought listing) has not been previously observed. These observations are articulated in the last hypothesis:

Hypothesis 3 (H3): Compatibility between message frame and motivational orientation results in higher message recall.

Method

Participants and Design

The sample (N = 106) consisted of undergraduate and master’s degree students (34 men, 72 women, M_age = 25.26 years) who participated in exchange for course extra credit. In terms of ethnicity, 58% were Caucasian, 20% African American, 4% Hispanic, 14% Asian, and 4% Other. The experiment occurred in a designated computer lab where participants individually interfaced a web program, which was designed specifically for this study. Participants were randomly assigned by the web program to one of four conditions in a 2 (motivational orientation: promotion vs. prevention orientation) × 2 (message frame: gain vs. loss frame) between-subjects design. Participants accessed the program using a unique login and password. Clickstream data from the session log file captured user behavior, including the health messages selected during the preparation stage of information seeking. Participants were unaware that their online behavior (i.e., the messages they selected by clicking on the computer mouse) were being recorded on a server-side log file, which matched the unique login with the online behavior. The log file also recorded the experimental condition for each participant. Transactional log files permit unobtrusive collection of online behavior that is superior to participant recall of online activities (Han, 2011). The combination of log files with survey data, which was used in the current study to assess information seeking behavior, is one method for assessing eHealth applications (Han, 2011). The current study was approved by the IRB at Tulane University.

Procedure

Prior to assignment of treatment groups, participants completed a questionnaire that contained demographic questions and questions pertaining to their consumption of fruits and vegetables for the past 30 days, which is from the Behavioral Risk Factor Surveillance Survey (BRFSS) (CDC, 2010).

Regulatory orientation induction

Participants were then given a unique login and password to access the web program, which randomly assigned participants to one of the four treatment groups. On the web browser, participants were exposed for 2 minutes to either a promotion or prevention orientation message, which was designed to induce temporarily a motivational orientation.

Message frame

After 2 minutes, either a gain- or loss-framed message appeared directly below the motivational orientation message. The message frame appeared for 30 seconds, after which the web program instructed participants to complete a self-administered survey that contained items pertaining to the manipulation check and attitudinal measures.

Preparation stage of information seeking

Upon completion of the survey, the web program brought participants to a new webpage that had a list of six taglines that pertained to six additional messages on fruit and vegetable consumption. The six taglines were worded with concerns associated with promotion orientation (i.e., advancement) or prevention orientation (i.e., safety). The participants checked three of the six taglines, which were linked to messages worded with their respective motivational orientation. The three promotion-oriented taglines were as follows: create more energy, improve digestive health, and strengthen bones. The three prevention-oriented taglines were as follows: reduce risk of cancer, lower blood pressure, and prevent heart disease. The web program randomly listed the taglines. Instructions on the screen informed participants that they should select three informational messages based on the six taglines that were of interest to them. The web program was designed such that participants could proceed only if they selected exactly three taglines. No time restrictions were placed on the selection process.

Exploration stage of information seeking

Upon selection of the three informational messages based on the taglines, the web program brought participants to a new webpage that displayed the actual messages. A menu on the left side of the screen displayed the taglines for the three selected informational messages. By clicking on the menu buttons, messages appeared one at a time and appeared to the right of the menu. Participants could click on menu buttons and reread their three messages as many times and in any order that they wanted. All six messages were written at the Flesch-Kincaid grade level of 12 and were approximately 101 to 112 words in length. No time limits were given for participants’ reading their three selected messages. When participants completed this stage of information seeking, they clicked the “finish” tab on the menu bar, which closed the web program. Participants were then asked to recall what they remembered from the three informational messages they had selected. Participants wrote these thoughts on a blank piece of paper that was subsequently coded for analysis. Participants then completed a questionnaire that had items on attitudes toward fruits and vegetables and intention to eat more fruits and vegetables.

Dependent Variables

Selected information

The proportions of promotion-oriented and prevention-oriented informational messages were calculated for participants from log files that captured their online user behavior, including their message selection during the exploration stage of information seeking. For instance, the proportion of promotion-oriented information messages was the number of promotion-oriented informational messages selected divided by 3 (i.e., the total number of informational messages selected).

Attitudes

Postmanipulation attitudes toward eating fruits and vegetables was measured after participants were exposed to the motivational orientation and message frame messages (α = .77)—and again at the end of the experiment when participants had read the informational pieces they had selected (α = .78). That is, attitudes were measured before the preparation stage of information seeking and again after the exploration stage of information seeking. Attitudes were measured twice to determine whether attitudinal differences were attributed to exposure to the motivational and message frame messages or to exposure to the three selected informational pieces. Because these attitudinal measures were highly correlated (r = .97, p ≤.01), the attitudinal measure assessed at the end of the experiment was used in the analysis. Attitudes were assessed by four questions on a 5-point scale anchored by the adjectives boring-fun, important-unimportant, unpleasant-pleasant, un-enjoyable-enjoyable, with higher scores reflecting more favorable attitudes.

Behavioral intention

A 7-point item that assessed participants’ intention to eat more fruits and vegetables was worded as, “How likely are you to eat more fruits and vegetables in the coming next month?” Response ranged from 1 (very unlikely) to 7 (very likely).

Recall

Message recall was measured by the proportion of thoughts correctly recalled from the thought listing task. In coding the correctness of thoughts provided by participants, two independent coders reliably coded (Cohen’s Kappa = .91) both the total number of thoughts and the number of thoughts correctly recalled. Coders determined the correctly recalled thoughts by referring to the information selected by participants, which was captured by the log files on the server. Recall was calculated by the number of thoughts correctly recalled divided by the total number of thoughts recalled.

Independent Variables

Motivational orientation

Motivational orientation (e.g., promotion vs. prevention orientation) was temporarily induced by highlighting different types of concerns (e.g., accomplishment vs. safety) within a message. This technique has been previously utilized for health topics (Lee & Aaker, 2004; Spiegel et al., 2004). For example, Lee and Aaker induced promotion orientation with a message that emphasized the energy-enhancing properties of a fruit juice and induced prevention orientation with a message that emphasized disease prevention properties of fruit juice. Thus, motivational orientation can be temporarily induced by emphasizing either the accomplishment concerns (i.e., promotion orientation) or safety concerns (i.e., prevention orientation) of a specific health issue. In the current study, the promotion orientation message emphasized the accomplishment concerns of eating fruits and vegetables whereas the prevention orientation message emphasized the safety concerns. The messages that were pilot tested contained approximately the same number of words (146 words for promotion orientation vs. 144 words for prevention orientation) and had the same level of reading difficulty (Flesh-Kincaid grade level 10).

Participants in the promotion orientation condition read the following message:

We need a wide range of vitamins and minerals for good health. Fruits and vegetables provide the necessary nutrients to buffer our bodies from the physical demands of the world, including pollution and stress. For example, medical research suggests that eating fruits and vegetable may contribute to healthy eyesight. Growing evidence suggests that diets rich in fruits and vegetables may create better eyesight, including improved night vision. Night vision is the ability to see in dark environments and is important for night driving. Many of the vitamins and minerals found in fruits and vegetables, including vitamin C and vitamin A, have been associated with creating better vision. In addition, dark green leafy vegetables contain pigments that appear to be able to help improve eyesight. Fruits and vegetables are clearly an important part of a good diet. You can choose from a variety of fruits and vegetables.

Participants who were randomly assigned to the prevention orientation treatment read the following message:

We need a wide range of vitamins and minerals for good health. Fruits and vegetables provide the necessary nutrients to buffer our bodies from the physical demands of the world, including pollution and stress. For example, medical research suggests that eating fruits and vegetable may contribute to healthy eyesight. Growing evidence suggests that diets rich in fruits and vegetables may reduce the risk of developing eyesight problems, including cataract. Cataract is the gradual clouding of the eye’s lens and can lead to blindness. Many of the vitamins and minerals found in fruits and vegetables have been associated with protecting eyes from developing cataract . In addition, dark green leafy vegetables contain pigments that appear to be able to help combat eyesight problems. Fruits and vegetables are clearly an important part of a good diet. You can choose from a variety of fruits and vegetables.

Two questionnaire items assessed the promotion orientation induction (r = .77, p ≤ .01): (1) “Most of my thoughts were focused on how I can improve my eyesight,” and (2) “I thought mostly of how I can get better vision.” The following two questionnaire items assessed the prevention orientation induction: (r = .97, p ≤ .01): (1) “Most of my thoughts were focused on how I can prevent eyesight problems,” and (2) “I thought mostly of how I can guard against vision problems.” Response options ranged from 1 (strongly disagree) to 7 (strongly agree).

Message frame

There are four possible permutations of gain- and loss-framed messages (Dillard & Marshall, 2003; Levin et al., 1998; O’Keefe & Jensen, 2006; Rothman & Salovey, 1997). Gain-framed messages can emphasize obtaining a desirable/positive outcome or avoiding an undesirable outcome. Likewise, loss-framed messages can emphasize obtaining an undesirable outcome or avoiding a desirable/positive outcome. To balance the messages, the message frames for both the promotion and prevention orientations were manipulated by depicting the goal as obtaining a desirable/positive outcome (i.e., gain frame) or avoiding a desirable/positive outcome (i.e., loss-framed). Eye vision was chosen as the health outcome because, among the health issues (e.g., cancer, heart disease) associated with vegetable and fruit consumption, eyesight issue was deemed to be most relevant across age groups. For participants who read the promotion orientation message, the gain frame message was, “Get clear vision. Eat more fruits and vegetables.” The loss-framed message was, “Don’t miss out on getting clear vision. Eat more fruits and vegetables.” For participants who read the prevention orientation message, the gain-framed message was, “Prevent vision problems. Eat more fruits and vegetables.” The loss-framed message was, “Don’t miss out on preventing vision problems. Eat more fruits and vegetables.” Two questionnaire items assessed the message frame induction (r = .60, p ≤ .01): (1) “Was the information in the message you just read positive or negative?,” and (2) “Was the information in the message you just read favorable or unfavorable?” Response options ranged from 1 (negative/unfavorable) to 7 (positive/favorable).

Control Variables

Prior consumption of fruits and vegetables

To control for potential confounding variables, the daily consumption (by serving size) of fruits and vegetables was assessed with six questions from the Centers for Disease Control and Prevention–sponsored BRFSS survey (CDC, 2010), which is commonly used to ascertain fruit and vegetable intake (M = 3.75, SD = 2.42).

Selection of incongruent information

To address the hypotheses related to the effect of exposure to information selected that is incongruent with the assigned motivational orientation condition, the number of incongruent information selected was calculated from the log files (M = 1.07, SD = 0.64, range: 0-2). Selection of incongruent message occurs when participants select messages (in the preparation stage of information seeking) that do not match their motivational orientation. Of the six potential messages participants could select, three were worded with concerns associated with promotion orientation and three were worded with concerns linked to prevention orientation. Participants selected only three of the six information messages.

Results

Manipulation Check

All data for manipulation checks were analyzed as a 2 (promotion or prevention orientation) × 2 (gain framed or loss framed) between-subject factorial, with prior consumption of fruit and vegetables as a covariate. The motivational orientation induction through message priming occurred as anticipated. Participants with the promotion orientation induction (M = 5.32, SD = 1.15) reported that their thoughts were focused more on ways to improve their eyesight than did participants in the prevention orientation induction (M = 4.28, SD = 1.52), F(1, 101) = 16.02, p < .01, $η_{p}^{2}$ = .14. Participants in the prevention orientation induction (M = 4.69, SD = 1.62) reported that their thoughts were more focused on ways to prevent eyesight problems than did participants in the promotion orientation (M = 2.41, SD = 1.49), F(1, 101) = 55.69, p ≤ .01, $η_{p}^{2}$ = .36. Induction of message frames occurred as anticipated. Finally, participants who read the gain-framed message (M = 6.24, SD = 0.85) reported more favorable thoughts than did participants who read the loss-framed message (M = 6.04, SD = 0.77), F(1, 101) = 18.79, p ≤ .05, $η_{p}^{2}$ = .16.

Hypotheses

For H1a, 2 (motivational orientation) × 2 (message frame) ANCOVA (analysis of covariance), with prior consumption of fruit and vegetables as a covariate, indicated a main effect for motivational orientation for the proportion of selected information on promotional concerns, F(1, 101) = 48.60, p < .01, $η_{p}^{2}$ = .33. Confirming H1a, participants in the promotion orientation condition (M = 0.64, SD = 0.22) selected proportionately more information on promotional concerns than did participants in the prevention orientation condition (M = 0.35, SD = 0.22). Likewise, for H1b, participants in the prevention orientation condition (M = 0.65, SD = 0.22) selected more information on preventative concerns than did participants in the promotion orientation condition (M = 0.37, SD = 0.21), F(1, 101) = 48.35, p < .01, $η_{p}^{2}$ = .32. Means for the outcome measures by treatment group are displayed in Table 1.

Table 1.

Means for Outcome Measures by Treatment Group.

	Prevention Orientation		Promotion Orientation
	Loss Frame	Gain Frame	Loss Frame	Gain Frame
Selected information	0.29 (0.21)	0.40 (0.21)	0.62 (0.23)	0.64 (0.20)
Attitudes	3.79 (0.48)	3.14 (0.54)	3.15 (0.38)	3.91 (0.55)
Behavioral intention	3.37 (1.39)	2.76 (1.66)	2.89 (1.55)	3.81 (1.75)
Message recall	0.88 (0.24)	0.67 (0.10)	0.60 (0.16)	0.72 (0.14)

Note: Standard deviations are reported in parentheses. Selected information reflects the proportion of selected informational pieces that highlighted promotional concerns.

For H2a, H2b, and H3, all data were analyzed with a 2 (motivational orientation) × 2 (message frame) ANCOVA, with prior consumption of fruit and vegetables and number of incongruent information selected as covariates. For H2a, analysis indicated that the covariate of prior consumption of fruits and vegetables significantly predicted attitudes toward fruits and vegetables, F(1, 100) = 9.54, p < .01, $η_{p}^{2}$ = .09. The number of incongruent information selected did not significantly influence attitudes. Taking into account the covariates, there was a significant interaction between motivational orientation and message frame, F(1, 100) = 57.58, p < .01, $η_{p}^{2}$ = .37. The Sidak post hoc test (p < .05) revealed the hypothesized direction of effect whereby the gain-framed message (M = 3.91, SD = 0.55) elicited more positive attitudes than the loss-framed message (M = 3.15, SD = 0.38) in the promotion orientation condition, and loss-framed message (M = 3.79, SD = 0.48) resulted in more positive attitudes than the gain-framed messages (M = 3.14, SD = 0.54) in the prevention orientation condition. The Sidak post hoc test is a pairwise comparison test with strict control for family-wise error rate (i.e., probability of making a Type 1 error).¹ The significant interaction is depicted in Figure 1.

Figure 1.

Two-way interaction of regulatory orientation and message goal frame on attitudes toward eating fruits and vegetables.

For H2b, analysis indicated that neither of the covariates predicted intention to eat more fruits and vegetables. There was, however, a significant interaction between motivational orientation and message frame for intention to eat more fruits and vegetables, F(1, 100) = 5.16, p < .05, $η_{p}^{2}$ = .05. However, the Sidak post hoc test (p < .05) revealed significant differences between gain- and loss-framed messages for only the promotion orientation condition. Specifically, the gain-framed message (M = 3.81, SD = 1.76) elicited greater behavior intentions than loss-framed message (M = 2.89, SD = 1.55) for only the promotion orientation condition. In the prevention orientation condition, although the loss-framed message (M = 3.37, SD = 1.39) was higher in behavioral intention than the gain-framed message (M = 2.76, SD = 1.66), the difference was not statistically significant.

For H3, analysis indicated that neither of the covariates predicted message recall. Analysis, however, revealed a significant interaction between motivational orientation and message frame, F(1, 100) = 25.38, p < .05, η_p2 = .20. The Sidak post hoc test (p < .05) revealed significant differences between gain- and loss-framed messages for both motivational orientation conditions. In the promotion orientation condition, higher message recall was reported with the gain-framed message (M = 0.73, SD = 0.14) than with the loss-framed message (M = 0.60, SD = 0.16). Likewise, in the prevention orientation condition, higher message recall was reported with the loss-framed message (M = 0.88, SD = 0.24) than with the gain-framed message (M = 0.67, SD = 0.10). The significant interaction is depicted in Figure 2.

Figure 2.

Two-way interaction of motivational orientation and message goal frame on message recall.

Discussion

While the promotion orientation/gain frame and prevention orientation/loss frame effects have been previously documented (Cesario et al., 2004; Lee & Aaker, 2004; Spiegel et al., 2004), the results of the current study indicate that the effects of message frame and motivational orientation occur irrespective of exposure to information that was incongruent with the activated motivational orientation. The findings are in line with earlier research that suggests individuals tend to select and recall information that is consonant with their preexisting attitudes and beliefs (Oliver, 2002). For H1a and H1b, that motivational orientation was the only significant predictor in the analysis suggests that, per selective exposure, participants were selecting information that was congruent with their motivational orientation. While previous studies have demonstrated that the momentary activation of regulatory orientation is possible, the current study indicates that such momentary activation remains even after exposure to information that is incongruent with motivational orientation. This is demonstrated by the findings related to H2a. Specifically, the promotion orientation/gain frame and prevention orientation/loss frame elicited more favorable attitudes. When the message frame and the motivational orientation were unaligned, attitudes were significantly less favorable.

It is an intriguing finding that compatibility effects were seen for behavioral intentions for the promotion orientation conditions but not for the prevention orientation condition. Although the loss-framed message elicited greater behavioral intentions than the gain-framed message in the prevention orientation condition, the mean difference was not statistically significant. There are a number of potential explanations. First, meta-analytic reviews found a gain-framed advantage for promotion behaviors (O’Keefe & Jensen, 2006), but loss frames were not more persuasive for detection behaviors (O’Keefe & Jensen, 2006) or were only slightly more persuasive (O’Keefe & Jensen, 2009). Thus, it appears that framing effects are more pronounced for promotion behavior. Relatedly, the emphasis of action or compliance in a message tends to prompt behavioral adoption (O’Keefe & Jensen, 2006). Gain frames focus on compliance whereas loss frames highlight noncompliance. This can explain why compatibility effects were seen for attitudes (H2a) but not for behavioral intentions (H2b). Another potential explanation is that framing effects occur with attitudes but not with behaviors because attitudinal changes tend to precede behavior (Fishbein & Ajzen, 1975). This would explain why framing effects occurred for attitudes but not for behavior. However, it should be noted that, while a previous meta-analytic review (O’Keefe & Jensen, 2007), which parsed out different types of promotion behavior, found that framing effects did not occur with nutrition behavior, the current study suggests that some framing effects are seen in the context of the nutrition.

While previous research found self-reported processing fluency for the promotion orientation/gain frame and prevention orientation/loss frame participants (Lee & Aaker, 2004), such research was limited to self-reported ease of processing rather than actual message recall. Processing fluency is a form of enhanced engagement that results from compatibility effects. While the current study did not directly test what explains this mechanism, the enhanced engagement that occurs during compatibility of motivational orientation and message frame resulted in more accurate recall of messages, which is similar to the processing fluency finding of Lee and Aaker. Moreover, such enhanced recall of message content occurred irrespective of exposure to messages that were incongruent with the activated motivational orientation. Of interest is that under compatibility effects, prevention-orientated individuals recalled more message attributes than did promotion-oriented individuals (see Figure 2). One possible explanation is that individuals with a promotion orientation are generally not concerned with making mistakes whereas the individuals with a prevention orientation are highly sensitive to making mistakes. Said in another way, promotion-oriented individuals utilize advancement strategies and are more concerned about errors of omission (i.e., forgetting something) than about errors of commission (i.e., making mistakes) (Higgins, 1997). Thus, the message-recall results align with the general characteristics of these two separate motivational orientations.

In terms of practical implications for health communication, this study suggests that, as campaigns and communication-based interventions become more prevalent on the Internet, a motivational orientation can be temporarily elicited through health messages. That regulatory orientation can be momentarily activated allows for more nuanced design of health communication messages. In particular, upfront knowledge of individual’s motivational orientation can inform the types of health messages (i.e., message goal frames) that are most likely to be persuasive. Promotional and prevention orientations should be matched with gain- or loss-framed messages, respectively, to maximize the persuasive effects of health messages. Thus, to some extent, concerns about how individuals construe health behaviors (i.e., prevention vs. detection behavior) can be controlled by the communicator through careful design of health messages. This study also suggests that, if higher message recall is a desired outcome, Internet health information consumers should be given health messages in the prevention orientation with loss-framed messages.

Three limitations of this study should be noted. First, behavioral intention was a single-item measure. A multiitem measure should be used by future research. Second, only two discrete stages of information seeking were assessed in the current study. Whereas information seeking on the Internet is cyclical with reiteration of discrete information seeking steps (David et al., 2007; Rice, McCreadie, & Chang, 2002), it remains unclear whether the momentary activation of motivational orientation can be sustained for longer periods of time and for exposure to a greater amount of information that is incongruent with the activated motivational orientation. The current study only assessed the first two stages of information seeking (i.e., preparation and exploration stages). The last stage (i.e., consolidation) entails evaluating the results based on the goals set in the preparation stage and is used to determine the goals for the next cycle of information seeking. Thus, it is unclear whether the momentary activation of motivational orientation can be sustained for subsequent cycles of information seeking. Third, this study’s findings are limited to the context of fruit and vegetable consumption. Relatedly, however, significant effects have been found for fruit juice (Lee & Aaker, 2004) and consumption of fruits and vegetables (Cesario et al., 2004; Spiegel et al., 2004). Future research should conduct tests in other topic areas to ascertain whether the framing-effects pattern is limited to the context of fruit and vegetable consumption.

Footnotes

Author’s Note

The author is grateful for the constructive comments of the two anonymous reviewers.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Traci Hong (PhD, Annenberg School for Communication, University of Southern California) is an Assistant Professor at the Department of Communication, Texas A&M University. Her research interests include Internet effects on health behavior.

References

Afifi

W. A.

Weiner

J. L.

(2004). Toward a theory of motivated information management. Communication Theory, 14(2), 167-190. doi:10.1111/j.1468-2885.2004.tb00310.x

Apanovitch

A. M.

McCarthy

Salovey

(2003). Using message framing to motivate HIV testing among low-income, ethnic minority women. Health Psychology, 22, 60-67. doi:10.1037/0278-6133.22.1.60

Brannon

L. A.

Tagler

M. J.

Eagly

A. H.

(2007). The moderating role of attitude strength in selective exposure to information. Journal of Experimental Social Psychology, 43, 611-617.

Brasher

D. E.

Goldsmith

D. J.

Hsieh

(2002). Information seeking and avoiding in health contexts. Human Communication Research, 28(2), 258-271. doi:10.1111/j.1468-2958.2002.tb00807.x

Brinol

Petty

R. E.

(2006). Fundamental processes leading to attitude change: Implications for cancer prevention communications. Journal of Communication, 56, S81-S104.

Centers for Disease Control and Prevention. (2010). State-specific trends in fruit and vegetable consumption among adults—United States, 2000-2009. Morbidity and Mortality Weekly Report, 59(35), 1125-1130.

Cesario

Grant

Higgins

E. T.

(2004). Regulatory fit and persuasion: Transfer from “feeling right.” Journal of Personality and Social Psychology, 86, 388-404. doi:10.1037/0022-3514.86.3.388

Chaffee

S. H.

Saphir

M. N.

Graf

Sandvig

Hanh

K. S.

(2001). Attention to counter-attitudinal messages in a state election campaign. Political Communication, 18, 247-272.

David

C. C.

(2009). Learning political information from the news: A closer look at the role of motivation. Journal of Communication, 59(2), 243-261. doi:10.1111/j.1460-2466.2009.01414.x

10.

David

Song

Hayes

Fredin

E. S.

(2007). A cyclic model of information seeking in hyperlinked environments: The role of goals, self-effcacy, and intrinsic motivation. International Journal of Human-Computer Studies, 65, 170-182. doi:10.1016/j.ijhcs.2006.09.004

11.

Dillard

J. P.

Marshall

L. J.

(2003). Persuasion as a social skill. In Greene

J. O.

Burleson

B. R.

(Eds.), Handbook of communication and social interaction skills (pp. 479-513). Mahwah, NJ: Lawrence Erlbaum.

12.

Dutta-Bergman

(2005). Developing a profile of consumer intention to seek out additional information beyond a doctor: The role of communicative and motivation variables. Health Communication, 17(1), 1-16.

13.

Elliot

A. J.

McGregor

H. A.

(2001). A 2X2 achievement goal framework. Journal of Personality and Social Psychology, 80(3), 501-519. doi:10.1037//0022-3514.80.3.501

14.

Festinger

(1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson.

15.

Fishbein

Ajzen

(1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.

16.

Fisher

Jonas

Frey

Schulz-Hardt

(2005). Selective exposure to information: The impact of information limits. European Journal of Social Psychology, 35, 469-492.

17.

Fox

(2006). Online health search 2006. Pew Internet & American Life Project. Retrieved from http://www.pewinternet.org/Reports/2006/Online-Health-Search-2006.aspx

18.

Fox

(2011). Health topics. Pew Internet & American Life Project. Retrieved from http://www.pewinternet.org/~/media//Files/Reports/2011/PIP_HealthTopics.pdf

19.

Fredin

E. S.

David

(1998). Browsing and the hypermiedia interaction cycle: A model of self-efficacy and goal dynamics. Journalism and Mass Communication Quarterly, 75, 35-54.

20.

Frey

(1986). Recent research on selective exposure to information. In Berkowitz

(Ed.), Advances in experimental social psychology (Vol. 19, pp. 41-80). San Diego, CA: Academic Press.

21.

Gamson

W. A.

(1992). Talking politics. New York, NY: Cambridge University Press.

22.

Han

J. Y.

(2011). Transaction logfile analysis in health communication research: Challenges and opportunities. Patient Education and Counseling, 82(3), 301-312. doi:10.1016/j.pec.2010.12.018

23.

Hecht

M. L.

Marsiglia

F. F.

Elek

Wagstaff

D. A.

Kulis

Dustman

Miller-Day

(2003). Culturally grounded substance use prevention: An evaluation of the keepin’ it R.E.A.A. curriculum. Prevention Science, 4(4), 233-248.

24.

Herek

Gillis

J. R.

Glunt

E. K.

Lewis

Welton

Capitanio

J. P.

(1998). Culturally sensitive AIDS educational videos for African American audiences: Effects of source, message, receiver, and context. American Journal of Community Psychology, 26, 705-743.

25.

Higgins

E. T.

(1997). Beyond pleasure and pain. American Psychologist, 52(12), 1280-1300.

26.

Higgins

E. T.

(1998). Promotion and prevention: Regulatory focus as a motivational principle. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 30, pp. 1-46). San Diego, CA: Academic Press.

27.

Higgins

E. T.

(2000). Making a good decision: Value from fit. American Psychologist, 55, 1217-1230.

28.

Higgins

E. T.

(2006). Value from hedonic experience and engagement. Psychological Review, 113(3), 439-460. doi:10.1037/0033-295X.113.3.439

29.

Iyengar

(1991). Is anyone responsible? How television frames political issues. Chicago, IL: University of Chicago Press.

30.

Johnson

J. D.

(1997). Cancer-related information seeking. Cresskill, NJ: Hampton Press.

31.

Jonas

Schulz-Hardt

Frey

Thelen

(2001). Confirmation bias in sequential information search after preliminary decisions: An expansion of dissonance theoretical research on selective exposure to information. Journal of Personality and Social Psychology, 80(4), 557-571. doi:10.103700022-3514.80.4.557

32.

Kahneman

Tversky

(1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263-291.

33.

Kim

Y. M.

(2007). How intrinsic and extrinsic motivations interact in selectivity: Investigating the moderating effects of situational information processing goals in issue publics’ web behavior. Communication Research, 34(2), 185-211.

34.

Klapper

J. T.

(1960). The effects of mass communication. New York, NY: Free Press.

35.

Knobloch-Westerwick

Meng

(2009). Looking the other way: Selective exposure to attitude-consistent and counterattitudinal political information. Communication Research, 36(3), 426-448.

36.

Kreps

G. L.

Neuhauser

(2010). New directions in eHealth communication: Opportunities and challenges. Patient Education and Counseling, 78(3), 329-336. doi:10.1016/j.pec.2010.01.013

37.

Lee

Aaker

J. L.

(2004). Bringing the frame into focus: The influence of regulatory fit on processing fluency and persuasion. Journal of Personality and Social Psychology, 86(2), 205-218. doi:10.1037/0022-3514.86.2.205

38.

Lerman

Ross

Boyce

Gorchov

P. M.

McLaughlin

Rimer

B. K.

Engstrom

(1992). The impact of mailing psychoeducational materials to women with abnormal mammograms. American Journal of Public Health, 82, 729-230.

39.

Levin

I. P.

Schneider

S. L.

Gaeth

G. J.

(1998). All frames are not created equal: A typology and critical anlaysis of framing effect. Organizational Behaviour and Human Decision Processes, 76(2), 149-188.

40.

Liu

Gwizdka

Liu

Belkin

N. J.

(2011). Predicting task difficulty for different task types. Proceedings of the American Society for Information Science and Technology, 41(1), 1-10. doi:10.1002/meet.14504701173

41.

Noar

S. M.

(2006). A 10-year retrospective of research in health mass media campaigns: Where do we go from here? Journal of Health Communication, 11, 21-42. doi:10.1080/10810730500461059

42.

O’Connor

A. M.

Pennie

R. A.

Dales

R. E.

(1996). Framing effects on expectations, decisions, and side effects experienced: The case of influenza immunization. Journal of Clinical Epidemiology, 49(11), 1271-1276.

43.

O’Keefe

D. J.

Jensen

J. D.

(2006). The advantage of compliance or the disadvantage of noncompliance? A meta-analytic review of the relative persuasive effectiveness of gain-framed and loss-framed messages. In Beck

C. S.

(Ed.), Communication yearbook (Vol. 30, pp. 1-43). Mahwah, NJ: Lawrence Erlbaum.

44.

O’Keefe

D. J.

Jensen

J. D.

(2007). The relative persuasiveness of gain-framed loss-framed messages for encouraging disease prevention behaviors: A meta-analytic review. Journal of Health Communication, 12, 623-644. doi:10.1080/10810730701615198

45.

O’Keefe

D. J.

Jensen

J. D.

(2009). The relative persuasiveness of gain-framed and loss-framed message for encouraging disease detection behaviors: A meta-analytic review. Journal of Communication, 59, 296-316. doi:10.1111/j.1460-2466.2009.01417.x

46.

Oliver

M. B.

(2002). Individual differences in media effects. In Bryant

Zillman

(Eds.), Media effects: Advances in theory and research (2nd ed., pp. 507-524). Mahway, NJ: Lawrence Erlbaum.

47.

Oliver

M. B.

Krakowiak

K. M.

(2009). Individual differences in media effects. In Bryant

Oliver

M. B.

(Eds.), Media effects: Advances in theory and research (3rd ed., pp. 517-531). New York, NY: Routledge.

48.

Petty

R. E.

Cacioppo

J. T.

(1986). Communication and persuasion: Central and peripheral routes to attitude change. New York, NY/Berlin, Germany: Springer-Verlag.

49.

Rice

R. E.

Katz

J. E.

(Eds.). (2001). The Internet and health communication: Expectations and experiences. Thousand Oaks, CA: SAGE.

50.

Rice

R. E.

McCreadie

Chang

S. L.

(2002). Accessing and browsing information and communication. Cambridge: MIT Press.

51.

Rothman

A. J.

Bartels

R. D.

Wlaschin

Salovey

(2006). The strategic use of gain- and loss-framed messages to promote healthy behavior: How theory can inform practice. Journal of Communication, 56, 202-220. doi:10.1111/j.1460-2466.2006.00290.x

52.

Rothman

A. J.

Salovey

(1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121(1), 3-19.

53.

Salovey

Schneider

T. R.

Apanovich

A. M.

(2002). Message framing in the prevention and early detection of illness. In Dillard

J. P.

Pfau

(Eds.), The persuasion handbook: Theory and practice (pp. 391-408). Thousand Oaks, CA: SAGE.

54.

Scheufele

D. A.

(1999). Framing as a theory of media effects. Journal of Communication, 49(1), 103-122.

55.

Shah

D. V.

Kwak

Schmierbach

Zubric

(2004). The interplay of news frames on cognitive complexity. Human Communication Research, 30(1), 102-120. doi:10.1111/j.1468-2958.2004.tb00726.x

56.

Shen

Dillard

J. P.

(2007). The influence of behavioral inhibition/approach systems and message framing on the processing of persuasive health messages. Communication Research, 34(4), 433-467. doi:10.1177/0093650207302787

57.

Smith

S. M.

Fabrigar

L. R.

Powell

D. M.

Estrada

M.-J.

(2007). The role of information-processing capacity and goals in attitude-congruent selective exposure effects. Personality and Social Psychology Bulletin, 33(7), 948-960. doi:10.1177/0146167207301012

58.

Spiegel

Grant-Pillow

Higgins

E. T.

(2004). How regulatory fit enhances motivational strength during goal pursuit. European Journal of Social Psychology, 34, 39-54. doi:10.1002/ejsp.180

59.

Tversky

Kahneman

(1981). The framing of decisions and the psychology of choice. Science, 211, 453-458.

60.

Williams-Piehota

Schneider

T. R.

Pizarro

Mowad

Salovey

(2003). Matching health messages to information-processing styles: Need for cognition and mammography utilization. Health Communication, 15(4), 375-392.

61.

Wilson

Wallston

King

(1990). Effect of contrast framing, motivation to quit, and self-efficacy on smoking reduction. Journal of Applied Social Psychology, 20, 531-547.

62.

Zillman

Bryant

(1985). Selective exposure phenomena. In Zillman

Bryant

(Eds.), Selective exposure to communication (pp. 1-10). Hillsdale, NJ: Lawrence Erlbaum.