Journal of the Academy of Marketing Science

, Volume 36, Issue 4, pp 522–537

The effect of incidental affect on preference for the status quo

Authors

  • HsiuJu Rebecca Yen
    • Department of Information ManagementNational Central University
    • Department of Business AdministrationNational Chung Cheng University
Original Empirical Research

DOI: 10.1007/s11747-008-0084-2

Cite this article as:
Yen, H.R. & Chuang, S. J. of the Acad. Mark. Sci. (2008) 36: 522. doi:10.1007/s11747-008-0084-2

Abstract

The authors report results from four experiments that describe the influence of incidental affect on preference for the status quo. The first and second experiments, each using different affect induction methods, demonstrated that positive affect increased and negative affect reduced the choice of the status quo alternative. The third experiment replicated the effect of incidental affect on a separate dimension of status quo bias, the preference of omission. The last experiment investigated whether preference for the status quo was affected by four specific emotions that differed along valence and certainty appraisal dimensions. Compared with affective states associated with uncertainty, affective states associated with certainty led to a greater magnitude of status quo bias. The certainty appraisals mediated the effect of specific emotions on preference for the status quo, regardless of the emotion valence. The findings extend our knowledge by showing that incidental affect can influence consumer choice through emotion-related appraisals.

Keywords

EmotionsStatus quo biasIncidental affectOmission biasAppraisal tendency model

A major issue in the decision-making literature has been the problem of how consumers make choices among a set of available alternatives. The theory of rational decision making assumes that consumers select the alternative that offers the highest utility. That is, only preference-relevant features of the alternatives influence an individual’s decision. However, in the real world, decision makers often stick with the choice they have made (status quo option) even when faced with new options. Such a preference is inconsistent with the rational choice model, which predicts that a previously selected option should have no effect on choice (Luce 1959; see also Chernev 2004). Research provides abundant evidence for such a bias (i.e., status quo bias) in both hypothetical and real-world decision tasks (Samuelson and Zeckhauser 1988).

In this article, we explore the influence of emotion and, in particular, incidental affect on the status quo. Incidental affect refers to affective states that typically arise from factors external to the judgment about the decision being made. Following the current literature (Bagozzi et al. 1999), we use the affect-related terms (namely, mood, affect, emotion) interchangeably while referring to incidental affect in the rest of this article. While most research examining mood effects on consumer behaviors (e.g., Kahn and Isen 1993; Lee and Sternthal 1999) deal with incidental affect, its role in status quo bias has been largely ignored (see Garg et al. 2005, for an exception). Although prior research has effectively examined how anticipated and anticipatory emotions influenced the selection of the status quo option (Inman and Zeelenberg 2002; Kahnenman and Tversky 1982; Luce et al. 1999; Riis and Schwarz 2000), these studies focused on emotions that are a subjective experience relevant to the choice task. Such emotional experiences are considered to be a form of task-related affect that are elicited by the task or process of making judgments and decisions (Cohen et al. 2007).

Given the susceptibility of consumption choice to contextual factors and the pervasiveness of incidental affect, our focus on incidental affect addresses an important gap in the status quo literature. The major goal is to advance our understanding regarding how valence-based and discrete affective states influence status quo bias. In the first step, we manipulated positive and negative affect using two different induction techniques and tested the effect of incidental affect on preference for the status quo option in two separate experiments. The results of both experiments support our predictions, showing that positive affect is associated with a greater preference for the status quo option than is negative affect. This extends our knowledge regarding the status quo bias by demonstrating that it is a function of incidental affect, which has been under-explored in the extant literature.

Second, we included preference for omission as another dependent variable in the third experiment, and the results showed that incidental affect influences preference for omission in the same way as it influences preference for status quo. Although Ritov and Baron (1992) argued that status quo bias is an extension of omission bias, Schweitzer (1994) produced evidence for independence between status quo and omission biases. In spite of this paradoxical result, our findings support the argument that status quo and omission biases are closely interrelated and share the same causes (Anderson 2003).

Third, we adopted the appraisal tendency approach (Lerner and Keltner 2000) to examine the effect of discrete affective states. By moving beyond valence and looking into the role of emotion-related appraisals, this research followed the emerging trend in examining specific emotions (DeSteno et al. 2000; Lerner and Keltner 2000). The fourth experiment reported in this article showed that certainty appraisals of the affective states influenced the magnitude of status quo bias and mediated the emotional impact on the preference for the status quo. With the combination of the valence-based and appraisal tendency approaches, this research is able to delineate the role of global affective states and certainty appraisal in explaining status quo bias.

The remaining content of this article is organized as follows. We begin by summarizing relevant research on status quo bias and incidental affect. Then, we derive predictions that address the effect of incidental affect on status quo bias from the valence-based and emotional appraisal perspectives. We report results from four controlled experiments that test these hypotheses and conclude with a discussion of the implications of our findings.

Status quo bias and incidental affect—the valence-based approach

Defining status quo bias

In choosing among alternatives, a decision maker has a tendency to prefer options that cause no change in the state of the world (the status quo; Samuelson and Zeckhauser 1988) and/or require no action on their part (omissions; Ritov and Baron 1990). Loss aversion (Kahneman and Tversky 1979; Thaler 1980) and norm theory (Kahneman 1992; Tversky and Kahneman 1991) explain the essential causes of the status quo bias. Loss aversion describes people’s tendency to weight potential losses greater than potential gains of the same amount. Thus, it seems likely that individuals have a strong tendency to remain at the status quo or to not take action because the perceived loss associated with leaving the current state is greater than the perceived advantages. Norm theory states that outcomes are perceived as worse when decision makers imagine that a better outcome could have occurred. When an adverse outcome is caused by inaction, it is more difficult to mentally simulate the outcome of action; the emotional response is thus not as strong. Due to this tendency to blame actions that produce adverse outcomes more than inactions that produce poor results, individuals are inclined to select status quo choices to avoid anticipated regret.

Let us illustrate the status quo bias with the following example. Consider a situation in which a consumer is choosing between two apartments (A and B) with uncertain outcomes, such that neither alternative clearly dominates the other. For example, A has the advantages of a color TV, cable, and wall-to-wall carpeting and B has the advantages of a dishwasher, refrigerator, and nice new furniture. If the consumer already owns A, then it becomes a reference point in the evaluation of B. In exchanging choice A for B, although the consumer will gain the advantages of B, the advantages of A that are given up in the exchange are experienced as losses to the consumer because he or she already has them. Because “losses loom larger than gains” (Kahneman and Tversky 1979, p. 279), the consumer’s evaluation of losses and gains associated with Option A (the status quo option) is likely to be asymmetric in nature. That is, the losses when giving up Option A are weighted more heavily than the corresponding advantages of obtaining Option B. As a result, the comparative attractiveness of Option A would rise and result in a lower probability of switching.

The effect of incidental affect on status quo bias

The relationship between affect and cognition has been an interesting research topic for several decades. The extant literature on affect and information processing has revealed asymmetric effects of positive and negative affect on judgment and choice. Research on incidental affect, in general, indicates that positive affect leads to more heuristic information processing, whereas negative affect results in more systematic processing (Bless et al. 1990; Fiedler 1990; Isen 1987; Mackie et al. 1992; Schwarz 1990). Existing explanations for the asymmetric effects of incidental affect primarily posit that positive moods lead to a reduction in either ability (e.g., Mackie and Worth 1989) or motivation (e.g., Bodenhausen et al. 1994a; Wegener et al. 1995).

The position that positive affect is associated with deficits in either ability or motivation, however, seems at odds with other existing theories and research findings. For example, the broaden-and-build theory (Fredrickson 1998, 2001) stipulates that positive emotions should broaden an individual’s thought–action repertoires and build their enduring personal resources. Consistent with this notion, research also demonstrated that people in a happy mood made faster and more efficient decisions (Isen and Means 1983) and were more creative in problem solving (Isen et al. 1987). Interestingly, the cognitive advantages demonstrated by happy individuals tended to appear only in processing of positive information (e.g., Isen et al. 1992; Isen and Simmonds 1978). Thus, Isen and Simmonds (1978) suggested that the difference may stem from motivation by people with positive affect to maintain their pleasant mood. In this study, we chose to focus on two theoretical models to understand the effects of incidental affect on status quo bias: the revised affect-as-information model and the mood-maintenance model.

The affect-as-information (AAI) model argues that emotions may serve as information to signal whether a situation is favorable or unfavorable with regard to one’s personal goals and current concerns (Schwarz and Clore 1983). Positive moods signal a benign environment in which the individual may not be motivated enough to engage in effortful cognitive processes, while negative moods signal the presence of a problem that motivates an individual into careful and systematic processing of information. Recent revisions of the AAI model (e.g., Bless et al. 1996b; Bless 2001; Clore et al. 2001) argued that affect does not necessarily influence motivation or ability to process information but, rather, leads to a different processing style and influences one’s confidence in relying on general knowledge structures (e.g., scripts, heuristics, prior judgments). That is, a positive affect signaling that the environment is benign may increase the individual’s confidence in relying on general knowledge structures, given that it has been effective in the past. In contrast, a negative affect signals a problematic environment, which is usually a deviation from normal, routine situations. As such, people will feel it necessary to attend to the specifics of the situation because relying on general knowledge structures is not appropriate. The differential reliance on general knowledge structures in positive and negative affective states has been found in various content domains, including the processing of persuasive messages (e.g., Schwarz et al. 1991), the reliance on scripts for behavior sequences (e.g., Bless et al. 1996a), and the use of global trait information in impression formation (e.g., Isbell 2004).

Evidence from past research suggests that the status quo option is a decision maker’s inflated preference for the current state of affair (Samuelson and Zeckhauser 1988). That is, the status quo option might be the most accessible knowledge or the default for the decision makers in most situations. Thus, happy individuals who feel confident in following a “business as usual” procedure should be more likely to select the status quo option. On the other hand, a signal of a problematic environment through a negative affect leads individuals to attend to the specifics of the present situation (i.e., carefully scrutinizing the different features of the choices). The lack of confidence to respond “as usual” (i.e., maintaining one’s previous choice) should lead to a lower magnitude of status quo bias among individuals feeling negative.

In addition, the choice of the status quo bias may be affected by the individual’s mood-maintenance motivation. The mood-maintenance explanation suggests that people in a positive mood are motivated to maintain their mood and may avoid cognitive activities that could interfere with their good mood (Isen 1987; Mackie and Worth 1989). Wegener and Petty’s (1994) hedonic contingency model resonates this view, based on the assumption that “behaviors that result in more positive or less negative feelings are rewarded, and behaviors that result in less positive or more negative feelings are punished”. To experience rewards, happy individuals must carefully choose activities that will prolong or enhance their positive affective states. Wegener and Petty (1994) found that people feeling positive are motivated to process uplifting messages and avoid depressing or negative information. However, studies also show that people with a positive mood have more thoughts about losses than controls when the level of risk is moderate to high (Isen and Geva 1987; Isen et al. 1988). That is, compared to people having a negative mood, people with a positive mood care more about whether information will spoil their mood. In order to prolong the good mood, people in a positive mood are more sensitive to the mood-changing consequences of their actions than people in a negative or neutral mood. Given that status quo options may be seen as less threatening (Riis and Schwarz 2000), people in a positive mood will show a greater preference for status quo options because the affective qualities of such options meet the need to maintain the good mood. Selecting status quo choices will also allow happy individuals to avoid the loss that is usually believed to derive from changing the current state. Conversely, people with a negative affective are less prone to maintain their current option because their need for mood-repairing may direct their attention toward the potential gains associated with the alternative.

Based on the conceptualization of the two mechanisms, we proposed the first hypothesis which was tested in Experiments 1 and 2.
  1. H1:

    The status quo bias is a function of affect valence, such that preference for the status quo option is stronger for people in a positive affective state than for those in a negative affective state.

     

Omission bias as a function of incidental affect

According to Ritov and Baron (1992), the current state of affairs might be maintained as a result of two different biases. The first is a bias toward maintaining the current state of affairs, and the second is reluctance to take action that would change this state. Thus, an alternative explanation of status-quo bias is that changing the status quo requires an act whereas preserving the status quo requires only an omission; a failure to act. When a decision leads to a negative outcome, relative to what might have been, people think the decision was worse if the outcome resulted from action as opposed to inaction (Baron and Ritov 1994; Kahneman and Tversky 1982; Spranca et al. 1991). The preference for harmful inaction over equally harmful action is captured by the concept of omission bias (Spranca et al. 1991).

Additionally, Ritov and Baron (1992) argued the two biases are a unitary phenomenon. In spite of research showing the two biases are independent of each other (Schweitzer 1994), status quo and omission biases still work in concert in the majority of real-world cases and may share underlying causes (Anderson 2003). This all suggests that omission bias is another dimension of status quo bias. On the basis of these commonalities between the two biases, we expect that preference for omission will vary as a function of emotion valence in a manner identical to that of status quo bias. Similar to the status quo option, omission is the preferred choice of the decision makers (Ritov and Baron 1990) and individuals persist in seeking default no-action options (Anderson 2003). Thus, it may serve as the general knowledge structure or “business as usual” procedure for the individual. According to the mechanism posited by the revised AAI model, people in a positive mood have a higher likelihood to commit the omission bias than people in a negative mood. The mood-maintenance model, based on different reasoning, also predicts the same pattern of results. That is, positive emotion should be associated with a greater emphasis on avoiding errors of commission (making a wrong decision) because doing so is in accordance with the motivation to maintain a good mood and avoid uncertainty risk. People in a negative feeling state have a need to repair their mood, thus, they should emphasize avoiding errors of omission (missing an opportunity for improvement). The preceding discussion leads to our second hypothesis, which was tested in Experiment 3:
  1. H2:

    The omission bias is a function of affect valence, such that preference for the omission option is stronger for people in a positive affective state than for those in a negative affective state.

     

Specific emotions and status quo bias—the role of certainty appraisals

According to the valence-based approach, it, by default, predicts specific emotions of the same valence will exert similar influences on judgment and choice. Although prior research has primarily examined positive and negative affects, researchers have also found that valence (i.e., positive vs. negative affective states) alone cannot explain the relationship between incidental affective states and judgment (e.g., DeSteno et al. 2000; Keltner et al. 1993; Tiedens and Linton 2001). Recent research showed that emotions of the same valence may differ in their physiology (Levenson 1992), facial expression patterns (Ekman 1993), and cognitive components (Smith and Ellsworth 1985). It appears that sadness encourages systematic processing but anger promotes heuristic processing (Bodenhausen 1993; Bodenhausen et al. 1994b; Lerner et al. 1998). Because this research examined the cognitive consequences of emotions, it should be promising to focus on the cognitive appraisal component to elucidate the effects of specific emotions.

Appraisal theorists propose that emotions can be meaningfully differentiated at a more gradated level than merely positive and negative (e.g., Lazarus 1991; Ortony et al. 1988; Smith and Ellsworth 1985; Weiner 1980). Studies of emotion-appraisal associations showed that the experience of specific emotions is consistently associated with specific sets of appraisals. For example, anger is associated with appraisals of: (a) other-responsibility for negative events, (b) individual control, and (c) a sense of certainty about what happened (Averill 1983; Smith and Ellsworth 1985; Weiner et al. 1982). The appraisals of the current states produce judgments and actions that are congruent with their constitute appraisals. For instance, fear as an emotion is associated with the appraisal that the current situation is risky, and increased perceptions of risk in subsequent circumstances (DeSteno et al. 2000; Lerner and Keltner 2000, 2001). The congruency between incidental affective states and subsequent judgment is called appraisal tendency (Lerner and Keltner 2000), and can exist along any appraisal dimension.

Due to the important role that certainty and uncertainty have played in the study of information processing (Kahneman et al. 1982; Weary and Jacobson 1997), we focus on the certainty–uncertainty appraisal dimension in this research. Prior studies have shown that the experiencing of some emotions is reliably associated with certainty–uncertainty appraisal (Ellsworth and Smith 1988; Smith and Ellsworth 1985). Emotions such as contentment, happiness, anger, and disgust are characterized by feelings of certainty; understanding what is happening in the current situation and feeling able to predict what will happen next. Other emotions like hope, surprise, worry, and fear are accompanied by feelings of uncertainty; not understanding what is happening and feeling unsure about what will happen next. Drawing on the appraisal tendencies, we posit that certainty appraisal of specific emotions is another important factor determining the role of specific emotions on the magnitude of the status quo bias.

Certainty appraisal and status quo bias

Early theories of social cognition considered the role of passing uncertainty as a motivator for engaging in effortful cognitive processes. Festinger (1954) asserted that individuals feeling uncertain engage in social comparison. Pelham and Wachsmuth (1995) found support for this argument by showing that people undertake more systematic processing and social comparison when they are uncertain about their self-views. People engage in more effortful processing when an actual level of certainty is below the desired level, because certainty-appraisal serves as an internal cue to signal whether further processing is necessary (Tiedens and Linton 2001). Indeed, Martin et al. (1993) argued that positive mood leads to heuristic processing rather than systematic processing because it provides a sense of subjective certainty. Tiedens and Linton (2001) extended that argument and demonstrated that certainty-related emotions resulted in more heuristic processing than do uncertainty-associated emotions, which promote systematic processing.

Building on Tiedens and Linton’s (2001) work and the concept of “affect as input” (Forgas 1995), which postulates that people use an affective state as a signal about the current situation or problem, we predict that certainty-appraisals promote status quo bias. Experiencing an uncertainty-related emotion evokes a perceived threat to the achievement of desired goals, thus the situation calls for systematic and attentive processing. It makes sense that willingness to engage in systematic and effortful processing will reduce the individual’s probability of committing to a status quo bias while neither alternative in the choice set clearly dominates the other. In contrast, experiencing an emotion of certainty signals the situation is safe, and people are unlikely to sense a need for changing the current state, thus they prefer to stay with the current choice.

Furthermore, to extrapolate the appraisal tendency perspective and the nature of the status quo bias, a logical hypothesis would suggest the influence of specific emotions on cognitive processing function through the mediation of certainty appraisals for subsequent situations. That is, if specific emotions result in differential certainty appraisals, and those appraisals also affect the preference for the status quo option, then certainty might be the mechanism to bridge the effects of specific emotions on the status quo bias. This mediation can be illustrated as follows: People experiencing disgust are expected to have a greater level of the status quo bias than those experiencing fear, because feelings of disgust result in a higher degree of certainty than being fearful.

On the basis of this discussion, we proposed the 3rd and 4th hypotheses that will be tested in Experiment 4.
  1. H3:

    Certainty-appraisals of emotions influence the level of status quo bias, such that preference for the status quo option is stronger for people feeling certain than for those feeling uncertain.

     
  2. H4:

    Certainty-appraisals mediate the influence of affect on preference for the status quo.

     

Experiment 1

The objective of this experiment is to test the first hypothesis that consumer preference for the status quo is a function of emotions, such that preference for the status quo is more pronounced for positive emotions than for negative emotions.

Design and procedure

One hundred and eighty-six undergraduates of a university in central Taiwan participated in this study as a requirement for a marketing management course. This study used a mixed experimental design, incorporating both between-subjects and within-subjects components. The participants were randomly assigned to one of the 3 (valence: positive, baseline, negative) × 2 (counterbalance: 1st vs. 2nd choice) conditions. The factor “counterbalance”, a design that can reduce the order effect, involved counterbalancing the perceived status quo option such that in one case the status quo option was the first choice and in the other case the status quo option was the second choice. One additional factor (product category: apartment vs. health club vs. vacation hotel) was a within-subjects factor.

When the subjects came to the laboratory, an experimenter told the participants that they would be engaged in a number of different tasks because of the brief nature of each task. The first experimenter introduced the first task and distributed the materials for affect induction. A second experimenter introduced the second task, which was the decision problems, and the participants completed the task at their own pace. Every subject made three choice-decisions, one for each product category. This design is supposed to dissociate the affect induction and the choice task.

Affect induction

Incidental affect was manipulated using a methodology adapted from the procedure used by Smith and Ellsworth (1985). During the affect induction procedure, the participants were initially asked to relate a personal incident that had evoked the target affect (happy or sad). In the positive affect (i.e., happy) condition, the participants were asked to remember, relive, and vividly recall an event that had made them feel happy. The instruction for negative affect (i.e., sad) induction was identical except that the participants had to remember a sad event or experience, while subjects in the baseline condition described in detail what their typical day is like and the activities they undertake on a routine day. After the participants were introduced to the foregoing scenario and had recalled the situation for a few minutes, they were asked a series of questions about the recalled experience. The subjects then wrote down their responses to the questions. The questions, first used by Smith and Ellsworth (1985) to encourage participants to recall their experiences in details, should maximize the effectiveness of affect induction. For example, the subjects were asked to describe in detail why the things that happened in the recalled situation made them feel happy (or sad).

Choice task

Once the emotion induction and the memory writing were complete, the participants moved to the second task in which they had to make a choice between two options, one of which was designated as their current (status quo) option, in the product categories of apartment, health club, and vacation hotel. We adopted the design of the choice task from the work of Nowlis et al. (2002). For instance, in the apartment category, two apartments with different advantages were described. The current (status quo) apartment had the advantages of being in a new apartment building and having a color TV, cable, and new wall-to-wall carpeting, whereas the new apartment (alternative) had a dishwasher, a refrigerator, nice new furniture, and the cost of heating was included in the rent. The subjects were told, “Imagine that you have been renting a one-bedroom apartment (Current Apartment). Your current lease is up and you have the chance to stay in your current apartment or move to a different apartment (New Apartment). What will you do?” and had to select one of the two choices (see Appendix). We designed the attributes in each choice set for the other two product categories based on information that was taken from a 2005 Consumer Reports (see the product categories and their attributes in the Appendix). The dependent variable was the product choice that a participant made (status quo versus alternative) across the three product categories.

Results and discussion

Manipulation check

Immediately after the emotion induction, the participants answered two questions taken from Smith and Ellsworth’s (1985) study. The two items assessed the pleasantness of the recalled experience with a question that read, “How happy was it to be in the situation you wrote about?” and another question asked the subjects to rate how enjoyable it was to be in the situation they wrote about. The two items, rated on a 7-point scale (1 = not at all, 7 = extremely), showed adequate internal consistency (α = .81) and were averaged to provide the measure of perceived pleasantness.

A 3 (valence) × 2 (counterbalance) analysis of variance (ANOVA) revealed only a main effect for affect manipulation on the perceived pleasantness of the event, F (2, 183) = 10.69, p < .0001). Participants who had undergone the happiness induction rated their feeling state as more pleasant than did those in the neutral condition (M = 5.1 vs. 4.87, F (1, 122) = 2.81, p = .09), whereas participants in the control group reported a higher degree of pleasantness than did those who had undergone the sad induction (M = 4.87 vs. 4.12, F (1, 122) = 9.06, p < .01). Additionally, the analysis showed no significant interaction effect, F (2, 180) = 0.5, p = .61. These results confirmed the effectiveness of inducing different affective states.

Effects of incidental affect on status quo bias

Table 1 presents the percentage of respondents who selected the current option across the three product categories in which the status quo was counterbalanced to be either the 1st or the 2nd option. The data showed that participants’ preference for the status quo varied as a function of emotion valence. An average of 74% of the participants in the positive affect condition stayed with their current option, and 47% of the participants in the negative affect condition displayed the same pattern. The current option was preferred by 59% of the respondents in the control group.
Table 1

Percentages choosing status quo option in Experiments 1 and 2

 

Positive emotion

Control condition

Negative emotion

Study 1

   

 First option as status quo

   

  Apartments

75a

58

43

  Health clubs

78

64

47

  Vacation

72

60

49

 Second option as status quo

   

  Apartments

69

55

44

  Health clubs

72

60

48

  Vacation

77

58

50

 Average

74

59

47

Study 2

   

 First option as status quo

   

  Apartments

79

60

50

  Health clubs

70

58

43

  Vacation

75

59

45

 Second option as status quo

   

  Apartments

72

61

40

  Health clubs

74

50

48

  Vacation

79

55

51

  Average

75

57

46

aAll values represent percentages

To examine the significance of the difference, we perform a 3 (valence) × 2 (counterbalance) × 3 (product categories) loglinear modeling using SPSS 13.0. Loglinear analysis is a statistical technique for analyzing data when both independent and dependent variables are categorical or nominal (Tansey et al. 1996). The participant’s choice was modeled as a 0–1 dummy variable, where 1 denoted the choice of the current (status quo) option. Both the emotion valence and the product category were treated as a 0–1–2 variable, where 0 denoted a positive affect or the apartment, 1 denoted control condition or health club, and 2 denoted negative affect or vocational hotel. We treat the order of current option as a 0–1 variable where 0 denoted the first choice as the current option and 1 denoted the second choice as the status quo choice.

The analysis revealed no significant interaction effect between incidental affect and the other two variables \(\left( {x_{{\text{affect}}\,{\text{x}}\,{\text{counterbalance}}\,{\text{x}}\,{\text{product}}}^2 \left( 4 \right) = 1.32,p = .857;x_{{\text{affect}}\,{\text{x}}\,{\text{counterbalance}}}^2 \left( 2 \right) = 0.70,p = .705;x_{{\text{affect}}\,{\text{x}}\,{\text{product}}}^2 \left( 2 \right) = 1.19,p = .871} \right)\). The influence of affect on preference for the current option (χ2 (2) = 18.55, p < .0001) was the only main effect found in the analysis. The proportion of participants choosing the status quo was significantly higher in the positive affect condition than in the negative affect condition (χ2(1) = 18.43, p < .0001). The analysis also showed significant differenced in preferences for the current option between the two induced affect states and the neutral group \(\left( {x_{{\text{positive}}\,{\text{vs}}{\text{.}}\,{\text{neutral}}}^2 \left( 1 \right) = 7.44,p < .01,x_{{\text{negative}}\,{\text{vs}}{\text{.}}\,{\text{neutral}}}^2 \left( 1 \right) = 3.95,p < .05} \right)\).

The results reported in this experiment support our first hypothesis that consumers’ preference for the status quo was a function of their incidental affect, such that the status quo was more preferred by happy consumers than by sad ones. This finding was replicated across the three different product categories. However, there is always the possibility that the preference of status quo is primarily caused by the involuntary reaction of participants to the stimuli rather than objective evaluation (Forgas and Ciarrochi 2001). For example, the recollection of positive or negative events used to induce incidental affect may also have produced self-enhancing motivations in the participants other than producing an affect change (Sedikides 1995). As such, secondary consequences may occur in all affect-induction procedures, thus it is important to use diverse methods of affect induction to establish that incidental affect is indeed the critical variable that influences preference for the status quo. To prevent this potential research limitation, we conducted Experiment 2 with the goal to replicate and extend the findings of Experiment 1 by employing a different, and not self-related, affect-induction procedure.

Experiment 2

In Experiment 2, we tested H1 employing a different affect-induction procedure. This approach has been applied in prior studies to “triangulate” mood effects (Forgas 1994, 1995) because most mood-induction methods showed a tendency to have unintended cognitive and motivational consequences. Thus, it is desirable to use different procedures in a series of related experiments to establish that the observed effects are due to affect and not other variables. Experiment 2 employed an audiovisual affect-induction procedure that was successfully used by Forgas and Ciarrochi (2001) and Adaval (2003) to elicit the required affective states.

Design and procedure

A total of 240 undergraduates (129 females, 111 males) in central Taiwan participated in return for course credit. Experiment 2 employed the same mixed-experimental design as Experiment 1, with two between-subjects and one within-subjects factors. Again, the subjects were randomly assigned to each cell of a 3 (valence: positive, neutral, negative) × 2 (counterbalance) design.

Procedure and materials

The procedure of this experiment was identical to Experiment 1 except for the affect-induction method. In the positive affect condition, a segment from the movie Pretty Woman was shown to induce a happy mood. In the negative affect condition, a clip from the feature file Ordinary People was shown to induce a sad mood. The clips lasted 20 min 22 s and 21 min and 45 s respectively, and prior research has shown that the film clips reliably induced the feeling of happiness and sadness (Adaval 2003). The participants were instructed to watch the film clips as if they were watching video at home, and to become fully involved in the film. No film clip was shown to the subjects in the control group, which served as the baseline for comparison. Film as an emotion manipulator helps exclude potential motivational and cognitive preoccupations in memory (the so-called autobiographical task).

Results and discussion

Manipulation checks

Cronbach alpha for the two items measuring participants’ perceived pleasantness provided adequate convergence (α = .90), indicating that it was appropriate to use the averaged response for the manipulation check. In a 3 × 2 ANOVA, we found a main effect of incidental affect (F (2, 237) = 9.62, p < .0001) while the interaction effect was not significant (F (2, 234) = 1.123, p = .327). Participants reported feeling more pleasant after they had watched the happy film clip than those in the control group (M = 4.82 vs. 4.45 for positive vs. control condition), F (1, 158) = 3.334, p = .069. Additionally, perceived pleasantness was higher in the control group (M = 4.45) than it was in the negative affect condition (M = 3.92), F (1, 158) = 6.789, p < .05. These results indicated that the films were successful in inducing the desired affective states.

Incidental affect and preference for the status quo

As illustrated in the lower half of Table 1, the averaged selection of the current option was 75% in the positive emotion condition versus only 46% in the negative emotion condition. We conducted a 3 × 2 × 3 loglinear analysis that is identical to that of Experiment 1. Analysis showed that the three-way interaction as well as the other interactions tested in the model were all non-significant (all χ2’s<1, all p’s > .65), indicating the effect of incidental affect on status quo bias was not influenced by the product category and/or the presentation order of the current option. As predicted, the impact of emotion valence on status quo bias was significant (χ2(2) = 8.69, p < .01), showing that subjects in the three affect conditions had significantly different tendencies in selecting the status quo option. Further analysis showed that the happy respondents’ preference for the current option was significantly stronger than those of sad respondents (χ2(1) = 23.56, p < .001). Similar results patterns also appeared for the “happiness vs. control” (75% vs. 57%, respectively, χ2 (1) = 8.86, p < .005) and the “control vs. sadness” (57% vs. 46%, respectively, χ2(1) = 3.99, p < .05) contrasts.

In this experiment, we induced happiness and sadness using movie clips rather than an event-recall task. Although the use of film clips limited the types of affective states we could manipulate, it afforded greater experimental control over the affect induction. Together, Experiments 1 and 2 provided clear and convergent support for our first hypothesis that affective state influences the magnitude of the status quo effect. The results consistently demonstrated that preference for the status quo was stronger in a positive affective state than in a negative affective state. This tendency to “stick with the status quo alternative,” such as purchasing the same brand and staying in the same job (Samuelson and Zeckhauser 1988), has been explicitly linked to inertia tendencies. People’s tendency towards inertia represents the very real behavioral tendency to resist change, to put off decisions, and to prefer inaction over action. This implies that people’s preference for not acting, as opposed to acting, is one of the explanations for the advantages that the status quo notoriously has over alternative states of affairs. Thus, we next examined whether preference for inaction over action, as another dimension of status quo bias, also varies as function of incidental affect.

Experiment 3

To extend the generalizability of our findings regarding emotional effect on status quo bias, this experiment replicated the first two experiments but focused on preference for omission as the dependent variable. To capture the omission bias, we added a no-choice option in which the respondents could avoid making a choice. To further demonstrate this effect, we removed the default option (the status quo), thus controlling for the potential status quo bias relating to a preference for a specific option in the choice set. This experiment tests H2 that predicts that preference for omission will vary as a function of incidental affect in a way similar to preference for the status quo. That is, people in a positive affective state are more prone to omission bias than are people who are in a negative affective state.

Design and procedure

One hundred and forty-seven students received extra credit to participate in the experiment. This experiment employed the same mixed experimental factors (i.e., valence × product category) except the counterbalance design. The subjects were randomly assigned to one of three between-subject experimental conditions (valence), whereas the experimental procedure and the affect manipulation were identical those employed in Experiment 1. The choice task given to subjects asked them to select from a set of three choices, which included a no-choice option. The other two options were the same as Option A and B used in Experiment 1, but neither of them was designated as the status quo. For example, in the apartment category, participants were told “Imagine that you have been renting a one-bedroom apartment. Your current lease is up and you have the chance to choose one of the two apartments or to choose neither. What will you do?”

Results and discussion

Manipulation check

A one-way ANOVA showed that the affect induction significantly influenced respondents’ ratings of pleasantness based on the average of the two manipulation-check items (α = .83), F (2, 144) = 11.25, p < .0001. Participants exposed to positive affect induction rated their mood as being significantly more pleasant than did those in the control group (M = 4.83 vs. 4.30, F (1, 96) = 3.98, p < .05). Participants’ moods were also significantly more pleasant in the control group than in the negative affect condition (M = 4.30 vs. 3.67, F (1, 96) = 5.81, p < .05). The results confirmed the effectiveness of the affect manipulation in this experiment.

Effects of affect valence on omission bias

The dependent variable was the participants’ preference for omission, operationalized as the selection of the no-choice option. We treated the dependent variable as a 0–1 dummy variable; whereas 1 denoted the selection of the no-choice option and 0 denoted the selection of Option A or B in the choice set. A 3 × 3 loglinear analysis examined the effects of incidental affect and product category on the selection of the no-choice option. Results showed neither main effect for product category (χ2 (2) = 0.24, p = 0.89) nor the interaction between affect and product category (χ2 (4) = 0.19, p = 0.97) were significant, suggesting that product type does not moderate the effect of incidental affect on omission.

Consist with the prediction of H2, there was a significant main effect for incidental affect (χ2 (2) = 7.6, p < 0.05), indicating that selection of the no-choice option varied as a function of individual affective states. While the average selection of the no-choice option was 34%, only 18% of the participants in the negative emotion condition selected the no-choice option. The percentage choosing the no-choice option was 51% and 33% for the positive affective condition and the control condition, respectively (see Table 2). Follow-up contrasts performed on the selection of no-choice option were all significant \(\left( {x_{{\text{positive}}\,{\text{vs}}{\text{.}}\,{\text{control}}}^2 \left( 1 \right) = 9.48,p < .05;x_{{\text{positive}}\,{\text{vs}}{\text{.}}\,{\text{negative}}}^2 \left( 1 \right) = 37.38,p < .001;x_{{\text{negative}}\,{\text{vs}}{\text{.}}\,{\text{control}}}^2 \left( 1 \right) = 9.59, < p < .05} \right)\), providing further support for the observed effect of affective state on omission.
Table 2

Percentages choosing status quo option in Experiment 3

 

Positive emotion

Control group

Negative emotion

Product

Option A

Option B

No choice

Option A

Option B

No choice

Option A

Option B

No choice

Apartments

20a

27

53

33

32

35

40

42

18

Health clubs

23

22

55

40

24

36

44

36

20

Vacation

26

28

46

33

39

28

41

43

16

Average

23

26

51

35

32

33

42

40

18

aAll values represent percentages

Results from this experiment confirmed our prediction regarding the effect of incidental affect on propensity for omissions. Although the link between incidental affect and the tendency to avoid the errors of omission (missing an opportunity for improvement) and commission (making a wrong decision) has been a part of conceptualizing the effect of incidental affect on status quo, this experiment has shown that such an effect can occur independently of the status quo bias. The findings complement the first two experiments in supporting our claim regarding the relationship between incidental affect and status quo bias, because incidental affect influences proneness toward the current state and toward preference for not undertaking a decision.

The first three experiments consistently demonstrated that positive affect led to greater status quo bias than did the negative affect. The design of these experiments, however, focused on emotion valence without looking into the impact of specific affective states on status quo bias. Experiment 4 extended the results from the first three studies by examining four specific emotions and their differential influence on status quo bias.

Experiment 4

In Experiment 4, we extended our investigation of the affective influence on status quo in two important ways. First, we examined four affective states that differ along both valence and certainty dimensions (Smith and Ellsworth 1985): disgust (negative valence, certainty), fear (negative valence, uncertainty), happiness (positive valence, certainty), and hopefulness (positive valence, uncertainty). This allows us to simultaneously investigate how emotion valence and certainty-appraisal of specific emotions can influence consumer choice of the status quo. The findings will test our third hypothesis which predicted that higher certainty would result in greater preference for the status quo. Second, we investigated whether certainty-appraisal mediated the effects of incidental affect on status quo bias, which was addressed by our fourth hypothesis. This extension enables us to test the role of certainty-appraisal as another mechanism that explains the observed effects of affect on status quo bias.

Design and procedure

Three hundred and eight subjects, recruited from the undergraduate population of a university in central Taiwan, participated in this study in exchange for extra credit. We used a mixed experimental design, with emotion valence (positive vs. negative), certainty (high vs. low), and counterbalance (1st vs. 2nd choice) as the between-subject factors. The product type, again, was treated as a within-subject variable. The cover story, affect manipulation, counterbalance, choice task context, and experimental procedure were similar to those followed in Experiment 1. After writing about the event, the participants responded to three items, taken from Smith and Ellsworth’s (1985) appraisal questionnaire, which assessed the certainty of the emotion on a seven-point scale ranging from 1 (not at all) to 7 (extremely). The three items asked the participants to rate the degree to which they understood what was happening around them, how well they could predict what would happen next, and how certain they were about what was happening when they were feeling the target emotion. The participants also answered the two-item scale that measured the pleasantness of the recalled event. Both measures showed adequate internal consistency (certainty: α = 0.83; pleasantness: α = 0.81).

Results and discussion

Manipulation check

For certainty appraisals, a 2 (valence) × 2 (certainty) × 2 (counterbalance) ANOVA indicated that the different certainty-appraisals were significantly different between certainty and uncertainty emotions, F (1, 300) = 9.06, p < .01. Participants who underwent happiness and disgust rated the event as more certain (M = 4.54) than did those who underwent hopefulness and fear (M = 4.21). The analysis revealed no significant interaction effect (all F’s < 1.57, all ps > .21).

As for pleasantness, a 2 × 2 × 2 ANOVA revealed only a main effect for valence of affect on perceived pleasantness, F (1, 300) = 43.86, p < .0001. Participants who were exposed to positive mood induction rated their feeling state as more pleasant (M = 5.08) than did those who were exposed to negative mood induction (M = 4.03). No other main effect or interaction effect was significant (all Fs < 2.5, all ps > 0.12).

Effects of emotions and certainty appraisal on status quo bias

The data given in Table 3 show that the percentage of respondents choosing the current option, averaged across different orders of the status quo option (i.e., 1st vs. 2nd option) and product categories, is a function of affect valence and certainty. Similar to the findings of Experiments 1 and 2, the current option was selected by 62.5% of the respondents in the positive affect conditions and by only 39.5% in the negative affect conditions. Approximately 59.5% of the respondents chose the status quo alternative in certainty conditions, and 42.5% displayed preference for the status quo in uncertainty conditions.
Table 3

Percentage choosing status quo option in Experiment 4

 

Negative emotion

Positive emotion

 

Certainty (disgust) n = 90

Uncertainty (fear) n = 90

Certainty (happy) n = 64

Uncertainty (hopeful) n = 64

First option as status quo

    

 Apartments

42a

35

69

56

 Health clubs

50

30

63

60

 Vacation

46

31

75

53

Second option as status quo

    

 Apartments

44

38

72

50

 Health clubs

54

25

75

50

 Vacation

54

27

69

53

Average

48

31

71

54

aAll values represent percentages

To test the significance of the data pattern, we conducted a 2 (valence) × 2 (certainty) × 2 (counterbalance) × 3 (product categories) loglinear analysis. As we predicted, the differential preference for the status quo alternative observed in different affect conditions (χ2 (1) = 12.77, p < 0.001) and certainty–uncertainty conditions (χ2 (1) = 8.08, p < 0.001) were both significant. On the other hand, none of the other main effects or interactions were significant (all χ2s < 2.2, all ps > .16). Further planned paired contrasts on the preference for the status quo choice revealed a significant difference between certainty and uncertainty conditions for participants who were induced to feel positive emotions (χ2 (1) = 12.19, p < 0.001). The analysis also indicated a similar significance pattern for those who were induced to feel negative affect (χ2 (1) = 16.45, p < 0.001). These results provide support for H3.

H4 posits that certainty-appraisals mediate the influence of the affect on preference for the status quo choice. We followed the procedure and conditions suggested by Baron and Kenny (1986) to test H4. If certainty-appraisals mediate the relationship between affective states and status quo bias then, when both affective states and certainty appraisals are simultaneously used to predict the preference for the status quo choice, the effect of affective states should become non- or less significant (Baron and Kenny 1986). A series of logistic regression analyses were performed for affective states of the same valence. In the positive affect conditions, the probability of selecting the current state as the dependent variable was regressed on two positive affective states (0 = hopefulness, 1 = happiness) and certainty-appraisals in Models 1 and 2, respectively. The analysis showed that happy respondents had a significantly higher rating of certainty (M = 4.80, SD = 0.93) than did the hopeful respondents (M = 4.51, SD = 0.62) t(178) = 2.02, p < .05), thus supporting the first condition. We also found that affective states of positive valence and certainty-appraisal both significantly predicted the selection of the status quo option (see Models 1 and 2 in Table 4), thereby confirming the second and third conditions. A final analysis showed that the coefficient of affective states dropped from 0.52 (p = 0.0005) to 0.41 (p = 0.098) but certainty appraisals remained significant when both positive affect and certainty-appraisal were simultaneously entered as predictors (see Table 4). The results demonstrated that certainty mediated the impact of positive-valenced discrete emotions on status quo bias. Similar results appeared for disgust and fear (see Table 4) and confirmed the mediation of certainty appraisals for the impact of the negative affective states on status quo bias.
Table 4

Mediating effect of certainty in positive (A) and negative (B) affective states

 

Variables

Model 1

Model 2

Model 3

A

    

Positive affective states

Happiness–hopefulness

0.5211 (0.0005)2

 

0.411 (0.098)

Certainty-appraisal

 

0.333 (0.005)

0.333 (0.0056)

R square

0.021

0.49

0.51

B

    

Negative affective states

Disgust–fear

0.566 (0.0001)

 

0.465 (0.061)

Certainty-appraisal

 

0.309 (0.005)

0.308 (0.0056)

R square

0.019

0.43

0.461

aβ

bp value

The results from Experiment 4 lead us to conclude that choice of the status quo is not only a function of affect valence, but is also affected by the certainty appraisal associated with specific affective states. The mediation of certainty appraisals appeared across the two valence conditions, establishing the robustness of this mediation mechanism of certainty dimension. Previous studies that have examined discrete emotional states focus mostly on those of the negative affect (e.g., Lerner and Keltner 2000; Raghunathan and Pham 1999; Rucker and Petty 2004; Tiedens and Linton 2001), thus revealing no significant effect for valence. The inclusion of specific emotions that differ along valence and certainty appraisals allows us to disentangle the individual effect of emotion valence and emotion-associated appraisals on status quo bias.

Although no interaction effect appeared, we compared the percentages of respondents choosing the status quo in the four emotional conditions to further explore the data. As Table 5 shows, people feeling discrete emotions exhibited significantly different preferences for the status quo, except for the contrast between disgust and hopefulness (p = 0.214). This suggests that emotions differing along valence could result in similar magnitudes of status quo bias, depending on the compound effect brought in by the certainty appraisals of the emotions. We consider this finding to be further evidence to support the notion that the effects of specific emotions on decision making are beyond the prediction of affect valence.
Table 5

Percentage choosing the status quo in the four emotional states

Comparison

df

χ2

p value

Con. 1 vs. 2

1

16.447

0.001

Con. 1 vs. 3

1

5.809

0.016

Con. 1 vs. 4

1

1.541

0.214

Con. 2 vs. 3

1

36.639

0.001

Con. 2 vs. 4

1

6.125

0.013

Con. 3 vs. 4

1

12.186

0.001

Con Condition, Con. 1 disgust (negative/certain), Con. 2 fear (negative/uncertain), Con. 3 happiness (positive/certain), Con. 4 hopefulness (positive/uncertain)

General discussion and conclusions

This article examines the effect of incidental affect on preference for the status quo. Although prior research has investigated incidental affect and status quo effect individually, our research illustrates the need to examine these two factors simultaneously. The data from the four experiments produced cumulative evidence for the role of incidental affect on status quo bias. Consistent with our predictions, positive affect increases and negative affect reduces the magnitude of the status quo bias. The use of a completely different affect-induction procedure (Experiment 2) and the inclusion of a neutral affective state (Experiments 1 and 3) establish the robustness of this observed effect of incidental affect. This article also demonstrates this effect in two choice contexts: preference for the choice perceived to be the status quo (Experiments 1, 2, and 4) and preference for inaction over action (Experiment 3), which is rare in the literature with few exceptions (Chernev 2004; Ritov and Baron 1992). This research contributes to the literature by offering empirical evidence that incidental affect is another common cause shared by the two decision-avoidance biases (Anderson 2003).

In this research, we suggest that the asymmetric effects of positive and negative affect on status quo bias are prompted by the informational and motivational properties of mood. With regard to the informational property of mood, we refer to the revised AAI model and posit that the heuristic processing (i.e., status quo and omission biases) exhibited by happy individuals does not necessarily reflect their cognitive constraint, but a processing strategy appropriate in given situations. That is, selecting the options that cause no change in the current state is presumably the most effective and functional cognitive style when affect signals a benign environment. The heuristic processing that allocates less attention to specifics of the given situation will allow the individual to direct the remaining resources toward other tasks when necessary. More importantly, such cognitive processing, although biased, may ultimately make individuals in a good mood more adaptive to the environment than those in a bad mood. This is supported by research finding that happy individuals, while working simultaneously on a second task, performed better at that task than sad participants (Bless et al. 1996a). Additional evidence showed that individuals feeling positive are particularly sensitive to information inconsistent with general knowledge structures than those feeling negative (Bless et al. 1996b), indicating that happy individuals are willing and able to allocate sufficient resources to process detailed information.

The higher magnitude of status quo bias exhibited by happy individuals, if conceived as the outcome of reduced motivation and/or efforts, may seem to contradict the position of other existing theories, such as the broaden-and-build theory (Fredrickson 1998, 2001). To resolve this potential incongruity, we focus on the revised AAI model which has attempted to uncover the potential advantages of positive emotion without disregarding the widely found association between positive mood and heuristic processing. As noted earlier, the argument of the revised AAI model has received support in studies showing that individuals with a positive mood, although activating heuristic processing, perform better at a simultaneously ongoing task than sad individuals (Bless et al. 1996a, b). It is plausible to consider the better performance of happy individuals on additional tasks as an illustration of the resource-building function of positive emotions. Based on that reasoning, we surmise that the effect of incident affect on status quo bias predicted by the revised AAI model should reconcile with that speculated by the broaden-and-build theory.

We proposed that incident affect also influences one’s choice of the status quo option via the mechanism of mood-maintenance motivation. Although the hedonic contingency model postulates that people in a positive mood are more motivated to process uplifting messages and engage in more systematic processing of information (Wegener and Petty 1994), it also holds that people in a positive mood are more sensitive to mood-changing outcomes of their actions than people in a negative mood. Because the decisions we studied in this research involve tradeoffs between valued attributes, and people in general believe that the status quo leads to less regret (Inman and Zeelenberg 2002), the greater preference for the status quo option found in happy respondents may merely indicate the avoidance of mood-deflating activities. If so, this motivational mechanism of incidental affect is essentially coherent with the underlying assumption of hedonic contingency on the motivation to achieve and maintain a pleasant mood (Wegener and Petty 1994).

The findings reported in this article also indicate that the proposed causes of the status quo effect (Ritov and Baron 1992; Samulson and Zeckhauser 1988) may not be able to adequately explain the bias if the effect of incidental affect is not taken into consideration. In particular, the status quo bias has largely been attributed to an individual’s tendency to weigh losses heavier than gains and the inclination to associate loss with action in making decisions (Kahneman and Tversky 1979, 1984). If the current results follow from the explanation of loss aversion, all participants should have exhibited a general preference for the status quo. In this research, we suggest that the process underlying status quo bias is rather complex and the effect of incident affect on status quo bias may involve multiple mechanisms that also moderate the role of loss aversion in such a process. Taking the motivational mechanism as an example, positive emotion makes loss aversion more apparent to individuals because happy individuals have high sensitivity to loss (Isen and Geva 1987; Aspinwall and Brunhart 1996; Reed and Aspinwall 1998). Accordingly, happy individuals will avoid any risky actions because their primary concern is to minimize possible losses. In contrast, individuals in a negative affective state may strive to repair their negative mood, thus their interest will be maximizing their gains; weakening the influence of loss aversion. The differential focus on loss and gains thus results in varied preference for the status quo. In conclusion, the inclusion of incidental affect with loss aversion has complemented the existing explanations of the status quo bias.

Another important contribution of the findings from Experiment 4 concerns the direct impact and mediation effect of certainty appraisals on status quo bias. This research extends prior research regarding the cognitive consequence of certainty-associated emotions (Tiedens and Linton 2001) to illustrate the impact of certainty appraisals on status quo bias. While previous studies examining effects of discrete emotions on cognitive processing have primarily focused on negative affect (Bodenhausen et al. 1994b; Keltner et al. 1993; Lerner and Keltner 2000; Tiedens and Linton 2001), this research also extends to examine specific emotions of both valences. The exploratory finding that a positive/uncertain and a negative/certain emotion exerts a similar influence on preference for the status quo suggests that the interplay of emotion valence and emotion appraisals merits further investigations.

Taken together, our results offer insight into the mechanism of status quo bias and identify incidental affect as well as certainty as important determinants of status quo bias. In many cases, individuals’ preference for the status quo is susceptible to the influence of incidental affect because they are unaware of the impact deriving from their affective states. Finally, status quo bias may arise in similar magnitudes when people are experiencing emotions of different valence, and this could be attributed to how people appraise the certainty of their affective states.

Implications for marketing practice

The findings of this study have some implications for marketing practice. The status quo effect explains why customers are reluctant to give up their existing service choice even when the alternatives incur cheap costs. For firms that attempt to switch their customers to new services, a common inducement is the trial purchase without obligation (Thaler 1980). During the trial period, consumers abandon the search for better alternatives, while increasing their psychological investment in the trial services. Marketing managers may consider inducing mild positive and certainty-associated affective states during the period near the end of the trial subscription, such as offering small gifts or a service guarantee, to enhance the status quo effect because positive affect and certainty appraisal promotes the status quo bias. Conversely, marketing managers who intend to convince customers of other firms to switch will need to induce a mildly negative feeling state or uncertainty-associated feeling state, such as revealing the disadvantages of using their existing supplier or brand, to reduce the consumers’ proneness to the status quo.

Limitations and future research

In the four experiments, we focused on consumers’ preference for the status quo, conceptualized as the tendency to keep the current state of affairs and reluctance to take action that will change this state. We generalize the findings across different product categories and across different presentation orders of the status quo choice using different manipulations. However, it would be useful to validate this phenomenon in a field setting to enhance the external validity of our findings. For example, future research could investigate status quo bias by evaluating the effect of naturally occurring (not induced) incidental affect in a real-life setting. Also, we relied on the findings of previous research to infer that positive mood facilitates one’s adaptation to the environment by freeing up resources for handling the primary task. This assumption should be further verified in future studies.

Although we showed the influence related to the valence and certainty appraisal of incidental affect, a future study should identify other factors that interact with incidental affect. For example, previous research has found that need for cognition may influence the degree of risk taking (Kuvaas and Kaufmann 2004; Lin et al. 2006), and it would thus be worth investigating whether need for cognition will moderate the relationship between incidental affect and preference for status quo. Other moderators might also include differences in affect intensity (Schimmack and Diener 1997) that vary from situations to situations. Finally, other appraisal dimensions, such as risk, control, effort, and agency, might also influence individuals’ choice (Smith and Ellsworth 1985). The other appraisal of the emotional experience can inform decision makers of relevant aspects of the situation, and influence their motivational force toward more desirable states. Thus, an extension of our work would be to understand the role of other appraisal dimensions on choice.

Copyright information

© Academy of Marketing Science 2008