Try Something New Together: Joint Consumption Fosters Choice of Unfamiliar Products

RAN Yaxuan¹, ZHANG Puyue¹, CHEN Siyun², XIANG Diandian¹

(1 School of Business Administration, Zhongnan University of Economics and Law, Wuhan 430073, China)
(2 School of Management, Jinan University, Guangzhou 510632, China)

Abstract

Joint consumption—where multiple consumers participate in and share the costs of a consumption activity—is a common phenomenon. Drawing on risk-shift theory, this research investigates the mechanisms and boundary conditions through which consumption context (individual vs. joint consumption) influences preferences for unfamiliar products. Across five experiments, we find that consumers in joint consumption contexts are more willing to try unfamiliar products compared to those consuming alone. This occurs because joint consumption transfers consumption risk, thereby reducing perceived risk associated with unfamiliar products. However, this effect only emerges for low-risk products and disappears or even reverses when consumers share the experience with socially distant companions.

Keywords: joint consumption, unfamiliar product, perceived risk, collective decision-making, risk-shift theory
Classification Code: B849: F713.55

1. Introduction

Imagine you and a friend decide to dine out together. After searching and filtering options on your phone, you're left with two choices: a restaurant you've visited before with decent food, or a newly opened establishment with comparable ratings. Which would you choose? This scenario illustrates a typical joint consumption situation that people encounter daily, where consumers experience the consumption process together and share expenses.

Research indicates that over half of consumers participate in joint consumption activities at least three times per month (Wu et al., 2019). In practice, an increasing number of companies are designing marketing strategies around joint consumption—for example, McDonald's "second drink half-price" promotion, Happy Valley's discounted couple tickets for couples or best friends, and Coca-Cola's "share a Coke" bottles that require two people to open together. Given the prevalence and popularity of joint consumption, scholars have begun examining the core mechanisms of this consumption model and its behavioral effects.

Existing research on joint consumption primarily addresses three topics: (1) drivers of participation in joint consumption (Ran et al., 2018); (2) decision-making processes and influencing factors in joint consumption (Garcia-Rada et al., 2019); and (3) consequences of joint consumption (Wu et al., 2021). Among these, studies on decision-making processes have received particular attention. However, most such research uses joint consumption as a backdrop to examine how group characteristics influence decision outcomes (Etkin, 2016; Lowe et al., 2019; Wu et al., 2019), without directly answering the fundamental question of what key differences exist between joint and individual consumption. Only Nikolova et al. (2018) have examined this distinction in moral decision-making, finding that individuals in joint consumption contexts are more likely to engage in unethical behavior to build intimate relationships. Whether joint consumption (vs. individual consumption) influences consumer product preferences and decisions remains a black box. Therefore, this research focuses on comparing joint and individual consumption to answer the opening question—how does joint consumption (vs. individual consumption) affect consumer preferences and choices regarding unfamiliar products?

1.1 Joint Consumption

Joint consumption refers to situations where multiple consumers participate together in a consumption process (Liu & Min, 2020; Yang et al., 2015), emphasizing two characteristics: "co-purchasing" and "co-using." Specifically, each participant pays a certain cost for the purchase (i.e., "co-purchasing") and subsequently shares ownership of the product (i.e., "co-using"), such as splitting the bill when dining with friends. Joint consumption is fundamentally distinct from other forms of multi-person consumption such as group buying, collaborative consumption, and shared consumption (see Table 1). First, group buying emphasizes aggregating multiple familiar or unfamiliar consumers to increase bargaining power and obtain discounts (Jing & Xie, 2011), such as group purchases on Pinduoduo. Second, collaborative consumption refers to situations where consumers obtain usage rights from third parties without ownership, with consumption involving one or multiple people (Benoit et al., 2017), such as bike-sharing services. Third, shared consumption emphasizes sharing behavior, where consumers share their own goods or services with others without transferring ownership (Belk, 2010), such as sharing snacks with others.

Table 1 Distinctions Between Joint Consumption and Related Concepts
Source: Compiled by the authors

Joint and individual consumption differ significantly in decision-making processes. In individual consumption, consumers primarily base decisions on their own personality, attitudes, and preferences (Simpson et al., 2012). In joint consumption, however, consumers make decisions not only for themselves but also for others. Research shows that joint consumption reflects collective preferences and generates systematic biases (Gorlin & Dhar, 2012; Yang et al., 2015; Garcia-Rada et al., 2019). For example, people in group decisions tend to avoid their own preferences and choose compromise options (Simonson, 1989). Accordingly, we hypothesize that preferences for unfamiliar products in joint consumption will differ from those in individual consumption.

1.2 Joint Consumption and Unfamiliar Product Preference

Choosing between familiar and unfamiliar products represents one of the most common forms of exploratory behavior, requiring people to trade off between certain benefits and uncertain costs. Following existing literature (Walter et al., 2020), we define unfamiliar products as those consumers have never purchased or used, including but not limited to newly launched brands and new products from established brands. Most existing research on promoting unfamiliar product choice has examined individual factors such as sensory experience (Lashkova et al., 2019), curiosity (Golman et al., 2019), and consistency-maximizing psychology (Riefer et al., 2017). A few studies have explored how contextual factors (e.g., product display) influence unfamiliar product preferences (Walter et al., 2020). This research follows this line of inquiry to analyze how consumption mode—joint vs. individual consumption—affects unfamiliar product preferences.

Research on exploratory behavior shows that perceived risk from unfamiliar products is a critical factor influencing consumer decisions (Ariffin et al., 2018). Bettman (1973) proposed that perceived consumption risk depends on five factors: (1) lack of product information and experience; (2) product newness; (3) low brand confidence; (4) high price; and (5) importance of the purchase decision. Consequently, consumer decisions about unfamiliar products are generally considered high-risk because consumers cannot retrieve reference information from memory to predict usage consequences. In summary, the key to choosing unfamiliar products lies in risk assessment, with lower perceived risk increasing the likelihood of selecting unfamiliar products.

Numerous studies have analyzed antecedents of perceived risk, including demographic characteristics (e.g., gender, age, income; Spence et al., 1970), product category (Chaudhuri, 1998), corporate and endorser credibility (Soesilo et al., 2018), environmental cues (Celso Augusto de & Anderson, 2018), and consumption context (e.g., online vs. offline; Han & Kim, 2017). Although no research has directly examined how group size affects perceived risk, risk-shift theory can help answer this question. Risk-shift theory primarily explains the "group polarization" phenomenon in group decision-making, where group decisions become more risky and extreme than individual decisions (Hensley, 1977). For example, young people are more likely to engage in criminal activities when with friends (Gardner & Steinberg, 2005). The key mechanism is risk transferability: for risky decisions, group decision-making that shares outcomes among members weakens individual risk perception, making them bolder and more aggressive (Levinger & Schneider, 1969). Since joint consumption shares this collective responsibility attribute with group decision-making, consumers should perceive lower risk in joint consumption contexts than in individual contexts, leading them to prefer unfamiliar products. In other words, perceived risk explains the mechanism through which joint consumption (vs. individual consumption) promotes unfamiliar product choice. Accordingly, we propose:

Hypothesis 1: Compared to individual consumption, joint consumption contexts increase consumer preference for unfamiliar products.

Hypothesis 2: Perceived risk mediates the effect of consumption context (joint vs. individual) on unfamiliar product choice. Specifically, joint consumption (vs. individual consumption) reduces perceived risk, thereby increasing preference for unfamiliar products.

1.3 Boundary Conditions

Based on our explanatory mechanism—perceived risk—we propose two boundary conditions. The first is product risk (or product category risk), which reflects the inherent risk of a product category (Dowling & Staelin, 1994). This risk is determined by product attributes; for example, experiential products carry higher risk than search products (Lowengart & Tractinsky, 2001). Our core logic suggests that joint consumption (vs. individual consumption) reduces perceived risk, thereby promoting unfamiliar product choice. However, since a group's ability to transfer risk is limited, when product risk is high (e.g., prescription drugs, electronic devices), the residual risk after joint consumption transfer remains high—potentially exceeding consumers' maximum risk threshold. Consequently, consumers will still avoid these products due to concerns about serious consequences for companions and themselves (e.g., health hazards, financial loss). When product risk is low (e.g., chewing gum, tissues), the absence of severe consequences makes people more willing to "take a small risk" with companions. Therefore, we propose:

Hypothesis 3: Product risk moderates the effect of joint consumption (vs. individual consumption) on unfamiliar product choice. Specifically, when product risk is high, the effect of joint consumption (vs. individual consumption) on unfamiliar product preference will be weakened or eliminated.

The second boundary condition is relationship closeness, defined as the degree to which individuals include others in their self-concept. Social relationships can be categorized as strong ties or weak ties based on closeness levels (VanLear et al., 2006). Weak ties involve limited interaction and low emotional intensity (e.g., colleagues meeting for the first time), while strong ties involve the opposite (e.g., parents and close friends). According to impression management theory, when with unfamiliar companions, people adjust their behavior through self-monitoring to maintain positive self-presentation (Snyder & Monson, 1975), such as sharing positive news (Chen, 2017), eating less food (Herman et al., 2003), and avoiding products with negative market evaluations (Philp & Nepomuceno, 2020). Conversely, people feel freer and more relaxed with strong-tie companions and are more willing to express their true preferences. Since choosing unfamiliar products involves risk, consumers with distant companions will avoid risk to maintain a trustworthy image, preferring familiar, conservative options. With close companions, however, people can make choices more freely and bravely propose trying unfamiliar products. Thus, we propose:

Hypothesis 4: Relationship closeness within the group moderates the effect of joint consumption (vs. individual consumption) on unfamiliar product choice. Specifically, (a) joint consumption (vs. individual consumption) promotes unfamiliar product choice only when consuming with close companions; (b) the effect is weakened or eliminated when consuming with distant companions.

The research framework is illustrated in Figure 1.

1.5 Overview of Current Research

Figure 1 Research Framework

This research designs five experiments to test our hypotheses sequentially. First, Study 1a (a lab experiment) and Study 1b validate the main effect of joint consumption (vs. individual consumption) on unfamiliar product preference across different decision modes. Existing literature categorizes joint consumption into two types based on decision-making authority: joint decision-making and unilateral decision-making (Gorlin & Dhar, 2012). The former involves all group members participating in the decision process to reach a consensus (e.g., a couple deciding on a travel destination). The latter emphasizes one member making decisions for the entire group while others simply enjoy the outcome (e.g., a man preparing a surprise dinner for himself and his girlfriend). Although research suggests decision authority differences affect consumer preferences (Gorlin & Dhar, 2012), risk-shift theory posits that perceived risk is primarily influenced by the decision context—whether one is in a group—rather than decision mode (individual vs. joint). To verify this, Studies 1a and 1b employ joint and unilateral decision-making contexts, respectively, to ensure robustness of the main effect. Additionally, Study 1b excludes the influence of relationship type on the effect. Study 2 changes the manipulation method to verify the mediating effect of perceived risk and attempts to rule out emotional arousal as an alternative explanation. Study 3 then examines the moderating effects of relationship closeness and product risk. Study 4 extends the effect's applicability to non-consumption decisions. To ensure...

Study 1a

Study 1a aimed to test the effect of joint consumption (vs. individual consumption) on unfamiliar product preference (Hypothesis 1). The experiment used a single-factor two-level design (consumption group size: 1 person vs. 2 people). Conducted as a lab experiment simulating real consumption scenarios, it allowed participants to engage more immersively. Materials included familiar original-flavor Skittles and newly launched flower-fruit-flavor Skittles from the well-known Mars brand. The experiment lasted two days with 138 university student participants.

2.2 Procedure

Participants were randomly assigned to two conditions. Those in the individual consumption condition completed all tasks alone, while those in the joint consumption condition were randomly paired with another participant. To prevent experimenter demand effects—where participants consciously behave in ways they believe align with the experimental purpose (Zizzo, 2010)—we first required participants to complete an unrelated writing task, with paired participants completing it together. Writing task quality also served as an attention check. Afterward, participants were told they could select a gift as a reward. We placed equal quantities of products and corresponding posters on a table (see Figure 2). The experimenter then introduced both products: original-flavor Skittles as a well-known classic, and flower-fruit-flavor Skittles as a newly launched variety. Each participant received one pack, with both members of joint consumption pairs required to choose the same flavor (taking two packs of one flavor). After each selection, the experimenter immediately replenished the products to maintain equal quantities. To control for display position effects, the experimenter swapped the left-right order of flavors every hour. Throughout the experiment, the experimenter silently recorded participants' choices and demographic information.

Figure 2 Experimental Materials for Study 1a

2.3 Results

All 138 participants (Mage = 20.9 years, SD = 1.75 years; 74.6% female) passed the attention check, with 64 assigned to individual consumption and 74 (37 pairs) to joint consumption. Using G*Power 3.1 for chi-square tests (Faul et al., 2009), with two groups, effect size f = 0.5, and α = 0.05, a sample of 101 yielded power > 0.99, exceeding the 0.80 threshold and confirming adequate statistical power.

Using product choice as the dependent variable, chi-square analysis revealed a marginally significant main effect of group size (χ²(1) = 2.94, p = 0.086, φ = 0.17). Specifically, participants in joint consumption (67.57%) were more willing to choose the unfamiliar new product than those in individual consumption (50%; see Figure 3), supporting Hypothesis 1. Including gender, participation time, consumption context, and product choice in a binary logistic regression model, the Crude OR (0.48) remained relatively unchanged compared to Adjusted ORs (0.36, 0.47, and 0.34), confirming that gender and participation time did not affect the main effect. Additionally, chi-square analysis of product choice and acquaintance status in the joint consumption group showed no effect of acquaintance on unfamiliar product preference (p = 1.000).

Figure 3 Effect of Consumption Context on Unfamiliar Product Preference (Study 1a)

2.4 Discussion

Study 1a preliminarily supported Hypothesis 1: consumers in joint consumption (vs. individual consumption) were more willing to choose unfamiliar products. However, limitations existed: (1) familiarity manipulation relied solely on product properties and experimenter descriptions without validation; (2) the experiment allowed paired participants to discuss and make decisions jointly, only testing the main effect under joint decision-making; and (3) the all-student sample prevented examination of how relationship types (e.g., family vs. friends) might affect the relationship between consumption context and unfamiliar product preference. Research suggests that pleasure from being with friends activates a promotion focus, while responsibility from being with family activates a prevention focus (Fei et al., 2019), potentially causing consumers to avoid risk and choose familiar products when with family. Study 1b addresses these limitations by changing the scenario and testing whether relationship type and decision authority moderate the effect.

Study 1b

Study 1b aimed to retest the main effect (Hypothesis 1) with altered scenarios and materials to ensure robustness, while also excluding potential influences of relationship type and decision authority. Key design changes included: (1) shifting to an online experiment with stricter variable control; (2) examining joint consumption with both friends and family; (3) changing to a unilateral decision-making context where participants decided alone for the entire group; and (4) using fictional brands to exclude interference from existing brand attitudes and consumption experience. Study 1b employed a single-factor three-level design (consumption context: individual vs. with friends vs. with family), with joint consumption scenarios involving three people. We recruited 404 participants through a marketing research platform.

3.2 Procedure

Participants were randomly assigned to three conditions and read the scenario description. First, they read a passage describing the consumption context and viewed silhouette images showing corresponding group sizes (see Figure 4). To verify attention and reinforce scenario understanding, we asked participants to identify their consumption companions and complete a fill-in-the-blank question inferring why they were shopping alone (or with friends/family). Next, participants imagined choosing lunch between two restaurants: newly opened "Naiqu Western Restaurant" and frequently visited "Laisideng Western Restaurant," which were similar in price and menu. Using fictional brands, we manipulated familiarity through text descriptions (Choi & Ahluwalia, 2013; Skard & Thorbjørnsen, 2014). For the familiar restaurant, we stated: "This is Laisideng Western Restaurant that you visit frequently; you're very familiar with most dishes and come here regularly." For the unfamiliar restaurant: "This is Naiqu Western Restaurant, which just opened this Wednesday. You notice its decor, prices, and menu are similar to Laisideng, but you've never heard of it and don't know how it tastes." Similar methods were used in subsequent experiments.

After reading the scenario, participants reported their decision ("Which restaurant would you ultimately choose for lunch?" 1 = Naiqu, 2 = Laisideng) and completed a simulated ordering task by selecting dishes from a menu. All participants received the same menu containing 15 dishes, 5 marked as "new arrivals" (see Figure 5), with identical pricing for similar dishes to control for price effects. To simulate realistic ordering, participants could freely select dish types and quantities based on group size, receiving a virtual bill after submission. We recorded these choices. To exclude personal experience effects, participants rated their liking (1 = "dislike very much," 7 = "like very much") and knowledge (1 = "know very little," 7 = "know very much") of Western food. To control for relationship closeness, we also measured perceived closeness with the two friends (or parents) (1 = "very distant," 7 = "very close"). Finally, participants reported demographics and guessed the experimental purpose.

The marketing research platform is a WeChat public account co-established by marketing professors from several Chinese universities. Operating for many years with a mature and large sample pool, it provides trustworthy data quality.

Figure 4 Context Manipulation Images for Study 1b (from top: individual consumption, with friends, with family)

Figure 5 Example of New Dish Materials in Study 1b

3.3 Results

No participants guessed the experimental purpose. After excluding 135 questionnaires that failed attention checks or had duplicate IP addresses, we obtained 263 valid responses (Mage = 25.5 years, SD = 4.63 years; 49.8% female; group sizes ranged 70-110). Using G*Power 3.1 (Faul et al., 2009) for chi-square tests with three groups, effect size f = 0.5, and α = 0.05, the power for n = 263 was 0.99, exceeding the 0.80 threshold.

Main Effect. Binary logistic regression of restaurant choice on consumption context showed a significant effect (B = 0.30, SE = 0.15, Wald = 3.80, p = 0.05, OR = 1.35). Chi-square tests revealed a significant main effect of consumption context that was not influenced by relationship type. Pairwise comparisons (see Figure 4) showed that participants in joint consumption with friends (60.2%; χ²(1) = 3.53, p = 0.060, φ = 0.15) and with family (60.7%; χ²(1) = 4.23, p = 0.040, φ = 0.15) preferred the unfamiliar restaurant more than those in individual consumption (45.2%). No significant difference existed between the two joint consumption conditions (p = 0.943), indicating consumers prefer unfamiliar options regardless of whether they're with friends or family.

Similarly, chi-square analysis of new dish selection showed comparable results (see Figure 6). Compared to individual consumption (72.6%), joint consumption with friends (95.2%; χ²(1) = 15.21, p < 0.001, φ = 0.31) and with family (87.9%; χ²(1) = 6.74, p = 0.009, φ = 0.19) were more willing to try new dishes. A marginally significant difference existed between the two joint conditions (χ²(1) = 3.08, p = 0.079, φ = 0.13), possibly because participants considered parents' generally lower acceptance of Western food, making them more cautious and increasing perceived product risk. These results support Hypothesis 1.

Control Variables. No significant differences emerged across conditions in Western food liking (F(2, 260) = 0.35, p = 0.703) or knowledge (F(2, 260) = 0.43, p = 0.648). Including these variables plus gender, age, and income as covariates, the main effect remained significant (F(2, 244) = 2.37, p = 0.096, ηp² = 0.02). Additionally, no difference in perceived relationship closeness existed between the two joint consumption conditions (F(1, 188) = 0.08, p = 0.774).

3.4 Discussion

Comparing Studies 1a and 1b reveals that joint consumption contexts increase preference for unfamiliar products under both unilateral and joint decision-making, effectively ruling out decision authority as a potential moderator. Study 1b also demonstrates that relationship type is not a key factor affecting the main effect, while addressing Study 1a's limitations in variable control and expanding the generalizability of our findings.

Study 2

Study 2 had three main objectives: (1) test the mediating role of perceived risk (Hypothesis 2); (2) rule out emotional arousal as an alternative explanation, as previous research shows that being with others affects mood and behavior (Choi et al., 2016); and (3) control for individual traits more strictly, excluding potential influences of personal curiosity and openness (Steenkamp & Baumgartner, 1992). Key modifications included: (1) using ice cream as experimental material, which post-tests showed had moderate inherent risk; (2) revising the text-based scenario manipulation to elicit more realistic responses; (3) changing the binary choice to purchase intention measurement; (4) improving perceived risk measurement by assessing both failure probability and consequence severity; and (5) adding measures for emotion, curiosity, and openness. Study 2 used a single-factor two-level design (consumption context: individual vs. joint)...

4.2 Procedure

Participants were randomly assigned to two conditions and read instructions. To increase immersion, we included a ranking task: joint consumption participants ranked same-sex friends by relationship closeness and wrote the initials of their third-ranked friend, while individual consumption participants wrote their own initials (Polman et al., 2018). Participants then imagined shopping with that friend (or alone) for an upcoming outdoor picnic and described similar past experiences to reinforce the consumption context. After an unrelated shopping task, participants encountered the following scenario (see Figure 7):

"After finishing shopping, you notice ice cream in a freezer and decide to buy some. You find that the familiar vanilla flavor you (vs. both of you) always eat is sold out, leaving only a cinnamon apple pie flavor you've (vs. both of you) never heard of before. Both flavors cost the same: ¥28/330g."

Participants reported purchase intention for the unfamiliar flavor (1 = "very unwilling," 7 = "very willing") and answered an unrelated attention check question ("What is the price of the shirt? Please select nine pounds ten pence"). We then measured perceived risk with two items ("I think the likelihood of being dissatisfied is high"; "I think the consequences of dissatisfaction would be severe"; 1 = "strongly disagree," 7 = "strongly agree"; r = 0.41, p < 0.01; Johnson & Andrews, 1971). Emotional arousal was measured with three positive emotions (happy, joyful, excited; α = 0.86), one neutral emotion (peaceful), and three negative emotions (bored, sad, disappointed; α = 0.80) (Choi et al., 2016). Participants also reported curiosity ("I find learning new things interesting"; 1 = "strongly disagree," 7 = "strongly agree"; α = 0.88; Litman & Spielberger, 2003) and openness ("I am willing to change my worldview"; 1 = "strongly disagree," 7 = "strongly agree"; α = 0.78; adapted from Rudd et al., 2018). An attention check during the scales asked participants to select a specific value ("How satisfied are you with your attention? Please select '5'"). Finally, participants reported familiarity with both ice cream flavors (1 = "very unfamiliar," 7 = "very familiar") and demographics, and guessed the experimental purpose.

Figure 7 Experimental Materials for Study 2

4.3 Results

Manipulation Check. No participants guessed the purpose. After excluding 22 questionnaires that failed attention checks or had duplicate IPs, 150 valid responses remained (Mage = 21.09 years, SD = 2.67 years; 77.3% female; group sizes 74-76). Using G*Power 3.1 (Faul et al., 2009) for ANOVA with two groups, effect size f = 0.5, and α = 0.05, power for n = 150 exceeded 0.86, surpassing the 0.80 threshold.

Paired-samples t-tests confirmed successful familiarity manipulation (Mvanilla = 5.79, SD = 1.33; Mcinnamon = 2.18, SD = 1.26; t(149) = 22.43, p < 0.001, d = 2.79).

Mediation Analysis. We tested Hypothesis 2 using a mediation model (Model 4, 5000 bootstraps; Preacher & Hayes, 2008) with consumption context (0 = individual, 1 = joint), perceived risk, and purchase intention. As Figure 8 shows, the indirect effect of perceived risk was significant (effect = 0.33, SE = 0.16, 95% CI: [0.0137, 0.6536]).

Figure 8 Bootstrapping Mediation Analysis
Note: "ns" indicates p > 0.05, * p < 0.05, *** p < 0.001

Alternative Explanations. After reverse-scoring negative emotions, we calculated average emotional scores across seven states. Marginal differences emerged (Mindividual = 5.10, SD = 0.80; Mjoint = 5.31, SD = 0.71; t(148) = 1.69, p = 0.093, d = 0.28), possibly because being with close friends elicited more positive emotions. However, when entered as a mediator in PROCESS, emotion could not replace perceived risk (indirect effect = 0.088, SE = 0.071, 95% CI: [-0.0178, 0.2555]). As a covariate, the main effect of consumption context remained (F(1, 147) = 3.24, p = 0.074, η²p = 0.02).

Control Variables. Including gender, age, income, curiosity, and openness as covariates did not change the main effect (F(1, 148) = 3.05, p = 0.083, η²p = 0.02).

4.4 Discussion

Study 2 demonstrated the risk-transfer mechanism: reduced perceived risk drives unfamiliar product choice in joint consumption contexts (Hypothesis 2). The main effect was stronger than in Studies 1a and 1b, suggesting our ranking and recall manipulations increased immersion. Additionally, mood did not explain the effect, and individual curiosity and openness did not affect robustness.

Study 3

Study 3 had four objectives: (1) replicate the risk-transfer mechanism (Hypothesis 2); (2) demonstrate that joint consumption increases unfamiliar option preference only for low-risk products (Hypothesis 3); (3) manipulate relationship closeness to show the main effect occurs only in high-closeness groups (Hypothesis 4); and (4) rule out diffusion of responsibility as an alternative mechanism. Design changes included: (1) using ranking tasks to manipulate friend relationship perception, and (2) expanding beyond food consumption. Study 3 used a single-factor three-level design (consumption context: individual vs. with a close friend vs. with a distant friend), recruiting 271 participants through the marketing research platform.

5.2 Procedure

Participants were randomly assigned to three conditions. To manipulate relationship closeness, joint consumption participants ranked acquaintances by closeness and wrote the name of their first- (or fiftieth-) ranked friend (Polman et al., 2018). Joint consumption participants imagined watching a movie with that friend, while individual consumption participants imagined watching alone. Participants made four sequential decisions: movie selection, cinema selection, popcorn flavor, and gift brand (see Figure 9). First, they chose between two equally-rated comedy movies differing in director/actor familiarity. Second, they selected between two similar cinemas, one frequently visited and one newly opened. Third, joint consumption participants reported their friend's ranking to verify attention and reinforce the scenario. All participants completed an attention check question (same as Study 2). They then encountered:

"You (vs. you and your friend) arrived at the cinema half an hour early. After getting tickets, you (vs. both of you) want popcorn. You find the familiar caramel flavor is sold out, leaving only an unfamiliar fruit-cream flavor you've (vs. both of you) never tried."

Participants rated purchase intention for the unfamiliar flavor (1 = "very unwilling," 7 = "very willing"). Finally, they could exchange their ticket stub for a gift, choosing between two fictional brands of portable hand sanitizer with identical prices and cleaning effects, one familiar and one unfamiliar. All decisions used text-based familiarity manipulations similar to Study 1b. After all decisions, participants reported perceived risk (1 = "almost no risk," 7 = "very risky"; Cheron & Ritchie, 1982), diffusion of responsibility (1 = "almost no responsibility," 7 = "very responsible"; Whyte, 1991), and relationship closeness (1 = "very distant," 7 = "very close"). Finally, they reported demographics and guessed the experimental purpose.

Figure 9 Summary of Experimental Materials for Study 3

5.3 Results

Manipulation Check. No participants guessed the purpose. After excluding 58 duplicate IP or failed attention check responses, 213 valid questionnaires remained (Mage = 23.65 years, SD = 4.47 years; 71.4% female; group sizes 68-74). Using G*Power 3.1 (Faul et al., 2009) for ANOVA with three groups, effect size f = 0.4, and α = 0.05, power for n = 213 exceeded 0.99. Independent samples t-tests confirmed successful closeness manipulation (Mclosefriend = 6.73, SD = 0.45; Mdistantfriend = 3.91, SD = 0.81; t(140) = 26.05, p < 0.001, d = 4.33).

Product Risk Moderation. Comparing individual consumption and close-friend joint consumption across four product decisions using one-way ANOVA revealed no significant differences in movie (t(143) = 0.08, p = 0.934), cinema (t(143) = 0.56, p = 0.580), or hand sanitizer decisions (t(143) = 0.74, p = 0.458), but a main effect emerged for popcorn flavor. Close-friend joint consumption participants (M = 5.36, SD = 1.20) showed higher purchase intention for the unfamiliar flavor than individual consumption participants (M = 4.93, SD = 1.42; t(143) = 2.00, p = 0.048, d = 0.33). Compared to popcorn, movie, cinema, and hand sanitizer decisions involve higher perceived product risk due to greater consequences regarding price, efficacy, and health threats (Johnson & Andrews, 1971), requiring careful consideration. Thus, Hypothesis 3 is supported: the main effect weakens or disappears as product risk increases.

Relationship Closeness Moderation. Pairwise comparisons of popcorn choices across three conditions (see Figure 10) showed close-friend joint consumption (M = 5.36, SD = 1.20) had higher purchase intention than both individual consumption (M = 4.93, SD = 1.42; t(143) = 1.86, p = 0.048, d = 0.33) and distant-friend joint consumption (M = 4.96, SD = 1.19; t(140) = 2.04, p = 0.044, d = 0.36). No difference emerged between individual and distant-friend joint consumption (t(137) = 0.01, p = 0.906). This indicates the main effect occurs only with close companions, supporting Hypothesis 4.

Figure 10 Effect of Consumption Context on Unfamiliar Popcorn Flavor Purchase Intention (Study 3)

Perceived Risk Mediation. Using individual consumption as the reference (0) and creating dummy variables for close-friend and distant-friend joint consumption, mediation analysis (Model 4, 5000 bootstraps; Preacher & Hayes, 2008) showed that after adding perceived risk, the direct effect of consumption context became non-significant (β = -0.02, SE = 0.11, p = 0.853), while the indirect effect remained significant (effect = 0.04, SE = 0.03, 90% CI: [0.0013, 0.0864]). This indicates perceived risk fully mediates the effect, supporting Hypothesis 2.

Alternative Explanations. No significant differences emerged in diffusion of responsibility across three groups (F(2, 210) = 0.61, p = 0.543). Entering this variable as a mediator showed no mediation effect (indirect effect = -0.01, SE = 0.01, 90% CI: [-0.0347, 0.0112]). As a covariate, the main effect remained (F(2, 210) = 4.63, p = 0.059, ηp² = 0.03). Thus, perceived diffusion of responsibility does not explain the effect.

5.4 Discussion

A post-hoc experiment validated our descriptions of inherent risk differences across the four product categories. We recruited 63 university students (Mage = 21.3 years, SD = 1.44 years; 76.2% female) to rate perceived risk for 21 products. Paired t-tests showed popcorn (M = 2.68, SD = 1.47) differed significantly from movies (M = 3.38, SD = 1.60; t(62) = 3.14, p = 0.003) and cinemas (M = 3.16, SD = 1.60; t(62) = 2.22, p = 0.030). Although popcorn and hand sanitizer did not differ significantly (M = 2.78, SD = 1.41; t(62) = 0.53, p = 0.600), in Study 3's context, the free hand sanitizer prize may have aroused efficacy doubts, increasing risk. Overall product risk distribution (Figure 11) shows electronics, personal care, and medical products carry higher risk, while food and daily necessities carry lower risk, validating our material selection.

Figure 11 Post-Hoc Experiment Results

Study 3 identified two important boundaries: the effect occurs only for low-risk products (Hypothesis 3) and only with close companions (Hypothesis 4). The risk-transfer mechanism was replicated, showing joint consumption reduces perceived risk, thereby increasing purchase intention. Diffusion of responsibility was ruled out as an alternative explanation.

Study 4

Study 4 examined whether the joint consumption effect extends beyond dining decisions to other contexts, even non-consumption situations. Based on prior literature and Study 3, the preference for unfamiliar options in joint consumption appears only for low-risk decisions. For most consumers, ordinary food and affordable restaurant choices involve low risk because both failure probability and consequence severity are minimal (Johnson & Andrews, 1971). Similarly, physical products like hair clips, tissues, and cotton swabs are low-risk. Moreover, many daily non-consumption decisions involve familiar vs. unfamiliar choices, such as selecting previously heard songs vs. newly released tracks, or familiar vs. newly launched games. While these choices may not significantly impact consumers' lives, they are crucial for new or transitioning businesses. Study 4 tested the promotion effect in other decision contexts using a single-factor two-level design (consumption context: individual vs. with two close friends), recruiting 212 participants.

6.2 Procedure

The procedure mirrored Study 3, with context manipulation similar to Study 1b. Participants were randomly assigned and read scenario instructions, then made four choices: playlist selection ("Your Favorites" vs. "Daily 30 Songs"), tissue selection (Vinda vs. Murou), soda selection (classic Pepsi vs. white peach oolong Pepsi), and chip selection (original vs. rose flavor; see Figure 12). Each decision paired a familiar brand/flavor with an unfamiliar one, with text indicating equal prices and volumes—the only difference being prior trial experience. Attention checks were inserted, identical to Study 2. After all decisions, participants reported product familiarity to validate our manipulation, then provided demographics and guessed the purpose.

Figure 12 Summary of Experimental Materials for Study 4

6.3 Results

Manipulation Check. No participants guessed the purpose. After excluding 64 duplicate IP or failed attention check responses, 148 valid questionnaires remained (Mage = 24.61 years, SD = 4.30 years; 66.9% female; group sizes 69-80). Using G*Power 3.1 (Faul et al., 2009) for ANOVA with two groups, effect size f = 0.5, and α = 0.05, power for n = 148 exceeded 0.99.

Paired t-tests confirmed successful familiarity manipulation for tissues (MVinda = 6.43, SD = 0.72; MMurou = 1.90, SD = 0.97; t(146) = 38.88, p < 0.001, d = 5.31), soda (Moriginal = 6.46, SD = 0.68; Mpeach = 2.27, SD = 0.98; t(146) = 38.57, p < 0.001, d = 4.96), and chips (Moriginal = 6.41, SD = 0.75; Mrose = 1.99, SD = 0.97; t(146) = 39.09, p < 0.001, d = 5.10).

Main Effect. Since the joint consumption effect primarily manifests in low-risk decisions, it may be less pronounced in purchase intention than in choice behavior. To amplify this, we coded each decision (familiar = 0, unfamiliar = 1) and summed them into a continuous variable ranging 0-4 representing the number of unfamiliar options selected. Independent samples t-tests showed that joint consumption participants (M = 1.66, SD = 1.16) selected more unfamiliar options than individual consumption participants (M = 1.17, SD = 1.08; t(146) = 2.61, p = 0.010, d = 0.44). Hypothesis 1 was supported in broader contexts.

6.4 Discussion

Study 4 extended the main effect's applicability, showing that consumers with intimate companions prefer unfamiliar options across everyday decisions and low-risk product choices, even in non-consumption situations.

7. General Discussion

This research demonstrates how consumption context (individual vs. joint consumption) influences unfamiliar product preferences, revealing underlying mechanisms and boundary conditions. Across five experiments, we validated risk-transfer theory's application in joint consumption contexts through varied materials, manipulations, scenarios, and visual designs. We found that both joint and unilateral decision-making in joint consumption promote unfamiliar product choice compared to individual consumption. The mechanism involves reduced perceived risk when consuming with others, increasing willingness to explore unknown options. This effect occurs only in high-closeness groups and for low-risk products. We also excluded alternative explanations including decision authority, relationship type, diffusion of responsibility, and emotional arousal. Overall, this research enriches theoretical literature and provides practical guidance for new product promotion.

7.1 Theoretical Contributions

First, this research expands consumer behavior findings in joint consumption contexts. Existing literature is fragmented, examining how various factors within joint consumption affect decision outcomes (Etkin, 2016; Liu & Min, 2020; Lowe et al., 2019; Yang et al., 2015; Wu et al., 2019, 2021; Boothby, 2014; Parker et al., 2019). Drawing on group size research (Ran et al., 2017, 2020), we directly compare individual and joint consumption—common yet fundamentally different contexts—to examine how joint consumption qualitatively changes decision outcomes rather than focusing on trade-off behaviors during decision processes (Garcia-Rada et al., 2019), thereby extending joint consumption literature.

Second, this research expands factors influencing exploratory behavior. Unfamiliar product choice is a typical exploratory behavior (Mehlhorn et al., 2015). Existing research shows environmental factors (e.g., product display, choice set size; Walter et al., 2020; Lejarraga et al., 2012), individual factors (e.g., variety-seeking; Kahn, 1995), and social factors (e.g., information sources; Winet et al., 2020) affect exploration. As a salient decision factor (Ran et al., 2021), our findings supplement social factors by showing how consumption context, distinguished primarily by group size, changes unfamiliar product preferences. Specifically, consumers with intimate companions are more willing to explore unfamiliar options.

Finally, this research applies risk-shift theory to consumer behavior, enriching its perspective. While widely studied in moral decision-making (Levinger & Schneider, 1969; Ferguson & Vidmar, 1971), few scholars have applied it to product decisions (Johnson & Andrews, 1971; Woodside, 1974), and those examined groups of strangers discussing new products—rare in daily life with limited practical implications. Our research connects risk-shift theory to common joint consumption forms, finding that groups of close companions exhibit risk shift even without discussion, perceiving less risk than individuals.

7.2 Practical Implications

Our findings provide practical guidance for promoting new brands or product lines. First, managers should identify product risk levels before designing context-specific promotions. For low-risk products, especially in restaurants and food industries, marketers can launch new product gift boxes or sets for multiple people, encouraging consumers to bring friends. Once products gain acceptance and word-of-mouth, consumers will repurchase regardless of companionship due to inertia (Dubé et al., 2010). During promotion, marketers should subtly integrate "joint" contexts, combining products with images of friend gatherings and family reunions to strengthen purchase intention. For high-risk products, avoid multi-person packages initially, as consumers will avoid risk across contexts, making joint promotion less effective and potentially wasteful.

Second, for products unsuitable for joint consumption (e.g., facial cleanser, razors), managers should recognize that risk perception is key. While context manipulation is impossible, providing diagnostic information like authentic reviews and professional endorsements can build trust (Xu et al., 2020), reducing risk and promoting purchase. Overall, companies should comprehensively analyze and position new products, understanding their risk levels in consumers' minds to target marketing activities accordingly.

7.3 Limitations and Future Directions

This research has limitations. First, due to pandemic restrictions, only Study 1a was conducted offline; others were online, limiting ecological validity. Although Study 1a simulated real consumption, the lab environment may have influenced choices. Second, while Studies 2 and 3 demonstrated perceived risk mediation, our measurement design prevented examining how different risk dimensions (financial, performance, social; DelVecchio & Smith, 2005) change in joint consumption. Conceptually, all three dimensions relate to joint consumption, but we cannot confirm their relative impacts.

Third, our conclusions apply only to risky decisions; in extremely low-risk decisions (e.g., free trials), perceived risk is no longer key, violating our premise. Future research could examine joint context effects in risk-free decisions through other mechanisms. Fourth, our joint consumption scenarios involved only 2-3 people; we did not examine group size effects. We limited size to 2-3 because maintaining intimacy with many people is difficult, but future research could explore behavior in larger groups (e.g., classes, clubs). Fifth, we used different decision modes across experiments—Study 1a used joint decision-making while others used unilateral decisions. Given that decision modes may affect joint consumption outcomes (Wu et al., 2019; Liu & Min, 2020), future research could examine their effects on other aspects of joint consumption. Sixth, our post-hoc risk ratings relied on self-reports, reflecting trends but lacking precision. Future research could use big data, text analysis, and machine learning for more accurate risk classification across diverse product categories.

Regarding research topics, existing joint consumption research focuses on hedonic consumption (Yang et al., 2015), shared dining (Clauzel et al., 2019), and moral decisions (Nikolova et al., 2018), while neglecting common forms like collective donations and crowdfunding. Future research could explore these specific contexts to examine how group size affects decision processes and outcomes. Additionally, individual traits (e.g., self-construal, gender) may moderate effects. Research shows low self-construal individuals balance preferences in groups, while high self-construal individuals prioritize their own preferences in large groups (Wu et al., 2019). Whether our findings apply to self-construal individuals requires further investigation. Third, with metaverse technological changes, future research could examine whether new human-computer interaction-based joint consumption modes produce similar biases. Fourth, organizational teamwork shares mechanisms with joint consumption decision-making, suggesting our conclusions may generalize to organizational management and practice.

8. Conclusion

This research first applies risk-shift theory to consumer decision-making, examining how consumption context (individual vs. joint) affects unfamiliar product preferences through perceived risk and exploring the effect's boundary conditions. Studies 1a and 1b found significant effects under both unilateral and joint decision-making. Study 2 identified perceived risk as the key mechanism. Study 3 demonstrated the effect occurs only for low-risk products and with close companions. Study 4 extended the effect to non-consumption decisions. These conclusions expand joint consumption and exploratory behavior research while providing valuable guidance for companies selling unfamiliar products.

References

Ariffin, S. K., Mohan, T., & Goh, Y. N. (2018). Influence of consumers' perceived risk on consumers' online purchase intention. Journal of Research in Interactive Marketing, 12(3), 309–327.

Belk, R. (2010). Sharing. Journal of Consumer Research, 36(2), 715–734.

Benoit, S., Baker, T. L., Bolton, R. N., Gruber, T., & Kandampully, J. (2017). A triadic framework for collaborative consumption (CC): Motives, activities and resources & capabilities of actors. Journal of Business Research, 79, 219–227.

Bettman, J. R. (1973). Perceived risks and its components—A model and empirical test. Journal of Marketing Research, 10(2), 184–190.

Boothby, E. J., Clark, M. S., & Bargh, J. A. (2014). Shared experiences are amplified. Psychological Science, 25(12), 2209–2216.

Celso Augusto de, M., & Anderson, K. (2018). The effects of environmental factors on B2B e-services purchase: Perceived risk and convenience as mediators. Journal of Business & Industrial Marketing, 34(7), 767–778.

Chaudhuri, A. (1998). Product class effects on perceived risk: The role of emotion. International Journal of Research in Marketing, 15(2), 157–168.

Chen, Z. (2017). Social acceptance and word of mouth: How the motive to belong leads to divergent WOM with strangers and friends. Journal of Consumer Research, 44(3), 613–632.

Cheron, E. J., & Ritchie, J. R. B. (1982). Leisure activities and perceived risk. Journal of Leisure Research, 14(2), 139–154.

Choi, B., & Ahluwalia, R. (2013). Determinants of brand switching: The role of consumer inferences, brand commitment, and perceived risk. Journal of Applied Social Psychology, 43(5), 981–991.

Choi, N. H., Jung, J. M., Oyunbileg, T., & Yang, P. (2016). The impact of emotional arousal levels and valence on product evaluations: From regulatory goal perspective. European Journal of Marketing, 50(1/2), 78–99.

Clauzel, A., Guichard, N., & Riché, Caroline. (2019). Dining alone or together? The effect of group size on the service customer experience. Journal of Retailing and Consumer Services, 47, 222–228.

DelVecchio, D. & Smith, D. C. (2005). Brand-extension price premiums: The effects of perceived fit and extension product category risk. Journal of the Academy of Marketing Science, 33(2), 184–196.

Dowling, G. R. & Staelin, R. (1994). A model of perceived risk and intended risk-handling activity. Journal of Consumer Research, 21(1), 119–134.

Dubé, J. P., Hitsch, G. J., & Rossi, P. E. (2010). State dependence and alternative explanations for consumer inertia. Rand Journal of Economics, 41(3), 417–445.

Etkin, J. (2016). Choosing variety for joint consumption. Journal of Marketing Research, 53(6), 1019–1033.

Faul, F., Erdfelder, E., Buchner, A., & Lang, A. G. (2009). Statistical power analyses using GPower 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41*(4), 1149–1160.

Fei, X., You, Y., & Yang, X. (2019). "We" are different: Exploring the diverse effects of friend and family accessibility on consumers' product preferences. Journal of Consumer Psychology, 30(3), 543–550.

Ferguson, D. A., & Vidmar, N. (1971). Effects of group discussion on estimates of culturally appropriate risk levels. Journal of Personality and Social Psychology, 20(3), 436–445.

Garcia-Rada, X., Anik, L., & Ariely, D. (2019). Consuming together (versus separately) makes the heart grow fonder. Marketing Letters, 30(1), 27–43.

Gardner, M., & Steinberg, L. (2005). Peer influence on risk taking, risk preference, and risky decision making in adolescence and adulthood: An experimental study. Developmental Psychology, 41(4), 625–635.

Golman, R., Loewenstein, G., Molnar, A., & Saccardo, S. (2019). The demand for, and avoidance of, information. Management Science, 1–19.

Gorlin, M., & Dhar, R. (2012). Bridging the gap between joint and individual decisions: Deconstructing preferences in relationships. Journal of Consumer Psychology, 22(3), 320–323.

Han, M. C., & Kim, Y. (2017). Why Consumers Hesitate to Shop Online: Perceived Risk and Product Involvement on Taobao.com. Journal of Promotion Management, 23(1), 24–44.

Hensley, W. E. (1977). Probability, personality, age, and risk taking. The Journal of Psychology, 95(1), 139–145.

Herman, C. P., Roth, D. A., & Polivy, J. (2003). Effects of the presence of others on food intake: A normative interpretation. Psychological Bulletin, 129(6), 873–886.

Jing, X., & Xie, J. (2011). Group buying: A new mechanism for selling through social interactions. Marketing Science, 57(8), 1354–1372.

Johnson, D. L., & Andrews, I. R. (1971). Risky-shift phenomenon tested with consumer products as stimuli. Journal of Personality and Social Psychology, 20(3), 382–385.

Kahn, B. E. (1995). Consumer variety-seeking among goods and services. Journal of Retailing and Consumer Services, 2(3), 139–148.

Lashkova, M., Antón, C., & Camarero, C. (2019). Dual effect of sensory experience: Engagement vs diversive exploration. International Journal of Retail & Distribution Management, 48(2), 128–151.

Lejarraga, T., Hertwig, R., & Gonzalez, C. (2012). How choice ecology influences search in decisions from experience. Cognition, 124(3), 334–342.

Levinger, G., & Schneider, D. J. (1969). Test of the "risk is a value" hypothesis. Journal of Personality and Social Psychology, 11(2), 165–169.

Litman, J. A., & Spielberger, C. D. (2003). Measuring epistemic curiosity and its diversive and specific components. Journal of Personality Assessment, 80(1), 75–86.

Liu, P. J., & Min, K. E. (2020). Where do you want to go for dinner? A preference expression asymmetry in joint consumption. Journal of Marketing Research, 57(6), 1037–1054.

Lowe, M. L., & Haws, K. L. (2014). (Im)moral support: The social outcomes of parallel self-control decisions. Journal of Consumer Research, 41(2), 489–505.

Lowe, M., Nikolova, H., Miller, C. J., & Dommer, S. L. (2019). Ceding and succeeding: How the altruistic can benefit from the selfish in joint decisions. Journal of Consumer Psychology, 29(4), 652–661.

Lowengart, O., & Tractinsky, N. (2001). Differential effects of product category on shoppers' selection of web-based stores: A probabilistic modeling approach. Journal of Electronic Commerce Research, 2(4), 142–156.

Mehlhorn, K., Newell, B. R., Todd, P. M., Lee, M. D., Morgan, K., Braithwaite, V. A., Hausmann, D., Fiedler, K., & Gonzalez, C. (2015). Unpacking the exploration–exploitation tradeoff: A synthesis of human and animal literatures. Decision, 2(3), 191–215.

Nikolova, H., Lamberton, C., & Coleman, N. V. (2018). Stranger danger: When and why consumer dyads behave less ethically than individuals. Journal of Consumer Research, 45(1), 90–108.

Parker, J. R., Nita, U., & Schleicher, M. G. (2019), How and why the collaborative consumption of food leads to overpurchasing, overconsumption, and waste. Journal of Public Policy & Marketing, 38(2) 154–171.

Philp, M., & Nepomuceno, M. V. (2020). When the frugal become wasteful: An examination into how impression management can initiate the end-stages of consumption for frugal consumers. Psychology & Marketing, 37(2), 260–275.

Polman, E., Effron, D. A., & Thomas, M. R. (2018). Other people's money: Money's perceived purchasing power is smaller for others than for the self. Journal of Consumer Research, 45(1), 109–125.

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891.

Ran, G., Zhang, Q., & Huang, H. (2018). Behavioral inhibition system and self-esteem as mediators between shyness and social anxiety. Psychiatry Research, 270, 568–573.

Ran, Y. X., Liu, J. N., Zhang, Y. S., & Wei, H. Y. (2020). The magic of one person: The effect of the number of endorsers on brand attitude. Acta Psychologica Sinica, 52(3), 371–385.

Ran, Y. X., Niu, Y. X., & Chen, S. Y. (2021). "More" is less: Why multiple payment mechanism impairs individual donation. Acta Psychologica Sinica, 53(4), 413–430.

Ran, Y. X., Wei, H. Y., Maglio, S. J., Huang, M., & Li, Q. (2017). How and when the size of apology representative affects consumer forgiveness. Nankai Business Review, 20(4), 38–48.

Riefer, P. S., Prior, R., Blair, N., Pavey, G., & Love, B. C. (2017). Coherency-maximizing exploration in the supermarket. Nature Human Behaviour, 1, 0017.

Rudd, M., Hildebrand, C., & Vohs, K. D. (2018). Inspired to create: Awe enhances openness to learning and the desire for experiential creation. Journal of Marketing Research, 55(5), 766–781.

Simonson, I. (1989). Choice based on reasons: The case of attraction and compromise effects. Journal of Consumer Research, 16(2), 158–174.

Simpson, J. A., Griskevicius, V., & Rothman, A. J. (2012). Consumer decisions in relationships. Journal of Consumer Psychology, 22(3), 304–314.

Skard, S., & Thorbjørnsen, H. (2014). Is publicity always better than advertising? The role of brand reputation in communicating corporate social responsibility. Journal of Business Ethics, 124(1), 149–160.

Snyder, M., & Monson, T. C. (1975). Persons, situations, and control of social-behavior. Journal of Personality and Social Psychology, 32(4), 637–644.

Soesilo, P. K. M., Gunadi, W., & Arimbi, I. R. (2018). The effect of endorser and corporate credibility on perceived risk and consumer confidence: The case of technologically complex products. Journal of Marketing Communications, 26(6), 1–21.

Spence, H. E., Engel, J. F., & Blackwell, R. D. (1970). Perceived risk in mail order and retail store buying. Journal of Marketing Research, 7(3), 364–369.

Steenkamp, J. B. E. M., & Baumgartner, H. (1992). The Role of Optimum Stimulation Level in Exploratory Consumer Behavior. Journal of Consumer Research, 19(3), 434–448.

VanLear, C. A., Koerner, A., & Allen D. M. (2006). Relationship typologies. In Vangelisti AL, Perlman D (Eds.), The Cambridge handbook of personal relationships (pp. 91–112). Cambridge: Cambridge University Press.

Walter, M., Hildebrand, C., Häubl, G., & Herrmann, A. (2020). Mixing it up: Unsystematic product arrangements promote the choice of unfamiliar products. Journal of Marketing Research, 57(3), 509–526.

Whyte, G. (1991). Diffusion of responsibility: Effects on the escalation tendency. Journal of Applied Psychology, 76(3), 408–415.

Winet, Y. K., Tu, Y., Choshen-Hillel, S., & Fishbach, A. (2020). Social exploration: When people deviate from options explored by others. Journal of Personality and Social Psychology. Advance online publication, https://doi.org/10.1037/pspi0000350.

Woodside, A. G. (1974). Is there a generalized risky shift phenomenon in consumer behavior? Journal of Marketing Research, 11(2), 225–226.

Wu, E. C., Moore, S. G., & Fitzsimons, G. J. (2019). Wine for the table: Self-construal, group size, and choice for self and others. Journal of Consumer Research, 46(3), 508–527.

Wu, Y., Hamilton, R. W., Kim, N. Y. J., & Ratner, R. K. (2021). Navigating shared consumption experiences: Clarity about a partner's interests increases enjoyment. Journal of Marketing Research, 58(3), 439–455.

Xu, L., Jiang, Y. R., Cui, N., Zhang, L. X., Zhao, S. Q. (2020). Hometown is the most reassuring place: The impact of ontological security threat on preference of hometown brands. Acta Psychologica Sinica, 52(4), 513–527.

Yang, L. W., Chartrand, T. L., & Fitzsimons, G. J. (2015). The influence of gender and self-monitoring on the products consumers choose for joint consumption. International Journal of Research in Marketing, 32(4), 398–406.

Zizzo, D. J. (2010). Experimenter demand effects in economic experiments. Experimental Economics, 13(1), 75–98.