A factor that could affect consumers’ variety-seeking behaviors is individual demographics, such as gender and age. For the effect of gender on variety-seeking behavior, researchers focused on the feminine menstrual cycle and gender differences. For example, across the reward domains of mating and hedonic food, Faraji-Rad et al. (2013) showed that women seek more variety in rewards when they are closer to ovulation because of their increased reward sensitivity caused by hormonal shifts during the fertile phase of the menstrual cycle. Similarly, Durante and Arsena (2015) revealed that women select a greater number of unique options from consumer product sets at high fertility, which is particularly strong for those in committed relationships. Chen et al. (2016) focused on two genders and demonstrated that men’s variety-seeking behavior in the product consumption domain increases in the presence of short-term, not long-term mating cues; by contrast, women’s variety-seeking behavior decreases in the presence of long-term but not short-term mating cues. For the effect of age on variety-seeking behavior, Novak and Mather (2007) found that younger adults selected similar levels of variety when choosing between what to consume immediately and later. By contrast, older adults consistently selected less variety when choosing something to be consumed later than immediately.

Individual characteristics could influence consumers’ variety-seeking behaviors. Consumers who feel powerful (Jiang et al., 2014) are chronically indecisive (Jeong et al., 2016; Jeong and Drolet, 2016) and are novices (Sela et al., 2019) present more variety-seeking behaviors. First, building on an action-orientation perspective of power, Jiang et al. (2014) demonstrated that because high power is associated with a readiness to act and switching behavior generally requires taking actions in some form, consumers who feel powerful are more likely to switch in choice tasks. Second, Jeong et al. (2016) and Jeong and Drolet (2016) highlighted that chronic indecisiveness is associated with increased variety-seeking behavior. Chronically indecisive consumers (vs. not) feel less anxious and more positive after selecting a mix of products. Finally, consumers can acquire knowledge and signal their status in the marketplace during variety-seeking. Sela et al. (2019) argued that novices (vs. experts) perceive greater (vs. less) variety-seeking to indicate expertise because of perceived category breadth knowledge (vs. within-category discernment). Thus, novices (vs. experts) seek more (vs. less) variety to signal expertise. However, privately self-aware consumers are less inclined to opt for a varied choice set (Goukens et al., 2009).

In recent years, researchers explored luck beliefs, mindset traits, and self-oriented perfectionism in consumers’ variety-seeking behaviors. For instance, Zhao et al. (2021b) analyzed data from 593 respondents and showed that personal luckiness and belief in luck positively affect variety seeking. Li and Sun (2021) investigated 364 participants in the United States and found that consumers with a growth (vs. fixed) mindset are more likely to engage in variety seeking. As a purchasing strategy, variety-seeking also could be positively influenced by self-oriented perfectionism (Fu et al., 2021; N = 312).

Other personality traits, such as goal orientation and trait anger, influence variety-seeking behaviors depending on the situations. Considering decision tasks, Wu and Kao (2011) found that in the sequential choices for sequential consumption conditions, promotion-focused consumers tend to select a greater variety of items than prevention-focused consumers. The effect reversed in the simultaneous choices for sequential consumption conditions for prevention-focused consumers. Considering state anger, Zhao et al. (2021a) showed that people from relatively resource-abundant environments generally tend to seek variety when they are temporarily in an angry mood, independent of trait anger; although those with low trait anger tend to choose more variety compared to those with high trait anger. For people growing up in relatively resource-scarce environments, those with a low trait of anger tend to choose less variety when they feel angry than those with a high trait of anger.

Early researchers mainly explored the relationship between broad emotions (positive and negative feelings) and variety-seeking. For example, Kahn and Isen (1993) explored the influence of the positive effect on variety-seeking among safe and enjoyable products. The findings revealed that the positive affect induced by a gift bag of candy or sugarless gum enhanced consumers’ variety-seeking in choice behavior in three food categories (i.e., crackers, soup, and snack food) when circumstances did not make negative features of the items. However, the different degrees of positive feelings could produce distinct effects. Roehm and Roehm (2005) believed that more extreme positive moods might reduce variety-seeking—unlike mild positive moods—because the moderate stimulation obtained from variety-seeking is insufficient to meet people’s demands of extreme positive moods. The results of two pilot studies and two experiments showed that participants who viewed an ad cultivating an extremely positive mood switched less between candy bar snack brands on successive choices and selected fewer brands.

Then, researchers discussed how specific emotions and physical conditions, including positive and negative emotions (Chuang et al., 2008), sadness and happiness (Lin and Lin, 2009; Chien-Huang and Hung-Chou, 2010, 2012; Lin et al., 2011; Lin, 2014), local optimism and pessimism (Yang and Urminsky, 2015), and winning-losing perception (Chang et al., 2021), affect consumers’ decision-making behaviors when faced with multiple choices. In these moods and states, seeking variety helps people change their current status. For example, a study with 124 subjects demonstrated that people are likely to include more variety in their consumption decisions when they are induced to a negative emotion than a positive emotion (Chuang et al., 2008). Moreover, a series of research discussed the effect of two specific emotional states (sadness and happiness) on variety-seeking behaviors and found similar conclusions (Lin and Lin, 2009; Chien-Huang and Hung-Chou, 2010, 2012; Lin et al., 2011; Lin, 2014). These studies used choice task scenarios and revealed that participants with a sad mood selected more variety than those with a happy mood. Furthermore, Yang and Urminsky (2015) demonstrated that local optimism increases sequential choice consistency, whereas local pessimism increases sequential variety-seeking. Finally, Chang et al. (2021) found that consumers who have failed in a competition or not achieved a goal tend to seek less variety in their later consumption than consumers who have succeeded because losing feedback weakens consumers’ perception of their control of personal mastery.

Interestingly, some special physiological states have effects on variety-seeking, such as hunger (Goukens et al., 2007) and sleepiness (Huang et al., 2019). When people felt hunger or thirst, visual food or drink cues encouraged them to seek variety in relevant domains because these cues were more attractive to consumers who were in hunger or had just finished a fitness (Goukens et al., 2007). Another physiological state influencing variety-seeking is sleepiness. Huang et al. (2019) used multiple methods and revealed that sleepier consumers tended to seek more variety because of the need for arousal to maintain wakefulness. Particularly in Study1, a natural experiment based on the change of DST policy provided practical evidence for the positive effect of DST (decreasing short-term sleeping time and increasing sleepiness) on variety-seeking in products purchased by using Nielsen panel data (approximately 60,000 U.S. households data).

Individuals’ perception of the external circumstances depends on their keen sensory system, which receives various stimuli from the outside and then influences individuals’ mindset and decision making. People seek various choices when consuming to satisfy the sensory demand of vision (Maimaran and Wheeler, 2008; Deng et al., 2016; Huang and Kwong, 2016) and taste (Inman, 2001; Mukherjee et al., 2017), which have been discussed more in the current research.

Initially, the structural and superficial features of vision affect consumers’ variety-seeking behaviors subconsciously. First, individuals’ choices could be causally influenced by novel visual stimuli. For example, Maimaran and Wheeler (2008) demonstrated that exposure to variety arrays (arrays of differing shapes) increases variety-seeking, whereas exposure to uniqueness arrays (e.g., one circle among six squares) increases the choice of unique over common objects. Second, the display of products further influences variety-seeking in consumption because of the direction match between displays and eye movements. For example, Deng et al. (2016) used multiple methods (e.g., field study, laboratory study, and eye-tracking study) and demonstrated that consumers chose more variety (i.e., distinct fragrances, different candies, unique chocolates, and different types of lollipops) when alternatives were horizontally assorted or displayed. Third, a superficial feature can affect various perceptions even when the actual content or structure of an assortment remains unchanged. Huang and Kwong (2016) revealed that when the menu or catalog of an assortment is more difficult to read, the individuals perceived a higher variety. This readability effect stems from the subjective interpretation of the feeling of difficulty, that is, consumers generally endorse a lay belief that it is more difficult to make choices when they face a greater variety of options.

Subsequently, people might seek variety of taste stimuli to satisfy their needs. Inman (2001) believed that people switch more on sensory attributes (e.g., flavor) than nonsensory attributes (e.g., brand) to seek more pleasure. Inman (2001) used ACNielsen wand panel data for purchases of tortilla chips and cake mixes from almost 2000 consumers over 3 years (Study 1) and examined actual consumption behavior using a six-week consumption diary panel from over 850 consumers in two cities (Study 2) and employed a survey methodology (Study 3; 1056 responses) to verify his hypotheses: the difference of variety-seeking based on sensory and brand could be explained by “sensory-specific satiety,” that is, because of the high correlation between sensory-specific satiety and variety-seeking on sensory attributes, consumers switched more on flavors than brands. The research of Inman (2001) on sensory is broad, and subsequently, Mukherjee et al. (2017) discussed the relationship between a more specific taste—spicy and variety-seeking consumption. Based on embodied cognition and the metaphor “variety is the spice of life,” the authors found that spicy gustatory sensations (e.g., spicy vs. mild potato chips) activate a desire to be interesting that leads to greater variety in the subsequent unrelated choices (e.g., candy bars).

Ultimately, Lee and Sergueeva (2017) demonstrated an interesting “chewing effect” and argued that chewing more increases the viewing time and consumers’ thought-engagement while shopping and then increases variety-seeking behavior among consumers.

Variety-seeking could also be the behavioral result of spontaneous thinking. The priming mindset influences variety-seeking in follow-up consumption, including past experiencing priming (Shen and Wyer, 2010) and semantic concept priming (Fishbach et al., 2011; Huang and Wyer, 2015; Zhang and Guo, 2019).

First, people’s past experiences can affect variety-seeking in the future. When individuals’ past behaviors associated with “same” were primed, they would get the feeling of boredom and then switch to a “different” decision rule (e.g., various types of herbal tea for four consecutive days) when performing a later task to eliminate this negative feeling (Shen and Wyer, 2010).

Second, the influence of semantic concepts on variety-seeking is nonconscious. For example, Fishbach et al. (2011) showed that when the negative concept related to “repetition” (e.g., boredom) was primed, it triggered an individual’s consumption structure based on satisfaction, that is, encouraging them to seek variety in order selection (e.g., buying smaller bottles of different shampoo, preferring CDs from different artists, staying in different hotels in the same city, visiting different cities in Europe, shopping at different stores, and choosing different snacks). Moreover, the influence of semantics is not only manifested in words related to choice behaviors but also has the same effect in words unrelated to choice behavior. For instance, Huang and Wyer (2015) found two opposite effects of mortality on variety-seeking: anxiety-inducing and concept-activation effects. The former was driven by the desire for stability and decreased the variety of individuals’ choices in an unrelated multiple-choice decision situation, whereas the latter induced a global processing style and increased variety-seeking. In addition, individuals’ temporal perspectives also trigger different seeking mindsets and affect variety-seeking behavior. Zhang and Guo (2019) demonstrated that past thinking brings familiar seeking and decreases variety-seeking, whereas future thinking induces novelty seeking and increases variety-seeking.

The social environment’s influence on people’s daily behavior is subtle and has potential that is not easy to detect. Social factors that influence consumption variety are mainly from the two aspects of social relationships and social culture.

People would like to make various decisions to maintain well social relationships. The first social relationship comes from social pressure. Ratner and Kahn (2002) demonstrated that people choose more variety when they make decisions in public than in private because they expect to receive positive evaluations from others (perceived as “social pressure”; Ratner and Kahn, 2002). The second social relationship comes from interpersonal motivation. According to Ratner and Kahn (2002), Choi et al. (2006) showed that people have a stronger tendency to seek variety when they make choices for others. The explanations are as follows: (a) people should be responsible for their choices (the interpersonal mechanism; b) people expect to be satisfied more quickly when they choose for others (the intrapersonal mechanism). In addition, to maintain the self and interpersonal relationships, individuals’ perceived relational threat affects variety seeking in snack choices. Across three studies, Finkelstein et al. (2019) experimentally manipulated relational self-threat and found that those who experience high (vs. low) threat seek less variety, even when the same choice set is construed as having more (vs. less) variety. The third social relationship comes from the acquisition of interpersonal resources, that is, social influence. Ariely and Levav (2000) showed that the original groups choose more varied dishes than created groups, which is attributable to the interaction among group members and help individual satisfy goals of information gathering and self-presentation in the form of uniqueness in the group context. Chuang et al. (2013) maintained that to derive more enjoyment from a shared product, people show less variety and make choices consistent with the opinions of others in online information.

Furthermore, people in love form a special social relationship. For example, Etkin (2016) argued that consumers prefer more variety for joint consumption with their partners (e.g., going out to dinner, a movie, and a concert on a weekend), when they perceive more (vs. less) time ahead in a committed relationship. Huang and Dong (2019) found that a salient relationship state—romantic crush—can increase consumers’ variety-seeking tendency in unrelated consumption situations.

Variety-seeking behaviors in consumption could be influenced by the root of social culture. Kim and Drolet (2003) highlighted that as a choice rule, people in a unique culture display greater variety. Similarly, Yoon et al. (2011) reported that because members of a collectivist culture tend to follow group members’ choices, their choices in snacks are associated with a higher uniformity-seeking tendency than those of individualistic cultural backgrounds. Moreover, building on the compensation consumption literature, Yoon and Kim (2018) demonstrated that consumers with low socioeconomic status and perceive low economic mobility (e.g., economically stuck consumers) seek more variety than others to compensate for their lack of personal control. Finally, political ideology has a counterintuitive effect on variety-seeking. Fernandes and Mandel (2014) showed that conservatism is positively related to variety-seeking because of social normative concerns.

The physical environment factors that affect variety-seeking in consumption mainly include the space environment and time point.

First, the constraints of a physical space enhance variety-seeking in consumption. Based on resistance theory, Levav and Zhu (2009) found that consumers confined by space make more various and unique choices to resist the invasion of their private space and seek freedom. The authors revealed that people in narrower aisles sought more varied candies than people in wider aisles (Study 1), and this effect of confinement in narrow aisles is extended to more unique choices in charities (Study 2), particularly in those with high chronic reactance tendency (Study 3). Moreover, the field study (94,110,967 usable transactions) used crowding as a proxy for confinement and found a positive relationship between crowding and variety-seeking in real grocery purchases.

As another type of space environment, the restaurant atmosphere, store environment, and web feature also could influence consumers’ variety-seeking. For example, Ha and Jang (2013a) collected 309 useable responses and pointed out that consumers’ desired hedonic and utilitarian values of the restaurant positively influence their variety-seeking intentions. Similarly, according to 617 usable responses to the restaurant experience, Ha and Jang (2013b) showed that atmospheric quality, overall boredom, and boredom with atmospheric attributes significantly influence dinners’ variety-seeking intentions positively. For the off-line store environment, Mohan et al. (2012) investigated 350 shoppers in Dubai and established that the store environment (including lighting, scent, and music) affects variety-seeking positively. For the online web feature, with 698 usable responses, Hung et al. (2011) demonstrated that quality web features affect interpersonal trust and platform credibility positively, and both constructs drive a user’s online community usage and brand variety-seeking behavior.

Second, the objective time of day could further influence variety-seeking in consumption. Given the influence of physical laws, people exhibit different levels of variety-seeking in consumption at different time points. For example, Roehm and Roehm (2004) found that people are more likely to seek variety in candy choices at low arousal (e.g., 9 AM; 10:00 AM–11:20 AM) than peak arousal (e.g., 4 PM; 3:10 PM–4:20 PM) moments of the day. However, the latest research provided an inconsistent result of diurnal variation in variety-seeking. Based on circadian rhythms in chronobiology, Gullo et al. (2019) applied four studies, including an empirical analysis of millions of purchases, and stated that individuals pick less varied flavors of yogurt when choosing in the morning. Furthermore, different external environments and changes in life events can change people’s variety-seeking. Koschate-Fischer et al. (2018) showed that consumers reduce their variety-seeking tendency after experiencing a life event (1,475 panelists).

The marketing strategy influences consumers’ variety-seeking behaviors primarily in the purchase stage. Kahn and Louie (1990) first studied the relationship between retail stores’ promotion strategy and variety-seeking. They found that if only one shampoo brand is promoted and people are generally loyal to the last brand purchased, they tend to switch among shampoo brands when the promotion is withdrawn.

In the later stages, the research on the impact of marketing strategy has become in-depth, such as product packaging, product bundle strategy, product category and information, and product assortment. For example, product packaging uniformity is associated with arousal potential and influences consumers’ variety-seeking. Roehm and Roehm (2012) showed that consumers’ variety-seeking is greater in product categories where packaging is similar among competitors.

Furthermore, the product bundle strategy affects consumers’ variety-seeking when they experience multiple products. Mittelman et al. (2014) found that consumers seek more variety when choosing from single offerings (e.g., a choice of two individual candy bars) than from bundled offerings (e.g., a choice of a bundle of two candy bars), which is termed “offer framing effect.” Kim et al. (2018) based on the decision-framing effect and found that travelers show higher variety-seeking in travel package decisions when the bundle package is selected from a combined decision rather than from two single decisions.

Moreover, product category and product information affect variety-seeking behavior. For the product category, several researches were conducted from various perspectives. Based on a specific-abstract categorization strategy, Kim and Yoon (2016) showed the “category specificity effect” and revealed that individuals are likely to order a greater variety of dishes when the menu contains no category labels or abstract category labels due to the enhanced perception of variety offered in the menu. Baltas et al. (2017) indicated that in hedonic product categories, consumers seek more variety in sensory attributes, whereas, in utilitarian product categories, they seek more variety in functional attributes. What is the difference between digital and consumable goods? Adomavicius et al. (2015) showed a reduction in behavioral effects of bundle cohesion and timing on variety of preferences for digital goods. For the product information, Lin et al. (2017) indicated that when people purchase products for themselves, the presence of risky information and health claims, and high product involvement promote more variety-seeking.

Finally, as detailed in Section “Theoretical Perspective and Underlying Mechanism”, the displays and assortments of products affect consumers’ variety-seeking behaviors. For example, the display of novel geometric figure arrangement combinations (various shapes) increases consumers’ variety-seeking (Maimaran and Wheeler, 2008). The horizontal assortment is easier to process and can increase individuals’ perceived variety, thereby ultimately leading to greater variety-seeking (Deng et al., 2016; Table 1).

Optimal stimulation level theory is an early and fundamental theory to explore variety-seeking behavior in consumption, which is widely applied in the existing literature. One reason consumers seek variety in product selection is to meet their demand for stimulation (Menon and Kahn, 1995). According to optimal stimulus level theory, the relationship among internal individual factors, external environmental factors, and consumer preference response can be represented by an inverted U-shaped curve function. In this curve function, the peak vertex of the curve is the optimal stimulus level, the attribute set under this level can cause the consumer’s satisfaction to reach the highest level, and the stimulus level on both sides of the vertex is too low or too high to satisfy the consumer (McAlister, 1982). If consumers often buy the same product or category, their effective stimulus level in decision-making decreases. Therefore, to obtain greater stimulation, consumers attempt to buy different products or products to achieve their goals (Roehm and Roehm, 2005). In addition, because of physiological stimulation and arousal (e.g., body temperature), consumers receive the least stimulation in the morning and produce a lower variety-seeking (Gullo et al., 2019). Arouse theory was also applied by Roehm and Roehm (2004) and Huang et al. (2019) to explain consumers’ need for stimulation.

As the internal influencing factors, much research focus on the effect of personality characteristics on variety-seeking behavior in consumption from the individual perspective. As a result, theories and underlying mechanism of these effects are in varied forms, which are mostly based on the consumers’ personality traits. For example, Self-awareness theory and Goal orientation theory were adopted to explore how consumers’ self-awareness and promotion–prevention orientation affect their variety-seeking behaviors (Goukens et al., 2009; Wu and Kao, 2011). According to Implicit Theory, consumers with a growth (vs. fixed) mindset are more likely to engage in variety seeking due to their changing preferences (Li and Sun, 2021). Based on Signal theory, Sela et al. (2019) found that variety-seeking behavior can serve as a signal to indicate expertise. Personality characteristics also can shape consumers’ variety-seeking mindset and then promote variety-seeking behaviors (Kim and Yoon, 2016; Zhang and Guo, 2019).

Emotions are the psychological states that people need to face every day. Different emotional states bring different stimulation levels to consumers. Based on the Mood evaluation framework, compared with positive emotions (such as happiness), negative emotions (such as sadness) bring low satisfaction to consumers; therefore, consumers experiencing negative emotions increase their satisfaction through variety-seeking behaviors (Roehm and Roehm, 2004; Lin and Lin, 2009; Chien-Huang and Hung-Chou, 2010, 2012; Lin et al., 2011; Lin, 2014). Building on Processing style theory, mortality salience increases variety-seeking behaviors in consumption by influencing an individual’s global processing style (Huang and Wyer, 2015). Variety-seeking behavior in consumption is observed to help cope with and alleviate the negative effects of negative emotions. Optimal stimulus level theory also can help explain this. In an extremely positive mood state, consumers reduce their variety-seeking behaviors because the stimulus provided by variety-seeking behaviors in consumption belongs to the middle level, which is not enough to meet the demand for extreme positive emotions for stimulation (Roehm and Roehm, 2005). However, consumers’ variety-seeking behavior when in a mildly positive mood (moderate degree) is influenced by product characteristics, such as security and pleasantness (Roehm and Roehm, 2005).

The theory of sense of control is the core element in the compensatory consumption perspective. Compensatory consumption means that consumers engage in certain consumption behaviors to make up for the lack of psychological needs because of the lack of overall self-esteem or self-realization (Gronmo, 1988). The essential feature of compensatory consumption is to make up for psychological defects or threats through consumption behavior, emphasizing consumption behavior as an alternative means and tool—rather than functional value—to meet demand. Compensatory consumption is a kind of pure psychological consumption and self-presentation of psychological imbalance. Therefore, in a variety of scenarios in which psychological defects and threats might occur, variety-seeking in consumption can be used as an alternative means to meet psychological needs and cope with threats. For example, because consumers with low social status and perceived low social mobility tend to have a low sense of personal control, they show more variety-seeking behaviors in consumption to compensate for their psychological defects (Yoon and Kim, 2018). If people in love are “left out,” their sense of control in a romantic relationship is reduced—to restore a sense of control, they seek a variety of choices in consumption (Huang and Dong, 2019).

As mentioned earlier, environmental psychology focuses on the relationship between the environment and individuals’ psychology and behavior. The environment includes the physical and social environments, both of which have an important impact on people’s behavior.

First, spatial perception is a physical environment. According to the Resistance theory, if consumers feel constrained (such as in narrow aisles and among crowded people), they resist the invasion of private space through more various and unique choices, which is equivalent to resisting the constraint (Levav and Zhu, 2009). In addition, according to the spontaneous effect, a diversified display of commodities stimulates consumers’ variety mindset, leading to the emergence of variety-seeking in consumption (Maimaran and Wheeler, 2008). Finally, because of a match between the human binocular vision field and the dominant direction of eye movements (which are both horizontal in direction), it is easier for horizontal (vs. vertical) displays to be processed. This processing fluency allows people to browse information more efficiently, which increases perceived assortment variety and ultimately leads to more variety being chosen (Deng et al., 2016).

Second, social groups and the cultural and political factors in the social environment affect variety-seeking behaviors in consumption from different aspects. The influence of society on consumer behavior is mainly constrained by social norms, which could be generalized by interpersonal and intrapersonal motivation. To maintain consistency with the group (normative constraints) and given the influence of group norms or opinion leaders, people might change their original consumption habits that are inconsistent with the reference group (to promote variety-seeking in consumption) or insist that the original consumption habits are consistent with the group (to prevent variety-seeking in consumption; Ariely and Levav, 2000; Ratner and Kahn, 2002; Choi et al., 2006; Fernandes and Mandel, 2014; Etkin, 2016; Finkelstein et al., 2019). Cross-culture theory explains the individual difference in variety-seeking from the root cultural perspective, and collectivism vs. individualism is the main cultural difference. Members of a collectivist culture tend to consist of group members, and their choices are associated with a less variety-seeking tendency than those of individualistic cultural backgrounds (Kim and Drolet, 2003; Yoon et al., 2011).

Evolutionary psychology research focuses on the influence of women’s ovulation period and gender differences, and scholars use the carry-over effect to investigate variety-seeking behavior in consumption between men and women. Given the influence of hormonal changes during the physiological cycle and to meet reproduction needs, women may be more sensitive to rewards and seek variety when seeking a spouse; therefore, they seek various and novel choices extend to irrelevant consumption choice tasks (Faraji-Rad et al., 2013; Durante and Arsena, 2015; Chen et al., 2016). From an evolutionary perspective, Life-history theory demonstrates that people from relatively resource-abundant or relatively resource-scarce childhoods (i.e., childhood SES) often respond differently when faced with an environmental threat (Griskevicius et al., 2013). Variety-seeking may be a risk reduction strategy against uncertainty about future taste preferences in simultaneous choices for future sequential consumptions among people from different degrees of resource childhoods (Zhao et al., 2021a; Table 2).

In the survey method, five-point and seven-point Likert scales are applied to measure participants’ variety-seeking. The items in the scales were adopted from the previous studies. Participants assess how much they would like to purchase or consider new and unfamiliar brands and products. To test the hypothesized relationships, structural equation modeling (SEM) is performed in research. For example, Hung et al. (2011) measured variety-seeking from Kahn et al. (1986), which used five-point Likert scales (1 = strongly disagree, 5 = strongly agree). In the research of Fu et al. (2021), variety seeking was measured by three items from Grünhagen et al. (2012), with a five-point probability scale ranging from 1 (not probable) to 5 (very probable). A sample item is “I am willing to see different food products and brands.” In the research of Zhao et al. (2021b), variety seeking was measured with the five-item Variety Seeking Scale (Helm and Landschulze, 2009). A sample item is “Buying the same product or brand is boring, even if the product or brand is good.” Furthermore, Van Trijp et al.’s (1996) seven-point Likert type scale is also used by Ha and Jang (2013a,b) and Liu et al. (2021). A sample item is “I am very cautious in trying new or different products.”

In the experimental method, researchers usually describe a consumption scenario and ask participants to imagine a choice in this scenario. The two common choices are simultaneous selection (multiple products or services choices at one time) and sequential selection (one product or service at a time, multiple choices in a row). These choice scenarios include food consumption, purchasing behavior, tourism consumption, and so on. Typically, researchers use the number of products or services selected by participants as the variety-seeking index.

Purchasing and selecting products tasks are frequently used as selection scenarios in the research, the majority of which are used for the food selection task. For example, Simonson (1990) asked participants to imagine that they are going to the supermarket, and their shopping list contained eight products, each a different type of good. The author asked the participants to choose one good every day or choose for three days at a time (Simonson, 1990). Many studies followed this research design (Mukherjee et al., 2017; Gullo et al., 2019), such as the purchasing socks task (five out of nine; Yoon and Kim, 2018), the outing task (potato chips choose three out of four; Chen et al., 2016), the teatime reservation task (25 snacks, 20 options; Roehm and Roehm, 2005), and the sandwich pre-arranged task (seven out of nine; Goukens et al., 2007). The number of brand categories that participants selected is recorded as variety-seeking. In addition, the drinks choosing task is also applied in the research. For example, Goukens et al. (2007) designed a drink-selection scenario, in which participants imagined that they received a gift basket and could choose six drinks among eight flavors. Similarly, volunteer tasks (five out of six; Chen et al., 2016) and the tea beverage task (four options) exist (Shen and Wyer, 2010).

Some studies also adopted other forms of selection scenarios. For example, Levav and Zhu (2009) designed the charitable donation task, in which participants can donate their reward for participating in the experiment to one, several, or all six charities. Goukens et al. (2007) designed a holiday scenario, in which participants imagined that they had won a free trip to Sri Lanka, including air tickets, accommodations, and four experience activities. They could choose four out of 16 activities: four beach activities, four outdoor adventures, four sports activities, and four cultural experiences (Goukens et al., 2007, 2009; Huang and Wyer, 2015). Furthermore, other studies considered cross-product categories’ choice tasks, such as food and stationery categories (tea drinks, potato chips, and books; Shen and Wyer, 2010; Huang and Wyer, 2015), daily necessities categories (lipstick, high heels, yogurt, candy, nail polish, and restaurant; Durante and Arsena, 2015), and entertainment activities (drinks, movies, weekend activities; Etkin, 2016; Gullo et al., 2019). In addition, some studies also used behavior switching to measure the variety-seeking in consumption (Jiang et al., 2014). For example, Yang and Urminsky (2015) used magazines, music, and movies as experimental materials and measured their preference conversion through participants’ choices before and after.

This measurement method requires participants to make real choices during the experiment, but participants were not aware that their choices were influenced and recorded. This measurement method makes the variety-seeking behavior appear in a more realistic scenario, reflecting people’s relatively real, and potential choices and increasing the validity of the research results. Researchers usually let participants choose by selecting experimental rewards or compensation.

In the real selection task, many studies use candies or chocolate as the selection stimuli that are finally selected as rewards or compensation for participation considering the convenience of the experiment and the sample. For example, Simonson (1990) rewarded participants with snacks and asked them to choose between sweet and salty snacks (three total groups). Similarly, five rewards were available for choosing among nine snacks (Choi et al., 2006; Durante and Arsena, 2015), up to five desserts, candies, or yogurts (Yoon and Kim, 2018; Huang et al., 2019), the candy list selection task (Ratner and Kahn, 2002; Roehm and Roehm, 2004), choosing three out of six types of candies (Levav and Zhu, 2009), rewarding three of four lollipop flavors (Chen et al., 2016), and the chocolate selection design for three out of four choices and six choices (Maimaran and Wheeler, 2008; Yoon and Kim, 2018). In addition, Deng et al. (2016) fabricated a research purpose as investigating the influence of virtual store lighting on shopping patterns and gave each participant two dollars to buy the displayed candies.

In addition to snack choice tasks, researchers also asked participants to choose stationery frequently used by college student samples. For example, Levav and Zhu (2009) asked participants to choose three out of six color highlighters as rewards in Experiment 4, which was also applied in Gullo et al. (2019). Another distinct and interesting selection task was the flower arrangement task designed by Mittelman et al. (2014), who provided participants with differently colored roses that needed to be put in vases, and used the number of selected colors as a variety-seeking measure.

In recent years, researchers began to call for the study of consumer behavior in the real environment. Scholars used purchasing data generated by consumers to measure the variety-seeking in consumption and analyzed variety-seeking using data obtained from various methods. Among them, consumer panel data from Nielsen and retail stores are often used by researchers (Inman, 2001; Levav and Zhu, 2009; Yoon et al., 2011; Gullo et al., 2019); in such research, researchers typically used the ratio of the number of categories purchased to the total number of categories as a variety-seeking measure. In addition, some researchers conducted field studies among cities (Koschate-Fischer et al., 2018), field experiments (Yoon et al., 2011; Deng et al., 2016), or natural experiments (Huang et al., 2019) to obtain real behavior data. Kahn et al. (1986) provided an analytical framework for how to use panel data to define and measure variety-seeking and offered seven simple and verifiable models commonly used in the marketing domain.

Universal Product Codes (UPCs) are useful and helpful when adopted to calculate consumers’ variety-seeking behaviors. Inman (2001) used UPCs to construct two indexes to measure consumers’ observed switching (the observable flavor or brand switching percentage) and expected switching (which is calculated based on Zero Order; Grover and Srinivasan, 1987). The switching index is then calculated as: Relative Switching Intensity = (Observed – Expected) _Flavor–(Observed – Expected) _Brand.

Levav and Zhu (2009) used purchasing data in Study 5 to compute a variety-seeking index that captured the extent of variation in a transaction. This was computed for each customer by dividing the number of unique UPCs purchased in a category by the category’s total purchases. The authors used its log odds to conduct an OLS regression using this variety index, with log (variety/(1- variety)), as the dependent variable. Gullo et al. (2019) followed Levav and Zhu (2009), using scanner panel data from a major grocery chain’s single California location. They defined variety as the number of unique UPCs purchased in a category relative to the number of total items purchased. Similarly, Huang et al. (2019) used the Chicago Nielsen consumer panel data set and the number of UPCs per trip to measure variety-seeking.

In addition, Koschate-Fischer et al. (2018) combined two datasets, an individual-level consumer panel and a survey, collected over 3 years. They used the change in SOW and SOU to compute variety-seeking. SOW is the share of wallet, defined as the percentage of money a customer allocates to the preferred brand in a category (our unit of analysis). SOU is the share of units, defined as the percentage of units purchased for the preferred brand in a particular category, controlling for price level effects (Table 3).