Effects of Visual Cues and Social Density on Beverage Consumption: A Field Experiment in a Bar

Abstract

We conducted a randomized controlled field study to explore the effectiveness of sensory marketing on beverage consumption patterns in a real bar setting. Specifically, we examined (a) the effect of visual elements (i.e., consumption-inducing text messages on coasters), (b) the effect of social density, and (c) the joint effect of visual elements and social density. We manipulated coaster type (visual consumption-inducing messages either present or absent), measured social density, and collected sales data. The results show that visual elements have a significant effect on beverage consumption, but social density does not. The joint effect between the two factors is significant such that the effect of visual elements is higher when social density is low. This study contributes to the sensory marketing literature by revealing the interaction between visual and spatial cues in a field setting.

Keywords

sensory marketing visual elements social density field study

Introduction

The popularity of sensory marketing reflects the effectiveness of using consumers’ senses to enhance consumption experiences and sales (Hinestroza & James, 2014; Hultén, 2011; Krishna, 2012; Streicher & Estes, 2016). Consumers are sensitive to marketing messages involving physical sensations, and such sensations have a tremendous influence in their attitudes and behaviors (Krishna, 2012; Lowe & Haws, 2017). Many successful companies, including hospitality firms, use sensory stimuli to enhance customer satisfaction with products or retail environments (Krishna et al., 2017; Spence et al., 2014).

Krishna (2012) defined sensory marketing as “marketing that engages the consumers’ senses and affects their perception, judgment, and behavior” (p. 333). Prior research has examined various sensory modalities such as vision, touch, hearing, smell, and taste, and shown that sensory cues create holistic consumer experiences (Brasel & Gips, 2014; Krishna et al., 2017). For instance, prior research shows that visual elements such as color (Labrecque & Milne, 2012; Lick et al., 2017) and visual presentation (Cornil et al., 2017; M. Kim & Lennon, 2008) influence consumer expectations, perceptions, attitudes, and behavioral intention. Olfactory cues, such as unique scents, increase purchase intention and pleasure (Doucé & Janssens, 2013; Moore, 2014). Moreover, Ryu and Jang (2007) demonstrate the effect of sound as well as lighting on customer pleasure. Zhu and Meyer (2017) show the tactile effect on purchase intention. Despite the popularity of research on sensory marketing, very few studies involved actual business performance such as the number of purchases or revenues.

Sensory cues interact with other physical elements in the Servicescape. One such element is social density. Often used interchangeably with social crowdedness (see O’Guinn et al., 2015), social density reflects to consumers’ perceptions of spaciousness in relation to other consumers in a retail setting (Consiglio et al., 2018). Social density has received increasing attention in the marketing literature (Madzharov et al., 2015; O’Guinn et al., 2015). Hospitality operators need to understand the impact of social density on consumer responses (Machleit et al., 1994; Shirai, 2017). In particular, the recent Covid-19 pandemic has intensified the importance of social density in the Servicescape. Does lower social density due to social distancing increase or hamper the effectiveness of sensory cues? Addressing this question has serious implications to the hospitality industry facing severe negative shocks. In addition, previous research hints to the potential interactive effects between sensory cues and social density, yet empirical investigations of such effects between sensory cues and social density are scant (Hock & Bagchi, 2018; Line et al., 2018). In particular, there is a call for field experiments to examine the joint impact of such cues on consumer behavior (Hinestroza & James, 2014; Shirai, 2017).

To address these research gaps, we examined (a) the effect of visual elements (i.e., consumption-inducing messages on coasters), (b) the effect of social density, and (c) their joint effect on patrons’ beverage consumption in a real bar. Compared to previous studies focusing on perceived crowding, (e.g., D. Kim et al., 2016; Wang & Ackerman, 2019), this study captured the actual density in the bar. The findings of this field experiment show that the consumption-inducing message on the coaster had a positive impact on beer consumption while social density failed to have a direct impact on beer consumption. However, the impact of the consumption-inducing coaster on beverage orders was stronger with decreasing social density. This result highlights the importance of designing and implementing sensory marketing when the hospitality industry is under density control measures such as social distancing.

This study contributes to the hospitality literature in several ways. First, this study explores the impact of visual elements on consumption behaviors. Specifically, this study adopts a text message cue, which has been largely ignored in the hospitality literature. Second, we examine the role of social density in the bar setting using a randomized controlled field experiment. Finally, this field experiment contributes to theory development as the current literature is silent on the joint effect of visual elements and social density on actual consumption behaviors rather than proxies such as behavioral intention.

Background

Visual Elements

As consumers use their senses to perceive, experience, and interact with the retail environment, subconscious information processes influence their thoughts and decisions (Krishna et al., 2017; Krishna & Schwarz, 2014). Prior research shows that effectively manipulating sensory cues (i.e., information that is heard, smelled, touched, tasted, and/or seen) influences consumers’ cognitions and emotions, thus inducing positive purchase behaviors (Krishna, 2012; Michon et al., 2005). Vision is considered the most dominant sense (Hecht & Reiner, 2009; Lowe & Haws, 2017). Accordingly, vision is heavily used in marketing, and more than 80% of marketing communications involve visual senses (Ali & Ahmed, 2019; Aprilianty & Purwanegara, 2016). Property comprises of various elements including images, shapes, colors, lightings, and short phrases (Esmailpour & Zakipour, 2016; Labrecque et al., 2013; Raghubir, 2010), and such elements are incorporated in brand logos, symbols, interior and exterior design elements, packaging, and advertisements (Hultén, 2015; Krishna et al., 2016).

The focus of this study is on short written messages. Specifically, we investigated whether a short message printed on a beverage coaster influences bar patrons’ beverage consumption. Written phrases stimulate a sense of sight and as such they are considered elements of visual properties in a broader sense (Krishna et al., 2016; Raghubir, 2010). Brief and simple phrases are commonly used in advertisements. Advertising copies or slogans provide consumers with descriptive and persuasive cues about the product/service (Barisic & Blazevic, 2014; Elder & Krishna, 2010). Due to their concreteness, written massages are easy to recall, and they are particularly effective for motivated consumers who can understand the meaning of the text (Childers & Houston, 1984; Tang et al., 2004). Different types of information induce various information processing strategies (Li et al., 2019). Visual stimulation with written messages might evoke imagery processing (M. Kim & Lennon, 2008). Lutz and Lutz (1978) suggest that “imagery is a mental event involving visualization of a concept or a relationship” (p. 611). Imagery information processing helps consumers visually imagine a situation of owning products/services (MacInnis & Price, 1987). Therefore, it is effective in enhancing consumer attitudes and behavioral intentions (M. Kim & Lennon, 2008; Lutz and Lutz, 1978; McMahon, 1973).

Out experimental stimulus, a short text message, is expected to evoke images of a sensory experience. We thus examine if such evoked imagery will induce beverage consumption. Thus, we hypothesize:

Hypothesis 1 (H1): The presence (vs. absence) of consumption-inducing text messages on coasters will have a positive impact on patrons’ beverage consumption (i.e., the number of beverages).

Social Density

Copious evidence shows that both the physical and the social environment influence consumers’ purchase decisions (Hwang et al., 2018; Pine & Gilmore, 1999). Social density or crowding refers to consumers’ holistic assessments of overall spaciousness, the number of people in a given space, and the physical distance between individuals (Consiglio et al., 2018; Madzharov et al., 2015). Social density is essential to the retail environment and a driver of consumers’ shopping experiences (Eroglu, Machleit, & Chebat, 2005). However, it is also a very complex concept. Social density can influence consumers’ emotions and behaviors both positively and negatively (e.g., Harrell et al., 1980; Kim, Lee, & Sirgy, 2016). The effect of social density on consumers’ cognition, emotions, and behaviors may depend on situational goals (Michon et al., 2005), expectations (Kim, Lee, & Sirgy, 2016; Pons & Laroche, 2007), and types of consumption venues (Hanks et al., 2017; Machleit et al., 2000).

Numerous research findings suggest that a high level of social density has a negative impact on customer satisfaction in a retail setting (e.g., Eroglu, Machleit, & Barr, 2005). When a person recognizes that his or her personal space is encroached, anxiety and fear might be evoked as a human defense system response (Jacobsen et al., 2019; McNaughton & Corr, 2004). In crowded environments, people are more likely to feel psychological stress, a loss of control (Hui & Bateson, 1991), and a loss of personal space (Stokols, 1972), thus leading to feelings of confusion and frustration (Hyun & Kim, 2015; Machleit et al., 1994). Consequently, social density might have a negative impact on consumer satisfaction (Eroglu & Machleit, 1990; Eroglu, Machleit, & Chebat, 2005) and induce negative product and service evaluations (Huang et al., 2018; O’Guinn et al., 2015).

However, social density may also have a positive impact on consumer perceptions, but such a notion has received less attention (Shirai, 2017). Belonging is a basic human need (Maslow, 1943), and interacting with other consumers in a crowded retail setting might satisfy such a need (Huang et al., 2018). Moreover, consumers might have different levels of tolerance for crowding depending on the consumption context. For example, Hui and Bateson (1991) discovered that consumers’ tolerance for crowding was higher in a bar than in a banking context. The positive impact of social density on satisfaction might be particularly salient in hospitality, tourism, and entertainment contexts such as hotels, restaurants, bars, theme parks, festivals, and sports events (Hanks et al., 2017; Hyun & Kim, 2015; Yarnal & Kerstetter, 2005).

In this article, we argue that in bar environments where the hustle and bustle is expected, crowding is likely to enhance consumer experiences. Specifically, we propose that social density will have a positive impact on bar patrons’ beverage consumption.

Hypothesis 2 (H2): The level of social density is positively correlated with beverage consumption.

Sensory cues are perceived holistically (Lowe & Haws, 2017). The Flow Theory suggests that holistic experiences reflecting the five senses are important for effective marketing communications (Csikszentmihalyi, 2014; Li et al., 2019). The Servicescape model suggests that “the built environment (i.e., the manmade, physical surroundings as opposed to the natural, or social environment)” influences both consumers’ and employees’ emotional reactions and consequent behaviors (Bitner, 1992, p. 58). The Servicescape is composed of ambient conditions such as temperature, music, and scent; space such as layout, furnishings, and equipment; and symbols, signs, and artifacts (Bitner, 1992, p. 60). Tombs and McColl-Kennedy (2002) introduced the notion of social Servicescape showing that social and cultural factors greatly influence consumer behaviors via direct and indirect interactions with other consumers.

Recently, Pizam and Tasci (2019) introduced the term “Experiencescape” defined as

the sensory, functional, social, natural, and cultural stimuli in a product or service environment, surmounted with a culture of hospitality, all of which accrue to an experience for different stakeholders and result in positive or negative cognitive, affective, and behavioral reactions toward products, services, brands, and firms. (p. 26)

The Experiencescape extends the notion of the Servicescape by adding social and cultural components to describe a holistic retail environment. A retail environment is composed of a variety of atmospheric elements, including sensory attributes, and consumers process information by integrating data from various sources (Babin & Attaway, 2000; D. Kim & Perdue, 2013). The Dual coding theory is helpful to understand consumers’ information processing of multisensory stimuli. The Dual coding model assumes a synergistic effect of two independent, but partly related stimuli, on storage, encoding, organizing, and retrieval of information (D’Agostino et al., 1977; Paivio, 1971). For instance, presenting written information with a related picture can enhance memory compared to written information presented alone (Lwin et al., 2010). According to the dual coding theory, two independent cues might produce additive effects on performance (D’Agostino et al., 1977; Paivio, 1971). Accordingly, multiple environmental cues in a retail setting might result in additive effects.

Hypothesis 1 suggests that a consumption-encouraging text message on a coaster as visual stimulation should increase patrons’ beverage consumption. Hypothesis 2 predicts that high social density should increase beverage consumption due to its elevated atmosphere. Each element—visual stimulation and social density might be a part of Experiencescape as they influence consumers’ cognition and behavior in the given retail situation. Relying on the additive effects, we expect that pairing a consumption-inducing message with a socially dense environment should boost consumers’ beverage consumption. We thus put forth the following:

Hypothesis 3 (H3): The positive effect of consumption-inducing text messages is maximized in socially dense environments.

Method

Study Design and Experimental Stimuli

To test our hypotheses, we performed a field experiment in a bar in a metropolitan city in the Republic of Korea. We conducted a 2 (visual elements: present vs. absent) × 2 (social density: low vs. high) true field experiment using a randomized controlled trial (RCT) method.

We chose coasters to test the effect of visual elements because every beverage order in the bar was placed on a coaster. Half of the coasters had a printed message “One More Drink!” in Korean, while the other half had no message. This in-store promotion was appropriate as the bar enforces legal drinking age and has strict policies regarding overconsumption. Moreover, there was no verbal encouragement of ordering alcoholic beverages by the servers. Also, responsible drinking is enforced by drunk driving laws and supported by public education related to health and drinking (Kim, JeKarl, & Lee, 2016). Since visual features other than the message itself also can affect consumer responses, the other design elements were kept constant (see Appendix A).

To measure social density, we documented the level of crowdedness from the perspective of new customers. Our measure of social density was recorded as integer values at the table level. It captures the number of existing customers at the bar from the perspective of a new set of customers when they ordered beverages. Our social density measure was recorded for each table at a specific time when new customers entered the bar and were seated at a table. For example, assume that three new customers entered at 9:30 pm and were seated at Table 1. At that time, there were 22 existing customers at the bar. For these three customers occupying Table 1 at 9:30 pm, the social density value was 22. Now assume that one customer left the bar at 9:42 pm, and a couple was seated at Table 5 at 9:45 pm. For this couple, the social density value will be 22 + 3 − 1 = 24 because there were 22 existing customers plus three occupying Table 1 minus the one customer who just left. If a new set of customers arrived when the bar was completely empty, then the social density for these customers is zero. Since the beer consumption was recorded at the table level, we were able to evaluate the association between social density and beverage consumption for each table.

Table 1.

Description of Participants and Tables (n = 1,074).

Variable	N	%
Gender
Female	549	51.12
Male	525	48.88
Age (observed)
20s	210	50.2
30s	151	36.1
40s	48	11.5
Over 50	9	2.2
Table location (n = 418)
Table	285	68.2
Terrace	114	27.3
Bar	19	4.5
Number of customers per table (n = 418)
1	17	4.1
2	266	63.6
3	77	18.4
4	35	8.4
5	11	2.6
6 or more	12	2.8
Customer composition per table (n = 418)
Alone	17	4.1
2 (1 male and 1 female)	158	37.8
Only males	82	19.6
Only females	93	22.2
Mix (more than 3)	68	16.3

Field Experiment Procedure and Data Collection

The bar had 25 to 30 tables with a capacity to serve 70 to 100 customers, depending on the season. It offered typical beverages and bar food, with beer being the most popular choice. We performed the experiment in late autumn. As this was a medium to low volume period, certain areas of the bar were closed, and the total capacity was 70 customers. The experiment was run on 14 consecutive days by two trained research assistants (RAs).

This was a randomized controlled field experiment. We generated a long list with a random sequence of 0’s and 1’s. Using the list, we randomly assigned the tables to the control group (0 = coasters without a message) and the treatment group (1 = coasters with a message). When new customers were seated at a table, the RAs used the list and distributed coasters only when the next number in the list was one. Once the table has been assigned to either the treatment group or the control group (the same coasters for the entire party), the used number in the list was crossed out, and the next number in the list was used for the next table assignment. In addition to the treatment status of each party at the table, the RAs also recorded the number of guests per table, the time of the first order, the gender composition of customers, their approximate ages, and time of payment. We also collected data from the point-of-sale (POS) system, including table numbers, items ordered, time of payment, price per item, total amount charged, and payment method. For the analysis, we matched the POS data with the table-level data documented by the RAs.

Data Description

During the study period, we collected data on 1,158 customers seated at 442 tables. The most popular beverage was beer, served in either a glass or a bottle. Either a bottle or a glass of beer counted as one beverage order. Patrons at 11 tables ordered wine or whiskey, and we excluded these data because it is difficult to equate such consumption quantities with beer. We also excluded nine tables with foreigners since the visual elements were printed in Korean, and three tables with missing data. Thus, the final sample included data from 1,074 customers at 418 tables. Each table had 2.56 customers on average. In Table 1, we provide detailed information about participants and tables.

Results

Hypotheses Testing

Our independent variable, social density, is a continuous variable; therefore, we performed a regression analysis with the following regression equation:

\begin{array}{l} Y_{i, t} = b_{0} + b_{1} T R E A T_{i, t} + a_{0} D E N S I T Y_{i, t} \\ + a_{1} T R E A T_{t} * D E N S I T Y_{i, t} + b_{2} X_{i, t} + e_{i, t} \end{array}

where Y_i,t is the outcome variable of interest, beer consumption per customer at of table i, in time t, and TREAT_i,t takes a value of 1 for a table with visual elements, and 0 otherwise. DENSITY_i,t is the social density measured when customers at table i ordered at time t.¹ X _i,t represents table-level control variables, including the approximate age of customers, the number of customers in the party, day of the week dummies, and table dummies. The day of the week dummies and table dummies are included to account for effects attributable to the day of the week and time-invariant characteristics of each table, such as ocean views;² e_i,t is the error term.

The coefficients of interest are b₁ and a₁. When a table is treated, that is, TREAT_i,t = 1, the causal effect of treatment is b₁ + a₁ * DENSITY_i,t. Therefore, if DENSITY_i,t = 0, the causal effect is just b₁. However, if DENSITY_i,t is 15, for example, the treatment effect is b₁ + a₁ * 15. This shows that the causal effect of treatment may vary depending on the value of DENSITY_i,t as long as a₁ is not 0. Furthermore, assuming b₁ > 0, the sign of a₁ either boosts or dampens the treatment effect, given the value of DENSITY_i,t. If a₁ > 0, a higher value of DENSITY_i,t boosts the treatment effect. On the other hand, if a₁ < 0, a higher value of DENSITY_i,t dampens the treatment effect.

We first verified that randomization of the treatment was achieved. As in any RCT, we did this by comparing the means of the various table-level characteristics between the treatment and control groups. Large differences in the means of table-level characteristics imply a failure of proper randomization. On the other hand, having balanced means for observable characteristics under RCT imply that the randomization was well-executed, and therefore, unobserved variables are expected to be balanced as well.

Table 2 shows means and standard deviations for table-level observed characteristics for the treatment and the control group (Column 1 and 2), as well as associated p values resulting from t-tests for mean differences for the two groups (Column 3). Most of the observed variables have mean differences that are not statistically significant, suggesting that randomization was successful in balancing observable characteristics of the treatment and the control group. For example, the share of male customers in the treatment group is 0.5, while the corresponding number is 0.48 in the control group (p = .70). All the other variables have high p values except for a dummy variable which equals 1 if the table is at terrace and has an ocean view (p value of .07). In the regression analysis, we controlled for a variety of table characteristics as explained in the next paragraph.

Table 2.

Mean Comparisons Between Treatment and Control Group.

Table-Level Characteristics	(1)	(2)	(3)
Table-Level Characteristics	Treatment GroupMean (SD)	Control GroupMean (SD)	Mean differencet-test p value
Share of male	0.50 (0.35)	0.48 (0.34)	.70
Number of customers per table	2.6 (1.3)	2.5 (1.5)	.48
Social density	16.5 (11.7)	15.5 (11.7)	.36
=1 if table is right next to the street (ocean view)	0.23 (0.03)	0.31 (0.03)	.07
=1 if table is bar table	0.05 (0.02)	0.04 (0.01)	.72
=1 if table is interior table	0.72 (0.03)	0.65 (0.03)	.11
=1 if approximate age of customers: 20s	0.54 (0.5)	0.47 (0.5)	.11
=1 if approximate age of customers: 30s	0.34 (0.47)	0.38 (0.49)	.38
=1 if approximate age of customers: 40s	0.03 (0.01)	0.01 (0.01)	.27
=1 if approximate age of customers: 50s	0.03 (0.17)	0.05 (0.21)	.37
Number of observations	203	215

Table 3 shows the regression results based on the estimation equation above using STATA 14.2. Column 1 shows the causal effect of visual elements (i.e., text messages on coasters) without any control variables. Column 2 adds table-level characteristics: number of customers, share of females, and approximate ages. Column 3 adds day of the week dummies to control for day of the week effects. For example, drinks per customer might be higher on Fridays. Column 4 adds table dummies to control for specific table effects. For example, some tables have a better view of the ocean and customers sitting at those tables may drink more, on average.³

Table 3.

Effects of Coaster on Drinks per Customer.

Variables	(1)	(2)	(3)	(4)
Treatment	.555***	.588***	.584***	.608***
Treatment	(0.164)	(0.163)	(0.158)	(0.174)
Density	−.00259	−.00207	−.00374	−.00104
Density	(0.00348)	(0.00348)	(0.00426)	(0.00536)
Treatment*density	−.0137**	−.0139**	−.0138**	−.0166**
Treatment*density	(0.00685)	(0.00676)	(0.00660)	(0.00814)
Constant	1.521***	1.325***	1.381***	1.404***
Constant	(0.0737)	(0.126)	(0.171)	(0.357)
Observations	418	418	418	418
R ²	.053	.082	.086	.153
Controls	N	Y	Y	Y
Week dummies	N	N	Y	Y
Table dummies	N	N	N	Y

Notes. Robust standard errors in parentheses. The dependent variable in all columns is beer consumption per customer.

p < .1. **p < .05. ***p < .01.

Since the randomization was properly executed, regression estimates should not change much, regardless of whether controls are included or not. The empirical results confirm this. As shown in Column 1, the causal effect of the coaster with the message is 0.56 additional drinks per customer if the social density value is 0. The estimate is also highly statistically significant with a p value of less than 0.01. In Columns 2, 3, and 4, estimates range from 0.58 to 0.61, quite similar to the estimates in Column 1.

These results provide support for H1, which states that the presence of consumption-inducing text messages on coasters will have a positive impact on patrons’ beverage consumption. On the other hand, the coefficient for density is not statistically significant, implying that the regression results do not lend support for H2, which states that the level of social density is positively correlated with beverage consumption. This might reflect the fact that the linear specification of density does not capture a potential concave relationship between density and the outcome (see Figure 1).

Figure 1.

The Relationship between Social Density and Beverage Consumption, with and without Visual Cues.

The regression results fail to support H3, which states that the positive effect of consumption-inducing text messages is maximized in socially dense environments. In contrast, the results show that the treatment effect is stronger when the bar is less crowded. Estimates for the joint effect of visual elements and density are significant across Columns 1 to 4. Since the sign is negative, an increase in social density dampens the causal effect of the coaster with the message. More precisely, based on the estimates from Column 4, the causal effect of the coaster with the message is an additional 0.601−(0.017*DENSITY) drinks per customer. Therefore, if the density is 10, the causal effect is 0.601−0.17 = 0.431 additional drinks per customer. These finds are consistent with Figure 1 visualizing that the causal effect decreases with social density.

Figure 1 shows the relationship between social density and drinks per customer for the treatment group (coaster with the message) and control group (coaster without the message) using locally weighted scatterplot smoothing.⁴ The line depicts the nonparametric relationship between social density and drinks per customer using locally weighted regressions. Without treatment (dashed line), the figure shows that an increase in social density up to 30 is associated with a slight increase in the number of drinks per customer. However, at around 30, a higher value of social density is associated a slight decrease in the number of drinks per customer. Thus, as opposed to H2, higher social density is associated negatively with the outcome, and not in a linear fashion. The solid line represents the relationship between these two variables when treated (i.e., a table is given coasters with the message).

Essentially, the vertical gap between the solid line and the dashed line can be interpreted as the causal effect of the treatment at any given social density. For example, at a social density value around 5, the solid line shows about two drinks per customer, and the dashed line indicates about 1.5 drinks. Therefore, the causal effect at that social density value is about 0.5 more drinks. As shown in Figure 1, the causal effect is strongest when social density is less than 15. Above 15, the causal impact becomes weaker. Figure 1 also provides graphical evidence that H3 is not supported. In contrast to H3, the effect of visual elements weakens as social density increases, especially when social density values exceed 30.

Robustness Checks for Regression Results

We performed a battery of robustness checks to rule out potential biases and threats to internal validity from uncontrolled factors.⁵ More precisely, we additionally controlled for variables and checked if our estimates do not drastically change (Table 4). As a benchmark, Column 1 of Table 4 displays the same estimates from the main regression results in Column 4 of Table 3.

Table 4.

Robustness Checks for the Main Results.

Variables	(1)	(2)	(3)	(4)	(5)
Treatment	.608***	.589***	.605***	.603***	.601***
Treatment	(0.174)	(0.173)	(0.159)	(0.174)	(0.230)
Density	−.00104	−.0001518	−.00366	−.00410	.00227
Density	(0.00536)	(0.00538)	(0.00427)	(0.00679)	(0.01565)
Treatment*density	−.0166**	−.0173**	−.0155**	−.0168**	−.0210**
Treatment*density	(0.00814)	(0.00825)	(0.00656)	(0.00803)	(0.01059)
Constant	1.404***	1.438***	1.387***	1.610***	1.609***
Constant	(0.357)	(0.362)	(0.187)	(0.509)	(0.704)
Observations	418	418	418	418	418
R ²	.153	.133	.093	.171	.464
Full controls	Y	N	Y	Y	Y
Week dummies	Y	Y	Y	Y	N
Table dummies	Y	Y	N	Y	Y
Full controls except age variable	N	Y	N	N	N
Bar and ocean view table dummies	N	N	Y	N	N
Order hour dummies	N	N	N	Y	N
Order date-hour dummies	N	N	N	N	Y

Notes. Robust standard errors in parentheses. The dependent variable in all columns is beer consumption per customer.

p < .1. **p < .05. ***p < .01.

Our regression results were robust and stable. First, we removed the average age of customers in a table from the regression. In our main regression, we used this variable as a control variable. Removing the age variable does not change the results much in terms of the key estimated coefficients and statistical significance (Column 2). The estimated coefficient for TREAT_i,t changes from .608 in the original coefficient to .589 when the age variable is dropped. Similarly, the estimated coefficient for TREAT_t*DENSITY_i,t does not change much, from −.0166 to −.0173. Second, to control for potential differences in beverage consumption depending on the location of tables, we controlled for the table location using three categories, interior tables, terrace tables, and bar counter (Column 3). Third, the hour of the beverage orders may matter. We included the hour of the order dummies as controls (Column 4). Fourth, outside and inside temperature may also influence beverage consumption. Consequently, we included the table location and order date and order hour as controls (Column 5). In all cases, the estimated coefficients remain robust and stable, suggesting that it is not likely that unobserved characteristics drastically influence our results.

In addition, we used an alternative measure of social density to check for robustness of the results. With this new measure, we took into account the bar’s total capacity by dividing the original density measure by the official maximum capacity, 92 people. Appendix B Table B1 replicates our main regression results shown in Table 3, replacing the original social density variable with this capacity-scaled density measure. The estimates were qualitatively similar to those found in Table 3, both in terms of the signs and statistical significance. The main coefficients of interest were still highly statically significant at the 5% level.

Discussion

Theoretical Implications

We investigated the impact of two Servicescape factors—visual elements and social density—on consumers’ beverage consumption in a bar. Previous research has examined the effectiveness of visual elements in the context of interior design cues such as color and lighting (e.g., Biswas et al., 2017; Lick et al., 2017; Rimkute et al., 2016). However, there is scant research on written message cues. Our findings indicate that coasters with a consumption-inducing text message had a positive impact on beer orders. Specifically, when coasters had a written message (vs. no message), the number of drinks consumed increased by approximately 0.6 drinks per customer. This finding demonstrates the positive impact of visual elements in influencing consumer behaviors, namely beverage orders. Short text messages used in this study might induced imagery-evoking (Babin et al., 1992; MacInnis & Price, 1987). The message, “One more drink!,” can be considered as an invitation to imagine ordering an additional beverage. Therefore, the findings of this study add to the literature on the influence of visual elements on consumers’ information processing strategies.

Higher social density, on the other hand, does not seem to increase patrons’ beverage consumption. Although high social density is linked to negative responses (e.g., stress and frustration) in retail environments (Eroglu, Machleit, & Chebat, 2005; Huang et al., 2018), the positive effect is also documented in hospitality settings (Hanks et al., 2017; Hyun & Kim, 2015). However, our study findings indicate that high social density does not lead to increased beverage consumption. When considered jointly, our results indicate that the effect of visual elements depends on the level of social density. Contrary to our predictions, the positive effect of the text message was stronger when social density was low. The notion of information processing can explain this unexpected finding. The message on the coaster contained information, thus prompting cognitive responses (Li et al., 2019; Wilson & Gilbert, 2008). People select and process information in the environment by paying attention to external stimuli (Hutchinson et al., 2016). However, people’s cognitive resources are limited, and therefore, attention to all multisensory cues might be limited (Kwan et al., 2017). This mental processing limitation can be explained by simultaneous information processing. Prior studies suggest that simultaneous information processing can improve or limit mental processing. Previous research shows that presenting two or more stimuli simultaneously can enhance mental processing (Chowdhury et al., 2008; Das et al., 1975). However, there is also research suggesting that stimuli presented in a sequential manner are easier to process than a simultaneous presentation of multiple cues (e.g., Hogarth & Einhorn, 1992). This might be explained by the notion of information overload. Similarly, sensory overload can mask important sensory cues (Krishna, 2012; Miller, 1956). In sum, in crowded environments, high sensory stimulation may reduce customers’ attention to cognitive cues such as visual images. Our findings demonstrate that not all combinations of marketing stimuli necessarily enhance the Experiencape. Consumers perceive marketing stimuli holistically when forming overall impressions of a brand or a company (Labenz et al., 2018; Lindstrom, 2005). Hence, it is critical to coordinate sensory stimuli in a way that they can reinforce the effect of one another, thus leading to optimal outcomes (Guzman & Iglesias, 2012).

Managerial Implications

The findings of this field experiment have important implications for practitioners. First, hospitality companies can use sensory stimuli as a differentiating and positioning strategy. In our field experiment, coasters with a consumption-encouraging text message had a positive effect on bar patrons’ beverage consumption. Hospitality managers need to understand the importance of sensory marketing. Sensory stimulations are powerful and immediate, so they can influence consumers’ attitude significantly (Gobe, 2001; D. Kim & Perdue, 2013). As sensory stimulations influence customers’ emotions and memory, they help companies to build long-term relationship with customers.

The cost to produce the coasters was very small compared to the increase in beverage sales. This clearly demonstrates the importance of allocating even minimal business resources to developing and implement marketing strategies that incorporate sensory cues. In addition, it is relatively easy to manipulate design and ambient cues such as colors, scents, and music rather than issues related to operations management such as employee relations (Demoulin, 2011). Therefore, practitioners should realize the value of sensory marketing which is cost-effective and flexible (Fisk et al., 2011).

Consumers are surrounded by various sensory inputs both in the real world and online. This study stresses the importance of coordinating environmental elements. Although stimulating senses can be critical to enhance customer experiences, businesses need to carefully choose environmental cues that produce positive interactive effects (Krishna, 2013). Only a few studies have revealed the interactive effects of two or more environmental stimuli, and our findings demonstrate such an effect with visual elements and social density. We found that the impact of the text message on beverage sales was stronger when there were fewer customers in the bar, and this is opposite to the proposed synergy effect between a visual element and social density. This finding indicates that hospitality managers need to understand environmental characteristics such as crowding to maximize the effectiveness of sensory stimuli.

In addition, we should consider the implications of our findings in light of the COVID-19 pandemic. The COVID-19 outbreak hit the hospitality industry severely, and various regulations have been applied to overcome the situation. Restaurants, bars, clubs, and many other hospitality businesses have been asked to enforce social distancing by closing the business or by reducing the operating hours and capacity (Bartik et al., 2020; Gursoy & Chi, 2020). Our findings indicate that short text messages as visual stimuli might be effective in increasing beverage sales in less-crowded environment. Therefore, this type of promotion may work well in the current social distancing climate. Taken together, our study findings emphasize the importance of applying effective sensory marketing strategies. Even more importantly, hospitality practitioners must understand how to select and coordinate diverse sensory cues to boost sales and to enhance the consumer’s experience.

Limitations and Future Research

Some limitations inherent in the design of this study present avenues for future research. Our real-world setting implies external validity and realism (Cook & Campbell, 1979) and thus is a major strength of the study. However, field studies tend to be plagued with noise. Although we controlled for possible effects of factors such as table location and day of the week, we cannot rule out the possibility that other factors might have affected our results. Moreover, the generalizability of our findings is limited by the fact that our study involved a single hospitality establishment. Also, because the visual text messages were written in Korean, we had to exclude nonnative speakers from our analyses. It would be interesting to replicate our visual element manipulation in a different language such as English to test for culture-related effects as language is a part of signs or symbols that yields different interpretations depending on ethnic groups (Rosenbaum, 2005). This study showed a negative aspect of social density. In other multisensory environments, positive effects of social density may exist. This is an interesting avenue for future research. In addition, it might be meaningful to extend this study to other sensory cues. Testing the joint effects of visual elements and other sensory cues such as scent or sound in a similar setting might yield interesting findings.

Footnotes

Appendix A

Appendix B

Table B1.

Effects of Coaster on Drinks per Customer Using the Social Density Ratio.

Variables	(1)	(2)	(3)	(4)
Treatment	.555***	.588***	.584***	.608***
Treatment	(0.164)	(0.163)	(0.158)	(0.174)
Social density ratio	−.238	−.191	−.344	−.096
Social density ratio	(0.320)	(0.320)	(0.392)	(0.492)
Treatment* social density ratio	−1.264**	−1.274**	−1.272**	−1.532**
Treatment* social density ratio	(0.630)	(0.622)	(0.607)	(0.749)
Constant	1.521***	1.325***	1.381***	1.404***
Constant	(0.0737)	(0.126)	(0.171)	(0.357)
Observations	418	418	418	418
R ²	.053	.086	.086	.153
Controls	N	Y	Y	Y
Week dummies	N	N	Y	Y
Table dummies	N	N	N	Y

Notes. Robust standard errors in parentheses. The dependent variable in all columns is beer consumption per customer.

p < .1. **p < .05. ***p < .01.

Acknowledgements

The authors thank Minwoo Kim and Jinhwan Seol for their excellent help on data collection.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, or publication of this article: This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (grant no. NRF-2017S1A5A8021353).

ORCID iD

Hyunjoo Yang

Notes

Author Biographies

Min Gyung Kim, PhD, is an assistant professor in the College of Business Management at Hongik University, Republic of Korea. Min holds a PhD in hospitality management at the Pennsylvania State University. Her research interests are consumer emotions, sensory marketing, communication, and branding for enhancing consumer experiences in the service setting.

Hyunjoo Yang, PhD, is an assistant professor in the School of Economics at Sogang University. He holds a PhD in economics from Brown University. His research interests are applied microeconomics and development economics.

Anna S. Mattila, PhD, is a Marriott Professor of lodging management at The Pennsylvania State University. She holds a PhD in services marketing from Cornell University. Her research interests focus on service encounters with a particular interest in healthy eating, service recovery, corporate social responsibility, social media, and cross-cultural research.

References

Ali

E. H. M.

Ahmed

M. O.

(2019). Sensory marketing and its effect on hotel market-share: Perception of hotel customers. Journal of Tourism and Hospitality Management, 7(1), 116–126.

Aprilianty

Purwanegara

M. S.

(2016). Effects of colour towards underwear choice based on electroencephalography (EEG). Australasian Marketing Journal, 24(4), 331–336.

Babin

B. J.

Attaway

J. S.

(2000). Atmospheric affect as a tool for creating value and gaining share of customer. Journal of Business Research, 49(2), 91–99.

Babin

L. A.

Burns

A. C.

Biswas

(1992). A framework providing direction for research on communications effects of mental imagery-evoking advertising strategies. Advances in Consumer Research, 19(1), 621–628.

Barisic

Blazevic

(2014). Visual identity components of tourist destination. International Journal of Social, Management, Economics and Business Engineering, 8(7), 2155–2159.

Bartik

A. W.

Bertrand

Cullen

Z. B.

Glaeser

E. L.

Luca

Stanton

C. T.

(2020). How are small businesses adjusting to COVID-19? Early evidence from a survey (No. w26989). National Bureau of Economic Research.

Biswas

Szocs

Chacko

Wansink

(2017). Shining light on atmospherics: How ambient light influences food choices. Journal of Marketing Research, 54(1), 111–123.

Bitner

M. J.

(1992). Servicescapes: The impact of physical surroundings on customers and employees. Journal of Marketing, 56(2), 57–71.

Brasel

S. A.

Gips

(2014). Tablets, touchscreens, and touchpads: How varying touch interfaces trigger psychological ownership and endowment. Journal of Consumer Psychology, 24(2), 226–233.

10.

Childers

T. L.

Houston

M. J.

(1984). Conditions for a picture-superiority effect on consumer memory. Journal of Consumer Research, 11(2), 643–654.

11.

Chowdhury

R. M.

Olsen

G. D.

Pracejus

J. W.

(2008). Affective responses to images in print advertising: Affect integration in a simultaneous presentation context. Journal of Advertising, 37(3), 7–18.

12.

Consiglio

De Angelis

Costabile

(2018). The effect of social density on word of mouth. Journal of Consumer Research, 45(3), 511–528.

13.

Cook

Campbell

(1979). Quasi-experimentation, design and analysis issues for field settings. Houghton Mifflin.

14.

Cornil

Chandon

Krishna

(2017). Does Red Bull give wings to vodka? Placebo effects of marketing labels on perceived intoxication and risky attitudes and behaviors. Journal of Consumer Psychology, 27(4), 456–465.

15.

Csikszentmihalyi

(2014). Play and intrinsic rewards. In Csikszentmihalyi

(Ed.), Flow and the foundations of positive psychology (pp. 135–153). Springer.

16.

D’Agostino

P. R.

O’Neill

B. J.

Paivio

(1977). Memory for pictures and words as a function of level of processing: Depth or dual coding? Memory & Cognition, 5(2), 252–256.

17.

Das

J. P.

Kirby

Jarman

R. F.

(1975). Simultaneous and successive synthesis: An alternative model for cognitive abilities. Psychological Bulletin, 82(1), 87–103.

18.

Demoulin

N. T.

(2011). Music congruency in a service setting: The mediating role of emotional and cognitive responses. Journal of Retailing and Consumer Services, 18(1), 10–18.

19.

Doucé

Janssens

(2013). The presence of a pleasant ambient scent in a fashion store: The moderating role of shopping motivation and affect intensity. Environment and Behaviour, 45(2), 215–238.

20.

Elder

R. S.

Krishna

(2010). The effects of advertising copy on sensory thoughts and perceived taste. Journal of Consumer Research, 36(5), 748–756.

21.

Eroglu

S. A.

Machleit

K. A.

(1990). An empirical study of retail crowding: Antecedents and consequences. Journal of Retailing, 66(2), 201–221.

22.

Eroglu

S. A.

Machleit

K. A.

Barr

T. F.

(2005). Perceived retail crowding and shopping satisfaction: The role of shopping values. Journal of Business Research, 58(8), 1146–1153.

23.

Eroglu

S. A.

Machleit

K. A.

Chebat

J. C.

(2005). The interaction of retail density and music tempo: Effects on shopper responses. Psychology & Marketing, 22(7), 577–589.

24.

Esmailpour

Zakipour

(2016). The sensory stimuli model; Engage with the consumer senses for brand distinguishes. Journal of Management Sciences, 2(4), 212–218.

25.

Fisk

R. P.

Rosenbaum

M. S.

Massiah

(2011). An expanded servicescape perspective. Journal of Service Management, 22(4), 471–490.

26.

Gobe

(2001). Emotional branding: The new paradigm for connecting brands to people. Allworth Press.

27.

Gursoy

Chi

C. G.

(2020). Effects of COVID-19 pandemic on hospitality industry: Review of the current situations and a research agenda. Journal of Hospitality Marketing & Management, 29(5), 527–529.

28.

Guzman

Iglesias

(2012). The multi-sensory and multi-experiential brand challenge. Journal of Product & Brand Management, 21(6), Article 09261.

29.

Hanks

Line

Kim

W. G. W.

(2017). The impact of the social servicescape, density, and restaurant type on perceptions of interpersonal service quality. International Journal of Hospitality Management, 61, 35–44.

30.

Harrell

G. D.

Hutt

M. D.

Anderson

J. C.

(1980). Path analysis of buyer behavior under conditions of crowing. Journal of Marketing Research, 17(1), 45–51.

31.

Hecht

Reiner

(2009). Sensory dominance in combinations of audio, visual and haptic stimuli. Experimental Brain Research, 193(2), 307–314.

32.

Hinestroza

N. B.

James

P. T. J.

(2014). The effects of sensory marketing on the implementation of fast-food marketing campaigns. Journal of Management & Marketing Research, 14, 1–11.

33.

Hock

S. J.

Bagchi

(2018). The impact of crowding on calorie consumption. Journal of Consumer Research, 44(5), 1123–1140.

34.

Hogarth

R. M.

Einhorn

H. J.

(1992). Order effects in belief updating: The belief-adjustment model. Cognitive Psychology, 24(1), 1–55.

35.

Huang

Wyer

R. S.

Jr. (2018). The influence of social crowding on brand Attachment. Journal of Consumer Research, 44(5), 1068–1084.

36.

Hui

M. K.

Bateson

J. E.

(1991). Perceived control and the effects of crowding and consumer choice on the service experience. Journal of Consumer Research, 18(2), 174–184.

37.

Hultén

(2011). Sensory marketing: The multi-sensory brand-experience concept. European Business Review, 23(3), 256–273.

38.

Hultén

(2015). Sensory marketing: Theoretical and empirical grounds. Routledge.

39.

Hutchinson

J. W.

Weingarten

(2016). Visual attention in consumer settings. Routledge International Handbook of Consumer Psychology, 61, 1–53.

40.

Hwang

Shin

Mattila

A. S.

(2018). So private, yet so public: The impact of spatial distance, other diners, and power on solo dining experiences. Journal of Business Research, 92, 36–47.

41.

Hyun

S. S.

Kim

M. G.

(2015). Negative effects of perceived crowding on perceived value and brand identification among luxury cruise passengers. Journal of Travel & Tourism Marketing, 32(3), 241–259.

42.

Jacobsen

J. K. S.

Iversen

N. M.

Hem

L. E.

(2019). Hotspot crowding and over-tourism: Antecedents of destination attractiveness. Annals of Tourism Research, 76, 53–66.

43.

Kim

Lee

C. K.

Sirgy

M. J.

(2016). Examining the differential impact of human crowding versus spatial crowding on visitor satisfaction at a festival. Journal of Travel & Tourism Marketing, 33(3), 293–312.

44.

Kim

Perdue

R. R.

(2013). The effects of cognitive, affective, and sensory attributes on hotel choice. International Journal of Hospitality Management, 35, 246–257.

45.

Kim

K. K.

JeKarl

Lee

J. H.

(2016). Drinking behaviors and policies to reduce harms caused by alcohol use and health promotion policy. Korean Journal of Health Education and Promotion, 33(4), 21–34. (In Korean)

46.

Kim

Lennon

(2008). The effects of visual and verbal information on attitudes and purchase intentions in internet shopping. Psychology & Marketing, 25(2), 146–178.

47.

Krishna

(2012). An integrative review of sensory marketing: Engaging the senses to affect perception, judgment and behavior. Journal of Consumer Psychology, 22(3), 332–351.

48.

Krishna

(2013). Customer sense: How the 5 senses influence buying behavior. Palgrave Macmillan.

49.

Krishna

Cian

Aydınoğlu

N. Z.

(2017). Sensory aspects of package design. Journal of Retailing, 93(1), 43–54.

50.

Krishna

Cian

Sokolova

(2016). The power of sensory marketing in advertising. Current Opinion in Psychology, 10, 142–147.

51.

Krishna

Schwarz

(2014). Sensory marketing, embodiment, and grounded cognition: A review and introduction. Journal of Consumer Psychology, 24(2), 159–168.

52.

Kwan

C. M.

Dai

Wyer

R. S.

Jr. (2017). Contextual influences on message persuasion: The effect of empty space. Journal of Consumer Research, 44(2), 448–464.

53.

Labenz

M. S. F.

Wiedmann

K. P.

Bettels

M. S. J.

Haase

M. S. J.

(2018). Sensory stimuli in print advertisement: Analyzing the effects on selected performance indicators. Marketing, 3(2), 7–15.

54.

Labrecque

L. I.

Milne

G. R.

(2012). Exciting red and competent blue: The importance of color in marketing. Journal of the Academy of Marketing Science, 40(5), 711–727.

55.

Labrecque

L. I.

Patrick

V. M.

Milne

G. R.

(2013). The marketers’ prismatic palette: A review of color research and future directions. Psychology & Marketing, 30(2), 187–202.

56.

Wang

C. R.

Meng

Zhang

(2019). Making restaurant reviews useful and/or enjoyable? The impacts of temporal, explanatory, and sensory cues. International Journal of Hospitality Management, 83, 257–265.

57.

Lick

König

Kpossa

M. R.

Buller

(2017). Sensory expectations generated by colours of red wine labels. Journal of Retailing and Consumer Services, 37, 146–158.

58.

Lindstrom

(2005). Broad sensory branding. Journal of Product & Brand Management, 14(2), 84–87.

59.

Line

N. D.

Hanks

Kim

W. G.

(2018). An expanded servicescape framework as the driver of place attachment and word of mouth. Journal of Hospitality & Tourism Research, 42(3), 476–499.

60.

Lowe

M. L.

Haws

K. L.

(2017). Sounds big: The effects of acoustic pitch on product perceptions. Journal of Marketing Research, 54(2), 331–346.

61.

Lutz

K. A.

Lutz

R. J.

(1978). Imagery-eliciting strategies: Review and implications of research. Advances in Consumer Research, 5(1), 611–620.

62.

Lwin

M. O.

Morrin

Krishna

(2010). Exploring the superadditive effects of scent and pictures on verbal recall: An extension of dual coding theory. Journal of Consumer Psychology, 20(3), 317–326.

63.

Machleit

K. A.

Eroglu

S. A.

Mantel

S. P.

(2000). Perceived retail crowding and shopping satisfaction: What modifies this relationship? Journal of Consumer Psychology, 9(1), 29–42.

64.

Machleit

K. A.

Kellaris

J. J.

Eroglu

(1994). Human versus spatial dimensions of crowding perceptions in retail environments: A note on their measurement and effect on shopper satisfaction. Marketing Letters, 5(2), 183–194.

65.

MacInnis

D. J.

Price

L. L.

(1987). The role of imagery in information processing: Review and extensions. Journal of Consumer Research, 13(4), 473–491.

66.

Madzharov

A. V.

Block

L. G.

Morrin

(2015). The cool scent of power: Effects of ambient scent on consumer preferences and choice behavior. Journal of Marketing, 79(1), 83–96.

67.

Maslow

A. H.

(1943). A theory of human motivation. Psychological Review, 50(4), 370–396.

68.

McMahon

C. E.

(1973). Images as motives and motivators: A historical perspective. The American Journal of Psychology, 86(3), 465–490.

69.

McNaughton

Corr

P. J.

(2004). A two-dimensional neuropsychology of defense: Fear/anxiety and defensive distance. Neuroscience and Biobehavioral Reviews, 28(3), 285–305.

70.

Michon

Chebat

J. C.

Turley

L. W.

(2005). Mall atmospherics: The interaction effects of the mall environment on shopping behavior. Journal of Business Research, 58(5), 576–583.

71.

Miller

G. A.

(1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63(2), 81–97.

72.

Moore

D. J.

(2014). Is anticipation delicious? Visceral factors as mediators of the effect of olfactory cues on purchase intentions. Journal of Business Research, 67(9), 2045–2051.

73.

O’Guinn

T. C.

Tanner

R. J.

Maeng

(2015). Turning to space: Social density, social class, and the value of things in stores. Journal of Consumer Research, 42(2), 196–213.

74.

Paivio

(1971). Imagery and verbal processes. Holt, Rinehart, & Winston.

75.

Pine

J. B.

Gilmore

J. H.

(1999). The experience economy: Work is a theatre and every business a stage. Harvard Business School Press.

76.

Pizam

Tasci

A. D.

(2019). Experienscape: Expanding the concept of servicescape with a multi-stakeholder and multi-disciplinary approach (invited paper for “luminaries” special issue of International Journal of Hospitality Management). International Journal of Hospitality Management, 76(B), 25–37.

77.

Pons

Laroche

(2007). Cross-cultural differences in crowd assessment. Journal of Business Research, 60(3), 269–276.

78.

Raghubir

(2010). Visual perception: Sensory marketing—Research on the sensuality of products. In Krishna

(Ed.), Sensory marketing: Research on the sensuality of products (pp. 201–215). Routledge.

79.

Rimkute

Moraes

Ferreira

(2016). The effects of scent on consumer behaviour. International Journal of Consumer Studies, 40(1), 24–34.

80.

Rosenbaum

M. S.

(2005). The symbolic servicescape: Your kind is welcomed here. Journal of Consumer Behaviour: An International Research Review, 4(4), 257–267.

81.

Ryu

Jang

S. S.

(2007). The effect of environmental perceptions on behavioral intentions through emotions: The case of upscale restaurants. Journal of Hospitality & Tourism Research, 31(1), 56–72.

82.

Shirai

(2017). Underdog effects: The role of consumption domain and retail crowding. Journal of Consumer Marketing, 34(5), 384–392.

83.

Spence

Puccinelli

N. M.

Grewal

Roggeveen

A. L.

(2014). Store atmospherics: A multisensory perspective. Psychology & Marketing, 31(7), 472–488.

84.

Stokols

(1972). On the distinction between density and crowding: Some implications for future research. Psychological Review, 79(3), 275–277.

85.

Streicher

M. C.

Estes

(2016). Multisensory interaction in product choice: Grasping a product affects choice of other seen products. Journal of Consumer Psychology, 26(4), 558–565.

86.

Tang

Fryxell

G. E.

Chow

C. S.

(2004). Visual and verbal communication in the design of eco-label for green consumer products. Journal of International Consumer Marketing, 16(4), 85–105.

87.

Tombs

McColl-Kennedy

J. R.

(2002, December). Beyond the servicescape: Customer to customer interactions in the social servicescape [Conference session]. Australian and New Zealand Marketing Academy Conference, Melbourne, Victoria, Australia.

88.

Wang

I. M.

Ackerman

J. M.

(2019). The infectiousness of crowds: Crowding experiences are amplified by pathogen threats. Personality and Social Psychology Bulletin, 45(1), 120–132.

89.

Wilson

T. D.

Gilbert

D. T.

(2008). Explaining away: A model of affective adaptation. Perspectives on Psychological Science, 3(5), 370–386.

90.

Yarnal

C. M.

Kerstetter

(2005). Casting off: An exploration of cruise ship space, group tour behavior, and social interaction. Journal of Travel Research, 43(4), 368–379.

91.

Zhu

Meyer

(2017). Getting in touch with your thinking style: How touchscreens influence purchase. Journal of Retailing and Consumer Services, 38, 51–58.