Sage Journals: Discover world-class research

Abstract

Ambient scents are being increasingly used in different service environments. While there is emerging research on the effects of scents, almost nothing is known about the long-term effects of consumers’ repeated exposure to ambient scents in a service environment as prior studies on ambient scents have been lab or field studies examining short-term effects of scent exposure only. Addressing this limitation, we examine the short- and long-term effects of ambient scents. Specifically, we present a conceptual framework for the short- and long-term effects of nonconsciously processed ambient scent in olfactory-rich servicescapes. We empirically test this framework with the help of two large-scale field experiments, conducted in collaboration with a major German railway company, in which consumers were exposed to a pleasant, nonconsciously processed scent. The first experiment demonstrates ambient scent’s positive short-term effects on consumers’ service perceptions. The second experiment—a longitudinal study conducted over a 4-month period—examines scent’s long-term effects on consumers’ reactions and demonstrates that the effects persist even when the scent has been removed from the servicescape.

Keywords

ambient scent servicescape long-term effects field experiment sensory marketing

The persuasive power of an odor cannot be fended off, it enters into us like breath into our lungs, it fills us up, imbues us totally. There is no remedy for it.

—Patrick Süskind, Perfume: The Story of a Murderer (p. 79) The use of ambient scents has become an increasingly distinctive trend in servicescapes (Henshaw et al. 2016; Krishna 2012). Singapore Airlines, for instance, has holistically integrated its signature scent, Stefan Floridian Waters, into its corporate communication concept as the ambient scent on its planes and its lounges, as its employees’ perfume, and as the cosmetic articles on all of its aircraft (Henshaw et al. 2016). Similarly, Abercrombie & Fitch has used its brand scent, Fierce, for several years, applying it extensively at its stores (Lindstrøm 2005). Overall, with an annual growth rate of 10% (Elejalde-Ruiz 2014) and a volume of over US$200 million in 2017 (Girard 2017), the market for ambient scents is growing rapidly. Nevertheless, the managerial goals of companies seeking to leverage ambient scents in their physical environments are very diverse, comprising aspects such as masking bad smells at a service delivery location, creating a pleasant atmosphere, enhancing a brand’s recall value, and advancing the consumers’ well-being or service experience. Their ultimate objective is, however, to increase their sales and customer loyalty (Goldkuhl and Styvén 2007). In this regard, Spence et al. (2014, p. 478) note that “understanding how and when specific scents enhance the consumer experience will be key for commercial success in the years to come.”

Prior research in this domain has revealed that pleasant ambient scents have a positive influence on consumers’ perceptions of, for example, the physical servicescape (Mattila and Wirtz 2001), on their brand evaluations (Morrin and Ratneshwar 2003), and time perceptions (Spangenberg, Crowley, and Henderson 1996). Furthermore, this literature also shows that ambient scents regulate consumers’ emotions, such as those related to pleasure, arousal, and mood (e.g., Mattila and Wirtz 2001; Spangenberg, Grohmann, and Sprott 2005), and even behaviors (e.g., Biswas and Szocs 2019; Herrmann et al. 2013; Hirsch 1995).

While past studies provide important insights into consumers’ perceptions of and reactions to ambient scents, almost nothing is known about the long-term effects of consumers’ repeated exposure to ambient scents in a service environment. To the best of our knowledge, only three studies, all outside marketing and in fields as diverse as geriatrics and developmental psychology, have examined the long-term effects of scent. These studies show that repeated scent exposure positively affects individuals’ arousal level (Delaunay-El Allam et al. 2010), inclination to agitation behavior (Moorman-Li et al. 2017), and their sense of balance (Sakamoto et al. 2012). Prior service and marketing research has, however, almost exclusively relied on static experiments focused on short-term effects by examining consumers’ reactions to a newly introduced olfactory stimulus. Specifically, research has not yet broached the issue of whether consumers’ responses to an ambient scent wear off over time or whether they persist even after the scent has been removed. In addition, prior research suggests that service atmospherics are particularly important for long-term customers (Dagger and Sweeney 2007). Thus, “[i]gnoring the time-variant character of servicescape effects may lead to inappropriate conclusions” (Brüggen, Foubert, and Gremler 2011, p. 72).

Also, research has not examined consumers’ reactions when an introduced ambient scent is later withdrawn or is absent. This is also critical since companies frequently implement ambient scents as a temporary promotional tool, for example, during the holiday seasons (e.g., Spangenberg, Grohmann, and Sprott 2005), or may discontinue its use for budget reasons (Spence et al. 2014). In addition, in a retail setting, ambient scents might, for example, be present in one section of the retail space but not in others (Hirsch 1995).

Finally, those studies that examined ambient scents’ short-term effects usually relied on highly controlled lab or other controllable environments, such as classrooms (see Rimkute, Moraes, and Ferreira 2016). As a consequence, there is need for research on ambient scent in olfactory-rich servicescapes characterized by diverse sensory cues (e.g., Canniford, Riach, and Hill 2018; Mattila and Wirtz 2001), such as during train journeys when it is difficult to avoid malodors and olfactory distraction (Canniford, Riach, and Hill 2018).

This article addresses these gaps in research and makes several important contributions to the literature. We first present a conceptual framework describing the short- and long-term effects of a pleasant, but nonconsciously processed, ambient scent on consumers’ service perceptions in a servicescape. We subsequently tested the framework’s central tenets by examining ambient scent’s short-term effects on consumers’ service perceptions in an olfactory-rich environment characterized by many sensory cues. For this purpose, we exposed the customers of a major German railway company to a pleasurable, nonconsciously processed ambient scent diffused via the train’s air conditioning system. The results of our first cross-sectional field experiment support ambient scents’ positive short-term effects on consumers’ service perceptions, even in an olfactory-rich servicescape.

To further test our conceptual framework, we measured the reactions of the same railway company’s customers to a pleasant ambient scent over a 4-month period in a second large-scale field experiment based on a longitudinal control group design. This study investigates the long-term effects of ambient scents on consumers’ service perceptions but also examines the possible aftereffects by measuring consumers’ reactions after the ambient scent was removed. The results show that the use of a nonconsciously processed long-term ambient scent has an enduring, positive impact on consumers’ service evaluations. Furthermore, our results indicate that ambient scents’ positive effect on service evaluations persists for at least 2 weeks after the ambient scent has been withdrawn. As such, this research has strong relevance for companies in the service and retail sectors, suggesting that nonconsciously processed ambient scent can be a powerful tool in servicescapes.

Theoretical Background and Hypotheses

A servicescape acts as a backdrop to consumers’ evolving emotions, cognitions, and behavioral intentions toward a service (Baker et al. 2002; Brüggen, Foubert, and Gremler 2011). Olfactory stimuli, an integral part of a service environment, influence consumers’ psychological responses such as by affecting approach or avoidance behavior (Bitner 1992; Krishna 2010; Spence et al. 2014). The olfactory system is unique in being the only sensory system relying solely on inputs of chemical molecules; in contrast, the visual system relies on light waves, while the auditory system uses sound waves. The olfactory bulb does the initial processing of olfactory molecules and transmits the information to the brain’s piriform cortex, which projects onto the orbitofrontal cortex via the amygdala (Rolls 2015; Soudry et al. 2011). The amygdala plays a critical role in emotions (Phelps and LeDoux 2005) and memory (Hamann et al. 1999). The presence of ambient scent has therefore been linked to positive experiences, mood, and memory (Krishna 2012; Krishna, Lwin, and Morrin 2010). In fact, the loss of smell often leads to mood disorders, depression, and an overall deterioration of the quality of life (Deems et al. 1991).

In essence, when chemical molecules from a pleasant ambient scent enter the human olfactory system, this triggers the brain’s reward circuitry area (Camerer, Loewenstein, and Prelec 2005), which in turn leads to an overall higher level of pleasant feelings (Biswas and Szocs 2019; Ressler 2004). Freiherr (2017) highlights that, in contrast to other sensory modalities, the thalamic relay does not process olfactory information on its way from the receptors to the neocortex, which might be one of the reasons for the multitude of nonconscious processes involved in olfactory perception. Using ambient scents to influence consumers has a special appeal for practitioners, particularly when processed nonconsciously, since consumers are more susceptible to the effects of scents, in such scenarios (Bradford and Desrochers 2009; Canniford, Riach, and Hill 2018).

The Effects of Nonconsciously Processed Ambient Scent on Consumers’ Service Perceptions

Building on extant literature, we theorize a pleasant ambient scent’s differing short- and long-term effects in an olfactory-rich servicescape on the basis of the exposure’s duration (Bradford and Desrochers 2009). Scents are usually processed nonconsciously (Li et al. 2007), either due to their subliminal nature (i.e., their low intensity, see Laird 1932) or because they match personal experiences and expectations, and are therefore integrated into an olfactory background (e.g., Biswas and Szocs 2019; Forster and Spence 2018; Krishna, Lwin, and Morrin 2010). Our theoretical framework therefore focuses on nonconsciously (rather than consciously) processed scents. Nonconscious processing entails that a scent influences cognition and behavior without consumers being aware of this influence (Holland, Hendriks, and Aarts 2005). This differs from consumers’ general ability to detect a scent stimulus when focusing on detection (Li et al. 2007). Subliminal scent stimuli, however, are processed nonconsciously only (Holland, Hendriks, and Aarts 2005).

Figure 1 illustrates our conceptual model. In the following sections, we first discuss pleasant, but nonconsciously processed, ambient scents’ initial effects after a one-time exposure (Stage I), followed by the effects of the scent after repeated exposure (Stage II), and, finally, by the aftereffects on consumers’ service perception following the discontinuation of the ambient scent stimulation (Stage III).

Figure 1.

Effects of nonconsciously processed ambient scents in a service environment.

One-time exposure effects

One positive effect of a pleasant ambient scent effect is its ability to mask other scents and especially unpleasant smells from food items, other passengers, and pets (Canniford, Riach, and Hill 2018; Henshaw et al. 2016), which can be quite prevalent in olfactory-rich servicescapes (e.g., in a train). Adding a pleasant ambient scent that masks malodors should therefore lead to favorable service perceptions (Goldkuhl and Styvén 2007).

Besides their ability to displace malodors, numerous studies demonstrate that pleasant scents have an additional positive effect on consumer perceptions when processed nonconsciously (e.g., Holland, Hendriks, and Aarts 2005; Laird 1932; Li et al. 2007). Affective priming, which affective primacy theory covers (Zajonc 1980, 1984), offers a potential explanation for this effect. According to this theory, a subliminal stimulus (i.e., a scent prime) evokes an emotional response, which is transferred to the processing of a situation and modifies its affective evaluation (Li et al. 2007). Consumers can therefore form positive affective reactions to a situation without a prior cognitive interpretation of the factors determining the positive affect (Winkielman, Zajonc, and Schwarz 1997).

Finally, research has shown that pleasant ambient scents can trigger the reward circuity areas of the brain, which in turn can lead to feelings of pleasantness (Biswas and Szocs 2019; Camerer, Loewenstein, and Prelec 2005). In other words, a pleasant ambient scent can lead to a more affective state for consumers.

Repeated exposure effects

What happens when consumers are repeatedly exposed to the same ambient scent? A constant masking of malodors and ongoing affective priming will certainly impact consumers’ service perceptions positively. The positive effects are less likely to diminish, as consumers process the scent nonconsciously and therefore do not become accustomed to the preferential scent treatment (Bradford and Desrochers 2009).

In fact, consumer psychology research suggests that an additional mechanism—associative learning (Biswas et al. 2014; Herz 2005)—reinforces consumers’ positive evaluation of a service episode in such a situation. By repeatedly being exposed to a pleasant ambient scent in the course of service delivery, consumers learn to associate the scent stimulus with the service itself (Canniford, Riach, and Hill 2018). Put differently, the scent stimulus’s positive valence spills over to the service environment. This is in line with research on evaluative conditioning (e.g., Hofmann et al. 2010; Houwer, Thomas, and Baeyens 2001), which suggests that the repeated pairing of a conditioned stimulus (i.e., a service delivery episode) with an unconditioned stimulus (i.e., a pleasant ambient scent) improves customers’ liking of the former over time (Hasford et al. 2018). Research has specifically shown that this effect also occurs if the unconditioned stimulus is processed nonconsciously (e.g., Baeyens et al. 1996; Jones, Fazio, and Olson 2009). Some studies have shown that evaluative conditioning effects are stronger if individuals are unaware of the unconditioned stimulus (Walther and Nagengast 2006). Numerous studies likewise underscore the importance of odor-induced associative learning and its effects on human perceptions and behaviors (e.g., Herz 2005; Reekum, de Berg, and Frijda 1999; Todrank et al. 1995). Therapeutic approaches, such as aromatherapy, rely on this mechanism (Herz 2009).

Aftereffects

Following a successful pairing of an unconditioned stimulus (i.e., the pleasant ambient scent) and a conditioned stimulus (i.e., a service delivery episode), evaluative conditioning theory posits that the mere presence of the conditioned stimulus elicits responses comparable to those observed in the presence of the unconditioned stimulus (Hofmann et al. 2010; Reekum, de Berg, and Frijda 1999). This postconditioning effect usually persists for a certain time until “extinction.” Extinction implies that experiencing the conditioned stimulus (the service delivery episode) without the unconditioned stimulus (the scent) leads to a gradual decline in the conditioned response (Houwer, Thomas, and Baeyens 2001). Thus, after having mentally disassociated the two stimuli, consumers are no longer expected to show positive affective reactions to the service environment. Prior research in the chemoreception field has shown that scents can induce postconditioning effects for a week or longer (Baeyens et al. 1996; Herz 2005), suggesting that service providers can benefit from prolonged aftereffects when discontinuing an ambient scent.

Furthermore, when consumers process an ambient scent nonconsciously, they will not recognize its discontinuation, thus making positive aftereffects possible. In contrast, when processing a scent consciously, consumers may recognize its sudden absence, which could potentially result in a devaluation of the service and the provider (Henshaw et al. 2016).

Hypotheses

To test the central tenets of our conceptual model, we evaluated whether the implementation of a pleasant, but nonconsciously processed, ambient scent leads to the proposed positive short-, long-term, and aftereffects on consumers’ evaluation of situational service perceptions, even in an olfactory-rich servicescape such as a train journey. A train journey is a suitable service setting for studying ambient scent effects because it allows for nonconscious scent evaluation and also because in most countries around the world, consumers frequent this type, or similar types, of servicescape regularly.

We examined the effects of scent on three dependent variables frequently used to assess consumers’ situational service evaluations: (1) their perceived service quality (i.e., “a consumer’s evaluative perceptions of a service encounter at a specific point in time”; Cronin and Taylor 1994, p. 127), which prior research on short-term scent effects has often considered (McDonnell 2007; Morrin and Chebat 2005), (2) consumers’ service experience (i.e., “the outcomes of the interactions between organizations, related systems/processes, service employees and consumers”; Bitner et al. 1997, p. 193), and (3) service value perceptions (i.e., “consumers’ overall assessments of product and service utility based on what is exchanged”; Harris and Goode 2004, p. 145). These constructs are important and relevant from conceptual and practical perspectives (e.g., Cronin, Brady, and Hult 2000; Hightower, Brady, and Baker 2002). See Table 1 for a complete list of item wordings.

Table 1.

Constructs, Measurement Items, and Quality Criteria (Study 1).

Construct	Item #	Questionnaire Item Wording (1 = fully disagree/7 = fully agree)	Factor Loading	KMO	Variance Explained	Cronbach’s α	Discriminant Validity via HTMT Criterion (Henseler, Ringle, and Sarstedt 2015)\|Bootstrapped Mean (Bias-Corrected 95% CIs)	Item-to- Total Correlation
Perceived service quality (Dabholkar, Shepherd, and Thorpe 2000)	1	The overall rail journey quality is excellent	.876	.834	.818	.925	Perceived service quality ↔ service experience: .852 [.178, −.986] Perceived service quality ↔ perceived service value: .598 [.022, −.913]	.782
	2	The rail journey is of a very high quality	.946					.894
	3	The rail journey has a high-quality standard	.915					.842
	4	The rail journey is superior in every way	.879					.787
Service experience (Brady and Cronin 2001)	1	When I leave the train today, I will feel that I had a good experience	.858	.706	.723	.809	Service experience ↔ perceived service quality: .852 [.178, −.986] Service experience ↔ perceived service value: .329 [.015, −.798]	.668
	2	I believe the railway company tries to give me a good experience today	.871					.692
	3	I believe the railway company knows the type of experience its customers want	.823					.615
Perceived service value (Harris and Goode 2004)	1	The services of the railway company are excellent value for money	.856	.779	.701	.854	Perceived service value ↔ perceived service quality: .598 [.022, −.913] Perceived service value ↔ service experience: .329 [.015, −.798]	.737
	2	The services of the railway company are excellent value	.683					.514
	3	I am happy with the value for money the railway company offers	.899					.793
	4	The services I purchase from the railway company are worth every penny	.884					.759

Note. KMO = Kaiser-Meyer-Olkin test for sampling adequacy; HTMT = Heterotrait-monotrait ratio of correlations.

In line with prior research (e.g., Lunardo and Mbengue 2013; Mattila and Wirtz 2001; Morrin and Chebat 2005), we expect a pleasant ambient scent to influence all three types of service perceptions after onetime exposure. Consistent with our theoretical framework, we expect the effects of repeated exposure to a nonconsciously processed scent to have continued positive effects. Finally, as discussed earlier, an ambient scent’s positive effects will persist, as consumers do not recognize the discontinuation of a nonconsciously processed scent. Formally stated:

Hypothesis 1_a–c: The presence (vs. absence) of a pleasant, nonconsciously processed, ambient scent in a service environment will lead to consumers perceiving higher levels of (a) service quality, (b) service experience, and (c) service value.

Hypothesis 2_a–c: The positive short-term impact of a pleasant, nonconsciously processed, ambient scent on consumers’ (a) service quality, (b) service experience, and (c) service value perceptions will continue with prolonged exposure.

Hypothesis 3_a–c: The positive impact of a pleasant, nonconsciously processed, ambient scent on consumers’ (a) service quality, (b) service experience, and (c) service value perceptions will persist after its removal.

Figure 2 provides an overview of our hypotheses.

Figure 2.

Overview of hypotheses.

Overview of Studies

First, we conducted a series of pretests, including one in a field setting with 330 actual consumers, to select an appropriate scent that consumers would predominantly process nonconsciously. We then ran two field experiments to test our hypotheses. The first experiment was a cross-sectional field study (204 actual consumers) to assess the scent’s short-term effects on consumers’ affective responses (Hypothesis 1). This study demonstrates ambient scent’s effectiveness in the specific service context of train journeys, an olfactory-rich environment. The second experiment was a longitudinal field study with 100 regular commuters; this study assessed ambient scent’s long-term and aftereffects (Hypotheses 2 and 3). Another 74 consumers were used in unscented conditions to bolster the internal validity of the longitudinal study’s results. We conducted all the experiments in cooperation with a major German railway company. Table 2 offers an overview of the pretests and field studies.

Table 2.

Study Overview.

Study	Environment and Sample	Design	Goals/Hypotheses Tested
Pretest 1	Sensory lab, n = 68 customers of the service provider (43% females, mean age: 36 years, n_{first scent} = 35, n_{second scent} = 33)	Two between-subjects conditions (scent #1 vs. scent #2)	Explorative: Identifying an appropriate ambient scent stimulus
Pretest 2	Field study, n = 330 customers on their train journey (56% females, between 15 and 77 years old)	Four between-subjects conditions (no scent vs. scent intensity via four, six, or eight cartridges per coach)	Explorative: Select optimal intensity just below the threshold that differentiates between a collective conscious versus nonconscious scent processing (50% recognition) Results: All three scent conditions below threshold but highest intensity leads to best perceived air quality
Pretest 3^a	Field study, n = 64 customers on their train journey (29% females, between 15 and 77 years old)	Two between-subjects conditions (scent intensity: 6 vs. 13 cartridges per coach)	Explorative: Evaluate how intense a scent has to be in order to induce conscious processing Results: 13 Cartridges lead to conscious processing
Study 1	Field study, n = 204 customers on their train journey (51% females, 88% nonsmokers, aged between 16 and 79 years, with a mean age of 35.4 years)	Two between-subjects conditions (no scent vs. scent with eight cartridges per coach)	Hypothesis 1a–c (ambient scents’ positive short-term effects) between-subjects: All hypotheses supported
Study 2	Field study, n = 174 regular commuters on their train journey (net sample n_scent = 35: 49% females, 86% nonsmokers, aged between 16 and 61 years, with a mean age of 34.6 years)	Nine within-subjects measurement points over a 4-month period (including one pre- and one postscent exposure), eight cartridges per coach	Hypothesis 1a–c (positive short-term effects) within subjects: All hypotheses supported Hypothesis 2a–c (positive long-term effects) within subjects: All hypotheses supported Hypothesis 3a–c (positive after effects after scent removal) within subjects: All hypotheses supported, but less pronounced aftereffects for service quality (Hypothesis 3a)

^aOnly reported in Note 2.

Pretest 1: Finding an appropriate Scent Stimulus

A professional fragrance manufacturer developed two scents that fit the railway company’s servicescape. According to optimal stimulation theory, consumers aim for an optimal target-arousal level when entering a certain service environment (Das and Hagtvedt 2016; Wirtz, Mattila, and Tan 2000). Since rail commuters are likely to seek a relaxing service experience (i.e., low arousal, as confirmed through interviews with customers), the railway company sought to apply a pleasurable but moderately arousing, ambient scent to avoid creating a strong discrepancy between the consumers’ experience of the servicescape and their target-arousal level (Mattila and Wirtz 2001). For the pretest, the first scent that the fragrance manufacturer developed was mainly comprised of jasmine, melon, and violet leave essences, while the second scent was mainly comprised of lemon, grapefruit, green tea, and sandalwood essences.

We conducted a lab-based pretest with 68 customers, whom the railway company helped recruit. All the participants received a gift worth approximately €5. The test was based on a between-subjects design, with participants evaluating the first or the second scent. The two between-subjects groups did not differ significantly in terms of age or gender (smallest p value = .543). On a bipolar scale (−3 = unpleasant, +3 = pleasant), the consumers liked both scents equally, M_scent1 = 1.11 versus M_scent2 = 0.73, t(66) = 1.051, p = .297, but the first scent yielded significantly lower levels of arousal, −3 = relaxing, +3 = stimulating, M_scent1 = −0.26 versus M_scent2 = 0.52, t(66) = 2.105, p = .039. The first scent comprised of a great number of essential ingredients, which means that its production process is elaborate. These two aspects act as a safeguard against other service providers copying the scent, as it can be difficult to register and protect a signature scent (Schifferstein and Spence 2008). We consequently chose the first scent for further assessment in the field studies.

Pretest 2: Optimal Intensity and Perception in a Field Setting

To determine the scent’s optimal intensity level, we conducted an additional field pretest with a total of 330 consumers on their train journey, during which we diffused the scent via the train’s air conditioning system.¹ All the participants received gifts worth €5 for their participation. This experiment comprised four between-subjects conditions, one without scent application (two coaches, n = 132, 54% female, between 16 and 77 years old) and three with different scent intensities (each in a separate coach, n_low = 86, n_medium = 50, n_high = 62, 57% female, between 15 and 64 years old), depending on the number of scent cartridges used per coach (i.e., four, six, or eight cartridges).

In the scented conditions, we first asked the consumers about their unaided recognition of any special scent and then asked about their aided recognition. Olfactory research has found that thresholds for scent recognition vary inter- and intraindividually (e.g., Biswas and Szocs 2019; Cain and Gent 1991). Researchers, therefore, routinely use 50% of the subjects who can recognize a scent as the threshold that differentiates between a collective conscious versus nonconscious scent processing (e.g., Burdack-Freitag, Heinlein, and Mayer 2017; Doty 2001; Marin, Acree, and Barnard 1988). In our study, 27% of the subjects recognized the scent (unaided and aided combined), which corresponds exactly to the rate observed in related research (Krishna, Lwin, and Morrin 2010), clearly indicating that the majority of consumers processed the scent nonconsciously.² Furthermore, we found no significant relationship between scent recognition and scent intensity, low 29%, medium = 30%, and high = 21%, χ²(2) = 1.563, p = .458. All 53 consumers who recognized the scent rated its perceived pleasantness (5 items, 7-point bipolar scales ranging from −3 to +3, Cronbach’s α = .95, Fisher 1974). With a significant positive deviation from the scale midpoint, M = 0.75, t(39) = 3.348, p = .002, η² = .223, the results confirm that the scent stimulus has a positive valence.

Finally, we tested whether the nonconsciously processed scent evokes positive evaluations of the servicescape. For this purpose, we allowed all 330 consumers to rate the perceived air quality (a 7-point bipolar single item, −3 = very bad/+3 = very good; Zhang, Arens, and Pasut 2011). A one-way analysis of variance (ANOVA) indicated an improved air quality perception in all three scent-intensity levels (M_low = 0.72, SD_low = 1.44; M_medium = .84, SD_medium = 1.38; M_high = 1.45, SD_high = 1.30), as compared to the control group without scent, M_control = .19, SD_control = 0.135; F(3, 326) = 12.422; p < .001; η² = .103. Pairwise comparisons (Hochberg’s procedure) showed significant differences between the no scent condition and all three scent intensity levels (p_control/low = .034; p_{control/medium} = .027; p_control/high = .000), as well as between the low and high intensities (p = .010), indicating that the unconscious manipulation of our olfactory cues was successful. Consequently, we applied an intensity of eight cartridges in the two subsequent field studies (Online Appendix A provides further details).

Study 1: Cross-Sectional Field Experiment

Study Design

Study 1 was a field experiment with consumers traveling between two midsize towns in Germany. The study had a between-subjects design with two manipulated conditions (scented vs. unscented coaches). We selected different train coaches randomly to serve as the experimental conditions. In the first coach, we diffused the scent via the train’s air conditioning system (scented condition), while consumers in two other coaches were not exposed to the scent (unscented condition).

Sample and Sampling Procedures

Ten minutes after the train had left the initial station, our assistants went through the two coaches and asked the consumers to participate in a market research study. We did not alert the respondents to the ambient scent diffusion during the study. All the participants received gifts worth €5 as an incentive.

The study sample consisted of 204 consumers. The respondents in the scented condition (n = 65) and the unscented condition (n = 139) did not differ significantly in terms of age, M_unscented = 36.0 versus M_unscented = 34.1 years, t(179) = 0.721, p > .10; smoking habits, smokers_unscented = 21.8% versus smokers_scented = 22.4%, χ²(1) = 0.009, p > .10); education, χ²(4) = 6.380, p > .10; occupation, χ²(4) = 0.930, p > .10; and train usage, commuter versus noncommuter, commuter_unscented = 48.9% versus commuter_scented = 38.6%, χ²(1) = 1.697, p > .10. However, the scented condition contained significantly more male participants (64.4%) than the unscented condition, 45.5%, χ²(1) = 5.562, p = .015.

Results

We first analyzed the construct measures’ reliability and validity. The results (reported in Table 1) support all the measures’ adequacy. Next, we ran a series of ANOVAs to compare the differences in the consumer perceptions of the scented and unscented conditions. The results show that consumers in the scented condition had significantly more favorable perceptions of the service quality, M_scented = 4.24, SD = 1.40 versus M_unscented = 3.77, SD = 1.27, F(1, 195) = 5.533, p = .020, η² = .028; service experience, M_scented = 3.82, SD = 1.43 versus M_unscented = 3.19, SD = 1.41, F(1, 199) = 8.687, p < .001, η² = .042; and service value, M_scented = 3.69, SD = 1.37 versus M_unscented = 3.05, SD = 1.38, F(1, 195) = 14.521, p < .001, η² = .069, than those in the unscented condition, with small to medium effect sizes (η²).³ In line with Hypothesis 1_a–c, these results support ambient scent’s positive short-term effect on consumers’ service perceptions, even in an olfactory-rich servicescape.

Study 2: Longitudinal Field Experiment

Study Procedure and Design

Study 2 examines ambient scent’s long-term effects and aftereffects when the scent is removed (Hypotheses 2_a–c and 3_a–c). This study was a field experiment conducted over a 4-month period with the same scent stimulus and a panel of consumers from the German railway company. The study had a scented condition (n = 100), an unscented control panel (n = 25), and two further unscented samples of additional consumers (total n = 49) to control for panel conditioning (see Figure 3).

Figure 3.

Experimental design of Study 2. The comparison of the experimental panel and the control panel, and subsequently, the control panel and the two control groups allow for an evaluation of mere measurement effects (i.e., adverse effects that arise from repeatedly interviewing the same customers; Dholakia and Morwitz 2002).

On a different route section that used in Study 1 (both directions), all operating trains were equipped with eight cartridges that diffused the ambient scent into all the passenger compartments. We applied a pretest/posttest control group design with a total of nine waves spread over a 4-month period. Wave 1 was run without ambient scent, thus serving as the baseline measurement. Waves 2–8 involved the use of an ambient scent, and we ran the final Wave 9, again without scent diffusion.⁴ In each of the waves, we gathered data on the same constructs as in Study 1. Table B1 in Online Appendix B offers an overview of the reliability and validity statistics, which support all the construct measures’ adequacy.

In line with Hypotheses 2_a–c and 3_a–c, we define long-term effects as changes observed in the dependent variables after repeated exposure in Waves 3–8 (Figure 2) and define aftereffects as any observed changes after the removal of the olfactory stimulus (Wave 9). Wave 9 was scheduled 2 weeks after the scent removal. While research on human chemoreception shows that aftereffects persist for at least 1 week (Baeyens et al. 1996; Herz 2005), 2 weeks offer greater scope for the effects’ extinction (e.g., Krishna, Elder, and Caldara 2010). To reiterate, Study 2’s focus is on long-term effects and aftereffects; nevertheless, analyzing the changes in the service perceptions between Waves 1 and 2 allows the short-term effects (Hypothesis 1_a–c) to be further tested in a more conservative, within-subjects design.

Sample and Sampling Procedures

The main sample (n = 100, Figure 3) comprised actual commuters on a specific rail route section between two midsize towns in Germany with a daily (only working days) one-way commute of at least 15 minutes in order to ensure sufficient exposure to the scent and to also give the participants enough time to answer the survey questions properly. We either recruited our respondents directly on the trains on the specific track section or contacted subscribed, regular customers via mail. None of the respondents participated in the pretests or in Study 1. Again, we did not alert the participants to the ambient scent diffusion during the study. After the completion of the final wave, all the participants received an explanation of the study’s goals and the role of the scent stimulus by mail. In Wave 2, we controlled for the participants’ awareness of the study background by means of a probing question. None of the participants could guess the study purpose or identify the scent stimulus, which further supports our contention that the consumers were unaware of the scent manipulation. Upon completion of the entire study, the participants received travel vouchers and giveaways worth €150. To ensure flexibility, we offered each participant the option of receiving the questionnaire via mail, e-mail, or from assistants on all the trains running on the specific track section every Wednesday between 5 a.m. and 11 a.m. Regardless of how the questionnaire was distributed, all the participants had to complete the survey on the train during their Wednesday morning commute.

Of the 100 respondents who participated in our baseline measurement (Wave 1), 35 participated in all nine study waves (i.e., 315 data points), representing an average panel mortality of 12% per wave, which is relatively low for a consumer panel (Tortora 2009). An a posteriori power analysis reveals that our sample size was sufficiently large to detect even small effect sizes (η² = 0.036) with a power of at least .95 at α = .05 (Faul et al. 2007).⁵ The respondents in the base and final data sets do not differ significantly with respect to gender, χ²(1) = 0.204, p > .10; educational degrees, χ²(5) = 10.342, p > .05; smoking habits, χ²(1) = 0.454, p > .10; direction of commute, χ²(1) = 0.089, p > .10; and type of questionnaire receipt, χ²(1) = 0.572, p > .10. However, the respondents in the final data set are significantly older, t(133) = −2.207, p = .029, and more often employed, rather than being students/pupils/apprentices, χ²(4) = 11.030, p = .017, than those in the base data set.

Of the 35 participants who participated in all the waves, 49% were women, 86% nonsmokers, and their ages ranged from 16 to 61 years (mean = 34.6). There were no significant differences in the survey mode used (13 on the train, 11 by mail, and 11 by e-mail). Most of the participants commuted between 6 a.m. and 9 a.m.; 16 respondents always used the same train, and 13 participants had a maximum variation in their departure time of ±1 hour. Only 6 of 315 data points (1.9%) fall clearly outside the range of usual traveling times. While these results show that the variation in travel times is relatively small, further analyses confirm that the within-subject variation is not an issue, as we did not find a systematic covariation between the study wave number and the travel times. Finally, with a single exception (an 18-minute delay, equaling four data points), none of the trains on any of the survey days arrived at their final stop with a delay of more than 10 minutes.

Results

Preliminary analysis of panel conditioning

By comparing the experimental and the control panels, we could rule out possible adverse effects arising from repeatedly interviewing the same consumers (i.e., mere measurement effects, Dholakia and Morwitz 2002). First, we evaluated if repeated observations of the control panel not experiencing scent led to an improvement of their situational service perceptions (Friedman’s ANOVA by rank). While this did not apply to the service value (p = .283), the respondents’ assessment of the service quality, χ²(2) = 7.586, p = .023, and the service experience, χ²(2) = 9.477, p = .009, increased significantly over time. More precisely, we observed an increase in the service quality assessment (p = .059, η² = .109) between Waves 1 and 2 and in the service experience assessment between Waves 1 and 3 (p = .017, η² = .152). Consequently, we proceeded with another in-depth analysis involving two additional control groups in the same trains in Waves 2 (n = 39) and 3 (n = 10; Figure 3). The results of a Mann-Whitney U tests show that the control panel and the control groups did not deviate significantly regarding the service quality (Wave 2: U = 416.50, p = .328, η² = .015) or the service experience (Wave 3: U = 98.50, p = .339, η² = .027). We conclude that our results are not biased by panel conditioning effects due to mere measurement effects (Online Appendix C provides more details).

Short-term effects

We used separate repeated-measures ANOVAs for each dependent variable, applying Greenhouse-Geisser corrections. Table 3 presents detailed descriptive statistics of each construct across all the study waves. Figure 4 summarizes the development of situational service perception evaluation over time.

Table 3.

Construct Values per Wave in Study 2.

Construct	Measurements (Ranging From 1 to 7)	Wave 1	Wave 2	Wave 3	Wave 4	Wave 5	Wave 6	Wave 7	Wave 8	Wave 9
Perceived service quality (Dabholkar, Shepherd, and Thorpe 2000)	N	35	35	35	35	35	35	35	35	35
	Mean	3.80	4.19	4.31	4.29	4.29	4.23	4.30	4.39	4.14
	Standard deviation	1.25	1.41	1.20	1.18	1.36	1.42	1.37	1.43	1.48
	Minimum	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
	Maximum	6.50	6.75	6.50	6.00	6.25	6.50	6.25	6.50	6.50
Service experience (Brady and Cronin 2001)	N	35	35	35	35	35	35	35	35	35
	Mean	2.84	3.41	3.75	3.71	3.80	3.73	3.80	3.84	3.75
	Standard deviation	1.05	1.47	1.38	1.33	1.53	1.52	1.56	1.57	1.52
	Minimum	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
	Maximum	5.00	6.67	6.00	6.00	6.00	6.00	6.00	7.00	6.00
Perceived service value (Harris and Goode 2004)	N	35	35	35	35	34	35	35	35	35
	Mean	3.38	3.69	3.83	3.71	3.85	3.80	3.79	3.89	3.94
	Standard deviation	1.04	1.20	1.22	1.10	1.21	1.27	1.24	1.23	1.35
	Minimum	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
	Maximum	5.50	5.75	6.00	5.50	5.75	5.75	5.75	6.00	6.00

Figure 4.

Temporal structure of scent effects as deviation from baseline (Study 2).

To isolate ambient scent’s short-term effects, we compared Wave 2 with the baseline measurement without scent (Wave 1), using follow-up tests with planned comparisons as simple contrasts. The results in Table 4 (Panel A) support the positive short-term effects of all three service perception constructs (Hypothesis 1_a–c).

Table 4.

Ambient Scent Effects (Study 2).

				Perceived Service Quality (Dabholkar, Shepherd, and Thorpe 2000)	Service Experience (Brady and Cronin 2001)	Perceived Service Value (Harris and Goode 2004)
			N	35	35	34
Ambient scent’s overall within-subjects effect	Mauchly’s test of sphericity		χ²	84.139	66.340	71.590
			df	35	35	35
			p	.000	.001	.000
	Greenhouse-Geisser ε			.576	.621	.644
	Test of overall within-subjects effects (rANOVA)		F	2.154	7.042	3.878
			df_M	4.610	4.971	5.148
			df_R	156.738	169.022	169.895
			p	.067	.000	.002
			η²	.060	.172	.105
Panel A short-term effect	Tests of within- contrasts (simple contrast Wave 1)	Wave 2 versus Wave 1	F	3.131	6.442	5.446
			p	.086	.016	.026
			η²	.084	.159	.142
Panel B Long-term effect	Tests of within- contrasts (simple contrast Wave 1)	Wave 3 versus Wave 1	F	4.155	17.817	12.138
			p	.049	.000	.001
			η²	.109	.344	.269
		Wave 4 versus Wave 1	F	4.327	12.894	6.855
			p	.045	.001	.013
			η²	.113	.275	.172
		Wave 5 versus Wave 1	F	4.336	15.580	14.159
			p	.045	.000	.001
			η²	.113	.314	.300
		Wave 6 versus Wave 1	F	3.550	13.826	9.035
			p	.068	.001	.005
			η²	.095	.289	.215
		Wave 7 versus Wave 1	F	4.551	19.685	10.649
			p	.040	.000	.003
			η²	.118	.367	.244
		Wave 8 versus Wave 1	F	6.625	18.331	12.827
			p	.015	.000	.001
			η²	.163	.350	.280
Panel C Aftereffect	Tests of within-contrasts to last scented wave (simple contrast Wave 8)	Wave 9 versus Wave 8	F	3.381	.333	.088
			p	.075	.567	.768
			η²	.090	.010	.003
	Tests of within-contrasts to baseline (simple contrast Wave 1)	Wave 9 versus Wave 1	F	2.107	13.849	13.334
			p	.156	.001	.001
			η²	.058	.289	.282

Note. df_M =model degrees of freedom; df_E = error degrees of freedom; ANOVA = analysis of variance.

The increases of service experience, F(1, 34) = 6.442, p = .016, η² = .159, and service value, F(1, 33) = 5.446, p = .026, η² = .142, are significant, with medium to large effect sizes. We found that the scent had a marginally significant effect on the perceived service quality, F(1, 34) = 3.131, p = .086, η² = .084. We also applied a single-paper meta-analysis (McShane and Böckenholt 2017), a method that facilitates the summary of multiple studies in a research paper by providing a significance test of the aggregated effects across the studies. The results confirm that the scent’s cumulative short-term effect on the service quality was significant (z = 2.600, p = .005), which also applies to its cumulative effects on the service experience (z = 3.452, p < .001) and the service value perceptions (z = 2.702, p = .003).

Long-term effects

Supporting Hypothesis 2_a–c, the results in Table 4 (Panel B) demonstrate that ambient scent improves consumers’ service experience significantly over time and that this effect remains stable at a high level, with a large effect size (.159 < η² < .367). Similarly, with respect to service value, we observed a positive and significant effect, medium to large in size (.142 < η² < .300), across all the waves. Finally, we observed the scent’s enduring positive effect on the service quality, which unfolds with a time lag of three waves. Specifically, the consumers’ service quality assessments improved significantly, with a medium effect size (.084 < η² < .163), from Wave 3 onward.

Aftereffects

We compared the results between Waves 8 and 9 by using repeated measure ANOVAs and a simple contrast (Table 4, Panel C). We find that, even in comparison with the last scented wave, none of service perceptions’ three facets decreased significantly 2 weeks after the scent had been removed: service quality, F(1, 34) = 3.381, p = .075, η² = .090; service experience, F(1, 34) = 0.333, p = .567, η² = .010; and perceived service value, F(1, 34) = 0.259, p = .614, η² = .008. In sum, these results indicate that ambient scent continues to have a positive aftereffect, supporting Hypothesis 3_a–c. For the service quality construct, however, Figure 4 presents a pronounced visual drop. A comparison of Wave 9 with Wave 1 (baseline) regarding this construct indicates that extinction works faster, F(1, 34) = 2.107, p = .156, η² = .058, meaning that the positive aftereffect lies between the Wave 8 level and the baseline level.

Discussion and Implications

Main Findings and Theoretical Implications

This research provides a framework for predicting the effects of nonconsciously processed ambient scents on consumers’ service perceptions. Importantly, this research focuses on ambient scents’ long-term effects, which is a novel focus since prior research on effects of ambient scent have been static in nature and only examined short-term effects of ambient scent. Our proposed model focuses on olfactory-rich service environments (such as train journeys), where ambient scents compete with other olfactory sensations that multiple sources emit. We argue that a nonconsciously processed ambient scent implemented over a longer period can improve consumers’ situational service perceptions via affective priming effects. Furthermore, since consumers process the scent nonconsciously, they do not recognize a scent campaign’s discontinuation. As such, our conceptual framework suggests that service providers should install nonconsciously processed ambient scents and pursue an enhanced situational evaluation through a long-term scent campaign.

The results of two field pretests and two field studies support the model’s central tenets. These studies were realistic and were conducted in the field rather than in lab settings with contrived scenarios. That is, we exposed actual consumers of a major German railway company to ambient scents that a professional fragrance manufacturer prepared. The scents were diffused through the trains’ air conditioning and ventilation system.

A field pretest confirms (1) that the scent stimulus used in the main studies was processed at predominantly nonconscious levels and (2) that its application led to higher levels of perceived air quality. The first field experiment (Study 1) demonstrates the positive effects of nonconsciously processed ambient scents on consumers’ perceived service quality, service experience, and perceived service value after a one-time exposure, even in an olfactory-rich environment. The second field experiment provides a longitudinal assessment of ambient scents’ effects over 4 months, providing insights into these scents’ impact on service perceptions over a longer period. Our results demonstrate nonconsciously processed ambient scents’ long-lasting positive effects on consumers’ service perceptions. In line with an associative learning account, these effects persist for a while, even after the ambient scent has been removed.

Our results clearly underline ambient scents’ contribution to consumers’ perceived service experience every time they reuse the service (Verhoef et al. 2009). Zemke and Shoemaker (2007, p. 937) conclude that “a proprietary scent can ‘tangibilize’ a company’s service.” Extending this notion, environmental cues—such as ambient scents—can make a significant and lasting contribution to consumers’ service value perception. Our findings are particularly relevant for servicescapes that consumers frequent regularly, such as café shops, public transportation, sport club facilities, and supermarkets (Berry, Wall, and Carbone 2006). With regard to the removal of scent diffusion from a servicescape, our results suggest that aftereffects of a nonconsciously processed scent persist for at least 2 weeks, as we did not identify a significant drop in the evaluations after the scent’s removal.

Managerial Implications

Our results suggest that service managers can use nonconsciously processed ambient scents to enhance consumers’ situational evaluation of their service experience, its quality, and its value, and not only temporarily but also over an extended period. Marketers should therefore not only rely on ambient scent diffusion’s short-term commercial benefits but also carefully consider the influence of long-term scent exposure on their consumers in the servicescape, including the effects after discontinuation. Our results suggest that, at least right after the removal of a nonconsciously processed ambient scent, there are no negative aftereffects that jeopardize the previous investments.

It is important to note that an ambient scent’s influence probably varies across service industries. For example, ambient scents appear particularly relevant for services where the situational experience is key (e.g., casinos, hotels, movie theaters, and amusement parks; see Berry, Wall, and Carbone 2006; Biswas and Szocs 2019; Hirsch 1995). Similarly, our results suggest that the positive effects are long-lasting in olfactory-rich servicescapes with many potential sources of olfactory sensations (e.g., buses, trains) and in those where there may be various unpleasant odors (e.g., in elevators or gyms).

In light of the positive effects and the low costs associated with ambient scents’ use vis-à-vis alternative service elements (e.g., new interior design), they have the potential to offer good returns on investment. According to a professional fragrance manufacturer (Scentcommunication, https://scentcommunication.com/en/), the development of an appropriate ambient scent requires an investment of between US$5,000 and 55,000, depending on its purpose, its exclusiveness, and its ingredients. The ambient scent itself is mostly sold in scent carriers, for instance, scent cartridges, costing up to US$28 per cartridge and lasting for weeks. Ambient scent is applied by placing the scent carrier in existing air conditioning systems or via scent diffusing devices, costing up to US$100. In sum, scenting multiple trains in the presented way (i.e., the development of a particular signature scent and placing eight cartridges in an existing air conditioning system) costs roughly US$30,000, with a pronounced cost reduction over time after repeated application.

Limitations and Future Research

As the first study on long-term ambient scent effects, our research offers multiple future research opportunities. First, while we identified ambient scents’ positive long-term and aftereffects on consumers’ situational service perception, future research should shed more light on consumer responses by analyzing a scent’s impact on a company’s bottom line (e.g., in terms of money spent, cross-buying, and up-buying; Hirsch 1995) and examining potential interactions with other environmental factors relevant to the consumers’ service assessment (e.g., lighting; Biswas et al. 2017 and music; Spangenberg, Grohmann, and Sprott 2005). Future research should also focus more strongly on aftereffects, particularly on their persistence over a longer period (i.e., more than 2 weeks).

Second, it is impossible to fully exclude all confounding environmental influences in a field experiment (Gneezy 2017). For instance, in Study 2, we scented all the coaches in all the trains operating on the same line and considered only those consumers who made the same commute every working day. Thus, although we ensured that participants were exposed to the scent during their daily commute, we cannot rule out that they had other train journey experiences (e.g., on the weekends on a different track section). Likewise, we did not account for potential differences in temperature, crowding, and noise in the different train compartments, which may have influenced ambient scent’s effects.

Third, future research should consider scent effects on brand-related constructs, such as consumers’ attitude toward the brand and their brand liking. Attribution theory (Kelley 1973) suggests that such sensory branding activities require consumers to be able and willing to first process the ambient scent consciously (Bradford and Desrochers 2009). In addition, consumers need to relate the scent stimulus to the correct source (Bradford and Desrochers 2009). However, as consumers rarely attribute ambient scents correctly, particularly in olfactory-rich environments (Krishna, Lwin, and Morrin 2010), implementing sensory branding activities appears to be very challenging. Future research should test this notion.

Fourth, future research should aim at generalizing our findings to other service settings. External environmental information cues, such as scents, might be less important for services with high search qualities (Zeithaml 1981). It is therefore reasonable to assume that such service settings show weaker effects. In contrast, we expect stronger effects in servicescapes that consumers encounter in a high construal-level processing mode (e.g., Liberman and Trope 2008), such as travel agencies or companies selling retirement insurances, because recent research suggests that intangible service aspects have a stronger impact in such situations (Ding and Keh 2017). Similarly, while our research is particularly relevant to regularly frequented servicescapes, future research should replicate our findings in infrequently visited sites such as hospitals.

Future research should also investigate whether different ambient scent effects depend on the prevailing olfactory situation within the servicescape (i.e., whether the air quality is negative, neutral, or positive). In this context, assessing ambient scents’ effects in settings where smoking is still allowed would be a fruitful area for future research. Similarly, our investigation took place in winter; the study should therefore be repeated in the summer, or over a full year, to control for different weather conditions, temperatures, and associated olfactory factors. In fact, cold weather might have led to an underestimation of the scent effects.

While we applied a complex ambient scent stimulus, a simpler stimulus might trigger even more pronounced positive effects (Herrmann et al. 2013), since it may induce more pronounced long-term gains in consumers’ service perceptions due to enhanced perceptual fluency. The perceptual fluency concept builds on the processing fluency theory (Schwarz 2004), according to which consumers constantly monitor the effort required to process information cues prevalent in an environment. This effort decreases when the sensory stimulation in a servicescape is simple instead of complex and if consumers process a situation repeatedly (Orth, Wirtz, and McKinney 2016). As fluency per se feels positive (Landwehr, Golla, and Reber 2017), repeated exposure to a simple scent stimulus should translate into improved situational service evaluations. While prior service research has used the concept of perceptual fluency in the context of short-term scent exposure (Herrmann et al. 2013) and for the visual perception of servicescapes (Orth and Wirtz 2014), future studies should consider processing fluency in the context of long-term scent exposure.

Finally, we only investigated ambient scent’s effects on consumers. However, in service companies, the servicescape and all environmental cues also strongly affect those employees working in a servicescape for at least 7 hours per day and 5 days per week (e.g., Bitner 1992). Social psychology research has demonstrated that employees’ mood states likely spill over to consumers via emotional contagion effects (Grandey, Goldberg, and Pugh 2011), thus enhancing scent marketing’s positive effects on the servicescape via an employee-mood state mediation.

Supplemental Material

Supplemental Material, JSR_Scent_Executive_Summary_final - Short- and Long-Term Effects of Nonconsciously Processed Ambient Scents in a Servicescape: Findings From Two Field Experiments

Supplemental Material, JSR_Scent_Executive_Summary_final for Short- and Long-Term Effects of Nonconsciously Processed Ambient Scents in a Servicescape: Findings From Two Field Experiments by Anna Girard, Marcel Lichters, Marko Sarstedt and Dipayan Biswas in Journal of Service Research

Supplemental Material

Supplemental_Material - Short- and Long-Term Effects of Nonconsciously Processed Ambient Scents in a Servicescape: Findings From Two Field Experiments

Supplemental_Material for Short- and Long-Term Effects of Nonconsciously Processed Ambient Scents in a Servicescape: Findings From Two Field Experiments by Anna Girard, Marcel Lichters, Marko Sarstedt and Dipayan Biswas in Journal of Service Research

Footnotes

Acknowledgments

The authors would like to thank Dr. Marc Girard for his support with the data collection for Study 1.

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Marko Sarstedt

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Baeyens

Frank

Wrzesniewski

Amy

Houwer

Jan de

Eelen

Paul

(1996), “Toilet Rooms, Body Massages, and Smells: Two Field Studies on Human Evaluative Odor Conditioning,” Current Psychology, 15 (1), 77–96.

Baker

Julie

Parasuraman

Grewal

Dhruv

Voss

Glenn B.

(2002), “The Influence of Multiple Store Environment Cues on Perceived Merchandise Value and Patronage Intentions,” Journal of Marketing, 66 (2), 120–141.

Berry

Leonard L.

Wall

Eileen A.

Carbone

Lewis P.

(2006), “Service Clues and Customer Assessment of the Service Experience: Lessons from Marketing,” Academy of Management Perspectives, 20 (2), 43–57.

Biswas

Dipayan

Szocs

Courtney

(2019), “The Smell of Healthy Choices: Cross-modal Sensory Compensation Effects of Ambient Scent on Food Purchases,” Journal of Marketing Research, 56 (1), 123–141.

Biswas

Dipayan

Szocs

Courtney

Chacko

Roger

Wansink

Brian

(2017), “Shining Light on Atmospherics: How Ambient Light Influences Food Choices,” Journal of Marketing Research, 54 (1), 111–123.

Biswas

Dipayan

Szocs

Courtney

Krishna

Aradhna

Lehmann

Donald R.

(2014), “Something to Chew On: The Effects of Oral Haptics on Mastication, Orosensory Perception, and Calorie Estimation,” Journal of Consumer Research, 41 (2), 261–273.

Bitner

Mary J.

Faranda

William T.

Hubbert

Amy R.

Zeithaml

Valarie A.

(1997), “Customer Contributions and Roles in Service Delivery,” International Journal of Service Industry Management, 8 (3), 193–205.

Bitner

Mary J.

(1992), “Servicescapes: The Impact of Physical Surroundings on Customers and Employees,” Journal of Marketing, 56 (2), 57–71.

Bradford

Kevin D.

Desrochers

Debra M.

(2009), “The Use of Scents to Influence Consumers: The Sense of Using Scents to Make Cents,” Journal of Business Ethics, 90 (S2), 141–153.

10.

Brady

Michael K.

Cronin

Joseph J.

(2001), “Some New Thoughts on Conceptualizing Perceived Service Quality: A Hierarchical Approach,” Journal of Marketing, 65 (3), 34–49.

11.

Brüggen

Elisabeth C.

Foubert

Bram

Gremler

Dwayne D.

(2011), “Extreme Makeover: Short- and Long-term Effects of a Remodeled Servicescape,” Journal of Marketing, 75 (5), 71–87.

12.

Burdack-Freitag

Andrea

Heinlein

Anja

Mayer

Florian

(2017), “Material Odor Emissions and Indoor Air Quality,” in Springer Handbook of Odor, Buettner

Andrea

, ed. Dordrecht, the Netherlands: Springer, 863–868.

13.

Cain

William S.

Gent

Janneane F.

(1991), “Olfactory Sensitivity: Reliability, Generality, and Association with Aging,” Journal of Experimental Psychology: Human Perception and Performance, 17 (2), 382–391.

14.

Camerer

Colin

Loewenstein

George

Prelec

Drazen

(2005), “Neuroeconomics: How Neuroscience Can Inform Economics,” Journal of Economic Literature, 43 (1), 9–64.

15.

Canniford

Robin

Riach

Kathleen

Hill

Tim

(2018), “Nosenography: How Smell Constitutes Meaning, Identity and Temporal Experience in Spatial Assemblages,” Marketing Theory, 18 (2), 234–248.

16.

Cronin

J. Joseph

Brady

Michael K.

Hult

Tomas M.

(2000), “Assessing the Effects of Quality, Value, and Customer Satisfaction on Consumer Behavioral Intentions in Service Environments,” Journal of Retailing, 76 (2), 193–218.

17.

Cronin

J. J.

Taylor

Steven A.

(1994), “SERVPERF versus SERVQUAL: Reconciling Performance-based and Perceptions-minus-expectations Measurement of Service Quality,” Journal of Marketing, 58 (1), 125–131.

18.

Dabholkar

Pratibha A.

David Shepherd

Thorpe

Dayle I.

(2000), “A Comprehensive Framework for Service Quality: An Investigation of Critical Conceptual and Measurement Issues through a Longitudinal Study,” Journal of Retailing, 76 (2), 139–173.

19.

Dagger

Tracey S.

Sweeney

Jillian C.

(2007), “Service Quality Attribute Weights,” Journal of Service Research, 10 (1), 22–42.

20.

Das

Gopal

Hagtvedt

Henrik

(2016), “Consumer Responses to Combined Arousal-inducing Stimuli,” International Journal of Research in Marketing, 33 (1), 213–215.

21.

Deems

Daniel A.

Doty

Richard L.

Settle

R. G.

Moore-Gillon

Victoria

Shaman

Paul

Mester

Andrew F.

Kimmelman

Charles P.

Brightman

Vernon J.

Snow

James B.

(1991), “Smell and Taste Disorders, A Study of 750 Patients from the University of Pennsylvania Smell and Taste Center,” Archives of Otolaryngology—Head and Neck Surgery, 117 (5), 519–528.

22.

Delaunay-El Allam

Maryse

Soussignan

Robert

Patris

Bruno

Marlier

Luc

Schaal

Benoist

(2010), “Long-lasting Memory for an Odor Acquired at the Mother’s Breast,” Developmental Science, 13 (6), 849–863.

23.

Dholakia

Utpal M.

Morwitz

Vicki G.

(2002), “The Scope and Persistence of Mere-measurement Effects: Evidence from a Field Study of Customer Satisfaction Measurement,” Journal of Consumer Research, 29 (2), 159–167.

24.

Ding

Ying

Keh

Hean T.

(2017), “Consumer Reliance on Intangible versus Tangible Attributes in Service Evaluation: The Role of Construal Level,” Journal of the Academy of Marketing Science, 45 (6), 848–865.

25.

Doty

Richard L.

(2001), “Olfaction,” Annual Review of Psychology, 52 (1), 423–452.

26.

Elejalde-Ruiz

Alexia

(2014), “For Branding, Many Places Adopt Signature Scents,” Los Angeles Times, December 18, 20.

27.

Faul

Franz

Erdfelder

Edgar

Lang

Albert-Georg

Buchner

Axel

(2007), “G* Power 3: A Flexible Statistical Power Analysis Program for the Social, Behavioral, and Biomedical Sciences,” Behavior Research Methods, 39 (2), 175–191.

28.

Fisher

Jeffrey D.

(1974), “Situation-specific Variables as Determinants of Perceived Environmental Aesthetic Quality and Perceived Crowdedness,” Journal of Research in Personality, 8 (2), 177–188.

29.

Forster

Sophie

Spence

Charles

(2018), ““What Smell?”: Temporarily Loading Visual Attention Induces a Prolonged Loss of Olfactory Awareness,” Psychological Science, 29 (10), 1642–1652.

30.

Freiherr

Jessica

(2017), “Cortical Olfactory Processing,” in Springer Handbook of Odor, Buettner

Andrea

, ed. Dordrecht, the Netherlands: Springer, 759–768.

31.

Girard

Cassandra

(2017), “Meet the Scent Marketing Firm Winning the Battle for Your Nose,” NBC News, (accessed June 5, 2018), [available at https://www.nbcnews.com/business/your-business/meet-scent-marketing-firm-winning-battle-your-nose-n783761].

32.

Gneezy

Ayelet

(2017), “Field Experimentation in Marketing Research,” Journal of Marketing Research, 54 (1), 140–143.

33.

Goldkuhl

Lena

Styvén

Maria

(2007), “Sensing the Scent of Service Success,” European Journal of Marketing, 41 (11/12), 1297–1305.

34.

Grandey

Alicia A.

Goldberg

Lori S.

Pugh

S. D.

(2011), “Why and When Do Stores With Satisfied Employees Have Satisfied Customers?,” Journal of Service Research, 14 (4), 397–409.

35.

Hamann

Stephan B.

Ely

Timothy D.

Grafton

Scott T.

Kilts

Clinton D.

(1999), “Amygdala Activity Related to Enhanced Memory for Pleasant and Aversive Stimuli,” Nature Neuroscience, 2 (3), 289–293.

36.

Harris

Lloyd C.

Goode

Mark M. H.

(2004), “The Four Levels of Loyalty and the Pivotal Role of Trust: A Study of Online Service Dynamics,” Journal of Retailing, 80 (2), 139–158.

37.

Hasford

Jonathan

Kidwell

Blair

Hardesty

David M.

Angabe

Keine

(2018), “Emotional Ability and Associative Learning: How Experiencing and Reasoning about Emotions Impacts Evaluative Conditioning,” Journal of Consumer Research, 45 (4), 743–760.

38.

Henseler

Jörg

Ringle

Christian M.

Sarstedt

Marko

(2015), “A New Criterion for Assessing Discriminant Validity in Variance-based Structural Equation Modeling,” Journal of the Academy of Marketing Science, 43 (1), 115–135.

39.

Henshaw

Victoria

Medway

Dominic

Warnaby

Gary

Perkins

Chris

(2016), “Marketing the ‘City of Smells’,” Marketing Theory, 16 (2), 153–170.

40.

Herrmann

Andreas

Zidansek

Manja

Sprott

David E.

Spangenberg

Eric R.

(2013), “The Power of Simplicity: Processing Fluency and the Effects of Olfactory Cues on Retail Sales,” Journal of Retailing, 89 (1), 30–43.

41.

Herz

Rachel S.

(2009), “Aromatherapy Facts and Fictions: A Scientific Analysis of Olfactory Effects on Mood, Physiology and Behavior,” The International Journal of Neuroscience, 119 (2), 263–290.

42.

Herz

Rachel S.

(2005), “Odor-associative Learning and Emotion: Effects on Perception and Behavior,” Chemical Senses, 30 (Suppl. 1), i250–i251.

43.

Hightower

Roscoe

Brady

Michael K.

Baker

Thomas L.

(2002), “Investigating the Role of the Physical Environment in Hedonic Service Consumption: An Exploratory Study of Sporting Events,” Journal of Business Research, 55 (9), 697–707.

44.

Hirsch

Alan R.

(1995), “Effects of Ambient Odors on Slot-machine Usage in a Las Vegas Casino,” Psychology and Marketing, 12 (7), 585–594.

45.

Hofmann

Wilhelm

Houwer

Jan de

Perugini

Marco

Baeyens

Frank

Crombez

Geert

(2010), “Evaluative Conditioning in Humans: A Meta-analysis,” Psychological Bulletin, 136 (3), 390–421.

46.

Holland

Rob W.

Hendriks

Merel

Aarts

Henk

(2005), “Smells Like Clean Spirit: Nonconscious Effects of Scent on Cognition and Behavior,” Psychological Science, 16 (9), 689–693.

47.

Houwer

Jan de

Thomas

Sarah

Baeyens

Frank

(2001), “Association Learning of Likes and Dislikes: A Review of 25 years of Research on Human Evaluative Conditioning,” Psychological Bulletin, 127 (6), 853–869.

48.

Jones

Christopher R.

Fazio

Russell H.

Olson

Michael A.

(2009), “Implicit Misattribution as a Mechanism Underlying Evaluative Conditioning,” Journal of Personality and Social Psychology, 96 (5), 933–948.

49.

Kelley

Harold H.

(1973), “The Processes of Causal Attribution,” American Psychologist, 28 (2), 107–128.

50.

Krishna

Aradhna

, ed. (2012), “An Integrative Review of Sensory Marketing: Engaging the Senses to Affect Perception, Judgment and Behavior,” Journal of Consumer Psychology, 22 (3), 332–351.

51.

Krishna

Aradhna

, ed. (2010), Sensory Marketing: Research on the Sensuality of Products. New York, NY: Taylor & Francis.

52.

Krishna

Aradhna

Elder

Ryan S.

Caldara

Cindy

(2010), “Feminine to Smell But Masculine to Touch? Multisensory Congruence and Its Effect on the Aesthetic Experience☆,” Journal of Consumer Psychology, 20 (4), 410–418.

53.

Krishna

Aradhna

Lwin

May O.

Morrin

Maureen

(2010), “Product Scent and Memory,” Journal of Consumer Research, 37 (1), 57–67.

54.

Laird

Donald A.

(1932), “How the Consumer Estimates Quality by Subconscious Sensory Impressions,” Journal of Applied Psychology, 16 (3), 241–246.

55.

Landwehr

Jan R.

Golla

Benedikt

Reber

Rolf

(2017), “Processing Fluency: An Inevitable Side Effect of Evaluative Conditioning,” Journal of Experimental Social Psychology, 70 (May), 124–128.

56.

Wen

Moallem

Isabel

Paller

Ken A.

Gottfried

Jay A.

(2007), “Subliminal Smells Can Guide Social Preferences,” Psychological Science, 18 (12), 1044–1049.

57.

Liberman

Nira

Trope

Yaacov

(2008), “The Psychology of Transcending the Here and Now,” Science, 322 (5905), 1201–1205.

58.

Lindstrøm

Martin

(2005), Brand Sense: Build Powerful Brands through Touch, Taste, Smell, Sight, and Sound. New York, NY: Free Press.

59.

Lunardo

Renaud

Mbengue

Ababacar

(2013), “When Atmospherics Lead to Inferences of Manipulative Intent: Its Effects on Trust and Attitude,” Journal of Business Research, 66 (7), 823–830.

60.

Marin

A. B.

Acree

T. E.

Barnard

(1988), “Variation in Odor Detection Thresholds Determined by Charm Analysis,” Chemical Senses, 13 (3), 435–444.

61.

Mattila

Anna S.

Wirtz

Jochen

(2001), “Congruency of Scent and Music as a Driver of In-store Evaluations and Behavior,” Journal of Retailing, 77 (2), 273–289.

62.

McDonnell

John

(2007), “Music, Scent and Time Preferences for Waiting Lines,” International Journal of Bank Marketing, 25 (4), 223–237.

63.

McShane

Blakeley B.

Böckenholt

Ulf

(2017), “Single Paper Meta-analysis: Benefits for Study Summary, Theory-testing, and Replicability,” Journal of Consumer Research, 43 (6), 1048–1063.

64.

Moorman-Li

Robin

Gilbert

Brian

Orman

Anna

Aldridge

Petra

Leger-Krall

Sue

Anderson

Clare

Castillo

Juan Hincapie

(2017), “Evaluating the Effects of Diffused Lavender in an Adult Day Care Center for Patients with Dementia in an Effort to Decrease Behavioral Issues: A Pilot Study,” Journal of Drug Assessment, 6 (1), 1–5.

65.

Morrin

Maureen

Chebat

Jean-Charles

(2005), “Person-place Congruency: The Interactive Effects of Shopper Style and Atmospherics on Consumer Expenditures,” Journal of Service Research, 8 (2), 181–191.

66.

Morrin

Maureen

Ratneshwar

(2003), “Does It Make Sense to Use Scents to Enhance Brand Memory?” Journal of Marketing Research, 40 (1), 10–25.

67.

Orth

Ulrich R.

Wirtz

Jochen

McKinney

Amelia

(2016), “Shopping Experiences in Visually Complex Environments: A Self-regulation Account,” Journal of Service Management, 27 (2), 194–217.

68.

Orth

Ulrich R.

Wirtz

Jochen

(2014), “Consumer Processing of Interior Service Environments,” Journal of Service Research, 17 (3), 296–309.

69.

Phelps

Elizabeth A.

LeDoux

Joseph E.

(2005), “Contributions of the Amygdala to Emotion Processing: From Animal Models to Human Behavior,” Neuron, 48 (2), 175–187.

70.

Reekum

Carien M. van

Berg

Helma vann de

Frijda

Nico H.

(1999), “Cross-modal Preference Acquisition: Evaluative Conditioning of Pictures by Affective Olfactory and Auditory Cues,” Cognition & Emotion, 13 (6), 831–836.

71.

Ressler

Newton

(2004), “Rewards and Punishments, Goal-directed Behavior and Consciousness,” Neuroscience and Biobehavioral Reviews, 28 (1), 27–39.

72.

Rimkute

Justina

Moraes

Caroline

Ferreira

Carlos

(2016), “The Effects of Scent on Consumer Behaviour,” International Journal of Consumer Studies, 40 (1), 24–34.

73.

Rolls

Edmund T.

(2015), “Taste, Olfactory, and Food Reward Value Processing in the Brain,” Progress in Neurobiology, 127-128 (April), 64–90.

74.

Sakamoto

Yuko

Ebihara

Satoru

Ebihara

Takae

Tomita

Naoki

Toba

Kenji

Freeman

Shannon

Arai

Hiroyuki

Kohzuki

Masahiro

(2012), “Fall Prevention Using Olfactory Stimulation with Lavender Odor in Elderly Nursing Home Residents: A Randomized Controlled Trial,” Journal of the American Geriatrics Society, 60 (6), 1005–1011.

75.

Schifferstein

Hendrik N. J.

Spence

Charles

(2008), “Multisensory Product Experience,” in Product Experience, Schifferstein

Hendrik N. J.

Hekkert

Paul

, eds. San Diego, CA: Elsevier Science, 133–164.

76.

Schwarz

Norbert

(2004), “Metacognitive Experiences in Consumer Judgment and Decision Making,” Journal of Consumer Psychology, 14 (4), 332–348.

77.

Soudry

Lemogne

Malinvaud

Consoli

S.-M.

Bonfils

(2011), “Olfactory System and Emotion: Common Substrates,” European Annals of Otorhinolaryngology, Head and Neck Diseases, 128 (1), 18–23.

78.

Spangenberg

Eric R.

Grohmann

Bianca

Sprott

David E.

(2005), “It’s Beginning to Smell (and Sound) a Lot Like Christmas: The Interactive Effects of Ambient Scent and Music in a Retail Setting,” Journal of Business Research, 58 (11), 1583–1589.

79.

Spangenberg

Eric R.

Crowley

Ayn E.

Henderson

Pamela W.

(1996), “Improving the Store Environment: Do Olfactory Cues Affect Evaluations and Behaviors?” Journal of Marketing, 60 (2), 67–80.

80.

Spence

Charles

Puccinelli

Nancy M.

Grewal

Dhruv

Roggeveen

Anne L.

(2014), “Store Atmospherics: A Multisensory Perspective,” Psychology & Marketing, 31 (7), 472–488.

81.

Todrank

Josephine

Byrnes

Deidre

Wrzesniewski

Amy

Rozin

Paul

(1995), “Odors Can Change Preferences for People in Photographs: A Cross-modal Evaluative Conditioning Study with Olfactory USs and Visual CSs,” Learning and Motivation, 26 (2), 116–140.

82.

Tortora

Robert D.

(2009), “Attrition in Consumer Panels,” in Methodology of Longitudinal Surveys, Lynn

Peter

, ed. Chichester, England: John Wiley, 235–250.

83.

Verhoef

Peter C.

Lemon

Katherine N.

Parasuraman

Roggeveen

Anne

Tsiros

Michael

Schlesinger

Leonard A.

(2009), “Customer Experience Creation: Determinants, Dynamics and Management Strategies,” Journal of Retailing, 85 (1), 31–41.

84.

Walther

Eva

Nagengast

Benjamin

(2006), “Evaluative Conditioning and the Awareness Issue: Assessing Contingency Awareness with the Four-picture Recognition Test,” Journal of Experimental Psychology: Animal Behavior Processes, 32 (4), 454–459.

85.

Winkielman

Piotr

Zajonc

Robert B.

Schwarz

Norbert

(1997), “Subliminal Affective Priming Resists Attributional Interventions,” Cognition & Emotion, 11 (4), 433–465.

86.

Wirtz

Jochen

Mattila

Anna S.

Tan

Rachel L.P.

(2000), “The Moderating Role of Target-arousal on the Impact of Affect on Satisfaction: An Examination in the Context of Service Experiences,” Journal of Retailing, 76 (3), 347–365.

87.

Zajonc

Robert B.

(1984), “On the Primacy of Affect,” American Psychologist, 39 (2), 117–123.

88.

Zajonc

Robert B.

(1980), “Feeling and Thinking: Preferences Need no Inferences,” American Psychologist, 35 (2), 151–175.

89.

Zeithaml

Valarie A.

(1981), “How Consumers Evaluation Processes Differ between Goods and Services,” in Marketing of Services, Donnelly

James H.

George

W. R.

, eds. Chicago, IL: American Marketing Association, 186–190.

90.

Zemke

Dina M. V.

Shoemaker

Stowe

(2007), “Scent across a Crowded Room: Exploring the Effect of Ambient Scent on Social Interactions,” International Journal of Hospitality Management, 26 (4), 927–940.

91.

Zhang

Hui

Arens

Edward

Pasut

Wilmer

(2011), “Air Temperature Thresholds for Indoor Comfort and Perceived Air Quality,” Building Research & Information, 39 (2), 134–144.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB

0.21 MB