Textual information on food packaging addressing several sensory modalities

Problem description

Setting the scene

Food packaging refers to “all the ways food and beverages are boxed, wrapped, arranged, and presented to consumers in retail stores” [1]. Product packaging can be regarded as a highly important factor in consumer decision-making processes in contemporary market environments. This is particularly so when considering that a substantial number of purchasing decisions are made in-store [2] and that consequently, product packages serve as important extrinsic cues. However, they are also important cues for Internet shoppers when browsing webpages showing wrapped food. Product packaging affects consumer responses such as product selection, intention to purchase, and perception of quality [3]. The concept of packaging design incorporates multiple elements such as verbal information, pictures, shape, size, colors, and graphic design. In particular, the information displayed on the front of the packaging, the so-called front-of-package information, is relevant in marketing, as this type of information can, in contrast to mandatory back-of-package information, be controlled by marketers [1]. In general, packaging often communicates the brand name and logo, price, or origin of a product but fails to provide any information about its sensory characteristics, although trying to take advantage of consumers’ senses could positively affect buying behavior positively [4].

Krishna [5] defines sensory marketing as that which “engages consumers’ senses and affects their perception, judgment, and behavior.” In recent literature, the concept of multisensory marketing has been used to describe marketing measures that appeal to multiple senses, as addressing multiple senses is acknowledged as being capable of evoking stronger and more favorable consumer responses (see, for instance, [6]).

Sensory modalities play a vital role in all the stages of consumption.

However, the relevance of separate senses varies between these stages. When consumers choose and buy products, the dominant sensory modality is vision, which enables them to gather relatively large amounts of information in a comparatively short time [7].

During the purchase of food products, consumers lack reliable information regarding the sensory experience of the product; thus, extrinsic cues are vital [8]. However, it is possible to code information regarding the sensory attributes of products in language, and thus communicate them as cognition [9]. Cognitively conveyed information about a product’s sensory properties is also referred to as amodal sensory information [5]; that is, information lacking the typical stimulus and perception conditions for a sensory modality. In this study, we refer to verbally conveyed amodal sensory information as verbal sensory information.

The relevance of verbal sensory information arises from several factors, two of which are emphasized here. First, as previously mentioned, the dominance of sight over other sensory modalities during the product selection and purchase stages is acknowledged [7]. Cognitive neuroscience research provides a second explanation for the relevance of verbal sensory information, and current studies show connections between verbal information and the activation of certain parts of the brain, with these brain regions not just tied to the modality in which the information has been presented [10]. Indeed, sensory stimuli are not processed in isolation; rather, manifold associations (i.e., cross-modal correspondences) between sensory modalities may occur. Cross-modal correspondences constitute “nonarbitrary associations that appear to exist between different basic physical stimulus attributes, or features, in different sensory modalities” [11]. Several empirical studies have provided evidence of consistent cross-modal correspondence between verbal information and sensory attributes in non-synesthetic individuals, reporting, for example, the correspondence between verbal stimuli and tastes or flavors [12], as well as odors [13].

Moreover, empirical work gives reason to assume that our senses shape language: Studies demonstrate that sensory metaphors (such as describing a person as ‘cold’ instead of unfriendly) are more memorable, tend to be used more frequently than non-sensory (semantic) equivalents, and evoke more associative cues because they involve the senses [14]. Furthermore, recent research indicates that language can be used to construct synaesthetic metaphors (connections between concepts referring to different sensory modalities) in print advertisements. An example of a synaesthetic metaphor is a ‘sweet voice,’ in which a hearing-related concept is described through a taste-related attribute [15].

Building on these considerations, scholars have shown that verbal sensory information, as an extrinsic cue, is highly relevant to sensory and hedonic consumer responses [16]. Support for this assumption stems, inter alia, from neuroimaging studies indicating that verbal sensory information may affect brain responses. For instance, such information may modulate cortical representations of taste [17] and activate the olfactory brain regions [10]. However, the effects of verbal sensory information on consumer responses have not received much academic attention in the field of consumer research. To date, studies have examined the effects of descriptive sensory food names on the evaluation and choice of food [18], as well as the effects of verbal sensory information on the choice of product [4], and food acceptance [19]. However, the research on the influence of verbal sensory information is limited.

Responses to verbal sensory information

Extrinsic information is capable of stimulating the anticipation of a product’s attributes, and thus generates sensory and hedonic expectations [3]. These preconceived beliefs about product attributes are highly relevant as they may create bias in the sensory perception of products [20] and evoke expectations regarding future pleasure when consuming the product. This may, in turn, affect decisions about consumption and make the consumption experience more enjoyable [21].

The imagination of emotions, which “are expected to be experienced in the future if certain events do or do not occur” [22], is referred to as the construct of anticipated emotions. These are “spontaneous reactions to an external stimulus” that describe consumers’ expected feelings in the future [23]. The extant literature suggests that the tendency of consumers to anticipate the experience of consumption and the emotions associated with it becomes more relevant and likely, the more products possess experiential qualities [24]. Thus, anticipated emotions seem particularly relevant to food products.

Moreover, research provides reason to assume that consumers will not only anticipate emotions, but also simulate prior perceptions associated with products when confronted with relevant information; that is, viewing visual or verbal information about a product may induce similar behaviors to interact with the product. Scholars conclude that “our initial perceptions of objects, both conscious and nonconscious, are stored in memory and are simulated or played back on subsequent encounters with not only the object itself but also representations of that object, such as verbal and visual depictions” [25].

The concept of embodied mental simulation suggests that when interacting with a product, our brains encode and integrate all sensory perceptions (e.g., how it feels, looks, or smells) related to the product. When we are subsequently confronted with (verbal or visual) information regarding a product, we are likely to mentally simulate perceptions previously associated with the product. This process of embodied mental simulation results in neural activation of the same sensory brain regions that are active during sensory perception [26]. These assumptions are based on embodiment theory, which aims to explain the linkage between ’low’ cognitive processes as perception and ‘high’ cognition as language processing and comprehension [27].

Embodied approaches to language comprehension suggest that verbal descriptions of our perceptual (sensory) world–that is, descriptions of what we see, feel, hear, taste, or smell–are mentally simulated within the brain’s perception and action systems [28]. In other words, understanding verbal sensory information may initiate processes used to perceive the real-world referent of a word. However, there is no conclusive empirical evidence of such mental simulation processes for all human senses. While the literature indicates that sound words are mentally simulated, there is no clear evidence for the mental simulation of olfactory information [29], which highlights the importance of gaining a more profound understanding of the potential effects of verbal sensory information on embodied mental simulation.

In addition to anticipated emotions and embodied mental simulation, extrinsic cues may affect buying behavior [30], particularly the intention to purchase [31].

Research agenda

Regarding the composition of verbal sensory information, the number of senses addressed by sensory informationmayaffect consumer responses in the same way that multiple-sense information (compared to information stimulating only a single sense) enhances perceptions of taste [32]. This has been confirmed and extended by work reporting the positive effects of (verbally presented) multiple-sense information on the valence and number of sensory associations of a product [33]. However, owing to the lack of studies that systematically approach this topic, more comprehensive insights into the effects of the number of senses addressed are lacking.

Given the findings above, we put forward the following research question:

RQ 1: In which way does verbal sensory information displayed on product packaging addressing either a single sensory modality or more than one modality affect consumer responses?

Drawing on previously reported findings, we hypothesize that:

H1 to RQ 1: Verbal sensory information will favorably affect a) anticipated emotions, b) embodied mental simulation, c) intention to purchase,and that, d) this effect will be stronger for multiplesense information than for single-sense information.

Sensory information does not always operate in the same manner. Although the literature points to the effectiveness of extrinsic food cues [34], several empirical studies report surprising results, which provide a reason to assume that the effects of extrinsic cues could be biased by numerous factors. Bialkova et al. [35] assume that the effects of verbal sensory information depend on the product category for which sensory information is employed. One possible explanation for the relevance of the product category may be found in the fact that knowledge of the product category can affect consumers’ use of extrinsic cues such as verbal sensory information [36]. Moreover, considering that hedonic expectations can affect sensory perceptions [37], as well as behavioral intentions [38], the sensory appeal associated with the product category could also be decisive in modulating the effects of verbal sensory information on consumer responses.

However, the effects of product categories have not been subject to systematic investigation; thus, it remains unclear whether and in what way different product categories may cause heterogeneous responses to verbal sensory information. Therefore, we propose the following research question:

RQ 2: Does sensory information have a different effect on consumer responses across different categories of packaged food products which vary regarding their sensory appeal?

Scientific work suggests that various demographic, psychographic, and socioeconomic predictors may influence the use of information provided in food packaging [39]. In particular, the gender and age of respondents are acknowledged to affect the processing of sensory cues, and extant research indicates that women tend to use information provided on food labels more frequently and conscientiously than men [40]. Additionally, scholars point to the fact that thresholds for elaborative processing of sensory stimuli differ by gender [41].

Younger and older consumers differ in their perceptions and interpretations of food labels [42]. Research suggests that older consumers are more vigilant about sensory attributes and consider more characteristics when evaluating foods [43]. Consistent with this finding, age has been found to be a significant predictor of food label use in some studies. However, while some scholars conclude that older consumers tend to use food labels more intensively [40], others argue that younger consumers are more influenced by information on labels than middle-aged or older consumers [44].

The current state of hunger during participation in a study may bias consumer responses because it relates to sensory sensitivity [45] and may affect brain responses to food cues [46], thus potentially influencing consumers’ reactions to the stimuli. Finally, sensory and hedonic expectations, such as the expected liking for a product, may shape the effects of extrinsic cues on perceptions of a product [47]. Using these considerations, we employed gender, age, feelings of hunger, and liking a product as covariates. Figure 1 illustrates this conceptual model.

 

Figure 1: Conceptual model.

Empirical study

Method

We employed a mixed 3-9 two-factor experimental design, with the number of sensory modalities addressed by sensory information serving as a between-subjects factor, resulting in three experimental conditions (CG: control group, no sensory information; EG1: Single-Sense Information; EG2: Multiple-

Sense Information). In the single-sense condition, amodal sensory information targets only one sensory modality at a time (taste, texture, or smell), whereas in the multiple-sense condition, two different sensory modalities are stimulated simultaneously (e.g., taste and smell). Nine different products (cereal bars, chocolate, cookies, crisps, honey, ketchup, raisins, rice cakes, and smoothies) served as the within-subjects factor which varied with respect to their sensory appeal (e.g., crisps vs. rice cakes). The selection of these products is based on several considerations. First, recent literature suggests that the strength of the association between food products and sensory modalities depends on the perceived health of the food. Speed [48] showed that unhealthy foods are more strongly associated with all perceptual modalities (particularly taste, smell, and interoception) than are healthy food products. As unhealthy food is believed to be more attractive than healthy food, investigating products that differ in sensory appeal could be enlightening in this context. Second, all products can be classified as snacks or snack-related (ketchup and honey) rather than as full meals. Third, all the products were presumably familiar to and affordable for our respondents (regarding honey, the product used in our study was a standard honey and not an exclusive variety). Moreover, the use of verbal sensory information seems meaningful for different products as all of them contain different flavors/types. Finally, our selection of products is pragmatic as they are all packaged with labels that can be manipulated.

Images of real product packaging were manipulated using the graphics software package Adobe Photoshop to design singlesense, multiple-sense, and control conditions for each product. The stimuli varied only regarding the manipulated factor (i.e., the sensory information provided) but were kept identical across the groups in all other respects (e.g., color, shape, or size). To ensure the same length (word count) for all stimuli, we employed two sensory attributes referring to the same sensory modality in the singlesense conditions and combined two sensory attributes referring to different sensory modalities in the multiple-sense conditions. For example, ketchup was described as “Mild ketchup. Delicate taste of sun-ripened tomatoes” (a single sense: flavor) or as “Mild ketchup. Delicate smell of sun-ripened tomatoes” (multiple senses: smell and flavor) (Figure A1 of Appendix A). For chocolate, either the information “Pleasantly bitter. Balanced cocoa and mint flavor” (a single sense: flavor) or “Pleasantly smooth. Balanced cocoa and mint flavor” (multiple senses: texture and flavor) were provided. Sensory attributes were selected based on an online pretest (n = 100 consumers in a Central European country), in which we asked respondents to list all sensory attributes associated with the different categories. After checking for plausibility and valence (only attributes that could be regarded as positive/desirable for the respective category were included), the most frequently mentioned attributes per sensory modality were used to develop the stimuli. The stimuli of the control condition included the product name (e.g. “Ketchup”), but no sensory or other information in order not to confound responses. Appendix A provides the exact wording (in German with a translation into English) of the sensory information used for data collection under the two experimental conditions for each product. Appendix B presents all labels.

The study was conducted in a laboratory located at a Central European university to ensure that all participants encountered the same sensory and environmental conditions (e.g., lighting, setting, sounds, and room temperature). The data collection form was a paper-and-pencil questionnaire. A team of trained research assistants supported the data collection. They recruited respondents in this Central European country based on a predefined quota for age and gender who were invited to participate in a study on ‘packaging design.’ The subjects were randomly assigned to the single-sense condition, multiple-sense condition, or control group. In each condition, participants received photo albums with images of the packaging of nine products and were asked to answer questions regarding each product based on their packaging. To avoid sequence bias, the order of products was manipulated systematically. Respondents’ anonymity was protected, and they could terminate responding at any time. After completing the questionnaire, they were debriefed, dismissed, and did not receive compensation for their participation.

Anticipated emotions, embodied mental simulation, and intention to purchase were response variables. Self-assessment was performed using well-established scales. For embodied mental simulation, although this construct is related to bodily reactions (i.e., mental simulation may lead to neural activation of many of the same sensory regions of the brain that are active during perception), the aim of our study was not to measure and substantiate the activation of brain regions, but rather to investigate whether verbal sensory information is likely to evoke or enhance mental simulation.

Operationalization as self-assessment is a common practice in the literature (see, for instance, [25] and is based on the understanding of mental simulation as “product-related imagery or imaginary product experiences that consumers project in their minds based on relevant marketing information they receive” [49]. Therefore, we considered self-assessment as an appropriate measure in this context. Appendix C presents the measurement details (also for the covariates) and psychometric properties, which were highly satisfactory.

Altogether, 421 respondents (quota sample), 51% female and 49% male, participated in this study. Respondents’ ages ranged from 15 to 91 years, with an average of 44 years. Participants indicated that they were moderately hungry (a mean of 3.15 on a 7-point scale, increasing with feelings of hunger) at the beginning of the study. The average liking for the products ranged from 3.52 (for rice cakes) to 5.69 (for chocolate). Group sizes for the experimental conditions were balanced (approximately 140 per group). Appendix D displays the averages for all the variables used in this study.

Results

Tables 1–3 present the results provided by a mixed design ANCOVA for each response variable; being product specific, the covariate liking could not be included in this analysis. The marginals of response variables have been estimated by a series of univariate ANCOVAs, also including the liking of the product as a covariate. We note that the differences between these marginals and the marginals based on the mixed design ANCOVA (ignoring liking) are negligible, i.e. the significant influence of liking affected all the experimental groups to about the same extent.

Anticipated emotions. Table 1 shows the significant main effects of both, experimental condition and product category. Moreover, hunger, as a covariate, significantly enhanced anticipated emotions. The influence of gender and age is product-specific (right column of Table 1) and the results have face validity. Regarding the number of senses stimulated, sensual information (i.e., information stimulating one or two senses simultaneously) is advantageous over a stimulus without any verbal sensory information. Evaluations of anticipated emotions in general, and the effects of verbal sensory information in particular, differ across different products. Consistently, marginals are smaller for the CG than for the EGs; addressing the two sensory modalities is preferable in most but not all (e.g., smoothies) cases. Rice cakes had the lowest anticipated emotions. Raisins, smoothies, honey, and rice cakes deserve particular mention, as marginal anticipated emotions were significantly different across the experimental groups.

Response Variable: Anticipated Emotions (AE)
Effect Variable EG	F	p-level	Report on more detailed (statistically significant, type I error of 5%) results
	4.97	<0.01	EG: overall, AE are higher for EG1 and EG2 than for CG (pairwise comparisons)
Product Gender Age Hunger Product × EG Product × Gender Product × Age Product × Hunger	14.34 1.88 0.87 20.29	<0.01 0.17 0.35 <0.01	Product(1): in particular, for raisins, smoothies, honey, and rice cakes AE differ over experimental conditions Gender: honey, rice cakes, and smoothies induce higher AE emotions for female (than for male) participants Age: cereal bars, crisps, ketchup, and smoothies induce higher AE for younger, but raisins for older respondents Hunger: AE increase with hunger (not for honey, raisins, and rice cakes) Liking(2): AE increase with the liking of a product consistently
	0.92 5.74 12.87 1.23	0.54 <0.01 <0.01 0.28
Notes:     Mixed design ANCOVA: between factor - experimental group EG (3 levels); within factor – product (9 levels); covariates: gender, age, hunger (1)based on a series of univariate ANCOVA -tests (for each product category) with gender, age, hunger, and liking as covariates (2)results for liking according to a series of separate univariate ANCOVAs

Table 1: Anticipated emotions, results of mixed design ANCOVA.

Response variable: Embodied Mental Simulation (EMS)
Effect Variable EG Product Gender Age Hunger Product × EG Product × Gender Product × Age Product × Hunger	F	p-level	Report on more detailed (statistically significant, type I error of 5%) results
	2.24 11.02 0.41 2.53 15.50	0.11 <0.01 0.52 0.11 <0.01	EG: overall, EMS is higher for EG1 and EG2 than for CG (but at a 10% level only, pairwise comparisons) Product(1): in particular, for raisins and honey EMS differ over experimental conditions Gender: rice cakes and smoothies induce higher EMS for female (than for male) participants Age: cereal bars, crisps, ketchup, and smoothies induce higher EMS for younger, but raisins and chocolate for older respondents Hunger: EMS increases with hunger (not for honey, ketchup, raisins, and rice cakes) Liking(2): EMS increases with the liking of a product consistently
	1.15 4.37 13.47 1.05	0.31 <0.01 <0.01 0.39
Notes:     Mixed design ANCOVA: between factor – experimental group EG (3 levels); within factor – product (9 levels); covariates: gender, age, hunger (1)based on a series of univariate ANCOVA -tests (for each product category) with gender, age, hunger, and liking as covariates (2)results for liking according to a series of separate univariate ANCOVAs

Table 2: Embodied mental simulation, results of mixed design ANCOVA.

Embodied mental simulation. Table 2 presents results similar to those for anticipated emotions; however, the effect of the experimental manipulation was weaker. We consider similar patterns of results remarkable, as both variables represent different constructs. While anticipated emotions refer to predictions of the emotional consequences of a behavior [50], the concept of mental simulation relates to cognition, particularly the theory of grounded cognition [49], postulating that cognition is based on sensorimotor processing. Mental simulation is the mechanism for grounding cognition and is understood as a type of mental imagery that reinforces modal processing. The mechanism of mental imagery can be described as “the ability to reactivate and manipulate modality-specific mental representations when current sensory stimulation or overt motor action is absent, and this ability can be associated with the subjective experience of perceiving or acting within one’s mental world” [51]. The concept of embodied mental simulation proposes that based on a link between language and the sensorimotor system, our bodies interact with their environment [52] and simulate their perceptions of objects [25]. However, both anticipated emotions and mental simulation may be understood as prefactuals, which “capture the idea of something that is not yet a fact but could well become a fact” [53]. This common base may account for the similar patterns in the results.

Response variable: Purchase intention (PI) Effect Variable EG F p-level Report on more detailed (statistically significant, type I error of 5%) results 2.21 0.11 Product Gender Age Hunger Product × EG Product × Gender Product × Age Product × Hunger 15.36 7.42 0.05 20.34 <0.01 <0.01 0.82 <0.01 EG: overall, PI is higher for EG2 than for CG (pairwise comparisons) Product(1): in particular, for raisins and honey PI differ over experimental conditions Gender: females are more inclined than males to purchase honey, rice cakes, and smoothies Age: younger respondents prefer crisps, ketchup, and smoothies; older respondents prefer chocolate, honey, and raisins Hunger: PI increases with hunger (not for honey, ketchup, and rice cakes) Liking(2): PI increases with the liking of a product consistently 1.11 3.94 14.43 1.77 0.34 <0.01 <0.01 0.08 Notes:     Mixed design ANCOVA: between factor - experimental group EG (3 levels); within factor – product (9 levels); covariates: gender, age, hunger (1)based on a series of univariate ANCOVA -tests (for each product category) with gender, age, hunger, and liking as covariates (2)results for liking according to a series of separate univariate ANCOVAs Table 3: Intention to purchase, results of mixed design ANCOVA.

Purchase intention: Table 3 reveals the highly significant withinsubject effects of product, hunger, and gender as covariates on purchase intention. However, the effect of the number of senses was not statistically significant. A closer investigation of the marginals reveals a similar (but somewhat weaker) pattern of results as for anticipated emotions. Overall, the simulation of multiple senses is (directionally) advantageous compared to addressing a single sense, which is, again, more favorable than the absence of verbal sensory information.

Conclusions

Results concerning RQ1 – verbal sensory information on product packaging

Effects of verbal sensory information in general. Our results demonstrate the positive effects of verbal sensory information on consumer response. Interestingly, as regards the strength of the effects of verbal sensory information on the response variables investigated here, these effects are the stronger, the more ‘sensory-related’ (i.e. the closer to the visualization of the sensory experience of the product) the response variables are. While highly significant effects of the number of senses stimulated by verbal sensory information on anticipated emotions emerge, this effect is only marginally significant for embodied mental simulation and intention to purchase. Thus, our findings provide (directional) support for H1a, b, and c.

Effects of number of senses addressed by sensory information. As hypothesized, our results clearly demonstrate that the number of senses addressed by the sensory information doesaffectconsumer responses. This is because the stimulation of multiple senses generates more positive consumer responses than singlesense information, which again proves more advantageous than the absence of verbal sensory information (EG2 > EG1 > CG). However, while the effects are admittedly weak, surprising deviations for certain products can be observed (i.e., partial support for H1d).

Results concerning RQ2 – Effects of product category

Unambiguous product effects exist across all the categories. The highly significant within-subject effect of the product across all the response variables investigated leaves no doubt that the product category might bias the effects of verbal sensory information. As the interpretation of these biasing effects is complex, a closer look at certain categories that reveals surprising response patterns seems worthwhile.

Verbal sensory information is particularly useful for raisins, rice cakes, and honey. Remarkably, an inspection of the product categories reveals certain commonalities, as all these categories represent polarizing products with, at least to some extent, a lower sensory appeal inherent to the product (raisins, rice cakes), or can be regarded as challenging depending on their ambiguous classification as healthy or unhealthy, with the resulting implications (honey, raisins). Hence, our results indicate that the use of verbal sensory information may be particularly effective for polarizing product categories. This finding is particularly remarkable as recent literature provides valid arguments supporting the opposite by showing stronger associations for unhealthy (hence sensorially appealing) food than for healthy (and less appealing) food for all perceptual modalities [48]. It is possible that the provision of sensory information might have helped establish stronger sensory associations, especially for products that could not be clearly classified as sensorially appealing. Importantly, it should be emphasized that the use of verbal sensory information might be intricate and demanding in such cases. Therefore, particular attention should be paid to the careful development and testing of this information (the next subsection describes such targeting in more detail).

Results Concerning Covariates

Responses to certain products differed by gender; honey, rice cakes, and smoothies induced stronger anticipated emotions for female than male participants. Moreover, rice cakes and smoothies elicited higher embodied mental simulation for female respondents. These findings may, to some extent, be explained by the fact that according to the extant literature, women are more likely than men to purchase these products (which is in line with our results). This might point to a more extensive knowledge of these products for women than for men, possibly accounting for the more favorable responses of the female participants regarding these (see, for example, [54] for the gender effects regarding honey consumption, and [55] for a discussion of the role of gender with respect to chocolate). Remarkably, studies have also revealed a more critical attitude of women (compared to men) towards sugar, gluten, dairy, red meat, white flour, alcohol, and food additives [56] and indicating that women tend to prefer healthier food.

Regarding age, certain products caused different responses, depending on the respondents’ age. Overall, chocolate, honey, and raisins prompt more favorable responses from older consumers. Raisins evoke stronger anticipated emotions, a higher degree of embodied mental simulation, and higher purchase intentions for older respondents than for younger consumers. Chocolate induces a stronger degree of embodied mental simulation and a higher intention to purchase, while honey causes a higher intention to purchase for older consumers.

A review of the literature in the fields of nutrition and eating behavior may account for these findings. Honey may be regarded as an extraordinary and highly complex product that differs remarkably in terms of consumers’ appearance, composition, and sensory perception by consumers [57]. Literature suggests that the intention to purchase honey increases with increasing consumer age. According to some authors, one explanation for this effect might be found in the fact that older consumers tend to pay more attention to their health, and honey is a traditional health food [54]. Thus, the health benefits of honey seem to be of vital importance regarding consumer responses, with the propensity to consume honey for health purposes being influenced by consumer characteristics such as age and gender [58].

Interestingly, in addition to behavioral intentions, the relevance and quality of the sensory perceptions of honey seem to depend on age. Older consumers are better at describing and perceiving the sensory properties of honey [59], whereas younger consumers consider price a more relevant criterion than sensory characteristics when purchasing honey [60] (this remark is based on further results not presented in detail in this paper). We consider this a noteworthy finding that might account for the positive responses of older respondents towards honey in our study.

Raisins represent a similarly polarizing category, being frequently referred to in everyday discourse as a controversial snack or ‘love or hate product.’ Although raisins have traditionally been suspected of being cariogenic (i.e., promoting dental caries), recent studies have proposed the opposite and showed that raisins have antibacterial properties that may reduce oral pathogens, which could even make them beneficial for oral health [61].

Regarding chocolate, socio-demographic factors are acknowledged to play a vital role, with the literature proposing more favorable responses towards chocolate for older and female respondents [55]. While we did not find any gender effects for chocolate, our results confirmed the favorable effect of increasing age on responses to chocolate.

Younger respondents tended to rate cereal bars, crisps, ketchup, and smoothies the most favorably. As these categories tend to represent “snack products,” our findings may be indicative of the recognition that the food preferences of younger consumers are often for “fast food” [62]. Additionally, scholars emphasize the tendency of younger cohorts to consume more industrial and semiindustrial food products, which are characterized by their often intensive and sweet taste [59]. This may have biased the sensory perceptions of younger consumers. Moreover, younger cohorts consider other criteria such as price, which is more relevant than the sensory characteristics of products [60]. We consider both findings worthy of remark, as they might account for age differences in the processing and perception of the verbal sensory information. This could be a promising avenue for future research to investigate how sensory information should be designed to enhance the attractiveness of healthier products to younger consumers.

Moreover, as put forward in our conceptual model, feelings of hunger significantly affected anticipated emotions, embodied mental simulation, and the intention to purchase, in particular, for food edible on its own (i.e., snacks rather than honey or ketchup).

Naturally, liking a product had a positive impact on the response variables, but we did not observe any product-specific patterns.

Implications

This study has scientific and practical implications because it investigates the effects of verbal sensory information, particularly the impact of the number of senses addressed by verbal sensory information. Moreover, this research sheds light on the effects of product category, which remains largely ignored in the academic literature. Our empirical work provides the first argument for why the provision of verbal sensory information might be particularly promising for product categories characterized by either low sensory appeal or polarizing associations regarding their benefits. We consider this a noteworthy finding, as, against this background, verbal sensory information might contribute to the reduction of barriers to the consumption of sustainable/climate-friendly products (which share the prejudice of being less sensorially appealing than their conventional counterparts), or novel foods.

Moreover, our research highlights the role of socio-demographically induced particularities in consumers’ perceptions of different products, with important implications for the development of appropriate sensory information for different categories and target groups. Our results can be easily applied to packaging design and marketing for food products and contribute to a general understanding of the potential range of use of verbal sensory information.

Limitations and further research

In this study, the participants responded to pictures of food packaging. We report the responses before the product was tasted. Bearing in mind that extrinsic cues might be capable of altering sensory perception and responses after tasting, empirical work employing real products, incorporating the opportunity to taste these products, and assessing responses after tasting would be an interesting expansion of this experiment. Beyond that, and against the background that we find significant effects of product category, the inclusion of further categories and the systematic investigation of the potential causes for these effects, possibly including real and fictitious brands, offers a promising avenue for future research. Although we distinguished between single- and multiple-sense information, our results did not provide reliable insights into the effectiveness of separate sensory modalities. Thus, there is a strong need for future research on the effects of the stimulation of different senses.

References

1. Chandon P (2013) How package design and packaged-based marketing claims lead to overeating. Applied Economic Perspectives and Policy 35: 7-31.
2. Stilley KM, Inman JJ, Wakefield KL (2010) Spending on the fly: Mental budgets, promotions, and spending behavior. Journal of Marketing 74: 34-47.
3. Deliza R, MacFie HJH (1996) The generation of sensory expectation by external cues and its effect on sensory perception and hedonic ratings: A review. Journal of Sensory Studies 11: 103-128.
4. Swahn J, Mossberg L, Öström A, Gustafsson I (2012) Sensory description labels for food affect consumer product choice. European Journal of Marketing 46: 1628-1646.
5. Krishna A (2012) An integrative review of sensory marketing: Engaging the senses to affect perception, judgment and behavior. Journal of Consumer Psychology 22: 332-351.
6. Knoeferle KM, Knoeferle P, Velasco C, Spence C (2016) Multisensory brand search: How the meaning of sounds guides consumers’ visual attention. Journal of Experimental Psychology: Applied 22: 196-210.
7. Schifferstein HNJ, Fenko A, Desmet PMA, Labbe D, Martin N (2013) Influence of package design on the dynamics of multisensory and emotional food experience. Food Quality and Preference 27: 18-25.
8. Hoppert K, Mai R, Zahn S, Schwarz PEH, Hoffmann S, et al. (2014) Is there a fit in cognitive and sensory evaluation of yogurt? The moderating role of nutrition training. Food Quality and Preference 31: 65-68.
9. Mittal B (2004) Lack of attribute searchability: Some thoughts. Psychology & Marketing 21: 443-462.
10. González J, Barros-Loscertales A, Pulvermüller F, Meseguer V, Sanjuán A, et al. (2006) Reading cinnamon activates olfactory brain regions. NeuroImage 32: 906-912.
11. Spence C (2011) Cross-modal correspondences: A tutorial review”, Attention, Perception, & Psychophysics 73: 971-995.
12. Crisinel AS, Jones S, Spence C (2012) The sweet taste of maluma: Cross-modal associations between tastes and words. Chemosensory Perception 5: 266-273.
13. Herz RS, von Clef J (2001) The influence of verbal labeling on the perception of odors: evidence for olfactory illusions?. Perception 30: 381-391.
14. Akpinar E, Berger J (2015) Drivers of cultural success: The case of sensory metaphors. Journal of Personality and Social Psychology 109: 20-34.
15. Bolognesi M, Strik Lievers F (2020) How language and image construct synaesthetic metaphors in print advertising. Visual Communication 19: 431-457.
16. Li J, Streletskaya NA, Gómez MI (2019) Does taste sensitivity matter? The effect of coffee sensory tasting information and taste sensitivity on consumer preferences. Food Quality and Preference 71: 447-451.
17. Okamoto M, Dan I (2013) Extrinsic information influences taste and flavor perception: A review from psychological and neuroimaging perspectives. Seminars in Cell & Developmental Biology 24: 247-255.
18. Amsteus M, Liljegren F, Markovic S, Månsson M (2016) Chocolate or succulent chocolate: The impact of sensory descriptions on choice. International Journal of Business and Social Science 7: 213-224.
19. Tuorila HM, Meiselman HL, Cardello AV, Lesher LL (1998) Effect of expectations and the definition of product category on the acceptance of unfamiliar foods. Food Quality and Preference 9: 421-430.
20. Schifferstein HNJ (2001) Effects of product beliefs on product perception and liking. In: Frewer LJ, Risvik E, Schifferstein H (eds) Food, People and Society: A European perspective of consumers’ food choices. Springer, Berlin, Heidelberg Pg No: 73-96.
21. Alba JW, Williams EF (2013) Pleasure principles: A review of research on hedonic consumption. Journal of Consumer Psychology 23: 2-18.
22. Baumgartner H, Pieters R Bagozzi RP (2008) Future-oriented emotions: Conceptualization and behavioral effect. European Journal of Social Psychology 38: 685-696.
23. Moore DJ, Lee SP (2012) How advertising influences consumption impulses: The role of visualization, anticipated emotions, taste anticipation, and hedonic rationalization. Journal of Advertising 41: 107-120.
24. Underwood RL, Klein NM, Burke RR (2001) Packaging communication: Attentional effects of product imagery. Journal of Product and Brand Management 10: 403-422.
25. Elder RS, Krishna A (2012) The ‘visual depiction effect’ in advertising: Facilitating embodied mental simulation through product orientation. Journal of Consumer Research 38: 9881003.
26. Simmons WK, Martin A, Barsalou LW (2005) Pictures of appetizing foods activate gustatory cortices for taste and reward. Cerebral Cortex 15: 1602-1608.
27. Jirak D, Menz MM, Buccino G, Borghi AM, Binkofski F (2010) Grasping language - A short story on embodiment. Consciousness and Cognition 19: 711-720.
28. Barsalou LW (1999) Perceptual symbol systems. Behavioral and Brain Sciences 22: 637-660.
29. Speed LJ, Majid A (2018) An exception to mental simulation: No evidence for embo-died odor language. Cognitive Science 42: 1146-1178.
30. Kahn BE, Wansink B (2004) The influence of assortment structure on perceived variety and consumption quantities. Journal of Consumer Research 30: 519-533.
31. Lado Cousté N, Martos-Partal M, Martínez-Ros E (2012) The power of a package: Product claims drive purchase decisions. Journal of Advertising Research 52: 364-375.
32. Elder RS, Krishna A (2010) The effects of advertising copy on sensory thoughts and perceived taste. Journal of Consumer Research 36: 748-756.
33. Steiner E, Wagner U (2016) Einfluss amodaler, sensorischer Verpackungsinformationen auf Assoziationen zum betreffenden Produkt. Marketing ZFP 38: 119-134.
34. Piqueras-Fiszman B, Spence C (2015) Sensory expectations based on product-extrinsic food cues: An interdisciplinary review of the empirical evidence and theoretical accounts. Food Quality and Preference 40: 165-179.
35. Bialkova S, Sasse L, Fenko A (2016) The role of nutrition labels and advertising claims in altering consumers’ evaluation and choice. Appetite 96: 38-46.
36. Chocarro R, Cortiñas M, Elorz M (2009) The impact of product category knowledge on consumer use of extrinsic cues – A study involving agrifood products. Food Quality and Preference 20: 176-186.
37. Blackmore H, Hidrio C, Yeomans MR (2021) A taste of things to come: the effect of extrinsic and intrinsic cues on perceived properties of beer mediated by expectations. Food Quality and Preference 94: 104326, pp. 1-12.
38. Chang IC, Liu CC, Chen K (2014) The effects of hedonic/ utilitarian expectations and social influence on continuance intention to play online games. Internet Research 24: 21-45.
39. Bryła P (2020) Who reads food labels? Selected predictors of consumer interest in front-of-package and back-of-package labels during and after the purchase. Nutrients 129: 2605.
40. Stran KA, Knol LL (2013) Determinants of food label use differ by sex. Journal of the Academy of Nutrition and Dietetics 113: 673-679.
41. Meyers-Levy J, Sternthal B (1991) Gender differences in the use of message cues and judgments. Journal of Marketing Research 28: 84-96.
42. Piqueras-Fiszman B, Ares G, Varela P (2011) Semiotics and perception: Do labels convey the same messages to older and younger consumers?. Journal of Sensory Studies 26: 197-208.
43. Asioli D, Canavari M, Castellini A, de Magistris T, Gottardi F, et al. (2011) The role of sensory attributes in marketing organic food: Findings from a qualitative study on Italian consumers. Journal of Food Distribution Research 42: 1-6.
44. Bae YW, Lee SM, Kim KO (2016) Age and gender differences in the influence of extrinsic product information on acceptability for RTD green tea beverages. Journal of the Science of Food and Agriculture 96: 1362-1372.
45. Stafford LD, Welbeck K (2011) High hunger state increases olfactory sensitivity to neutral but not food odors. Chemical Senses 36: 189-198.
46. Charbonnier L, van Meer F, Johnstone AM, Crabtree D, Buosi W, et al. (2018) Effects of hunger state on the brain responses to food cues across the life span. NeuroImage 171: 246-255.
47. Blackmore H, Hidrio C, Yeomans MR (2022) How sensory and hedonic expectations shape perceived properties of regular and non-alcoholic beer. Food Quality and Preference 99: 1-47.
48. Speed LJ, Papies EK, Majid A (2023) Mental simulation across sensory modalities predicts attractiveness of food concepts. Journal of Experimental Psychology: Applied 29: 557571.
49. Barsalou LW (2008) Grounded Cognition. Annual Review of Psychology 59: 617-645.
50. Bettiga D, Lamberti L (2018) Exploring the role of anticipated emotions in product adoption and usage. Journal of Consumer Marketing 35: 300-316.
51. Schendan HE, Ganis G (2012) Electrophysiological potentials reveal cortical mechanisms for mental imagery, mental simulation, and grounded (embodied) cognition. Frontiers in Psychology 3: 1-22.
52. Gallese V, Sinigaglia C (2011) What is so special about embodied simulation?. Trends in Cognitive Sciences 15: 512519.
53. Epstude K, Scholl A, Roese NJ (2016) Prefactual thoughts: Mental simulations about what might happen. Review of General Psychology 20: 48-56.
54. Zeng M, Yan WY, Zeng ZJ (2023) Analysis of consumers’ willingness to pay for honey in China. Sustainability 15: 1-10.
55. Del Prete M, Samoggia A (2020) Chocolate consumption and purchasing behaviour review: Research issues and insights for future research. Sustainability 12: 1-17.
56. Bärebring L, Palmqvist M, Winkvist A, Augustin H (2020) Gender differences in perceived food healthiness and food avoidance in a Swedish population-based survey: A cross sectional study. Nutrition Journal 19: 140.
57. Bogdanov S, Jurendic T, Sieber R, Gallmann P (2008) Honey for nutrition and health: A review. Journal of the American College of Nutrition 27: 677-689.
58. Zanchini R, Blanc S, Pippinato L, Di Vita G, Brun F (2022) Consumers’ attitude towards honey consumption for its health benefits: First insights from an econometric approach. British Food Journal 124: 4372-4386.
59. Šedík P, Kňazovická V, Horská E, Kačániová M (2018) Consumer sensory evaluation of honey across age cohorts in Slovakia. Potravinarstvo Slovak Journal of Food Sciences 12: 673-679.
60. Żak N (2017) Honey market in the opinion of young consumers. Handel Wewnętrzny 366: 424-438.
61. Wong A, Young DA, Emmanouil DE, Wong LM, Waters AR, et al. (2013) Raisins and Oral Health. Journal of Food Science 78: 26-29.
62. Brown K, McIlveen H, Strugnell C (2000) Nutritional awareness and food preferences of young consumers. Nutrition & Food Science 30: 230-235.