Intrinsic and Extrinsic Cue Words of Locally Grown Food Menu Items and Consumers’ Choice at Hyper-Local Restaurants: An Eye-Tracking Study

Abstract

This research examined whether intrinsic (fresh) and extrinsic (local) cue words of locally grown food menu items were associated with consumers’ visual attention (i.e., fixation count and total fixation duration) and menu choice using an eye-tracking methodology and an online survey. Fifty consumers were randomly assigned to read a hyper-local restaurant menu that featured four items (with or without intrinsic or extrinsic cue words) while being eye-tracked. The participants of this study showed a somewhat favorable attitude towards locally grown food though their behaviors were not incongruent with their attitudes. The results showed that there was a relationship between fixation counts and the extrinsic cue word, local, for subsequent menu choices in the way that participants who chose the menu item with the word local appeared to look at it more frequently before making their final menu selection. The participants of this study further identified that factors such as attractive menu descriptions, personal preference, and healthfulness were important when making menu selections. The findings of this study suggested that restaurant operators could use well-crafted descriptions with the word local in their menu, as this approach could influence consumers’ menu selections.

1. Introduction

Consumers’ interest in locally grown food has increased rapidly in recent years, as the “locavore” movement in the United States continues [1]. The word locavore is defined as “a person who makes an effort to eat food that is grown, raised, or produced locally, usually within 100 miles of home” [2]. Data showed that more than 20% of consumers in a survey reported eating locally grown foods twice per week [1]. Moreover, hyper-local food sourcing, which means that the restaurants grow the food in their own gardens or make their products in-house [3], has been a leading trend since 2012 [4]. Many restaurants have been including local food in their menu to take advantage of this trend [1].

Consumers associate locally grown products with sustainable consumption, meaning that their food-consumption behaviors align with the triple bottom line, a business structure that promotes environmental, economic, and social responsibility [5,6]. For example, consumers believe that locally grown products are more energy efficient and can reduce carbon footprints and greenhouse gas emissions [5]. They also want to support the local economy through their purchases and changes in their consumption behaviors [5,6]. Further, the entrepreneurial activities associated with the increased demand for, and consumption of, locally grown food can invite more investment, create more local job opportunities, reduce food imports, and strengthen local food systems, thus supporting the sustainable development of the local community [6,7]. The consumer’s interest in purchasing locally grown food is also shaped by the perception that these ingredients are perceived as being fresher, thereby yielding better-tasting, higher quality meals [8]. More specifically, consumers’ positive perceptions of, and purchase intentions for, locally grown foods are influenced by both intrinsic and extrinsic cues.

Extrinsic cues are the items that somehow relate to the product but are not part of the product itself. Examples of extrinsic cues include branding, product labeling, information available at the point of sale, product information, packaging, region of origin, and price [9]. Additionally, consumers have recognized and associated the value of the region of origin (local) with locally grown food [10,11,12,13,14]. On the other hand, intrinsic cues are defined as the inherent attributes of a product, such as freshness, appearance, color, odor, taste, and texture [15,16,17,18,19]. Among these characteristics, freshness was recognized as an attribute that consumers often acknowledge when it comes to local food [15,16,17,18,19]. Hyper-local restaurants can potentially leverage the use of these extrinsic and intrinsic cues in their menus to strategically promote menus with locally grown ingredients. For example, previous studies in the literature showed that restaurants could use carbon labels (a form of extrinsic cue) to encourage consumers to purchase more environmentally friendly food options [20]. It may also be possible for restaurants to use certain cue words to effectively highlight a locally grown menu item since words could shape customers’ knowledge and perceptions [21]. However, the list of cue words for locally grown food is exhaustive [9,15,16,17,18,19]. This study chose the word local to represent an extrinsic cue and the word fresh to represent an intrinsic cue, as these are frequently mentioned attributes of locally grown food, based on a review of the literature [9,10,11,12,15,16,17,18,19]. Therefore, two questions that remain to be answered are (1) are these extrinsic and intrinsic cue words effective in communicating about locally grown menus? and (2) do these extrinsic and intrinsic cue words influence subsequent menu choice? These cue words can only be recognized as effective if customers pay attention to them.

Most of the previous research related to local products used observations or surveys to investigate the preferences, attitudes, and purchase intentions of consumers [10,22,23,24,25,26], which are prone to self-reporting biases [27]. Hence, Wedel and Pieters [28] highlighted the importance of capturing the visual attention of consumers by using an approach such as eye-tracking to understand the focus of the visual attention of consumers during their decision-making process [28]. The advantages of using an eye tracker in research include its ability to provide objective and unbiased data and its ability to detect where, and for how long, the viewers are actually searching and focusing on something [29]. Food and sustainability research has utilized eye-tracking as a method to capture consumers’ visual attention to products and services [20,29,30,31,32,33,34,35], but its application with regard to locally grown food has been limited.

Therefore, this study fills the previously mentioned research gaps by combining eye-tracking and survey methods to investigate consumers’ menu choices in response to intrinsic (fresh) and extrinsic (local) cue words on the menu of a hyper-local restaurant. The specific objectives of this study were to: (1) describe participants’ purchase behavior and attitudes towards locally grown food; (2) explore words associated with locally grown food and hyper-local; and (3) examine how the intrinsic (fresh) and extrinsic (local) cue words of locally grown food menu items are associated with consumers’ visual attention; total fixation durations (TFD); the number of fixations (FC); and subsequent menu choices from hyper-local restaurant menus. This current study contributes to the sustainable and locally grown food literature by showing how the display of extrinsic and intrinsic cue words in menu descriptions relates to consumers’ menu choices at restaurants.

2. Literature Review

2.1. Sustainable Consumption and Locally Grown Food

There are several benefits related to sustainable consumption and locally grown food. According to the National Park Service, food transportation by planes, trucks, and ships emits nearly 250,000 tons of greenhouse gasses into the atmosphere each year. Using locally grown produce and food products in their business establishments, retailers can reduce their carbon footprint [36]. The U.S. Department of Energy stated that food in the United States travels, on average, 1500 to 2500 miles from farm to table. Furthermore, the U.S. Department of Energy communicated that carbon emissions caused by food transport can be reduced by consumers buying food locally [37].

Previous studies in the literature on locally grown food also addressed consumers’ ability to connect sustainability with locally grown food [5,38,39,40,41]. According to Mirosa and Lawson [38], one of the societal motivations for consumers to purchase local food rests in the perception of environmental sustainability (e.g., fewer food miles from farm to market required). Likewise, Yue and Tong [39] researched the importance of various characteristics on consumers’ decisions to buy fresh, locally grown food. For 61% of participants, being environmentally friendly was important for various reasons in their decision to buy fresh, locally grown food. Furthermore, Kovacs et al. [40] studied the importance of food attributes and motivators for young local-food consumers in Hungary. Consumers from their study also mentioned that one of the characteristics of local products is environmental friendliness. More recently, Nakajima [41] researched the demand for local produce in the context of sustainable food consumption in Singapore. Nakajima [41] stated that, although a large majority of Singaporeans support the environment and think that purchasing locally grown food is more environmentally friendly, there is still a dearth of demand for local produce. Based on her research, Nakajima proposed that it may be effective to educate the public about the sustainability factor of local produce in order to improve the willingness-to-pay (WTP) for local produce [41].

Definitions of local food are essential to better understand the context of this research; however, a legally or internationally recognized definition of local food does not exist [42]. For example, a study of the attitudes and preferences of consumers concerning local food by Aprile et al. [10] found no single, commonly recognized definition of local food from their respondents in Naples, Italy. In general, local food is defined as food items that travel 400 miles or less from their original place or the state where it is produced [42]. The Canadian Food Inspection Agency endorses an interim policy identifying local as “food produced in the province or territory in which it is sold, or food sold across provincial borders within 50 km of the originating province or territory” [43]. Similarly, Canadian university students utilized physical distance to define locally grown food, including the terms ‘100 km’ and ‘province-wide’ (e.g., Ontario) [44]. In conclusion, the two definitions are similarly based on the travel distance, although the distances vary.

2.2. Intrinsic and Extrinsic Cues

Intrinsic and extrinsic cues were introduced as terms and defined to explain their relationship with locally grown food. Consumers purchase locally grown food for a multitude of reasons. A variety of intrinsic and extrinsic cues are used by consumers to evaluate the quality of food products [45,46,47,48]. The inherent attributes of a product, including its appearance, color, odor, taste, and texture, define intrinsic cues. Consumers can appraise these inherent attributes objectively before and after trying food products [9].

In contrast, lower level cues, also known as extrinsic cues, include branding, provided information at the point of sale, information on the package, and price [9]. The consumer’s selection of a product is influenced and strengthened by these extrinsic cues [49]. Previous research on locally grown food found that intrinsic cues such as freshness [15,16,17,19,50], and extrinsic cues such as health benefit [10,19,26] and region of origin (local) [10,11,12,16,23], develop positive perceptions and influence consumers’ purchase intentions.

2.2.1. Intrinsic Cue (Freshness) of Locally Grown Food

Consumers were shown to acknowledge the freshness of local food in previous research [15,16,17,18,19,39,41]. One of the early studies by Bruhn et al. [15] examined the perceptions of consumers on locally grown food at major grocery stores in two communities in California. Bruhn et al. [15] discovered that support for local agriculture and the selling of fresher produce in supermarkets were the main reasons why consumers buy locally grown food. Likewise, Feagan et al. [17] studied the motivations of consumers who support farmers’ markets in the Niagara region of Ontario, Canada. Feagan et al. [17] found that the availability of fresh food was the most common reason for consumers to shop at these markets. A study by Yue and Tong [39] on the importance of various characteristics in consumers’ decisions to buy fresh, locally grown food also highlighted the significance of freshness. Eighty-three percent of participants from their study rated freshness as “very important” for fresh, locally grown food. In the UK, Chambers et al. [16] used a qualitative (focus group) method to analyze the perceptions and behaviors of consumers on local, national, and imported foods. According to their findings, local food was perceived as being fresher than imported foods to the consumers due to the food traveling a shorter distance [16]. Based on a study conducted in Singapore, Nakajima [41] indicated that what mainly impedes the demand for local produce among consumers who have positive perceptions of sustainability is not the price of a product nor the incomes of consumers. Rather, Nakajima found that product-specific factors (e.g., freshness and produce quality), as well as ease of identification at the store were positively related to locally grown food purchases. In general, the freshness of local food is important to consumers, based on the evidence of these previous studies. Therefore, this research observed the visual attention of consumers to the intrinsic cue word, fresh.

2.2.2. Extrinsic Cue (Region of Origin) of Locally Grown Food

Researchers regularly study country of origin information and the brand names of products, as they are influential extrinsic cues [51,52,53,54]. For instance, Lawley et al. [9] observed the role and relationship of intrinsic and extrinsic cues, and their comparative relevance and effects on the consumption of fish products by Australian consumers. Lawley et al. discovered that the extrinsic cue, country of origin (e.g., Australian-grown), was especially used as a substitute indicator by respondents to assess the quality of fish [9]. The region of origin (local) has replaced country of origin or brand names in the current study, as previous findings have shown that consumers recognize the value of region of origin (local) [10,11,12,13,14]. For example, an early study by Eastwood et al. [11] showed that among produce products, tomatoes were the most frequently purchased produce, and the region of origin was the greatest concern for consumers in Knox County, TN, United States [10]. Research by Gallons et al. [12] on consumers’ characteristics at direct markets, such as the roadside stand, farmers market, tailgate market, and pick-your-own farm in Delaware, United States, discovered that 92.9% of respondents would shop for state-certified produce. Furthermore, 75.2% of respondents preferred produce with the state logo compared to ones without the logo; locally grown was rated as a very important reason, as indicated by 49% of the respondents who bought from these markets [12]. Likewise, 74.4% of respondents in a study by Patterson et al. [13] preferred produce from Arizona, United States compared to produce from other regions.

In the case of Europe, Vieregge et al., [14] examined the perceptions of consumers on the use of locally grown foods in the standard menu items at McDonald’s in Switzerland. They found that 67.4% of respondents in general preferred the use of local produce [14]. In a study based in Naples, Italy, Aprile et al. [10] stated that 95% of respondents attempted to locate and purchase local food. In conclusion, these findings may confirm that the local origin of food (local) has been established as being crucial to consumers. In order to clarify that the extrinsic cue, region of origin (local), is valid enough to be tested in this research, a discussion of the country of origin labelling regulations is necessary. As mentioned previously in the introduction, there is no requirement for restaurants to include such information, based on a review of menu labelling requirements. Thus, the region of origin (local) is, as a construct, valid enough to be tested. Overall, this research observed the visual attention of consumers to the extrinsic cue word local.

2.3. Theoretical Framework

The theory of attention in visual marketing is described by two situations: the determination of attention to visual marketing stimuli reflected in consumers’ eye movements and the influence of attention on the communication effects of direct marketing interest (e.g., memory, preference, and choice) [55]. According to studies in the previous literature, there are two different types of attention: goal-directed attention and stimulus-driven attention [56,57]. Goal-directed attention is influenced by top-down factors, and stimulus-driven attention is influenced by bottom-up factors [55]. Top-down factors originate from the previous conditions and characteristics of consumers (e.g., expectations, prior knowledge, goals, and moods). Bottom-up factors originate from the characteristics or sensory aspects of the stimuli to which consumers are exposed [55]. In order to determine the covert patterns of attention in overt eye movements, two factors of influence are combined. The theory of attention in visual marketing distinguishes between the informativeness of stimuli originating from goal-directed attention (top-down) and the salience of stimuli originating from stimulus-driven attention (bottom-up). The theory identifies that attention and other processes are influenced by a combination (additively or multiplicatively) of the informativeness and salience of objects. Even if consumers are not sensitive to them, highly visible stimuli will still capture their attention. On the other hand, depending on the consumers’ expectations and goals, consumers’ attention to stimuli will be varied, and consumers will discover information different ways. The attention of consumers is independent of the configurations of specific stimuli because each consumer has different goals and expectations. However, the greatest amount of attention is paid when the salient stimuli are informative, and the least amount is paid when the stimuli are not informative. The theory of attention in visual marketing systematically reconciles the different views of attention determinants on informativeness and salience. Furthermore, the theory associates these views with the appropriate visual marketing effects [55].

In marketing research, eye-tracking research grew rapidly because visual attention provides an important channel for determining consumers’ visual stimuli, such as product labelling [58,59]. Therefore, eye-tracking techniques showed excellent potential in accurately evaluating consumer perceptions of food product labels and other visual stimuli [28].

2.4. Top-Down Effect on Visual Attention

2.4.1. Familiarity

Familiarity is the most intensely researched top-down effect on visual attention [55], as experiments conducted in psychology have shown that human vision is biased towards familiar objects [60]. Familiarity has been shown to influence the amount of attention paid to an ad (as a whole) with differential influences on attention to ad objects from previous research [61,62,63,64]. According to Wedel and Pieters [55], familiar ads are easier to process relative to unfamiliar ads, thus familiar ads require less attention. For instance, Pieters et al. [65] analyzed the visual attention of consumers through three repeated exposures to print ads using the eye-tracking method. Pieters et al. [65] found that the total amount of attention paid by consumers declined by more than 50%. Familiarity decreased the amount of attention paid to a brand although it did not influence the amount of attention paid to pictures and increased the amount of attention paid to the text [62]. Clement et al. [66] examined the effect of the design features of jam packages on visual attention. They found no correlation between the visual search patterns of consumers and familiarity with the category of jam products or the specific brand of jams. However, they did discover decreases in gazing time in initial visual attention and in further attention times when consumers had previous experience at the grocery store.

On the other hand, Fenko et al. [67] studied the effect of familiarity on visual attention by means of a health label placed on yogurt products. The researchers found that higher familiarity with the health label did not lead to higher amounts of visual attention being paid to the label. Janiszewski and Warlop [68] determined that, overall, the effects of familiarity on attention are linked to the effects of associative memory. In other words, the learning environment that makes semantic information available from memory is generated by brand conditioning while simultaneously promoting attention to a brand. Brand familiarity is a relevant influential visual factor when considering locally grown food as a brand. In previous research, consumers’ positive perceptions of, and purchase intentions towards, locally grown foods were influenced by both intrinsic (e.g., freshness) and extrinsic cues (e.g., region of origin). However, these cues, and the term hyper-local, were not utilized in the descriptions of menu items of hyper-local restaurants to verify whether consumers’ familiarity of these cue words are associated with consumers’ visual attention (TFD and FC). Therefore, the following hypotheses were proposed:

H1: 

The higher the familiarity with the intrinsic cue word fresh, the greater the amount of visual attention paid to this word on the menu.

H2: 

The higher the familiarity with the extrinsic cue word local, the greater the amount of visual attention paid to this word on the menu.

2.4.2. Visual Attention Influence on Product Choice

Several gazing parameters have been shown to be correlated with choice decisions across various tasks [69,70]. Prior research on eye-tracking revealed that the longer the participants fixated on an item, the more likely they were to choose that item [71]. Pieters and Warlop [61] observed the relationship between consumers’ attention and their brand choice. They discovered that consumers are more prone to selecting the brands they fixated on longer compared to other brands. Likewise, Duerrschmid and Danner [72] examined the relationship between the gazing behavior of respondents and their food selection. Respondents were asked to pick the product most appealing to them from four options presented as pictures. Duerrschmid and Danner [72] discovered a strong correlation between the respondents’ probability of choice and their visual attention (i.e., number of fixations and fixation duration). The likelihood of a product being selected was significantly higher when more visual attention was paid to it [72]. In the context of sustainability label research, Van Loo et al. [73] reported on the importance of sustainability attributes (e.g., Carbon Footprint, Fair Trade, Rainforest Alliance, and USDA Organic) on coffee product labels and their influence on visual attention and the willingness-to-pay of consumers in the United States. Van Loo et al. found that the respondents for whom sustainability labels were of higher importance paid more visual attention to these sustainability-attribute labels. Furthermore, the greater amount of visual attention paid to sustainability labels by respondents led to a higher willingness-to-pay for items with these labels. However, the previous findings are limited to observing the consumers’ visual attention and product labels, so the following hypotheses were proposed:

H3: 

Visual attention to the intrinsic cue word fresh is associated with subsequent menu choice.

H4: 

Visual attention to the extrinsic cue word local is associated with subsequent menu choice.

3. Materials and Methods

3.1. Participants

Before the study, the research protocol was reviewed and approved by the University Research Ethics Board (Protocol #20-270-MR 2006). The participants of this study consisted of the general public, such as faculty members, staff, and graduate and undergraduate students (19 years or older) from a university in the southeastern United States. Pernice and Nielsen [74] recommended a minimum sample size of 39 participants for the analysis of eye-tracking gaze point analysis. Therefore, this study aimed to recruit 50 participants for this research. The participants were recruited through multiple approaches, including flyers posted on bulletin boards throughout the university campus, as well as electronic invitations through the college LISTSERV. Participants who completed this study were also asked to share information about this study with their friends and/or family members. Both the flyer and electronic invitation contained the exact same information, explaining the purpose of the study, eligibility criteria, and expectations of the study. The main recruitment criteria consisted of participants’ general interest in foods (stated on the recruitment invitations) and a minimum age of 19 years. To be eligible, participants were required to have normal or corrected-to-normal vision and full color vision. Interested participants contacted the researchers by email. Once their eligibility was determined, they were scheduled for a 20 min visit to the eye-tracking laboratory to complete the study.

3.2. Study Materials

3.2.1. Restaurant Menu with the Modified Entrée Dinner Menu Items

The stimuli in this study were restaurant menus that featured four entrée dinner menu items, namely Arabica Crusted Steak, Fresh Herb Grilled Chicken, Wood Fire Grilled Salmon, and Local All-Natural Beef Burger. These menu items were selected based on a review of several menus from casual mid-scale restaurants, as well as mid-scale hyper-local restaurants. Two menu items were without intrinsic and extrinsic cues, one menu item had one intrinsic cue word (fresh), and one menu item had one extrinsic cue word (local). An intrinsic cue word (fresh) was chosen as consumers were shown to acknowledge the freshness of local food in previous research [15,16,17,18,19,39,41]. An extrinsic cue word (local) was chosen as previous findings [10,11,12,13,14] have shown that consumers recognize the value of region of origin (local).

Four different restaurant menus (Menu A, B, C, and D) were developed by putting the four menu items in different orders based on the Latin square design (See Appendix A). A Latin square design is the arrangement of t treatments, each one repeated t times, in such a way that each treatment appears exactly one time in each row and each column in the design [75]. This method of design is used to minimize the systemic errors caused by rows (treatments) and columns [75]. Several eye-tracking studies conducted in a similar context (e.g., food labels) also implemented a Latin square design [76]. This method was also chosen to give validity to specific words, rather than to specific locations. For instance, if the intrinsic cue word (fresh) was always listed in the first row at the top and received high fixations, the high fixations result could have been due to the cue location. The Latin square method controlled for other variables that could have affected this study through the use of other designs [76]. Other than the order of the menu items, the design, font, font size, color of the font, and overall layout of all the menus were identical (see Figure 1). To support internal validity, prior to conducting research the menu items and design were verified by two external researchers who have expertise in hospitality research and in eye-tracking methods.

3.2.2. Tobii X2-60 Eye Tracker

The eye-tracking device, a Tobii X2-60 Eye Tracker with 1280 × 1024 pixels, was placed at the base of a 17” computer monitor to record the participants’ eye movements. The Tobii X2–60 is a lightweight eye tracker with a length of 184 mm (7.2 in.). It is a discreet eye tracker that can be used to provide in-depth analysis of natural visual behavior. Also, its wide range of head movements enable the subject to move while preserving accuracy and precision during recording. Portable eye-tracking labs and studies involving an eye tracker to monitor even large objects at close range (up to 36° gaze angle) are feasible. With a variety of software and stimuli setup choices, the Tobii X2 Eye Tracker provides full flexibility [77] and was thus chosen for use in this study.

3.3. Measures

3.3.1. Eye-Tracking

Eye-tracking data, in general, consist of the number of fixations and fixation durations (in seconds) on the Areas of Interest (AOI) [78,79]. This research measured the FC and the TFD of participants on the AOI by using the Tobii X2-60 Eye Tracker. The number of fixations measures how many times participants fixated on an AOI [67], while the fixation duration(s) measures the total durations participants fixated on an AOI [67]. This research focused on measuring consumers’ visual attention while reading menu items. Thus, the ‘minimum fixation duration’ parameter is set to the default value to allow for relatively short fixations (60 ms), which are common during readings [80]. This means that the duration of every fixation is compared to the ‘minimum fixation duration’ parameter independently. The fixation is recategorized as an unknown eye movement when the duration is less than the parameter value. The AOI examined in this research included menu items with the cue words fresh and local (Figure 2). The heat map, which provides an intuitive visualization of the eye-tracking data, is commonly used to depict fixation durations on the AOI. The green spots represent shorter fixation durations, and the red spots represent longer durations. Figure 3 depicts an example of a heat map created during an eye-tracking session.

3.3.2. Online Qualtrics Survey

The post-study online Qualtrics questionnaires consisted of 18 items. Five items were used to gather the demographics of the participants, including their gender, age, ethnicity, work status, and income level. Three items were included to understand participants’ locally grown food shopping and dining behaviors, including the frequency of locally grown food purchases and the locations where the items were purchased, as well as the frequency with which participants dined out at a hyper-local restaurant. Four items were used to investigate participants’ attitudes towards locally grown foods (e.g., “Buying locally grown food is beneficial and a wise choice” and “Buying locally grown food is beneficial to the environment”). Familiarity (4 items) with the extrinsic (local), intrinsic (fresh) cue words, as well as the concepts of locally grown food and hyper-local were measured on a 7-point Likert scale, with 1 being “strongly disagree” and 7 being “strongly agree”. Each participant was also asked to provide three words they could relate to “locally grown food” and the word “hyper-local” within two open-ended questions.

3.4. Procedure

3.4.1. Data Collection

The participants were asked to come to the eye-tracking lab on the specific date and time they had signed up for. Each participant received a reminder email from the researcher 24 h in advance. The following standard procedures were conducted in each experimental session. First, the participants were greeted at the door of the lab and were invited to be seated in front of a computer equipped with an eye tracker. Next, participants were instructed to read and sign, at their discretion, an informed consent form that presented a summary of the research, foreseeable risks, expected benefits, precautions taken to ensure data confidentiality and anonymity, and the voluntary nature of the study. Third, the participants completed a calibration task in front of the computer screen to adjust their eyes to the eye tracker. Participants’ eye movements followed a red dot across the computer screen to complete this task. Fourth, the participants began the formal study by reading a given scenario on a computer screen. The scenario informed the participants that they were dining at a hyper-local restaurant that served locally grown food for dinner. After reading the menu, they were asked to choose one item at the end of an eye-tracking session. All sessions were recorded to collect the TFD and FC on AOI. Fifth, the participants scanned the QR code at the end of the eye-tracking session to complete the post-study online questionnaire from Qualtrics. No time restraint was implemented so that the participants could complete the experiment at their own pace. The entire process took about 20 min. Each participant received $10 (US) as compensation for their time in this study.

3.4.2. Data Analysis

Visual attention data were analyzed using Tobii Studio analysis software (version 1.152). Data from the survey were analyzed using the Statistics Package for Social Sciences (SPSS, version 25.0). Descriptive statistics (e.g., frequency, mean, and standard deviation), were used to describe, for example, the demographics of the participants, their locally grown food purchase behaviors, and familiarity with locally grown food. Inferential statistics, including correlation analysis and point biserial correlations, were used to test the hypotheses. Word clouds were generated using Pro Word Cloud, which is a plug-in application for Microsoft Word to visually represent other words the participants associated with the words local and hyper-local. In order to test H1 and H2, a linear regression was used to verify whether intrinsic (fresh) and extrinsic (local) cue words of locally grown food menu items associated with consumers’ visual attention (TFD and FC). In order to test H3 and H4, point biserial correlation was used to validate whether consumers’ visual attention (TFD and FC) to cue words (fresh and local) was associated with subsequent menu choice.

4. Results

4.1. Participants’ Profile, Attitudes, and Behaviors of Locally Grown Food

Since Pernice and Nielsen [74] recommended a minimum sample size of 39 participants for the analysis of eye-tracking gaze point analysis, this study included 50 participants in the final sample. The participants consisted of an almost equal number of females (n = 26, 52%) and males (n = 24, 48%). The ages of the participants ranged from 19 to 64 years, with a mean age of 30.76 ± 12.26 years. Most of the participants were White or Caucasian (n = 36, 72%), followed by Asian (n = 6, 12%), Black or African (n = 3, 6%), Hispanic or Latino (n = 3, 6%), American Indian or Alaska Native (n = 1, 2%), and other (n = 1, 2%). More than half of the participants classified themselves as undergraduate or graduate students (n = 28, 56%), followed by non-students (n = 22, 44%). The most frequently selected household incomes were below $25,000 (n = 14, 28%) and above $100,000 (n = 14, 28%), followed by $25,000–$49,999 (n = 8, 16%), $50,000–$74,999 (n = 7, 14%), and $75,000–$99,999 (n = 4, 8%). The remaining participants indicated a preference not to answer to this question (n = 3, 6%) (

Table 1. Demographics of the Research Participants (N = 50).

Variable	Category	n (%)
Gender	Female	26 (52.0)
	Male	24 (48.0)
Ethnicity	White or Caucasian	36 (72.0)
	Asian	6 (12.0)
	Black or African	3 (6.00)
	Hispanic or Latino	3 (6.00)
American Indian or Alaska Native		1 (2.00)
	Other: Please specify	1 (2.00)
Current classifications	Students	28 (56.0)
	Non-students	22 (44.0)
Household income	Below $25,000	15 (30.0)
	$100,000 and above	14 (28.0)
	$25,000–$49,999	8 (16.0)
	$50,000–$74,999	7 (14.0)
	$75,000–$99,999	6 (12.0)

).

In the three months preceding the experiment, participants purchased (including grocery pick-up) food labeled as locally grown (e.g., Alabama-grown) an average of 4.08 ± 5.25 times, ranging from 0 to 24 times. Participants were asked where they normally purchased locally grown food (including grocery pick-up). Over half of the participants selected grocery store chains, such as Publix (n = 27, 54%), followed by farmers’ markets (n = 17, 34%). Participants dined out (including curbside pick-up and take-out orders) an average of 1.98 ± 4.47 times (range = 0 to 22 times) at hyper-local restaurants promoting locally grown food in the last 3 months.

Participants’ attitudes to locally grown food were also investigated. The results showed that the participants demonstrated somewhat positive attitudes towards locally grown food, with a total average mean score of 5.21 ± 0.82. Participants somewhat agreed (n = 21, 42%) and strongly agreed (n = 12, 24%) that buying locally grown food instead of non-locally grown food is beneficial to the environment (5.56 ± 1.03). A total of 22 participants (44%) somewhat agreed that buying locally grown food instead of non-locally grown food is a wise choice (5.30 ± 0.953). Close to a third of the participants somewhat agreed (n = 15, 30%) that buying locally grown food instead of non-locally grown food makes them feel good (5.24 ± 1.33).

Overall, the participants were familiar with the word fresh (6.18 ± 1.18), local (6.10 ± 1.00), and the concept of locally grown food (6.04 ± 1.10). However, the participants appeared to be less familiar with the word hyper-local (3.40 ± 1.76), as close to a third (n = 32) of them disagreed that they were familiar with this particular word.

4.2. Words Related to Local and Hyper-Local Foods

Figure 4 represents two different word clouds generated using the keywords provided by the participants to both local and hyper-local, respectively. The size of a word was positively related to the frequency with which this word was mentioned. The larger the size of the word, the higher the frequency with which it was mentioned.

Participants were asked to write three words related to the word local and hyper-local. The words most frequently associated with the word local were “fresh” (n = 38), followed by “organic” (n = 10). Additionally, participants also wrote down farm-related words such as “farm-to-table”, “farmer”, “farm grown”, “farm”, “farmed close”, “farm-raised”. Nine of the words contained “farm” in them. “Healthy” and “expensive” were each written down by six participants. Four participants related the word “support” to the word local, and another four participants further wrote down the word “community”. Participants also associated the word local with distance, for example, “near” (n = 3), proximity (n = 1), and accessibility (n = 1) (See Figure 4).

The words most frequently associated with the word hyper-local were also “fresh” (n = 24); “local” and “tasty” were each stated six times. Participants also associated hyper-local with farmers (n = 5), healthy (n = 5), and organic (n = 5). Hyper-local also made the participants think about “market” (n = 4) and “expensive” (n = 4) (See Figure 4).

4.3. Results of Visual Attention

On average, the FC for the word fresh was 18.54 (10.77) and the FC for local was 19.30 (14.10). The participants spent 4.25 s on the word fresh and 4.47 s on the word local (See

Table 2. Mean number of FCs and TFDs on the menu items with the words fresh and local (N = 50).

AOI	FC Mean (SD)	TFD (Seconds) Mean (SD)
Intrinsic cue word Fresh	18.54 (10.77)	4.35 (2.82)
Extrinsic cue word Local	19.30 (14.10)	4.47 (3.64)

).

4.4. Effect of Familiarity

A linear regression analysis was used to verify the relationship between familiarity with the words fresh and local and visual attention. The results of the linear regression analysis showed that the word fresh did not statistically relate to the FC (β= −0.20, p = 0.16) or TFD (β = −0.18, p = 0.20). Therefore, Hypothesis 1 (H1) was not supported. The same method was applied to investigate the correlation between the word local and visual attention. Once again, the results were statistically insignificant for both the FC (β = −0.17, p = 0.23) and TFD (β = −0.18, p = 0.20). Therefore, Hypothesis 2 was also not supported.

4.5. Visual Attention and Product Choice

A point-biserial correlation analysis was used to verify the relationship between visual attention to the intrinsic cue word fresh and subsequent menu choice. The results of FCs ((r_pb = 0.26, n = 50, p = 0.07) and TFD were statistically insignificant (r_pb = 0.22, n = 50, p = 0.13), indicating that visual attention to the word fresh was not associated with the subsequent menu choice. Therefore, H3 was not supported. Hypothesis 4 (H4) was partially supported as the results of the point biserial correlation indicated that FCs to the extrinsic cue word local were positively associated with subsequent menu choice (r_pb = 0.295, n = 50, p = 0.038) although results of the TFD were insignificant (r_pb = 0.27, n = 50, p = 0.06).

4.6. Menu Choice and Reasons of Selection

Due to the randomization, Menu A and Menu B were each read by 13 participants while Menu C and Menu D were each read by 12 participants. The participants’ menu choices were further examined to understand which menu item was most frequently selected by them and the reasons behind their selection decision.

For the participants who viewed Menu A, only two of them selected the menu item that included the intrinsic cue word fresh (i.e., Fresh Herb Grilled Chicken), and three of them selected the item that included the extrinsic cue word local (i.e., Local All-Natural Beef Burger). For Menu B, only two participants selected the menu item that included the intrinsic cue word fresh, and another two participants selected the item containing the extrinsic cue word local. Most of the participants who viewed Menu B also selected the Arabica Crusted Steak (N = 6). For Menu C, three of the participants selected the menu item that included the intrinsic cue word fresh, and three of the participants selected the item that included the extrinsic cue word local. For Menu D, two participants selected the menu item that included the intrinsic cue word fresh, and another two participants selected the item that included the extrinsic cue word local. In general, it appeared that the menu items containing the words fresh and local were not the most-favored choices of the participants in this study.

The reasons for their choices were further analyzed to gain a deeper understanding of the participants’ menu selection decisions. Overall, only three participants stated that they selected a particular menu item because it was described as local, and three other participants chose a particular menu item because it contained the word fresh. The results of the analysis implied that the descriptions of the menu item which made the food sound appealing and appetizing played a large role in participants’ menu selections. Other reasons included personal preference, healthiness, combinations of ingredients, familiarity with the food, and new food items.

Table 3. Participants’ final menu choices and examples of the reasons.

Final Menu Choices	n (%)	Reasons of Menu Choice Provided by the Participants
Arabica Crusted Steak	19 (38.0)	“The crusted steak sounded the most appealing, I usually try steak out when trying new restaurants. The sides sounded the most appealing as well.” “The ingredients in the Arabica Crusted Steak dish appeared to be the most appetizing.” “The description of the steak made it even more tantalizing to order.” “I like steak a lot, but what drew me to that menu item initially was its name. Then, the description under it made me very hungry, so I opted for that one.” “It sounded like a unique menu item that I haven’t had before. I like to try new things, and I like a variety of flavors.”
Wood Fire Grilled Salmon	12 (24.0)	“I really enjoy salmon but rarely have an opportunity to afford it at a restaurant.” “Without the price of each item, I prefer to eat salmon because I like it most. It also provides the salad.” “I am a very picky eater and fine myself sticking to the basics. The salmon was a bit out of my comfort zone, but I would have been willing to try it. I seemed to like everything that would have come with it or willing to try it.”
Local All-Natural Burger	10 (20.0)	“This selection was very familiar to me, and I knew I could trust it. I know that I have had burgers in the past that I have liked.” “The burger and fries seemed tasty. The other items had unusual ingredients.” “I like meat and like the idea of knowing the source of the meat.” “The burger just sounded good at the time. Also, it said it was local and all natural, so I figured it would be fresh and good quality.”
Fresh Herb Grilled Chicken	9 (18.0)	“I chose the chicken option due to being a lean meat. It also sounded the most appetizing with fresh ingredients.” “I really like chicken to begin with, but I felt like it included a really wide array of vegetables, letting me get different flavors and textures.” The ingredients were the most appealing to me. Pearl onions are not something you see very often, but I like them a lot, so that was one thing that caught my eyes as I read through the menu. The herb chicken sounded the best. I liked how it was very descriptive and sounded the freshest/lighter.

illustrates the participants’ final menu choices and examples of the reasons provided by the participants when explaining their menu choice.

5. Discussion

The main objective of the research was to investigate the relative effects of various intrinsic and extrinsic cue words on consumers’ visual attention and purchase intentions by using the eye-tracking method. Overall, the participants of this study did not appear to frequently purchase locally grown food or dine out at restaurants that serve locally grown products. Even so, their attitudes towards locally grown food were positive, and were consistent with previous studies in the literature [81,82,83,84].

The participants most frequently linked the word local with “fresh” (n = 38). This finding was consistent with previous research that indicated consumers acknowledged freshness as one of the characteristics of locally grown food [15,16,17,18,19,39,41], and they perceived locally grown as being fresher and healthier [16]. Although the findings of this study showed that participants could identify the word fresh with locally grown food, the results of the linear regression showed that high familiarity did not translate to higher amounts of attention being paid to this word. Similarly, a study by Wedel and Pieters [55] found that familiar ads were easier to process relative to unfamiliar ads, indicating that familiar ads needed and attracted less attention. The findings of this study are also consistent with Clement et al. [66], who found no correlation between the visual search patterns of consumers and familiarity with a category of jam products or a specific brand of jams. Likewise, Fenko et al. [67] studied the effect of familiarity, as it related to visual attention, with a health label placed on yogurt products. They found that higher familiarity with the health label did not lead to higher amounts of visual attention being paid to this label. As a result, it appears that consumers’ familiarity with these words did not translate into higher amounts of visual attention in the study.

The results of this study further revealed that there was no relationship between visual attention (TFD and FC) to the intrinsic cue word fresh and subsequent menu choice. Likewise, there was no relationship between visual attention (TFD) to the extrinsic cue word local and subsequent menu choice. However, the proposed relationship between visual attention (FC) to the extrinsic cue word local and subsequent menu choice was supported. Participants who chose the menu item with the word local seemed to look at it more frequently before making their final menu choice. This finding was consistent with previous research, as the likelihood of a product being selected was significantly higher when more visual attention was paid to it [72]. Likewise, Van Loo et al.’s [73] study showed that respondents who placed importance on sustainability labels paid more visual attention to these labels. These results imply that visual attention is the beginning of the information acquisition process leading to selection decision [73].

In general, the cue words local and fresh do not seem to be the main reasons determining menu item selections. Only three participants stated that they selected a particular menu item because it was described as local, and three other participants chose a particular menu item because it contained the word fresh. These findings are inconsistent with studies in the previous literature on locally grown food. A study conducted by Kim, Rahman, and Bernard [85] showed that patrons at hyper-local restaurants in the UK and US stated the importance of the region of origin (local) on their online restaurant reviews. Also, Baiomy, Jones, and Goode’s [86] research recommended that restaurants increase their identification of local products to attract customers. Nakajima’s [41] study found that product-specific factors (e.g., freshness and produce quality, as well as ease of identification at the store) were positively related to locally grown food purchases. In this research, participants somewhat agreed and strongly agreed that buying locally grown food instead of non-locally grown food is beneficial to the environment. Although the participants of this study perceived the environmental benefit of locally grown food, product-specific factors, such as the freshness highlighted in Nakajima’s [41] study, did not appear to influence their menu choice. Overall, it is worthwhile to note that the participants of this study may not represent avid consumers of locally grown products, based on their reported purchasing behavior of locally grown food.

A more detailed analysis of why certain menu items was chosen by the participants uncovered several key factors that were associated with consumers’ menu choice. Menu item descriptions that made the food sound appealing and appetizing played a large role in participants’ menu selections. This was made evident in the study as steak, which had a vivid menu description, was chosen the most often. Previous studies in the literature showed that consumers may like or dislike a dish based on its description [87]. A detailed menu description could entice customers and increase their perception of food quality and satisfaction [86,88]. It is possible that the insignificance of the results of those intrinsic and extrinsic cue words may be because participants were distracted by superfluous descriptions. Thus, this study could serve as a preliminary study for more comprehensive research in the future.

Other reasons, such as personal preference, healthfulness, ingredients, familiarity with the food, combinations of ingredients, and interest in trying new food items also affected menu selection. These reasons have been stated in previous studies as being related to customers’ menu choice [89,90]. For example, Peters and Remaud [89] indicated that restaurants should emphasize combinations of ingredients rather than solely focusing on sustainable produce. Regarding healthfulness, Lee and Carnage [90] emphasized the importance of nutritional aspects on menu items as it was one of the most important attributes influencing consumers’ menu selections. Overall, the findings from this study supported these studies.

Implications

Based on the major findings from this research, some practical implications are provided for restaurateurs who serve locally grown food. There was a relationship between the (FC) of the extrinsic cue word local and subsequent menu choice. Participants who selected the menu item with the word local appeared to have looked at it more frequently before making their final menu selection. This might indicate the importance of the presence of the extrinsic cue word local in menu items. This study further provided two additional pieces of evidence to support the significance of the word local as an extrinsic cue. First, three participants who selected ‘Local All-Natural Burger’ as their menu choice specifically mentioned local as being the reason for their choice. Second, results from the analysis of words showcased local as being frequently associated with hyper-local. Combining these findings together, restaurants wanting to promote and market locally grown food could consider including local as part of the name of the menu item or as part of the item description.

In the current research, many participants picked steak because the description made it look appealing. Therefore, it is important to note that descriptions of food items play an important role in menu selection, as well-crafted descriptions increase the likelihood of a menu item being selected. Based on this, restaurant owners who use proper descriptions with the word local can expect to increase menu item exposure and sales. Additionally, other descriptive words, such as fresh, organic, healthy, support, and community that are closely related to the word local can be used together in menu descriptions. It has been shown that a story marketing strategy can be applied to menu descriptions to better promote a product. Story marketing is defined as a strategic marketing tactic that uses audio, visual, and immersive storytelling to generate a brand experience that puts the consumer at the core of the story, resulting in profitable interactions [91]. For instance, restaurants can define what the word local means specifically for their operations. For example, The Shed restaurant located in London, UK, incorporates story marketing in their menu descriptions. The Shed’s menu states, “We use all things wild, foraged, and locally grown, including sustainable livestock from the Gladwin’s family farm in West Sussex–we call this local and wild” [92]. Incorporating such messages visually in the menu would enable restaurants to provide a better brand experience for their customers, as customers can connect with the restaurants’ commitments and values. As a result, restaurants can expect a potential increase in sales for items with locally grown food ingredients.

Policymakers can also contribute to promoting hyper-local restaurants in the following ways. For a policy implication, each city or province can host a restaurant week to promote hyper-local restaurants in its community. For example, the city of Birmingham, Alabama, in the United States, hosts Birmingham Restaurant Week (BRW) every summer for ten days [93]. By encouraging attendees to order delicious local food and drinks as well as to support Birmingham’s exceptional restaurants and bars, this event aims to support the local culinary scene and community. The participating establishments will provide prix fixe breakfast, lunch, and/or dinner options, along with specials on beer, wine, and cocktails. Participation is limited to locally owned and operated eateries, coffee shops, bars, and food trucks [93]. Implementing this local event may help hyper-local restaurants to communicate the characteristics of locally grown food (e.g., freshness and localness) actively to their consumers. For an educational implication, university communities can host farmers’ markets with food trucks that offer locally grown food. Nakajama [41] suggested that it may be effective to educate the public about the sustainability factor of local produce in order to improve the WTP for local produce. For example, some universities in the United States are hosting farmers’ markets on their campuses, including Auburn University, the University of Michigan, and the University of Southern California. Hosting farmers’ markets at universities will be an effective way to educate a variety of consumers on locally grown foods, as personal interactions between vendors and consumers will take place.

Theoretically, this research used familiarity with the intrinsic (fresh) and extrinsic (local) cue words of locally grown food to guide top-down attention. Top-down attention is related to the subject’s prior knowledge and experience, which was translated to familiarity with the cue words examined [60]. However, the insignificant findings of Hypotheses 1 and 2 revealed that attention was not directed towards cue words that were shown on a restaurant menu featuring locally grown food. In this case, bottom-up attention, which highlights the characteristics or sensory aspects of menu items, may be the first step in catching the attention of restaurant customers, which subsequently leads to their menu choice. These findings further complemented the data from open-ended questions that alluded to the importance of having good descriptive terms for menu items.

Over the different tasks, it has been shown that a variety of gazing parameters are correlated with choice decisions [69,70]. However, Hypothesis 4 was partially supported as the results of the point biserial correlation indicated that only FCs to the extrinsic cue word local were positively associated with subsequent menu choices. This suggests that FCs and TFD may be perceived differently to customers when it comes to reading a menu, as the duration of fixation on cue words is not as important as the frequency of fixation. Therefore, the results of FCs and TFD may be different depending on the reading materials, which calls for more future research for verification.

6. Limitations

This study has several limitations. First, the samples obtained for the current research were recruited from the southeast of the United States. Samples from different regions might have different perspectives on tested cue words. Thus, the findings may not be generalizable beyond this geographical location. Second, this research observed the visual attention of consumers to the words fresh, local, and hyper-local. While consumers recognized the words fresh [15,16,17,18,19,39,41] and local [10,11,12,13,14] as important factors, other words written down by the participants, such as healthy, organic, and price, might attract consumers’ attention as well. Therefore, future studies should expand the scope of this research by measuring the TFDs and FCs of words more familiar to consumers. Additionally, no pilot study was conducted to test the intrinsic and extrinsic cue words; rather, these words were chosen based on studies in the previous literature. Incorporating a pilot study could have verified if those cue words were eye-catching for the participants. It also appeared that the participants were not familiar with the word hyper-local. Therefore, asking them to provide three words associated with this concept might not be an accurate reflection of how well they understood this concept. Hence, the survey instrument can be improved by defining this term to the participants before the open-ended question is shown. This study did not consider other confounding factors that could have affected consumer food choice selection, such as personal preference and health. Fifth, there was a limited number of menu items (four items) used in this research. As this was exploratory research, more menu items can be included in the future. Lastly, the Tobii X2-60 Eye Tracker used in this research was attached to the computer to collect the TFDs and FCs. As a result, findings may not be reflective of real customer behavior when using other devices such as mobile phones, tablets, and laptops. Moreover, the study was conducted in a lab setting and was not a complete reflection of a real commercial restaurant. It is recommended that future studies be conducted in a restaurant to be more realistic.