**Simone Mancini 1,\*, Giovanni Sogari 2, Davide Menozzi 2, Roberta Nuvoloni 1,3, Beatrice Torracca 1, Roberta Moruzzo <sup>1</sup> and Gisella Paci 1,3**


Received: 26 June 2019; Accepted: 18 July 2019; Published: 19 July 2019

**Abstract:** This study provides a framework of the factors predicting the intention of eating an insect-based product. As part of the study, a seminar was carried out to explore how the provision of information about ecological, health, and gastronomic aspects of entomophagy would modify consumer beliefs regarding insects as food. Before and after the informative seminar, two questionnaires about sociodemographic attributes and beliefs about the consumption of insects as food were given. Participants were then asked to carry out a sensory evaluation of two identical bread samples, but one was claimed to be supplemented with insect powder. Results showed that perceived behavioral control is the main predictor of the intention, followed by neophobia and personal insect food rejection. The disgust factor significantly decreased after the participants attended the informative seminar. Sensory scores highlighted that participants gave "insect-labelled" samples higher scores for flavor, texture, and overall liking, nevertheless, participants indicated that they were less likely to use the "insect-labelled" bread in the future. Our findings provide a better understanding of insect food rejection behavior and help to predict the willingness to try insect-based products based on some important individual traits and information.

**Keywords:** entomophagy; novel food; neophobia; disgust; edible insects

#### **1. Introduction**

In Europe, the habit of eating insects, or entomophagy, has not become widespread, with the exception of some specific and local cases [1,2]. Traditionally, most people in Western countries consider insects to be a threat and a health risk rather than a food source [3], despite the fact that insect protein and fat contents, mineral contents, and amino acids profiles confirm that they are a good source of nutrients [4].

Most of the studies carried out in Europe about consumers' acceptance of insects as food illustrate a low consumer willingness to try eating insects [5], with the exception of the Netherlands, Belgium, and few other Northern countries, where some insect-based products have been commercially available in the past few years [6–8]. Particularly, in urban and Western societies, insects are rarely eaten or their consumption is perceived to be culturally inappropriate [2]. As reported by Deroy et al. [9], one of the greatest barriers to adopting entomophagy is cognitive disgust linked to the perception that insects are related to undesirable thoughts such as dirt, death, disease, and contamination [10]. Only a few works have investigated how potential consumers from Southern European countries might respond to entomophagy [11].

The theory of planned behavior (TPB) [12], has been proven to be a proper theoretical framework for understanding sustainable and ethical consumer behaviors concerning food [13,14]. According to TPB, the perceived behavioral control (PBC), along with subjective norms (SN) and attitude, impact a person's intention. Perceived behavioral control is the measure of perceived control over the behavior (i.e., how easy or difficult performing the behavior will be). Perceived behavioral control can also have a direct impact on behavior because performance of a behavior not only depends on motivation, but also the individual control of the behavior. If an individual has limited control over an activity, the activity might not be implemented, even in the presence of strong motivational factors.

Menozzi et al. [15], examined whether the TPB could be employed to understand young adults' behavior when faced with the prospect of eating products containing insect flour, and suggested that attitude and, to a lesser extent, PBC play a significant role in affecting the intention of performing the behavior, while the SN is not a significant factor in forming the behavioral intention. This result is common in several other studies [16], and the SN is generally a weak predictor of intentions [17].

From another point of view, when a new food product is introduced into a culture, it generally induces feelings of fear and refusal that is called food neophobia [18], which refers to the unwillingness to try and the tendency to avoid novel food [19].

Food neophobia is expected to reduce the likelihood of readiness to incorporate insects into the diet [20–22]. Pliner and Hobden [19] developed an instrument to measure food neophobia which consisted of ten statements, five positively worded (neophilic) and five negatively worded (neophobic), and was called a food neophobia scale (FNS).

Recently, many studies have indicated that food neophobia, as an individual trait, is one of the most important predictors for understanding consumer willingness to try insects [22–25]. The relationship between entomophagy and food neophobia may depend on the knowledge of their origin and habitat or may be based on negative consequences following their consumption [26].

Food exposure increases product familiarity and influences the willingness to accept new food, and thus reduces neophobic reactions [21,27–29]. At the same time, information is a factor that plays a role in consumers' acceptance [30,31] and can be used to promote entomophagy [22]. Information about insects, such as production technology, safety risk, and scientific-like information [32,33], tailored to consumers can modify their perception of items, both novel and familiar.

Despite the huge interest in factors specifically related to the intention of eating insects, there is a lack of research that investigates the link between the mechanism of rejection of insects as food (both at a personal and social level), the level of food neophobia for unfamiliar food, and the perceived behavior control of introducing insects to the diet.

On the basis of these assumptions, the first aim of this study was to evaluate the willingness of consumers to adopt insects as food and to investigate the main factors (e.g., sociodemographic variables, food neophobia, and other behavioral constructs) that affect the intention to eat insects. The second aim was to investigate whether and how information can influence the willingness to taste insects and their acceptability. This was achieved by comparing consumers' willingness to adopt insects before and after a seminar on insects, and investigating how an informative seminar on the benefits of including insects in human diets influenced the consumer acceptance. In addition, an analysis about the perception of the sensory properties of two identical breads ("insect-labelled" vs. "no insect-labelled") was performed.

#### **2. Material and Methods**

#### *2.1. Theoretical Model*

The model includes two concepts, food neophobia and insect food rejection, with the introduction of some elements of the TPB.

In accordance with previous studies, given the premises mentioned in the introduction, we tested the following hypotheses (Figure 1).

**Figure 1.** The conceptual model corresponding to the current research. Our study tested the following hypotheses (H): H1, food neophobia will positively affect rejection factors of insect as food (e.g., disgust) [10]; H2, food neophobia and the perceived behavioral control (PBC) will be correlated; H3, PBC will have a significant positive effect on the intention of eating an insect-based food in the coming months [15]; H4, a higher level of food neophobia will have a strong negative effect on the behavioral intention of eating an insect-based food in the coming months [21]; H5, a higher level of insect food rejection will have a strong negative effect on the intention of eating an insect-based food in the coming months; H6, demographic attributes will have a moderate effect on the intention of eating an insect-based food in the coming months [34]; H7, food neophobia will negatively affect the willingness to eat an insect-based product; H8, intention of eating an insect-based food in the next months will be a strong predictor of the willingness to eat an insect-based product [15].

#### *2.2. Participants and Data Collection*

The survey was conducted at the University of Pisa (Italy) in March 2018. Participants were students attending a seminar entitled "Insects as Food and Feed: Future Prospects". The three-hour seminar was structured as oral presentations from different speakers with visual support. The aim of the seminar was to describe benefits, disadvantages, and future prospects of insects as food and feed. In particular, aspects were delivered related to species and breeding technologies, food safety and regulatory concerns, and consumers' approach.

A total of 165 students with a bachelor's degree and a master's degree participated in this study. They did not receive monetary compensation for their participation. Prior to the seminar session, participants were informed about the questionnaires and an informed consent form was signed.

Our protocol was based on previous studies where an informative session about entomophagy was held among university students followed by an insect tasting session [34,35]. Two questionnaires were proposed chronologically to the students: one before (questionnaire 1, Q1) and another after (questionnaire 2, Q2) the seminar. Furthermore, at the end of Q2, the participants were invited to take part in a tasting session immediately after the seminar ended.

#### *2.3. Questionnaire Design and Measures*

The structure of the questionnaire for both Q1 and Q2 was based on previous study designs [15,34]. The first part of the questionnaire (Q1) included three questions concerning the demographic profile of the respondents (i.e., age, gender, and Italian region of origin) and 20 individual statements, divided into two different sections. Section A included 10 items about a person's tendency to reject or avoid eating unfamiliar food (neophobia). Section B, which followed A, included 10 items focused on

beliefs towards the consumption of insects (i.e., disgust, distaste, fear, and social acceptance). For all the statements, responses were given on a seven-point agreement scale, ranging from "1 = strongly disagree" to "7 = strongly agree".

In addition, participants were asked whether they had ever eaten insects, and if yes, they were asked to indicate on what occasion and their liking of the experience (i.e., "If you have already tried eating insects in the past, how much did you like it?"). Next, they were asked to evaluate the importance of possible barriers to the consumption of insects (e.g., negative taste and texture, low level of food safety, being vegetarian).

The second questionnaire (Q2) was composed of three sections. The first section was composed of three items that measured behavioral beliefs toward eating products containing insect powder. The aim was to evaluate whether this practice would have: (1) "positive effects on health", (2) "positive effects on the environment", and (3) "a familiar taste compared to known products". The second part of the questionnaire was structured in the same way as section B of Q1 in order to compare the participants' beliefs before and after the informative seminar. The last part included the intention of eating insect-based food in the coming months (behavioral intention) and the perceived behavioral control, i.e., "I believe that eating products containing insect powder in the coming months is possible." Finally, participants were asked about their willingness to eat insects (i.e., whether they wanted take part in the tasting session). If participants decided to not attend the tasting section, they were asked why (e.g., short of time, afraid to being allergic, afraid of health risks, vegetarian or other).

#### *2.4. Tasting Session*

Although the students who agreed to take part in the tasting session had been told that edible insects would be included, no further detail was given about what food they were going to taste at the end of the session, as suggested by Spence [36]. Participants were asked to score five sensory attributes of each sample (appearance, odor, flavor, texture, overall liking) using a nine-point balanced hedonic scale (1, extremely negative; 5, neither negative nor positive; 9, extremely positive) and to express the probability of consuming the product in the future using a similar nine-point scale (1, extremely improbable; 9, extremely probable). The two bread samples were identical, except one was claimed to be supplemented with insect powder, i.e., "insect-labelled" bread, although it did not contain any insect ingredients. Bread was chosen as the food product because consumers were familiar with it, an attribute which has been suggested to reduce food neophobia among Western consumers [23,27,35]. The tasting session was organized in a separate room, and once they had tasted the bread samples, participants were asked to not share information or impressions with others that had not yet tried them. Participants were also informed about the safety of the bread preparations and the allergenic potential of insects, even though the "insect-labelled" bread did not contain insect.

The two types of samples came from the same pre-sliced wholemeal bread (Lidl Stiftung & Co. KG, Neckarsulm, Germany), and they were the same size (5.7 cm × 5.7 cm × 1.1 cm, 9 g), measuring one quarter of a whole slice.

Samples were presented monadically in white plastic dishes, and the samples of "insect-labelled" bread were clearly indicated to the participants. Participants were instructed to drink water between sample tasting to neutralize the taste. Ordering of samples was randomized, so that approximately half of participants tasted the "control" bread first while the other half tasted the "insect-labelled" bread first (32 and 34 participants, respectively).

#### *2.5. Statistical Analyses*

A comparison of mean scores, i.e., independent samples *t*-test, was used to assess associations between behavioral beliefs about eating products containing insect powder (interval variables, seven-point scale) and willingness to try an insect-based food (dummy variable: yes or no).

Additionally, correlational analyses (Pearson correlation) between behavioral beliefs and behavioral intention to eat a product containing insect powder in the coming months was conducted.

Exploratory factor analyses (EFA), principal component extraction method, and varimax (orthogonal) rotation, were run to assess the unidimensional structure of the food neophobia and food rejection constructs [37]. The Cronbach's alpha was used to assess the internal consistency of the construct scales, where a value greater than 0.7 is usually recommended. Finally, a structural equation modelling (SEM) technique was employed on the data to test for the model identified in Figure 1. The SEM allows for the specification of model structure with both latent and observed variables. Latent variables, i.e., abstract phenomena that cannot be directly measured by the researcher, were analyzed using confirmatory factor analysis (CFA). CFA, often referred to as the measurement model, is used to test for the underlying latent variable structure. The statistical significance of parameter estimates was tested with the critical ratio (C.R.), i.e., the parameter estimate divided by its standard error. It operates as a z-statistic for testing that the estimate is statistically different from zero [37]. The model fit of both CFA and SEM was assessed using the chi-square (χ2), the comparative fix index (CFI), the Tucker–Lewis index (TLI), the standardized root mean square residual (SRMR), and the root mean square error of approximation (RMSEA) with 90% confidence. Since the chi-square is affected by sample size, the application of multiple fit indices is usually recommended. An acceptable fit based on these indexes is considered with CFI and TFI ≥0.9 and RMSEA ≤0.08. The width of the RMSEA confidence interval is informative about the precision in the estimate of the RMSEA. For SRMR a value less than 0.08 is generally considered a good fit [38,39]. The coefficient of determination (R2) measured the explained variance of the endogenous variables (intention and willingness to try). We applied the Bayesian estimation routine in IBM® SPSS® AMOS 24.0 (IBM, Armonk, NY, USA) recommended for analyzing categorical data [37].

The statistical difference of the correlation between attendance at the tasting session and gender of participants was tested using the chi-squared test of independence. Generalized linear model (GLM) statistics was used on the tasting session scores to evaluate the significance of the bread sample type, gender, and tasting order effects. Moreover, for each participant, delta scores were calculated subtracting for each evaluated parameter the score of the "control" bread from the score of the "insect-labelled" bread; thus, positive deltas indicate a higher score for the "insect-labelled" sample. Then the difference in responses given by each participant to the two types of samples was investigated using a Wilcoxon rank sum test.

#### **3. Results**
