A descriptive sensory evaluation of edible insects in Czechia: do the species and size matter?

ABSTRACT Our recently conducted questionnaire survey focused on the acceptance of insects by Czech respondents and revealed different preferences between the consumption of whole and ground insects. To identify whether the species and size may affect the acceptance of insects, small and large species of Blattodea, Coleoptera, and Orthoptera were assessed in their whole form and as ground meals in white bread and chocolate cookies by a panel of 16 trained panelists. The hedonic ranking test of whole fried and salted insects showed that Tenebrio molitor was the best choice. According to the paired preference test, smaller insects were preferred in the case of Orthoptera and Coleoptera, but larger cockroaches were rated better than smaller ones. The overall acceptability of insect meal in food products was higher than that of whole insects. Only bread containing Tenebrio molitor and Gryllus assimilis meals were comparable to those of the control, whereas no significant differences were detected among the cookie samples. Conclusively, it was proven that the acceptance of whole insects is influenced more by species than size, and all tested meals can be used to fortify confectionary products with no adverse effects on their liking.


Introduction
During the last few decades, intensive research has shown that insects are one of the most promising alternative foods.Generally, insects are rich sources of high-quality proteins and lipids, as well as micronutrients. [1][3] Because of the low environmental impact, high biomass gain, and valorization of by-products or waste in the food industry, the production of insects might also be both feasible and sustainable. [4]According to Jongema, [5] more than 2100 species of insects are consumed in more than 100 countries.However, most of these countries are in tropical zones.Insect consumption is uncommon in Western civilization.However, there is some evidence of entomophagy in the past.For instance, sheep cheese cazu marzu containing Phiophila casei larvae is a regional delicacy in Sicily, Italy. [6]Furthermore, cockchafer soup was a popular dish in Central Europe, including Czechia, in the first decades of the 20 th century. [7]While marine invertebrates are commonly consumed throughout the continent, their terrestrial counterparts had gained little interest from a human consumption perspective until the beginning of the 21 st century.[11] From this perspective, the awareness campaign calling for sustainable agriculture and the positive experiences gained while traveling to countries familiar with entomophagy may be identified as the main drivers of this phenomenon. [12]However, despite increased awareness of the benefits of entomophagy, the availability of insect products is limited.Our previous research in Czechia [9] showed that the majority of people who ate insects described their experiences as positive, but this does not automatically mean that they are willing to adopt insects in their regular diets.Considering the low-cost, convenient, readily available, and tasty foods established in Western dietary habits, insect-derived foods must deliver at least the same immediate satisfaction without a trade-off between quality and sustainability. [13]o promote insects as food, the introduction and establishment of insect foods in the European market represents a serious challenge in the near future.The willingness to buy insect-based products can be influenced by many factors such as food type, labeling, price, and taste.In general, insects are available either in whole or ground form.While whole insects are typically offered as dried, grilled, or pickled snacks, insect meal is used as an additive in wheat flour, with the intention of enhancing the nutritional value of familiar products such as bread, cookies, chocolate bars, pasta, etc. [14] Whole insects, particularly those collected from the wild, are dominant in tropical areas. [15]In Europe, the majority of customer surveys revealed that respondents preferred products containing insect meal compared to products in which insect parts were visible. [16]Therefore, offering products containing insect meals seems to be a promising business approach in this region of the world.
Generally, the perception of insects by humans varies; therefore, species selection is one of the first and most crucial steps when designing insect-based food.The differences in product evaluation in relation to insect type were recently described by Adamczyk et al., [17] whose research showed that products containing crickets on the label were evaluated more highly than those with larvae on the label.In our previous research, we observed significant preferences for whole insect consumption, and Orthoptera was the most favorable order.In contrast, no species preference was detected in the case of "hidden" consumption. [9]his may indicate that even "less fancy" insects can be used for insect meal production.Therefore, in this study, we aimed to verify this hypothesis and evaluate the acceptance of both whole and ground insects, with emphasis on their species and size.The insects selected for this trial have recently been investigated as promising alternative foods by several authors [18][19][20][21] and are commercially available in the Czech Republic.

Insect samples
Six insect species from three orders (Blattodea, Orthoptera, and Coleoptera) were tested in this study.From each of the selected orders, one species represented the "small" species and the other one the "large" species (Blaberus discoidalis subadults > Blatta lateralis adult females; Locusta migratoria adults > Acheta domesticus adults; Zophobas morio larvae > Tenebrio molitor larvae).All insects were obtained from the rearing facilities at the Czech University of Life Sciences Prague (CZU) and the National Institute of Public Health.Prior to the analysis, the samples were starved for 24 h and freeze-dried.For the insect meal, the insects were dried in a thermal oven (Memmert, Schwabach) for 5 h at 80°C and homogenized using a laboratory mill (Grindomix 200, Haan).Both whole and ground insects were stored in a freezer (−80°C) for analysis.Twelve hours prior to sensory evaluation, chocolate cookies (crinkles) and white bread containing insect meal (10% replacement of flour) were prepared.The recipes are provided in Supplement 1.For the whole insect samples, all insects were provided to the assessors fried in sunflower oil (for approximately 4-8 min, depending on their size) and slightly salted (a pinch of table salt per batch of each insect species in the pan).The fried insects were prepared directly in a laboratory prep room, separated from the sensory lab, and served immediately after the experiment.

Sensory evaluation
Sensory analysis of the samples was carried out under laboratory conditions (ISO 8589:2007) [22] by 16 trained assessors according to ISO 8586:2012. [23]They were university students of the Nutrition and Food program (11 women and 5 men, 19-26-years old).The assessors were informed in advance that they would be evaluating insects and products that contained insects.Three sensory methods were used for sample evaluation: the sensory profiling method (ISO 13299:2016), [24] discrimination hedonic pair comparison test (ISO 5495:2005), [25] and ranking test (ISO 8587:2006). [26]Linear graphically oriented unstructured 100 mm scales were used for the sensory profile.The orientation of hedonic scales was 0% = rejection and 100% = excellent; in the case of intensity descriptors, the orientation was 0% = imperceptible and 100% = very strong.The samples for evaluation were served in the following order: hidden ground form of insects (white bread, chocolate cookies) and then the visible form of whole fried insects.For bread and cookies, all seven samples (six insects and one control without insects) were used together.First, the panelists were asked to evaluate the samples (in random order) by the sensory profiling method and then by ranking the preference in order from the worst to the best.For whole-fried insects, the samples were served in pairs in the following order: Coleoptera, Orthoptera, and Blattodea.The assessors were asked to evaluate each pair using the profile method before tasting (overall pleasantness of sample, pleasantness of color, and pleasantness of smell) and after degustation (pleasantness of texture, overall pleasantness of taste, overall intensity of taste, intensity of off-flavor, overall sample acceptance, and flavor description -voluntary).Samples of the same order were evaluated using paired preference tests.Finally, the assessors were asked to evaluate all insect samples together using a ranking test.This research was approved by the Ethics Committee of the Czech University of Life Sciences Prague (05062021).

Statistical analysis
The data obtained from the sensory profiling assessment were statistically evaluated using the Statistica 12 software (StatSoft, Inc., Tulsa, OK, USA) using one-way factorial analysis of variance (ANOVA) and Scheffe's post hoc analyses, with a significance level of α = 0.05.The results are expressed as arithmetic mean (x) ± standard deviation (SD).The ranking test results were evaluated using the Friedman test.The number of assessors preferring one sample in the pair-comparison test was at the 5% significance level.

Results
Table 1 shows the results of the ranking tests of all samples.In the case of whole insects, T. molitor, B. craniifer, and G. assimilis were the top rated insects.On the other hand, Z. morio larvae were significantly the least favored species.In the comparison of large and small species from the same order using a paired test (Table 2), significant differences were recorded between the species in all three orders (according to Melgaard et al., [27] 12 answers out of 16 was the minimum required at the 5% significance level).The panelists preferred "small" species over "large" species in the case of Coleoptera and Orthoptera.Conversely, larger species were preferred in the case of Blattodea.All panelists were willing to taste the samples (and continue the sensory profiling evaluation after assessing the visual characteristics) (Table 2).The evaluation prior to the tasting showed moderate to low acceptability of the provided insects when T. molitor, B. lateralis, and G. assimilis were the most acceptable samples, with scores of 37-59%.In contrast, the lowest acceptability score of 22% was recorded for B. discoidalis.While the sensory evaluation showed no significant differences in pleasantness of color, texture, smell, overall intensity of taste, and intensity of off-flavor of whole fried insects, the highest values (64-67%) of overall pleasantness of taste were recorded for T. molitor, G. assimilis, and B. discoidalis.On the contrary, the worst score for this descriptor was obtained for B. lateralis (32%).Regarding overall sample acceptance, the highest values were obtained for B. discoidalis (67%), followed by G. assimilis (65%) and T. molitor (65%).The lowest overall evaluation score was observed for B. lateralis (41%).Unlike the ranking and preference paired test, the ANOVA showed that the differences in overall sample acceptance among all insects were insignificant (see Table 2).
Regarding the sensory evaluation of white bread, the ranking test showed significant differences between the bread samples when those containing no insect meal were the top-rated option.In the case of cookies, the samples containing L. migratoria meal had the highest scores.Unlike white bread, the statistical evaluation revealed no significant differences among the cookies.The profiling method showed a similar trend, where no significant differences were found in the sensory properties of the cookies (Table 3), whereas the scores of white breads (Table 4) differed significantly.According to the panelists, the acceptance of cookies and bread containing ground insects was greater than that of whole fried insects.When the panelists were given the opportunity to voluntarily describe the taste of the whole insects, the comments varied from mushrooms, nuts, and chicken up to sillage or dust (see Table 5).In the case of white bread, the overall acceptance of the samples containing insect meal from G. assimilis (72%), T. molitor (63%), and Z. morio (63%) was comparable to that of the control bread (82%).On the other hand, both cockroach meals (Table 4) had the lowest scores for overall acceptance (35% for B. discoidalis and 48% for B. lateralis) among the samples, especially due to low values of the pleasantness of smell (30% and 47%) and high values of the overall intensity of off-flavors (72% and 47%).

Discussion
Sensory properties such as textural characteristics, taste, color, and aroma are important determinants of consumers' attitudes toward food choices [28,29] ; therefore, determining these properties is essential for the introduction of new products containing insects into the market.[33] According to our data, it is clear that species play a more important role in the liking of whole insects than size.This can be explained by the fact that look and taste are represented by the morphological appearance and chemical composition, which are species-specific.Wilkinson et al. reported that these properties influence consumers' willingness to eat insects. [34]egarding tasting insects, appearance and smell can be the most important drivers influencing selection and preferences.While properties of an insect's look, such as shape, color, or texture, are easy to describe for the panelists, the definition of "insect taste" or "insect smell" seems challenging.In our study, the panelists described various flavors of insects, and nuts, chicken, yards, and crisps were reported more than once.Through the available sources, the participants often found insects to have a neutral or mild taste, mainly of mushrooms or nutty aromas. [28,35]The bland taste of insects was also confirmed by the study of Meyer-Rochow and Hakko, [36] where almost half of the participants who were unable to see and smell the food failed to distinguish edible insects from other products.Depending on cultural and geographical factors, the descriptions of the same insects may also differ.For instance, Grabowski et al. [37] reported that color descriptions of Mediterranean crickets were similar in Cambodia, Germany, and Thailand, whereas smell and taste ratings differed.Finally, the perception of food can also vary between regions when some foods are appreciated by some and avoided by others.In the case of insects, stinkbugs, known for their pungent odor, are popular food items in Africa and Asia. [29]egarding appearance, the visibility of an entire insect as an ingredient was previously reported to negatively impact their liking, [38,39] and their willingness to consume whole insects may increase after previous positive experiences. [40]For instance, house crickets gained a better score after tasting by the assessors in a sensory study previously conducted by Kouřimská et al. [41] This corresponds to the data obtained in this study, wherein all insects except B. lateralis had lower scores before tasting than after tasting.Interestingly, B. discoidalis was the worst choice prior to tasting but received the highest rating after tasting.Conversely, the other tested cockroach, B. lateralis, seemed to be quite acceptable to the panelists prior to the test, but this species dropped to the bottom of the rankings after tasting.The success of B. discoidalis contrasts with our previous survey results, [9] in which whole cockroaches were marked as the least popular choice from a preference point of view.In the case of whole fried mealworms and crickets, the panelists preferred species of a smaller size prior to as well as after their consumption.To understand the differences in preferences between mealworms, two obvious causes have been suggested.First, Z. morio was previously reported to contain high amounts of fat; therefore, [42] it was very oily.Second, it has a harder cuticle than T. molitor.Both properties contribute to the favorability of smaller T. molitor.However, it is also possible that eating smaller worms is more acceptable for Westerners than consuming larger superworms.In the case of Orthoptera, the long wings and legs of locusts might be the reason for their preference for crickets.From the perspective of insect producers, this finding seems to be good news since both "small" species (T.molitor and G. assimilis) have a significantly shorter developmental cycle and lower demands on rearing facilities than their larger counterparts, and their production thus seems to be more feasible.To the best of our knowledge, a comprehensive comparative study of the sensory properties of whole edible insects is currently lacking.However, the results of our study may be compared to those reported in the literature.For instance, the mealworm T. molitor in our study was proven to be an insignificant but top-rated species among the samples, which partly corresponds with the results of Caparros Megido et al., [43] who reported a higher liking of baked mealworms than baked crickets.Similarly, sensory evaluation by Żołnierczyk [44] showed that roasted T. molitor larvae scored better than Z. morio.
The consumption of familiar products containing insect meal is considered to be a way to overcome consumers' concerns and prejudices and may help to promote insects as regular foods in Western society. [34]However, increased familiarity itself does not necessarily improve willingness to eat insects.According to Barbera et al., [45] the quality of the experience is a decisive factor; therefore, it is important to invest in the sensory aspects.Usually, insect meal is used as a partial replacement for wheat flour to enhance the nutritional quality of the product, with no negative impact on its properties.In this study, we followed recent studies [46,47] that reported 10% supplementation as an acceptable addition of ground insects to bakery products.Mafu et al. [48] reported that the acceptance of bread decreased when more than 10% wheat flour was replaced with ground crickets.Our data showed that the insect meal influenced the sensory and physical properties of white bread, such as smell, color (see Supplement 2), and acceptance.The substantially lower values of lightness and yellowness compared to those of the control bread are common in insect-based bread and do not necessarily indicate a lower quality of the bread.The dark color of insect-based bread may be preferred by some consumers because of its similarity to that of whole-grain bread.
The results obtained in our study indicated that samples containing cockroach meal were the least popular choice among the panelists.This partly corresponds with Oliveira et al., [47] who also reported a decrease in the quality score of bread with the addition of 5-15% ground cockroach Nauphoeta cinerea compared to that of control bread.In contrast, the scores of cockroach-containing breads in their study ranged from 76.2-82.2%,which is higher than that obtained in our study (35 and 48%).According to Dutcosky, [49] breads with a score of 81-100 points can be classified as good quality, 61-80 as regular, and 31-60 as bad quality.From this point of view, only the meals prepared from G. assimilis, Z. morio, and T. molitor in this study met the quality requirements for regular bread of moderate quality.This is in agreement with the majority of previously published studies, in which control bread often had the highest average values for global liking and overall acceptability.For instance, the global liking of breads containing 10% cricket powder obtained an intermediate average, but significantly lower values in comparison with control breads were reported by Osimani et al. [50] Kowalski et al. [46] did not find any significant differences in the overall acceptability between breads containing 10% A. domestica, T. molitor, and Z. morio insect meals.The same study showed that the overall acceptability of A. domestica, Z. morio, and the control sample was lower, but did not differ significantly.The highest preference for control breads with respect to breads fortified with T. molitor meal was reported by Roncolini et al. [51] Similar values of acceptability in breads with no and 10% inclusion of house cricket meal were reported by Bawa et al. [52] Therefore, our hypothesis that consumers would not mind the species when consuming insect meal-based products was not confirmed for white bread.Excluding appearance, species-specific patterns of insects may cause differences in taste, smell, and flavor, depending on various factors, such as pheromones on their surface, diet, and presence of an exoskeleton. [51]In the case of low-rated insect meals, their addition to bread at a lower percentage than that used in this study may be considered.This may increase the acceptability of products with improved nutritional compositions.For instance, breads fortified with L. migratoria powder at levels of 1-4% contained higher values of essential amino acids than the WHO limits and had high overall acceptability. [53]nlike bread, our results showed that 10% replacement of wheat flour by insect meal had no significant effect on any of the sensory properties of the cookies.In light of this, all the sampled insect meals may be used for this purpose.This is in agreement with Castro Delgado et al., [54] who observed no negative impact of the substitution of flour with 15% cricket powder in cookies.Min et al. [55] revealed that cookies containing mealworm powder had up to 50% better overall quality, as determined by the panelists.A high willingness to buy cookies containing mealworm meals was also reported by Lucchese-Cheung et al. [56] Vanilla muffins containing 15% mealworm were comparable to control muffins, as reported by Çabuk. [57]From this point of view, it seems that the strong sweet tastes of other components in cookies mask the specific taste of insects, at least in the case of a 10% addition of insect meal.

Limitations
The samples were evaluated by trained panelists who knew in advance that they would be tasting insects and products containing them.Thus, their attitudes toward entomophagy were not a priori neutral or negative.Therefore, outputs cannot be applied to ordinary consumers and the general public, and a consumer test is recommended to follow our laboratory sensory trial.

Conclusion
In this study, sensory analysis of fried whole insects revealed that insect species is a more important factor for acceptance than size.The significant increase in the overall acceptability of cockroach B. discoidalis after tasting indicated that appearance should not be the decisive factor, and customers may change their opinion after a positive experience.It was also proven that familiar products containing insect meal are better accepted than whole insects.We found significant species-specific preferences among white bread samples, whereas no such differences were detected in cookies.From this point of view, even less popular insects or insects with worse sensory properties may be used to enrich the nutrient composition of sweet bakery confectionaries, with no side effects regarding acceptance.For white bread, the addition of ground insects at concentrations lower than 10% should be considered.Overall, G. assimilis and T. molitor seemed to be the most promising insects, as they obtained high scores for both tested types of use.This research focused on the promotion of insectbased foods in the market, with an emphasis on their sensory attributes.In addition, we recommend focusing on visual cues such as naming or labeling, which may highlight the benefits and quality of the products.

Table 1 .
The sensory evaluation of whole insects, cookies, and white bread containing insect meals by the ranking test.
aRank sum = sum of order of samples from 16 evaluators (minimum, worst sample, 16; maximum, best sample, 96 for visible form and 112 for hidden form in food products).Numbers followed by different lowercase letters are statistically different (p ≤ 0.05), according to the Friedman test.
**Number of assessors preferring this sample in the pair comparison test.
*0 = no off-flavor, 100 = strong off-flavor **Number of assessors preferring this sample in the pair comparison test.
*0 = no off-flavor, 100 = strong off-flavor **Number of assessors preferring this sample in the pair comparison test.

Table 5 .
Description of flavor (or off-flavor) of the whole edible insects.