Nanofood insights: A survey of U.S. consumers ’ attitudes toward the use of nanotechnology in food processing

Novel applications of nanotechnology in food processing hold tremendous potential to revolutionize the food industry and address challenges in food security and public health. Understanding and addressing consumers ’ evolving attitudes toward the use of nanotechnology in food processing is important to promote the technology ’ s adoption and inform the development of regulatory frameworks that align with societal values. We used a survey research design to explore U


Introduction
Nanotechnology refers to the production, characterization, and manipulation of nano-sized materials-less than 100 nm in diameter, with 1 nm being one millionth of a millimeter-that contain novel properties (Ramachandraiah et al., 2015;Weiss et al., 2006).Compared to bulk materials, nanomaterials demonstrate differences in their "optical properties, catalytic properties, porosity, electromagnetic, mechanical (stiffness and elasticity), material and surface properties (strength, weight reduction, increased stability, altered reactivity, improved functionality and absorption potential)" (Cockburn et al., 2012(Cockburn et al., , pp. 2225(Cockburn et al., -2226;;Ramachandraiah et al., 2015).Because of these novel properties, nanotechnology has led to the development of innovative food formulations (Weiss et al., 2006).
Two key areas of nanotechnology applications with tremendous potential for continued development are in food processing and food packaging (Sahani & Sharma, 2021).In food processing, nanotechnology can be used to increase the nutritional value of food, creating functional culinary options for consumers with improved health benefits (Baker et al., 2022;He et al., 2019;Thiruvengadam et al., 2018).In this sense, nanostructures serve as "encapsulation and delivery systems that carry, protect, and deliver functional food ingredients to their specific site of action" in the human body (Sahani & Sharma, 2021;Weiss et al., 2006, p. 108).When functional ingredients (e.g., proteins, antioxidants, vitamins) are delivered in nanocapsules, their taste is masked and their release is controlled (Sahoo et al., 2021).In food packaging, nanotechnology can be used to increase safety of food manufacturing, processing, and shipping through the detection of pathogens and contaminants (Weiss et al., 2006, p. 108).In this sense, nanostructures serve as sensors that monitor food quality (Sahani & Sharma, 2021).For example, when nanostructures encounter food pathogens in packaged dairy products, they fluoresce in various colors indicating spoilage.In addition, nanostructures can be used in food packaging to increase durability, temperature resistance, and flame resistance, extending the shelf-life of food and reducing food waste during transportation, especially (Thiruvengadam et al., 2018).
Nanotechnology has the potential to "transform the entire agrifood area" and "elevate global food production" by improving the nutritional value, safety, and availability of many foods (Thiruvengadam et al., 2018, p. 1).Because of this potential, nanotechnology has significant implications for improving public health and well-being worldwide.However, because some of the same novel properties in nanotechnology that strengthen food systems may pose risks to human and environmental health, extensive research is needed before the technology can be widely adopted (Sahoo et al., 2021;Sekhon, 2014).Relatedly, the U.S. Food and Drug Administration (FDA) is still in the initial stages of developing legislation to regulate nanotechnology applications in food, also slowing adoption.This is partially because some approvals are granted through risk assessments that consider the traditional particle size of a substance and, therefore, engineered nanoparticles are evaluated on a case-by-case basis (He et al., 2019).
Another key barrier slowing the adoption of nanotechnology in the agri-food sector, and arguably the most complex, is consumer acceptance (Ronteltap et al., 2011;Tian et al., 2016;Weiss et al., 2006;   Yawson & Kuzma, 2010).Because applications of nanotechnology in food are new, consumers are still developing their opinions (Frewer, 2017;Siegrist & Hartmann, 2020).However, this could change quickly with increasing media exposure (Duncan, 2011;Frewer, 2017).Among the studies that have examined acceptance of nanotechnology, many have found that global consumers (e.g., Australian, Spanish, German, British, American) have generally positive attitudes (e.g., Cobb & Macoubrie, 2004;Cormick, 2009;Gómez-Llorente et al., 2022;Reisch et al., 2011).Others, however, have found that consumers in Italy, Switzerland, France, and Germany are reluctant to purchase or consume nanofoods (e.g., Bieberstein et al., 2012;Casolani et al., 2015;Siegrist et al., 2009;Sodano et al., 2016).These contradictory results could reflect varying cultural and social norms, differences in regulatory environments, or diverse levels of public knowledge about the technology.Consumers' perceptions of nanotechnology also differ depending on its application in food (Bieberstein et al., 2012;Henchion et al., 2019).For example, Bieberstein et al. (2012) investigated French and German consumers' acceptance of two applications: nano-packaging and nano-fortification with vitamins.They found that French consumers were more reluctant to accept nano-packaging than nano-fortification, and that German consumers were more reluctant to accept nano-fortification than nano-packaging, suggesting acceptance may differ depending on nationality.Henchion et al. (2019) similarly found that Irish consumers were more accepting of nanotechnology used in chicken fillet packaging rather than when it was used in cheese processing.These results are also consistent with those from Siegrist et al. (2008) who found that, although consumers were reluctant to buy nano-packaged foods and nanofoods, they believed nano-packaged foods were more beneficial.
It is important to re-evaluate consumers' attitudes toward nanotechnology constantly because they evolve and are influenced by external events, such as media exposure and societal expectations (Duncan, 2011;Frewer, 2017;Gupta et al., 2013;Rathore & Mahesh, 2021;van Giesen et al., 2018).When forming attitudes toward nanotechnology, consumers rely more on affect than cognition because they have little knowledge about the technology, and therefore rely on feelings to determine their overall sentiment.Over time, however, consumers' attitudes toward nanotechnology integrate with knowledge, and as knowledge accumulates, their reliance on affect decreases and their reliance on cognition increases (van Giesen et al., 2018).Because media exposure of nanotechnology has increased in recent decades, and such trends indicate exposure will continue to rise with research advancements, re-evaluating shifts in attitude is important for developing effective communication strategies that address public concerns (Fisk et al., 2014).
The current study sought to identify the cognitive and affective factors influencing U.S. consumers' attitudes toward nanotechnology.
According to Edwards (1990), affect and cognition are the basis of attitude formation.This is an important area of research because attitudes toward nanotechnology influence consumers' intentions to consume nanofoods and will help inform the development of regulatory frameworks (He et al., 2019;Thyroff, 2011).To the best of our knowledge, our study is the first to explore consumers' attitudes toward nanotechnology on a national scale in more than a decade.By identifying influential factors, we can recommend communication strategies that may increase consumers' attitudes toward nanotechnology applications in food and contribute to the discourse on societal impact and acceptance.

Literature review and Theoretical background
Several studies reported that institutional trust directly influenced consumers' acceptance of nanofoods (Roosen et al., 2015;Sodano et al., 2016).Specifically, Sodano et al. (2016) found that trust in industry and distribution chains positively influenced Italian consumers' willingness to buy nanofoods, as did trust in research and public institutions.Roosen et al. (2015) similarly found that Canadian and German consumers' willingness to pay for food processed and packaged using nanotechnology increased with trust in various institutions responsible for food safety (e.g., agriculture, pharmaceutical industry, food industry).Institutional trust is often investigated as a function of perceived risks, with higher levels of trust leading to lower perceived risks (Capon et al., 2015;Siegrist et al., 2000).
Perceived risks have a negative influence on consumers' willingness to buy nanofoods, while perceived benefits have a positive influence (Sodano et al., 2016).The positive effects of perceived benefits outweigh the negative effects of perceived risks (Sodano et al., 2016;Steenis & Fischer, 2016).In studies conducted by Cobb and Macoubrie (2004) and Siegrist et al. (2008), U.S. and Swiss consumers, respectively, associated more benefits with nanotechnology than they did risks.More recently, Sodano et al. (2016) found that Italian consumers associated more risks with nanotechnology than benefits.It is possible that perceived risks are increasing with time, or that perceived risks differ based on consumers' nationality.
Additionally, Siegrist et al. (2008) found that preference for naturalness significantly influenced Swiss consumers' perceived risks and benefits of nanotechnology.When consumers perceive food as natural, they tend to hold it in higher regard (Rozin, 2005;Siegrist & Hartmann, 2020).Swiss consumers who held naturalness in higher regard perceived more risks of nanotechnology and less benefits when compared to those who held naturalness in lower regard (Siegrist et al., 2008).The process food goes through, from farm to table, is more important to consumers than food contents when they determine level of naturalness (Rozin, 2005;Siegrist & Hartmann, 2020).Because nanotechnology is a food processing method, there is evidence to suggest that preference for naturalness negatively influences consumers' attitudes toward nanofoods (Sodano et al., 2016).
Preference for naturalness tends to be associated with health consciousness (i.e., awareness of personal health and diet)-a sustained trend among consumers (Granqvist & Ritvala, 2016).Nanotechnology may appeal to health-conscious consumers because of its positive implications for food safety and its ability to create functional foods that provide added health benefits (Das et al., 2009).Numerous studies have reported results suggesting health consciousness positively influences consumers' acceptance of functional foods (e.g., Huang et al., 2019;Jung et al., 2020;Moons et al., 2018).Beyond functional foods, health consciousness has been found to influence consumers' acceptance of novel food processing technologies negatively (Okello et al., 2022).
Food technology neophobia, inclusive of preferences for naturalness, perceptions of risk, and beliefs in health benefits, is "a personality trait that affects consumers' willingness to accept new food technologies" (Maxim et al., 2021, p. 29).Sodano et al. (2016) and Vitigal et al. (2015) found food technology neophobia significantly influenced Italian and Brazilian consumers' behavioral intentions toward nanofoods.Matin et al. (2012) also found food technology neophobia significantly influenced Canadian consumers' attitudes toward nanotechnology.Interestingly, the more consumers feel concerned about sustainability aspects of their food, the more severe food technology neophobia they experience (De Steur et al., 2016).
Consumers' acceptance of nanotechnology may be related to their views of nature (Vandermoere et al., 2011(Vandermoere et al., , 2012)).Vandermoere et al. (2011) found that the more French consumers believed in the need to preserve nature to ensure a future for humankind the less supportive they were of nanofoods.Vandermoere et al. (2012) also found that the same belief negatively influenced German consumers' general attitudes toward nanotechnology.Interestingly, Matin et al. (2012) found that environmental attitudes significantly explained Canadian consumers' general attitudes toward nanotechnology but not their attitudes toward nanotechnology applications in food.Therefore, the belief that using nanotechnology puts the environment at risk may depend on its application.
Moreover, some speculate that consumers' reluctance to accept nanotechnology is influenced by their low knowledge of the technology (Beaudrie et al., 2013;Besley et al., 2008;Braman et al., 2009;Casolani et al., 2015;Gupta et al., 2013;Ho et al., 2011).For some consumers, low knowledge "may encourage early adoption of the technology, for others it can create concerns" (Giles et al., 2015, p. 467).Studies report varying results for the influence of knowledge on consumers' acceptance of nanofoods.Vandermoere et al. (2011) found no relationship between French consumers' familiarity with nanotechnology and their support for nanofoods.Greehy et al. (2011) found the opposite-the more that Irish consumers were familiar with nanotechnology, the more positive their attitudes were toward consuming nanofoods.
Because most consumers are unfamiliar with nanotechnology, they are likely to rely on subjective norms (i.e., the influence of family, friends, or peer groups' perceptions of one's behavior) when considering perceived risks (Fishbein & Ajzen, 1975;Siegrist & Cvetkovich, 2000;Thyroff, 2011).Thyroff (2011) confirmed this after finding that subjective norms significantly influence U.S. university students' intentions to consume nanofoods.Yet, since 2011, no studies have investigated this relationship.Therefore, there is a need to determine how subjective norms influence consumers' acceptance of nanotechnology using a national sample because attitudes toward the technology may have evolved in the last decade (Gupta et al., 2013).Through our review of the literature, we identified various cognitive and affective factors that have influenced, or have the potential to influence, consumers' attitudes toward nanotechnology, and we used those to develop a comprehensive regression model.The inclusion of such factors in the regression model is supported by several prominent theories, including Ajzen (1985) theory of planned behavior (TPB) and Homer and Kahle's (1988) value-attitude-behavior (VAB) model.Both theories have been widely applied to study consumers' attitudes and behaviors toward novel food technologies (e.g., Lee et al., 2021;Ma & Chang, 2022;Thyroff, 2011).
The TPB posits that attitudes, subjective norms, and perceived behavioral control are key determinants of an individual's behaviors (Ajzen, 1985).In our study, the role of subjective norms aligns with TPB, as it highlights how social norms can directly influence consumers' attitudes.The VAB posits that values influence attitudes, which in turn shape behaviors (Homer & Kahle, 1988).In our study, we included values such as environmental concern, health consciousness, preference for naturalness, and food technology neophobia, and sought to determine how they influence attitudes toward nanotechnology (Lee et al., 2021;Parrella et al., 2023) The comprehensive nature of our regression model allowed us to assess the combined effects of cognitive and affective factors on consumers' attitudes (i.e., positive or negative views) toward the use of nanotechnology in food processing.Attitudes are critical precursors to behavioral intentions and consumer acceptance (Ajzen, 1985;Homer & Kahle, 1988;Nardi et al., 2019;Parrella et al., 2023).The following research question guided the study: What effect do institutional trust, perceived risks, perceived benefits, preference for naturalness, health consciousness, food technology neophobia, environmental concern, subjective knowledge, and subjective norms have on U.S. consumers' attitudes toward the use of nanotechnology in food processing while controlling for age, education, ethnicity, gender, and political ideology?Note.Categorical variables were dummy coded.The reference group for subjective knowledge was nothing; age was 65-74 years; education was some college; ethnicity was white; gender was female; political ideology was moderate.
To answer the research question, we propose nine hypotheses.

H1.
Institutional trust has a statistically significant, positive effect on attitudes.
H2. Perceived risks have a statistically significant, negative effect on attitudes.
H3. Perceived benefits have a statistically significant, positive effect on attitudes.

H4.
Preference for naturalness has a statistically significant, negative effect on attitudes.

H5.
Health consciousness has a statistically significant, negative effect on attitudes.

H6.
Food technology neophobia has a statistically significant, negative effect on attitudes.
H7. Environmental concern has a statistically significant, negative effect on attitudes.
H8. Different levels of subjective knowledge (know nothing, know a little, know a lot) have varying effects on attitudes.
H9. Subjective norms have a statistically significant, positive effect on attitudes.

Methods
Because our study was part of a larger project comparing consumers' attitudes toward various agri-food technologies, similar methods may appear elsewhere (e.g., Parrella et al., 2023).

Questionnaire
We developed the questionnaire using Qualtrics.The measurement scales and items were selected from previous literature based on their relevance and adaptability to the current study and context.Before answering questions, participants read the following statement: Nanotechnology is used to process foods with nanoparticles that have unique properties.We opted for this concise statement to establish a clear context for the questions, emphasizing that nanotechnology is a food processing technology.By maintaining brevity, our intention was to assess consumers' attitudes based on their current awareness, while minimizing any potential influence that additional information might introduce.
We measured subjective knowledge using one question and treated it as a categorical variable in the model: How much do you know about the use of nanotechnology in food processing?They selected from three response options-nothing (1), a little (2), and a lot (3).We measured attitudes toward the use of nanotechnology in food processing using three items adapted from Chen et al. (2013).The first item, Applying nanotechnology in food processing is …, included response options ranging from extremely bad (1) to extremely good (6).The second item, Applying nanotechnology in food processing is …, included response options ranging from extremely foolish (1) to extremely wise (6).The third statement, I am … applying nanotechnology in food processing, included response options ranging from strongly against (1) to strongly for (6).
We adopted Thyroff's (2011) definition of subjective norms-"the perceived social pressure to consume or not to consume nanofoods" (p.150)-and measured participants' perceived subjective norms relating to foods processed using nanotechnology using three items.We adapted two from Thyroff (2011) and developed one.The items were 1) People who are important to me would approve of me consuming foods processed using nanotechnology; 2) People who are important in my life think I should consume foods processed using nanotechnology; and 3) People I know would consume foods processed using nanotechnology.
Participants responded to each item using a 7-point Likert scale from strongly disagree (1) to strongly agree (7).
We measured perceived benefits of nanotechnology using three items adapted from McCarron (2016): 1) Nanotechnology can reduce the likelihood of food causing an illness; 2) Nanotechnology can increase the nutritional value of food; and 3) Nanotechnology can keep food fresh for longer periods of time.Participants responded to each item using a 6-point Likert scale from strongly disagree (1) to strongly agree (6).
We measured perceived risks of nanotechnology using three items, one adapted from Hu et al. (2020), one from Urala and Lähteenmäki (2004), and one from Vasquez et al. (2022).The items were 1) Eating foods processed using nanotechnology can damage human health; 2) Foods processed using nanotechnology can negatively impact the environment; and 3) The new properties of nanoparticles carry unforeseen risks.Participants responded to each item using a 6-point Likert scale from strongly disagree (1) to strongly agree (6).
We measured institutional trust using Roosen et al.'s (2015) six-item scale.The items in the scale are various institutions-agriculture, food industry, science/research, pharmaceutical industry, government agencies/public authorities, and consumer organizations.Participants indicated their trust in the six institutions' responsibility for food safety.They responded to each item using a 6-point Likert-type scale from no trust (1) to complete trust (6).
We measured preference for naturalness using three items from Bearth et al. ( 2014): 1) I feel good when I eat natural foods; 2) The more natural the products are, the higher the quality of nutrients and vitamins; and 3) Natural foods are better for my health.Participants responded to each item using a 6-point Likert scale from strongly disagree (1) to strongly agree (6).
We measured health consciousness using three items from Huang et al. ( 2019): 1) I am prepared to do anything that is good for my health; 2) I often dwell on my health; and 3) I take health into account a lot in my life.Participants responded to each item using a 6-point Likert scale from strongly disagree (1) to strongly agree (6).
Like Verbeke (2015), we measured food technology neophobia using four items from the 'new food technologies are unnecessary' dimension of Cox and Evans's (2008) food technology neophobia scale.This modification aligns with established practices in the literature, where different versions of the food technology neophobia scale have been used to fit different research contexts (Wendt & Weinrich, 2023).We also conducted a confirmatory factor analysis, which is methodologically necessary to ensure convergent validity of the modified scale (Wendt & Weinrich, 2023).The four items were 1) There are plenty of tasty foods around so we don't need to use new food technologies to produce more; 2) The benefits of new food technologies are often grossly overstated; 3) New food technologies decrease the natural quality of food; and 4) There is no sense in trying out high-tech food products because the ones I eat are already good enough.Participants responded to each item using a 7-point Likert-type scale from totally disagree (1) to totally agree (7).
We measured environmental concern using four items from Dunlap et al. 's (2000) new ecological paradigm scale.To determine which items to include, we conducted an exploratory factor analysis and selected those with the highest factor loadings on factor one.The items we chose were 1) We are approaching the limit of the number of people the earth can support; 2) Humans are severely abusing the environment; 3) The earth is like a spaceship with very limited room and resources; and 4) If things continue on their present course, we will soon experience a major ecological catastrophe.Participants responded to each item using a 5-point Likert-type scale from strongly disagree (1) to strongly agree (5).

Validity and reliability
We pilot tested the questionnaire using a sample of Texas A&M University faculty, staff, and students (N = 237).We conducted a confirmatory factor analysis to assess the construct reliability (e.g., Composite Reliability [CR]) and validity (e.g., Average Variance Extracted [AVE] for convergent validity), since most of the scales were adapted from previous literature.The goodness-of-fit indices for the model are as follows: the chi-square test was significant (χ 2 (515) = 1547.382,p < 0.001).Importantly, the test's significance is sensitive to sample size, and give our study's large sample size, the significant chisquare test does not determine the model's goodness of fit.We used other fit indices to collectively evaluate the model's fit.
The Comparative Fit Index (CFI) was 0.955, exceeding the recommended threshold of 0.90, and the Tucker-Lewis Index (TLI) was 0.947, also surpassing the recommended threshold of 0.90.The Root Mean Square Error of Approximation (RMSEA) was 0.043, indicating a good fit as it was below the recommended cutoff of 0.08.The Standardized Root Mean Square Residual (SRMR) was 0.039, which is below the acceptable limit of 0.08.Factor loadings for all items on their respective constructs were substantial, ranging from 0.54 to 0.95, and all were statistically significant (p < 0.001), suggesting that each item was a good indicator of its underlying factor (see Table 1).The AVE for all but one construct (environmental concern) exceeded the minimum criterion of 0.50, confirming convergent validity, while the CR for each factor was above the recommended threshold of 0.70, indicating good internal consistency.Although the AVE for environmental concern was slightly lower than 0.50, the construct validity can still be considered adequate since the CR was higher than 0.70 (Fornell & Larcker, 1981).Collectively, these indices support the validity and reliability of our adapted scales (Bollen, 1989;Marcoulides & Schumacker, 2013).
To assess the reliability of each scale, we calculated Cronbach's alpha coefficients.The values resulting from the final data collection ranged from 0.75 to 0.94, indicating strong internal consistency (Cronbach, 1951).

Data collection
The study was approved for data collection by the Texas A&M University Institutional Review Board (IRB2022-0967M) on November 7, 2022.Participants gave informed consent by selecting 'Continue' before answering questions.We purchased a sample of U.S. consumers aged 18 and above through Qualtrics and collected data via an online questionnaire from December 8, 2022, to January 4, 2023.The sample adhered to U.S. Census Bureau quotas, reflecting regional and gender distributions: approximately 17% from the Northeast, 21% from the Midwest, 24% from the West, 38% from the South, 48% male, and 52% female.We collected N = 1071 complete responses.Using the Qualtrics Sample Size Calculator, we aimed for 1068 responses to generalize our results to 1 billion consumers, with a 95% confidence level and 3% margin of error.Although our large sample size allows for generalizability, it is important to note that we collected data from a paid participant pool, possibly introducing selection bias.The Qualtrics representative we worked with scrubbed the data, performing various quality checks to ensure the validity of each response.Participants also had to answer a quality check question correctly at the questionnaire mid-point for their response to be deemed valid.

Sample
Table 2 presents a detailed overview of the sample's characteristics.Participants represented a wide age range, with those aged 65-74 years (22.04%) and those aged 25-34 years (16.43%)constituting substantial portions.About half identified as female (50.33%) and most identified as white (75.91%).Large proportions of participants had completed some college (23.72%), high school (22.97%), and a bachelor's degree (21.94%).In addition, 31.47%earned between $35,000 and $74,999 annually, and 33.46% identified as moderate in their political ideology.Nearly half of participants were married (42.11%), and most supported no children (69.65%).Finally, 85.15% of participants deemed themselves the primary grocery shopper in their household.

Data analysis
Because all responses were complete, we performed no missing data treatments.We analyzed the data with Stata/SE 17.0 and conducted a simultaneous multiple regression analyses using ordinary least squares as the estimation method.The regression model included institutional trust, perceived risks, perceived benefits, preference for naturalness, health consciousness, food technology neophobia, environmental concern, subjective knowledge, subjective norms, and various sociodemographic variables-age, education, ethnicity, gender, and political ideology.
Prior to conducting the regression, we checked if the assumptions of regression analyses, including linearity, normality of residuals, and homoscedasticity, were violated by the data.Residual versus Predictor Plots depicted Lowess lines that were closely aligned with the linear reference line for most of the distribution, supporting that assumption of a linear relationship between the continuous predictors and the outcome.Density Plot and QQ Plots revealed a close to normal distribution of residuals (skewness: − 0.41, kurtosis: 4.06).The Residual versus Fitted Plot demonstrated a consistent dispersion of residuals across fitted values, confirming homoscedasticity.The non-significant Breusch-Pagan test (X 2 (1) = 0.63, p = 0.427) provided further support for the met assumption of homoscedasticity.
We also measured the degree of multicollinearity among predictors using the Variance Inflation Factor (VIF).The mean VIF across predictors was 1.34, well below the commonly accepted threshold of 10 (Hair et al., 2009).This indicates that multicollinearity was not a concern in our regression model, further ensuring the reliability of our results.

Results
Participants preferred natural food and were somewhat food technology neophobic (see Table 3).They were also somewhat health conscious and somewhat environmentally concerned.Participants were unsure about how people close to them felt about their consumption of foods processed using nanotechnology (i.e., subjective norms).In addition, they associated moderate risks and benefits with nanotechnology.Participants had neutral attitudes toward the use of nanotechnology in food processing, and they had only some trust in institutions responsible for food safety.
Most participants (f = 768; 71.71%) perceived they knew nothing about the use of nanotechnology in food processing (see Table 4).
A correlational analysis revealed positive, statistically significant associations and substantial in strength, between participants' attitudes and subjective norms (r = 0.69, p < 0.001), attitudes and perceived benefits (r = 0.67, p < 0.001), and subjective norms and perceived benefits (r = 0.65, p < 0.001), according to Davis's (1971) interpretation criteria (see Table 5).Positive and negative associations, statistically significant and of varying strengths, were identified between most of the variables.
The regression model including institutional trust, perceived risks, perceived benefits, preference for naturalness, health consciousness, food technology neophobia, environmental concern, subjective knowledge, subjective norms, age, ethnicity, gender, education, and political ideology as predictors accounted for 64.22% of the variance in attitudes toward the use of nanotechnology in food processing (see Table 6).Accounting for the number of predictors, the adjusted percentage of variance explained is 62.94%.The F-test shows that the model explained a statistically significant amount of variation in the dependent variable (F (37, 1033) = 50.12,p < 0.001).
Holding all other variables constant, each additional point in institutional trust is associated with a statistically significant average increase in attitude of 0.116 (t (1017) = 4.17, p < 0.001); each additional point in perceived risk is associated with a statistically significant average decrease in attitude of 0.233 (t (1017) = − 8.84, p < 0.001); each additional point in perceived benefit is associated with a statistically significant average increase in attitude of 0.329 (t (1017) = 11.59,p < 0.001); each additional point in subjective social norm is associated with a statistically significant average increase in attitude of 0.246 (t (1017) = 12.05, p < 0.001); and each additional point in food technology neophobia is associated with a statistically significant average decrease in attitude of 0.074 (t (1017) = − 3.75, p < 0.001).
Furthermore, holding all other variables constant, the mean attitude toward the use of nanotechnology in food processing of participants who perceived to know a lot about nanotechnology was statistically significantly higher (0.370) when compared to participants who perceived to know nothing about nanotechnology (t (1017) = 3.33, p = < 0.001); the mean attitude of participants who completed some high school was statistically significantly lower (0.230) when compared to participants who completed some college (t (1017) = − 2.14, p = 0.033); the mean attitude of participants who identified as American Indian or Alaska Native was statistically significantly lower (0.465) when compared to participants who identified as white (t (1017) = − 2.25, p = 0.024); and the mean attitude of participants who identified as Black or African American was statistically significantly lower (0.153) when compared to participants who identified as white (t (1017) = − 2.33, p = 0.020).

Discussion
Participants in the current study had neutral attitudes toward the use of nanotechnology in food processing, which contradicts results previously reported by Frewer et al. (2014) and Cobb and Macoubrie (2004).It is possible that, with time, attitudes toward nanotechnology are decreasing as media exposure is increasing (Duncan, 2011;Frewer, 2017;Gupta et al., 2013).In addition, participants' objective and subjective knowledge of nanotechnology was low, reflecting results from previous studies that also reported low knowledge levels among consumers (e.g., Beaudrie et al., 2013;Besley et al., 2008;Braman et al., 2009;Casolani et al., 2015;Gupta et al., 2013;Ho et al., 2011).Therefore, it is reasonable to infer that consumers seldom engage with or seek information regarding the use of nanotechnology in the agri-food sector because they know little about the subject.
Perceived benefits had the largest positive effect on participants' attitudes toward the use of nanotechnology in food processing.The advantages of nanotechnology that directly benefit consumers include enhanced food safety, increased nutritional value, and prolonged shelflife (Sahani & Sharma, 2021;Thiruvengadam et al., 2018;Weiss et al., 2006).To increase consumers' attitudes, consumer-centric benefits of nanotechnology should be the focus of communication efforts.Such advantages of nanotechnology also have societal implications, such as increased food availability, improved public health, and more sustainable food systems (Thiruvengadam et al., 2018).Future research should explore whether communicating the benefits of nanotechnology to consumers, emphasizing either consumer-level or societal-level impacts, leads to different effects on attitudes toward the technology.That way, better insight can be gained into the types of benefit-related messages that might resonate most with consumers.
In line with previous research (e.g., Sodano et al., 2016;Steenis & Fischer, 2016), we found that the positive effects of perceived benefits outweigh the negative effects of perceived risks.Still, mitigating perceived risks should be an intended outcome of communication efforts.The negative effects of perceived risks and food technology neophobia are related as perceived risk is a function of food technology neophobia (Maxim et al., 2021).The statistically significant correlations between perceived benefits, perceived risks, and food technology neophobia suggest that efforts to increase perceived benefits of nanotechnology would also likely decrease perceived risks and mitigate neophobic reactions.To further decrease perceived risks and mitigate neophobic reactions, communication strategies should directly address specific concerns that consumers associate with nanotechnology and demystify its applications in food.More research is needed to identify the specific risks related to human, environmental, and economic health and inform strategic messaging campaigns.It is important to interpret the results and insights we obtained about food technology neophobia with the understanding that our analysis was based solely on the 'new food technologies are unnecessary' dimension of the original scale.
Following perceived benefits, the large positive effect of subjective norms on attitudes was anticipated.Because participants' knowledge of nanotechnology was low, they were more likely to rely on the perceived opinions of those significant in their social circles when evaluating the technology (Fishbein & Ajzen, 1975;Siegrist & Cvetkovich, 2000;Thyroff, 2011).Thyroff (2011) first established a significant relationship between subjective norms and consumers' consumption intentions toward nanofoods.The current study emphasizes the important role subjective norms play in shaping attitudes more than a decade later and the relevance of subjective norms over other factors.To increase consumers' attitudes, communication strategies should leverage the role of social influence by highlighting social approvals and endorsements and featuring relatable figures and testimonials (Parrella et al., 2023).Strategically integrating messages about nanofoods into everyday narratives and showcasing how influencers benefit from nanotechnology, communicators can foster a sense of social approval, making it more relatable and acceptable.
Following perceived benefits and subjective norms, institutional trust also had a large positive effect on participants' attitudes toward nanotechnology.We operationally defined institutional trust through an assessment of participants' trust in various institutions regarding their responsibility for food safety, including agriculture, the food industry, science/research, the pharmaceutical industry, government agencies/ public authorities, and consumer organizations.These results mirror those of Roosen et al. (2015) and Sodano et al. (2016) who found that institutional trust directly influenced consumers' purchase intentions toward nanofoods.In light of these consistent results, institutions overseeing nanotechnology in the agri-food sector should prioritize food safety-related messaging, emphasizing their commitment and responsibility for ensuring food safety.Trust in these institutions' responsibility for food safety likely extends beyond technical considerations, such as adherence to scientific protocols and industry standards, to encompass broader perceptions of reliability and accountability within the societal framework.Future research should explore the factors influencing consumers' trust in these institutions by investigating how they are perceived in terms of technical competence, ethical behavior, transparency, and sense of responsibility to the larger community.
In the past, more objective knowledge has been linked to more favorable attitudes toward nanotechnology (Greehy et al., 2011).We found that subjective knowledge had the same positive effect on attitudes.Those who perceived to know a lot about the use of nanotechnology in food processing had significantly more positive attitudes toward nanotechnology when compared to those who perceived to know nothing.Therefore, ongoing research on how knowledge affects consumers' attitudes is important, especially as more information about nanotechnology in the agri-food sector emerges.Our results suggest, however, that current communication efforts should target the affective domain, supporting the notion that consumers' responses to nanotechnology are driven by affect rather than cognition (van Giesen et al., 2018).
Preference for naturalness did not significantly influence participants' attitudes toward nanotechnology.It was significantly and moderately associated with perceived risks, suggesting that the more consumers preferred natural, unprocessed foods, the more risk they linked to nanotechnology (Siegrist et al., 2008).Nanotechnology, by nature, may not align with the concepts of natural or minimally processed.However, communicating about food safety and the responsible applications of nanotechnology in specific food products while reassuring consumers about nanotechnology's compatibility with healthy, nutritious options may be an effective strategy to decrease perceived risks among consumers who prefer naturalness.
Like preference for naturalness, health consciousness also did not significantly influence participants' attitudes toward nanotechnology.Participants' knowledge of the technology was low, suggesting they were unfamiliar with specific applications of nanotechnology and its association with functional foods that provide added nutritional values and health benefits (Baker et al., 2022;Das et al., 2009;He et al., 2019;Thiruvengadam et al., 2018).Given that the current study explored attitudes toward the use of nanotechnology in food processing broadly, we recommend future research explore the effects of health consciousness on consumers' attitudes toward nanotechnology-enhanced functional food products.When consumers perceive nanotechnology as having potential advantages to their personal health through novel functional foods, it might become more appealing to those who are health-conscious (Huang et al., 2019;Jung et al., 2020;Moons et al., 2018).
Contradictory to results from Matin et al. (2012) and Vandermoere et al. (2011Vandermoere et al. ( , 2012)), environmental concern did not significantly influence participants' attitudes toward nanotechnology, and it was not strongly correlated with other variables.This could be because our measure of environmental concern focused more on global perspectives and existential threats rather than specific environmental issues that consumers may associate with agri-food technologies.The long-term effects of nanotechnology on the environment are not fully understood (Sahoo et al., 2021;Sekhon, 2014).Therefore, as research advances, it is important to concurrently explore how environmental concern influences consumers' attitudes and acceptance, especially because applications of the technology in food processing and packaging will become more prevalent and so will information about those applications.

Conclusions
Consumers' attitudes toward nanotechnology applications in food are constantly evolving due to its novelty and rapid development.Results from our study provide a robust snapshot into the cognitive and affective factors shaping consumers' attitudes toward nanotechnology at this point in time as the regression model accounted for most of the variance in the outcome.Perceived benefits, subjective norms, and institutional trust had the greatest effects on attitudes, suggesting that communication strategies implemented by researchers and industry stakeholders (e.g., policymakers, food manufacturers, regulatory agencies) should target these factors.We recommend they emphasize the benefits of nanotechnology, leverage social influences through subjective norms, and foster food safety-related trust in institutions overseeing nanotechnology applications.Although the positive effects of perceived benefits on attitudes outweighed the negative effects of perceived risks, addressing perceived risks and food technology neophobia through targeted communication efforts could also help foster more favorable attitudes.
Furthermore, preference for naturalness and health consciousness did not have a significant influence on consumers' attitudes.However, their effects should continue to be explored when attitudes toward specific nanotechnology-enhanced functional foods are considered and clearer connections are made between nanotechnology, nutrition, and human health.Future studies should also include different measures of environmental concern because, even though it did not significantly influence attitudes in the current study, the measure we used may not have captured specific environmental issues that consumers perceive as relevant to nanotechnology.As nanotechnology continues to advance, research on consumers' attitudes should also progress, especially as more information about the technology becomes available and knowledge (objective and subjective) increases.

Limitations and strengths
The study has three primary limitations and two primary strengths that are important to acknowledge.Regarding limitations, the first is the reliance on self-reported data from a non-probability sample, which may introduce response bias and affect the accuracy of the results.The second is the sample being skewed toward a predominantly white demographic, limiting the diversity of consumer perspectives represented.The third limitation is the treatment of Likert-scale data as continuous.Although our assumption analyses indicated acceptable linearity, normality of residuals, and homoscedasticity, and the large sample size helps to justify the approach, it is important to acknowledge that treating Likert-scale data as continuous may affect the precision of regression coefficients.Regarding strengths, the first is the comprehensive exploration of cognitive and affective factors influencing consumers' attitudes toward the use of nanotechnology in food processing that contributes meaningfully to the literature and builds on established theory.The second is the large, national sample, which enhances statistical rigor and the potential for generalizable results.

Table 1
Confirmatory factor analysis results.
Note.Item numbers correspond to those listed in Section 2.1.J.A.Parrella et al.

Table 2
Socio-demographic characteristics of participants.
(continued on next page) J.A.Parrella et al.

Table 3
Means and standard Deviations of variables (N = 1071).

Table 4
Participants' subjective knowledge of nanotechnology in Food processing (N = 1071).
Their framework suggests that trust in the food industry influences consumers' reliance on emotions (affect heuristic), which in turn influences their risk and benefit perceptions of nanotechnology.These perceptions directly affect consumers' purchase intentions toward nanofoods.By including institutional trust, perceived risks, and perceived benefits, our model aligns with Siegrist et al.'s framework by highlighting the role of these variables in influencing consumers' responses to the nanotechnology.
.Siegrist et al.'s (2007)model explaining consumers' willingness to buy nanofoods also supported the inclusion of factors in the regression model.

Table 6
Coefficients from the regression model with attitude toward the use of nanotechnology in Food processing as the dependent variable (N = 1071).