Recency negativity: Newer food crops are evaluated less favorably

Abstract

Food crops produced by new technologies such as genetic engineering are widely opposed (Gaskell, Bauer, Durant, & Allum, 1999; Scott, Inbar, Wirz, Brossard, & Rozin, 2018). Here, we examine one reason for this opposition: recency. More recently-developed crops are evaluated less favorably, whether they are produced by artificial selection (i.e., conventional breeding), natural or man-made irradiation, or genetic engineering. Negative effects of recency persist in a within-subjects design where people are able to explicitly compare crops developed at different times, and an internal meta-analysis shows that the negative effect of recency is robust across measures and stimuli. These results have implications for the evaluation of crops produced using new modification techniques, including the widespread opposition to genetic engineering.

Introduction

Humans have been altering the plant genome through artificial selection for roughly the past 10,000 years (Zohary, Hopf, & Weiss, 2012), but the past century has seen significant advances in crop production resulting from new modification techniques. These include more intense forms of human selection, including well-known technologies such as genetic engineering (GE), but also older, lesser-known technologies such as forced mutation through irradiation or chemical exposure (Ahloowalia, Maluszynski, & Nichterlein, 2004; NASEM, 2016; Wieczorek & Wright, 2012). Scientists are also developing next-generation gene editing techniques such as CRISPR-Cas9, which will allow more precise changes to organisms’ genomes (Adli, 2018). New crop production technologies have already shown significant benefits—such as herbicide and pest resistance—to agricultural producers in both developing and developed countries (Klümper & Qaim, 2014). However, future applications could particularly benefit people in the developing world by making staple crops more pest-resistant, drought-tolerant, or vitamin-rich (Guleria, Kumar, & Guleria, 2017; Narayanan et al., 2019; Tang, Dolnikowski, Russel & Grusak, 2009; Yang et al., 2017).

One of the major obstacles to the use of these technologies has been not technical, but social. Particularly for GE crops, there has been widespread public skepticism, which is prevalent at least in the developed world, and particularly strong in Europe (Gaskell et al., 1999; Frewer et al., 2013; Scott, Inbar, & Rozin, 2016; Pew Research Center, 2016). This skepticism has persisted and even increased despite the scientific consensus that 1) the risks of GE crops to the environment are acceptable; 2) and the risks to human health are no greater than those posed by conventional agriculture (NASEM, 2016). Because of the salience and importance of this public debate, there has been a substantial body of research examining attitudes towards GE food in particular (for a review, see Scott et al., 2018). This research has uncovered a number of determinants of these attitudes, which can be divided into three broad areas: perceptions of risks (which are higher among opponents; Siegrist, 2000; Moon & Balasubramanian, 2004); perceptions of benefits (which are lower among opponents; Gaskell et al., 2004; Traill et al., 2006); and trust in institutions that evaluate and regulate GE products (which is lower among opponents; Siegrist, 1999; Siegrist, 2000). All of these can be seen as rational or quasi-rational in that they involve weighing normatively defensible attitude inputs (for example, risk and benefit beliefs) to arrive at a considered judgment about the attitude object. More recently, research has been examining less normatively-defensible bases of GE food opposition, including moral objections to genetic engineering (Sjöberg, 2008; Scott et al., 2016) and lack of knowledge about biology and genetics (Fernbach, Light, Scott, Inbar, & Rozin, 2019).

Here, we examine a hitherto-unexplored explanation for skepticism of novel crop production technologies, including genetic engineering: their recency. We were stimulated to explore this hypothesis by our puzzlement at the fact that domesticated entities are rated as much more natural than recently genetically modified entities (Rozin, 2005) even though domesticated plants (or animals) usually undergo greater genetic change than genetically engineered ones. (For example, a cocker spaniel is rated as much more natural than a pig with a single gene insertion.) We thought that one reason could be that for domesticates, the genetic change happened a long time ago.

We therefore hypothesized that the mere fact that a particular crop was developed more recently (as opposed to longer ago) would render it less acceptable to consumers. In the current research, we test this hypothesis for crops produced using a variety of techniques, including genetic engineering, natural and artificial irradiation, and even conventional breeding. Next, we outline the theoretical rationale behind the “recency negativity” hypothesis.

There is converging evidence suggesting that recency might play a role in aversion to some crop production technologies. First, food is an area in which cultural traditions and heritage are particularly important. Cultures have strong norms around what can be eaten and what theoretically edible entities must be avoided, as well as how foods are prepared (Rozin, Cohen, & Ruby, 2019), and many of these traditions are hundreds or even thousands of years old. Observant Hindus, for example, find the idea of eating beef repugnant, observant Jews will eat beef but are repulsed by pork, and many secular Americans eat beef and pork but are repulsed by the notion of eating horse (Roth, 2007). Indeed, adherence to traditional cultural norms regarding what can be eaten and how food is prepared may have served an adaptive function by helping our ancestors avoid pathogens and parasites and consume a time-tested nutrient-balanced diet—at the cost, of course, of limiting the food repertoire (Billing & Sherman, 1998; Boyd & Richerson, 1985; Henrich & Henrich, 2010; Navarrete & Fessler, 2003). In the developed world, this food traditionalism manifests as a preference for “natural” or “heirloom” crops and minimal processing (Román, Sánchez-Siles, & Siegrist, 2017; Rozin, 2005; Scott & Rozin, 2017).

Second, there is a substantial literature on the importance of food novelty as a basis for food rejection.¹ All else equal, people like a food more the more they have tried it (Pliner, 1982), which is a specific case of the more general phenomenon of “mere exposure,” where repeated exposure to a stimulus tends to increase liking for it (Zajonc, 1968). The individual-differences variable of “food neophobia” further suggests that food novelty is inherently aversive to some people. Food neophobia is defined as the avoidance of novel or unfamiliar (i.e., unknown) foods, and is correlated with avoidance and negative expectations of novel foods as well as lower liking of them (Arvola, Lähteenmäki, & Tuorila, 1999; Pliner & Hobden, 1992; Pliner, Lähteenmäki, & Tuorila, 1998). Food neophobic people are less likely to enjoy novel foods and integrate them into their diets even after being persuaded to try them. Unsurprisingly, then, the diets of those high in food neophobia are relatively homogenous (Koivisto & Sjödén, 1996; Perry et al., 2015). The related Food Technology Neophobia Scale, which is intended to specifically measure skepticism of new food technologies (e.g., “The benefits of new food technologies are often grossly overstated”; Cox & Evans, 2008) does indeed predict reluctance to consume foods produced using new technology, including GE foods (Evans, Kermarrec, Sable, & Cox, 2010; Matin et al., 2012; Vidigal et al., 2015.

Finally, when it comes to risk perceptions more broadly, perceived newness seems to play a role in risk judgments. The “psychometrics of risk perception” literature (Fischhoff, Slovic, Lichtenstein, Read, & Combs, 1978; Slovic, Fischhoff, & Lichtenstein, 1984, 1985) has found that perceptions of newness or novelty of a hazard correlate with perceived lack of knowledge about it, involuntariness of exposure, and delayed effects. Together, these perceptions constitute what is known as the “unknown risk” factor in risk perceptions. Higher perceptions of unknown risk are associated with greater perceived risk overall, greater desire to reduce risk, and greater endorsement of regulations of the hazardous activity (Slovic, Fischhoff, & Lichtenstein, 1985).

There is thus theoretical motivation for the idea that recency of development will decrease consumer acceptance of a food crop—a phenomenon we call “recency negativity.” Note that this is distinct from wariness of new food technologies (as measured, for example, by the Food Technology Neophobia Scale), because we hypothesize that even keeping the production technology constant, more recently-developed crops will be evaluated less favorably. We define “favorability” primarily as willingness to consume the crop in question, because of this question's direct relevance to consumer behavior. However, across studies we also ask participants to evaluate the crops on other dimensions including risk, benefits, moral acceptability, and naturalness.

Across four studies (and another 7 in the Supplemental Material), we find negative effects of recency on evaluations for crops produced by conventional breeding (Studies 1–2), gene insertion (Study 3), and natural and artificial radiation (Studies 1–4). We also show that the recency negativity effect holds across different lengths of time, different crop traits, and between-vs. within-subjects designs.

Open Data and Materials. Data, materials, and analysis scripts are available at https://osf.io/937hd/.