How Natural Is “More Natural”? The Role of Method,Type of Transfer,and Familiarity for Public Perceptions of Cisgenic and Transgenic Modification

Abstract

A frequent expert assumption is that the public will consider cisgenics more “natural” and therefore more acceptable than transgenics. Experimental (Studies 1 and 2) and representative survey (Eurobarometer) data highlight that public concerns indeed are stronger when the boundaries of species are crossed. However, genetic combinations that could come into existence naturally are not always considered unproblematic. Human intervention in the process amplifies concern while familiarity with the method and its products explains little of the variance. Although cisgenics is more supported than transgenics, a majority of respondents across countries considers cisgenic products to be genetically modified food that must be labeled.

Keywords

genetic modification cisgenics naturalness morality theory of hybrids

Introduction

The history of genetically modified organisms (GMOs) has been a troubled one; a “global controversy” in the 1990s (Bauer & Gaskell, 2002) involved a broad range of actors and concerns. One problem was that industry and decision makers tended to ignore consumer benefits, taking for granted that benefits from GMOs would be obvious to the public (Mather et al., 2012). Perceiving a relative advantage of a development over existing alternatives, however, is an important factor for the acceptance of innovations (Rogers, 1995). A considerable part of the public, at least in Europe, perceived a lack of benefits along with pronounced risks, resulting in strong skepticism against GMOs (Gaskell et al., 2004). Another factor identified by research on innovations, which closely relates to the perception of a relative advantage, is a development’s compatibility with values, past experiences, and needs (Rogers, 1995). Again, perceived compatibility of GMOs with prior practice was low. The techniques and products were seen as unnatural and aroused concerns about interfering and tampering with nature (Kronberger et al., 2001; Nagata, Hibino, Sugiman, & Wagner, 2006; Sjöberg, 2000a, 2000b; Sparks & Shepherd, 1994; Townsend, Clarke, & Travis, 2004; Wagner et al., 2001). Such concerns continue to play an important role in recent debates over GMOs, in which different forms of genetic modification are distinguished.

Up to date, most applications of biotechnology have been based on transgenics, in which genetic traits from one plant or animal are transferred to an unrelated species. Cisgenics (or intragenics), in contrast, is the genetic modification of a recipient organism with a gene from a crossable—sexually compatible—variety of largely the same species. The distinction has achieved quite a bit of attention because it is closely related to questions on regulation (Kuzma & Kokotovich, 2011; Reardon, 2011; Waltz, 2011, 2012). In 2006, a group of researchers proposed that cisgenic plants should be exempted from the scrutiny and safety tests that are demanded for transgenic plants on a case-by-case basis (Schouten, Krens, & Jacobsen, 2006a, 2006b). The authors argued that cisgenic transfer is no more dangerous than traditional breeding, in which huge chunks of DNA are swapped. The proposal incited considerable debate (Kuzma & Kokotovich, 2011; Reardon, 2011; Waltz, 2011, 2012). If governments define and regulate cisgenics similarly to conventional breeding, some hoped, it could herald a new future for biotechnology. Others criticized the proposal as an attempt to mislead the public into believing that they face a natural product, while another objection was based on the view that giving cisgenics special treatment would unfairly suggest that transgenics is less safe. In 2011, the debate was reignited when the U.S. Environmental Protection Agency invited comments on a draft rule that would exempt cisgenic organisms from the requirement of being registered before being field-tested or marketed. While decisions are pending, it is clear that there are far-reaching implications. Regulatory systems around the world are under pressure to formulate standards (Russel & Sparrow, 2008).

There are different motivations to consider public participation—from ideals of enhanced citizenship and democratic participation, to the hope that opportunities for participation will prevent controversy, to the more cynic view that participation might provide legitimation for (already taken) decisions (Bucchi & Neresini, 2008). While we do not know what are the motivations of actors involved in the cisgenics debate, it seems noteworthy that cisgenics has hardly been discussed in public (Kuzma & Kokotovich, 2011; Waltz, 2012). This is surprising as the argument is regularly heard that the public will consider cisgenics natural and consequently acceptable. It seems that—while rhetorically it is acknowledged that public views are important—there is little interest in actual engagement. Because experts are not necessarily good in correctly estimating lay people’s views (Segall & Roberts, 1980), it may be as problematic to presume perceived naturalness for cisgenics as it was to presume consumer benefits for GMOs in the 1990s.

The aim of this contribution is to take a step back and to investigate, in a more differentiated way, how lay people approach the topic of cisgenics, whether and to what degree they favor cisgenics over transgenics, and whether they consider cisgenics comparable to traditional breeding. Before addressing these questions in three empirical studies, we review the theoretical literature on what is known with regard to the perception of cisgenic and transgenic organisms.

What Factors Might Trigger Concern?

Proponents of cisgenics argue that the technique respects species boundaries and that the resulting organisms could have evolved by mutation or traditional breeding. Since they might—at least theoretically—come into existence naturally, they should not raise public concerns, the reasoning goes (Myskja, 2006; Schouten et al., 2006a). The argument builds on important assumptions on what is perceived to be natural and morally acceptable by lay people. In the following we examine what prior research tells on the issue.

Crossing Species Boundaries

The “essentialist theory of hybrids” provides insights on the way people think about living nature (Wagner et al., 2010). Natural kinds are categories referring to objects (plants, animals, human beings) that are thought of as existing independently of human behavior (Haslam, Rothschild, & Ernst, 2000; Quine, 1969). Their category membership and phenotype is assumed to rest on an underlying “essence,” which is seen to be fixed, inherent, and identity-defining (Ahn et al., 2001). Not only psychological but also anthropological work stresses that thinking in terms of such nonoverlapping, mutually exclusive categories represents the world as a well-structured and stable place (Douglas, 1966; Leach, 1972). Although humans continuously modify their environments (including the world of living beings), the symbolic order tends to be thought of as a stable system (Tetlock, 2002). The assumption that the world and its inhabitants can be reliably classified seems to provide a reassuring sense of orientation. Genetic hybrids, which obviate easy classification, not only arouse moral concern but also are experienced as cognitively vexing because ambiguity about their identity makes it difficult to know what to expect from them (Ahn & Kim, 2001; Gelman & Kremer, 1991). We want to call the resulting feeling of confusion threat to the symbolic order. In the context of biotechnology, this threat most clearly is documented in popular culture depictions of the technology and its products. In media pictures and focus group deliberations, hybrids—especially when they cross the borders between species—often are given a “monstrous” touch: They are depicted as bigger than their natural cousins, as threatening, and as chimeras (Castro & Gomes, 2005; Nerlich, Clarke, & Dingwall, 2000; Wagner, Kronberger, Berg, & Torgersen, 2006; Wagner, Kronberger, & Seifert, 2002).

For public perceptions, it is not only relevant what kind of organism (plant, animal, human) is involved (Connor & Siegrist, 2011; Siegrist & Bühlmann, 1999) but also how distant gene recipient and gene donor are in terms of phylogenesis (Mielby, Sandøe, & Lassen, 2013). In focus group discussions with U.S. participants, for example, plant-to-plant gene combinations received more support than animal-to-plant combinations (Knight, 2009). While cultures may differ in the ways in which they name and classify natural kinds, societies across the globe mobilize moral feelings to defend their culture’s classification system against deviant cases (Leach, 1972). This line of theorizing suggests that abnormalities—such as hybrids—should be abhorred and met with moral repugnance (Douglas, 1966), no matter whether they have come into existence naturally or by genetic modification.

Familiarity and Habituation

However, negative evaluation of hybrids may also result from a lack of familiarity. Proponents of biotechnology often claim that people abhor the new and unknown, and if only they got used to the technology and its products, they no longer would oppose it. Research in fact indicates that familiarity generally tends to be associated with positive affect (Mandler, 1983), and in the absence of negative memories, familiar stimuli are more likely to be considered safe (Zajonc, 1968). A hazard that has been present for a long time tends to be attenuated due to habituation, even though the technical risk remains the same (Slovic, Fischhoff, & Liechtenstein, 1986). Consequently, positive affect could at least in part be due to what is regularly observable and familiar, while negative affect could constitute a reaction to what is uncommon and unknown. Familiarization also affects symbolic orders as language systems reflect what is considered normal in a given culture (Leach, 1964/2000). The mule, for example, although being a hybrid, has been given a proper name in its history; the hybrid occupies a clear place in the symbolic order. Other (nameless and less familiar) hybrids might arouse more concern, not so much because they cross the borders between species but rather because they are unknown.

Human Intervention and Method

A third factor that may affect the perception of genetic hybrids relates to the methods involved. Generally, anthropogenic interference is more prone to summon moral evaluation than natural causation, and the “natural” mostly is seen to be an entity’s state before human intervention (Böhm & Pfister, 2005). Other research suggests that genetic modification has a particularly pronounced potential to reduce perceived naturalness; physical and chemical transformation or traditional breeding and domestication are less likely to be deemed problematic (Rozin, 2005). This line of reasoning suggests that genetic modification—no matter whether cisgenic or transgenic—should result in some degree of concern, even if the results of the modification could have come into existence also naturally.

Overview on Studies

In summary, research has identified different sources of concern. To what degree the evaluation of cisgenic and transgenic modification is affected by factors such as the crossing of species boundaries, human intervention, or habituation is to be addressed in the empirical part of this article. If it is human intervention at the genetic level per se that arouses concern, then cisgenics and transgenics should be evaluated similarly. If it is the crossing of species, in contrast, then transgenics should arouse considerably more concern than cisgenics. However, it is also possible that negative reactions are the result of unfamiliarity with the method and its products so that familiar hybrids—even if crossing species boundaries—are more supported than unfamiliar ones.

The factors that are varied in the experimental design of the first two studies (method, type of gene transfer, familiarity) are primarily thought to affect appraisals that have little to do with the consequences—the risks and benefits—resulting from the application of technologies. Rather, such evaluations concern aspects inherent to the technology or to its symbolic dimensions (Reiss & Straughan, 1996; Verhoog, 2003), including moral intuitions, negative imagery, or a feeling of confusion related to a hybrid’s status in a culture’s symbolic order. Given that perceived compatibility with values and prior practices tends to correlate with perceived advantage (Rogers, 1995), however, we also expect the experimental factors to affect appraisals such as perceived risks and benefits.

While Study 1 (conducted in Austria) and Study 2 (conducted both in Austria and Japan) examine the role of method, type of transfer, and familiarity for evaluation with the help of experimental designs, in Study 3 representative survey data are used to explore the scope of the experimental results. To what degree do respondents across Europe differentiate between cisgenic and transgenic options? Will cisgenics be equated with more traditional forms of breeding?

Study 1

Hypotheses

In this study, we examine the evaluation of naturally occurring genetic combinations, varying the method by which they come into existence (natural procreation vs. genetic modification) and the familiarity of the genetic combination. More specifically, we address the following hypotheses.

Method of Gene Transfer

While the “natural existence” argument suggests that naturally occurring gene combinations should meet support no matter by which method they come about, research suggests that the method of gene transfer does make a difference for everyday evaluation (Rozin, 2005; Rozin et al., 2004). Gene combinations resulting from natural procreation should be evaluated more positively than the same gene combinations brought about by genetic modification.

Awareness of Natural Existence

People may not know that some genetic hybrids do actually exist in nature. The “natural existence” argument would suggest that reminding people of the natural existence of a genetic hybrid should increase its acceptance.

Familiarity

We expect respondents to abhor hybrids that cannot be easily named and classified. While the mule (horse*ass) is a well-known hybrid that can be readily labeled, the geep (goat*sheep) and zorse (zebra*horse)—hybrids that also occasionally come into existence by way of natural procreation—are far less known. If familiarity is a relevant factor to influence evaluation, the latter hybrids should arouse more concern than the mule.

Method

Participants

Participants were 188 randomly assigned Austrian students. Age ranged from 18 to 44 years (M = 23.78, SD = 4.15). Forty-nine percent of the respondents were male.

Design and Materials

The study was introduced as a study on the perception of animals. First the respondents were asked to rate two animals (ass-horse, goat-sheep, or horse-zebra, each pair corresponding to one of the hybrids mule, geep, and zorse) on a number of adjective scales. The sequence of presentation and the role the animals played as either gene donor or gene recipient in the following text were balanced:

Please imagine genes from a [gene donor] being transferred onto a [gene recipient] by way of [method]. This produces a genetic combination of both animals. It is called a hybrid. [explicit reference to natural existence vs. no further information]

A 3 × 2 × 2 between-subjects experimental design was used (with a minimum cell size of 15) in which the following factors were varied:

Hybrid (mule/geep/zorse)

Method of gene transfer (sexual procreation/genetic modification)

Reference to natural existence (reference/no reference): While for half of the respondents the natural existence of the hybrids was explicitly mentioned, for the other half no such information was provided¹

Respondents provided the following measures, which all used seven-point scales.

Negative imagery

Respondents rated the hybrid on the same bipolar adjective scales that were used to evaluate the animals serving as gene donor and recipient at the beginning of the study. The adjectives included ugly-beautiful, dangerous-gentle, and pure-impure. The items were recoded so that higher values indicate more negative imagery (agreement to words in italics). Mean imagery across the adjectives was calculated separately for donor, recipient, and hybrid. Subsequently, the variable “negative imagery” was created by subtracting the average imagery of the genetic parents from the hybrid imagery score. The variable represents a relative measure, indicating an imagined negativity “surplus” of the hybrid compared to the genetic parents. A value of zero indicates that the hybrid is imagined in a way that corresponds to the blending of gene donor’s and recipient’s evaluation. A positive value, in contrast, indicates that the hybrid is imagined more negatively than the parents. The values in the sample range from −2.33 to 5.33 (M = 0.58, SD = 0.94).

Symbolic order threat

Respondents indicated their agreement to the items (7 = full agreement): The existence of such hybrids will confuse the order of nature; if such experiments become the rule I fear to lose track of nature; by way of such experiments the natural boundaries between living beings become blurred. A mean score was calculated to indicate symbolic order threat.

Moral acceptability

Respondents rated their agreement to the question, “Do you find such experiments morally acceptable?” (7 = high acceptability).

Perceived usefulness

Respondents indicated their agreement to two items (7 = full agreement): It is good to change nature to our advantage; and by way of such experiments we will create beings that are more useful to us. A mean score was calculated to indicate perceived usefulness.

Perceived risk

Respondents rated the question, “To what degree do you think such experiments bear risks?” (7 = extremely risky).

Hybrid familiarity

Toward the end of the survey, respondents read: This creature can come into existence by natural procreation. It is called [mule/geep/zorse]. Have you ever heard of this animal before? (7 = high familiarity).

Results

Manipulation Check

To test whether the hybrids indeed vary in familiarity, we run a 3 × 2 × 2 ANOVA with hybrid familiarity as the dependent variable and the variables hybrid (mule vs. geep vs. zorse), reference to natural possibility (mentioned vs. not mentioned), and method (sexual procreation vs. genetic modification) as factors. Besides the hybrid factor (F = 270.23, p < .000, η² = .76), no other effect or interaction reaches significance. Scheffé post hoc comparisons indicate that the mule (M = 5.68, SD = 0.99) is considerably more familiar to respondents than both the geep (M = 1.29, SD = 1.01, p < .000) and the zorse (M = 1.50, SD = 1.45, p < .000); zorse and geep do not differ in familiarity (p = .62).

Analyses

Table 1 presents descriptive statistics and correlations. Proposals that are seen to be morally acceptable are considered more useful and less risky, while evoking less symbolic order threat and negative imagery. In order to address the role of hybrid, method, and reference to natural occurrence we run a 3 × 2 × 2 MANCOVA with all evaluations as dependent variables (perceived riskiness, usefulness, moral acceptability, negative imagery, and symbolic order threat); method (natural procreation vs. genetic modification), hybrid (mule vs. zorse vs. geep), and reference to natural possibility (mentioned vs. not mentioned) as independent variables; and gender and age as covariates.

Table 1.

Correlations and Descriptive Statistics.

							Study 1
	Morally acceptable	Negative imagery	Symbolic order threat	Perceived risk	Perceived usefulness	Like breeding	M	SD
Morally acceptable	—	−.15*	−.61**	−.61**	.51**		3.12	1.73
Negative imagery	−.26**	—	.20**	.12	−.17*		.53	.93
Symbolic order threat	−.45**	.19**	—	.54**	−.42**		4.90	1.38
Perceived risk	−.51**	.07	.28**	—	−.45**		5.74	1.32
Perceived usefulness	.38**	−.18**	−.43**	−.28**	—		3.03	1.57
Like breeding	.26**	−.21**	−.37**	−.19**	.35**	—
Study 2
M	2.65	.85	5.47	5.57	2.75	2.90
SD	1.57	1.25	1.19	1.36	1.37	1.64

Note: Numbers above the diagonal refer to Study 1 while those below refer to Study 2.

p < .05. **p < .01.

There are no significant effects for age. Gender affects all of the evaluations (all Fs > 11.00, ps < .002, η²s between .06 and .11) with the exception of negative imagery (F = 0.01, p = .95); women rate the proposal generally less favorable than men.

The method of gene transfer clearly matters for evaluations. For hybrids resulting from genetic modification rather than natural procreation, proposals are considered less morally acceptable (F = 6.53, p = .01, η² = .04) and more threatening to the symbolic order (F = 9.32, p = .003, η² = .05); such hybrids also are imagined more negatively (F = 5.15, p = .02, η² = .03; see Figure 1). Although the introduction does not provide information on consequences, proposals involving genetic modification tend to be considered more risky (F = 3.70, p = .06, η² = .02) and marginally less useful (F = 2.89, p = .09,η² = .02) than those involving natural procreation.

The hybrid factor affects imagery (F = 4.99, p = .008, η² = .05), a finding that is qualified by a significant Method × Hybrid interaction (F = 4.00, p = .02, η² = .04). The unfamiliar hybrids arouse more negative imagery than the more familiar mule, with the zorse resulting from genetic modification evoking the most pronounced negative imagination.² The hybrid factor affects none of the other evaluations, and no other main effect or interaction in the model reaches statistical significance. Reference to the natural occurrence of the hybrid does not significantly affect any of the evaluations (all Fs < 2.36, ps > .12, η²s < .02).

Discussion

Study 1 addressed the role of unfamiliarity and of method for the evaluation of genetic hybrids. Although respondents imagined unfamiliar hybrids (geep or zorse) more negatively than familiar hybrids (mule), the effect of familiarity on other appraisals seems limited. The method of genetic modification, in contrast, considerably amplifies concern. The method makes respondents feel uncomfortable, even if it is used to create organisms that can also come into existence naturally or by breeding. Explicitly reminding respondents of the natural existence of the hybrids does not affect evaluation.

While Study 1 highlights that the method of genetic modification is likely to amplify negative evaluation (even if combining sexually compatible genes), it does not show how cisgenics—that is, within-species gene transfers—are evaluated in comparison to cross-species transfers. If the prime factor of evaluation is the method used, then the effects of transgenics and cisgenics on evaluation should hardly differ. However, respondents may differentiate between different forms of genetic intervention. This question is addressed in Study 2.

Study 2

Hypotheses

In Study 2, we focus on the type of gene transfer (within-species vs. cross-species), a distinction that is crucial to the differentiation between cisgenics and transgenics. More specifically, the following hypotheses are addressed.

Type of Gene Transfer

Cross-species gene transfers are expected to arouse more concern than within-species gene transfers.

Gene Recipient

Study 2 not only addresses gene transfers from animal to animal but also includes transfers involving human genes. The involvement of human genes is expected to arouse particular concern.

Culture

Although natural kind categories may carry different meanings in different cultures, we assume that cross-species gene transfers arouse concern in most cultures because they challenge symbolic orders. Folk taxonomies of nature converge considerably across cultures (Bailenson, Shum, Atran, Medin, & Coley, 2002). Although we expect general cross-cultural validity, differences may appear in detail. The relationship between humans and animals is expected to differ psychologically in Japan and Europe (Lovejoy, 1936; Mito, 1995). Western thought tends to align animals and humans in a hierarchy of value, in which humans tend to be thought of as the “creation’s crowning glory.” Japanese culture, in contrast, is animistic in nature and traditionally emphasizes the equality of all living beings including humans. As a consequence, Western respondents are expected to react more harshly to genetic combinations involving human genes.

Familiarity

We include familiarity as a covariate. Thereby, it does not address familiarity with the hybrid but rather with the method of genetic modification. Increased familiarity is expected to be associated with more positive evaluation.

Perceived Similarity With Natural Breeding

Finally, we address the question of how cisgenics and transgenics relate to traditional breeding in public perception. Are there indicators suggesting that cisgenics is equated with traditional breeding?

Method

Participants

The Austrian sample comprised 83 nonstudent respondents who were approached on the street, in supermarkets, and at a train station. Forty-eight percent were male. Age ranged from 17 to 63 (M = 35.88, SD = 11.94). Similarly, the Japanese sample comprised a total of 123 primarily nonstudent respondents. Thirty-seven percent were male. Age ranged from 18 to 75 (M = 30.59, SD = 13.74).

Design and Material

The study was introduced as research on the perception of animals, and respondents read the sentence, “According to scientists it is possible to transfer a gene from one living being to the fertilized ovum of any other.” Subsequently, participants responded to the item: “Are you already familiar with this kind of genetic modification?” on a 7-point scale (7 = high familiarity). Then a specific gene transfer was introduced: “Imagine an experiment in which a gene of a [donor] is transferred onto the fertilized ovum of a [recipient].” The questionnaires included permutations of the following two experimental factors:

Gene recipient (human/animal): The gene recipient was either human or a domesticated animal. For the latter category, we used three different exemplars (cow, pig, and sheep).

Type of transfer (within/cross): Within-species transfers included human-to-human and animal-to-animal transfers, while cross-species transfers included human-to-animal and animal-to-human transfers. More specifically, within-species transfers included human-to-human, cow-to-cow, pig-to-pig, or sheep-to-sheep combinations, while cross-species transfers included human-to-cow, human-to-pig or human-to-sheep, and cow-to-human, pig-to-human, or sheep-to-human combinations. Combinations across different domestic animals were not included.

Both an Austrian and a Japanese sample were included, resulting in a 2 × 2 × 2 (Recipient × Transfer × Country) between-subjects experimental design (with a minimal cell size of 18).

Respondents evaluated the gene donor, the gene recipient, and the hybrid on adjective scales.³ Wording, question format, and variable composition matched the following dependent variables of Study 1: Negative imagery; Symbolic order threat⁴; Perceived usefulness; Perceived risk; and Moral acceptability (for details, see the Method section of Study 1). Additionally, respondents in Study 2 indicated their agreement to the statement, “Actually genetically modified organisms are no different from organisms created by traditional breeding” (7 = strong agreement).

Results

Descriptive statistics and correlations are presented in Table 1. To examine the role of the factors gene recipient, transfer, and culture, we run a 2 × 2 × 2 MANCOVA with all evaluations as dependent variables; transfer (within vs. cross-species transfer), gene recipient (animal vs. human), and country (Austria vs. Japan) as independent variables; and gender, age, and familiarity as covariates.

Age affects moral acceptability (F = 4.96, p = .03, η² = .03) but none of the other evaluations; older respondents voice more moral concern than younger respondents. Men and women differ in the evaluation of moral acceptability, usefulness, and symbolic order threat (all Fs > 8.00, ps < .01, η²s between .04 and .06), with women being more skeptical. Familiarity with the method of genetic engineering affects all evaluations (all Fs > 3.50, ps < .06, η²s between .02 and .10) with the exception of equalization with breeding (F = 1.43, p = .23, η² = .01). Unexpectedly, the evaluations tend to be more negative the more familiar a respondent feels with the technique of genetic modification.

Transfer affects all appraisals with the exception of perceived usefulness (F = 1.86, p = .17, η² = .01): Compared to within-species gene transfers, cross-species transfers are met with more moral concern (F = 30.22, p = .000, η² = .14), more negative imagery⁵ (F = 13.84, p = .000, η² = .07) and more pronounced symbolic order threat (F = 4.99, p = .03, η² = .03). They also are more likely to be considered risky (F = 18.44, p = .000, η² = .09) and less likely to be equated with breeding (F = 4.80, p = .03, η² = .03).

Figure 2 illustrates that the pattern of results is highly similar in Austria and Japan. In fact, there are no country main effects with the exception of higher risk perceptions in Japan compared to Austria (F = 4.80, p = .03, η² = .03). Surprisingly, the kind of gene recipient (animal vs. human) also has little effect on evaluations; only risk perception is higher when the gene recipient is human rather than animal (F = 5.08, p = .03, η² = .03). A Recipient × Country interaction for symbolic order threat further qualifies the results (F = 5.13, p = .03, η² = .03). In Austria, symbolic order threat is more pronounced when the recipient is human rather than animal (F = 4.67, p = .03); in Japan, in contrast, symbolic order threat is comparable for the two kinds of recipient (F = 0.19, p = .67). No other main effect or interaction in the model reaches statistical significance.

As mentioned above, cross-species transfers are less likely to be considered comparable with traditional breeding than within-species transfers. To further explore the finding, we test equalization with breeding for within- and cross-species transfers in Austria and Japan against a test value of 4, which represents the midpoint of the seven-point scale (1 = definitely disagree, 7 = definitely agree that equal to breeding). The four comparisons indicate that in both countries the mean values are significantly lower than 4, indicating that both types of transfer are perceived to be different from breeding (Austria within: t = −2.83, p = .007; Japan within: t = −4.01, p = .000; Austria across: t = −5.64, p = .000; Japan across: t = −6.69, p = .000).

Discussion

Both in Austria and in Japan cross-species gene transfers evoked more negative appraisals than within-species combinations. Familiarity with the method of genetic modification, unexpectedly, increased rather than attenuated negative evaluation. Although equalization with breeding was more harshly denied for cross-species transfers than for within-species transfers, none of the two options was considered comparable to breeding. Besides minor differences, the patterns of evaluation were highly similar in the two countries.

Study 3

In Study 3, we turn to the question to what degree people differentiate between cisgenic and transgenic gene transfers. We focus on plants and include a large set of European countries.

Method

We use data from the 73.1 Eurobarometer “Biotechnology and the Life Sciences,” which is a representative survey fielded in 2010 with samples of about 1,000 respondents per European country (smaller samples were drawn in Luxembourg, Cyprus, Malta, and Iceland). The relevant questions were part of a split ballot design, and so were posed to only half of the respondents, resulting in a sample of N = 13,529. We report findings across the 27 European member states, with each country’s contribution weighted according to its population size.

In the survey, respondents were presented the following scenario:

Some European researchers think there are new ways of controlling common diseases in apples—things like scab and mildew. There are two new ways of doing this. Both mean that the apples could be grown with limited use of pesticides, and so pesticide residues on the apples would be minimal. The first way is to artificially introduce a resistance gene from another species such as a bacterium or animal into an apple tree to make it resistant to mildew and scab. [evaluation transgenics] The second way is to artificially introduce a gene that exists naturally in wild/crab apples which provides resistance to mildew and scab. [evaluation cisgenics]

Respondents evaluated the transgenic and the cisgenic option on the following statements (1 = totally disagree, 2 = tend to disagree, 3 = tend to agree, 4 = totally agree): it will harm the environment; it is fundamentally unnatural; It makes you feel uneasy; it should be encouraged.⁶ Furthermore, respondents were asked to indicate which of the following statements was closest to their view for both the cisgenic and the transgenic scenario:

Apples created by this technique would be like GM food and should be clearly identified with a special label.

Apples created by this technique would be the same as ordinary apples and would not need special labeling.

All questions included an “I don’t know” option. Such responses were excluded from the analyses.

Results

First analyses of the data indicate that across Europe apples produced by cisgenics tend to be evaluated as more useful, less risky, less harmful to the environment, less unnatural, and as evoking less feelings of uneasiness (Gaskell et al., 2010; Gaskell et al., 2011). The authors also reported the combined percentages of respondents indicating that they “tend to agree” or “totally agree” that the two options should be encouraged, highlighting higher support for cisgenic (55%) than for transgenic apples (33%). In the following, we complement the result by presenting the corresponding percentages for all other evaluations (see Table 2 for correlations and descriptive statistics). While 78% of the respondents perceive the transgenic option as unnatural, 57% consider the cisgenic option unnatural. Although the latter option is considered more natural than the former, a majority of respondents consider both options to be unnatural. Cisgenics makes 44% of respondents feel uneasy, while the percentage is 63% for transgenics. Respondents also expect less harm for the environment to result from the cisgenic compared to the transgenic procedure (37% vs. 55%).⁷

Table 2.

Correlations and Descriptive Statistics.

	Useful^a	Risky^a	Environ. Harm	Unnatural	Uneasy	Encourage	M (SD) [Cisgenic]
Useful^a	—	−.53	−.56	−.50	−.57	.72	2.85 (.94)
Risky^a	−.75	—	.73	.64	.66	−.57	2.47 (.94)
Will harm the environment	−.47	.48	—	.64	.69	−.56	2.28 (.94)
Unnatural	−.38	.43	.54	—	.66	−.58	2.69 (1.00)
Makes feel uneasy	−.52	.53	.59	.55	—	−.62	2.40 (1.02)
Should be encouraged	.72	−.74	−.49	−.46	−.57	—	2.50 (1.01)
M (SD) [Transgenic]	2.39 (1.00)	2.84 (.94)	2.63 (.96)	3.14 (.91)	2.81 (1.02)	2.05 (.97)

Note: Numbers below the diagonal refer to transgenics while numbers above refer to cisgenics. All ps < .000.

The item wording differed for the cisgenic and the transgenic options (see Note 6).

In a next step, we address the question to what degree individuals differentiate between the two variants of genetic modification. For this purpose, we cross-tabulate support for cisgenics and transgenics. “No support” includes the responses “totally disagree” and “tend to disagree,” while “support” includes the responses “tend to agree” and “totally agree” to the item indicating encouragement. Across Europe, 72% of respondents do not differentiate between cisgenic and transgenic options to a degree indicating differential encouragement (42% oppose and 30% support both procedures). Only 28% of respondents differentiate between the two options, out of which the majority (24%) falls into the expected category: They oppose transgenics but support cisgenics. Four percent indicate support for transgenics but not for cisgenics.

Across the 27 European member states, 90% of the respondents hold that transgenic apples are genetically modified food that needs to be labeled; for the cisgenic apple it is 78% (there is no country with a majority of respondents considering either option ordinary food that need not be labeled). In the group of respondents opposing both cisgenics and transgenics, unsurprisingly, almost all say that both the cisgenic and the transgenic variants are GM food in need of labeling (93% for cisgenics, 94% for transgenics). In the group differentiating between the two options (opposing transgenics while supporting cisgenics) labeling is considered more necessary for transgenics (96%) than for cisgenics (61%). In the group of those supporting both options 77% call for labeling of transgenics and 66% for cisgenics. In the small group of those opposing cisgenics but supporting transgenics, 82% call for labeling cisgenic and 75% for labeling transgenic produce. These results indicate that in all groups—even in those encouraging cisgenics—a clear majority of respondents considers labeling necessary, no matter whether the product was created by transgenic or cisgenic techniques.

Discussion

In expert discussions the question how cisgenics relates to breeding plays an important role, and some have suggested that cisgenics—because of being comparable to breeding—should not require labeling (Schouten et al., 2006a, 2006b). The results from the representative survey presented above indicate that citizens across Europe indeed view cisgenic products more positively than transgenic alternatives. However, while cisgenics is considered more natural than transgenics, the majority of respondents consider both options unnatural. Similarly, cisgenic produce is commonly classified as GM food that should be labeled.

General Discussion

Based on theoretical considerations, the presented studies systematically examined how different aspects related to the process of mixing genes affect the appraisals of lay people. It was hypothesized that if it is human intervention at the genetic level per se that arouses concern then cisgenics and transgenics should be evaluated similarly. If it is the crossing of species, in contrast, then transgenics should arouse considerably more concern than cisgenics. Finally, it was acknowledged that negative reactions could also result from unfamiliarity with the method and its products. The distinctions are relevant for science communication in that they address frequently heard expert expectations on how public opinion works in the context of cisgenic and transgenic modification. Proponents of cisgenics argue that the technique respects species boundaries and that the resulting organisms could have evolved by mutation or traditional breeding. As they may—at least theoretically—come into existence naturally, they should not raise public concern, the reasoning goes.

Study 1 illustrated that the method of genetic modification makes respondents feel uncomfortable, even if it is used to create organisms that can come into existence naturally or by breeding. Explicitly reminding respondents of the natural existence of the hybrids did not affect evaluation. Advice that organisms resulting from cisgenics could have evolved naturally hence does not seem to be a convincing argument for support. Study 2 showed that both in Austria and in Japan cross-species gene transfers evoked more negative appraisals than within-species combinations, a result corroborated in Study 3. Representative survey data indicate that citizens across Europe view cisgenic products more positively than transgenic alternatives. However, in both Studies 2 and 3, cisgenics is considered to be different from breeding and commonly classified as a form of genetic modification. For products resulting from such processing, labeling is considered necessary.

Overall, the result that cisgenic modification is received more positively than transgenics but at the same time is unlikely to be considered “natural” or comparable to breeding can be best explained by a combination of theoretical accounts that stress the role of method and human interference (Rozin, 2005) and the role of symbolic orders and essentialist thinking (Wagner et al., 2010). The mixing of genes evokes a feeling of confusion about the stability and reliability of the symbolic order that goes hand in hand with increased moral alert and a tendency to imagine the outcome in a negative way, an effect that is amplified if the transfer is brought about by means of genetic modification rather than by natural means. There is less evidence for effects of habituation. Familiarity with the genetic hybrid hardly affected evaluation in Study 1, and surprisingly, in Study 2, familiarity with genetic modification was associated with more negative rather than positive evaluation both in Japan and Austria. It is possible that memories about gene technology’s troubled history make habituation less likely to occur (Zajonc, 1968).

What do the results mean for the future of cisgenics? In line with the few existing studies available, our results suggest that transgenic procedures are seen as less natural and less acceptable than cisgenic alternatives (Gaskell et al., 2010; Gaskell et al., 2011; Haller, 2009; Mielby & Lassen, 2009; Schenk et al., 2011). However, not all people discriminate between the procedures. In a Swiss sample, about 40% of respondents differentiated between cisgenic and transgenic apples (Haller, 2009). Our results suggest an even smaller percentage; across Europe about a quarter of respondents distinguished between the two kinds to a degree reflected in differential encouragement. In a Dutch sample, the effect size of the preference for cisgenics over transgenics was six times smaller than the difference of the two forms of genetic modification to conventional breeding (Schenk et al., 2011). Our results also suggest that both in Europe and in a Japanese sample, cisgenics is very unlikely to be equated with breeding. A large majority of European respondents considers cisgenic produce as GM food in need of labeling.

Observers have criticized that cisgenics has hardly been discussed in public (Kuzma & Kokotovich, 2011; Waltz, 2012). Our study adds a further caveat by highlighting that experts in science, industry and regulation are not necessarily experts in correctly estimating public views. Circumventing broader discussion and silently regulating cisgenics as a form of breeding is not only problematic from a democratic point of view but it might also be a risky approach that in terms of public reactions could backfire.

Footnotes

Acknowledgements

The authors gratefully acknowledge the help of Marlen Jamie-Lee Campbell, Andrea Huemer-Mayr, Milena Marinova, Kazumasa Mori, Hideo Saitou, Yoshiko Terazawa, Stefan Totter, and Aino Yamamoto in collecting the data.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Austrian part of this research was supported by the European Commission (Grant QLG7-CT-1999-00286) and by the Austrian Bundesministerium für Bildung, Wissenschaft und Kunst (Contract 20.088/03-VI/1/03, GEN-AU/ELSA K 15, Subproject 5). The Japanese part was supported by the Japanese Ministry of Education, Culture, Sports, Science and Technology (Grant 14710079, Grant-in-Aid for Young Scientists).

Notes

Author Biographies

Nicole Kronberger is an associate professor at the Department of Social and Economic Psychology at Johannes Kepler University, Linz, Austria. Her research interests focus on social psychological phenomena such as stereotypes, risk perceptions, moral reasoning, and values, mostly in the context of public sensemaking processes of technologies.

Wolfgang Wagner is the head of the Department of Social and Economic Psychology at Johannes Kepler University, Linz, Austria. His research interests focus on public understanding of science, psychological essentialism, social representation theory, identity construction, stereotyping, and religious fundamentalism.

Motohiko Nagata is an associate professor of social psychology at Kyoto University, Japan, where he has been since 2007. He received his PhD in social psychology from Kyoto University in 2002. His research interests are in public reaction to science and technology as well as community development.

References

Ahn

W. K.

Kalish

Gelman

S. A.

Medin

D. L.

Luhmann

Atran

Shafto

(2001). Why essences are essential in the psychology of concepts. Cognition, 82(1), 59-69.

Ahn

W. K.

Kim

N. S.

(2001). The causal status effect in categorization: An overview. In Medin

D. L.

(Ed.), The psychology of learning and motivation: Advances in research and theory (pp. 23-65). San Diego, CA: Academic Press.

Bailenson

J. N.

Shum

M. S.

Atran

Medin

D. L.

Coley

J. D.

(2002). A bird’s eye view: Biological categorization and reasoning within and across cultures. Cognition, 94(1), 1-53.

Bauer

M. W.

Gaskell

(Eds.). (2002). Biotechnology. The making of a global controversy. Cambridge, England: Cambridge University Press.

Böhm

Pfister

H.-R.

(2005). Consequences, morality, and time in environmental risk evaluation. Journal of Risk Research, 8, 461-479.

Bucchi

Neresini

(2008). Science and public participation. In Hackett

E. J.

Amsterdamska

Lynch

Wajcman

(Eds.), The handbook of science and technology studies (pp. 449-472). Cambridge: MIT Press.

Castro

Gomes

(2005). Genetically modified organisms in the Portuguese press: Thematization and anchoring. Journal for the Theory of Social Behaviour, 35(1), 1-17.

Connor

Siegrist

(2011). Sorting biotechnology applications: Results of multidimensional scaling (MDS) and cluster analysis. Public Understanding of Science, 22, 128-136.

Douglas

(1966). Purity and danger. London, England: Routledge.

10.

Gaskell

Allansdottir

Allum

Castro

Esmer

Fischler

Wagner

(2011). The 2010 Eurobarometer on the life sciences. Nature Biotechnology, 29, 113-114.

11.

Gaskell

Allum

N. C.

Wagner

Kronberger

Torgersen

Hampel

Bardes

(2004). GM Foods and the misperception of risk perception. Risk Analysis, 24, 185-194.

12.

Gaskell

Stares

Allansdottir

Allum

Castro

Esmer

Wagner

(2010). Europeans and biotechnology in 2010: Winds of change? A report to the European Commission’s Directorate-General for Research. Retrieved from http://ec.europa.eu/research/biosociety/pdf/eb_64_3_final_report_second_edition_july_06.pdf

13.

Gelman

S. A.

Kremer

K. E.

(1991). Understanding natural causes: Children’s explanations of how objects and their properties originate. Child Development, 62, 396-414.

14.

Haller

(2009). Apples compared to apples: Attitudes towards cisgenic and transgenic breeds. Yearbook of Socioeconomics in Agriculture, pp. 3-34.

15.

Haslam

Rothschild

Ernst

(2000). Essentialist beliefs about social categories. British Journal of Social Psychology, 39, 113-127.

16.

Knight

A. J.

(2009). Perceptions, knowledge and ethical concerns with GM foods and the GM process. Public Understanding of Science, 18, 177-188.

17.

Kronberger

Dahinden

Allansdottir

Seger

Pfenning

Gaskell

Mortensen

(2001). “The train departed without us.” Public perceptions of biotechnology in ten European countries. Notizie di Politeia, 7, 26-36.

18.

Kuzma

Kokotovich

(2011). Renegotiating GM crop regulation. EMBO Reports, 12, 883-888.

19.

Leach

(2000). Animal categories and verbal abuse. In Hugh-Jones

Laidlaw

(Eds.), The essential Edmund Leach. Volume I: Anthropology and society (pp. 322-342). New Haven, CT: Yale University Press. (Original work published 1964)

20.

Leach

(1972). Humanity and animality. 54th Conway Memorial Lecture. London, England: South Place Ethical Society.

21.

Lovejoy

A. O.

(1936). The great chain of being. A study of the history of an idea. Cambridge, MA: Harvard University Press.

22.

Mandler

(1983). Knowing and liking. Motivation and Emotion, 7, 125-144.

23.

Mather

D. W.

Knight

J. G.

Insch

Holdsworth

D. K.

Ermen

D. F.

Breitbarth

(2012). Social stigma and consumer benefits: Trade-offs in adoption of genetically modified foods. Science Communication, 34, 487-519.

24.

Mielby

Lassen

(2009). Cisgenic crops: More natural, more acceptable. In Millar

West

P. Hobson

Nerlich

(Eds.), Ethical futures: Bioscience and food horizons (pp. 261-264). Wageningen, Netherlands: Wageningen Academic.

25.

Mielby

Sandøe

Lassen

(2013). The role of scientific knowledge in shaping public attitudes to GM technologies. Public Understanding of Science, 22, 155-168.

26.

Mito

(1995). Nihonzaru no bunpuhensen ni miru nihonjin no doubutsukan no hensen [The changes of the Japanese perception of animals examined by distribution change of Japanese macaque]. In Kawai

Hanihara

(Eds.), Doubutsu to bunmei (pp. 89-105) [Animals and civilization]. Tokyo, Japan: Asakurashoten.

27.

Myskja

B. K.

(2006). The moral difference between intragenic and transgenic modification of plants. Journal of Agricultural and Environmental Ethics, 19, 225-238.

28.

Nagata

Hibino

Sugiman

Wagner

(2006). The Japanese experience. In Gaskell

Bauer

M. W.

(Eds.), Genomics & society. Legal, ethical & social dimensions (pp. 212-227). London, England: Earthscan.

29.

Nerlich

Clarke

D. D.

Dingwall

(2000). Clones and crops: The use of stock characters and word play in two debates about bioengineering. Metaphor and Symbol, 15, 223-239.

30.

Quine

W. V.

(1969). Ontological relativity and other essays. New York, NY: Columbia University Press.

31.

Reardon

(2011). EPA proposal would exempt some GMOs from registry. Science, 332, 652.

32.

Reiss

M. J.

Straughan

(1996). Improving nature. The science and ethics of genetic engineering. New York, NY: Cambridge University Press.

33.

Rogers

E. M.

(1995). Diffusion of innovations (4th ed.). New York, NY: Free Press.

34.

Rozin

(2005). The meaning of “natural.” Process more important than content. Psychological Science, 16, 652-658.

35.

Rozin

Spranca

Krieger

Neuhaus

Surillo

Swerdlin

Wood

(2004). Preference for natural: Instrumental and ideational/moral motivations, and the contrast between foods and medicines. Appetite, 43, 147-154.

36.

Russel

A. W.

Sparrow

(2008). The case for regulating intragenic GMOs. Journal of Agricultural and Environmental Ethics, 21, 153-181.

37.

Schenk

M. F.

van der Maas

M. P.

Smulders

M. J. M.

Gilissen

L. J. W. J.

Fischer

A. R. H.

van der Lans

I. A.

Frewer

L. J.

(2011). Consumer attitudes towards hypoallergenic apples that alleviate mild apple allergy. Food Quality & Preference, 22(1), 83-91.

38.

Schouten

H. J.

Krens

F. A.

Jacobsen

(2006a). Cisgenic plants are similar to traditionally bred plants: International regulations for genetically modified organisms should be altered to exempt cisgenesis. EMBO Reports, 7, 750-753.

39.

Schouten

H. J.

Krens

F. A.

Jacobsen

(2006b). Do cisgenic plants warrant less stringent oversight? Nature Biotechnology, 24, 753-753.

40.

Segall

Roberts

L. W.

(1980). A comparative analysis of physician estimates and levels of medical knowledge among patients. Sociology of Health and Illness, 2, 317-334.

41.

Siegrist

Bühlmann

(1999). Die Wahrnehmung verschiedener gentechnischer Anwendungen: Ergebnisse einer MDS-Analyse [The perception of different applications of genetic engineering: Results of a MDS analysis.]. Zeitschrift für Sozialpsychologie, 30(1), 32-39.

42.

Sjöberg

(2000a). Factors in risk perception. Risk Analysis, 20(1), 1-12.

43.

Sjöberg

(2000b). Perceived risk and tampering with nature. Journal of Risk Research, 3, 353-367.

44.

Slovic

Fischhoff

Liechtenstein

(1986). The psychometric study of risk perceptions. In Covello

V. T.

Menkes

Mumpower

(Eds.), Risk evaluation and management (pp. 3-24). New York, NY: Plenum Press.

45.

Sparks

Shepherd

(1994). Public perceptions of the potential hazards associated with food production and food consumption: An empirical study. Risk Analysis, 14, 799-806.

46.

Tetlock

P. E.

(2002). Social functionalist frameworks for judgment and choice: Intuitive politicians, theologians, and prosecutors. Psychological Review, 109, 451-471.

47.

Townsend

Clarke

D. D.

Travis

(2004). Effects of context and feelings on perceptions of genetically modified food. Risk Analysis, 24, 1369-1384.

48.

Wagner

Kronberger

Berg

Torgersen

(2006). The monster in the public imagination. In Gaskell

Bauer

M. W.

(Eds.), Genomics & society: Legal, ethical & social dimensions (pp. 150-168). London, England: Earthscan.

49.

Wagner

Kronberger

Gaskell

Allansdottir

Allum

Cheveigné

Rusanen

(2001). Nature in disorder: The troubled public of biotechnology. In Gaskell

Bauer

M. W.

(Eds.), Biotechnology 1996-2000. The years of controversy (pp. 80-95). London, England: National Museum of Science and Industry.

50.

Wagner

Kronberger

Nagata

Sen

Holtz

Palacios

F. F.

(2010). Essentialist theory of “hybrids”: From animal kinds to ethnic categories and race. Asian Journal of Social Psychology, 13, 232-246.

51.

Wagner

Kronberger

Seifert

(2002). Collective symbolic coping with new technology: Knowledge, images and public discourse. British Journal of Social Psychology, 41, 323-343.

52.

Verhoog

(2003). Naturalness and the genetic modification of animals. Trends in Biotechnology, 21, 294-297.

53.

Waltz

(2011). Cisgenic crop exemption. Nature Biotechnology, 29, 677.

54.

Waltz

(2012). Tiptoeing around transgenics. Nature Biotechnology, 30, 215-217.

55.

Zajonc

R. B.

(1968). Attitudinal effects of mere exposure. Journal of Personality & Social Psychology, 9, 1-27.