Changes in the influence of affect and cognition over time on consumer attitude formation toward nanotechnology: A longitudinal survey study

Abstract

Insights into how consumer attitudes toward nanotechnology are formed and develop are crucial for understanding and anticipating possible barriers in consumer acceptance of nanotechnology applications. In this study, the influence of affect and cognition on overall opinion is investigated longitudinally for emerging nanotechnologies, and compared with conventional technologies. Overall, in attitude formation toward nanotechnology applications, people rely relatively more on affect than cognition. Over time, reliance on affect decreases whereas reliance on cognition increases for nanotechnology. This suggests that over time nanotechnology applications have become somewhat more integrated within people’s already existing knowledge structure. However, for conventional technologies the influence of affect and cognition on overall attitude remains stable over time. The current study shows that it is essential to address both affective and cognitive aspects of public opinion of nanotechnology.

Keywords

affect attitudes cognition longitudinal survey nanotechnology

Nanotechnology is a promising yet little known novel technology. Nanotechnology enables the creation of completely new products, as well as substantial improvement of properties of existing products. Consumers lack knowledge and experience with nanotechnology making it difficult for them to understand the full range of possible risks and benefits associated with nanotechnology and nanotechnology-based products (Siegrist, 2010). Previous technologies, such as genetic modification, biotechnology, and nuclear energy have met with considerable public resistance, leading to rejection of these technologies by the public at large (Currall et al., 2006; Einsiedel and Goldenberg, 2004). Certain applications of nanotechnology hold the risk of running into similar issues (Frewer et al., 2011; Gupta et al., 2012; Siegrist et al., 2008). Consumer response may thus significantly influence the development of nanotechnology. Therefore, it is important to understand public attitudes, and particularly how these develop over time (Schenk et al., 2011).

Nanomaterials are developed in a wide range of applications including consumer products. In 2009, RijksInstituut voor Volksgezondheid en Milieu (RIVM) surveyed consumer products in Dutch markets, and found more than 140 products with nanoparticles already available, mainly in cleaning products and paints (Wijnhoven et al., 2009). Despite the potential of nano-materials, there is limited knowledge on the health risks of using nano-products. The associated uncertainty might thus influence consumer response. In 2009, the Dutch government launched the ‘Nanopodium’ initiative to facilitate societal dialog about nanotechnology through bottom-up initiatives of Dutch citizens.¹ During the 2 years of its existence a broad range of public information campaigns, debates, exhibitions, and other initiatives around nanotechnology proposed by the public were supported. Nanopodium resulted in increased media attention for nanotechnology in the Dutch printed press. In 2007, there were 222 reports on nanotechnology in the printed media, which increased to 589 in 2008, 734 in 2009, and 541 in 2010, see Figure 1.

Figure 1.

Intensity graph for media attention to nanotechnology, specified per news source.

In spite of the considerable media attention during 2009 and 2010, by the end of the Nanopodium initiative Dutch public knowledge on nanotechnology remained fairly limited (Stijnen et al., 2011) and not very different from such knowledge in, for example, the United Kingdom (Fischer et al., 2013; Gupta et al., 2015). Similar to the United States (Priest et al., 2011), no evidence for any amplification of risks associated with nanotechnology was present. By the end of Nanopodium in 2011, it was expected that the initiated societal dialog would continue and even intensify through public and institutional initiatives,² and that this would lead to increasing knowledge and awareness of the Dutch public about nanotechnology.

There is, however, no report to what extent the ongoing coverage and dialogue reach out to consumers and influence Dutch people in gaining knowledge and in forming an informed opinion about nanotechnology and to what extent attitudes of the Dutch population have evolved in the last few years. The latter is relevant as, even if information and knowledge remain low in the Dutch context it is relevant to gather how people form more or less informed opinions about nanotechnology with this limited knowledge (Cobb, 2005; Scheufele and Lewenstein, 2005; Siegrist, 2010).

Previous research has identified that among US and French citizens, value predispositions, religious beliefs, and heuristic cues are important in shaping consumer perception of nanotechnology applications (Cacciatore et al., 2011; Scheufele et al., 2009; Vandermoere et al., 2011). In studies in the United Kingdom, it has been shown that opinions and acceptance are dependent on the specific domain of nanotechnology applications (Pidgeon et al., 2011). Similarly, the Swiss (Siegrist et al., 2007, 2008) and Dutch (Gupta et al., 2015) perceived food-related nanotechnology applications less positive than applications in other domains. Nanotechnology applied to food products is considered more negative than those in packaging, in Switzerland (Siegrist et al., 2007), Mexico (López-Vázquez et al., 2012) and the Netherlands (Steenis and Fischer, 2016). Together, these studies align with articles comparing consumer views across different countries (e.g. Reisch et al., 2011) which show remarkable similarities: public response is characterized by little media attention, low knowledge, and reliance on non-cognitive attitudes. By and large, most of these studies have focused on comparison between different applications of nanotechnology but not comparison of nanotechnology with its conventional counterparts. In the light of the limited knowledge and limited media exposure to nanotechnology in many countries, it is not clear if and how the attitude structure of nanotechnology develops and changes over time (Pidgeon et al., 2011).

The present study aims to investigate how the underlying attitude structure in terms of influence of affect and cognition on the overall opinion of nanotechnology will change over time, and if so, whether changes can be explained by knowledge growth. This is done through a longitudinal survey study in which attitudes for nanotechnology applications and conventional applications are compared.

1. Attitudes

Consumer opinions are studied by attitudes – ‘associations in memory between a given object and one’s evaluation of that object’ (Ajzen, 2001; Fazio, 2007). In attitude research, a distinction is made between affect-based (e.g. feelings) and cognition-based (e.g. beliefs) attitude expressions (Edwards, 1990). In most situations, attitudes will jointly be determined by affect and cognition (Edwards, 1990). Depending on the context with the attitude object, overall attitude expressions will find their base either more in cognition or more in affect.

Having an established knowledge base, as for conventional technologies, means that relevant affective and cognitive information and experiences are available (Edwards, 1990; Plessner and Czenna, 2008). As people can draw on both affective and cognitive associations with conventional attitude objects, the attitude formation process is relatively straightforward. The attitude will consist of a mix of affective and cognitive evaluations (Shiv and Nowlis, 2004) that are combined into a summary evaluation or attitude (e.g. following Fishbein and Ajzen, 1975). Cognitive evaluation may relate to utility assessment or truth evaluation of the information (Gawronski and Bodenhausen, 2006). Compared with cognitive weighing of pros and cons affective evaluation requires less formal information (Slovic et al., 2002). Affective responses do not necessarily require conscious elaboration and can be created more quickly (Bornstein, 1989; Hansen and Wänke, 2009; Zajonc, 1980).

When individuals lack a priori evaluative associations (Fazio, 2007), evaluations toward the attitude object need to be constructed on the spot (Schwarz, 2007). In low information situations, people can still easily access a broad range of affects, further contributing to the heavier weighting of affect compared with cognition (Clore and Huntsinger, 2007; Clore and Schnall, 2005). Affect experienced at the moment of evaluation thus plays an important role in people’s early judgments of unfamiliar applications (Loewenstein et al., 2001). This would explain the relatively dominant role of affect for nanotechnology (Kahan et al., 2007; Satterfield et al., 2009) and predicts a contrast with conventional technologies:

H1. For nanotechnology compared with conventional technology, there are differences in the influence of affect and cognition on overall attitude, with affect being relatively more influential for nanotechnology.

Prior knowledge influences information search and information processing and is also expected to influence the attitude formation process. Having a knowledge base of relevant information about a technology allows the formation of informed attitudes (Edwards, 1990; Plessner and Czenna, 2008). People with higher domain knowledge and expertise are better able to use recalled evidence and are influenced by the content (Ofir, 2000).

When prior knowledge is limited, as is the case with nanotechnology, it tends to be structured in a rudimentary fashion. The new (nanotechnology) information is not yet integrated with previous knowledge (Peracchio and Tybout, 1996). As people learn more about the technology attitudes should become influenced more by values and cognitive beliefs (Reisch et al., 2011). When knowledge expands, previously unfamiliar nanotechnology applications become increasingly interconnected within knowledge structures and incongruities may be resolved with minimal effort (Peracchio and Tybout, 1996). Although media coverage did not increase after 2009, this does not rule out that people become more familiarized with nanotechnology as they can also get information about nanotechnology through other sources (e.g. workshops and debates).

As knowledge toward conventional technology is more developed than toward nanotechnology, the attitude structure of conventional technological applications is more accessible and expected to be relatively stable over time (Fazio, 1990) compared with nanotechnology. For conventional technologies both affective and cognitive evaluations are available to draw upon, even if motivation to evaluate the technology is low and if there is no opportunity to extensively deliberate on the technology (see Fazio, 1990). With increasing knowledge about nanotechnology, it will be easier to integrate information and connect it in already existing knowledge structures. Over time, cognitive associations (information and facts) will become part of the attitude structure which creates the opportunity to also use cognition in evaluation (see Fazio, 1990). People are then able to access and use this knowledge to supplement their attitudes in a cognitive way. In the case of nanotechnology, therefore, a shift can be expected in overall attitudes becoming less affective over time, with increased knowledge growth. This leads to the following hypotheses:

H2a. The influence of affect and cognition on overall attitude toward conventional technology is stable over time.

H2b. For nanotechnology, affect becomes relatively less predictive for the overall attitude over time.

2. The present study

The aim of this study is to examine to what extent the relative influence of affect and cognition on overall attitude changes over time for nanotechnology and conventional technologies. In addition, the influence of subjective and objective knowledge is taken into account as a possible explanation for observed changes in affective and cognitive influence over time. Objective knowledge refers to accurate stored information, whereas subjective knowledge refers to self-beliefs about one’s own knowledge (Carlson et al., 2009).

Monitoring on the basis of longitudinal data brings advantages over cross-sectional data, as consumer attitudes can be compared with the ‘base level’ measurement of the first time point (de Jonge et al., 2010). At first, respondents are expected to not have much previous experience with nanotechnology, compared with conventional technology. Later on, changes in the knowledge about nanotechnology applications are monitored and are expected to converge more toward conventional technology knowledge, as with maturation of nanotechnology more information becomes available to consumers (Maynard, 2006). We expect that attitudes toward less-known nanotechnology will depend more on affect than cognition and that attitudes will become more cognitive over time as knowledge accumulates.

3. Method

Sample

Consumer attitudes toward nanotechnology and its applications were assessed in three surveys, about 10–11 months apart. Data collection took place during a 3-week period in October–November 2012, September 2013, and July 2014, respectively. Data were collected through Gesellschaft für Konsumforschung (GfK), a market research agency. The research complies with the Netherlands Code of Conduct for Scientific Practice and the Social Sciences Ethics Committee of Wageningen University waived the need for ethical consent. The authors did not have access to any identifying information about the participants as GfK anonymized and de-identified all data prior to author access. In-depth analyses of the first data collection round can be found in the work of van Giesen et al. (2015).

As socio-demographic information of panel members is known, the panel allowed for stratified random sampling of a nationally representative sample on gender, age, and education levels of the Netherlands. The GfK panel consisted of approximately 12,000 participants, who were repeatedly invited to participate in studies. Through a range of sampling techniques, the panel is maintained such that it remained representative for the population.³ Response rates and socio-demographic make-up of the samples regarding gender, age, and education levels can be found in Table 1.

Table 1.

Sample characteristics.

Data collection period	2012 (N = 1907) 16 October–6 November	2013 (N = 1335) 5 September–16 September	2014 (N = 856) 5 June 5–17 June	Population statistics (CBS, 2014) (1 January 1 2014)
Response rate	76% (of 2500)	74% (of 1870)	66% (of 1297)	–
Gender
Male	50.7%	51.7%	55.6%	49.5%
Female	49.3%	48.3%	44.4%	50.5%
Age
18–35	34.9%	29.0%	26.4%	34.7%
36–49	30.6%	30.0%	29.7%	31.2%
50–65	34.5%	41.0%	43.9%	34.1%
Education level
Low	27.8%	27.1%	27.9%	28.1%
Average	44.4%	44.4%	42.7%	43.1%
High	27.8%	28.5%	29.4%	28.8%

Note that 37 people in the first data collection round did not want to participate in follow up studies or were non-serious. Therefore in wave 2 the research agency approached 1870 people. Over time there is a higher drop out among females, χ²(2) = 9.34, p = .009, Cramer’s V = .048 and among the young age group, χ²(4) = 33.70, p < .001, Cramer’s V = .064. There are no differences in drop out across education levels, χ²(4) = 1.17, p = .88.

Materials

Respondents judged a familiar (conventional) and an unfamiliar (nanotechnology) application from the same application domain. This was repeated for two application domains (either water and energy or medicine and food) to limit response fatigue. In addition, the design was replicated with two different target products for each application domain to make sure that effects were not due to the choice of product. Each respondent therefore judged in total 4 of 16 available applications, representing an incomplete repeated measures factor across four domains. During the three annual surveys, respondents saw the same applications (see Supplementary Appendix 2, http://pus.sagepub.com/).

Attitude objects

Stimuli were 16 descriptions of technological applications. Four application domains were included that cover key areas of nanotechnology research and development: food, water purification, medicine, and energy (NWO, 2012). For each domain, two different applications were selected – food additives and food supplements, water purification and water quality monitoring, medical home tests and drugs, and solar energy and batteries – to provide replications, allowing to control for specific application and domain associations. In addition, for each application type, a conventional and a nano-based technology was used to manipulate familiarity between applications. Respondents received a short description of an application consisting of the following: (a) information about the technology behind the application, (b) examples in which the application can be used, and (c) advantages and disadvantages of the application. Descriptions were checked by an expert on nanotechnology and pilot tested.

Measures

Measures to assess the key constructs of the influence of affect, cognition, and knowledge on overall attitude are reported in Supplementary Appendix 1 (all measures were 7-point scales unless indicated). The affective attitude component was measured with four positive items (joy, desire, fascination, and satisfaction) and three negative items (fear, sadness, and disgust) (see Desmet, 2003; Russell, 1980). The cognitive component was measured with three positive items (useful, functional, and beneficial) and four negative items (useless, harmful, disadvantageous, and unusable) (based on Crites et al., 1994).⁴ For means per application and time point, see Supplementary Appendix 3. In the second and third wave, questions related to media exposure were added. Respondents were asked whether they have read, watched television programmes, and accessed the Internet for more information about nanotechnology and its applications (see Supplementary Appendix 1).

4. Results

Measurement model

First, the measurement model was tested using confirmatory factor analysis (CFA) with maximum likelihood estimation in the R package Lavaan (Rosseel, 2012). Comparative Fit Index (CFI) and Tucker Lewis Index (TLI) values above .95 and Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual (SRMR) values below .07 were adopted as indication of good fit. χ² is reported as customary, but not indicative of a model fit with large samples (Kline, 2005). The three waves of data were analysed using multi-group modelling, where each wave of data collection was considered as a separate group. By simultaneously estimating the model for the different time points it could be established whether the properties of the measurement model were stable over time. In the establishment of the measurement model, the relationships between the determinants (affect and cognition) and the dependent variable (overall attitude (ATT)) for conventional and nanotechnologies were not estimated because testing group differences between the structural parameters was part of the second step of analyses.

The measurement model included the latent variables for affect, cognition, and overall attitude across the three time points (affect T1/2/3, cognition T1/2/3, and overall T1/2/3). Both affect and cognition were defined by seven indicators, which were a mix of positive and negative items. Overall attitude was measured with one item in the first data collection round (ATT1) and three items in the second and third data collection round (ATT1, ATT2, and ATT3), see Figure 2.

Figure 2.

Measurement model.

To compensate for the response bias related to positive and negative framing of the items, the measurement model specified latent constructs ‘positive’ and ‘negative’ that were not related to any other construct and are defined across all time points (following Lattin et al., 2003). The CFA showed an acceptable to good fit across time points, χ²(1122) = 1246.27, p < .001; RMSEA = .054; SRMR = .077; CFI = .920; and TLI = .916, supporting the assumption that the measurement model is robust across time periods. Items were then averaged to form affective, cognitive, and overall attitude scales. This three-component structure is superior to a single attitude component of 17 items (Δχ² = 326.7, Δdf = 74, p = < .001) showing divergent validity, while the correlation between attitude components (average about .80 and range between .721 and .891) at each moment in time shows convergent validity.

Factor loadings were as follows for Affect T1, T2, and T3, respectively: joy (.69, .70, and .69), fascination (.66, .66, and .66), satisfaction (.77, .79, and .78), desire (.57, .59, and .59), fear (.24, .23, and .23), disgust (.38, .38, and .38), and sadness (.26, .26, and .26) and for Cognition T1, T2, and T3 as follows: useful (.90, .90, and .90), functional (.83, .82, and .83), beneficial (.73, .74, and .77), unusable (.51, .52, and .52), harmful (.32, .33, and .33), and useless (.57, .58, and .59).

Path models

Subsequently, the relative influence of affect and cognition on overall attitude over time was investigated in a multi-group model (conventional technology and nanotechnology), starting with the full model and subsequent trimming of the model. The overall model fit was good for the full model where only the covariances between the overall attitudes were constrained to be equal at the different time points, Model A in Table 2, χ²(28) = 145.91, p < .001; RMSEA = .050; SRMR = .029; CFI = .990; and TLI = .986 (see Figure 3). Next, the technology (Model B) and time effects were tested (Model C–E3). The model was trimmed by constraining parameters until the model with the least number of parameters was reached that showed no worse fit compared with the full model by comparing Δχ², with a significance level smaller than .10. In the last part, knowledge level and its interaction with affect and cognition was added to the model to check whether changes in affective and cognitive bs were due to changes in knowledge.

Table 2.

Path models with affect and cognition on overall attitude.

	Description	χ²	Df	CFI	TLI	RMSEA	RMSEA 90% LB	RMSEA 90% UB	SRMR	AIC	BIC	Δ χ²	Δdf	P
Full model
Model A:	Full model, no restrictions	145.91	28	.990	.986	.050	.042	.058	.029	71,631.67	71,791.27
Technology effect
Model B:	B_{A1_N} = B_{A1_C} B_{A2_N} = B_{A2_C} B_{A3_N} = B_{A3_C} B_{C1_N} = B_{C1_C} B_{C2_N} = B_{C2_C} B_{C3_N} = B_{C3_C}	158.98	35	.990	.988	.045	.038	.053	.029	71,630.37	71,747.38	13.08	7	.070*
Conventional
Model C:	B_{A1_C} = B_{A2_C} = B_{A3_C} B_{C1_C} = B_{C2_C} = B_{C3_C}	149.11	32	.991	.988	.046	.039	.054	.030	71,626.87	71,761.92	3.20	4	.523
Nanotechnology
Model D:	B_{A1_N} = B_{A2_N} = B_{A3_N} B_{C1_N} = B_{C2_N} = B_{C3_N}	155.06	32	.990	.987	.047	.040	.055	.030	71,632.82	71,767.87	9.16	4	.057*
Time effect nanotechnology (compared to Model C)T1 and T2 are equal
Model E1:	B_{A1_N} = B_{A2_N} B_{C1_N} = B_{C2_N} B_{A1_C} = B_{A2_C} = B_{A3_C} B_{C1_C} = B_{C2_C} = B_{C3_C}	150.97	34	.991	.988	.045	.038	.052	.030	71,624.73	71,747.50	1.86	2	.395
T2 and T3 are equal
Model E2:	B_{A2_N} = B_{A3_N} B_{C2_N} = B_{C3_N} B_{A1_C} = B_{A2_C} = B_{A3_C} B_{C1_C} = B_{C2_C} = B_{C3_C}	155.17	34	.990	.988	.046	.038	.053	.030	71,628.93	71,751.71	6.06	2	.048**
T1 and T3 are equal
Model E3:	B_{A1_N} = B_{A3_N} B_{C1_N} = B_{A3_N} B_{A1_C} = B_{A2_C} = B_{A3_C} B_{C1_C} = B_{C2_C} = B_{C3_C}	154.97	34	.990	.988	.046	.038	.053	.030	71,628.73	71,751.50	5.86	2	.053*

< .10; ** <.05.

CFI: Comparative Fit Index, TLI: Tucker Lewis Index, RMSEA: Root Mean Square Error of Approximation, LB: Lower Bound, UB: Upper Bound, SRMR: Standardized Root Mean Square Residual, AIC: Aikake Information Criterion, BIC: Bayesian Information Criterion; T1/2/3: time point 1/2/3; A1_N: affect time point 1 nano; A2_N: affect time point 2 nano; A3_N: affect time point 3 nano; C1_N: cognition time point 1 nano; C2_N: cognition time point 2 nano; C3_N: cognition time point 3 nano; A1_C: affect time point 1 conventional; A2_C: affect time point 2 conventional; A3_C: affect time point 3 conventional; C1_C: cognition time point 1 conventional; C2_C: cognition time point 2 conventional; C3_C: cognition time point 3 conventional; B; AIC; BIC; LB; UB.

Note that the fully unconstrained model includes 42 parameters, χ² = 12375.39.

Figure 3.

Path models for nanotechnology and conventional technology. Straight lines represent b values. Curved lines represent correlations across time points.

First, it was tested whether similar relations hold for conventional technology and nanotechnology by constraining all path coefficients to be equal for nanotechnology and conventional technology, as well as the covariances between overall attitudes (Model B). The chi-square difference between model A and B shows a decrease in the model fit when constraining technology, Δχ² = 13.08, Δdf = 7, p = .070 (see Table 2). This shows that it is meaningful to address differences between nanotechnology and conventional technology when predicting the influence of affect and cognition on overall attitude, in line with H1.

Next, the time effects for affect and cognition were estimated for nanotechnology and conventional technology to test the hypothesis that for nanotechnology affect becomes less predictive for overall attitude over time. First, a more restricted model was estimated where for conventional technology the relations for affect and cognition on overall attitude were constrained to be equal over time (all b values equal), while there were no restrictions on nanotechnology over time (Model C). This model shows no worse fit than the unconstrained model, Δχ² = 3.20, Δdf = 4, and p = .523 (Model C versus A, see Table 2). This indicates that for conventional technology the effects of affect and cognition on overall attitude are stable over time. Next, for nanotechnology the relations for affect and cognition on overall attitude were constrained to be equal over time (all b values equal), while there were no restrictions on conventional technology over time (Model D). This model showed a decrease in the model fit, Δχ² = 9.16, Δdf = 4, and p = .057 (Model D versus A, see Table 2), indicating that for nanotechnology there are differences in the effect of affect and cognition on overall attitude over time. Therefore, it is meaningful to take into account the time effect of affect and cognition for nanotechnology, but not for conventional technology, in line with H2a and H2b. So, for further model comparisons Model C is taken as the baseline model.

Then, it was further investigated whether differences in affect and cognition change across specific time points for nanotechnology, keeping conventional technology constrained. Constraining the relations for affect and cognition on overall attitude to be equal at T1 and T2 (Model E1) resulted in no worse model fit, Δχ² = 1.86, Δdf = 2, p = .395 (Model E1 versus C, see Table 2). A model where the relations for affect and cognition on overall attitude were constrained to be equal at T2 and T3 for nanotechnology resulted in a decrease in the model fit compared with Model C, Δχ² = 6.06, Δdf = 2, p = .048 (Model E2 and Table 2). A model where the relations for affect and cognition on overall attitude were constrained to be equal at T1 and T3 showed a decrease in the model fit compared with Model C, Δχ² = 5.86, Δdf = 2, p = .053 (Model E3 and Table 2). This shows that differences in affect and cognition over time exist for nanotechnology, and mainly between T3 and earlier waves.

Interpreting the results based on empirical relations shows that for nanotechnology the influence of affect decreases after T2 and stabilizes at T3 (b_T1 = .63, b_T2 = .58, and b_T3 = .59), which provides support for H2b (see Figure 3). Furthermore, the influence of cognition increases at T2 and drops again at T3 (b_T1 = .50, b_T2 = .54, and b_T3 = .48). The bs for conventional technology are relatively stable across time points for affect (b_T1 = .54, b_T2 = .54, and b_T3 = .54) and also for cognition (b_T1 = .55, b_T2 = .55, and b_T3 = .55) (see Figure 3), which provides support for H2a. The correlations between the constructs over time are low for affect, cognition, and overall attitude (Figure 3). Low correlations indicate that attitudes toward nanotechnology as well as conventional technology are not stable yet.

Knowledge

In the final step, the moderating effect of subjective and objective knowledge on the influence of affect and cognition on overall attitude for nanotechnology is investigated. In general, subjective and objective knowledge are low, but slightly increase over time, F_{subjective knowledge}(1, 5135) = 22.96, p < .001, M_T1 = 2.12, M_T2 = 2.26, and M_T3 = 2.47; F_{objective knowledge}(1, 4413) = 138.39, p < .001, M_T1 = 1.39, M_T2 = 2.33, and M_T3 = 2.58 (measured from -2 to +9, see Supplementary Appendix 5.1).

A model was estimated where the main effects of affect, cognition, subjective knowledge as well as the interactions of subjective knowledge with affect and cognition were allowed to differ over time for nanotechnology. For objective knowledge, a similar approach was used. A model where the interactions were not taken into account and the main effect of knowledge was constrained to be equal over time did not fit the data worse; subjective: Δχ² = 11.95, Δdf = 8, p = .150, objective: Δχ² = 11.23, Δdf = 8, p = .186. This shows that there is no moderating effect of subjective and objective knowledge. Thus, taking subjective and objective knowledge into account does not affect changes in the influence of affect and cognition on overall attitude over time. One reason for this might be that knowledge did not increase enough. Respondents did not read much about nanotechnology, and even less at T3 compared with T2, F(1, 5135) = 10.18, p < .001, M_T2 = 1.87, M_T3 = 1.81 (seven-point scale), did not watch more documentaries, F(1, 5135) = 18.30, p < .001, M_T2 = 1.78, M_T3 = 1.70, or searched the Internet for more information about nanotechnology (M = 1.50). Thus, over time people did not acquire more information on nanotechnology via the media.

5. Discussion

This study showed that the attitude formation process evolves differently for nanotechnology compared with conventional technology. For conventional technologies, the influence of affect and cognition on overall attitude stays stable over time. In attitude formation toward unfamiliar nanotechnologies, people rely relatively more on affect than cognition. Over time, for nanotechnology, reliance on affect decreases whereas reliance on cognition increases. At T2, the effect of cognition is higher for nanotechnology. Knowledge growth, neither objective nor subjective, does explain these changes.

Knowledge did not moderate the lower reliance on affect and the increase in reliance on cognition in attitude formation toward nanotechnology. Over time knowledge increased somewhat, but this has not been reflected in changes in attitude. This could be because knowledge toward nanotechnology and its applications is still low and within the 2.5 years of this study there has been no large increase in knowledge levels, in line with earlier findings (Reisch et al., 2011). A possible explanation is that respondents did not take up the information reported through different media sources or even because the amount of media coverage of nanotechnology was less between 2011 and 2014 than between 2009 and 2010.

Even though growth of formal knowledge was limited the results suggest that, over time, people started to think in a more cognitive way about nanotechnology. This suggests that valuation of nanotechnology applications have become somewhat more accessible in people’s knowledge structures. The relative influence of cognition for nanotechnology attitudes had, however, decreased once again by the end of the study. The reliance on affect at the start of the study can be understood as without knowledge individuals mostly need to rely on the few crystallized affective associations available (Slovic et al., 2007). If people are asked to respond to the same applications a year later, either media attention or other deliberations about nanotechnology are likely to have caused at least enough cognitions so that people can relate nanotechnology to existing products (Kahan et al., 2007). Moreover, people might be better able to draw cognition-based analogies with well-known products (as suggested for nanotechnology by Kuzma and Priest, 2010). Once such analogies are present, people classify and integrate unfamiliar nanotechnology into existing (technology) schemata (Davies, 2011; Kearnes et al., 2014). At first such classifications are likely to require cognitive reasoning, but once people’s classification has crystallized, more intuitive inferences may be extracted (Baylor, 2001). This means in turn that subsequent attitudes may be based more on affect (Pavelchak, 1989; Sujan, 1985), and affect becomes more dominant again. That affect remained more important for nanotechnology than for conventional theory suggests that the knowledge structure remains largely undeveloped, and perhaps even that with low media attention people revert to affect as the main guidance for their opinion (cf. Grobe and Rissanen, 2012). The current study did not assess whether, and if so, into which categories (schemata), nanotechnology was classified. It was neither studied whether it was the lasting lack of information that caused people to revert to affect in the later stages. In future research, a better understanding should be derived of the reasons behind changes in affect and cognition over time. For instance, experimental studies could be used to investigate how communication about nanotechnology and products with added nanotechnology can relate these products to other products (Siegrist et al., 2007). While the final explanation of why and how attitudes reverted to affect cannot be given, the current study showed that the evolution from an affect-based attitude for an unfamiliar product toward a more cognition-based attitude of similar products is not as straightforward as most theories would predict for a technology being part of the society for longer.

Attitudes toward both conventional and nanotechnology applications vary considerably over time (i.e. low correlations across time points). For conventional technology attitudes are anyway more crystallized than for nanotechnology; as for conventional technology, the internal weighing structure of affect and cognition remains stable over time. The lack of a stable long-term attitude for nanotechnology is an indication that attitudes are not strongly established yet and sensitive to change, which in turn makes it difficult for policy-makers and stakeholders to predict consumer behaviour (Petty et al., 1997). This also indicates that in case of negative incidents associated with nanotechnology it can lead to negative consumer opinion about the technology, and one single incident can trigger public backlash for other domains in which nanotechnology is utilized (Siegrist, 2010). A scrutiny of the media attention for 2013 shows that about 64% of the articles on nanotechnology are positive in tone, suggesting that such a negative trigger event had not yet occurred when data collection concluded.

The fact that consumer attitudes toward nanotechnology are not strongly established also means that attitudes still can develop in different ways and directions. One option is that knowledge will continue to grow while affect does not become dominant, in which case people rely most on cognitive cues. This does require motivation of consumers to inform themselves (Petty et al., 1997), which seems unlikely given that respondents in the current study indicated to not have searched for additional information on nanotechnology. So, it can be expected that besides cognition, affect will continue to play a major role. If a positive affective response develops, the technology perception will also be positive. Alternatively, a negative affective response may develop, which might lead to a general fear-response or aversion toward the technology. The fear of a fear-response is a scenario that is often reported when nanotechnology is paralleled to genetically modified organism (GMO) (Einsiedel and Goldenberg, 2004; Macoubrie, 2006; Sandler and Kay, 2006).

For policy-makers and stakeholders, to better connect with the general public, a design of information about nanotechnology and its applications is important. This study shows that currently public response is largely driven by affect. To align with this current dominance on affect, and at the same time contribute to a factual knowledge base, communication to the general public should be both cognitive and affective by nature. This combination seems important since it has been shown that factual information on its own is often of limited value in influencing consumer attitudes (Kahan et al., 2008). One way of designing this information is to provide people with information that connects to their existing knowledge structures, so that it becomes easier for people to integrate that unfamiliar instance in their current knowledge structure (Gregan-Paxton and John, 1997). For instance, the usefulness of nanotechnology applications in consumers’ daily life can be emphasized to increase their understanding. In the end, effective communication requires stepping back, assessing the extent of prior knowledge, and deciding how to communicate the basics of nanotechnology in such a way that it can be relevantly learned (Castellini et al., 2007). The media might operate as a primary source of science information for the public (Su et al., 2014). Effective communication strategies may lead to more consistent consumer opinions, which help policy-makers to anticipate trends that will dictate how the general public might react to new technology developments (Currall et al., 2006).

6. Conclusion

This longitudinal study investigated attitude formation for nanotechnology, an emerging technology, compared to conventional technologies. The affective–cognitive attitude structure for conventional technology was stable over time, whereas for nanotechnology this was not the case. This shows that consumer attitudes toward nanotechnology are not well-established yet, which in turn means that consumer attitudes are vulnerable to external impact. Therefore, the current study underlines the importance of addressing both affective and cognitive aspects of public opinion toward nanotechnology.

Footnotes

Acknowledgements

The authors thank the editors of Public Understanding of Science, and the journal’s anonymous reviewers for their helpful comments.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by NanoNextNL, a micro and nanotechnology consortium of the Government of the Netherlands and 130 partners. More information is available on .

Notes

Author biographies

Roxanne I. van Giesen is a Researcher in Consumer Behavior at CentERdata. She earned her PhD at the Marketing and Consumer Behavior Group at Wageningen University and has a (research) master’s degree in social and behavioural sciences from Tilburg University. Her research focuses on consumer response to unfamiliar technologies, focusing on intuitive, affective, and cognitive mechanisms in attitude formation.

Arnout R.H. Fischer is an Associate Professor at the Marketing and Consumer Behavior Group at Wageningen University. He studies consumer adoption of novel products and products produced with new technologies from the psychological perspective of the consumer.

Hans C.M. van Trijp is a Professor in Marketing and Consumer Behavior at Wageningen University. He is a Human Nutritionist by education, but moved early to the areas of consumer decision-making and choice behaviour, also in relation to marketing strategies and operations that facilitate uptake of products.

References

Ajzen

(2001) Nature and operation of attitudes. Annual Review of Psychology 52(1): 27–58.

Baylor

(2001) A U-shaped model for the development of intuition by level of expertise. New Ideas in Psychology 19(3): 237–244.

Bornstein

(1989) Exposure and affect: Overview and meta-analysis of research, 1968–1987. Psychological Bulletin 106(2): 265–289.

Cacciatore

Scheufele

Corley

(2011) From enabling technology to applications: The evolution of risk perceptions about nanotechnology. Public Understanding of Science 20(3): 385–404.

Carlson

Vincent

Hardesty

O’Bearden

(2009) Objective and subjective knowledge relationships: A quantitative analysis of consumer research findings. Journal of Consumer Research 35(5): 864–876.

Castellini

Walejko

Holladay

Theim

Zenner

Crone

(2007) Nanotechnology and the public: Effectively communicating nanoscale science and engineering concepts. Journal of Nanoparticle Research 9(2): 183–189.

CBS (2014) Bevolking; kerncijfers. Available at: www.cbs.nl

Clore

Huntsinger

(2007) How emotions inform judgment and regulate thought. Trends in Cognitive Sciences 11(9): 393–399.

Clore

Schnall

(2005) The influence of affect on attitude. In: Albarracín

Johnson

Zanna

MPE

(eds) The Handbook of Attitudes. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 437–489.

10.

Cobb

(2005) Framing effects on public opinion about nanotechnology. Science Communication 27(2): 221–239.

11.

Crites

Fabrigar

Petty

(1994) Measuring the affective and cognitive properties of attitudes: Conceptual and methodological issues. Personality and Social Psychology Bulletin 20(6): 619–634.

12.

Currall

King

Lane

Madera

Turner

(2006) What drives public acceptance of nanotechnology? Nature Nanotechnology 1(3): 153–155.

13.

Davies

(2011) How we talk when we talk about nano: The future in laypeople’s talk. Futures 43(3): 317–326.

14.

de Jonge

van Trijp

Renes

Frewer

(2010) Consumer confidence in the safety of food and newspaper coverage of food safety issues: A longitudinal perspective. Risk Analysis 30(1): 125–142.

15.

Desmet

PMA

(2003) Measuring emotion; development and application of an instrument to measure emotional responses to products. In: Blythe

AFMMA

Overbeeke

Wright

(eds) Funology: From Usability to Enjoyment. Dordrecht: Kluwer Academic Publishers, pp. 111–123.

16.

Edwards

(1990) The interplay of affect and cognition in attitude formation and change. Journal of Personality and Social Psychology 59(2): 202–216.

17.

Einsiedel

Goldenberg

(2004) Dwarfing the social? Nanotechnology lessons from the biotechnology front. Bulletin of Science, Technology & Society 24(1): 28–33.

18.

Fazio

(1990) Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. Advances in Experimental Social Psychology 23: 75–109.

19.

Fazio

(2007) Attitudes as object–evaluation associations of varying strength. Social Cognition 25(5): 603–637.

20.

Fischer

ARH

van Dijk

de Jonge

Rowe

Frewer

(2013) Attitudes and attitudinal ambivalence change towards nanotechnology applied to food production. Public Understanding of Science 22(7): 817–831.

21.

Fishbein

Ajzen

(1975) Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley.

22.

Frewer

Fischer

ARH

van Trijp

JCM

(2011) Communication of risks and benefits of nanotechnology: The issue of societal acceptance of emerging technologies. In: Frewer

Norde

Fischer

ARH

Kampers

FWH

(eds) Nanotechnology in the Agri-Food Sector. Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA, pp. 243–256.

23.

Gawronski

Bodenhausen

(2006) Associative and propositional processes in evaluation: An integrative review of implicit and explicit attitude change. Psychological Bulletin 132(5): 692–731.

24.

Gregan-Paxton

John

(1997) Consumer learning by analogy: A model of internal knowledge transfer. Journal of Consumer Research 24(3): 266–284.

25.

Grobe

Rissanen

(2012) Nanotechnologies in agriculture and food - An overview of different fields of application, risk assessment and public perception. Recent Patents on Food, Nutrition & Agriculture 4(3): 176–186.

26.

Gupta

Fischer

ARH

Frewer

(2012) Socio-psychological determinants of public acceptance of technologies: A review. Public Understanding of Science 21(7): 782–795.

27.

Gupta

Fischer

ARH

Frewer

(2015) Ethics, risk and benefits associated with different applications of nanotechnology: A comparison of expert and consumer perceptions of drivers of societal acceptance. NanoEthics 9(2): 93–108.

28.

Hansen

Wänke

(2009) Liking what’s familiar: The importance of unconscious familiarity in the mere-exposure effect. Social Cognition 27(2): 161–182.

29.

Kahan

Slovic

Braman

Gastil

Cohen

Kysar

(2007a) Nanotechnology risk perceptions: The influence of affect and values. In: Woodrow Wilson International Center for Scholars (ed.) Project on Emerging Nanotechnologies; Research Brief, vol. 5. Washington, DC: Woodrow Wilson International Center for Scholars, pp. 23–44.

30.

Kahan

Slovic

Braman

Gastil

Cohen

Kysar

(2008) Biased assimilation, polarization, and cultural credibility: An experimental study of nanotechnology risk perceptions. In: Woodrow Wilson International Center for Scholars (ed.) Project on Emerging Nanotechnologies; Research Brief, vol. 5. Washington, DC: Woodrow Wilson International Center for Scholars, pp. 45–56.

31.

Kearnes

Macnaghten

Davies

(2014) Narrative, nanotechnology and the accomplishment of public responses: A response to Thorstensen. NanoEthics 8(3): 241–250.

32.

Klerck

Sweeney

(2007) The effect of knowledge types on consumer-perceived risk and adoption of genetically modified foods. Psychology & Marketing 24(2): 171–193.

33.

Kline

(2005) Principles and Practice of Structural Equation Modeling. New York: The Guilford press.

34.

Kuzma

Priest

(2010) Nanotechnology, risk, and oversight: Learning lessons from related emerging technologies. Risk Analysis 30(11): 1688–1698.

35.

Lattin

Carroll

Green

(2003) Analyzing Multivariate Data. Pacific Grove, CA: Thomson Brooks/Cole.

36.

Lee

C-J

Scheufele

(2006) The influence of knowledge and deference toward scientific authority: A media effects model for public attitudes toward nanotechnology. Journalism & Mass Communication Quarterly 83(4): 819–834.

37.

Loewenstein

Weber

Hsee

Welch

(2001) Risk as feelings. Psychological Bulletin 127(2): 267–286.

38.

López-Vázquez

Brunner

Siegrist

(2012) Perceived risks and benefits of nanotechnology applied to the food and packaging sector in México. British Food Journal 114(2): 197–205.

39.

Macoubrie

(2006) Nanotechnology: Public concerns, reasoning and trust in government. Public Understanding of Science 15(2): 221–241.

40.

Maynard

(2006) Nanotechnology: Assessing the risks. Nano Today 1(2): 22–33.

41.

Nederlandse Organisatie voor Wetenschappelijk Onderzoek, NWO (2012) Roadmap nanotechnology in the top sectors. Available at: https://www.rvo.nl/sites/default/files/Roadmap%2010%20Nanotechnology%20Topsector%20HTSM%202013.pdf

42.

Ofir

(2000) Ease of recall vs recalled evidence in judgment: Experts vs laymen. Organizational Behavior and Human Decision Processes 81(1): 28–42.

43.

Pavelchak

(1989) Piecemeal and category-based evaluation: An idiographic analysis. Journal of Personality and Social Psychology 56(3): 354–363.

44.

Peracchio

Tybout

(1996) The moderating role of prior knowledge in schema-based product evaluation. Journal of Consumer Research 23(3): 177–192.

45.

Petty

Wegener

Fabrigar

(1997) Attitudes and attitude change. Annual Review of Psychology 48(1): 609–647.

46.

Pidgeon

Harthorn

Satterfield

(2011) Nanotechnology risk perceptions and communication: Emerging technologies, emerging challenges. Risk Analysis 31(11): 1694–1700.

47.

Plessner

Czenna

(2008) The benefits of intuition. In: Plessner

Betsch

(eds) Intuition in Judgment and Decision Making. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 251–266.

48.

Priest

Lane

Greenhalgh

Hand

Kramer

(2011) Envisioning emerging nanotechnologies: A three-year panel study of South Carolina citizens. Risk Analysis 31(11): 1718–1733.

49.

Reisch

Scholl

Bietz

(2011) ‘Better safe than sorry’: Consumer perceptions of and deliberations on nanotechnologies. International Journal of Consumer Studies 35(6): 644–654.

50.

Rosseel

(2012) Lavaan: An R package for structural equation modeling. Journal of Statistical Software 48(2): 1–36.

51.

Russell

(1980) A circumplex model of affect. Journal of Personality and Social Psychology 39(6): 1161–1178.

52.

Sandler

Kay

(2006) The GMO-nanotech (dis)analogy? Bulletin of Science, Technology & Society 26(1): 57–62.

53.

Satterfield

Kandlikar

Beaudrie

Conti

JEH

Herr Harthorn

(2009) Anticipating the perceived risk of nanotechnologies. Nature Nanotechnology 4(11): 752–758.

54.

Schenk

van der Maas

Smulders

MJM

Gilissen

LJWJ

Fischer

ARH

van der Lans

et al . (2011) Consumer attitudes towards hypoallergenic apples that alleviate mild apple allergy. Food Quality and Preference 22(1): 83–91.

55.

Scheufele

Corley

Shih

Dalrymple

(2009) Religious beliefs and public attitudes toward nanotechnology in Europe and the United States. Nature Nanotechnology 4(2): 91–94.

56.

Scheufele

Lewenstein

(2005) The public and nanotechnology: How citizens make sense of emerging technologies. Journal of Nanoparticle Research 7(6): 659–667.

57.

Schwarz

(2007) Attitude construction: Evaluation in context. Social Cognition 25(5): 638–656.

58.

Shiv

Nowlis

(2004) The effect of distractions while tasting a food sample: The interplay of informational and affective components in subsequent choice. Journal of Consumer Research 31(3): 599–608.

59.

Siegrist

(2010) Predicting the future: Review of public perception studies of nanotechnology. Human and Ecological Risk Assessment 16(4): 837–846.

60.

Siegrist

Cousin

M-E

Kastenholz

Wiek

(2007) Public acceptance of nanotechnology foods and food packaging: The influence of affect and trust. Appetite 49(2): 459–466.

61.

Siegrist

Stampfli

Kastenholz

Keller

(2008) Perceived risks and perceived benefits of different nanotechnology foods and nanotechnology food packaging. Appetite 51(2): 283–290.

62.

Slovic

Finucane

Peters

MacGregor

(2002) What’s fear got to do with it? It’s affect we need to worry about. In: Thomas Gilovich

Kahneman

(eds) Heuristics and Biases: The Psychology of Intuitive Judgment, Cambridge: Cambridge University Press, pp. 397–420.

63.

Slovic

Finucane

Peters

MacGregor

(2007) The affect heuristic. European Journal of Operational Research 177(3): 1333–1352.

64.

Steenis

Fischer

ARH

(2016) Consumer attitudes towards nanotechnology in food products: An attribute-based analysis. British Food Journal 118(5): 1254–1267.

65.

Stijnen

van Veggel

Fischer

ARH

Ronteltap

de Bakker

Minten

et al . (2011) Consumentenperceptie van nanotechnologieën in voedsel en landbouw: Een eerste verkenning [Consumer perception of nanotechnologies in food and agriculture: An initial exploration]. Wageningen: Wageningen UR Food & Biobased Research.

66.

LYF

Cacciatore

Scheufele

Brossard

Xenos

(2014) Inequalities in scientific understanding: Differentiating between factual and perceived knowledge gaps. Science Communication 36(3): 352–378.

67.

Sujan

(1985) Consumer knowledge: Effects on evaluation strategies mediating consumer judgments. Journal of Consumer Research 12(1): 31–46.

68.

van Giesen

Fischer

ARH

van Dijk

van Trijp

HCM

(2015) Affect and cognition in attitude formation toward familiar and unfamiliar attitude objects. PloS One 10(10): e0141790.

69.

Vandermoere

Blanchemanche

Bieberstein

Marette

Roosen

(2011) The public understanding of nanotechnology in the food domain: The hidden role of views on science, technology, and nature. Public Understanding of Science 20(2): 195–206.

70.

Wijnhoven

SWP

Dekkers

Hagens

de Jong

(2009) Exposure to nanomaterials in consumer products. RIVM letter report no. 340370001, 4 September, 2009. Bilthoven: RIVM.

71.

Zajonc

(1980) Feeling and thinking: Preferences need no inferences. American Psychologist 35(2): 151–175.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.29 MB