1.1 Central constructs and state of research

Before looking into the current state of research, the present study's key constructs “exposure perception” and “affect” as well as the related concept “risk perception” are defined.

1.1.2 State of research

With regard to the perception of mobile communications, RF-EMF risk perception can be considered well researched (e.g., Claassen et al., 2017; Cousin & Siegrist, 2010a, 2010b; Freudenstein et al., 2015; Freudenstein et al., 2015). Related concepts such as exposure perception or affective reaction have, by contrast, not been investigated much. However, empirical studies have shown that in the case of mobile communications, the affective reaction to a risk influences a persons’ risk perception (Freudenstein et al., 2015), and that exposure perception and risk perception correlate (Freudenstein et al., 2015). Nevertheless, it is unclear how the two constructs “affective reaction” and “exposure perception” regarding RF-EMF from mobile communications and particularly 5G relate to each other, that is, when laypeople explain their negative affect with perceived exposure.

The investigation of the perception of specifically 5G has so far primarily focused on general RF-EMF risk perception (Dilkova-Gnoyke et al., 2022; Frey, 2021; Koh et al., 2020) as well as on factors that influence the acceptance of 5G (Al-Maroof et al., 2021). Affective evaluation and RF-EMF exposure perception of 5G usage situations are therefore research gaps that the present study helps to close.

To learn more about the knowledge structures of laypeople, the mental models approach (Morgan et al., 2002) has proven to be useful. Mental models represent someone's existing ideas and beliefs about a certain topic that can be accurate or inaccurate. Consequently, they can be used to better understand what people (already) know, compared to expert knowledge. When people receive new information, their mental models can change. Therefore, various studies have investigated how the knowledge structures of laypeople and experts differ with regard to mobile communications in order to subsequently derive implications for science communication (Claassen et al., 2017; Cousin & Siegrist, 2010a). In the field of (RF-)EMF, various prevailing misconceptions have already been identified (Claassen et al., 2017; Cousin & Siegrist, 2010a, 2010b), some of which also allow assumptions about laypeople's RF-EMF exposure perception. A misconception that has been repeatedly identified is that the relative exposure to RF-EMF emitted by base stations is overestimated and that the relative exposure to RF-EMF emitted by mobile phones is underestimated (Claassen et al., 2017; Cousin & Siegrist, 2010b). For example, Cousin and Siegrist (2010a) found in a qualitative study that 12 out of 16 laypeople overestimated RF-EMF exposure from base stations.

With regard to mobile communications or other EMF sources, only few studies have so far focused on exposure perception (Claassen et al., 2017; Cousin & Siegrist, 2010a, 2010b; Freudenstein et al., 2015). Freudenstein et al. (2015) explicitly investigated the exposure perception regarding mobile communications by asking their respondents to rate a given set of EMF exposure characteristics that could potentially affect health risk perception. Those exposure characteristics included “strength of exposure,” “duration of exposure,” “distance to the EMF source,” “frequency,” and “number of exposure sources.” The authors concluded that the respondents correctly attributed a high level of importance to strength, duration, distance, frequency, and quantity. Other studies (Claassen et al., 2017; Cousin & Siegrist, 2010a) found that laypeople tend to misjudge the distance-exposure relationship. Additionally, Cousin and Siegrist (2010a) investigated the influence of the exposure source (antenna vs. phone) on exposure perception, and Claassen et al. (2017) looked into the location (indoor vs. outdoor). All those exposure characteristics are summarized in Figure 1. However, other characteristics not considered by the abovementioned authors could also influence laypeople's RF-EMF exposure perception.

1.2 The present study

In this study, we bring together two aspects that have hardly been linked in the literature so far: The affective reaction and the perceived exposure in situations where one is exposed to RF-EMF from mobile communications. To better understand the relationship between affect and exposure perception, two things are investigated: (1) How laypeople affectively evaluate everyday mobile phone usage situations (without explicitly being asked about RF-EMF exposure) and (2) how situational RF-EMF exposure is perceived. Besides, it is investigated whether 5G—as the latest mobile communication standard—is perceived differently from earlier mobile communication standards, for which studies on exposure perception are already available (Claassen et al., 2017; Cousin & Siegrist, 2010a; Freudenstein et al., 2015).

Besides the exposure characteristics investigated by the abovementioned authors, further exposure characteristics were identified through expert interviews for the present study and summarized in the form of a working model (see Figure 1). Although data transfer (upload vs. download) has an impact on exposure, it has not been researched as an exposure characteristic that may influence exposure perception yet. Due to the focus on 5G, network type and change (e.g., from 4G to 5G) have also been considered as exposure characteristics that can potentially influence exposure perception. Finally, visibility has no influence on RF-EMF exposure, but possibly on exposure perception.

The present study moves beyond the approach of Freudenstein et al. (2015) by asking participants to rank several RF-EMF exposure situations according to their perceived exposure as well as capturing their decision-making process. In this way, characteristics relevant to exposure perception can also be identified without specifically asking participants about them. Consequently, peoples reasoning is explored rather than imposed.

There are also important differences between our approach and the mental models approach (Morgan et al., 2002). The mental models approach aims to compare a lay model with an expert model. However, our focus was not on identifying knowledge gaps or misconceptions, but to capture the subjective, maybe incorrect perception of lay people regarding mobile communications. For this reason, a new, innovative approach was chosen for this study to capture participants perceptions on different dimensions.

2.1 Sample

A total of 35 individual interviews and 6 focus groups with the same participants (each group containing 5 –6 participants) were conducted between December 5 and 7, 2022. Participants were recruited from several locations in Southern Germany (the cities of Augsburg and Munich, surrounding area of Augsburg). Additionally, participants were recruited from areas where 5G rollout was at different stages at that time in order to obtain a sample that was as diverse as possible. The sample was composed of 18 women and 17 men. On average, they were 39 years old (age range 18–67). A total of 14 participants owned a 5G-enabled mobile phone. The distribution per group is presented in Table 1. Besides, participants differed in their primary occupation (e.g., full-/part-time employed, studying, and retired).

2.2 Data acquisition

Individual interviews and focus groups took place at the IU International University of Applied Sciences in Augsburg and were conducted by eight interviewers that were trained in advance. All focus groups were led by the same interviewer. Following Dilkova-Gnoyke et al. (2022), all participants of a focus group were first interviewed in parallel, before meeting in the focus group. This procedure enabled us to gradually explore the topic of “5G and mobile communications,” by first asking questions about individual opinions, impressions, and prior knowledge, without being influenced by the group. In contrast to the focus group, the individual interviews offered a more protected space in which the interviewees could share their views, as well as their knowledge and lack of knowledge, without potentially feeling social pressure from their peers. Consequently, the individual interviews captured the uninfluenced perception of 5G/mobile communications, whereas the focus groups required an active exchange of opinions and weighing up of arguments. The focus groups also provided an insight into how participants deal with contradictory arguments by other participants and react to explanations from others. Both the individual interviews and the focus group discussions were conducted using a (semi-) structured interview guide. This ensured that the same topics were discussed in all individual interviews and focus groups. Individual interviews were audio recorded, and focus group discussions were video recorded.

Among other things, the individual interviews assessed data, such as socio-demographic information, participants’ risk perception of certain pollutants, their understanding of risk, and 5G-related knowledge. The subsequent focus group aimed to assess participants’ affect toward mobile communications/5G, their mobile communications/5G exposure perception and risk perception, and base station siting preferences. Participants were also asked whether and which precautionary measures they take and how they assess scientific uncertainty. Participants were encouraged to share their personal opinions and were advised that there were no right or wrong opinions.

The results presented here focus on affective reaction and exposure perception, assessed in two interactive tasks in the focus groups (affective evaluation and ranking task), but also refer to statements regarding 5G and participants risk perception, captured in the individual interviews.

2.3 Materials

In a part of the individual interviews, participants were shown a sketch of a village with several potential risks depicted in or in the vicinity of it, such as a factory, a highway, an airplane, or a base station. They were instructed to describe the picture. Depending on whether the participants identified a potential hazard, they were asked to rate the risk associated with it: “How do you rate the various aspects [such as car fumes, noise, mobile communications, …] on a scale from 1 to 10, where 1 means absolutely harmless and 10 very dangerous?.” As we were mostly interested in the evaluation of mobile communications, for the subsequent analysis a distinction was made between people who mentioned the base station and rated it as >5 (high risk perception) and others.

The affective evaluation and exposure perception tasks were based on the same stimuli: pictures showing everyday situations where someone is surrounded by mobile communication technology, later on referred to as “mobile phone usage situations.” The aim was to capture participant's thoughts on these everyday mobile phone usage situations. Therefore, a diverse set of situations was shown (see Figure 2) in which a person is using a mobile phone (five pictures) or is in the vicinity of several mobile devices (one picture) or mobile phone base stations (three pairs of pictures). The base station situations were each showing a before-after-comparison, for example, an upgraded 4G base station, and consisted therefor of two pictures. Situations are shown as pictured in Figure 2—without titles or an explanatory text. However, some introductory context was given by the interviewer, like “imagine that you are the person in the picture and have not been home for three days. The view from your living room window has changed during this time.” Due to limited time, not all situations were discussed in all focus groups.

When selecting the situations, several exposure characteristics that may influence laypeople's exposure perception were considered (see Figure 1). As no complete variation of all characteristics identified was possible, it was decided to focus on quantity, change, proximity to particular body parts (head and reproductive organs), EMF source, network type, and data transfer. Still, most situations represented more than one exposure characteristic. For example, network type and data transfer were varied in the same mobile phone usage situations, with 5G reception and downlink in situation B, 5G reception and up- and downlink in situation C, 4G reception and uplink in situation H, and Wi-Fi reception and up- and downlink in situation A.

Network type and change were chosen because of our focus on 5G and its recent implementation, and data transfer was chosen as it had not been investigated yet. Quantity was chosen, because 5G requires more base stations than earlier mobile communication standards. EMF-source and proximity to particular body parts were chosen because a couple of misconceptions regarding these two exposure characteristics have been identified in the past. Further, only situations where devices or base stations are visible were shown, for example, through visible infrastructure instead of showing hidden base stations.

The order in which the pictures were introduced was the same in all groups (the order can be found in Figure 2). This procedure was deliberately chosen to obtain better comparability between the groups. Besides, the mobile phone situations were introduced first because “base stations are more closely related to risk concepts than home appliances” (Siegrist et al., 2006, p. 1021), and we were interested in the most intuitive evaluation of the situations, without imposing a focus on risk on the participants.

First, participants were asked to evaluate each situation affectively: “Please put yourself in the place of the person depicted in the picture. So, imagine that you are that person. Then please reach for the plus-card as quickly as possible if you have a good feeling. If you have a bad feeling, please reach for the minus-card.” After discussing their immediate affect toward each situation, participants were asked to rank the pictures according to their perceived exposure in the situations shown: “How would you arrange the pictures hierarchically? Please place the picture in which the person, i.e., you, is in your opinion most exposed to mobile communications in general in first place, and the picture in which the person, i.e., you, is least exposed to mobile communications in last place.” Once the participants had decided on an order, they could look at the rankings of the other participants. The reasons for their ranking were discussed afterward.

This two-step approach was chosen to initially receive an intuitive assessment from the participants of situations that are commonplace for them, but in which they probably think less about their exposure to RF-EMF. Thus, the concept of exposure was only introduced in a second step and the discussion was deliberately guided in this direction through the ranking task.

2.4 Data analysis

The recordings of the individual interviews and the focus groups were transcribed using the software F4transkript (dr. dresing & pehl GmbH, 2021) by three of the authors and two student assistants. A transcription guideline was used, which was based on Dresing and Pehl (2015) and adjusted to the study's circumstances. Afterward the qualitative data were structured using the software MAXQDA (VERBI Software, 2021). Guided by the initial RQs, categories were formed deductively and then extended inductively by additional categories, following the procedure recommended by Mayring (2015). The coding was carried out by one of the authors (SL), whereby each document was checked by a second person (ME or FA) and amended or discussed where necessary. Moreover, the results of the interactive tasks—the affective evaluation and ranking of exposure situations—were analyzed.

3.1 Affective evaluation and the role of exposure perception

The affective reaction to the different situations varied considerably. Although the videocall situation (A) was evaluated most positively, the additional base stations situation (F2) was evaluated most negatively. In general, each situation triggered positive and negative affect within our sample (see Table 2). Participants gave a variety of reasons for their evaluation, which are described in detail below. When talking about RF-EMF exposure, they were mostly referring to it as “radiation.”

Video telephony (A, discussed in six groups) was perceived positively by most participants because seeing each other is regarded as important (mentioned in four groups). It was reported that, compared to a conventional phone call, this form of telephony is closer to personal contact (mentioned in three groups), as people can see each other during the call. It was emphasized that video telephony is “more intimate” (female, 23, city), and one can have more profound conversations and better perceive feelings of the other person (mentioned in 3 groups). It is therefore not surprising that video telephony is mainly used in a private environment (mentioned in all six groups), but several participants also reported its use at work (mentioned in four groups) or at university/school (mentioned in three groups). “Radiation” was not mentioned in this context. The situation is different when talking on the phone (G, discussed in four groups) with a mobile phone to the ear. In this situation, “radiation” was mentioned in three focus groups, whereby participants from two groups were unsure whether their beliefs about exposure were correct: for example, “That's what they always say: At the ear is really bad. Most of the radiation. Is that true?” (female, 49, countryside). However, heat generation and headaches (each mentioned in one group) were also perceived as negative side effects independently from “radiation”. At the same time, the situation was perceived as ordinary (mentioned in two groups), and availability was also perceived as positive (mentioned in three groups).

Although some people emphasized the good 5G reception (mentioned in three groups) for video streaming (B, discussed in all six groups), participants from two of these focus groups pointed out that this is an activity they primarily perform with Wi-Fi. The fact that mobile use is generally possible, regardless of the Wi-Fi, was discussed in two focus groups. Although playing games on the phone was perceived as a waste of time (mentioned in three groups), informative videos—like instructions, learning and videos.—that can be accessed at any time were perceived positively (mentioned in four groups). Fun and distraction were discussed in two focus groups each as reasons for their positive affect, passing the time was discussed in one group. Regarding the upload situation (H, discussed in two groups), the fact that one can share information at any time was positively emphasized (mentioned in all two groups). 4G reception was rated positively by one group, with one person stating that: “5G would probably be even faster” (male, 27, countryside). However, one focus group also saw advertising and influencers as risks.

Triggered by the mobile phone in the front pocket (C, discussed in six groups), health concerns such as infertility as a result of RF-EMF were thematized in four focus groups, for example, “men shouldn't put the mobile phone in their trouser pockets, because otherwise the reproduction rate might drop a little” (female, 49, countryside). But participants were not sure whether their beliefs were right: “Whether that's true, I don't know” (same participant). In one group, “radiation” was mentioned, without deducing any consequences for health: “At first you think of the rays” (male, 22, countryside). The mobile phone was also experienced as too close to the body in this situation (mentioned in four groups). However, participants from four focus groups found it simply uncomfortable to carry the mobile phone in the front trouser pocket, and therefore had a negative affective reaction to the situation. On the other hand, having a mobile phone at hands reach made participants feel secure (mentioned in two groups).

The situation in which several people use a mobile phone (public transport, D, discussed in six groups) primarily triggered social concerns in the participants. The most common fear was a loss of sociability or a decrease in interpersonal communication (mentioned in five groups). Three groups discussed their concern of social decline, reflected in statements such as “head-down generation” (male, 42, city). However, participants from three focus groups stated that it was no different in the past and drew a comparison with reading books, among other things: “Well, I think when book printing really started to boom in practice, that there was exactly the same discussion” (male, 53, countryside).

The upgrade (E, discussed in six groups) of the base station from 4G to 5G was not perceived as a relevant change (mentioned in four groups), or it was emphasized that the base station had already been there (mentioned in three groups). Further, the upgrade made the network expansion visible, and participants were in favor of technological progress (mentioned in four groups). Visual aspects like defacement were a topic in two focus groups, and “radiation” was mentioned in four groups. On the one hand, “radiation” seemed to be a subconscious topic: “You always have the negative radiation in the back of your mind” (male, 32, countryside), but concrete ideas regarding exposure also influenced the evaluation: “I still have in my head that it is always said that the radiation is highest when you are right next to the mast” (male, 42, city). Again, participants were not sure whether their convictions were correct: “I don't know how it really is” (same participant). For the additional base stations (F, discussed in five groups), the visual aspect dominated the affective reaction (mentioned in four groups). Moreover, the usefulness of the additional antennas was questioned in four focus groups. One powerful base station would be preferred to three less powerful antennas by participants from three focus groups. Two focus groups talked about “radiation” that literally becomes visible here: “For me, this is visual: it radiates” (male, 56, city) and assumed that more base stations would lead to more “radiation”: for example, “I find this radiation more dangerous now, when there are three [antennae] instead of one” (female, 41, countryside). The new site (I, discussed in two groups) was perceived as too close (mentioned in all two groups). Likewise, participants would have liked to have a say in the construction of a new site (mentioned in one group) or expected to be informed in advance (mentioned in one group). “Radiation” was mentioned as a criterion in one focus group, whereby the respective participant referred to her own negative experiences with “electro smog.”

In particular, participants who showed high risk perception toward mobile communications in the individual interviews (values >5 on a scale from 1 to 10) rated those situations in which mobile phone base stations are central predominantly negatively. The upgrade situation only received two positive ratings from participants with high risk perception of mobile communications. In contrast, five participants evaluated the situation negatively. Among those with high risk perception, the additional base stations situation and the new site situation only triggered negative or neutral affect.

To sum up, situations that show the use of a mobile phone mostly led to a positive affective reaction by the participants. The evaluation was based on, among other things, the fact that they themselves use their mobile phone in the situations shown and that it enables them, for example, to stay in touch with family and friends, to entertain themselves or to access information at any time. Negative affect, on the other hand, was a reaction to the concern of social decline, esthetic aspects, or a lack of understanding for the new/expended site, and “radiation.” Although video telephony, streaming and uploading in particular evoked a positive affect, new or additional base stations triggered a negative affective reaction, as did situations in which the mobile phone is close to the body (RQ1). RF-EMF only played a subordinate role with 12 of the 35 participants naming “radiation” as reason for their negative affect at least once. Nevertheless, associations with RF-EMF seem to be triggered particularly by situations where the mobile phone is close to the body or base stations are visible (RQ2).

3.2 Exposure perception

The ranking task and the subsequent explanation of the exposure perception revealed a wide variety of beliefs among the participants, which were often very individual. The presentation of the results starts with the exposure characteristics that have already been investigated by other researchers, followed by the exposure characteristics we have identified, and exposure impact beliefs participants brought up themselves. Figure 3 shows three examples of how the situations were ranked by the participants.

3.3 5G versus Earlier mobile communication standards

Both the affective reaction to the situations and their ranking according to the perceived exposure (see Sections 3.2, 3.2.4, and 3.2.5) have shown that 5G and earlier mobile communication standards are perceived differently in some respects. Both the individual interviews and the focus groups revealed that the participants had the following convictions: The vast majority of participants (n = 30) believed that 5G will transmit data faster than previous mobile communication standards, resulting in a higher data transmission rate (n = 9). However, there was greater disagreement among participants with regard to the change in RF-EMF exposure. Although one participant assumed that RF-EMF exposure would decrease with 5G, participants from five focus groups came to the opposite conclusion. Several participants also assumed that 5G will require different (n = 7) and more (n = 5) mobile phone base stations than its predecessor, whereas the possibility of a decrease of the number of base stations was not mentioned at all.

To summarize, participants revealed beliefs about, for example, changing infrastructure or new frequencies for 5G. Although higher exposure compared to 4G (or Wi-Fi) was primarily associated with base stations, 5G reception on mobile phones played a rather subordinate role. However, 5G is not perceived as a fundamentally new technology by the participants, but rather as a continuation of mobile communications technology (RQ4).

4.1 Limitations

It should be noted that the sample is quite small due to the qualitative design and is not representative of relevant populations. In order to obtain the broadest possible spectrum of potential responses, the vast majority of questions were formulated in an open manner. Consequently, it is possible that more people than indicated share a certain view or support certain arguments but did not actively express them. Due to the focus group setting, it is possible that people may not have wanted to repeat an argument that had already been mentioned. At the same time, it is possible that an argument only became relevant for someone because another person brought it up. It is also possible that the topic only became relevant for the participants because they were confronted with it in the study situation (Zwick, 2005). Consequently, this challenge in measuring exposure perception could not be met with the study design. In addition, there was no complete variation of exposure characteristics within the situations that would allow for all systematic comparisons. This makes it difficult to attribute the exposure perception and results of the ranking task to a single criterion, such as 5G reception. These limitations could be overcome by collecting data quantitatively and systematically comparing situations that differ only in the criterion one is interested in.

Another limitation is the fact that the stimuli served as a starting point for the discussion and therefore also evoked associations related to them. In some cases, the stimuli were reinterpreted by the participants so that they were not exactly evaluating the situation depicted (both affectively and in terms of the associated exposure). For example, some participants assumed that they would lock their mobile phone when they had it in their (trouser) pocket, and it would therefore be inactive, whereas the sketch showed data being sent and received. This could possibly be avoided by asking participants not to put themselves in the situation and apply it to themselves, but to evaluate it for the person depicted.