Maximin principle, emotional aversion, and integrative judgment in the NIMBY context, including social dilemma and moral dilemma: The roles of the amygdala, angular gyrus, and ventromedial prefrontal cortex

ABSTRACT

Public facilities that have NIMBY (not in my backyard) structure involve both a social dilemma, in which individuals’ decisions to prevent the worst outcomes for themselves undermine the public interest, and a moral dilemma focused on the majority versus the minority. This study examined the cognitive-neural processes in judging whether to prioritize the site residents or the citizenry as a whole within the context of NIMBY. Our ROIs were the right angular gyrus being related to concern about the worst possible outcomes for others and oneself, the amygdala associating with emotional aversion to prioritizing the majority, and the vmPFC, which integrates the aversion into “all things considered” judgments. As a result of comparing ingroup conditions for which a NIMBY facility may make participants worst-off position and outgroup conditions for which this possibility is denied, the right angular gyrus was activated in both conditions. The amygdala was activated only in the ingroup, and the vmPFC exhibited a stronger tendency in the ingroup. We concluded that the cognitive-neural processes in judgments on NIMBY facilities are common to both decision-making to avoid the worst-off position for others and for oneself and moral judgments between the majority and the minority.

KEYWORDS:

• NIMBY
• right angular gyrus
• amygdala
• vmPFC

Introduction

NIMBY (Not in my backyard) refers to public facilities that bring public benefits to a majority of the society, but also pose various risks and damages to the locals of the region, as well as people’s reactions (judgments, behaviors, etc.) to these facilities. Public facilities such as airports and garbage disposal sites are considered NIMBY-type facilities. They provide public benefits to most of society but have negative external effects, such as safety and health hazards for residents of the location sites (O’Hare, 1977; Sakai, 2012). NIMBY facilities contain the structure of a social dilemma because the public interest is undermined by people’s opposition to placing them in nearby locations that would make those people worst-off position (Easterling, 2001). NIMBY contexts present structures of moral dilemmas focusing on the judgment between the majority and the minority, due to a conflict of interest between the majority in society and the residents of the location site (Smith & Desvousges, 1986). While numerous studies have investigated human decision-making in the NIMBY context over the past 2 decades (Foster & Warren, 2022; Wexler, 1996; Wolsink, 2007), no attempt has been made to examine the regions in the brain associated with such decisions.

Through presenting a vignette on site selection for a NIMBY facility, Nonami et al. (2022) showed that public benefits diminish if the rights of residents of the site are prioritized over government agencies and the facility cannot be sited. They described two scenarios in which the location of a NIMBY facility was a topic in the country to which the participants belonged (ingroup) or in a foreign country unrelated to them (outgroup). Participants in the ingroup condition that they could be worst off from the nearby location of NIMBY facility, while this could be denied for the outgroup. Although the participants understood the social dilemma structure inherent in these scenes, they prioritized the rights of the residents over government agencies in both conditions. According to Nonami et al. (2022), respondent judgments were based on two determinants: the first is the difference principle (maximin principle) that induces attention to the worst-off position between self and others (Rawls, 1999), and the other is empathic concern, which leads to non-utilitarian judgments in moral dilemmas (Gleichgerrcht & Young, 2013).

Kameda et al. (2016) showed that attention directed to minimizing the worst allocation to the self is associated with the angular gyrus in the right temporo-parietal junction (rTPJ). The rTPJ is also strongly associated with “theory of mind,” which is the capacity to infer other’s mental states (Decety & Lamm, 2007). The rTPJ is activated as a common cognitive-neural process not only for self-allocation but also for other-allocation (Kameda et al., 2016). Thus, the rTPJ is assumed to support decisions for minimizing the worst allocation to self and others in the NIMBY contexts.

In contrast, studies on moral judgments in the past two decades have increased our understanding of automatic and controlled processes that are involved in the production of moral judgments. First, non-utilitarian/deontological judgments in moral dilemmas are supported by automatic and emotional responses by the amygdala (Greene et al., 2004, 2008; Paxton et al., 2012). In particular, the amygdala is activated in personal dilemmas that require intentional harm to others and evokes moral aversion to utilitarian behaviors (Greene et al., 2004). Additionally, Shenhav and Greene (2014) tested whether moral judgment depends on integrating emotional responses with utilitarian assessments, based on the theory of integrative moral judgment suggested by studies of nonmoral decision-making in humans and other primates (Grabenhorst & Rolls, 2011; Rangel & Hare, 2010). Accordingly, the amygdala enables automatic emotional responses to utilitarian behaviors, whereas the ventromedial prefrontal cortex (vmPFC) integrates such responses into “all things considered” judgments weighing them against utilitarian consideration, and make assessments of overall moral acceptability. The judgments in the context of NIMBY, which presents structures of moral dilemmas, is also credited to be supported by the emotional responses evoked by the amygdala and the vmPFC integrating such responses.

Based on this, it is hypothesized that participants asked to give priority to the rights of either residents at the site of construction of NIMBY facilities or nationals/citizens as a whole would likely prioritize the residents, in both ingroup and outgroup scenarios (hypothesis 1). Because participants pay attention to neighborhood locations that could make themselves or others worst-off position, rTPJ (particularly the right angular gyrus) should be aroused in both ingroup and outgroup scenarios (hypothesis 2). Furthermore, Everett et al. (2016) reported that individuals who make utilitarian judgments in moral dilemmas are considered to be less trustworthy than non-utilitarian individuals and are less likely to be preferred as partners in resource exchanges. These negative assessments from others are more serious in the ingroup than in the outgroup, and the fear of such estimations may be more pronounced in an ingroup scenario. Therefore, we hypothesize that participants’ moral aversion to utilitarian actions may be more salient in the ingroup scenario, and thus the amygdala and vmPFC are more activated in that scenario than in the outgroup (hypothesis 3). These hypotheses are examined using tasks in which participants judge which of the decision-making rights of two actors (site residents or nationals/citizens) should be prioritized in the context of NIMBY facilities within or outside their own community (ingroup or outgroup).

Method and materials

Participants

The statistical sample size required to obtain a significance level of 5% on a corresponding t-test, 80% power, and an effect size of 0.5 on Cohen’s d was calculated to be N = 34. Assuming a 10% dropout, the final sample size was determined to be 38 participants. In all, 38 healthy undergraduates reporting no history of psychiatric or neurological disorders participated in the fMRI experiment. All of them were confirmed to be right-handed by the Edinburgh handedness inventory (Oldfield, 1971). Glasses/contact lens users were provided MRI-compatible glasses. Four participants who had more than 3 mm head movement were excluded. Ultimately, data from 34 participants (16 females, mean age 20.9 [19–22] years) were analyzed. Informed consent was obtained from each participant after providing an explanation of the experimental procedure, which approved by the Graduate School of International Cultural Studies, Tohoku University (2021 No.7).

Previous tasks

In this study, we first provide an overview of the participants’ judgment patterns for each of the moral dilemmas and the NIMBY dilemmas. For this purpose, three moral dilemmas, such as the trolley dilemma, were presented to participants as a previous task before undergoing fMRI. Unlike the NIMBY dilemma tasks presented later in the fMRI, these three were purely moral dilemma tasks that did not involve a social dilemma structure where nonutilitarian judgments to help the minority (one person) interferes with the overall interests involving both the majority (five persons) and the minority. In these moral dilemmas, participants were asked to make ethical evaluations of a utilitarian judgments to save five people in exchange for one (the participants could choose to response that pulling the lever to save five people is not wrong ethically or that it is wrong) and make decisions to indicate whether they would themselves actually perform utilitarian actions (one of two choices was made: “I pull the lever to switch the track” or “I do not pull the lever”). After completing these tasks, the participants underwent fMRI and were presented with the NIMBY dilemmas.

Task design

During fMRI, the participants were presented with 48 experimental tasks (NIMBY tasks for ingroup and outgroup conditions, 24 tasks each) describing the location of a NIMBY facility and 24 filler tasks. To manipulate the category factor (ingroup/outgroup), the NIMBY tasks were designed as follows. First, in the 24 tasks for the ingroup condition, the locations of NIMBY facilities, such as waste disposal sites and oil storage bases, were problematized in the countries or towns where the participants reside (screens 1–3, see Figure 1 for presentation time for each screen). Subsequently, in these tasks, it was presented that the NIMBY facilities might be located in the vicinity of the participants themselves (screen 4). Conversely, in the 24 tasks in the outgroup condition, we indicated that the location of the same NIMBY facilities was a matter of concern for foreign countries or distant towns, and there was no possibility that they would be located in their own neighborhood. In both the ingroup and outgroup conditions, it was presented to them that NIMBY facilities bring public benefits to the entire country or town, while having various adverse effects on the residents living at the site (screen 3). Finally, the participants were asked to judge which party should be given priority decision-making rights on the location of the NIMBY facility: nationals/citizens as a whole (majority), who are the ones who obtain the public benefits, or residents of the location site (minority), who are the ones who will suffer negative external effects (screen 5). Total presentation time per task ranged from 29.5 to 42.5 s, including the ITI between screens 4 and 5 (2–6 s). Random 3–7 se inter-item time pauses were provided between tasks. The participants also made judgments irrelevant to NIMBY facility questions in the filler tasks. Each task was presented on an MRI-compatible liquid crystal display (1920 × 1080 resolution).

Figure 1. Task design.Category factor (in group/outgroup) is manipulated in screen 1. Next, an outline of the NIMBY facility is presented (screen 2), followed by an outline of the minority (residents of the site) and majority (nationals/citizens as a whole) who would be affected by whether to locate (or not to locate) the facility (screen 3). After the category factor is re-emphasized in screen 4, the participants judge whether to give priority to the site residents or to the nationals/citizens as a whole (screen 5). Immediately before screen 1 and between screens 4 and 5, a fixation is inserted as an ITI.

The total 72 tasks were randomly divided into four blocks of 18 tasks each, and presented in block-randomization. To neutralize the position effect, 72 identical tasks were prepared, such that the left and right choices (majority and minority in the NIMBY tasks) were reversed in screen 3 and screen 5 and counterbalanced across participants.

Behavioral analysis

The behavioral data in this study is the percentage of each participant’s judgments (deciding questions of priority between the site residents and nationals/citizens as a whole) in NIMBY tasks. We calculated the percentage of judgments that prioritize the site residents or nationals/citizens for each ingroup and outgroup in total 48 NIMBY tasks. If a participant did not make a decision within the presentation time of the final stage (9 s for screen 5), the trial was excluded. An average of 1.1 trials were excluded among the 24 NIMBY tasks in the ingroup and outgroup for each participant (median = 1.0 in the ingroup, 0.5 in the outgroup, both 4.5%). We analyzed these behavioral data using a linear mixed-effects model with category (ingroup or outgroup) and judgment (site residents or nationals/citizens as a whole) as fixed effects.

Image acquisition

Functional images were acquired using a Philips 3T-MRI Achieva dStream. We adopted Echo Planar methods (EPIs) with parameters repetition time (TR) = 2500 ms, echo time (TE) = 30 ms, FA (flip angle) = 85°, FOV (field of view) = 192 mm², scan matrix = 64², and 3 mm iso voxel. The first two dummy volumes for each of the four blocks that were block-randomized and presented as described above were discarded to stabilize the T1-saturation effects. High-resolution T1-weighted anatomical images were also acquired using the following parameters; TR = 11.1 ms, TE = 5.1 ms, FOV = 256 mm², scan matrix = 368², 0.7 mm iso voxel.

fMRI analysis

The following preprocessing procedures were performed using Statistical Parametric Mapping (SPM12) software (Wellcome Department of Imaging Neuroscience; London, UK) implemented on MATLAB R2021b (MathWorks; Natick, MA, USA): realignment, slice timing correction, coregistration, segmentation, spatial normalization, and smoothing, using a Gaussian kernel with a full width at a half-maximum value of 8 mm.

A voxel-by-voxel, multiple-regression analysis of the expected signal change was applied to the preprocessed images for each participant. A standard event-related convolution model that used the canonical hemodynamic response function provided by SPM12 was employed. In this model, screens 1–5 (Figure 1) were set as explanatory variables in each of the two NIMBY conditions and the filler condition. The durations for screen 1–4 were the presentation time for each screen (2, 5.5, 14, and 6 s, respectively), and for screen 5, the median reaction time for all participants, 1.646 s, was applied. Realignment parameters for each session and a high-pass filter (128 s) were included in this model to remove low-frequency noise.

Then, hypothesis-driven region of interest (ROI) analyses were performed using parameters of choices (screen 5). We investigated the right angular gyrus, amygdala, and vmPFC as ROIs to examine the above hypotheses 1–3. Each ROI was created using an automated anatomical labeling atlas (Angular_R, Amygdala_L/R, and Frontal_Med_Orb_L/R, respectively) (Tzourio-Mazoyer et al., 2002). Parameter estimates for each ROI in ingroup and outgroup conditions were extracted by MarsBaR (http://marsbar.sourceforge.net/), and one-way repeated-measures ANOVAs was performed to examine the effects of the conditions.

Results

Previous tasks and behavioral results

First, we analyzed both the ethical evaluation of the utilitarian judgments and the decisions to perform utilitarian actions in the previous task to identify the moral judgment and actions of the participants. In the three trolley-type dilemmas, 83.3% of all responses (85/102) indicated that utilitarian judgment was “not wrong.” By contrast, only 44.1% (45/102) answered that they would actually perform the utilitarian action (McNemar – Test, P = 0.001). Although most participants considered utilitarian judgments to be ethically correct, they were less likely to actually perform the actions.

Next, we applied a linear mixed-effects model with fixed effects of category (ingroup or outgroup) and judgment (site residents or nationals/citizens) to the percentage at which either the site residents or nationals/citizens were prioritized in the NIMBY tasks. The fixed effect of category was not significant (t₍₉₉₎ = −0.83, P = 0.412, with a 95% confidence interval [CI] of −0.01 to 0.01), but the effect of judgment was significant (t₍₉₉₎ = 5.05, P = 0.001 with a 95% CI of 0.16 to 0.37), as well as their interaction term (t₍₉₉₎ = 4.06, P = 0.001 with a 95% CI of 0.07 to 0.19). The goodness of fit for the model was found to be significant (χ² = 74.26, df = 4, P = 0.001, AIC = 103.04). As shown in Figure 2, in both the ingroup and the outgroup conditions, nationals/citizens were given priority over the site residents. However, it was also observed that fewer judgments prioritized nationals/citizens in the ingroup condition than in the outgroup, with a significant group simple slope (t₍₉₉₎ = 4.04, P = 0.001). Conversely, judgments prioritizing residents of sites were less in the outgroup than in the ingroup, and the simple slope of the group was still significant (t₍₉₉₎ = −4.07, P = 0.001). In the ingroup condition, 39.8% (309/777) of judgments prioritized to the site residents, and 60.2% (468/777) gave priority to the nationals/citizens. In the outgroup, these values were 33.6% (261/777) and 66.4% (516/777), respectively. At the level of judgment rather than action intention, a higher percentage preferring the majority over the minority was common in both moral and NIMBY dilemmas.

Figure 2. Choice rates in favor of residents of the site or nationals/citizens after arcsine transformation under the in group and outgroup conditions (error bars show SEMs).

Neuroimaging results

We then analyzed the ROIs associated with our hypotheses. In this case; however, to place the judgment (site residents or nationals/citizens) factor as an independent variable, it is essential to have contrast estimates of the ROIs for the preference of either site residents or nationals/citizens. Accordingly, each of the four blocks of 72 tasks (24 each for in- and outgroups and 24 filler tasks), randomly divided into 18 each, is required to　contain ≥1 preference for site residents and nationals/citizens in each in- and outgroup. However, 24/34 participants did not complete all of these four preferences (two preferences in each in- and outgroup) in any blocks. Therefore, we removed the judgment (site residents or nationals/citizens) factor from the analysis and used only category (ingroup/outgroup) as an independent variable in subsequent analyses.

Right angular gyrus activity

Our hypothesis 2 predicts that the rTPJ (in particular, the right angular gyrus) is activated in both the ingroup and outgroup conditions. As a result of examination of the activity of the anatomically defined right angular gyrus (Figure 3(a)), the parameter estimate in the NIMBY tasks was M = 1.78 in the ingroup and M = 1.74 in the outgroup, and both values were significantly higher than the flat value (0) (one-sample t-test, both t₍₃₃₎ >3.25, p < 0.01, Figure 4). A linear mixed-effects model with a fixed effect of category (ingroup or/outgroup) did not show significant category fixed effect (t₍₃₃₎ = −0.15, P = 0.885 with a 95% CI of −0.72 to 0.62). The right angular gyrus was activated in both ingroup and outgroup.

Figure 3. Anatomically defined right angular gyrus (a), amygdala (b), and vmPFC (c).

Figure 4. Parameter estimates of the right angular gyrus under the in group and outgroup conditions (error bars show SEMs).

Amygdala activity

To investigate hypothesis 3 and examine whether the amygdala and vmPFC are more activated in the ingroup, we first analyzed the activity of the amygdala (both sides) during NIMBY tasks (Figure 3(b)). The parameter estimate for the amygdala was M = 0.81 in the ingroup condition and M = 0.05 in the outgroup condition (Figure 5). Only the former value was significantly higher than the flat value (0) (one-sample t-test, t₍₃₃₎ = 2.83, p < 0.01). A linear mixed-effects model with a fixed effect of category (ingroup or outgroup) was employed, and the fixed effect of category was found to be significant (t₍₃₃₎ = −2.18, P = 0.037 with a 95% CI of −1.46 to −0.05), supporting higher amygdala activity in the ingroup than in the outgroup. However, as mentioned above, we were unable to provide judgment (site residents or nationals/citizens) as an independent variable in this study, so the original purpose of hypothesis 3 that moral aversion to utilitarian judgments is stronger in the ingroup than in the outer group, could not be tested in detail.

Figure 5. Parameter estimates for the amygdala under the in group and outgroup conditions (error bars show SEMs).

vmPFC activity

Finally, we analyzed the activation of vmPFC (Figure 3(c)). The vmPFC is a region of the default mode network in the brain, and it is inactivated during task performance, so it can be predicted that the parameter estimate will be negative. The parameter estimates of vmPFC were M = −2.14 during NIMBY tasks in the ingroup condition, and M = −3.09 in the outgroup condition, both of which were significantly lower than the flat (0) (one-sample t-test, both t₍₃₃₎ < −3.52, p < 0.001; Figure 6). A linear mixed-effects model with a fixed effect of category (ingroup or outgroup) showed a fixed effect of category with a significant trend (t₍₃₃₎ = −1.89, P = 0.068 with a 95% CI of −1.96 to 0.07). As shown in the results, when the activity of vmPFC is negative in both the conditions, the value closer to the flat (0) is generally interpreted as having less deactivation, i.e., relative activation. Thus, the vmPFC during the NIMBY tasks tended to recruit in the ingroup rather than in the outgroup.

Figure 6. Parameter estimates for vmPFC under in group and outgroup conditions (error bars show SEMs).

Discussion

The NIMBY problem is a social dilemma, but at the same time, it can be regarded as a kind of moral dilemma that requires the judgment whether a majority or a minority should be prioritized. This study investigated three hypotheses in relation to the judgment between a majority and a minority, in an ingroup condition where the worst possible outcomes may be allocated to participants themselves and an outgroup condition where no possibility of being worst off follows from the location of the NIMBY facility.

The behavioral data showed that many judgments (about 60%) prioritized the rights of nationals/citizens as the majority in both ingroup and outgroup conditions. This is inconsistent with Nonami et al (2021, 2022). This does not support hypothesis 1 that the rights of the site residents would be prioritized for both ingroup and outgroup. This inconsistency may have arisen from the difference in choices between our study and those of Nonami et al (2021, 2022). The options in the present study were site residents or nationals/citizens in the context of the NIMBY facilities, whereas Nonami et al. (2021, 2022) presented pairs of residents and government agencies. Residents of the site and nations/citizens form a comparison on a criterion of minority/majority, while comparisons between residents and government agencies can be influenced by additional factors, such as ideological evaluation of the government. Furthermore, this study indicated that more than 80% of responses for ethical evaluation were “ethically not wrong” for the utilitarian judgments (saving five people in exchange for one) in the moral dilemmas completed as tasks before the presentation of the NIMBY tasks; however, only 44% of responses indicated that they would actually perform the utilitarian action. The participants in this study are Japanese, and the Japanese and Chinese view utilitarian judgments to save the majority by active intervention in typical moral dilemmas, such as the runaway trolley scenario, as “not wrong” to the same degree as Westerners in terms of their ethical evaluation. Conversely, their behavioral intentions toward utilitarian actions are reported to be lower than those of Western people (Gold et al., 2014; Yamamoto & Yuki, 2019). Social ecological reports indicate that East Asians such as Japan and China have less relational mobility in interpersonal situations (i.e., less freedom to form new interpersonal relationships and maintain or dissolve existing relationships) than North Americans (Yamada et al., 2017; Yuki et al., 2013). People in this region are oriented toward avoiding negative reputations from others in fixed relationships, and thus are reluctant to express artificial interventions such as “changing the course of the trolley” in moral dilemmas (Yamamoto & Yuki, 2019). Because the NIMBY tasks in this study only asked for judgments and did not involve the performance of the action, the ethical judgments in the moral dilemma were more relevant.

In the NIMBY dilemma, however, s/he who makes the “majority versus minority” judgment and takes action is in the position of a concerned party whose own interests are affected by their own judgments and actions. This is an important difference from a moral dilemma, such as a runaway train scenario, where the decision is made as a third party. Making judgments in the NIMBY dilemmas would anticipate the possibility that they themselves could be either the majority or the minority. Here, this was particularly apparent in the ingroup condition. In the NIMBY dilemma, a utilitarian judgment that prioritizes the rights of the majority (nationals/citizens) is a choice that places oneself in a free-rider position when on the majority side while maximizing the possibility of a worst-off position for oneself aligning with the minority (site residents). Since no cultural differences affect these effects in the NIMBY dilemma, there should be no social ecological differences in people’s judgments and behavioral intentions. Regardless, additional evidence is needed to determine whether the differences between judgments and behavioral intentions observed in the NIMBY dilemma in this study can be replicated in other cultures.

The right angular gyrus was activated in both the ingroup and outgroup conditions. This result supports hypothesis 2. According to Kameda et al. (2016), the right angular gyrus is associated with the maximin principle whereby people aim to avoid the worst-off position in resource allocations. Therefore, the activation of the right angular gyrus in the ingroup condition of the present study is considered to indicate an arousal of attention for the participants themselves in respect of their worst-off position. In addition, the right angular gyrus is associated with empathy for others and in paying attention to the potential of others becoming worst off as well as of the self (Decety & Lamm, 2007; Kameda et al., 2016). In this study, the activation of the right angular gyrus in the outgroup condition was interpreted to indicate the arousal of attention toward the payoff for others independent of the participants themselves, based on their empathy for others. Consistent activation of the right angular gyrus in both ingroups and outgroups suggests that this region functions as a common cognitive-neural process arousing attention to themselves taking of the worst-off position and to others in the context of NIMBY facilities, which includes social and moral dilemma structures.

Additionally, we predicted that moral aversion to utilitarian actions and assessments of overall moral acceptability integrating the aversion in the context of NIMBY facilities would be more salient in the ingroup condition (hypothesis 3), following Everett et al. (2016) and Greene et al. (2004). The results of our experiment showed that amygdala activation in the ingroup condition was significantly higher than in the outgroup condition, indicating that moral aversion in making a “majority versus minority” judgment in the NIMBY dilemma is more likely to manifest in the ingroup to which the individual making the judgment belongs. However, this study was unable to isolate the moral aversion corresponding to each of the “majority versus minority” judgment, so we cannot assert that this result is a strict pattern of moral aversion to utilitarian actions. Thus, hypothesis 3 above received only partial support with respect to moral aversion.

Furthermore, the vmPFC, which integrates moral aversion and utility to make “all things considered” judgments (Shenhav & Greene, 2014), showed negative values for both the out- and ingroup, with parameter estimates closer to flat (0) in the latter than in the former. If a region comprising the default mode network, such as the vmPFC or the parietal/posterior cingulate cortex, shows negative parameter estimates in either of the two conditions, it is generally interpreted as the region being relatively more active when the parameter is closer to flat (0) (Lombardo et al., 2010; Moran et al., 2009). Based on these suggestions, in this study vmPFC tends to be more strongly related to judgments concerning NIMBY facilities in the ingroup than in the outgroup. This result is consistent with hypothesis 3. However, the parameter estimates of vmPFC only showed a significant tendency between the ingroup and the outgroup. Therefore, the hypothesis that vmPFC arises more conspicuously in the ingroup than in the outgroup was not explicitly supported. In the case of the development of NIMBY facilities, it can also be concluded that the amygdala, associated with moral aversion to utilitarian actions, is more prominently activated in the ingroup, but the vmPFC, which integrates aversion and utility, arises consistently in both ingroup and outgroup. The function of vmPFC in relation to judgments on NIMBY facilities must be the object of close scrutiny in the future.

Our experiment indicated that the right angular gyrus, which is associated with the maximin principle involved in resource allocation for self and others (Kameda et al., 2016), is also activated in the context of the NIMBY problem. Furthermore, this study suggested the involvement of the amygdala and vmPFC in judgments of majority versus minority in the context of NIMBY facilities, as with studies on judgments of five people versus one person in moral dilemmas (Greene et al., 2004; Shenhav & Greene, 2014). From these results, we can conclude that the cognitive-neural processes concerned in judgments regarding NIMBY facilities are common for decision-making in resource allocations to the self and others and judgments in moral dilemmas.

In this study, the activation of the amygdala in the ingroup condition was considered to be due to the degree of high concern regarding the negative reputation from others for utilitarian judgment in the ingroup. However, another hypothesis is possible for interpreting this result, as follows. Judgments of majority versus minority regarding NIMBY facilities under ingroup condition were a task that involved the participants’ own payoff. If priority is given to the majority in such a task, the participants themselves may be able to obtain public benefits as members of the majority in exchange for the losses of the minority. Judgments to benefit themselves at the expense of others will evoke more serious emotional aversion for individuals than judgments to allocate payoffs between others unrelated to themselves. In other words, the reason why the activation of the amygdala was remarkable in the ingroup condition may be that the participants themselves were taking the standpoint of a dictatorial concerned party, making decisions for their own benefit in exchange for others’ loss. As noted earlier, this is a judgment that puts the participant in the position of a free rider himself. To examine this hypothesis, it is necessary to construct two types of dilemma: the concerned party dilemma, in which judgment by the decision makers influences their own payoff, and the third-party dilemma, in which one’s judgment does not affect the allocation for oneself. We manipulated concerned party and third party by locating NIMBY facilities variably with reference to the ingroup and the outgroup in the present study. However, as this manipulation method adopts group categories that include the effect of membership in addition to the involvement of self-interest, it made ambiguous which factors contributed to the results of this study. This is a major limitation of our experiment, and improvements are required for future study.

A further limitation of this study is that the inadequacy of the data did not allow making contrast estimates for each majority or minority judgment. This study hypothesized that moral aversion to utilitarian actions in the NIMBY dilemma would manifest in the ingroup condition (hypothesis 3). We could not fully test this hypothesis as we were unable to use judgment as an independent variable. In order to accurately test for moral aversion in the NIMBY dilemma, the paradigm of this study should be modified to reduce missing data. This is an important future requirement. In addition, since the participants in our experiment were primarily Japanese undergraduates in their twenties, a more rigorous and extensive data collection is essential to generalize the present results.

This study identified that the maximin principle, emotional aversion, and empathy for others are all involved in the choice of utilitarian and non-utilitarian judgments regarding NIMBY facilities. For individuals in the context of a NIMBY facility, it is assumed that a cognitive-neural process to create integrative judgments is activated while paying attention not only to one’s own payoff but also to others’, accompanied by aversion to utilitarian judgments. However, we have only just begun to develop an approach that regards the context of NIMBY facilities as a task that includes both social and moral dilemmas. This study constructed the context of NIMBY facilities in relation to the theoretical structure, but stricter conditions should be proposed for the selection of NIMBY facilities wherein two kinds of dilemmas are compounded. Further examinations need to be conducted to prove the effectiveness of this approach.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are openly available in “OSF HOME” at https://osf.io/x2jh5/?view_only=ddee9b3d72c944d782be1c58cd7b198f.

Additional information

Funding

This research was supported by a Grants‐in‐Aid for Scientific Research in Japan [20K20874].