Market mindset hinders interpersonal trust: The exposure to market relationships makes people trust less through elevated proportional thinking and reduced state empathy

Market relationships and their expected effect on trust

The contemporary consumer reality causes people to repeatedly establish relationships that are referred to as “market relations” (Fiske, 1991, 2004; Kasser et al., 2007; Sandel, 2013; Stanfield & Stanfield, 1997). As Fiske (1992) proposed almost 30 years ago,

Market pricing is so pervasive in Western society and so important in the Western cultural conceptions of human nature and society that many theorists have postulated that all human social behavior is based on more or less rational calculations of cost-benefit ratios in self-interested exchange. (p. 706)

When people engage in market relationships, they use a distinctive “mental grammar” that could be denominated as “market cognition” (Zaki et al., 2021). In the present work, we use the term “market mindset” to describe mental attributes people use to cognise, evaluate, and coordinate with each other and suggest that those with such a mindset primarily care about how much they get out of their investment and whether the repayment would be of comparable value. ¹

Market relationships are often interpreted as being in opposition to communal relationships in which people focus more on commonalities than on distinct features and individual goals (Fiske, 1992; McGraw & Tetlock, 2005). As Fiske (1992) suggested, the fact that people in a communal relationship recognise a shared identity makes them “think that it is natural to be relatively kind and altruistic to people of their own kind” (p. 691). The psychological characteristics that are typical of communal sharing are caring, kindness, altruism, and selfless generosity (Fiske, 1992, Table 1). In contrast, thinking in terms of the market mindset emphasises independence, individualistic efficiency, and self-control. Prior evidence indicates that market relationships do not allow for intimacy and emotional connectedness (Jiang et al., 2014; Mead & Stuppy, 2014; Vohs, 2015)—the qualities which could be considered as indispensable for showing trust (Mikulincer, 1998; Rempel et al., 1985). For example, Ma-Kellams and Blascovich (2013) found that when financial incentives were involved in a relationship, people tended to reveal reduced empathic accuracy. Furthermore, Molinsky et al. (2012) showed in a series of experiments that activating economic schemas had a negative effect on the compassion that people extended to others.

Trust involves having positive expectations of others and relying on their actions, despite uncertainty and a lack of control over their intentions and behaviour (Rousseau et al., 1998). This suggests that being trustful might facilitate and accelerate the building and regulating of relationships not only in a communal context but also in a transactional, market-related environment. As suggested by Arrow (1972) in one of his classical papers, “Virtually every commercial transaction has within itself an element of trust” (p. 357). Fukuyama (1995) indicated that trust is good for the economy because it significantly reduces transaction costs. Yet a careful analysis of some psychological features of market relationships suggests that trust might not function in concert with thinking in terms of the market mindset. For example, when someone lends a car to their friend (communal relationship), they trust it will be returned on time and in good condition. If a car is rented from a car rental business (market relationship), the salesperson will not rely on trust but will require a renter to sign a formal contract. Similarly, if consumers purchase different products (market relationship), they probably would prefer a formal guarantee that these products will be repaired quickly in the case of a defect rather than relying on the seller’s trustworthiness. On the contrary, in an intimate relationship, people prefer to trust one another to be faithful rather than signing a written contract clearly stating what would happen in the case of cheating (e.g., by introducing a system of fines and regulations). Therefore, even if trust may facilitate some transactions (Choi & Storr, 2020), we suggest that the market mindset does not make people spontaneously ready to trust others, especially when the situation goes beyond the strict market-related context.

Prior research that was not directly aimed at testing Fiske’s Relational models’ theory has shown that the market mindset signifies not only a preference for efficiency and structure but also the individualistic orientation towards effective but self-interested exchange (Caruso et al., 2013; Zaleskiewicz & Gasiorowska, 2017). In a similar vein, numerous studies using the money-priming paradigm have indicated that exposure to money—a prototypical medium of market pricing relations (Fiske, 1992)—and feelings of self-efficacy and self-sufficiency (Gasiorowska et al., 2016; Mukherjee et al., 2013; Vohs et al., 2006) strengthen the focus on self and impair a person’s propensity to take the perspective of others (Caruso et al., 2009), which may have a detrimental effect on decision-makers’ readiness to trust. If people with a salient market-type thinking believe that others are strongly motivated to endorse their own profits even at the cost of the interaction partners, they might be reluctant to show trust. Moreover, people can even suspect that if money is involved in social relations, trusting is highly risky because others would be eager to cheat to gain more. Indeed, prior evidence suggests that money primes (or wealth primes) unethical activities, such as lying to interaction partners to obtain higher profits or fraudulently overstating performance to receive better earnings (Gino & Pierce, 2009; Kouchaki et al., 2013; Piff et al., 2012). Connecting the market mindset with a reduced willingness to trust others also seems consistent with a common eagerness to associate money with greed, distrust, selfishness, suspiciousness, or evil (Furnham et al., 2014; Tang, 1992; Vohs, 2015; Wernimont & Fitzpatrick, 1972; Yang et al., 2013). Taking all this into account, we propose that the exposure to market relationships will cause people to show less trust compared with the exposure to nonmarket stimuli.

Although some evidence suggests a detrimental effect of market cognition on trust (e.g., Xin & Liu, 2013), the detailed mechanisms behind this effect have surprisingly never been explored. In this project, we propose two psychological mechanisms that could explain why the market mindset hampers trust: amplified proportional thinking and a reduced state empathy—both inevitably related to market cognition (Zaki et al., 2021; Zaleskiewicz et al., 2017, 2020). In the following section, we elaborate more on these two processes that might operate when an individual with the market mindset considers whether to trust another person or not.

Proportional thinking

Market relationships are based on clear, comprehensible rules which give people insight into the situation (Fiske, 1992, 2004). Market pricing ideology “presupposes a rationally calculating person operating in a predictable world of formally expressed expediential and moral rules; the conditions of action must be knowable, and outcome expectations must be calculable” (Fiske, 1992, p. 708). As previously noted, the core element present within market relationships is proportional thinking “directed toward calculating and acting in accord with ratios or rates for otherwise distinct goods to ensure that rewards or punishments for each party are proportional to their costs, contributions, effort, merit, or guilt” (Rai & Fiske, 2011, p. 64). All these characteristics may suggest that involvement in market relationships leaves little space for trust in the sense of both trusting and being trusted. On the one hand, thinking in terms of the market mindset is associated with the search for and orientation toward explicitly known and formally stated rules and proportions (Fiske, 2004; Gasiorowska & Zaleskiewicz, 2021; Zaleskiewicz et al., 2020). On the other hand, everyday encounters that require trust are rarely well-structured or predictable and do not supply clearly defined proportions between gains and losses that are highly valued in market exchanges. What is more, as the intentions of others are typically vague and hard to recognise, it is difficult or even impossible to directly calculate the proportions of costs and benefits resulting from the exhibition of trust in a relationship. We thus predicted that having a market mindset would increase one’s proportional thinking and, as a result, decrease one’s trust of others.

Reduced state empathy

As some authors propose, empathy, understood as an affective insight into the feeling state of another person (Spreng et al., 2009), might be an important factor in trust formation (Feng et al., 2004; Nooteboom, 2007). Studies investigating the effects of intranasal administration of oxytocin demonstrated that it facilitates human affiliation by increasing not only empathy (e.g., Hurlemann et al., 2010; Rodrigues et al., 2009) and social connection (Bosch & Young, 2017) but also the perceived facial trustworthiness of unfamiliar others (Theodoridou et al., 2009) as well as trusting behaviour (Baumgartner et al., 2008; Kosfeld et al., 2005; Mikolajczak et al., 2010). Empathy also plays an important social role because it is positively related to cooperation and social connectedness (de Vignemont & Singer, 2006), with the latter associated with trust (Glaeser et al., 2000; Sturgis et al., 2015). Therefore, we may conclude that trust formation is related to empathising, understood here as sensitivity to other people’s feelings (Cuff et al., 2016). Such a conclusion is important for our reasoning because the adoption of the market mindset may result in one’s situationally diminished sensitivity to affective cues (Caruso et al., 2013; Ma-Kellams & Blascovich, 2013; Molinsky et al., 2012; Zaleskiewicz et al., 2017), a reduced focus on others (e.g., Genschow et al., 2019), and a lower willingness to take the perspective of others (van Laer et al., 2013). All these characteristics may finally be associated with a diminished readiness to trust others.

If we propose that the negative relation between the market mindset and trust may be mediated by a reduced state empathy, we do not claim that the involvement in market relationships is inextricably linked with the absence of any emotional reactions. For example, when making their financial decisions, people may be affected by fear, anger, regret, and so on (Zaleskiewicz & Traczyk, 2020). What we instead suggest is that feelings associated with empathy which are inherently present in communal relationships (Zickfeld et al., 2017), reinforce social connectedness (de Vignemont & Singer, 2006), and can be linked to trust (Dunn & Schweitzer, 2005; Salazar, 2015) are not at the centre of market relationships. Therefore, we also predicted in our theoretical model that the market mindset may lessen people’s willingness to trust because it can situationally reduce empathy or, more specifically, sensitivity to others’ feelings.

The predictions we intended to investigate in this project refer not only to specific mediating effects of the two psychological mechanisms (i.e., increased proportional thinking and reduced empathy) but also to the configuration in which they are involved in the association between the market mindset and trust. Based on the extant literature, we proposed two competing hypotheses related to this issue.

The first hypothesis was that proportional thinking and state empathy operate in parallel as two relatively independent mechanisms. The rationale behind such a hypothesis may be the fact that brain regions associated with either emotional processing or cognition are characterised by a high degree of interconnectivity and operate by continuously feeding each other with relevant information (e.g., Ochsner et al., 2012; Pessoa, 2008). However, we also proposed an alternative hypothesis in which we assumed that the market mindset influences trust in a sequential manner via proportional thinking and state empathy. According to this prediction, people primed with a market mindset demonstrate elevated proportional thinking, which in turn reduces their state empathy and, consequently, lowers their trust. Such a sequential model would be consistent with evidence indicating that when an individual’s behaviour follows a certain pattern of thinking (in our case, proportional thinking; Zaleskiewicz et al., 2020), limited space remains for affect to influence their judgements (Forgas, 1995). In other words, when a given preference is salient, people are likely to “engage in highly selective, guided, and targeted information search and integration strategies designed to support a preexisting motivational objective” (Forgas, 1995, p. 47).

Overview of the present research

In the current project, we conducted a series of six experiments (including one presented in supplementary online materials [SOM]) to investigate how the market mindset affects people’s trust and the psychological mechanisms behind this effect. The main objective of the research was to investigate both the consequences of the market mindset for interpersonal trust and the psychological mechanisms behind this effect. We demonstrated that (a) participants’ exposure to market relations hindered their interpersonal trust (Experiments 1–2); (b) this effect was specific to the market but not to the communal mindset (Experiments 2–5); and (c) this effect was also observable for trust towards smiling people (Experiment 5). We excluded alternative explanations by demonstrating that the market mindset negatively affects trust but not caution or cynicism (Experiment S3 in SOM). In the next experiments, we examined the psychological mechanisms behind the detrimental effects of the market mindset on trust and found that these effects were mediated by enhanced proportional thinking (Experiments 3 and 5) and lowered state empathy (Experiments 4 and 5) for those in the market mindset. Finally, in a preregistered Experiment 5, we showed that these two psychological mechanisms operated relatively independently of each other.

We confirm that this project adheres to ethical guidelines specified in the APA Code of Conduct and that it was approved by the Ethics Committee for Human Research of the Robert B. Zajonc Institute for Social Studies, University of Warsaw, Poland (approval of 10 December 2012). All measures, manipulations, and exclusions as well as the method of determining the final sample size in the experiments in this project are disclosed in the respective “Method” sections.

Method

Procedure

After giving written informed consent and providing demographic information, participants were randomly assigned to one of the two experimental conditions: market mindset (n = 57) versus communal mindset (n = 51). To prime either market or communal mindset, we used the method that was successfully applied in earlier research in which the two mindsets were also activated (Gasiorowska & Zaleskiewicz, 2021; Zaleskiewicz et al., 2020). Participants were shown a picture presenting two black cartoon silhouettes facing each other on a white background, with the character on the left handing a paper bag with groceries to the character on the right (Figure 1). The picture was accompanied by a prepared description which was based on the interpretations provided by participants in other studies. Half of the participants received the following text describing the communal relationship: “The person on the left has gone to the store for an elderly neighbor and they brought them groceries. The elderly neighbor on the right is collecting the groceries and thanking them for their help.” The rest of the participants received the text describing the market relationship: “The person on the left is delivering groceries to the person on the right who ordered them online. The person on the right is buying the groceries online because their car broke down.”

OPEN IN VIEWER

Figure 1.

Picture used in Experiment 1.

Participants were then asked to think about the interpretation of the picture that had been provided to them on the previous page and to answer four questions measuring market orientation and communal orientation (Bohns et al., 2016; Gasiorowska & Zaleskiewicz, 2021). The two items used to measure market orientation were: “To what extent does this situation feel like a business transaction?” and “To what extent do you think one person is concerned about getting something from the other in this situation?” (M = 4.43, SD = 1.43, r = .35). The two items measuring communal orientation were: “To what extent does this situation feel like interacting with a close friend or family member?” and “To what extent is one person concerned about the other person’s needs in this situation?” (M = 4.83, SD = 1.73, r = .79). Participants answered these four questions in a random order using a seven-point scale (1 = not at all to 7 = very much).

Finally, participants answered three questions related to their generalised social trust: (a) “Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people?” (b) “Do you think that most people would try to take advantage of you if they got the chance or would they try to be fair?” and (c) “Would you say that most of the time people try to be helpful or that they are mostly looking out for themselves?” (e.g., Rosenberg, 1956; Zmerli & Newton, 2008). Participants rated each item on an 8-point scale with endpoints that corresponded to the attitudes indicated in the question (e.g., 1 = you can’t be too careful, 8 = most people can be trusted). The three answers were averaged to form an indicator of general social trust that served as the dependent variable (DV) for this experiment (M = 4.07, SD = 1.77, Cronbach’s α = .84).

Results and discussion

The results for the manipulation check can be found in the SOM. Our main hypothesis for this experiment was that the market mindset would negatively affect the level of generalised trust compared with the communal condition. The results of one-way analysis of variance (ANOVA) showed a significant effect of mindset manipulation, F(1, 106) = 7.56, p = .007, η² = .067. Participants in the market mindset condition declared lower generalised trust (M = 3.59, SD = 1.61) than participants in the communal mindset condition (M = 4.49, SD = 1.80). The results of Experiment 1 initially supported our prediction that priming people with the market mindset hampers their generalised trust towards others. The present experiment showed that merely exposing people to market relationships has the capacity to reduce their general social trust compared with priming them with communal relationships.

Method

Procedure

After giving written informed consent and providing demographic information, participants were randomly assigned to one of the three experimental conditions: market mindset (n = 40) versus communal mindset (n = 50) versus neutral condition (n = 60). We used the method that was successfully applied in earlier research in which these two mindsets were also activated (Gasiorowska & Zaleskiewicz, 2021; Zaleskiewicz et al., 2020) to prime either market or communal mindset. Depending on the experimental condition, participants were shown one of the three pictures which represented two black cartoon silhouettes facing each other (labelled A, B, or C in Figure 2). In the market mindset condition, one of the characters in the picture was holding a cardboard pizza box, while the other character was holding money (panel (a) in Figure 2). This picture was intended to prime a market relationship situation, in which one person uses money to pay for a certain product. In the communal mindset condition, one of the characters was holding a box wrapped like a gift, while the other character’s hands were empty (panel (b) in Figure 2). This picture primes close communal relationships, in which one person selflessly hands over goods to another person. In the neutral condition, the two hands of the two characters were empty in order not to cue any specific type of relationship mode (panel (c) in Figure 2). Participants were asked to take a careful look at the picture and imagine a real-life situation similar to the one presented in the picture. They were then asked to describe this situation in as much detail as they could. To sustain the manipulation, the respective picture was then present at the top of each of the subsequent screens. Following this manipulation task, participants were asked to answer three questions about the interpretation of the picture they had provided: (a) whether their interpretation of the picture was related to close relations between two people (e.g., between friends, family members); (b) whether their interpretation of the picture reflected commercial relations between two people (e.g., as in business or trade); and (c) whether in their opinion the two characters involved in the situation were close to each other. The questions were presented in a random order and used a 7-point response scale (from 1 = definitely not to 7 = definitely yes). The first two questions served as a manipulation check. The third question served as a control question to test whether the characters in the market and communal conditions were perceived as not closer to each other because their hands were extended in each other’s direction in comparison with the characters in the neutral condition whose hands were not extended. Finally, as in Experiment 1, participants answered three questions related to their general social trust. Participants rated each item on an 8-point scale on which the endpoints corresponded to the attitudes indicated in the question (e.g., 1 = you can’t be too careful, 8 = most people can be trusted). The three answers were averaged to form an indicator of general social trust that served as the DV for this experiment (M = 4.50, SD = 1.51, Cronbach’s α = .90).

OPEN IN VIEWER

Figure 2.

Pictures used to induce either market mindset (a) or communal mindset (b) in Experiments 2–6. Picture in (c) was shown in the neutral condition.

Results and discussion

The results for the manipulation check can be found in the SOM. Our main hypothesis for this experiment was that the market mindset would negatively affect the level of generalised trust compared not only with the communal mindset condition but also with the neutral condition. The results of one-way ANOVA showed a significant effect of mindset manipulation, F(2, 147) = 5.26, p = .006, η² = .067. Further planned contrasts revealed that participants in the market mindset condition declared lower generalised trust (M = 3.96, SD = 1.26) than participants in the communal mindset condition (M = 4.98, SD = 1.48), F(1, 147) = 10.48, p = .001, partial η² = .067. The difference between the market mindset and the neutral condition (M = 4.47, SD = 1.57) was not significant, F(1, 147) = 2.94, p = .088, partial η² = .020. Similarly, the difference between the communal mindset and neutral condition was not significant, F(1, 147) = 3.09, p = .081, partial η² = .021.

The results of Experiment 2 provided additional support for our prediction that priming people with the market mindset causes them to demonstrate lower generalised trust towards others compared with priming them with communal relationships. However, the results concerning the comparison of the two experimental conditions with the neutral condition were inconclusive.

Method

Results and discussion

The results for the manipulation check can be found in the SOM. We conducted a mediation analysis using Process 3.5 (Hayes, 2018) Model 4 with 10,000 bootstrapping samples (Figure 3) to test our research hypotheses that the market mindset hinders trust because it also triggers a preference for using proportions and numbers. We used effect coding with the neutral mindset as a reference category (Hayes & Preacher, 2014), with one dummy variable representing the market mindset versus neutral condition comparison and a second dummy variable representing the communal mindset versus neutral condition comparison. The effect-coded mindset manipulation served as an independent variable, level of trust as a DV, and indicator of proportional thinking as a mediator. The continuous variables were z-scored before analysis to allow for standardised coefficients.

OPEN IN VIEWER

Figure 3.

Mediation effect of communal and market relationships through proportional thinking on trust (Experiment 3).

The total effect that the market mindset versus neutral condition had on trust was significant and negative, path c₁: β = −.16, SE = .07, t = −2.09, p = .037, while the total effect of the communal versus neutral condition was significant and positive, path c₂: β = .18, SE = .07, t = 2.49, p = .013. People primed with the market mindset were less prone to declare that they would trust the people presented in the photos (M = 5.33, SD = 1.57) than those primed with either the neutral mindset (M = 5.52, SD = 1.46) or the communal mindset (M = 5.82, SD = 1.30).

The exposure to the market mindset manipulation (vs. neutral condition) had a significant and positive effect on proportional thinking, path a₁: β = .23, SE = .07, t = 3.07, p = .002, while the exposure to the communal manipulation (vs. neutral condition) had a significant and negative effect on proportional thinking, path a₂: β = −.18, SE = .07, t = −2.53, p = .012. Participants in the market mindset condition declared the highest motivation to rely on proportions and numbers (M = 3.36, SD = 1.38) compared with participants from the two other conditions (respectively, M = 2.79, SD = 1.24 for the communal mindset condition and M = 2.98, SD = 1.46 for the neutral condition). Proportional thinking was negatively related to the level of trust, path b: β = −.24, SE = .05, t = −4.74, p < .001. While controlling for the mediator, the direct effect of priming the market mindset versus the neutral mindset on trust was not significant, c ′ 1 : β = −.10, SE = .07, t = −1.37, p = .171, and the direct effect of priming the communal mindset versus the neutral mindset was also not significant, c ′ 2 : β = .14, SE = .07, t = 1.91, p = .056. The 95% bootstrapped confidence interval (CI) for the relative indirect effect from the market mindset versus the neutral condition did not contain zero, 95% boot CI = [−.107, −.017], similarly to the 95% bootstrapped CI for the relative indirect effect from the communal mindset versus the neutral condition, 95% boot CI = [.009, .093], with the former relative indirect effect being slightly stronger (β = −.06, boot SE = .02) than the latter (β = .04, boot SE = .02).

The present study not only replicated our previous results which showed the association between the exposure to market relationships and diminished trust but also demonstrated that this association is mediated by proportional thinking. In line with the argumentation that we presented in the Introduction, market-type interactions trigger a preference for using proportions and numbers that are not explicitly available in many social situations, and this is the reason the market mindset has a detrimental effect on trust. Unlike in our previous experiments, we also found a positive effect of the communal mindset on interpersonal trust. We believe this is due to the different operationalisation of our DV: while we used context-free measures of generalised trust in Experiments 1 and 2, we employed a measure involving social cues, such as human, faces in Experiment 3. Nevertheless, exposure to market-type cues diminished trust no matter if the trust was measured as generalised trust of other people or as trust of specific, identifiable strangers.

Method

Results and discussion

The results of the manipulation check for this experiment can be found in the SOM. To test our research hypotheses, we performed linear regression with mindset manipulation as the predictor, state empathy as the mediator, and the level of trust as the DV. For the multicategorical independent variable, we again used effect coding with the neutral condition as a reference category (Hayes & Preacher, 2014), with one dummy variable representing the market mindset versus neutral condition comparison and a second dummy variable representing the communal mindset versus neutral condition comparison (Figure 4). The total effect of the market mindset versus the neutral condition was significant and negative, path c₁: β = −.19, SE = .07, t = –2.57, p = .011, while the total effect of the communal mindset versus the neutral condition was significant and positive, path c₂: β = .19, SE = .07, t = 2.52, p = .012. Participants primed with the market mindset were less prone to declare that they would trust people presented in the photos (M = 5.39, SD = 1.28) than were participants in the neutral condition (M = 5.67, SD = 1.53) and those primed with the communal mindset (M = 5.93, SD = 1.47).

OPEN IN VIEWER

Figure 4.

Mediation effect of market and communal relationships through state empathy on trust (Experiment 4).

The exposure to market relationships (vs. neutral condition) had a significant and negative effect on state empathy, path a₁: β = −.26, SE = .07, t = –3.62, p < .001, while the exposure to communal relationships (vs. neutral condition) had a significant and positive effect on state empathy, path a₂: β = .26, SE = .07, t = 3.60, p < .001. Participants in the market mindset condition declared the lowest state empathy (M = 3.81, SD = 0.67) compared with participants from the two other conditions (respectively, M = 4.11, SD = 0.45 for the communal mindset condition and M = 3.98, SD = 0.57 for the neutral condition). State empathy was positively related to the level of trust, path b: β = .33, SE = .05, t = 6.59, p < .001. While controlling for the mediator, the direct effect of priming the market mindset versus the neutral condition on trust was not significant, c ′ 1 : β = −.10, SE = .07, t = −1.43, p = .152, which was the same as the direct effect of priming the communal mindset versus neutral condition, c ′ 2 : β = .10, SE = .07, t = 1.40, p = .163. The 95% bootstrapped CI for the relative indirect effect from the market mindset versus neutral condition did not contain zero, 95% boot CI = [–.154, –.035], the same as in the 95% bootstrapped CI for the relative indirect effect from the communal mindset versus neutral condition, 95% boot CI = [.040, .142]. The two relative indirect effects were similar in size, β = −.09, boot SE = .03, for the relative indirect effect from market mindset versus neutral condition and β = .09, boot SE = .02 for the relative indirect effect from the communal mindset versus the neutral condition.

The present study corroborated our previous results which showed that people’s exposure to market relationships has a detrimental effect on interpersonal trust. Furthermore, it demonstrated that this association is mediated by the presence of a lower state empathy when the market pricing mode is salient. In line with the argumentation that we presented in the Introduction, market-type interactions tend to decrease state empathy, which might be another reason the exposure to market relationships has a harmful effect on trust.

Method

Results and discussion

The results concerning the manipulation check can be found in the SOM. To test our research hypotheses, we conducted two types of mediation analysis (both preregistered) aimed at testing the configuration in which participants’ proportional thinking and their state empathy mediated the relation between market mindset priming and declared trust towards smiling people: (a) a mediation model in which the propensity to rely on proportions and empathy were parallel, uncorrelated mediators; and (b) a mediation model in which the propensity to rely on proportions and empathy were tested in a sequential order. For the multicategorical independent variable, we again used effect coding with the neutral condition as a reference category (Hayes & Preacher, 2014), with one dummy variable representing the market mindset versus neutral condition comparison and a second dummy variable representing the communal mindset versus neutral condition comparison. Continuous variables were z-scored prior to analysis.

The mediation analysis was conducted using MPlus (Muthén et al., 2017). We employed a path analysis to examine the hypothesised relationships among variables. Given that some variables were not distributed normally (e.g., high kurtosis and skewness), we used the robust full information maximum likelihood estimation method to estimate the model. We examined goodness of fit using multiple indices: the root mean square error of approximation (RMSEA) and its 90% CI, the standardised root mean square residual (SRMR), the comparative fit index (CFI), and the Tucker–Lewis index (TLI). We used multiple fit indices because they assess different types of model fit (e.g., model parsimony, absolute fit) and provide a more reliable, conservative evaluation when used together (Brown, 2015). The model was evaluated with criteria proposed by Hu and Bentler (1999). The lower bound for an acceptable fit for CFI and TLI was .90, while the upper bound for an acceptable fit for RMSEA and SRMR was .08. Indirect effects were tested by examining bootstrap CIs for the indirect effects using a bootstrap procedure with 10,000 samples.

The first regression analysis demonstrated that the market mindset had a significant and negative effect on participants’ trust, β = −.13, SE = .06, Z = −2.25, p = .025, while the effect of the communal mindset was not significant, β = .03, SE = .06, Z = 0.56, p = .577 (RMSEA = 0, SRMR = 0, TLI = 1, CFI = 1). After introducing proportional thinking and state empathy operating as parallel independent mediators, the model did not reach conventional thresholds for acceptable fit (RMSEA = .105, 90% CI = [.042, .184], SRMR = .042, CFI = 0.854, TLI = −0.318). All the hypothesised path coefficients were significant, except for the direct paths between the market mindset manipulation and the communal mindset manipulation and the DV (controlling for indirect effects). After excluding them,, the model fit the data well in the light of some but not all examined indices (RMSEA = .068, 90% CI = [.026, .115], SRMR = .037, CFI = 0.816, TLI = 0.447). The results (standardised path coefficients) for the two models are presented in Figure 5. The manipulations and mediators accounted for R² = .034 of the variance of participants’ declared trust.

OPEN IN VIEWER

Figure 5.

Path analysis for the effect of communal and market relationships on trust through a state empathy and proportional thinking as parallel mediators (Experiment 5).

We also examined total, direct, and indirect effects of the market and communal mindset manipulations on the DV in a more parsimonious model. The total indirect effect of the market mindset manipulation on trust via both mediators was significant and negative (β = −.04, boot SE = .02, 95% boot CI = [−.085, −.011]). Further investigation of the specific indirect effects revealed that both indirect effects from the market mindset manipulation on trust were significant, respectively β = –.02, boot SE = .01, 95% boot CI = [−.054, −.003] for the indirect effect via proportional thinking and β = −.02, boot SE = .01, 95% boot CI = [−.053, −.003] for the indirect effect via state empathy. In addition, the total indirect effect of the communal mindset manipulation on trust via both mediators was significant and positive (β = −.04, boot SE = 0.02, 95% boot CI = [−.085, −.011]), as both relative indirect effects from the communal mindset manipulation, respectively β = .01, boot SE = 0.01, 95% boot CI = [.001, .047] for the indirect effect via proportional thinking and β = .02, boot SE = 0.01, 95% boot CI = [.007, .055] for the indirect effect via state empathy.

Second, we tested an alternative preregistered model with proportional thinking and state empathy operating in a sequential order (RMSEA = 0, SRMR = 0, CFI = 1, TLI = 1). As in the first model, direct paths from the market mindset manipulation and the communal mindset manipulation to the DV (controlling for indirect effects) were not significant and, therefore, were excluded from the model. The resultant model fit the data well in the light of the most examined indices (RMSEA = .038, 90% CI = [.001, .101], SRMR = .023, CFI = .962, TLI = .830). The results (i.e., standardised path coefficients) for the two models are presented in Figure 6. The manipulations and mediators accounted for R² = .038 of the variance of participants’ declared trust, which indicates a small effect.

OPEN IN VIEWER

Figure 6.

Path analysis for the effect of the communal and market mindsets on trust through state empathy and proportional thinking as serial mediators (Experiment 5).

We examined total, direct, and indirect effects of the market and communal mindset manipulations on the DV in a more parsimonious alternative model. The total indirect effect of the market mindset manipulation on trust was negative and significant (β = −.04, boot SE = .02, 95% boot CI = [−.085, −.011]). Further investigation of the specific indirect effects revealed that the sequential indirect effect from the market mindset via proportional thinking and empathy on trust was not significant (β = −.003, boot SE = .002, 95% boot CI = [−.011, 0]). The indirect effect via only proportional thinking was significant and negative (β = −.020, boot SE = .012, 95% boot CI = [−.054, −.003]), as was the indirect effect via state empathy only (β = −.017, boot SE = .012, 95% boot CI = [−.048, −.001]). The two indirect effects via single mediators were of similar strength, 95% boot CI = [−.037, .029] for the effects comparison.

In addition, the total indirect effect of the communal mindset manipulation on trust via both mediators was significant (β = .04, boot SE = .02, 95% boot CI = [.014, .077]). The sequential indirect effect via proportional thinking and empathy was not significant (β = .002, boot SE = .002, 95% boot CI = [0, .008]), while both relative indirect effects via single mediators were significant and positive, respectively β = .01, boot SE = 0.01, 95% boot CI = [.001, .041] for the indirect effect via proportional thinking and β = .02, boot SE = .01, 95% boot CI = [.006, .051] for the indirect effect via state empathy. The two indirect effects via single mediators were of a similar strength, 95% boot CI = [−.040, .014] for the effects comparison.

In this experiment, we demonstrated that the salience of the market mindset priming had a detrimental effect on trust expressed towards smiling faces, which in our view provides strong support for our theoretical reasoning. We also examined two preregistered models of the link between the market and communal mindset manipulations, trust towards smiling persons and mediators—state empathy and proportional thinking. Although we did find that the model which assumed sequential mediation was a better fit to the data than the model which assumed independent parallel mediators, we did not find support for sequential mediation from the market mindset manipulation via enhanced proportional thinking and lowered state empathy on trust. Rather, our results suggest that although related to each other, the two mechanisms behind the detrimental effect of the market mindset on trust operate independently.

Summary of the results

In the present project, we examined the general hypothesis that when people’s thinking is determined by the market mindset, which is activated because of exposure to market relationship cues, they tend to trust others less. In addition, we proposed that merely making market relationships salient may be enough to negatively affect interpersonal trust and that two psychological mechanisms—hindered state empathy and enhanced proportional thinking—account for these effects.

We conducted a series of six experiments (including one presented in SOM) to test this hypothesis. In Experiment 1, we investigated two relational mode conditions—market relations and communal relations—and found that people with the market mindset revealed lower generalised trust than those with the communal mindset. In the following experiments, we also included a neutral condition to demonstrate that the difference between market and communal mindset conditions was caused by the detrimental effect of the market mindset rather than the beneficial effect of the communal mindset. We also excluded alternative explanations by demonstrating that the market mindset negatively affects trust but not caution or cynicism (Experiment S3). In further experiments, we examined two psychological mechanisms responsible for the detrimental effects on trust of the market mindset and found that these effects were mediated by enhanced proportional thinking (Experiments 3 and 5) and a reduced state empathy (Experiments 4 and 5). Finally, in a preregistered Experiment 5, we showed that these two psychological mechanisms are relatively independent of each other.

In the experiments conducted within this project, we employed different manipulation methods, DV measurements, and manipulation checks to avoid incidental results and to ascertain that the effects we found were not due to the specific procedures we used but are robust across various methodological approaches. By systematically modifying our experiments, that is, by changing manipulations, manipulation checks, and operationalisations of DVs, we were able to demonstrate high reliability (repeatability) of the effects, maximise internal and external validity, eliminate alternative explanations, and test for mediators that explain the psychological mechanisms behind the effects.

Does the market mindset inevitably reduce trust?

The direct effect observed in all the experiments presented in this article was that the market mindset significantly hampers declared trust towards others. The recurrent presence of such an effect in our experimental project might seem puzzling because theoretical considerations along with evidence found in research indicate that trust is both necessary and present in market settings as it catalyzes transactions and sustains the economic system as a whole (Aghion et al., 2010; Akerlof, 1970; Cohen, 2020; Fukuyama, 1995; Zak & Knack, 2001). Although business-type interactions are typically regulated by various written agreements, they rarely reduce uncertainty entirely as a complete substitute for trust (McCannon et al., 2018). For example, an economic entity may trust that the other party will comply with the terms of the formal contract and restrain from the temptation to cheat when an unfair action is accessible and hard to detect. If, in such a case, trust is reciprocated, both transaction sides avoid lengthy and costly litigation. We might even say that the economic system would be paralysed if everybody on the market strictly followed the principle of limited trust or did not trust others at all (Akerlof, 1970). One more example which supports the reasoning that trust is inevitably involved in the economic system is the Trust Game (Berg et al., 1995), which has been widely used to study trusting decisions in various financial contexts. If players participating in the Trust Game were extremely suspicious of each other, no money would be transferred between them and, consequently, no economic exchange would occur. Interestingly, studies using the Trust Game have consistently shown that people do, in fact, trust others even when they have no information about the partner that could suggest their decision will turn out to be profitable (Johnson & Mislin, 2011). In the context of the above argumentation and examples, can the results of our present project be considered as evidence of a paradox?

In our view, the results of Experiments 3–5, which had the aim of investigating the psychological mechanisms behind the negative effect of the market mindset on trust, seem to offer a solution to the puzzle. Based on these results, we propose that people may differently interpret trust depending on whether they make their decisions in the market or communal contexts. The market mindset increases people’s readiness to perceive reality in a market-consistent manner—by a tendency to use cost-benefit analysis (manifested through the motivation to use proportions and numbers) and the reluctance to focus on others’ emotions (as evidenced by diminished empathy). Unlike the communal context, people in the market-related context may be willing to reveal trust only if they have a chance to calculate whether and how much it pays off. In our experiments, we asked participants to answer general questions about trusting unknown others, providing them either with no contextual cues or a clearly communal cue—a smiling face in Experiment 5. In other words, participants could not engage in any calculations, but they could potentially base their decisions on their empathetic reactions. As Experiments 3–5 indicate, making the market mindset salient increases a person’s orientation towards proportional analysis but decreases empathy. Therefore, it is not surprising that those participants for whom the market mindset had been made salient declared reduced trust when faced with general questions related to trust in others. It is possible that more positive effects on trust would be observed as a result of the market mindset by introducing participants to a more market-relevant and information-rich situation. Indeed, a study by Al-Ubaydli et al. (2013) in which exposure to market cues positively affected trust towards others as measured by a decision in the Trust Game—the simulation of a situation that is well-structured and offers a chance to calculate how much it pays to trust the other party—provides initial support for this thesis. Similarly, Henrich et al. (2005) showed that the higher level of market integration predicted greater prosocial choices as measured by the use of different economic games.

Different types of trust

Although trust is a necessary component of well-functioning relationships in both market and communal spheres that share a number of commonalities from the perspective of trusting decisions (e.g., the presence of both a trustor and a trustee, the willingness to be vulnerable, and positive expectations regarding a trustee’s behaviour), the reasoning we have thus far proposed in this section suggests that trust could serve different purposes and be supported by different mechanisms depending on the specific relationship. In market relationships, people trust that others will complete tasks competently and on time, that they will receive fair remuneration, and that the other party will comply with the terms of the contract (Rousseau et al., 1998). In communal relationships, people trust that their partners or friends will not betray them and expect no tangible reward for their trust (Rempel et al., 1985). Trust in such relationships emerges through the pro-relationship behaviour of partners as a result of their departures from immediate self-interest and their provision of other-benefitting acts (Rusbult et al., 1998).

Some previous theorising does support the distinction we have introduced between psychological characteristics of trust in market relationships and in communal relationships. For example, Lewicki and Bunker (1995) differentiated between “calculus-based” or “calculative trust” and “relational trust” (also known as “cognition-based” and “affect-based trust”; see McAllister, 1995). Calculus-based trust is more aligned with the characteristics of market relationships—it occurs in economic exchanges and is based on rational considerations of the other party’s intentions as derived from the existence of deterrence (costly sanctions associated with breach of trust) and/or some credible information about the intentions, motives, and competence of the trustee (e.g., reputation, certification; Lewicki & Bunker, 1995). Relational trust seems to be closer to the definitional characteristics of communal relationships. It forms over time due to repeated informal interactions between two parties and is based on the experience of positive outcomes from those interactions as well as the care and attachment that arises within the relationship (McAllister, 1995). A particularly strong form of relational trust is that of “identification-based trust,” which forms between parties that identify with each other’s intentions and desires and in which “mutual understanding is developed to the point that each can effectively act for the other” (Lewicki & Bunker, 1995, p. 122). No deterrence or monitoring is necessary because both parties are certain that their interests will not be intentionally compromised (Lewicki et al., 1998). As calculus-based and relational trust seem to rely on distinct information, it is possible that they are differentially affected by the salience of the market mindset. It is also possible that while proportional thinking is more important for trust experienced in market relationships (such as calculus trust), emotional cues might be more predictive of trust experienced in communal relationships. This hypothesis seems to be indirectly supported by a recent study (Weiss et al., 2021) which showed that behaviour in the lab-based Trust Game (which contains more detailed information about potential returns) was predictive of only everyday trust with regards to more distant others (strangers, professional contacts, superordinates, and acquaintances) with whom people are more likely to have market-pricing relationships. Everyday trust towards closer others was not related to participants’ behaviour in the Trust Game.

To summarise, it is likely that people who are prompted to interpret reality through the lenses of the market mindset will have a significant drop in their general willingness to trust others because they spontaneously connect their thinking with communal (relational) trust which requires sensitivity to other people’s emotions and is not saturated with information about inputs and outputs. However, people who have the market mindset might be ready to reveal trust when they feel they are participating in a well-structured, business-type interaction that enables cost-benefit analysis. Such a divergent influence of the market mindset on trust seems to be an important issue to be addressed in future research.

The current research also offers possible directions for exploring interventions to enhance trust in different relational contexts. This is especially important given the possibility that the progressing process of marketisation in modern society might negatively affect people’s trust in others (Xin & Xin, 2017) and people may start to introduce market-pricing rules into communal relationships. While trust might increase in the communal context by amplifying emotional engagement, for example, by increasing the likeability of the trustee (in Experiment 3, communal mindset was shown to positively affect the state empathy and—as a result—trust), such intervention seems unlikely to be successful in the market context. We showed that the market mindset negatively affects trust by situationally blocking empathic concern; thus, the attempts to increase trust through affective cues might prove ineffective or even backfire. A more effective strategy oriented towards elevating trust in a transactional context could rely on the provision of information consistent with market rules, for example, a rule that is consistent with the preference for using proportions and numbers. In addition, specific information about the reputation of the transaction partner or the probabilities of gaining or losing on an investment could potentially boost trust in a market context. Further research should explore whether it is possible to nudge people to express more trust by framing a situation in a manner that is consistent with the active mindset or by providing information that satisfies the motivations that are active in the market or communal context.

Limitations and further research directions

Our findings should be interpreted with certain limitations and future research directions in mind. First, although we used various measures for our dependent and independent variables, we are aware that we did not cover a wide range of situations in which people might form trust towards another person. For that reason, even if we found consistent and robust effects, we still see the need to conceptually replicate our results using other behavioural measures of trust in externally valid contexts, most preferably in field experiments conducted in natural social environments. Second, we acknowledge that the obtained effects are rather small, which is probably due to the difficulty of experimentally manipulating and measuring mental states. Future research should address this issue and investigate whether other manipulation methods could produce larger effects.

Another limitation of this research is that most studies were conducted with Western samples (United States, United Kingdom, and Poland), including both student samples and online labour markets. Although the quality of data obtained from online labour markets has been questioned, different analyses have suggested that data collected on MTurk and Prolific Academic are valid and comparable to data collected via traditional methods (Buhrmester et al., 2011; Mason & Suri, 2012; Peer et al., 2017). Nevertheless, we believe that our theoretical ideas should be further tested with other non-Western samples. One prediction is that the effect that we found in the present project examining market relationships on trust might be weaker or even completely absent in cultures such as Japan in which market relationships are regulated by strict social norms and where formation of close social bonds is essential to maintain business relationships (De Mente, 1993; Lohtia et al., 2009).

In the set of experiments presented in this article, we used simple pictures to activate thinking about either market or communal relationships. To prime the latter, we presented participants with a situation in which one individual was giving a gift to another individual. Even if gift-giving most often occurs in the communal-sharing context, it can also sometimes happen in other social modes (e.g., equality matching). Although we excluded the alternative explanation that the gift-giving picture triggered relational modes other than communal sharing (see auxiliary studies in SOM), we see a need for further experimental research to investigate the psychological consequences of the activation of different modes of sociality which use other methods of inducing market or communal mindsets to avoid inaccurate interpretations of the manipulation effects.

Finally, in our experiments, we focused solely on situations in which a lack of clear-cut rules of interaction and sufficient information on costs and benefits does not allow people to use proportional thinking, which translates into reduced willingness to trust others. Therefore, we cannot (and do not) conclude that the market mindset reduces trust in all circumstances. Further research should address this issue and investigate potential interactions between the active mindset and the situational cues in predicting trust.