Surveillance Cues Enhance Moral Condemnation

Abstract

Humans pay close attention to the reputational consequences of their actions. Recent experiments indicate that even very subtle cues that one is being observed can affect cooperative behaviors. Expressing our opinions about the morality of certain acts is a key means of advertising our cooperative dispositions. Here, we investigated how subtle cues of being watched would affect moral judgments. We predicted that participants exposed to such cues would affirm their endorsement of prevailing moral norms by expressing greater disapproval of moral transgressions. Participants read brief accounts of two moral violations and rated the moral acceptability of each violation. Violations were more strongly condemned in a condition where participants were exposed to surveillance cues (an image of eyes interposed between the description of the violation and the associated rating scale) than in a control condition (in which the interposed image was of flowers). We discuss the role that public declarations play in the interpersonal evaluation of cooperative dispositions.

Introduction

“Man is least himself when he talks in his own person. Give him a mask, and he will tell you the truth.”

-Oscar Wilde

In the lead-up to the 1990 Texas gubernatorial election, the three Democrats seeking their party's nomination tried to outdo each other in expressing their support for capital punishment. State Attorney General James Mattox bragged about how many executions he had attended in his role, and former Governor Mark White appeared in an advertisement to showcase the executions he had overseen during his previous tenure (Slater, 2009). More recently, while campaigning for the US presidency, competing candidates Barack Obama and John McCain found themselves aligned in opposition to the Supreme Court's ruling on Kennedy v. Louisiana, which limited the power of the states to impose the death penalty against an individual for committing a crime (such as the rape of a child) that did not result in the death of the victim. In voicing their opposition to this ruling, both candidates employed strongly condemnatory language, using words such as “heinous”, “egregious” and “despicable” to describe the latter crime.

Social scientists have long noted the propensity of humans to actively manage their reputations (Goffman, 1956), and experiments demonstrate that people are more generous and cooperative when they know that their behavior is observed by others (e.g. Gächter and Fehr, 1999; Wedekind and Milinski, 2000). Recent studies suggest that even very subtle cues that one is being watched can increase cooperation (cf. Fehr and Schneider, 2010), especially with respect to in-group members (Mifune, Hashimoto, and Yamagishi, 2010). This phenomenon was initially demonstrated using stylized eyespots on a computer screen (Haley and Fessler, 2005), but other studies have found effects using an image of a pair of eyes on a notice (Bateson, Nettle, and Roberts, 2006; Ernest-Jones, Nettle, and Bateson, in press), an image of a robot with human eyes presented on a computer screen (Burnham and Hare, 2007), and even three dots in a schematic face configuration (Rigdon, Ishii, Watabe, and Kitayama, 2009).

To date, empirical investigations of how individuals modify their behavior when they know or sense that they are observed by others have neglected moral judgments (e.g. Haley and Fessler, 2005; Kurzban, DeScioli, and O'Brien, 2007; Piazza and Bering, 2008). Expressing our opinions about the morality of certain acts, however, is a very important means of advertising our cooperative dispositions, so one might expect private and public opinions to diverge somewhat. The latter, after all, are expressed in one's “own person”, without the “mask” of anonymity. For instance, when electing group leaders, we attend to the candidates' public declarations on moral issues. As a result, there is a strong pressure on such candidates to conform their expressed opinions to prevailing moral and political norms. On the campaign trail, politicians may denounce crime and corruption with a fervor that belies their own private views or behavior. For example, in discussing the fact that he now opposes capital punishment - despite the fact that he supported it every time he ran for office - former Texas Governor Mark White remarked “I'm not running for anything - it's a lot easier for me to say it” (Slater, 2009).

In the present study, we sought to investigate how subtle cues of being watched would affect judgments of the seriousness of moral transgressions. Our hypothesis was that such cues would induce in our participants an unconscious perception that their behavior was observed, thereby activating evolved reputation-maintenance mechanisms. We predicted that participants exposed to such cues would provide stronger endorsements of moral norms by rating moral transgressions as more serious than control participants.

Materials and Methods

Ninety-one participants were recruited in various libraries of the Campus Universitaire de Jussieu in Paris. Participation was voluntary and without remuneration. In an effort to minimize conscious concerns about the reputational consequences of their responses (which we felt might swamp the more subtle effects of our independent variable), we elected not to collect any personal (e.g., demographic) information from participants.

Each participant was given two vignettes to read, printed on opposite sides of a single A4 sheet of paper. The vignettes were taken from Schnall, Haidt, Clore and Jordan (2008), and each described a moral violation: finding a wallet and keeping the money in one case, and falsifying a resume in the other (see the appendix for the original English versions, which were translated into French for the present study). Participants were instructed to rate the moral acceptability of each vignette on a nine-point Likert scale printed at the bottom of the page (1 = Morally unacceptable; 9 = Morally acceptable). Participants were directed to complete the task alone. The experimenter returned a few minutes later to collect the completed sheets.

Participants were assigned randomly to one of two conditions. In the eyes condition (n = 43) a 47 × 17 mm image of a pair of eyes (see Figure 1a) was displayed on each side of the sheet between the vignette and the associated Likert scale. In the flowers condition (n = 48) the image was of white flowers (see Figure 1b).

Figure 1.

Images presented in the (a) eyes; and (b) flowers conditions.

Results

Results are depicted in Figure 2. As the data were not normally distributed, we used a non-parametric test for our analyses. For the ‘wallet’ vignette, the median rating of moral acceptability was 2 in the eyes condition and 3 in the flowers condition; the distributions in the two conditions were significantly different (Mann-Whitney U = 789, p = 0.024, one-tailed). For the ‘resume’ vignette, the median rating was 3 in the eyes condition and 5 in the flowers condition; the distributions in the two conditions again differed significantly (Mann–Whitney U = 750, p = 0.016, one-tailed).

Figure 2.

Mean ratings of the two vignettes across the two conditions (error bars represent 1 SE either side of the mean).

Discussion

Recent studies indicate that subtle surveillance cues can influence cooperative behaviors. We predicted that participants exposed to such cues would express greater disapproval of moral transgressions than control participants. Our results are consistent with our prediction: participants supplied lower ratings of the moral acceptability of two different moral violations when an image of eyes was interposed between the description of the violation and the associated rating scale than when the interposed image was of flowers.

How are we to explain this effect? ¹ At the proximate level, one possibility is that the surveillance cues actually affected our participants' perception of moral violations, perhaps by activating their awareness of internalized moral norms (an example of “private self-awareness”; Govern and Marsch, 2001; see also Batson, 1990). An alternative, or additional, proximate explanation (the two alternatives are not mutually exclusive) is that an image of a pair of eyes matches the input conditions for evolved mental mechanisms that detect when one's behavior is observed (Haley and Fessler, 2005). In this case the image of eyes activates “public self-awareness” – cued participants attend to the impression that they are making on others and calibrate their behavior accordingly (Govern and Marsch, 2001). Further research is needed to tease apart these alternative explanations. ² Both, however, are compatible with a reputation-maintenance function at the ultimate evolutionary level: participants who increase their explicit support for shared standards of behavior – for whatever proximate reason – in the presence of surveillance cues, may advertise their prosocial dispositions and thereby maintain their reputations.

Evidence that reputation-management effects ultimately regulate the public expression of moral judgments suggests that such public declarations play an important role in the interpersonal evaluation of cooperative dispositions. Indeed, failure to express our support for prevailing moral norms may arouse suspicion in our conspecifics. Talk, however, is cheap, so we are wise to take such declarations with the proverbial grain of salt, discounting them as appropriate. In consequence, individuals may attempt to compensate for this discounting by ramping up their rhetoric. To the extent that these compensatory efforts are discounted in turn, they may be ultimately futile; what sustains them is the fact that failure to send the inflated signal immediately brands the deviant as morally suspect (see Grafen, 1990; McKay, Mijović-Prelec, and Prelec, 2011). The situation is similar to grade inflation in academia or to the phenomenon whereby increasingly effusive letters of recommendation are necessary to avoid “damning” a student or former employee with “faint praise”. Exaggerated public denunciations of moral violations may be necessary if one is to avoid damning oneself with faint support for prevailing norms.