Would You Cheat? Cheating Behavior, Human Nature, and Decision-Making

Default Behavioral Predispositions: What are we Struggling With?

Having established what general mechanisms underlie decisions about cheating, we can move on to another relevant question – what is the default option when people face an opportunity to cheat? People are governed by two contradictory mechanisms: one that inclines them to exploit opportunities to gain and another that helps them form a sense of their own honesty. Because of these counteracting impulses, an irrational pattern of cheating emerges. But which side has greater pull?

Greene and Paxton (2009) attempted to answer this question by examining neural activity during honest and dishonest decision-making. They put forward two hypotheses that can explain honest decision when an individual is faced with temptation: (1) the will hypothesis which contends that honesty is a result of active resistance of temptation and therefore brain regions responsible for cognitive control are employed when people are honest; (2) the grace hypothesis which contends that for being honest, no additional cognitive control is required. The first one is more favorable toward the corrupt view of human nature, whereas the second is more compatible with the good-natured view.

Participants in their experiment had to predict an outcome of a fair coin-toss and were rewarded for doing it correctly. The cover story for this study was that it was an experiment examining paranormal abilities to foresee the future when one makes his or her predictions privately and is financially motivated to do it correctly. The real purpose was to test levels of dishonesty among subjects. In control group participants had to report their prediction before a toss so cheating was not possible, in experimental group they had to report after a toss whether they were correct so cheating was possible with zero probability of being caught.

Figure 2. Distribution of self-reported percent Wins in the Opportunity condition (Greene & Paxton, 2009, p. 12507).

Based on distribution of outcome declarations participants were divided into two groups which were analyzed separately. In the case of honest group, the crucial element of interest were loss trials in the opportunity condition as compared to loss trials in no-opportunity condition. Two competing hypotheses provide different predictions: the will hypothesis that in opportunity loss trials there should be increased activity in areas involved in cognitive control (anterior cingulate cortex, dorsolateral prefrontal cortex and ventrolateral prefrontal cortex) and that reaction time should be longer, as compared to no-opportunity loss trials; the grace hypothesis that no differences should appear.

Advertisement

It turned out that the grace hypothesis was valid in case of honest participants – there were no differences in reaction time and neural activity. These results suggest that there exist some people who are able to be honest by not even realizing their opportunity to gain. These participants did not take dishonest option into consideration. This result is surprising since it contradicts both intuitive understanding of human moral decision-making and standard economic model.

However, generalization of the grace hypothesis is not possible. Analysis of responses in the dishonest group provided reversed conclusions. It turned out that both reaction time in the opportunity-loss as compared to the opportunity-win was longer and that more brain activity in control-related regions were observed in the opportunity as compared to the control condition (both win and loss trials). Dishonest group seems therefore to provide a support for the will hypothesis. Their default response is the desire to cheat and they have to engage much cognitive control in such decisions.

A question arises which explanation applies to the majority of people? As Greene and Paxton (2009) point out, the [g]race hypothesis applies only to honest decisions in individuals who consistently behaved honestly and not to decisions reflecting limited honesty (p. 12509). In their experiment some participants did not even consider the possibility of cheating. But it is not implausible that in different circumstances they would also think according to dishonest pattern. Honesty seems to result from either lack of realization of opportunity to gain or active rejection of temptation when such opportunity is considered.

Abe (2011) comments these results that the exact role of prefrontal cortex in decisions about deception in not clear. It may be the case that it is engaged in active considerations about an action (cost-benefit analysis) or in attempts to resist temptation. It is also not certain what regions are responsible for predisposing people for different response patterns. As he argues, such decisions may result from some interaction between the prefrontal cortex and subcortical areas providing motivation (reward seeking).

Further research provide some evidence for the will hypothesis. Shalvi, Eldar and Bereby-Meyer (2012) conducted a die-under-cup study in which they examined which response is an automatic one: honesty which can transform into dishonesty when time and justifications are available; or dishonesty which can be resisted when people have enough time and lack of justifications. Authors hypothesized that cheating is a natural tendency that needs to be overcome by cognitive control.

In the first experiment they instructed participants to roll three times and to report the first outcome. The amount of time available was manipulated across conditions. One group had to report under time pressure, the other not. It turned out that those who had more time to deliberate were more honest than subjects in the time pressure condition. In the second experiment they reduced the availability of justifications by instructing participants to roll only once. Obtained results confirmed the hypothesis – in the low time pressure condition cheating was less pervasive than in the first experiment, but no differences were observed in high time pressure condition. As authors argue, people can behave ethically when they have time and do not have justifications.

Similar results were reached in a study conducted by Gunia and his collaborators (Gunia, Wang, Huang, Wang, & Murnighan, 2012) who had their participants play a modified version of Gneezy’s (2005) deception game. In this game 15 $ had to be divided between two players. Player 1 sees two options: one which earns him or her 10 $ and lets 5 $ to the other player; and another one which earns him or her 5 $ and lets 10 $ to the other player. Player 1 must send message to player 2 that indicates which option is more profitable for him or her and then player 2 makes decision which option to choose. Again, the amount of time available for player 1 decision was manipulated and it turned out that more immediate choices are less ethical – only about 56% of participants sent truthful message, whereas about 87 % did it in the contemplation condition (more time for deliberation).

Further exploration of the interplay between will and grace in the moral domain was done by Fosgaard and his collaborators (Fosgaard, Hansen, & Piovesan, 2013) who sought to distinguish between two effects in cheating: (1) grace effect – becoming aware that cheating is an option and therefore losing grace; (2) will effect – inferring a norm that cheating is acceptable and actively choosing to do it. They conducted an experiment in which they had participants toss a coin privately and report results on a sheet of paper. Participants could receive 10 DKK if coin-toss resulted in white or nothing if it resulted in black. What is important subjects were students of the same class. They introduced two manipulations in their study (2 x 2 design): suggestion whether cheating is an option – earlier reports on the sheet were either all whites or fairly distributed; suggestion whether reports were made by other classmates (previous reports handwritten) or by someone else (preprinted).Continued on Next Page »