We report on a two-stage experiment in which i) we first elicit the social network within a section of undergraduate students and ii) we then measure their altruistic attitudes by means of a standard Dictator game. We observe that more socially integrated subjects are also more altruistic, as betweenness centrality and reciprocal degree are positively correlated with the level of giving, even after controlling for framing and social distance, which have been shown to significantly affect giving in previous studies. Our findings suggest that social distance and social integration are complementary determinants of altruistic behavior

Research on moral cleansing and moral self-licensing has introduced dynamic considerations in the theory of moral behaviour. Past bad actions trigger negative feelings that make people more likely to engage in future moral behaviour to offset them. Symmetrically, past good deeds favour a positive self-perception that creates licensing effects, leading people to engage in behaviour that is less likely to be moral. In short, a deviation from a ‘normal state of being’ is balanced with a subsequent action that compensates the prior behaviour. We model the decision of an individual trying to reach the optimal level of moral self-worth over time and show that under certain conditions the optimal sequence of actions follows a regular pattern which combines good and bad actions. To explore this phenomenon we conduct an economic experiment where subjects play a sequence of giving decisions (dictator games). We find that donations in the previous period affect present decisions and the sign is negative: participants' behaviour in every round is negatively correlated to what they did in the past. Hence donations over time seem to be the result of a regular pattern of self-regulation: moral licensing (being selfish after altruistic) and cleansing (altruistic after selfish).

This study revisits different experimental data sets that explore social behavior in economic games and uncovers that many treatment effects may be gender-specific. In general, men and women do not differ in “neutral” baselines. However, we find that social framing tends to reinforce prosocial behavior in women but not men, whereas encouraging reflection decreases the prosociality of males but not females. The treatment effects are sometimes statistically different across genders and sometimes not but never go in the opposite direction. These findings suggest that (i) the social behavior of both sexes is malleable but each gender responds to different aspects of the social context, and (ii) gender differences observed in some studies might be the result of particular features of the experimental design. Our results contribute to the literature on prosocial behavior and may improve our understanding of the origins of human prosociality. We discuss the possible link between the observed differential treatment effects across genders and the differing male and female brain network connectivity, documented in recent neural studies.

Subjects’ decisions in multiple-choice tests are an interesting domain for the analysis of decision making under uncertainty. When the test is graded using a rule that penalizes wrong answers, each item can be viewed as a lottery where a rational examinee would choose whether to omit (sure reward) or answer (take the lottery) depending on risk aversion and level of knowledge. We formalize students as heterogeneous decision makers with different risk attitudes and levels of knowledge. Building on IRT, we compute the optimal penalty given students’ optimal behavior and the trade-off between bias and measurement error. Although MCQ examinations are frequently used, there is no consensus as to whether a penalty for wrong answers should be used or not. For example, examinations for medical licensing in some countries include MCQ sections with penalty while in others there is no penalty for wrong answers. We contribute to this discussion with a formal analysis of the effects of penalties; our simulations indicate that the optimal penalty is relatively high for perfectly rational students but also when they are not fully rational: even though penalty discriminates against risk averse students, this effect is small compared with the measurement error that it prevents.