The current thesis set out to investigate the suitability of the IRAP to assess attitudes in the moral domain and to predict cheating behaviour in a controlled context. Across six studies in two countries, we developed three IRAPs that targeted relations between actions and concepts of morality, reports of frequency of moral and immoral behaviour, and personal feelings towards engaging in moral or immoral actions, and interpreted our findings through the Relational Elaboration and Coherence (REC) Model. In the first part of the current research programme, correlations between the IRAPs and a cheating task suggested that individuals who are highly practised at immoral behaviour such as cheating, deceiving and lying are more likely to confirm that they do not engage in such behaviours, in itself an instance of that behavioural class. Further studies revealed that a history of bad feelings associated with engagement in immoral behaviour correlated with lower cheating, and that higher pro-moral biases in the IRAP correlated with lower reported psychopathic traits. In the latter part of the research programme described in the current thesis, a values-oriented intervention was shown to have an effect on IRAP performance and to produce a nonsignificant trend toward decreasing cheating levels. To conclude, strengths, limitations and opportunities for further research are discussed.