US President Barack Obama's executive order of 15 September, 'Using behavioral science insights to better serve the American people', is intended to strengthen the policymaking process. To prevent the Obama administration from falling victim to agendas backed by pseudoscience, I suggest that it should consider only studies that abide by certain established rules.

The rules were instituted in behavioural and experimental economics research in universities, largely in response to scientific misconduct by governments and academics. They stipulate that an independent ethics committee must review research involving human subjects before the start of any experiment; that participants should first give their informed consent; and that they should receive appropriate remuneration to ensure that their decisions reflect genuine preferences. Moreover, deception should never be used in experimental design or in the process of scientific enquiry.

In the interests of scientific transparency, studies that lead to policy change should be peer reviewed and experimental design details and data made publicly available.

Such guidelines would help to ensure that studies are relevant, rigorous, ethical and legal. They would also provide a means to replicate and test studies for bias, efficacy and accuracy.