We report the results of two laboratory experiments that study how university student and staff participants chose retirement savings investment options using ‘user-friendly' information prescribed by regulators. We demonstrate that choices of more than 20% of participants cannot be predicted...