The fact that human economic behaviour has a significant irrational element - one that is simultaneously hard-to-explain and highly predictable - has fascinated economists for decades from Fechner, 1860 to Shiller, 2005 and beyond. In this dissertation, I investigate the field from various perspectives: chapter 1 examines the impact that language describing irrational behaviour in the media has on stock markets; chapter 2 looks at whether musical harmonics can predict what choices participants in money-sharing games will make; and chapter 3 takes an existing theroetical model of stochastic decision-making and changes it to help explain phenomena such as the overweighting of small probabilities, the 'willingness-to-accept'-'willingess-to-pay' (WTA-WTP) disparity, and preference reversals.