Research conducted in the lab of Professor Christopher Summerfield has offered insights into why people sometimes make "irrational" choices about economic goods. For example, faced with choices among three goods - for example, apples, oranges and pears - humans who prefer apples over oranges, and oranges over pears should also prefer apples over pears. However, we sometimes show "intransitive" choices that violate this rational principle (e.g. preferring pears over apples).
The research, which was led by Dr. Konstantinos Tsetsos, showed that humans make intransitive choices because of a decision bias that they call "selective integration". Although this bias led to more intransitive chocies, Tsetsos and colleagues used computer-based simulations to demonstrate that it paradoxically increased monetary outcomes when the decision process was "noisy", or corrupted by variability in neuronal activity. In other words, it's sometimes "optimal" to be "irrational".