In decision making, people can be both rational and irrational. Recent research suggests that human decisions are optimal under cognitive resource constraints. This 'resource-rational' approach connects psychology and neuroscience with engineering, economics, and machine learning. By focusing on cognitive resources, the study sheds light on why we have biases like reference dependence, stochastic choice, and perseveration. These biases may actually be rational responses to the limits of our decision-making abilities.