Based on the criticism and development of expected utility theory, value function and weight function were proposed by prospect theory, giving a good prediction and explanation to the phenomena found in previous studies of risky decision-making. First of all, prospect theory proposes a value function over gains and losses relative to a reference point. According to prospect theory, the status quo generally serves as the reference point distinguishing losses from gains, so that the function is concave for gains and convex for losses. Secondly, value function of prospect theory is steeper for losses than for gains, which is a property known as loss aversion. People typically require more compensation to give up a possession than the amount that they would have been willing to pay to obtain it in the first place. At last, in prospect theory, the value of an outcome is weighted not by its probability, but instead by a decision weight which exhibit that moderate to high probabilities are underweighted and low probabilities are overweighed. Besides, some decision-making biases phenomena, such as framing effect, endowment effect and default bias also help to reveal the neural basis of prospect theory. The field of neuroeconomics is providing a rapidly increasing amount of data regarding the phenomena that lie at the heart of prospect theory. Recently, studies on neuroimaging, especially functional magnetic resonance imaging showed that prefrontal cortex, striatum, insula and amygdala are four main brain regions that relate to prospect theory. Specifically, studies on reference-dependence showed that prefrontal cortex (PFC), including orbitofrontal cortex (OFC) and medial prefrontal cortex (mPFC), and ventral striatum (vStr) may be involved in reference-dependent processing. Some argues that reference-dependence might be a general concept that includes many factors, including choice history, and may thus be dynamic. Therefore, PFC and vStr may represent different types of reference-dependence. Studies on loss aversion found striatum, ventromedial prefrontal cortex (vmPFC) and amygdala play key roles in enlarging losses. Researchers even found striatum and vmPFC exhibited a pattern of “neural loss aversion”, that these regions’ slope of the decrease in activity for increasing losses was greater than the slope of the increase in activity for increasing gains in a majority of participants. A study on individuals with focal bilateral amygdala lesions found that they showed a dramatic reduction in loss aversion compared to matched controls, suggesting that the amygdala plays a key role in generating loss aversion by inhibiting actions with potentially deleterious outcomes. Therefore, the neural basis under loss aversion may be a neural system consists of regions like striatum and amygdala. Amygdala first processed the information, and then transferred to the striatum, eventually led to the loss aversion. Finally, researchers found that weight function may be partly caused by anterior cingulate cortex (ACC) and dorsal striatum (dStr). But it still remained to be clarified whether these regions contribute to the pattern of weight function. In order to get full understanding of decisions under risk, with what have been already found, studies in the future may further explore prospect theory from different angles, like how prospect theory originates from evolutionary perspective and how genes play a part in it.