Max Planck Institute for Human Development, Berlin, Germany.
Neuroimage. 2010 Feb 1;49(3):2556-63. doi: 10.1016/j.neuroimage.2009.10.060. Epub 2009 Oct 29.
Many decisions people make can be described as decisions under risk. Understanding the mechanisms that drive these decisions is an important goal in decision neuroscience. Two competing classes of risky decision making models have been proposed to describe human behavior, namely utility-based models and risk-return models. Here we used a novel investment decision task that uses streams of (past) returns as stimuli to investigate how consistent the two classes of models are with the neurobiological processes underlying investment decisions (where outcomes usually follow continuous distributions). By showing (a) that risk-return models can explain choices behaviorally and (b) that the components of risk-return models (value, risk, and risk attitude) are represented in the brain during choices, we provide evidence that risk-return models describe the neural processes underlying investment decisions well. Most importantly, the observed correlation between risk and brain activity in the anterior insula during choices supports risk-return models more than utility-based models because risk is an explicit component of risk-return models but not of the utility-based models.
许多人们做出的决策可以被描述为风险决策。理解驱动这些决策的机制是决策神经科学的一个重要目标。有两类相互竞争的风险决策模型被提出来描述人类行为，即基于效用的模型和风险回报模型。在这里，我们使用了一种新颖的投资决策任务，该任务使用（过去的）回报流作为刺激，以研究这两类模型与投资决策（其中结果通常遵循连续分布）背后的神经生物学过程的一致性。通过表明（a）风险回报模型可以从行为上解释选择，以及（b）风险回报模型的组成部分（价值、风险和风险态度）在选择过程中在大脑中得到体现，我们提供了证据表明风险回报模型很好地描述了投资决策背后的神经过程。最重要的是，在选择过程中前脑岛的风险与大脑活动之间的观察到的相关性支持风险回报模型多于基于效用的模型，因为风险是风险回报模型的一个明确组成部分，而不是基于效用的模型的组成部分。
