Subjective Versus Objective Perception in Risk-Sensitive Foraging

当决策过程的结果未知，但可以通过某种发生的概率来指定时，通常必须做出决策。在这种不确定性条件下的决策分析已经成为动物行为研究的一个主要焦点。决策过程受到两个因素的影响：特定结果（例如，一定数量的食物）对生物体的益处以及该结果出现的可能性。迄今为止，大多数研究都考察了生物体评估和权衡利益的方式，但很少关注生物体如何评估和构建概率。相当多的经验证据表明，人们系统地对概率做出有偏见的估计。我们始终认为，低概率事件比实际发生的可能性更大，而高概率事件发生的可能性更小。人们似乎将概率均质化，使所有事件看起来比实际发生的可能性更均等。里尔博士的研究将确定动物（在本研究中是白足鹿鼠）在评估概率时表现出同样偏见的程度。这种系统性偏见会对我们解释行为实验的能力产生深远的影响。雷亚尔博士开发了一种方法，可以独立地建立生物体对概率的主观感知。通过让老鼠在不同的实验得出的概率中进行选择，他可以构建主观概率来解释观察到的选择行为模式。通过这种方式，他可以划分影响决策的双重因素。他也许还能说明一些限制人类和动物行为选择的共同认知结构。

Decisions often must be made when the outcome of the decision process is not known but can be specified by some probability of occurring. The analysis of decision-making under such conditions of uncertainty has become a major focus of research in animal behavior. The decision process is influenced by two factors: the benefit to the organism from a particular outcome (say, a certain quantity of food) and the probability of that outcome. Most research to date has examined the manner in which organisms assess and weigh benefits, but little attention has been paid to how organisms assess and construct probabilities. A considerable amount of empirical evidence suggests that people systematically make biased estimates of probabilities. We consistently believe that low-probability events are more likely and high-probability events are less likely to occur than they actually are. People seem to homogenize probabilities, making all events seem more equally likely than they actually are. Dr. Real's research will determine the degree to which animals (in this case, white-footed deer mice) show the same kind of bias in their assessment of probabilities. Such systematic bias would have profound consequences for our ability to interpret experiments on behavior. Dr. Real has developed a means of independently establishing the organism's subjective perception of probabilities. By allowing mice to choose over different experimentally derived probabilities, he can construct the subjective probability that accounts for the pattern of observed choice behavior. In this way he can partition the dual factors influencing decision-making. He may also be able to say something about the common cognitive structures that constrain behavioral choices in humans and animals.