Toddlers exhibit behaviours that suggest judicious responses to states of uncertainty (for example, turning to adults for help), but little is known about the informational basis of these behaviours. Across two experiments, of which experiment 2 was a preregistered replication, 160 toddlers (aged 25 to 32 months) identified a target from two partially occluded similar (for example, elephant versus bear) or dissimilar (for example, elephant versus broccoli) images. Accuracy was lower for the similar trials than for the dissimilar trials. By fitting drift–diffusion models to response times, we found that toddlers accumulated evidence more slowly but required less evidence for similar trials compared with dissimilar trials. By analysing eye movements, we found that toddlers took longer to settle on the selected image during inaccurate trials and switched their gaze between response options more frequently during inaccurate trials and accurately identified similar items. Exploratory analyses revealed that the evidence-accumulation parameter correlated positively with the use of uncertainty language. Overall, these findings inform theories on the emergence of evidence accumulation under uncertainty.

幼儿表现出的行为暗示对不确定状态做出明智的反应（例如，向成年人求助），但对这些行为的信息基础知之甚少。在两个实验中，其中实验2是预先注册的复制，有160个幼儿（年龄在25到32个月之间）从两个部分被遮盖的相似（例如，大象与熊）或不相似（例如，大象与西兰花）图像中识别出目标。相似试验的准确性低于相似试验的准确性。通过将漂移扩散模型拟合为响应时间，我们发现与不相似的试验相比，幼儿收集证据的速度较慢，但​​对相似试验所需的证据较少。通过分析眼睛的运动，我们发现，在不准确的试验中，幼儿需要更长的时间才能停留在选定的图像上，而在不准确的试验中，幼儿可以更频繁地在注视选项之间进行切换，并且可以准确地识别出类似的项目。探索性分析表明，证据积累参数与不确定性语言的使用呈正相关。总体而言，这些发现为不确定条件下证据积累的出现提供了理论依据。