Abstract: Recent researches indicated that the representations of time and numerosity are closely linked. Both the mode control model and the ATOM (a theory of magnitude) proposed that our brains use similar principles of information-accumulation to process time and numerosity. A great number of researches had revealed that the automatic processing of numerosity and numerical digits interferes with both duration and temporal order perceptions in a unidirectional way. Specifically, time distortions can be induced by task-irrelevant numerical information. For example, larger numerosity or numerical digits lead to temporal overestimations and smaller numerosity or numerical digits lead to temporal underestimations. On the other hand, larger digits are judged to occur later than the smaller digits inside a temporal sequence even both digits are presented simultaneously. Here, by summarizing recent behavioural and neuroscience studies, we try to explore the underlying cognitive and neural mechanisms of the unidirectional number-time interference. In terms of cognitive basis for the unidirectional interfering effect, mixed experimental facts had been demonstrated and diverse explanations proposed. For example, time estimation and numerosity perception were found to share common properties in that both obey the Weber’s law and show similar discrimination sensitivities. However, other studies by using brain damaged patients and healthy adults revealed a double dissociation of processing of numerosity and processing of time. The development of continuous magnitude prior to the development of discrete numerosity also explains the unidirectional interference effect. Other researches proposed that the congruency effect between time and numerosity could be attributed entirely to a response bias which is inherent in comparison tasks. Spatial attention was also suggested as a mediating factor to account for the unidirectional interfering effect in that the number system can activate a change of spatial attention more easily relative to the temporal system. The neural basis for the unidirectional interfering effect can be divided into three main theories. First, an activation superiority hypothesis posits that numerosity processing has an advantage relative to time processing in activating the parietal lobes which contain the joint brain centres of time and numerosity processing. Second, the coexistence hypothesis of common neural basis and independent brain mechanisms suggests that a common brain network, such as the thalamo-cortico-striatal circuits, is responsible for processing of both temporal and numerical information, and at the same time distinct brain activation patterns exist for the processing of time and numerosity, respectively. Third, the two-stage model of numerosity-time interaction hypothesis argues that the rIPC is the neural basis of interference effect at perceptual stage and IFG at decisional stage. Taken together, in order to answer the underlying mechanisms of the unidirectional interfering effect, future studies need concentrate on several important aspects. First, it would be interesting to ask in what specific stage, i.e., a perception stage or a decision stage, the unidirectional interference occurs. Second, does the unidirectional interference depend on the response expectation and perceptual set in explicit timing tasks mostly adopted by time-numerosity interference studies? Finally, it is worth to explore the specific temporal profiles of this unidirectional interfering effect.

Key words: numerical processing, time perception, unidirectional interference, cognitive mechanism, neural mechanism

摘要： 近年来，数量加工对时间知觉的单方向干扰效应受到广泛关注，然而关于该效应的认知和神经机制的研究却缺乏深度的文献梳理和理论分析。本文从抽象数量、数字符号等认知加工如何干扰时间知觉的角度，系统总结了近期数量加工单方向影响时间知觉的研究进展；并在对相关文献加以梳理的基础上，详细分析了该效应可能的认知和神经机制，并提出若干未来研究的设想。

关键词: 数量加工 时间知觉 单向干扰 认知机制 神经机制