Abstract: Probability discounting and temporal discounting are important phenomena in people’s daily economic decision making. Probability discounting occurs when people make a choice between different risky options. People tend to have a preference for the safe option in risky decision making. Temporal discounting refers that people tend to discount future benefits when facing decisions involving a smaller immediate gain and a larger future one. In recent years, researchers have had a strong interest in the relationship between probability discounting and temporal discounting. One hypothesis is that delayed rewards options might be evaluated as risky choices, leading to people’s preference for the prompter payoff option in inter-temporal choice tasks. This hypothesis has been supported by some behavioral studies. However, several behavioral studies challenge the single-process explanation. These behavioral studies revealed inconsistent conclusions for the hypothesis. The fMRI studies also suggested the dissociation between probability discounting and temporal discounting. However, the fMRI evidences are based on separate studies using different paradigms. We employed event-related potential (ERP) to explore whether probability discounting and temporal discounting reflected the same underlying process. ERP can provide a continuous measure of mental processing, making it possible to get a clear evidence for this question. The present study recorded ERPs while participants performed a modified version of the “Wheel of Fortune” gambling task. In the task, probability factor (Risky vs. Safe) and temporal factor (One Month vs. Today) was permuted to form four conditions: Today & Safe, Today & Risky, Month & Safe, and Month & Risky. The study design yielded four conditions based on the participants’ selections. Previous ERP studies on probability discounting have identified the error-related negativity (ERN) as a key index of probability discounting. While the ERP studies on temporal discounting showed inconsistent ERP components. We would examine whether the ERN effect would also appear in the temporal discounting process. In order to get the accurate conclusion, we also analyzed all the ERP components of the two processes. We then first examined whether the ERN can also be elicited by choices involving delays. We found that the ERN elicited by the Risky choice was more negative than that elicited by the Safe one, whereas ERN for Month choices did not differ from ERN for Today choices. We then found the different distributions of time process and the electrode between the two processes. Specially, probability discounting occurred in time window of -100~50ms on FCz and time window of 100~350ms on Pz, while the temporal discounting occurred in time window of 200~600ms on Pz. In conclusion, the current study showed that the probability discounting effect (Risky vs. Safe) and the temporal discounting effect (Month vs. Today) had different patterns in ERP waveforms. The differences suggested that the two processes may involve different neural mechanisms.

Key words: probability discounting, temporal discounting, event related potential (ERP), error-related negativity (ERN)

摘要： 本研究运用ERP技术来检验概率贴现和时间贴现是否反映了相同的神经机制过程。在本研究中，我们在一个刺激中同时操纵概率贴现（风险vs.安全）和时间贴现（1个月后vs.今天）水平。通过对两个过程上的ERN以及全时间段的ERP分析来确定它们是否反映了相同的认知过程。结果发现，这两者在ERN上存在差异，同时这两个效应在ERP的时程和出现的电极点上都存在差异。这些差异表明这两者可能包含着不同的神经机制过程。

关键词: 概率贴现, 时间贴现, 事件相关电位（ERP）, 错误相关负波（ERN）