时间压力下序贯决策研究进展——基于TAM扩展框架的评述

Statue of Sequential Decision-making with Time Pressure:Review with an Extentional TAM Framework

序贯决策代表了一大类决策问题,包括救灾决策、工作搜索、股票操作等,这类问题内生的时间压力使得"观望"和"跟随"策略不再适用。"截止日期"、"机会成本"带给决策者的不仅是时间压力,还迫使他们转向直觉决策,随之产生系统性可预见性偏差。本文通过文献梳理发现,时间压力下的序贯决策行为还存在亟需解决的问题:时间压力的影响机理还不清楚。主要原因在于缺乏统一的分析框架,孤立考察时间压力在特定问题情境下的影响,没有揭示作用的机理。因此,本文扩展了TAM框架,提出了"心理——行为——结果"的系统分析框架,对时间压力下的序贯决策研究文献进行梳理和回顾,结果表明,基于TAM扩展框架进行研究有助于揭示时间压力的影响机理。最后,文章为下一步的工作指出了有潜力的方向及研究方法。

Many real world decision-making problems are sequential in nature. A series of choices is made available over time, and it is often efficient （and sometimes even necessary） to make a selection without waiting to be presented with all of the alternatives. So two strategies wait to see and me too, may not be available. On mate choice task, for example, people choose their mate at one of a sequence of alternatives, without knowing the next al- ternatives is better or worse. This type of sequential decision has a coritinuous utility function. People aim to choose the best mate, but no way to measure their selection. There are many similar problems, including online DM, emer- gency DM, job search, stock operation, and so on. This decision-making scenario has the same essential features as a recreational mathematics problem known as the ＇sequential decision making＇ or ＇secretary problem＇ （see Fergu- son 1989 for a historical overview）. In sequential DM problems, an observer is presented with a sequence of possible choices, and must decide whether to accept or reject each possibility in turn. The number of choices in the complete sequence is fixed and known, and only the rank of each possibility, relative to those already seen, is presented to the observer. If the observer chooses the best possibility in the sequence, their decision is correct, and any other choice is regarded as incorrect. Such sequential decision problem has been know by many different names in the literature, but the most com- mon name is the ＂secretary problem＂, which appeared first in Scientific American and described by six assump- tions. Much studies investigated the optimal strategies of each assumption were relaxed. In recently, some studies in- dicated that decision maker may not relay on the optimal strategy to make decision, otherwise intuitive decision, un- der time pressure. Thus, some systematic and predictable biases accompanied with their search behavior. These bia- ses can be avoided as long as identification, so some research aim to identify the bias caused by endogenous time pressure of sequential problem. First of all, we should reveal effect of time pressure on decision making. There are some research found that a speed-accuracy tradeoff can occur with time constraints, and that individuals utilize many noncompensatory coping strategies,including acceleration and filtration of information. But there exist some problem inclusive, such as what happens, too late or too early, when decision maker stop searching information under time pressure？ There are two Hypotheses about decision maker＇ s search behavior. For one issue, according to intertempo- ral choice theory, human behavior is impatient, prefer the earlier results obtained, so they may under search and choose too early （Dong,2008）. For another issue, according to Loewenstein ＆ Perlec（1993） ,decision maker pref- erences increasing sequence, so they may over search and choose too late. This paper proposed that there are some gaps between established papers~ May be the reason is none of an integrated framework, each study survey the influ- ence of time pressure in appointed problem and the mechanization is not clear~ This paper proposed an integrated framework, psychology-behavior-result, based on TAM （Technology Accept- ance Mode） ,which is the most popular framework in IS. We extract the causal logic of TAM and proposed a system framework to analysis the influence mechanization and then reviews the literatures. Considering not only time pres- sure on psychology, behavior and results of individual impact, and considering the internal logic between the three issues. The aim of this study may answer two questions. First, why there are some inconsistent impact on DM＇ s search behavior in time pressure？ Second, how do we regulate this contradiction？ At last, some future areas of research bring this paper into a close. Firstly, some differences exist in search bias according to different theory hypothesizes, we should empirical study and give a reasonable expiation. Secondly, ex- plain the mechanic of time pressure on decision results is very important to improve quality of decision- making. Thirdly, laboratory experiment is feasible method for revealing the impact of time pressure on sequential DM.