臺灣博碩士論文加值系統

本文欲了解個體面對不確定的選項時選擇的行為，本文使用了 two-armed bandit 的問題來研究，我們參考了J.J. Goodnow (1955) 的實驗比較了不同資訊回饋對受測者選擇行為的影響。在實驗裡受測者需要在畫面裡選擇 “A” 按鈕或是 “B” 按鈕，在做完選擇後受測者將會進入結果畫面。我們所設計的四個實驗分別在結果畫面有不同的資訊回饋或有不同的報酬計劃。利用受測者在做出轉換行為前得到 “0” 報酬的次數，利用這個方式建構一個截斷點策略 (cut point) 當作是受測者行為指標。再者又考慮到受測者決策時可能不管我們所給的資訊回饋，個體決策的差別可能只是轉換次數的多寡，因此以四個實驗裡所有受測者轉換次數的中位數當作分組的標準。
研究的結果顯示：幾個不同資訊回饋的平均報酬，在我們的實驗裡看不出差異，顯示更多的資訊無法幫助受測者選擇有較高報酬的選項導致使平均報酬沒有差異。在轉換次數方面，轉換次數在受測者之間的差異很大但只有實驗二、三和實驗三、四的平均轉換次數不同。表示資訊回饋不同對受測者不確定行為上的轉換行為上影響不大。在實驗裡轉換次數的趨勢是會隨著嘗試 block 數增加而減少。另外我們檢視了兩個實驗的截斷點策略，四個實驗裡的受測者在這個表現上沒有一致的趨勢。最後我們把受測者分成了高轉換組與低轉換組之後，截斷點的趨勢在 block3 之前是相同的。資訊回饋對個體在不確定行為下的選擇影響沒有差異。個體行為受到個體本身差異的影響較多，相對的不同資訊回饋影響較小。

The study wants to comprehend individual choosing behavior under uncertainty, by using two-armed bandit problem. We compared different feedbacks impact on subjects choosing behavior. In four experiments subject decide click button “A” or button “B”, then subjects will enter the result phase. Four experiments have different feedbacks in result phase or different payoff schedules. We use times of button “A” being chosen payoff、switch number and cut point as indexes measuring behavior. The cut point index is constructed by counting how many times payoff “0” occurred between each switches. Also, considering that individual may ignore feedbacks that we gave. People using different strategies may simply a result of they have different switching frequency. We addressed this by using the median that switching times in all four experiments to category our subjects.
Results showed that different feedbacks have no effect on individual choosing behavior under uncertainty. Instead, we found that different among individuals has effect on choosing behavior under uncertainty.

一、緒論 1
1.1 研究背景與動機 1
1.2 研究問題 2
1.3 研究目的 3
二、. 文獻回顧 4
2.1 Two-armed bandit 實驗 4
2.2 Two-armed bandit 實驗結果與模型的對照 9
2.3 Two-armed bandit 資訊的獲取與學習 15
三、 實驗設計 18
3.1 實驗概述 18
3.2 實驗參數 19
3.2.1 實驗一 21
3.2.2 實驗二 22
3.2.3 實驗三 23
3.2.4 實驗四 23
3.3 使用軟體 24
3.4 實驗流程 24
四、實驗結果 27
4.1 平均表現 27
4.2 選擇的狀況 28
4.3 轉換次數 29
4.5 轉換次數分組 34
4.5.1 平均表現 34
4.5.2 選擇的狀況 35
4.5.3 截斷點 37
五、結論與展望 39
參考文獻 41
附錄一 實驗使用之指導語 44
附錄二 截斷點結果 45
附錄三 簡單回歸模型結果 48
附錄四 Box-plot 50

1.Banks, J., Olson, M. and Porter, D. (1997), “An experimental analysis of the bandit problem,” Economic Theory, 10, pp 55-77.
2.Becker, G .m., DeGroot, M.H., Marschak, J. (1986), “Measuring utility by a single-response sequential method,” Behavioral Science , 9, 226-232.
3.Berg, J. E., Daley, L. A., Dickhant, J. W., O’Brain, J. R. (1986), “Controlling Preferences for Lotteries on units of experimental exchange,” Quarterly Journal of Economic, 101, 281-306.
4.Camerer, C. (1987), “Do biases in probability judgment matter in markets? Experimental evidence,” American Economic Review, 77, 987-71.
5.Camerer, Colin F. and Robin M. Hogarth. (1999), “The effects of financial incentives in experiments: A review and capital-labor production framework,” Journal of Risk and Uncertainty, 19, pp. 7-42
6.Einhorn, H. and Hogarth, R. (1985), “Ambiguity and Uncertainty in Probabilistic Inference,” Psychological Review, 93, 433-461.
7.Fernie, G. and Tunney, R. J. (2006), “Some decks are better than others: The effect of reinforcer type and task instructions on learning in the Iowa Gambling Task,” Brain and Cognitior, 60, 60, pp. 94-102.
8.Fischbacher, U. (2006), “Z-tree: Zurich toolbox for readymade economic experiments,” Experimental Economics, pp. 171-178.
9.Gabiax, X., Laibson, D., Moloche, G. and Weinberg, S. (2006), “Costly information Acquisition: Experimental Analysis of a Boundedly Rational Model,” The American Economic Review, 96(4), pp. 1043-1068.
10.Gans N., Croson R. and Knox G. (2007), “Simple Models of Discrete Choice and Their Performance in Bandit Experiments,” Manufacturing Service Operations Management, 9, 383-408.
11.Grether, D. M. (1992), “Testing Bayes rule and the representativeness heuristic： Some experimental evidence,” Journal of Economic Behavior and Organization, 17, 31-57.
12.Herrnstein, R. J. and Loewenstein, G.F. (1993), “G. F. Utility Maximization and Melioration: internalities in Individual Choice,” Journal of Behavioral Decision Making, 6, pp. 149-185.
13.Gittins, J. C. and Jones, D. M. (1974), “A Dynamic Allocation Index for the Sequential Design of Experiments,” Progress in Statistics, pp. 241-266.
14.J.J. Goodnow (1955), “Determinants of Choice-Distribution in Two-Choice Situations,” The American Journal of Psychology, Vol. 68, No. 1, pp. 106-116.
15.Meyer, R. J. and Shi, Y. (1995), “Sequential Choice Under Ambiguity Intuitive Solutions to the Armed-bandit problem,” Management Science, 41 (5) ,pp. 817-834.
16.Read, G., Loewenstwin, G. and Kalyanaraman, S. (1999), “Mixing Virtue and Vice: Combining the Immediacy Effect and the Diversification Heuristic,” Journal of Behavioral Decision Making, 12, 12, pp. 257-273.
17.Tunney, R. J. and Shanks, D. R. (2002), “A Re-examination of Melioration and Rational Choice,” Journal of Behavioral Decision Making, 15, pp. 291-311.
18.Tunney, R.J. (2006), “Short article Preference reversals are diminished when gambles are presented as relative frequencies,” The Quarterly Journal of Experimental Psychology, 59(9), pp. 1516-1523.
19.Tversky, A., “Intransitivity of preference (1969),” Psychological Review, 76, 31-48.
20.Tversky, A., and Kahnrman, D. (1974), “Judgment under uncertainty: Heuristics and bias,” Science, 185, 1124-31.
21.Zauberman, G. (2003), “The Intertemporal Dynamics of Consumer Lock-In,” Journal of Consumer Research, 30, pp. 405-419.