本研究試圖探討「偏好逆轉現象」之所以產生或無法消除的動機性與認知性原因。利用期望值相等的彩券配對選擇遊戲，我們在實驗中進行下列變項的操弄：（1）以實際的獎金引發受試者的動機，（2）將彩券選擇數目減少成一對，（3）利用模擬市場交易方式，與受試者進行彩券交易；觀察偏好逆轉現象可否消除。此外，並測量受試者的內外控信念與偏好逆轉行為之關係及賭博遊戲對其控制錯覺的影響，結果顯示：（1）臺灣大學生與美國大學生有相同程度的偏好逆轉現象，（2）當彩券選擇數目只有一對時，受試者認為較易選擇，偏好逆轉現象也較少，（3）透過市場交易過程，受試者可以完全消除偏好逆轉行為，（4）無論是否發生偏好逆轉行為，受試者大多數認為其乃一合理之現象，因為「一券機率大，另一券金額高，判斷方法不同」，但此極不同選擇策略並非其實際發生逆轉行為的原因，（5）個人內控信念可以支持其作合理的偏好逆轉行為，（6）當認為輸贏結果較不可能來自「自己判斷」時也會解釋其逆轉行為是合理的，（7）贏錢經驗會促成對賭博遊戲的控制錯覺。

　　　　　The purpose of this study was to examine the preference reversal phenomenon which occurs systematically and frequently in human society. Three motivational and one cognitive factors were considered: (1) the monetary incentives, (2) the difficulty or complication of the game, (3) the market-like arbitrage and (4) the locus of control of the subject. The results indicated that: (1) the preference reversal frequencies of Taiwanese college students were consistent with that of American subjects in Reilly's study. (2) The subjects, who perceived the game of one pair was much easier than that of six pairs, would reduce the frequencies of preference revesal. (3) Under repeated arbitrage conditions, the reversers realized the amount of his token money was reducing, the preference reversal behavior would disappear. (4) Whether they did reversals or not, most subjects thought it would be reasonable to have reversals, according to the different strategies being taken between the preference choice and reservation prices. (5) The reverser who had more internal control in general, thought reversal behavior was more reasonable; those who thought he had less personal control inthe games, explained reversals could be reasonable too. (6) The winning of the games would induce the subjects' perceptions of personal control over the games-a kind of illusion of control.