功能性磁振造影之俄羅斯方塊中的競爭以及人際互動

Abstract

競爭行為包含了報酬（rewarding）、決策（decision making）（Engelmann et al., 2009; Hewig et al., 2009; Kuo et al., 2009）以及人際互動（Krach et al., 2008）。我們採用了兩個獨特的設計，第一、當遊戲進行時，受試者並不知道對手是人或者是電腦；第二、可以懲罰對手的設計增加了遊戲之中的互動性。在這種內隱的環境下受試者的競爭行為是此實驗所要觀察的，其中包括報酬、決策以及理解對手的意圖。 有二十一個受試者參與本實驗，並且此實驗採用二乘二設計。第一個變項是競爭與否。在競爭狀況下，受試者可以通過在對手的畫面上增加方塊來懲罰對手，反之亦然。在非競爭狀況下則無法懲罰對手。第二個變數則為對手是人抑或是電腦。電腦對手是一個設計好的人工智慧程式，可以自動的玩俄羅斯方塊（Tetris）。因此，本實驗關注『競爭與非競爭』和『對手是人與對手是電腦』這兩個對比。除此之外，第三個所關注的對比就是『懲罰對手與被對手懲罰』，其中『懲罰』只會在競爭的狀況下發生。實驗中所有發生的事件都會被記錄下來。 在『競爭與非競爭』的對比下，背側前扣帶迴（dorsal anterior cingulate cortex， BA 32）、下部頂葉（inferior parietal lobule，BA 40）以及背側前額葉皮質（dorsal lateral prefrontal cortex，BA 10）會有較強的反應。在『人類對手相較於電腦對手』的對比之下，發現右側下部頂葉的反應會與受試者的遊戲程度成正比。將受試者分成高程度群與低程度群。在高程度群的受試者中發現，在『人類對手相較於電腦對手』的對比之下，右側下部頂葉有比較強的反應。在『懲罰對手相較於被對手懲罰』的對比之下，發現背側前額葉皮質（BA 10）、背側前扣帶迴（BA 32）、前額葉基底區（orbital frontal cortex, BA 47）和紋狀體（striatum）有較強的反應。 結果顯示，報酬以及瞭解對手的意圖會涉入在俄羅斯方塊的競爭中。右側下部頂葉（BA 40）和瞭解對手的意圖有關（Krach et al., 2008）。背側前額葉皮質（BA 46,10）、背側前扣帶迴（BA 32）、前額葉基底區（BA 47）及紋狀體和報酬有關（Harber, 2009）。此外，在這種內隱的環境下，遊戲程度較佳的受試者，在『人類對手相較於電腦對手』的對比之下，右側下部頂葉會有較強的反應。

Competition involves rewarding, decision making (Engelmann et al., 2009; Hewig et al., 2009; Kuo et al., 2009), and interpersonal interaction (Krach et al., 2008). In this study, two novel manipulations were used to further study competition through a real-time competitive game. First, the participant did not know whether the opponent was a human being or a computer during the game. Second, punishment was added to strengthen the interaction with an opponent. Under such an implicit circumstance, participants’ behavioral responses and brain activations were collected to examine the neural correlates of mental processing, including rewarding, decision making and interpreting opponent’s intention. Twenty-one right handed healthy volunteers participated in this study. This study used a two-by-two design. The first factor was “competition”. The competition (or non-competition) was referred to a situation that the opponent could punish (or could not punish) the participant by adding rows to the participant’s screen during the Tetris game, vice versa. The second factor was “Opponent” referring to whether the participant confronted with either a human opponent or a computer opponent. Thus, there were primary contrasts in our design, “competition versus non-competition”, and “human opponent versus computer opponent”. In addition, the punishing events occurred in the competition condition but not non-competition condition. Hence, the third contrast was “punishing the opponent versus being punished by the opponent”. The computer opponent was an artificial intelligence program to play automatically the Tetris game. In an event related fMRI design, all participants’ responses were recorded. The contrast of competition versus non-competition produced greater activation in dorsal anterior cingulate cortex (BA 32), inferior parietal lobule (BA 40), and dorsal lateral prefrontal cortex (BA 10). In the contrast of “human opponent versus computer opponent”, signal intensity at right inferior parietal lobule was correlated with the level of the participant playing in the game. Moreover, participants were split into higher level and lower level players. For the higher level players, the contrast of human opponent versus computer opponent produced greater activation at the right inferior parietal lobule (BA 40). The contrast of “punishing the opponent versus being punished by the opponent” produced greater activation in dorsal lateral prefrontal cortex (BA 10), dorsal anterior cingulate cortex (BA 32), orbital frontal cortex (BA 47) and striatum. The result showed that interpreting opponent’s intention and rewarding were involved in the competitive Tetris game. Right inferior parietal lobule (IPL, BA 40) were related to perceive the intentional action (Krach et al., 2008). Frontal lobe including dorsal lateral prefrontal cortex (DLPFC, BA 46, 10), orbital frontal cortex (OFC, BA 47), dorsal anterior cingulate cortex (dACC, BA 24, 32), and striatum were related to rewarding (Haber, 2009). Under an implicit circumstance, participants with higher playing level engaged more competing behavior associated with greater activation in right-inferior parietal lobule (IPL, BA 40) when competing with a strategic and flexible human opponent.