測試不同打鬥經驗間之互相干擾效應

No Thumbnail Available

Date

2008

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

過去打鬥輸贏經驗被觀察到會影響個體下一次的打鬥行為:曾經贏過的個體再打鬥時獲勝機率會增高,而曾輸過的個體再打鬥時獲勝機率會降低。過去的輸贏經驗被認為是經由提供有關一個體自身打鬥能力的資訊而改變此個體再次打鬥時的行為決策與獲勝機率。過去打鬥經驗之影響力長久以來就被發現會隨著時間而衰退,但是一直沒有研究測試新舊打鬥經驗之間是否會互相干擾而造成彼此效應減弱。本研究使用紅樹林鱂作為實驗物種以了解新舊輸的經驗效應間是否互相干擾。本研究以輸的經驗來偵測干擾效應,因為過去研究顯示敗者效應比勝者效應易偵測,並且敗者效應維持時間較長。如果不同經驗的影響會相互干擾,此干擾應較易由輸經驗間的交互作用偵測出來。我進行了兩組打鬥實驗(LL_NL打鬥組和LN_NN打鬥組;L表示輸的經驗、N表示無經驗;NL個體表示第一天不給予經驗、第二天給予輸的經驗;LN個體表示第一天給予輸的經驗、第二天不給予經驗,LL個體和NN個體以此類推)。此兩組打鬥組,LL個體比其對手NL個體多一個(第一天的)輸經驗,而LN個體也比其對手NN個體多一個(第一天的)輸經驗。因此若LL個體之兩個輸的經驗間沒有任何干擾作用,LL個體(與NL個體對打)之獲勝機率應與LN個體(與NN個體對打)之獲勝機率相當。但若兩個新舊輸的經驗間互相干擾而使其效應總和減小,則LL個體(與NL個體對打)之獲勝機率應大於LN個體(與NN個體對打)。本研究結果顯示LL個體(與NL個體打鬥時)的獲勝機率顯著地比LN個體(與NN個體打鬥時)的獲勝機率來的高。此結果應來自於LL個體之兩個輸經驗間互相干擾,造成LL經驗效應顯著低於此二經驗單獨作用時之各自效應的總和。然而此干擾效應只被發現於一個月前兩個競爭者皆曾落敗之配對。在一個月前皆獲勝之配對中,LL個體(與NL個體打鬥時)獲勝的機率和LN個體(與NN個體打鬥時)獲勝機率沒有顯著差異,新舊經驗間未見明顯干擾效應。此外本實驗亦發現有敗者效應的存在,LL_NL打鬥組的打鬥時間較LN_NN打鬥組短,其發生激烈打鬥的機率較低而且組中競爭者提早認輸的機率較高,以上的結果都顯示競爭者有較多輸經驗時,其行為表現較為消極被動。
Considerable evidence indicates that experience in prior contests influences how individuals interact with each other in subsequent contests: a previous loss causes an individual to behave more submissively, while a previous win causes an individual to behave more aggressively. These experience effects are thought to be caused by individuals altering their estimated fighting ability and costs of fighting after wins and losses. The value of information from past experiences should be affected by the reliability of the information, which could decrease with the passage of time (time-driven decay) or with the occurrence of new events (event-driven decay). Although the time-driven decay of experience effects on contest behavior has been studied, the event-driven decay has never been addressed. Using Kryptolebias marmoratus as the study animal, I tested whether a new experience causes the effect of a previous losing experience to decay. I gave each individual two experiences and staged 2 different types of contests: LL-NL (LL individuals fighting against NL individuals) and LN-NN. If the second losing experience causes the effect of the first losing experience to decay, I would expect the LL individuals in the LL-NL contests to be more aggressive and win more contests against their opponents than the LN individuals in the LN-NN contests. A total of 195 LL-NL and 194 LN-NN contests were staged and LL individuals (in LL_NL contests) won significant more contests than the LN individuals (in LN_NN contests). This result suggests that the two losing experiences of the LL individuals interfere with each other such that the effect of the LL treatment appears to be smaller than the summed effect of a penultimate and a recent losing experience. Nonetheless, this interference effect was only observed in contests between contestants that had lost their last fight (more than a month ago) before being used for this experiment. The results also showed that there are loser effects in the LL_NL and LN_NN contests. The duration of LL_NL contests was shorter than that of LN_NN contests. Escalation rate was lower in LL_NL contests than in LN_NN contests. Besides, probability of early retreat in LL_NL contests was higher than in LN_NN contests. Individuals with more recent losing experiences tended to behave more submissively and retreat sooner.

Description

Keywords

打鬥經驗, 勝者效應, 敗者效應, 經驗衰退, 獲勝機率, 打鬥行為, 紅樹林鱂, contest experience, winner effect, loser effect, experience decay, event driven decay, probability of winning, contest behavior, fish fighting, fighting behavior, Kryptolebias marmoratus

Citation

Collections