Suppr超能文献

二元制网络街头斗殴视频能告诉我们什么以及不能告诉我们什么关于人类身体攻击行为的行为学、博弈论和性别差异现象学的内容。

What Dyadic Internet Street Fight Videos Can and Cannot Tell Us About the Ethological, Game Theoretic, and Sex-Differentiated Phenomenology of Human Physical Aggression.

作者信息

Potegal Michael, Li Siyuan, Kim Misu

机构信息

University of Minnesota, Minneapolis, Minnesota, USA.

University of Michigan, Ann Arbor, Michigan, USA.

出版信息

Aggress Behav. 2025 Jan;51(1):e70017. doi: 10.1002/ab.70017.

Abstract

Street fight videos on the internet may provide information about little known aspects of human physical aggression, but their reliability is unclear. Analyses of 100 dyadic fight videos addressing ethological, game theoretic and sex-differentiated questions derived from research on other animals found that prefight verbalizations or gestural signals of nonaggressive or aggressive intent loosely predicted who would strike first and who would win. The head is the preferred strike target. Ordinal severity rankings of different strikes ranged from 1 for spitting to 5 for choking. Half the videos showed briefer, unilateral assaults beginning with one or more high severity strikes, little evidence of escalation and fewer bystander interventions. A quarter of these were sneak attacks. The other videos showed longer fights with reciprocal strikes, some evidence of strike severity escalation and more bystander intervention. Both types were equally injurious. Winner/loser outcomes were reliably identified by postfight behaviors and/or signs of injury. Winners had advantageous prefight resource holding potential (RHP: greater height and/or vigor) significantly more often than losers. Consistent with tendencies for fights to occur between animals of the same sex, there were more male/male and female/female fights and fewer male/female fights than expected from random pairings of men and women in the videos. Female/female fights involved proportionally more hair-pulling, extended bouts of rapidly repeated strikes and longest durations. Bystanders intervened in over half the videos, attempting to separate fighters or help losers more often than they attacked the loser. Carefully selected internet street fight videos can provide important information.

摘要

互联网上的街头斗殴视频可能会提供有关人类身体攻击鲜为人知方面的信息,但其可靠性尚不清楚。对100个二元斗殴视频进行的分析,这些视频涉及从对其他动物的研究中得出的行为学、博弈论和性别差异问题,结果发现,打斗前非攻击性或攻击性意图的言语表达或手势信号大致能预测谁会先动手以及谁会获胜。头部是首选的攻击目标。不同攻击行为的严重程度排序从吐口水的1级到窒息的5级不等。一半的视频显示出较短的、单方面的攻击,以一次或多次高严重程度的攻击开始,几乎没有升级的迹象,旁观者干预也较少。其中四分之一是偷袭。其他视频则显示出较长时间的相互攻击打斗,有一些攻击严重程度升级的迹象,旁观者干预也更多。两种类型的打斗造成的伤害程度相同。通过打斗后的行为和/或受伤迹象可以可靠地确定胜负结果。获胜者比失败者更常具有有利的打斗前资源持有潜力(RHP:更高的身高和/或更强的活力)。与动物同性之间发生打斗的倾向一致,视频中男性/男性和女性/女性之间的打斗比随机配对的男性和女性预期的更多,而男性/女性之间的打斗则更少。女性/女性之间的打斗中,抓头发、快速重复攻击的长时间回合以及持续时间最长的情况比例更高。旁观者在超过一半的视频中进行了干预,他们试图分开打斗者或帮助失败者的次数多于攻击失败者的次数。精心挑选的互联网街头斗殴视频可以提供重要信息。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8583/11664032/be6cf5ba56a4/AB-51-e70017-g001.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验