Skip to main content
. 2015 Sep 22;5:14275. doi: 10.1038/srep14275

Table 1. Results of GLMM for affecting the probability of renewed aggression.

  Renewed aggression initiated by winners n = 186
Renewed aggression initiated by losers n = 175
β (SE) P β (SE) P
Presence of post-conflict affiliation
 Unaffiliated PC vs. PCAF –1.79 (0.57) 0.002 –3.89 (1.26) 0.002
 Unaffiliated PC vs. PCAWB by bystanders –1.77 (0.59) 0.003 –1.36 (0.63) 0.03
 Unaffiliated PC vs. PCALB by losers –1.64 (0.56) 0.003 –1.78 (0.71) 0.01
 Unaffiliated PC vs. PCALB by bystanders –2.49 (0.65)  < 0.001 –2.44 (0.85) 0.004
Duration of aggression (seconds) 0.002 (0.003) 0.40 0.005 (0.004) 0.18
Direction of aggression (unidirectional vs. bidirectional) –0.25 (0.40) 0.53 –1.10 (0.52) 0.04
New aggression –0.16 (0.43) 0.72 –0.80 (0.63) 0.21

GLMM = generalized linear mixed model, PCAF = post-conflict affiliation between former opponents, PCAWB by bystanders = post-conflict affiliation initiated by bystanders to winners, PCALB by losers = post-conflict affiliation initiated by losers to bystanders, PCALB by bystanders = post–conflict affiliation initiated by bystanders to losers, unaffiliated PC = PC in which any post-conflict affiliation did not occur, new aggression = aggression between previous former opponents and an uninvolved individual.