Prediction of aggressive tendencies from facial dimension and ratios: A Study on undergraduate students of Northeastern Nigeria

Abstract

Facial dimensions and ratios are important anthropometric parameters which have been used over the years to define and predict some physical and social characteristics notably among human behaviour like aggression. However, there is a paucity of data regarding this link among the African population, particularly in Northeastern Nigeria where high cases of insurgency, banditry and some aggressive related criminalities are reported daily. The study aimed to determine how linear facial dimensions or ratios can specifically predict a self-reported form of aggression (physical, verbal, anger and hostility) among undergraduate students. A cross-sectional study design was adopted whereby a total of 400 (200 male and 200 female) adult undergraduate students aged between 16 to 30 years were randomly selected. A 2D facial images were used to determine facial biometrics (four linear facial dimensions ((n-sn, sn-gn, go-go and zy-gy)  and five facial ratios {upper facial weight/lower facial height (UFW/LFH), upper facial weight/lower facial weight (UFW/LFW), upper facial weight/upper facial height (UFWUFH) and upper facial height/facial height (UFH/FH) and facial weight to height ratio (fWHR) - lower derived from the measured linear dimensions using a Sony digital camera DSC w380 and art face 3 software. Buss and Perry's aggression questionnaire was adopted, and scores for each aggression scale were recorded. Data analysis was done using IBM SPSS software version 22 and Cronbach’s alpha for each scale above 0.70. Pearson’s correlation analysis was used to determine the relationship of facial biometrics with aggressive tendencies. The study observed that lower facial height (sn-gn) significantly correlates with verbal aggression and anger irrespective of sex. Facial height (FH) also correlates with verbal aggression in both sexes, while fWHR correlates significantly with anger (AN) irrespective of sex. More so, lower facial height to facial height ratio (LFH/FH) correlates significantly with verbal aggression (VA) and anger (AN) irrespective of sex but correlates with the same in males only. Similarly, upper facial weight to lower facial height (UFW/LFH) correlates significantly with verbal aggression and anger irrespective of sex and with anger only in males. Stepwise multiple linear regression equations were formed for the parameters which showed a positive correlation with aggression and consequently UFW/LFH and UFW/LFW were the best predictors of aggressive tendencies (P = 0.01). Conclusively, facial characteristics are indicators of one’s mood and can weakly predict aggressive tendencies in an individual and may be useful to predict suitable partners in marriage, friendship, or recruitment exercise.