Wolves (Canis lupus) are frequently used as a model for describing the social behaviour of group-living animals. Communication within wolf packs reduces aggression and maintains pack cohesion and stability.
Wolves are considered to produce facial expressions which they use to convey ‘emotional’ states. However, to date, there is no quantitative research linking specific facial expressions to specific emotional states in wolves.

Using the Dog Facial Action Coding System (DogFACS) and Linear Discriminant Analyses (LDA), for the first time, this research quantifies and classifies nine emotional facial expressions of wolves; anger, anxiety, curiosity, fear, friendliness, happiness, joy, interest, and surprise.
This research demonstrates that wolves are more communicatively complex than previously considered and gives insights into how sentient wolves are.

In addition, the methods used in this research have the potential to provide a non-invasive tool to assess emotional states across mammalian species, which could prove beneficial for animal welfare purposes.
Moreover, this research provides an opportunity for humans to better understand wolves; it is hoped that increased knowledge of true wolf behaviour will allow humans to view wolves in a more positive light and thus, aid recovery and conservation programs of wild wolves.