It’s well known that pigs are highly expressive animals with a sophisticated range of vocal expressions – they can even produce a wider range of sounds more frequently than cows, goats, and sheep.
In a bid to decode the emotional state of pigs, a first-of-its-kind study set out to understand the meaning of the broad range of pig sounds, with the help of artificial intelligence (AI).
The study aimed to determine positive and negative emotions
In the study, the researchers designed an algorithm that can determine whether an individual pig is experiencing a positive emotion (when they feel ‘happy’ or ‘excited’), a negative emotion (when they feel ‘stressed’ or ‘scared’), or somewhere in between.
The study was recently published in an open-access journal, Scientific Reports, and was led by the University of Copenhagen, the ETH Zurich, and France’s National Research Institute for Agriculture, Food and Environment (INRAE).
An international team of researchers across Europe gathered more than 7,000 audio recordings of pig calls from 411 different pigs.
The audio recordings were collected throughout the lives of commercial pigs – right from their birth and up until they were slaughtered.
The study authors were able to capture the sounds made by pigs of different ages in a wide range of situations, eliciting positive and negative emotions.
Using an AI-driven algorithm, the researchers analyzed the audio sounds to differentiate and classify the oinks, grunts, and squeals into positive and negative emotional states.
The study found that there were clear differences in pig calls depending on positive and negative situations.
The findings generally validated prior observations that have linked high-frequency calls with negative emotional states and low-frequency sounds with positive emotional states.
Happy pigs had shorter grunts
In positive situations, where the pigs were reunited with their family after being separated or suckling their mothers, the pig calls elicited distinct sounds.
Pigs that expressed positive emotions produced far shorter calls, with minor fluctuations in amplitude. The grunts were short, beginning high and gradually reducing in frequency.
In contrast, emotionally negative situations included, among others, separation, starvation, castration, and handling and waiting in the slaughterhouse to be killed.
During these distressing situations, the pigs generally produced cries of agony in the form of screams, squeals, and barks that were longer in length. The researchers noticed that the cries were extra variable in frequency than the sounds of happy pigs.
The algorithm successfully decodes and understands the emotions of pigs
“With this study, we demonstrate that animal sounds provide great insight into their emotions. We also prove that an algorithm can be used to decode and understand the emotions of pigs, which is an important step towards improved animal welfare for livestock,” said Associate Professor Elodie Briefer, the study’s lead author, in a statement.
According to Briefer, the researchers’ algorithm correctly identified the animal’s emotion as positive or negative 92 percent of the time.
This groundbreaking research highlights the fact that animals, such as pigs, are indeed capable of experiencing emotions in a way that’s somewhat similar to our experience.
The researchers believe the AI-driven algorithm could be used to understand the emotions and even monitor the psychological well-being of other animals too.