Decoding pig emotions with a machine learning algorithm

We can now understand the feelings of pigs. An international team of researchers is the first in the world to translate pig grunts into actual emotions across an extended number of conditions and life stages. The research can be used to improve animal welfare in the future.

Is a pig grunt worth a lot of words? Maybe so. A group of researchers from around the world have translated pig grunts into feelings. Scientific Reports published the findings today.

The researchers used more than 7,000 audio recordings of pigs to create an algorithm that can tell if a pig is happy, excited, or scared. The recordings were collected in a wide range of situations encountered by commercial pigs, both positive and negative, from when they are born until their deaths.

We show that animal sounds give great insight into their emotions. Associate Professor Elodie Briefer of the University of Copenhagen's Department of Biology at the University of Copenhagen says that they have proved 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.

Short grunts are happy.

The researchers recorded pig sounds in both commercial and experimental scenarios, which are either associated with a positive or negative emotion, based on the behavior of the pigs. Sometimes positive situations include when piglets suckle from their mothers or when they are united with their family after being separated. Fights between piglets, castration and slaughter are some of the emotionally negative situations.

The researchers created mock scenarios for the pigs in order to evoke more nuanced emotions in the middle of the spectrum. There was an arena with toys or food and a corresponding arena without any stimuli. New and unfamiliar objects were placed in the arena for the pigs to interact with. When possible, the pigs calls, behavior and heartrates were recorded.

The researchers analyzed the more than 7,000 audio recordings to see if there was a pattern in the sounds as a function of the emotions, and if they could discern the positives and negatives from the negatives. The researchers collected more screams and shrieks in negative situations. Both barks and grunts were made in situations where the pigs experienced positive or negative emotions.

There were some interesting situations between the extremes. The researchers found a new pattern in the sound files that showed what the pigs experienced in certain situations.

There are clear differences in pig calls when looking at positive and negative situations. The calls are shorter in the positive situations. The grunts begin high and go lower in frequencies. We can classify almost all of the calls to the correct emotion by training an algorithm to recognize these sounds.

Farmers can keep an eye on animal emotions.

Over the last 20 years, the study of animal emotions has come a long way. The mental health of livestock is important for their well-being. Today's animal welfare focuses on the physical health of animals. Several systems exist that can be used to monitor an animal's physical health.

We have trained the program to decode pig grunts. We need someone who wants to develop an app that farmers can use to improve the welfare of their animals.

She says that the method could be used to better understand the emotions of other mammals.

There are facts.

• The researchers recorded 7,414 sounds from 411 pigs in different scenarios, from birth to death.
• A machine learning algorithm was trained to decode whether pig calls can be classified as a function of positive or negative emotions.
• The researchers defined the emotions of pigs based on how they naturally react to various positive and negative external stimuli, and whether stimuli can improve (positive) or threaten (negative) their lives.
• For example, typical signs of negative emotions in pigs are that they stand still, emit a lot of vocalizations, and try to escape, while positive ones include exploring their surroundings and having their ears postured forward.
• Positive situations included huddling with littermates, nursing, positive conditioning, enrichment, reunion with the mother, and freely running. Negative situations included, among others, missed nursing, short social isolation, piglet fights, piglet crushing by the mother, castration, and handling and waiting in the slaughterhouse.
• Sixteen researchers from Denmark, Switzerland, France, Germany, Norway and the Czech Republic participated in the study.

More information: Elodie F. Briefer et al, Classification of pig calls produced from birth to slaughter according to their emotional valence and context of production, Scientific Reports (2022). DOI: 10.1038/s41598-022-07174-8

Encoding of Situations in the Vocal Repertoire of Piglets is a study by C.line Tallet and colleagues. The journal was published in October.

Journal information: PLoS ONE , Scientific Reports Citation: Decoding pig emotions with a machine learning algorithm (2022, March 7) retrieved 7 March 2022 from https://phys.org/news/2022-03-decoding-pig-emotions-machine-algorithm.html This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.