An international team of researchers has reported the development of an algorithm that can translate the emotional state of pigs from the sound of their grunts. The researchers indicate the system could be used to monitor the well-being of pigs on a farm in real-time.
Domestic pigs display highly sophisticated varieties of vocal expression. Previous studies have established correlations between high frequency calls, such as squeals and screams, being associated with negative emotions, and low-frequency grunts being associated with positive or neutral emotions. But in between those two extremes are an assortment of sounds that are less understood.
The new research first set out to understand the broad array of pig vocalizations. To do this the researchers cataloged 7,414 different pig sounds, gathered from 411 animals.
Each pig sound was accompanied by close behavioral observations and, where possible, heart rate monitoring to establish positive or negative emotional associations. Positive scenarios were studied, such as piglets suckling or playing with toys, and emotionally negative scenarios were also tracked, including fights, separation from family and slaughter.
In general the findings validated prior observations linking high frequency calls with negative emotional states and low frequency sounds with positive emotional states. However, the researchers did discover a significant volume of calls that didn’t correlate with that simple distinction.
Two particular acoustic characteristics were found to be as important as frequency in understanding emotional valence: duration and amplitude modulation rate. For example, a high frequency squeal was determined to represent a positive emotion when it was short and contained few amplitude modulations.
"There are clear differences in pig calls when we look at positive and negative situations,” explained Elodie Briefer, an author on the study from the University of Copenhagen. “In the positive situations, the calls are far shorter, with minor fluctuations in amplitude. Grunts, more specifically, begin high and gradually go lower in frequency.”
The researchers then used a neural network to develop an algorithm that can translate the emotional characteristic of the pig sounds. In this proof-of-concept study the researchers claim the initial iteration of the algorithm can correctly translate pig emotions from their calls with 92 percent accuracy.
The long-term goal, according to the researchers, would be to develop some kind of app that can monitor the emotional well-being of commercial pigs in real-time. Briefer also hypothesizes their analytical method is transferable to other kinds of mammals, suggesting the possibility of a kind of universal translator that could track animal emotions by the sounds they make.
"We have trained the algorithm to decode pig grunts,” said Briefer. “Now, we need someone who wants to develop the algorithm into an app that farmers can use to improve the welfare of their animals.”
The new study was published in the journal Scientific Reports.
Source: University of Copenhagen
