Read on Twitter
It’s Sunday so let’s talk about patterns. Patterns in behaviour performed at the same time, over time,or between individuals, give us more information to help infer animal welfare. We need methods to help us detect patterns. A thread with some ole papers https://tinyurl.com/y4erayyv
For behaviour performed at the same time we can use (machine learning) algorithms developed to predict which your shopping habits. We apply this to behaviour to see what coordinated patterns there are in behaviour. E.g. in cows collab w @Sara69449861 https://tinyurl.com/y4uxlys4
To help understand individual patterns in behaviour over time we can use maths like Markov chains to understand repetitive sequences. These are sequences in which the probability of each event depends on the previous event. E.g. https://tinyurl.com/y3alnk65
Or we can compare patterns between individuals to see how unique each animal is. We’ve done this in chickens in work by @runzipunzi @MikeToscano2 https://tinyurl.com/yy99smjw
We can also look at patterns in interactions between animals. Mostly this is achieved using social network analysis. Interactions with others of the same species is an important part of welfare. E,g do hens have friends. https://tinyurl.com/y346vk8g
Lots of different ways of finding patterns in animal behaviour and relating these to welfare. Detecting patterns is big business in other areas of science and commence. As new methods come out we’ll always be looking at these for inspiration.
