Derek Murphy

Interested in acoustic ID in animals? 🐾🎤

We put 16 mammal datasets through a vocal showdown—MFCCs and Random Forests came out on top 🏆

Want cleaner, comparable caller ID across species?
We've got guidelines and ready-to-use functions for that🥳
#Bioacoustics #WildlifeTech #MachineLearning

Super interesting PhD position working with a great team @primatenzentrum.bsky.social And lemurs!

Computer vision for primate behavior analysis in the wild - Nature Methods In this Perspective, Vogg et al. discuss the progress and challenges of using computer vision approaches in the behavioral analysis of primates in their natural environment.

SFB Paper Alert!

That’s teamwork: 14 SFB members across nine projects put together a perspective paper on “Computer vision for primate behavior analysis in the wild”, which was published in @naturemethods.bsky.social

www.nature.com/articles/s41...

Computer vision for primate behavior analysis in the wild - Nature Methods In this Perspective, Vogg et al. discuss the progress and challenges of using computer vision approaches in the behavioral analysis of primates in their natural environment.

We highlight the latest advancements in multi-animal detection, tracking, identification and (inter)action understanding & identify challenges to be addressed #animalbehaviour #animalbehavior #computervision #machinelearning #wildlifeconservation
read more here: www.nature.com/articles/s41...
2/2

A figure from the paper showing, on the left, three frames from a video in which two lemurs move through an area of forest; in the middle, the same three frames with bounding boxes surrounding each lemur; and on the right, a plot of the model output, showing the location and behaviour of each lemur over time

It's a 2-paper kinda week 😎
Exciting news for animal behaviour researchers! Our new paper is out now in @naturemethods.bsky.social, exploring the intersection of computer vision and animal behaviour. Congrats to lead author @richardvogg.bsky.social and everyone who worked so hard on this one
🧵1/2

Thanks to the staff at the Forêt des Singes & everyone who contributed to this project! We hope that our findings will contribute to the development of more efficient and effective methods for studying animal social networks. Read the full paper to learn more! 9/9 authors.elsevier.com/a/1kv83mjMA3iZ

Our study highlights the importance of validating new methods against traditional approaches. By doing so, we can ensure that our research is robust and reliable, and that we're using the best tools available to understand and protect animal populations. #researchintegrity #researchmethods 8/9

So, what does this mean for researchers and conservationists? While single-file movement data might not be suitable for fine-grained social network analysis, it could be a useful "quick and dirty" method for pilot studies or monitoring population-level changes. #conservationtools 7/9

Plots showing the mean and range of values derived from increasing amounts of data from single-file movements, scan samples and focal observations. The values from each method converge to similar estimates of the number of distinct communities in the group

However, when we looked at the broader community structure, we found that the single-file movement data and traditional methods converged on similar estimates. This suggests that single-file movement data might be useful for coarse estimation of group-level structure #communitydetection 6/9

Plots showing the weak positive correlation between social network metrics derived from scan samples and single-file movement data, and between metrics derived from focal observations and single-file movement data.

The results were... mixed. While we found some similarities between the single-file movement data and traditional methods, the associations were weak and not always reliable 5/9

We collected data on the order of individuals in single-file movements and used it to estimate the social network of our macaque group. Then, we compared the results to more traditional methods: scan sampling and focal observations 4/9

a man and a woman are running through a tall grass field . Alt: a man and a woman are running through a tall grass field .

Traditional methods for collecting animal social network data can be challenging, taking a lot of time and effort. We wanted to see if observing single-file movements could provide a quicker, easier alternative. #wildliferesearch #conservation 3/9

a group of ducks walking across a crosswalk with a woman walking behind them . Alt: a group of ducklings walking in single-file across a crosswalk, following their mother.

We explored a recently suggested, less labour-intensive method for studying animal social networks: analysing single-file movements. But does it work? We put it to the test with a group of Barbary macaques. #animalbehaviour #primatebehaviour #socialnetworks 2/9

A Barbary macaque walks on all fours through an outdoor wire mesh tunnel towards the camera

👋🦋
First post AND first new paper on the butterfly app! authors.elsevier.com/a/1kv83mjMA3iZ

'Estimating social network metrics from single-file movements in Barbary macaques, Macaca sylvanus'
with @julxf.bsky.social @primatenzentrum.bsky.social out now in #AnimalBehaviourJournal
Read🧵 for more! 1/9