An R package for modelling the visual aspects of predator-prey interactions.

Visit this site for vignettes detaling the package: https://nicholascarey.github.io/attackR/index.html

### Summary

This package can be used to model how much of a prey’s visual field is taken up by an attacking predator, given its size, speed, and shape. It was used in the following publication in PNAS to determine how much of a prey’s visual field is occupied by an attacking humpback whale.

It can be used for virtually any predator-prey combination, but there is also specific functionality to determine how apparent size is affected by the timing of a whale’s mouth opening during a lunge, though this can be ignored for other predators.

David E. Cade, Nicholas Carey, Paolo Domenici, Jean Potvin, and Jeremy A. Goldbogen. 2020. Predator-informed looming stimulus experiments reveal how large filter feeding whales capture highly maneuverable forage fish. Proceedings of the National Academy of Sciences, 117 (1) 472-478. https://doi.org/10.1073/pnas.1911099116

While the code was applied specifically to humpback whales for this study, given some simple morphological parameters it can be used to model how any attacking predator appears to a prey. It can also be applied to other filter-feeding whales which engulf large volumes of water, such as blue whales, given some inputs regarding mouth morphology and lunge timings. The package takes full account of the predator’s three-dimensional shape when determining its perceived size in the prey’s visual field. This is because at close distances, the maximum width of the predator will not necessarily make up the widest apparent visual angle, and more anterior parts of the body will appear to the prey to be wider, and have a larger apparent size.

From the prey’s perspective the package calculates the visual angle of the attacker (α, in radians), and the rate of change of this angle (dα/dt, in radians/s), as well as distance and time to capture.

### Vignettes

There are two vignettes (see Articles above) describing the functionality of the package and how it was used in the above study:

Overview of attackR

attackR and whale attacks

See also the Reference page to view the detailed Help file for each function.

### Installation

attackR is not yet published on CRAN, but can be installed using the devtools package:

install.packages("devtools")
devtools::install_github("nicholascarey/attackR")

### Feedback

If you encounter any bugs or issues, or have feedback, send an email, or open an issue.