R-code and data from: Bi-modal alarm signals modulate responses to mono-modal alarm signals in Camponotus modoc carpenter ants
Description
This data set is a supplement to an article titled: “Bi-modal alarm signals modulate responses to mono-modal alarm signals in Camponotus modoc carpenter ants”. We investigated the interaction between two alarm modalities, pheromone and vibratory, in the carpenter ant, C. modoc. Our full description of the methodology, results, and their interpretation can be found in the publication. Abstract from article: "Distressed western carpenter ants, Camponotus modoc., produce alarm pheromone and substrate-borne vibrations. The alarm pheromone attracts nestmates but the effects of vibratory signals, or of bi-modal pheromonal and vibratory signals, are not known. Worker ants of two Camponotus congeners reportedly stand still (‘freeze’) or run fast in response to engineered drumming vibrations inputted on plastic, but many responses to ant-produced vibratory signals on wood have not yet been investigated. Generally, orientating towards signallers under vertebrate predator attack seems maladaptive and not beneficial to ant colonies. We tested the hypotheses (1) that vibratory alarm signals cause freezing, rapid running but not attraction of nestmates, and (2) that bi-modal alarm signals modulate responses to mono-modal alarm signals, thereby possibly reducing predation risk. Laser Doppler vibrometry recordings revealed that the ants’ vibratory signals readily propagate through ant nest lamellae, and thus quickly inform nest mates of perceived threats. With a speaker modified to record and deliver vibratory signals, we obtained drumming signals of distressed ants on a Douglas fir veneer, and bioassayed signal effects on ants in an arena with a suspended veneer floor. In response playback of vibratory signals, ants ran rapidly, or froze, but did not approach the vibratory signals. Exposed to alarm pheromone, ants frequently visited the pheromone source. However, concurrently exposed to both alarm pheromone and vibratory signals, ants visited the pheromone source less often but spent more time ‘frozen’. The ants’ modulated responses to bi-modal signals seem adaptative but the reproductive fitness benefits are still to be quantified" We have uploaded our data and scripts as an R studio project. Code used to wrangle data, analyze data and generate plots can be accessed in the project folder by opening the project file. The project contains: ---Data Data files of laboratory experiments to assess ant alarm behaviours. ---Outputs Plots and csv files generated from data analysis. ---Scripts Scripts of R code used to wrangle data, conduct analyses, and generate outputs.
Files
Steps to reproduce
Full methodology can be found in the associated article. Here, we describe steps to reproduce data analyses. For freezing behaviour ("mean.nm.total") we counted the total number of seconds each ant spent in freezing behaviour and then took the mean total for each replicate. We compared the overall mean response before ("pre") and after ("post") experiencing a given stimulus (e.g. vibratory) or across stimuli using generalized linear mixed models (GLMMs). For running speed ("mean.speed"), we quantified the mean running speed for the longest continuous run each ant performed pre and post stimulus using FIJI. For each replicate we then took the mean running speed and compared the mean running speed pre and post stimulus and across stimuli using GLMMs. For attraction we counted the number of visits to either the non-stimulus ("c") or stimulus speaker ("t") in total for each replicate. We then compared the total number of visits to either speaker for each replicate pre and post stimuli and across stimuli using GLMMs.