Data from: Behavioural mimicry among poison frogs diverges during close-range encounters with predators
Data files
Apr 11, 2025 version files 12.36 GB
-
ABI_VideoFiles.zip
4.23 GB
-
AHY_VideoFiles.zip
4.43 GB
-
AZA_VideoFiles.zip
3.69 GB
-
EscapeData2019.txt
22.47 KB
-
JumpAngleData2019.txt
269.22 KB
-
README.md
7.03 KB
Abstract
Aposematic species signal their unpalatability to potential predators with recognisable, and frequently conspicuous, colour patterns. These visual signals are often also associated with bold behaviour and a reduced propensity to escape from approaching predators. Bold behaviours may act as an aversive signal and allow defended prey to avoid the energetic/opportunity costs that arise from fleeing predators. For Batesian mimics, non-defended species which replicate the colours of defended models, behavioural mimicry may also improve mimic fidelity and reduce energetic/opportunity costs. However, as predators may test the honesty of aposematic signals through sampling behaviour, Batesian mimics can be at high risk during close-range interactions with predators. This raises the question of whether/when Batesian mimics should deviate from behavioural mimicry and initiate more extensive escape behaviour. Here, we exposed the chemically defended poison frog Ameerega bilinguis and its (non-toxic) Batesian mimic Allobates zaparo to a simulated predator encounter. We predicted Al. zaparo would escape to a greater distance and in a more erratic manner than Am. bilinguis. Yet, contrary to our predictions, Al. zaparo did not flee far from predators. It was, however, more likely to initiate escape prior to physical contact from the predator. We suggest that bold behaviour coupled with pre-emptive movement allows Al. zaparo to retain the benefits of behavioural mimicry while reducing the likelihood that predators will test signal honesty. Our data highlight, that when examining the evolution of mimicry, we must consider both morphological and behavioural traits, as well as how risk to the prey may change how they behave throughout the predation sequence.
https://doi.org/10.5061/dryad.bnzs7h4m8
Description of the data and file structure
Files and variables
File: EscapeData2019.txt
Description:
Data required for Analyses 1-4 & S1 (‘Total Jumps’, ‘Escape Distance’, ‘Escape Time’, & ‘Preemptive Jumps’)
See the main manuscript for further details.
Variables
- file: The video file from which data were extracted
- species: Species codes (ABI = Ameerega bilinguis, AZA = Allobates zaparo, AHY = Adenomera hylaedactyla)
- individual: Unique numeric code for each individual frog
- svl: Frog body length (snout-vent length) in mm
- run: The experimental session (1 to 4 for each frog)
- n_jumps: Total number of jumps performed during experimental run
- n_jumps_preemptive: Total number of ‘predator approaches’ that resulted in preemptive movement (out of 5)
- duration_s: Total time spent moving in seconds
- mean_abs_angle_change_rad: Mean change in angle between jumps in radians (absolute value)
- NA: where there is only 1 jump, there is no change in jump angle
- mean_distance_px: Mean distance travelled per jump in pixels
- jump_frequency_hz: The frequency of jumps in Hertz
- vidres_x_px: Resolution of the video file in pixels (X axis)
- vidres_y_px: Resolution of the video file in pixels (Y axis)
- vidscale_px_per_mm: Resolution of the video file scaled in pixels per mm
File: JumpAngleData2019.txt
Description:
Data required for Analyses 5-6 (‘Jump Angle’ & ‘Escape Trajectory’)
See the main manuscript for further details.
Variables
- file: The video file from which data were extracted
- species: Species codes (ABI = Ameerega bilinguis, AZA = Allobates zaparo, AHY = Adenomera hylaedactyla)
- individual: Unique numeric code for each individual frog
- run: The experimental session (1 to 4 for each frog)
- jump_n: The number of each jump in the jump sequence
- distance_px: Distance travelled during the jump in pixels
- duration_s: Time spend in motion during the jump in seconds
- angle_rad: Jump angle in radians relative to the starting point
- angle_change_rad: Jump angle in radians relative to the previous jump
- NA: the first jump in the sequence has no change in jump angle
- start_x_px: Frog location at start of the jump (X coordinate in pixels)
- start_y_px: Frog location at start of the jump (Y coordinate in pixels)
- end_x_px: Frog location at end of the jump (X coordinate in pixels)
- end_y_px: Frog location at end of the jump (Y coordinate in pixels)
- start_time_s: Time at the beginning of the jump (seconds from start of recording)
- end_time_s: Time at the end of the jump (seconds from start of recording)
File: ABI_VideoFiles.zip
Description:
Original video files of the behavioural trials with Ameerega bilinguis.
File names include the species (ABI = Am. bilinguis), individual species ID number, and round number (R1-4).
There are 80 videos corresponding to 4 rounds for each of the 20 Am. bilinguis.
File: AHY_VideoFiles.zip
Description:
Original video files of the behavioural trials with Adenomera hylaedactyla.
File names include the species (AHY = Ad. hylaedactyla), individual species ID number, and round number (R1-4).
There are 80 videos corresponding to 4 rounds for each of the 20 Ad. hylaedactyla.
File: AZA_VideoFiles.zip
Description:
Original video files of the behavioural trials with Allobates zaparo.
File names include the species (AZA = Al. zaparo), individual species ID number, and round number (R1-4).
There are 80 videos corresponding to 4 rounds for each of the 20 Al. zaparo.
Code/software
==== ESCAPE DATA - ANALYSES 1-4, S1 ====
Analyses of ‘Total Jumps’, ‘Escape Distance’, ‘Escape Time’, & ‘Preemptive Jumps’.
(See the main manuscript and Rmarkdown for further details)
Data file: EscapeData2019.txt
RMarkdown files:
Analysis 1 - Total number of jumps.Rmd | Analysis-1---Total-number-of-jumps.pdf
Analysis 2 - Total distance.Rmd | Analysis-2---Total-distance.pdf
Analysis 3 - Total time.Rmd | Analysis-3---Total-time.pdf
Analysis 4 - Preemptive jumps.Rmd | Analysis-4---Preemptive-jumps.pdf
Analysis S1 - BodyLength distance.Rmd | Analysis-S1---BodyLength-distance.pdf
A series of generalised linear mixed-effects models: Total Jumps (Poisson), Total Distance (log transformed, Gaussian), Escape Time (log transformed, Gaussian), and Preemptive Jumps (Binomial). Including the fixed effect of species and random effects of frog ID and run (session number).
Total Jumps, Escape Time used functions lmer or glmer from R package lme4 v.1.1.34 (Bates et al., 2015).
Escape Distance, Preemptive Jumps used function glmmTMB from R package glmmTMB v.1.1.8 (Brooks et al., 2017).
Model assumptions were checked using the R package DHARMa v.0.4.6 (Hartig, 2022).
Pairwise comparisons were conducted using the R package multcomp v.1.4.25 (Hothorn et al., 2008).
==== ESCAPE DATA - ANALYSES 5-6 ====
Analyses of ‘Jump Angle’ & ‘Escape Trajectory’
(See the main manuscript and Rmarkdown for further details)
Data file: JumpAngleData2019.txt
RMarkdown files:
Analysis 5 - Jump angle.Rmd | Analysis-5---Jump-angle.pdf
Analysis 6 - Escape trajectory.Rmd | Analysis-6---Escape-trajectory.pdf
Bayesian circular mixed effects models using function bpnme from R package bpnreg v.2.0.3 (Cremers, 2024).
Iterations = 5000. Burn in = 1000.
Jump Angle (the absolute angle of each jump between 0º and 180º) in radians.
Escape Trajectory (the overall escape direction between -180º and 180º) in radians.
Models included the fixed effect of species and the random intercept of frog ID.
References:
Bates, D., Mächler, M., Bolker, B. & Walker, S. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67: 1 - 48. https://doi.org/10.18637/jss.v067.i01
Brooks, M. E., Kristensen, K., Benthem, K. J. v., Magnusson, A., Berg, C. W., Nielsen, A., Skaug, H. J., Mächler, M. & Bolker, B. M. 2017. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. The R Journal 9: 378-400.
Cremers, J. (2024) bpnreg: Bayesian projected normal regression models for circular data. R package version 2.0.3. https://cran.r-project.org/package=bpnreg
Hartig, F. (2022) DHARMa: Residual diagnostics for hierarchical (multi-level / mixed) regression models. R package version 0.4.6. https://cran.r-project.org/package=DHARMa
Hothorn, T., Bretz, F. & Westfall, P. 2008. Simultaneous inference in general parametric models. Biometrical Journal 50: 346-363. https://doi.org/10.1002/bimj.200810425