## installing and/or loading from CRAN:
## installing and/or loading from CRAN:Data description
## [1] "Number of calls:"## [1] 6645## [1] "Experiment:"## [1] "FemGroup"## [1] "Number of calls per treatment:"| Novel | Stable |
|---|---|
| 2363 | 4282 |
## [1] "Number of calls per cage:"| 1 | 10 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|---|
| 1069 | 879 | 893 | 905 | 25 | 1119 | 20 | 398 | 546 | 791 |
## [1] "Number of calls per focal group:"| Focal | Group |
|---|---|
| 1563 | 5082 |
Correlation matrix
This graph shows the correlation between the parameters used to define the acoustic space. Parameters in which the absolute correlation was higher than 0.95 were removed. Note that random forest is very robust to colinearity. 
Confusion matrix
This is the accuracy of the random forest model at correctly clasifying call types on pre-defined categories based on spectrograph similarity for human observers.
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.794883 0.727975 0.784971 0.804536 0.373965
## AccuracyPValue McnemarPValue
## 0.000000 NaN
Variable importance
The relative contribution of each acoustic parameter to the random forest predictive model.Long story short: cross-correlation dereived parameters are the most important ones.
Random Forest acoustic space (based on bootstrap MDS)
Acoustic space defined by the 2 dimensions of a boostrap multidimensional scale analysis on the random forest proximity (see here for details).
Change in acoustic space area through time
These graphs show, for each individual, the cumulative (pink line) and relative (blue line) change in the area of its acoustic space (convex hull) as new calls are included. Relative change was defined as the change in acoustic space for the last 10 calls at any time. The calls were added in chronological order (as their were produced) and the period in which they occured is shown by the background color. If social group stability plays a role in acoustic “drift”, then a bump in the acoustic space would be expected at the start of the “post” period in birds in novel social groups.
Acoustic space plots with spectros
These plots show the acoustic space covered by each individual in 2 ways: - Using dots to show the position of each of its calls in the acoustic space - Using kernels to display the density of calls in acoustic space regions
Both plot types are shown consecutively for each individual for comparison.
Stats
GLMM model:
\[\ density.post.overlap \sim Treatment + n + Focal.or.group + (1 | Cage)\]
## [1] 34## Warning: 'rBind' is deprecated.
## Since R version 3.2.0, base's rbind() should work fine with S4 objects## Fixed term is "(Intercept)"## [1] "Best models"## [1] "P values"## Linear mixed model fit by maximum likelihood . t-tests use
## Satterthwaite's method [lmerModLmerTest]
## Formula: dens.ovlp ~ Treatment + n + Focal.Group + (1 | Cage)
## Data: dens.ovlp
##
## AIC BIC logLik deviance df.resid
## -84.3 -75.1 48.1 -96.3 28
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.793 -0.709 -0.255 0.302 2.130
##
## Random effects:
## Groups Name Variance Std.Dev.
## Cage (Intercept) 0.00000 0.0000
## Residual 0.00345 0.0587
## Number of obs: 34, groups: Cage, 9
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.197777 0.042074 34.000000 -4.70 4.2e-05 ***
## TreatmentStable 0.069407 0.022064 34.000000 3.15 0.0034 **
## n 0.001629 0.000937 34.000000 1.74 0.0912 .
## Focal.GroupGroup -0.047594 0.030759 34.000000 -1.55 0.1310
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) TrtmnS n
## TretmntStbl -0.427
## n -0.584 -0.147
## Focl.GrpGrp -0.808 0.314 0.167## [1] "Effect size and confidence intervals"
## [1] "Boxplot"
## [1] "test assumptions"
##
## # Overdispersion test
##
## dispersion ratio = 0.0042
## Pearson's Chi-Squared = 0.1173
## p-value = 1.0000## No overdispersion detected.Session information
## R version 3.4.4 (2018-03-15)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 16.04.4 LTS
##
## Matrix products: default
## BLAS: /usr/lib/openblas-base/libblas.so.3
## LAPACK: /usr/lib/libopenblasp-r0.2.18.so
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] compiler parallel stats graphics grDevices utils datasets
## [8] methods base
##
## other attached packages:
## [1] lmerTest_3.0-1 sjstats_0.14.3 MuMIn_1.40.4
## [4] cowplot_0.9.2 raster_2.6-7 spatstat_1.55-1
## [7] rpart_4.1-13 spatstat.data_1.2-0 adehabitatHR_0.4.15
## [10] adehabitatLT_0.3.23 CircStats_0.2-4 boot_1.3-20
## [13] MASS_7.3-50 adehabitatMA_0.3.12 deldir_0.1-14
## [16] kableExtra_0.9.0 corrplot_0.84 randomForest_4.6-14
## [19] caret_6.0-78 pbapply_1.3-4 smacof_1.9-6
## [22] plotrix_3.7 Hmisc_4.1-1 Formula_1.2-2
## [25] survival_2.42-3 lme4_1.1-15 Matrix_1.2-14
## [28] ecodist_2.0.1 nlme_3.1-137 AICcmodavg_2.1-1
## [31] vegan_2.5-2 lattice_0.20-35 permute_0.9-4
## [34] fossil_0.3.7 shapefiles_0.7 foreign_0.8-70
## [37] sp_1.2-7 gridExtra_2.3 ade4_1.7-10
## [40] ggplot2_2.2.1 RColorBrewer_1.1-2 readxl_1.0.0
## [43] warbleR_1.1.15 NatureSounds_1.0.0 seewave_2.1.0
## [46] tuneR_1.3.2 maps_3.3.0
##
## loaded via a namespace (and not attached):
## [1] estimability_1.3 SparseM_1.77 ModelMetrics_1.1.0
## [4] coda_0.19-1 tidyr_0.7.2 acepack_1.4.1
## [7] knitr_1.20 multcomp_1.4-8 data.table_1.10.4-3
## [10] RCurl_1.95-4.10 TH.data_1.0-8 mice_3.1.0
## [13] VGAM_1.0-4 proxy_0.4-22 xml2_1.2.0
## [16] lubridate_1.7.1 assertthat_0.2.0 gower_0.1.2
## [19] hms_0.4.0 bayesplot_1.5.0 evaluate_0.10.1
## [22] DEoptimR_1.0-8 htmlwidgets_0.9 reshape_0.8.7
## [25] stringdist_0.9.4.6 ddalpha_1.3.1 stats4_3.4.4
## [28] purrr_0.2.4 heplots_1.3-4 dplyr_0.7.5
## [31] backports_1.1.2 signal_0.7-6 pwr_1.2-2
## [34] quantreg_5.34 sjlabelled_1.0.10 abind_1.4-5
## [37] withr_2.1.1 sfsmisc_1.1-1 robustbase_0.92-8
## [40] checkmate_1.8.5 emmeans_1.1.3 weights_0.85
## [43] goftest_1.1-1 mnormt_1.5-5 cluster_2.0.7-1
## [46] lazyeval_0.2.1 crayon_1.3.4 candisc_0.8-0
## [49] ellipse_0.4.1 labeling_0.3 recipes_0.1.2
## [52] pkgconfig_2.0.1 slam_0.1-42 wordcloud_2.5
## [55] nnet_7.3-12 bindr_0.1.1 rlang_0.2.0
## [58] mitml_0.3-5 MatrixModels_0.4-1 sandwich_2.4-0
## [61] modelr_0.1.1 cellranger_1.1.0 rprojroot_1.3-2
## [64] polyclip_1.8-7 lmtest_0.9-36 zoo_1.8-1
## [67] pan_1.6 base64enc_0.1-3 ggridges_0.5.0
## [70] viridisLite_0.3.0 rjson_0.2.20 bitops_1.0-6
## [73] rgl_0.95.1441 DRR_0.0.3 stringr_1.3.1
## [76] coin_1.2-2 readr_1.1.1 jpeg_0.1-8
## [79] scales_0.5.0 magrittr_1.5 plyr_1.8.4
## [82] gdata_2.18.0 dimRed_0.1.0 snakecase_0.9.1
## [85] cli_1.0.0 dtw_1.20-1 Sim.DiffProc_4.0
## [88] TMB_1.7.13 htmlTable_1.11.2 mgcv_1.8-24
## [91] tidyselect_0.2.4 fftw_1.0-4 stringi_1.2.2
## [94] forcats_0.3.0 highr_0.6 yaml_2.1.19
## [97] latticeExtra_0.6-28 grid_3.4.4 polynom_1.3-9
## [100] tools_3.4.4 rstudioapi_0.7 foreach_1.4.4
## [103] prodlim_1.6.1 scatterplot3d_0.3-40 digest_0.6.13
## [106] pracma_2.1.4 lava_1.6 bindrcpp_0.2.2
## [109] Rcpp_0.12.17 car_2.1-6 broom_0.4.3
## [112] httr_1.3.1 psych_1.7.8 kernlab_0.9-25
## [115] Deriv_3.8.4 colorspace_1.3-2 rvest_0.3.2
## [118] tensor_1.5 CVST_0.2-1 splines_3.4.4
## [121] RcppRoll_0.2.2 spatstat.utils_1.8-0 xtable_1.8-2
## [124] nloptr_1.0.4 timeDate_3042.101 modeltools_0.2-21
## [127] ipred_0.9-6 R6_2.2.2 pillar_1.2.3
## [130] htmltools_0.3.6 prediction_0.3.2 nnls_1.4
## [133] glue_1.2.0 minqa_1.2.4 class_7.3-14
## [136] codetools_0.2-15 jomo_2.6-2 mvtnorm_1.0-6
## [139] tibble_1.4.2 numDeriv_2016.8-1 pbkrtest_0.4-7
## [142] unmarked_0.12-2 gtools_3.5.0 glmmTMB_0.2.0
## [145] rmarkdown_1.9 munsell_0.4.3 e1071_1.6-8
## [148] iterators_1.0.9 sjmisc_2.7.1 haven_1.1.1
## [151] reshape2_1.4.3 gtable_0.2.0