Information maintenance of food sources is associated with environment, spatial cognition and age in a food-caching bird
Description
These data were analyzed to explore to what extent information updating behaviors (i.e. sampling) may depend on environmental predictability (see Benedict et al. in Animal Behaviour). We tested this hypothesis by allowing wild food-caching mountain chickadees, Poecile gambeli, to visit and sample multiple feeders with temporarily stable, unlimited food for five consecutive winters. We used an index of feeder use breadth to explore how feeder visits across multiple feeders varied with environmental conditions. Each feeder visit is associated with information updating, and more information maintenance should be associated with distributing more visits across more feeders. While controlling for the total number of visits by each individual, we found that (1) chickadees redistributed feeder use among more feeders when environmental conditions were harsh and unpredictable, (2) juveniles had a higher feeder use breadth than adults, and individuals reduced their feeder use breadth as they aged, (3) better spatial learning and memory ability but not spatial cognitive flexibility was associated with smaller feeder use breadth and (4) learning associated with decreased food availability reduced subsequent feeder use breadth. Our data supported our predictions that factors affecting the predictability of resource information (environmental conditions and individual characteristics such as cognition and age) affect how individuals engage in information maintenance. The dataset is comprised of several types of data collected from 2015-2020. Behavioral data were collected in Sagehen Experimental Forest using radiofrequency identification (RFID)-equipped 'smart' feeders arranged in spatial arrays to allow for spatial learning and memory testing (Croston et al. 2015). These feeders allowed for detection of passive integrated transponder (PIT)-tagged chickadees foraging at the feeders. These behavioral data were collected during annual winter data collection periods and were summarized by the Levins' measure (i.e. the inverse of the Simpson's diversity index; Levins 1968). The Levins' measure uses proportion of visits to 8 feeders in a given array to estimate feeder use breadth and provide a metric of information maintenance. The dataset also includes (1) demographic data (age, sex) from long-term banding efforts in the study system, (2) cognitive data (spatial learning and memory ability, single reversal learning ability), and environmental data (from nearby California SNOTEL stations #539, #540, and #541). Feeder design by Eli S. Bridge & Vladimir V. Pravosudov: Bridge, E.S., & Bonter, D.N. (2011). A low-cost radio frequency identification device for ornithological research. Journal of Field Ornithology, 82, 52–59. https://doi.org/10.1111/j.1557-9263.2010.00307.x
Files
Steps to reproduce
Run script "Benedict_infoUpdating_ms_analyses.R" using R or RStudio. The R script contains all packages and all references to data needed for analyses and plots to reproduce the manuscript analysis. README contains detailed column descriptions for all data documents. All data documents should be in the same directory or the code adjusted accordingly.