Figures and data
Variation in the mineral and physical properties of quartz particles on food surfaces.
(A) Particle sizes followed a bimodal distribution, with most particles featuring metal inclusions. Nearly half the sample is < 25 μm, the “grittiness threshold” of the human oral cavity (Imai et al., 1995). (B) Circularity is a dimensionless shape factor (range: 0 to 1) based on two-dimensional microscopy and estimates for the projected area and perimeter of a given particle. Some examples are illustrated; overall, it is a convenient but imperfect proxy for sphericity, or deviations from spherical (Grace and Ebneyamini, 2021). Here, circularity varied as a function of mean Feret diameter, suggesting that larger particles hold greater potential for damaging attack angles during particle-enamel contact.
Experimental design and results.
(A) To elicit food-cleaning behaviors, we put sliced cucumbers in trays representing three treatments—food surfaces with low (
Raccoons (Procyon lotor) tend to live in wooded habitats near waterways, where they can be seen dousing foods before ingestion (photograph by Markus Zindl, reproduced with permission).
Predicted and observed cleaning times.
(A) Mean predicted time (large filled points vs. observed times (violin plots) for brushing and washing food (note log scale). The vertical line associated with predicted times represents the 5 to 95% confidence interval. (B) Predicted cleaning time as a function of cleaning inefficiency c, and handling time h, with mean predicted values (black points) for brushing and washing based on observed cleaning inefficiencies and handling times. The colored points (as in panel A) represent observed cleaning times. The trade-off between longer food handling times and efficient cleaning (oceanside food-washing; Region I) and shorter handling times and inefficient cleaning (immediate food-brushing; Region II) is depicted by the black curve.
The Aérospatiale-BAC Concorde 102 is an iconic aircraft manufactured from 1965-1979. This photo shows British Airways flight 002 on the eve of its final commercial flight on October 24, 2003 (photograph by Richard Vandervord, reproduced with permission).
Comparison of cleaning effectiveness.
JER simulated the brushing and washing behaviors of our study animals using the same three treatments of cucumber slices in our experiment. Each simulation was replicated three times (n = 18 simulations). We found that brushing was less efficient than washing across all treatments, eliminating an average of 76 ± 7% vs. 93 ± 4% of surface sands, respectively.
Individual variation.
(A) Monkeys put more time into brushing sandier treatments, X2 (2, n = 575 food-brushing events) = 194.7, p < 0.0001) with no difference across either dominance ranks or sex (Table S1). (B) Monkeys put more time into washing sandier treatments, X2 (2, n = 362 food-washing events) = 69.7, p < 0.0001), but we also detected an interaction effect with dominance rank that was independent of sex, X2 = 19.3, p < 0.0001 (Table S2). (C) Energy intake rates (kJ min-1) also varied across dominance ranks (ANOVA, LN-Transformed); F2,104= 10.0, p < 0.0001).
Standardized residual diagnostic plots for the brushing GLMM simulated using DHARMa.
(A) A qq-plot depicting deviations in the simulated standardized residuals from the overall expected distribution of the model. (B) Standardized simulated model residuals plotted against the predicted values of the model (rank transformed). Model outliers are represented by red stars. In both plots, no notable deviations were detected.
Standardized residual diagnostic plots for the washing GLMM, with ordinal rank modeled as a quadratic term, simulated using DHARMa.
(A) A qq-plot depicting deviations in the simulated standardized residuals from the overall expected distribution of the model. (B) Standardized simulated model residuals plotted against the predicted values of the model (rank transformed). Model outliers are represented by red stars. In both plots, no notable deviations were detected.
Standardized residual diagnostic plots for the washing GLMM, with ordinal rank modeled as a linear term, simulated using DHARMa.
(A) A qq-plot depicting deviations in the simulated standardized residuals from the overall expected distribution of the model. (B) Standardized simulated model residuals plotted against the predicted values of the model (rank transformed). Model outliers are represented by red stars. In both plots, no notable deviations were detected.
Predicted and observed cleaning times.
Mean predicted time (large filled points vs. observed times (violin plots) for brushing and washing food (note log scale). The vertical line associated with predicted includes uncertainty in handling time h, and represents the 5 to 95% confidence interval.
Summarized fixed effects for the food brushing GLMM (n = 575 events by animals with known rank) as an analysis of deviance table (Type II Wald Chi Square Tests).
Summarized fixed effects for the food washing GLMM (n = 362 events by animals with known rank) as an analysis of deviance table (Type II Wald Chi Square Tests) for the model that included both quadratic and linear ordinal rank terms.
Summarized fixed effects for the food washing GLMM (n = 362 events by animals with known rank) as an analysis of deviance table (Type II Wald Chi Square Tests) for the model that included just a linear ordinal rank term.
Full model fixed effects and confidence intervals for the food brushing GLMM. Modeled as a zero-inflated Poisson (ZIP) (n = 575 observations).
Full model fixed effects and confidence intervals for the food washing GLMM fixed effects with both a quadratic and linear ordinal rank term. Modeled as a zero-inflated Conway-Maxwell Poisson (ZICMP) (n = 362 observations).
Full model fixed effects and confidence intervals for the food washing GLMM fixed effects with only a linear ordinal rank term. Modeled as a zero-inflated Conway-Maxwell Poisson (ZICMP) (n =362 observations).
