Description
Supporting dataset and R code for the manuscript:
'Acrylic fibres do not affect nest building or incubation in laboratory zebra finches' published in 'Behaviour'.
In this study, we tested if, and how, the use of the synthetic acrylic fibres as nest material (n = 12) influenced 1) the nest building behaviour of pairs of laboratory zebra finches, 2) the insulative properties of the nest they build, 3) the temperature in their nest during incubation and 4) the resulting hatching success of their clutch. Control pairs (n = 12) were provided with wool fibres, the material of natural origin acrylic fibres typically replace. We also investigated whether hatching success had an influence on these.
We measured the insulation quality (i.e., how good a nest was at preventing cooling) of nests built with each type of material by pairs that either succeeded or failed at hatching at least one chick. We also measured the amount of material used to build each type of nest, the total weight of cardboard cage liners added to the nests, as well as the temperature in each nest during the week following the lay of the last egg of a clutch (i.e. 8 days in total). Finally, we determined whether the type of material used for nest building influenced hatching success and hatching rates.
##################
Dataset (‘Hebert_et_al_data.xlsx’ file):
• Pair_ID: Unique pair ID number [P21 omitted as never laid an egg; see article for details]
• Material: Type of nest-building material provided – either ‘wool’ or ‘acrylic’
• Pieces_used: Number of pieces used to build the nest
• Cardboard_g: Amount of ripped cardboard (in grams) added to the nest
• Hatching_success: A pair was classified as either ‘successful’ if it hatched at least one chick, or ‘unsuccessful’ if it failed.
• Clutches: Number of clutches laid by each pair – either ‘1’, or ‘>1’
• First_clutch_size: Number of eggs laid as part as the pair’s first clutch. A clutch is defined as a series of eggs laid with a maximum of one single-day gap between two eggs.
• Successful_initial_clutch_size: Initial number of eggs in clutches where at least one chick hatched [successful pairs only]
• Hatchlings: Number of chicks a pair hatched .
• Fledglings: Number of chicks the pair reared in the nest until fledging.
• Percent_eggs_hatched(success): Percentage of chicks that hatched from the successful clutch [successful pairs only]. Referred to as ‘Hatching rate’ in the article.
Calculated as follow: [‘Hatchlings’ / ‘Successful_initial_clutch_size’ ] x 100
• Average_incubation_temp: Average of the 1075-1152 temperature recordings within the nest during incubation (8-day period starting on the day the last egg of the clutch was laid)
• Average_daily_max_delta: Average of the eight values of daily maximal temperature variations within a nest during the same period (i.e., each daily value is [maximum temperature – minimum temperature recorded at the nest]; see Supplementary Figure S3). Referred to as ‘Max. Δ temperature’ in the article.
• Heat_retention: Referred to as ‘Nest insulation quality’ in the article. This is the average value of two heat retention capacity tests (see article for details) performed in the laboratory. The better a nest was insulated, the less heat its temperature logger lost per unit of time relative to its control located outside the nest. As a result, higher nest insulation quality values (in °C/20 s) indicate nests more effective at retaining heat, and vice versa.
Colour code:
Cells: Blue = wool; Yellow = acrylic
Text: Green = hatching success; Red = hatching failure
##################
R script (‘Hebert_et_al.R’ file) containing :
• The code used for statistical analyses.
• A code to split the dataset into sub-datasets by material type or hatching success and to plot data as in the article
All statistical analyses were performed using RStudio for Windows (2022.07.0 release; with R version 4.1.3) and plots were generated with GraphPad Prism. All tests were two-tailed and used with a significance level of p < 0.05.
[Results]
• Nest building (Figure 1)
Robust rank-based 2x2 ANOVAs (package Rfit) were used to determine whether the quantity of nest-building material (Figure 1a) and the amount (i.e. weight, in grams) of cardboard (Figure 1b) added to the nests differed between material type and/or hatching success. A Spearman correlation was also used to determine if these two variables were associated.
• Nest insulation quality (Figure 2 and Supplementary Figure S4)
A robust 2x2 ANOVA (package Rfit) was used to determine whether the capacity of the nests to retain heat (i.e., ‘Nest insulation quality’, or ‘Heat_retention’ in the excel datafile) differed between material type and/or hatching success (Figure 2a). Spearman correlations were used to determine if nest insulation quality was associated with (i) the quantity of nest-building material (Figure 2b) or amount of cardboard (Supplementary Figure S4) incorporated into each nest, and (ii) the number of fledglings each nest had produced (Figure 2c).
• Incubation temperatures (Figures 3-4 and Supplementary Figures S5-6)
Average incubation temperature - A robust 2x2 ANOVA (package Rfit) was used to determine whether the average incubation temperature (i.e., ‘Average_incubation_temp’) differed between material type and/or hatching success (Figure 3a). One-sample Wilcoxon signed-rank tests were used to determine whether the average incubation temperature differed from the ambient room temperature (i.e., 20°C), for each hatching outcome separately. Finally, Spearman correlations were used to determine if these average incubation temperatures were associated with the quantity of nest-building material (Figure 3b) or amount of cardboard (Figure 3c) incorporated into each nest. Spearman correlations were also used to determine if there was a relationship between incubation temperature and number of pieces (as in Figure 3b) for each type of material (Supplementary Figure S5a-b) and hatching success (Supplementary Figure S5c-d) separately.
Average daily maximal temperature difference (Max. Δ temperature) - A robust 2x2 ANOVA (package Rfit) was used to determine whether Max. Δ temperatures (i.e., ‘Average_daily_max_delta’) differed between material type and/or hatching success (Supplementary Figure S6). Spearman correlations were then used to determine if pairs’ Max. Δ temperatures were associated with the quantity of material the nests of failed (Figure 4a) and successful (Figure 4b) pairs had been built with.
• Clutch size and hatching success (Supplementary Figure S7)
Likelihood of hatching success - Fischer’s exact test was used to compare whether nests built with either material (i.e., wool or acrylic) differed in likelihood of hosting at least one hatchling (i.e. hatching success or hatching failure).
Clutch size - Wilcoxon rank sum tests were used to compare i) the size of the first clutch (i.e., ‘First_clutch_size ‘) between successful and failed pairs (Supplementary Figure S7a), and ii) the initial size of the successful clutch (i.e., ‘Successful_initial_clutch_size’) between the two material types (Supplementary Figure S7b).
Hatching rate - Hatching rates (i.e., ‘Percent_eggs_hatched(success)’) were compared between the two material types using a Wilcoxon rank sum test (Supplementary Figure S7c).
[Discussion]
We calculated the average difference in insulation quality (i.e., ‘Heat_retention’) between unsuccessful (n = 10) and successful (n = 13) nests. This was then used to determine how much additional heat successful nests lost per hour (°C/h) compared to nests in which no eggs hatched.
'Acrylic fibres do not affect nest building or incubation in laboratory zebra finches' published in 'Behaviour'.
In this study, we tested if, and how, the use of the synthetic acrylic fibres as nest material (n = 12) influenced 1) the nest building behaviour of pairs of laboratory zebra finches, 2) the insulative properties of the nest they build, 3) the temperature in their nest during incubation and 4) the resulting hatching success of their clutch. Control pairs (n = 12) were provided with wool fibres, the material of natural origin acrylic fibres typically replace. We also investigated whether hatching success had an influence on these.
We measured the insulation quality (i.e., how good a nest was at preventing cooling) of nests built with each type of material by pairs that either succeeded or failed at hatching at least one chick. We also measured the amount of material used to build each type of nest, the total weight of cardboard cage liners added to the nests, as well as the temperature in each nest during the week following the lay of the last egg of a clutch (i.e. 8 days in total). Finally, we determined whether the type of material used for nest building influenced hatching success and hatching rates.
##################
Dataset (‘Hebert_et_al_data.xlsx’ file):
• Pair_ID: Unique pair ID number [P21 omitted as never laid an egg; see article for details]
• Material: Type of nest-building material provided – either ‘wool’ or ‘acrylic’
• Pieces_used: Number of pieces used to build the nest
• Cardboard_g: Amount of ripped cardboard (in grams) added to the nest
• Hatching_success: A pair was classified as either ‘successful’ if it hatched at least one chick, or ‘unsuccessful’ if it failed.
• Clutches: Number of clutches laid by each pair – either ‘1’, or ‘>1’
• First_clutch_size: Number of eggs laid as part as the pair’s first clutch. A clutch is defined as a series of eggs laid with a maximum of one single-day gap between two eggs.
• Successful_initial_clutch_size: Initial number of eggs in clutches where at least one chick hatched [successful pairs only]
• Hatchlings: Number of chicks a pair hatched .
• Fledglings: Number of chicks the pair reared in the nest until fledging.
• Percent_eggs_hatched(success): Percentage of chicks that hatched from the successful clutch [successful pairs only]. Referred to as ‘Hatching rate’ in the article.
Calculated as follow: [‘Hatchlings’ / ‘Successful_initial_clutch_size’ ] x 100
• Average_incubation_temp: Average of the 1075-1152 temperature recordings within the nest during incubation (8-day period starting on the day the last egg of the clutch was laid)
• Average_daily_max_delta: Average of the eight values of daily maximal temperature variations within a nest during the same period (i.e., each daily value is [maximum temperature – minimum temperature recorded at the nest]; see Supplementary Figure S3). Referred to as ‘Max. Δ temperature’ in the article.
• Heat_retention: Referred to as ‘Nest insulation quality’ in the article. This is the average value of two heat retention capacity tests (see article for details) performed in the laboratory. The better a nest was insulated, the less heat its temperature logger lost per unit of time relative to its control located outside the nest. As a result, higher nest insulation quality values (in °C/20 s) indicate nests more effective at retaining heat, and vice versa.
Colour code:
Cells: Blue = wool; Yellow = acrylic
Text: Green = hatching success; Red = hatching failure
##################
R script (‘Hebert_et_al.R’ file) containing :
• The code used for statistical analyses.
• A code to split the dataset into sub-datasets by material type or hatching success and to plot data as in the article
All statistical analyses were performed using RStudio for Windows (2022.07.0 release; with R version 4.1.3) and plots were generated with GraphPad Prism. All tests were two-tailed and used with a significance level of p < 0.05.
[Results]
• Nest building (Figure 1)
Robust rank-based 2x2 ANOVAs (package Rfit) were used to determine whether the quantity of nest-building material (Figure 1a) and the amount (i.e. weight, in grams) of cardboard (Figure 1b) added to the nests differed between material type and/or hatching success. A Spearman correlation was also used to determine if these two variables were associated.
• Nest insulation quality (Figure 2 and Supplementary Figure S4)
A robust 2x2 ANOVA (package Rfit) was used to determine whether the capacity of the nests to retain heat (i.e., ‘Nest insulation quality’, or ‘Heat_retention’ in the excel datafile) differed between material type and/or hatching success (Figure 2a). Spearman correlations were used to determine if nest insulation quality was associated with (i) the quantity of nest-building material (Figure 2b) or amount of cardboard (Supplementary Figure S4) incorporated into each nest, and (ii) the number of fledglings each nest had produced (Figure 2c).
• Incubation temperatures (Figures 3-4 and Supplementary Figures S5-6)
Average incubation temperature - A robust 2x2 ANOVA (package Rfit) was used to determine whether the average incubation temperature (i.e., ‘Average_incubation_temp’) differed between material type and/or hatching success (Figure 3a). One-sample Wilcoxon signed-rank tests were used to determine whether the average incubation temperature differed from the ambient room temperature (i.e., 20°C), for each hatching outcome separately. Finally, Spearman correlations were used to determine if these average incubation temperatures were associated with the quantity of nest-building material (Figure 3b) or amount of cardboard (Figure 3c) incorporated into each nest. Spearman correlations were also used to determine if there was a relationship between incubation temperature and number of pieces (as in Figure 3b) for each type of material (Supplementary Figure S5a-b) and hatching success (Supplementary Figure S5c-d) separately.
Average daily maximal temperature difference (Max. Δ temperature) - A robust 2x2 ANOVA (package Rfit) was used to determine whether Max. Δ temperatures (i.e., ‘Average_daily_max_delta’) differed between material type and/or hatching success (Supplementary Figure S6). Spearman correlations were then used to determine if pairs’ Max. Δ temperatures were associated with the quantity of material the nests of failed (Figure 4a) and successful (Figure 4b) pairs had been built with.
• Clutch size and hatching success (Supplementary Figure S7)
Likelihood of hatching success - Fischer’s exact test was used to compare whether nests built with either material (i.e., wool or acrylic) differed in likelihood of hosting at least one hatchling (i.e. hatching success or hatching failure).
Clutch size - Wilcoxon rank sum tests were used to compare i) the size of the first clutch (i.e., ‘First_clutch_size ‘) between successful and failed pairs (Supplementary Figure S7a), and ii) the initial size of the successful clutch (i.e., ‘Successful_initial_clutch_size’) between the two material types (Supplementary Figure S7b).
Hatching rate - Hatching rates (i.e., ‘Percent_eggs_hatched(success)’) were compared between the two material types using a Wilcoxon rank sum test (Supplementary Figure S7c).
[Discussion]
We calculated the average difference in insulation quality (i.e., ‘Heat_retention’) between unsuccessful (n = 10) and successful (n = 13) nests. This was then used to determine how much additional heat successful nests lost per hour (°C/h) compared to nests in which no eggs hatched.
| Date made available | 14 Sept 2026 |
|---|---|
| Publisher | University of St Andrews |