https://doi.org/10.5061/dryad.gtht76htr

The dataset contains experimental data on cooperative behaviour in buff-tailed bumblebees (B. terrestris) through two tasks: a block-pushing task and a tunnel task.

Block-Pushing Experiment:

In this experiment, bees were trained to push a block to access a sugar water reward. Two experiments were conducted:

1. Experiment 1: Bees were trained either individually (Single control) or in pairs (Partnered). Delay tests were conducted to assess whether the presence of a partner influenced task performance.
2. Experiment 2: Similar to Experiment 1 but with variations in block size and training repetitions. Delay tests with longer delay times were conducted to further evaluate bee behaviour.

Results from both experiments were analysed using Generalised Linear Mixed Models (GLMMs). Key findings included:

• Partnered bees exhibited coordinated behaviour, as indicated by simultaneous block pushing.
• Delay tests showed that Partnered bees hesitated when their partner was delayed, suggesting a social influence on task performance.
• Training repetitions did not significantly affect the time taken for bees to push the block.

Tunnel Experiment:

In this experiment, bees were trained to walk through a transparent tunnel to access a sugar water reward. The experimental procedure included:

1. Training Phase: Bees were trained either individually (Single control) or in pairs (Partnered) to walk through the tunnel to access the reward.
2. Delay Tests: Bees were released into the tunnel with a delay between the first-released and second-released bees to assess their behaviour in reaching the reward door.

Results from the tunnel experiment indicated:

• Partnered bees showed coordinated behaviour, reaching the reward door simultaneously.
• Delay tests revealed that first-released bees hesitated when their partner was delayed, indicating social influence on task performance.
• Partnered bees were more likely to reverse their direction towards the reward door when encountering their partner in the tunnel, suggesting an understanding of their partner's role in the task.

Overall, the dataset provides insights into the cooperative behavior of bumblebees and the social dynamics underlying their task performance in both the block-pushing and tunnel experiments.

Description of the data and file structure

The dataset comprises several files containing behavioural observations and experimental variables related to buff-tailed bumblebees (B. terrestris) engaged in cooperative tasks. Each file serves a specific analytical purpose and contributes to understanding the mechanisms underlying cooperative behaviour in bees. Here's an overview of the data and file structure:

1. coop.csv (Models 2 and 3):

* This file contains data related to the block-pushing experiment.
* Variables include BeeID (Bee identity number), Pair (Pair number), Colony (Colony number), Treatment (Treatment group: 1 = partnered, 2 = controls), Experiment (Experiment number: 1 = Styrofoam, 2 = Lego), Waiting (Waiting time in the delay test), Order (Order of delay test), SecondBeePushedFirst (Whether the first-released bee pushed the block before the second-released bee), and beePushedWhenPartnerPresent (Whether the second-released bee’s location influenced the first-released bee’s first push on the block).
2. coop2.csv (Model 1):

* Data from the block-pushing experiment for Model 1 analysis.
* Variables include BeeID, Pair, Colony, Treatment, Experiment, Waiting, and Order.
3. Coop2ndReleasedVSControls.csv (Model 6):

* Data comparing second-released bees and control bees in the tunnel experiment.
* Variables include Colony, BeeID, Pair, Treatment, and Time (Time to touch the door in seconds).
4. CoopTime3.csv (Model 4):

* Contains data from bouts where bees do not turn in the tunnel.
* Variables include Phase (Phase number), Time (Time to touch the door), and Treatment.
5. CoopTimeSecond.csv (Model 5):

* Data from paired bees in the delay test comparing first-released and second-released bees.
* Variables include Phase (Test number of first-released bee).
6. CoopTraining.csv (Model 8):

* Data on whether bees were facing towards their partner's side at the door during training.
* Variables include Facing towards a partner at the door.
7. CoopTrainingData4PilotPartnered.csv (Initial analyses):

* Data to test differences in time to reach the reward door between the end of training and the delay test in the Partnered group.
* Variables include Phase (Phase number).
8. CoopTurningPropPhases1and3.csv (Model 9):

* Data on whether paired bees were more likely to turn back if they encountered their partner in the adjacent tunnel during bouts.
* Variables include PropTurn (Proportion of turnings when the partner was at the same division).
#### Additional Data Files:
ICCBlock.csv: Interclass Correlation Coefficients for the block-pushing experiment.
* Variables include Waiting (measured by different observers (abbreviations explained below)), SecondBeePushedFirst (measured by different observers), and beePushedWhenPartnerPresent (measured by different observers).
* InterreliabilityTunnel.csv: Interclass Correlation Coefficients values for the tunnel experiment.
* Variables include Time (measured by different observers), Nro of turns at the door (measured by different observers), Nro of turning back from the door when facing the partner (measured by different observers), Waiting time at the door before turning at first time (measured by different observers), Facing towards a partner at the door (measured by different observers).

#### Abbreviations:
* Nro = number
* AA = Anna Antinoja
* KM = Kaarle Mäkelä
* OL = Olli Loukola

#### Units and Interpretation Keys:
* Continuous Variables:

* Waiting: Waiting time in the delay test in seconds
* Time: Time to touch the door in seconds
* Nro of turns at the door
* Nro of turning back from the door when facing the partner
* Waiting time at the door before turning at first time: Waiting time in seconds
* Categorical Variables:

* BeeID: Bee identity number
* Pair: Pair number
* Colony: Colony number
* Treatment: Treatment group (1 = partnered, 2 = controls)
* Experiment: Experiment number (1 = Styrofoam, 2 = Lego)
* Order: Order of delay test
* SecondBeePushedFirst: Whether the first-released bee pushed the block before the second-released bee
* beePushedWhenPartnerPresent: Whether the second-released bee’s location influenced the first-released bee’s first push on the block
* Facing towards a partner at the door: Whether the bee was facing the other tunnel at the reward door (0 = no, 1= yes)

Relationships Between Data Files:

• The files contain complementary information from different experiments and analyses related to cooperative behaviour in bees.
• Bee identification numbers (BeeID) and experimental conditions (Treatment) serve as common identifiers across files, facilitating data integration and comparisons.
• Experimental variables such as Waiting time, Order, and Experiment number are consistent across multiple files, allowing for standardised analysis and interpretation.

Missing Data and Codes:

• Code is in the R script-file

Overall, the structured organisation of the dataset enables researchers to conduct detailed analyses exploring various aspects of cooperative behaviour in buff-tailed bumblebees, fostering a deeper understanding of social interactions in insect colonies.