Dynamic visual effects enhance flower conspicuousness but compromise colour perception

Dynamic visual effects such as glossiness are taxonomically widespread and have evolved repeatedly across the tree of life. Their changeable nature poses a challenge for reliable signalling, because for signals to be reliable, they should be consistent. Glossy visual effects defy that principle, because the bright, directional pulse of light that dominates their appearance is highly variable across space and time. Here, we investigate how dynamic light reflections influence signal efficacy using bumblebees as a model of insect vision and plant-pollinator interactions. We show that glossy floral signals occur across Angiosperm lineages. Through behavioural experiments with artificial stimuli that mimic the spectral and spatial reflectance properties of glossy and matte floral surfaces, we demonstrate that glossiness enhances long-range detectability but compromises fine-scale colour discrimination at close range. Glossiness thus poses an optical property by which organisms can attune the..., , , # Dynamic visual effects enhance flower conspicuousness but compromise colour perception

Dataset DOI: [10.5061/dryad.5tb2rbph3](10.5061/dryad.5tb2rbph3)

## Description of the data and file structure

Behavioural experiments to test the functional significance of surface gloss in visual signalling

### Files and variables

#### File: raw_data_plot_panel_A.csv

**Description:** Innate preference tests

##### Variables

* Colony_id
* Bee_id
* Date (Day/Month/Year)
* Stimulus_colour (b=blue, y=yellow)
* Count_matte
* Count_glossy
* Preference (m = matte, g = glossy, none=no preference)

#### File: raw_data_plot_panel_B.csv

**Description:** Learning test blue stimuli

##### Variables

* Colony_id
* Bee_id
* Date (Day/Month/Year)
* Stimulus_colour (b=blue, y=yellow)
* Count_matte
* Count_glossy
* Preference (i.e. average based on \"count_matte\" and \"count_glossy\", m=matte, g=glossy, none=no preference)
* Individual_training_visit (number depends on trainings required, max = 80 visits; 0=vis...