A biological image generator: cuttlefish camouflage and visual perception
| Project/Area Number |
20K15939
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 46030:Function of nervous system-related
|
|---|
| Research Institution | Okinawa Institute of Science and Technology Graduate University |
|---|
Principal Investigator |
Reiter Samuel 沖縄科学技術大学院大学, 計算行動神経科学ユ ニット, 准教授 (60869155)
|
|---|
| Project Period (FY) |
2020-04-01 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2021: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2020: ¥2,990,000 (Direct Cost: ¥2,300,000、Indirect Cost: ¥690,000)
|
|---|
| Keywords | Visual perception / Camouflage / Cephalopod / Eeep learning / Psychophysics / Motor control / Computation / Vision / Cuttlefish / cuttlefish / visual perception / camouflage / neuroscience / motor control |
|---|
| Outline of Research at the Start |
Cuttlefish possess a unique readout of their visual perception: they can change their skin camouflage pattern and color depending on their visual environment. To do this they use their complex brains to directly control up to millions of specialized pigment-filled skin cells called chromatophores. How exactly does the choice of camouflage pattern depend on the visual environment? This project will address this question by combining modern deep learning-based methods with a description of cuttlefish camouflage in terms of the activity of hundreds of thousands of individual chromatophores.
|
| Outline of Final Research Achievements |
Previous work suggested that cuttlefish camouflage to their environments by choosing one of a small number of camouflage patterns that best match prominent background image statistics. We built a high-resolution filming array and designed computational techniques to describe cuttlefish camouflage quantitatively and objectively for the first time. Using these high resolution filming and computational techniques, we found that cuttlefish camouflage is instead 1) high dimensional, 2) able to precisely match higher order background image statistics, 3) has dynamics characterized by a meandering search, most likely utilizing visual feedback, 4) is composed of groups of chromatophores that are flexibly reconfigured during different camouflage trajectories, and 5) differs substantially from simple visual threat behaviors. These results will guide mechanistic studies of this remarkable texture-matching system.
|
| Academic Significance and Societal Importance of the Research Achievements |
Our results push the boundaries of high-resolution filming of animal behavior, to study a system evolutionarily specialized to report the visual perception of animals which developed large brains independently of our own vertebrate lineage.
|
Report
(3 results)
Research Products
(1 results)
