Bevil Conway
@BevilConway
Followers
2K
Following
7K
Media
183
Statuses
2K
neuroscientist, artist, dad🇨🇦🇿🇼🏳️🌈 color, art, vision, cognition, & intrigued by social forces shaping science (he), views mine. [email protected]
Joined July 2019
A few postdocs asked me to read over their job applications. Here’s a few bits of advice: 🧵 /10. 1. If any advice doesn’t ring true for you, discard it (including this!). 2. People on the search committee are keen to welcome a colleague, they aren’t looking to take you down.
4
104
498
Congratulations on the book Nicole, and thanks for giving us a new and hopeful narrative for science today.
I hope Elusive Cures will help shift the current narrative (from scientists as corrupt). Most scientists I know are impassioned individuals throwing their best at complex challenges that matter for society; this is how Elusive Cures describes them. I'm humbled by this review.
1
0
5
The recent paper pertaining to the binding problem implies that there is no functional specialization in the visual system. I am agnostic about whether there is a binding problem, but there is no denying that neurons are typically not both direction selective and color selective.
There is no binding problem because vision—and cortex more broadly—is not modular. Phrenological theories create phrenological artifacts. Beyond binding: from modular to natural vision.#neuroscience.
3
0
15
Come join us as a postdoc to do some great science, collaboration between my lab and Dan’s, the best of both worlds (theory and neurophys)! See details in Dan’s post.
@BevilConway (NIH) and @NeuralCodeUMD (UMD) are seeking a brilliant postdoc to study high-acuity vision combining large-scale electrophys and deep computational modeling. Prior experience (broadly speaking) is recommended: email Dan Butts with CV and brief research statement.
0
1
4
RT @NEIDirector: Great @Nature news article about what is being done to slow, stop, or even reverse geographic atrophy (GA), the advanced f….
0
9
0
Thrilling work with far reaching implications not only for basic knowledge of how the jelly between your ears works but how to intervene when it doesn’t.
0
0
1
0
19
0
RT @j_g_allen: Know any kids in your life who want to be a doctor or scientist, with ambitions to cure cancer or find a cure for Alzheimer'….
0
95
0
The sky is rarely green. Why? It was fun confronting this surprising question with the smart folks at the BBC.
The sky can be beautiful fiery colours at sunset 🌅 & a cool blue during the day 🩵 but why is the sky never green? 💚. CrowdScience took a trip to the artic circle to find out. With @katieherli and @BevilConway. Producer: @hfishaay . Editor: @owlglass.
0
0
0
RT @pebonilla: NIH apparently axing the postbac program that was a launchpad for many thousands of future STEM PhD’s — my wife being one of….
0
4
0
The “e” on my keyboard is stuck, resulting in a vile interaction with my dyslexia on some of the text in the images. Sorry!.
1
0
0
Many studies have used color to study memory or cognition, assuming the color space is uniform. That work must be re-evaluated because the same distortions we observed in the color-matching data from monkeys are also apparent in the data collected in people. /END
4
0
2
What is the origin of color categories? The ability to form them is probably innate, but the emergence of consensus color categories likely depends on language. Lastly, the work has one major implication for many published studies in the psychology of human memory &cognition. 15/.
1
0
1
Macaque monkeys can spontaneously acquire color categories. One animal in our study had a category corresponding to pea green. Why? Who knows. But he couldn't share it with the other animals because monkeys don't have language. This work answers a 1000-year-old question. 14/
1
0
0
Lindsey & Brown sum it up: "human consensus [color] categories likely reflect the need for efficient and effective communication. as embodied in human language". But could monkeys learn color categories? 13/
1
0
1
Biases in color matches by monkeys are explained by unwitting distortions in color space. So, macaques do not have true consensus color categories. Since the visual system is similar in humans & macaques, the results show that humans don't have innate color categories either. 12/
1
0
1
We simulated color-match data characterized by a true category & by a space non-uniformity. Choice distributions of these 2 simulations, for every cue, can be plotted in a heatmap (thx @timothyfbrady!), and are very different. Which explains the monkey color-matching data? 11/
1
0
1
If you over sample the color space, you get a bias simply when there are more options on one side of the cue. While it is easy to order colors (. ROYGBIVROY. ), it is not easy to space them evenly. We used CIELUV, widely used by others & assumed to be uniform. But is it? 10/
1
0
1
The top panel is data from the study in humans, providing evidence for four color consensus color categories. The bottom panel shows data from monkeys. The monkeys seem to show evidence for two color categories. But before we jump to conclusions, we need to rule something out. 9/
1
0
1
Danny and co. adapted the paradigm as a four-alternative-forced choice, so on every trial there was a direct (rewarded) match for the cue. After thousands of trials, they had enough data to assess the errors. Did monkeys show color categories? If so, how many? and which ones? 8/.
1
0
1