Andrey Chetverikov
@achetverikov.bsky.social
Associate Professor in Cognitive Psychology at the University of Bergen, Norway. I study decision-making and biases in perception and visual working memory, with occasional forays into higher level decisions. https://andreychetverikov.org
This is my soul animal at the cover of the new Andrew Bird's EP (good music as well)
soundcloud.com/andrewbird/s...
soundcloud.com/andrewbird/s...
November 10, 2025 at 4:00 PM
This is my soul animal at the cover of the new Andrew Bird's EP (good music as well)
soundcloud.com/andrewbird/s...
soundcloud.com/andrewbird/s...
Huge congrats to @ben-kop.bsky.social on a successful PhD defense! Ben, it was my great pleasure to work with you, and I will also fondly remember our little chats at parties and conferences =) All the best in your career, wherever it takes you.
November 5, 2025 at 4:48 PM
Huge congrats to @ben-kop.bsky.social on a successful PhD defense! Ben, it was my great pleasure to work with you, and I will also fondly remember our little chats at parties and conferences =) All the best in your career, wherever it takes you.
Three days full of science in Frankfurt visiting @bledowski.bsky.social and @rademaker.bsky.social. Very happy to chat with the amazing students and colleagues in both labs! Now off to @ecvp.bsky.social in Mainz.
August 22, 2025 at 9:39 PM
Three days full of science in Frankfurt visiting @bledowski.bsky.social and @rademaker.bsky.social. Very happy to chat with the amazing students and colleagues in both labs! Now off to @ecvp.bsky.social in Mainz.
In the thread there people suggest openalex.org as an alternative. Seems like a good one, but will definitely require some getting used to. One thing I immediately miss is the excerpts from the paper where the search terms appear.
August 14, 2025 at 9:00 AM
In the thread there people suggest openalex.org as an alternative. Seems like a good one, but will definitely require some getting used to. One thing I immediately miss is the excerpts from the paper where the search terms appear.
My talk "Biased colors in biased space" is also not in the "Bias" section as you might have imagined but in "Color 2" on the last day. Oh well =) Would be happy to chat with the color crowd. And our PhD student Ekaterina Andriushchenko has a poster on Wednesday. Will post details later!
August 12, 2025 at 10:29 AM
My talk "Biased colors in biased space" is also not in the "Bias" section as you might have imagined but in "Color 2" on the last day. Oh well =) Would be happy to chat with the color crowd. And our PhD student Ekaterina Andriushchenko has a poster on Wednesday. Will post details later!
I think ScienceDirect implies that my interest in vision research is unhealthy
August 5, 2025 at 1:30 PM
I think ScienceDirect implies that my interest in vision research is unhealthy
Rewatched the first episode of Father Ted yesterday (30 years!) and I can't help thinking about the fellow imagery researchers when seeing this wonderful diagram. Isn't Father Dougal McGuire a great example of hyperphantasia? #imagery #fatherted
June 18, 2025 at 7:18 AM
Rewatched the first episode of Father Ted yesterday (30 years!) and I can't help thinking about the fellow imagery researchers when seeing this wonderful diagram. Isn't Father Dougal McGuire a great example of hyperphantasia? #imagery #fatherted
Am not bad at this game, am I?
June 8, 2025 at 8:12 AM
Am not bad at this game, am I?
Congrats nevertheless! You can do a lot of good with a bit more power =)
May 23, 2025 at 6:25 PM
Congrats nevertheless! You can do a lot of good with a bit more power =)
This is because the effect of condition is already visible in the very first frame, and the bias on the first frame is a strong predictor of the bias on the last frame. If a trajectory ended up with an attractive bias, it's more likely to start with an attractive bias and vice versa. 5/7
May 21, 2025 at 9:48 AM
This is because the effect of condition is already visible in the very first frame, and the bias on the first frame is a strong predictor of the bias on the last frame. If a trajectory ended up with an attractive bias, it's more likely to start with an attractive bias and vice versa. 5/7
What about response trajectories? We found that response trajectories are mostly repulsive for both conditions, ending with attraction in the Stroop condition. However, we point out that trajectories do not give strong evidence that initial biases are repulsive. 4/7
May 21, 2025 at 9:48 AM
What about response trajectories? We found that response trajectories are mostly repulsive for both conditions, ending with attraction in the Stroop condition. However, we point out that trajectories do not give strong evidence that initial biases are repulsive. 4/7
We replicated their work in a within-subject design instead of separate experiments. The latter also allowed us to look at individual differences, which were surprisingly stable across conditions (see panel B). Most people actually don't change the bias sign when the distracting task is added. 3/7
May 21, 2025 at 9:48 AM
We replicated their work in a within-subject design instead of separate experiments. The latter also allowed us to look at individual differences, which were surprisingly stable across conditions (see panel B). Most people actually don't change the bias sign when the distracting task is added. 3/7
Chen and Bae (www.sciencedirect.com/science/arti...) reported that 1) serial dependence is repulsive in a pointing direction estimation task; 2) it switches to attraction when a Stroop task is added to the delay period; 3) the trajectory of the mouse response starts with repulsion regardless. 2/7
May 21, 2025 at 9:48 AM
Chen and Bae (www.sciencedirect.com/science/arti...) reported that 1) serial dependence is repulsive in a pointing direction estimation task; 2) it switches to attraction when a Stroop task is added to the delay period; 3) the trajectory of the mouse response starts with repulsion regardless. 2/7
New preprint! Serial dependence is assumed to be attractive, but some studies consistently show repulsion. We tried to replicate a surprising repulsive serial bias that switches to an attractive one when people get distracted during the memory maintenance. It worked! osf.io/preprints/ps... 🧵
May 21, 2025 at 9:48 AM
New preprint! Serial dependence is assumed to be attractive, but some studies consistently show repulsion. We tried to replicate a surprising repulsive serial bias that switches to an attractive one when people get distracted during the memory maintenance. It worked! osf.io/preprints/ps... 🧵
There is a special place in hell for those who forces people to use checkboxes for continuous variables
May 5, 2025 at 11:03 AM
There is a special place in hell for those who forces people to use checkboxes for continuous variables
There are rector elections coming at UiB soon. There are two teams competing, and after reading their programs, I feel like they are quite similar. Didn't find much about basic research aside from these two bits. The first one looks self-contradictory, and the second is very vague.
February 24, 2025 at 9:01 AM
There are rector elections coming at UiB soon. There are two teams competing, and after reading their programs, I feel like they are quite similar. Didn't find much about basic research aside from these two bits. The first one looks self-contradictory, and the second is very vague.
It should create a plot like this, where the greenish curves are the results from a model sampling prior only and the red/lily is the data and the generative model. The prior for eta is set to 100 ms.
January 22, 2025 at 10:30 AM
It should create a plot like this, where the greenish curves are the results from a model sampling prior only and the red/lily is the data and the generative model. The prior for eta is set to 100 ms.
sure, I'm writing to "Begin forwarded message", thank you for this helpful suggestion Outlook
November 15, 2024 at 8:11 AM
sure, I'm writing to "Begin forwarded message", thank you for this helpful suggestion Outlook
And that’s how you know that VSS abstracts been published:) #visionscience
October 1, 2024 at 7:03 AM
And that’s how you know that VSS abstracts been published:) #visionscience
I spent a lot of time last weeks to find a way to fit my model to the data. The problem is that in absence of analytic solution, you need to simulate a lot. I finally managed to get it done with JAX on GPU, so my laptop can churn numbers pretty quickly. Feeling quite proud about this one!
August 22, 2024 at 7:13 PM
I spent a lot of time last weeks to find a way to fit my model to the data. The problem is that in absence of analytic solution, you need to simulate a lot. I finally managed to get it done with JAX on GPU, so my laptop can churn numbers pretty quickly. Feeling quite proud about this one!
Submitted circhelp package to CRAN! With recent updates, the functions to remove angular biases work even better, not just for motion and orientation, but also for other circular data, such as color.
achetverikov.github.io/circhelp/ind...
achetverikov.github.io/circhelp/ind...
July 3, 2024 at 12:30 PM
Submitted circhelp package to CRAN! With recent updates, the functions to remove angular biases work even better, not just for motion and orientation, but also for other circular data, such as color.
achetverikov.github.io/circhelp/ind...
achetverikov.github.io/circhelp/ind...
@ivntmc.bsky.social in the other reply pointed out their papers where they did something like that. It's not about stimulus noise per se, but at least suggests that the initial noise from short presentation time and set size is multiplied during the delay
elifesciences.org/articles/910...
elifesciences.org/articles/910...
June 19, 2024 at 7:25 AM
@ivntmc.bsky.social in the other reply pointed out their papers where they did something like that. It's not about stimulus noise per se, but at least suggests that the initial noise from short presentation time and set size is multiplied during the delay
elifesciences.org/articles/910...
elifesciences.org/articles/910...
@theresecollins.bsky.social @alexh.bsky.social apparently they posted online too soon =) should be working now
March 4, 2024 at 8:41 AM
@theresecollins.bsky.social @alexh.bsky.social apparently they posted online too soon =) should be working now