It was great to have you here!

Curious about the visual human brain, a vibrant and collaborative lab, and pursuing a PhD in the heart of Europe? My lab is recruiting for a 3-year PhD position. More details: www.rademakerlab.com/job-add

"Lawful kinematics link eye movements to the limits of high-speed perception" now out in @natcomms.nature.com.

Here is the link to the paper: rdcu.be/enLiB

You find a detailed thread below!

Big shoutout to my co-authors Richard Schweitzer, Éric Castet, Tamara Watson, & @svenohl.bsky.social. I want to thank more people than I can list here (see screenshot). And we are grateful for continued funding from @dfg.de , @erc.europa.eu , @daadworldwide.bsky.social over the past few years! 15/15

Our results point to a kind of perceptual invariance in vision, to self-imposed retinal motion. They suggest that the lawful kinematics of an actions’ sensory consequences by themselves might give rise to perceptual omission. Enjoy the paper! doi.org/10.1038/s414... 14/n

We show that a parsimonious model of early visual processing captures both the phenomenology of and sensitivity to high-speed stimulus perception in our task. It gives rise to perceptual omission of saccade-induced motion while maintaining sensitivity to high-speed motion.13/n

Second, saccade speeds vary across observers and we see consistent & specific covariations with individual visibility thresholds: An individual’s velocity of leftward saccades (which entail rightward retinal motion) predicted thresholds for rightward motion and vice versa. 12/n

We believe that this relation is not coincidental. First, we show that this relation holds only if the stimulus has static movement endpoints—as little as 13 ms are enough—much like the fixated image before and after a saccade induces saccadic omission. 11/n

That means that the high-speed stimulus was more visible when moving over larger distances. But this is not a consequence of longer movement durations: The shorter the motion path, the shorter the movement duration can be to render the stimulus visible. 10/n

… instead, visibility of continuous motion depends on a conjunction of the speed, duration, and amplitude of the movement and is exactly proportional to the main sequence of saccades. Plotting speed and duration thresholds as a function of amplitude shows this relationship. 9/n

At high speeds, this task becomes impossible to do, as the (physically continuous) motion looks just like a sudden displacement. The exciting finding regards the threshold at which motion becomes invisible: It is not simply a function of the speed or duration of the stimulus... 8/n

We used @vpixx.bsky.social's PROPixx for high-speed video projection to reproduce the lawful saccadic conjunction of speed/duration/amplitude in a moving stimulus, and assessed observers’ ability to see the stimulus’ motion by asking them if the stimulus moved with an up- vs downwards curvature. 7/n

The key prediction was now that the kinematics of saccades—as defined by the main sequence—impose a limit on the ability to perceive motion even in the absence of saccades. That is, we predicted that stimuli that rapidly change place when observers fixate their gaze should become invisible. 6/n

The saccade-induced retinal motion is subjectively invisible during natural vision; we call that saccadic omission. (We have some other work that investigates the [purely visual] natural-scene properties that drive saccadic omission in detail: doi.org/10.1101/2023... – more on that another time) 5/n

The kinematic law that governs saccade-induced motion on the retina is called the main sequence, and describes the relation of saccade speed and duration to the movement's amplitude: both peak velocity and duration of the movement increase systematically with the distance the eyes travel. 4/n

Saccadic eye movements incessantly impose rapid motion on the retina. We hypothesized that this abundance of input may shape if and how we perceive high-speed motion even during fixation. Importantly, the kinematics of saccade-imposed retinal motion are very systematic. In fact, they are lawful.

Here’s the idea: Perception relies on active sampling of the environment (think whisking, looking, sniffing, touching). These actions may leave traces in perceptual processes, which we should be able to find. (Here’s a deep dive into this perspective: go.nature.com/3wo8Htd)

"Lawful kinematics link eye movements to the limits of high-speed perception" finally out in @natcomms.nature.com! We started this work back in 2016 when Eric Castet & Tamara Watson visited my (then young) lab. This work is close to my heart & supported by a @erc.europa.eu Consolidator Grant. A🧵 1/n

2/n Sven Ohl, Martin Rolfs (@rolfslab.bsky.social), and I investigated how intention modulates awareness of one of the smallest human actions: microsaccades—tiny eye movements typically generated spontaneously during fixation.

New postdoc position on visual information sampling at our Science of Intelligence cluster. If you have a strong experimental background in eye movements/ active vision/ cognitive science & would like to do a postdoc for 20 months in Berlin, please shoot me a message or simply apply!
t.co/jo7v8s7awL