trained my first CNN today. am I an AI researcher now? 😂

Including driving variables improved out-of-sample prediction performance above a demographic-only model, most notably for physical health, life satisfaction, and social role constraints

Some driving behaviors are highly specific in their predictions. E.g., longer minutes per trip more strongly predicted physical decline than anything else.

Everyday driving patterns from 2,658 older adults across the US were recorded. They also completed self-reported measures of general well-being and health across cognitive, physical, social, and mental domains. We found widespread associations between driving behaviors and health:

Personal news: Today is my first day as an Assistant Research Scientist at the University of Michigan Transportation Research Institute. Excited to apply my expertise in attention and visual perception to an important real-world issue -- driving safety! #GoBlue

Why do abrupt visual onsets capture our attention? A common assumption is that they are highly salient. Here, we show that this assumption is flawed. Higher physical salience made a distractor easier to ignore, whereas a less salient onset distractor captured attention. doi.org/10.1037/xhp0...

An AOI drawer to flexibly define AOIs (6/8)

A Fixation Viewer to visualize fixation patterns (5/8)

In addition to its very intuitive API, PupEyes offers interactive tools such as a Pupil Viewer to inspect pupil traces (4/8)

Doing science is hard these days, but here’s our new review on attentional capture by abrupt onsets, published in @jephpp.bsky.social ! We hope to offer some food for thought for both newcomers and experienced researchers alike. Hope you enjoy it! psycnet.apa.org/record/2025-...

A computational model explains these data as two priority signals with different latencies and impacts on behavior. A distractor signal can be suppressed when the target signal is ready or, alternatively, when the distractor signal itself exerts less direct impact on behavior.

This allowed us to track the entire spatiotemporal dynamics of visual attention with respect to certain stimuli. E.g.,

To address this, we applied a "forced-response" method to saccadic behavior. This method forces saccade execution while systematically varying stimulus processing time, essentially forcing subjects to express their current attentional priority.

Visual attention is often thought to be guided by a latent "priority map". But this priority map also evolves over time. The deployment of attention (e.g., an eye movement) reflects only a snapshot of this evolution over time and space.