Thank you!

Here's my previous post for summary and preprint link: bsky.app/profile/yong...

Real-world Objects Scaffold Visual Working Memory for Features: Increased Neural Engagement When Colors Are Remembered as Part of Meaningful Objects Abstract. Visual working memory is a core cognitive function that allows active storage of task-relevant visual information. Contrary to the common assumption that the capacity of this system is fixed...

New paper with @timbrady.bsky.social and @violastoermer.bsky.social now out in JoCN! "Real-world Objects Scaffold Visual Working Memory for Features: Increased Neural Engagement When Colors Are Remembered as Part of Meaningful Objects" doi.org/10.1162/JOCN...

This highlights the role of semantics in VWM, challenging models that treat WM as primarily perceptual with fixed limits. Using naturalistic stimuli with careful evaluations may enable us to probe the rich structure of VWM, offering a deeper understanding of how VWM is used in the real-world. 11/

How can real objects enhance VWM? We hypothesize that semantic distinctiveness is the key: beyond visual features, real objects engage conceptual knowledge, yielding more distinctive and potentially stable memory representations, ultimately aiding performance. 10/

Lastly, a large-sample (n=300) correlation analysis showed that subjective familiarity rating predicted memory performance for real objects, while colourfulness scaled with memory performance for counterfeit objects, despite both stimulus sets matched in colourfulness scores. 9/

ERP pattern similarity analysis showed that encoding and remembering real objects resulted in earlier and more robust pattern stability than counterfeit objects. This may suggest that real objects result in quicker and more stable memory representations due to available visual-semantic templates. 8/

CDA is one way to look at neural engagement. But this only captures an average amplitude over predefined time windows. Another way we can examine neural engagement is to look at how the patterns of neural activity dynamically change over time. 7/

We also looked at underlying neural activities using EEG, specifically contralateral-delay-activity (CDA). The results showed increased neural engagement when remembering real than counterfeit objects, shown in both heightened and more spread-out lateralized ERP activities. 6/

Strikingly, despite this match VWM performance improved only for real objects, not counterfeit ones. Counterfeit objects resulted in similar performance to fully scrambled shapes. 5/

First, we tested whether these counterfeit objects are well matched in perceptual similarity with real objects. Using CNNs, we extracted visual features of each object and quantified similarities among them. This confirmed that counterfeit objects are indeed visually matched to real objects. 4/

Here, we tackled this using “counterfeit objects” (Cooper et al., 2023): GAN-generated images that match real objects in visual properties but are novel and unrecognizable. 3/

Oftentimes memory performance comparison is made across drastically different looking stimulus types such as real objects and colored circles or scrambled shapes. Specifically, real objects are visually more unique and discernible. Can these perceptual differences explain the memory benefit? 2/

New preprint with @SamJung @timbrady.bsky.social and @violastoermer.bsky.social: osf.io/preprints/ps.... Here we uncover what might be driving the “meaningfulness benefit” in visual working memory. Studies show that real objects are remembered better in VWM tasks than abstract stimuli. But why? 1/

🚨New paper altert🚨

As a synthesis of my PhD research, we revisited the prevailing assumption about the mechanisms underlying repetition learning, and re-evaluated these assumption in light of recent findings.

Now out in Perspectives on Psychological Science:
doi.org/10.1177/1745...

OpenWMData A collection of publicly available working memory datasets

Make it your New Year resolution to add a #workingmemory dataset to OpenWMData so that we can curate our field's precious data, start testing theories and benchmarking models across datasets, conduct secondary analyses and meta-research using the data itself, and help me feel like I'm, like, alive.

Has anyone attended any pre-data-collection poster sessions (i.e., poster sessions where people present their plans for experiments before data collection in order to get feedback when it's most useful) at conferences other than VSS?

Broadly, this suggests that the FFDE is relatively location-specific. Fun and quite shocking illusion to look at!

Our results consistently showed that the illusion was significantly disrupted with location shifts of the faces. We also look at how the illusion develops over time, something that hasn't been looked closely before.

FFDE is a fun illusion where peripherally presented faces start looking monstrous. Here we test whether this illusion can be transferred to new locations during the face streams. Importantly, we use joysticks so that people can continuously rate how weird the faces get, capturing the whole dynamics.

Testing location invariance of the flashed face distortion effect - Yong Hoon Chung, Nicole C. Anaya Sosa, Viola S. Störmer, 2025 Spatially aligned faces presented in a continuous stream in the periphery appear distorted and grotesque. This flashed face distortion effect (“FFDE”) was first...

New paper with NicoleAnayaSosa and @violastoermer.bsky.social now out in Perception! "Testing location invariance of the flashed face distortion effect" journals.sagepub.com/eprint/WVUFW...

Tomorrow afternoon I'll be presenting my symposium talk at #ESCoP2025 titled "Meaningful and familiar stimuli support visual working memory for simple features"! See you there!

We are now recruiting STEM mentors for the 2025-2026 graduate school application cycle!

⏰ Mentor applications close on July 31st, 2025 ⏰

✨ APPLY HERE: dashboard.project-short.com ✨

Questions? Email us: contact@project-short.com

#gradschool #phd #phdapplication #gradadmissions

And here is a preprint link: osf.io/preprints/ps...

New paper with LaurenWilliams, @timbrady.bsky.social, and @violastoermer.bsky.social now out in JEP: General! "Limits of verbal labels in cognition: Category labels do not improve visual working memory performance for obfuscated objects" psycnet.apa.org/record/2026-...

Today afternoon I’ll be presenting my poster at #vss2025 titled “Perceptual and conceptual contributions of the real-world object benefit in visual working memory: Is looking like an object good enough to enhance memory?” See you at Pavillion!

Overall, our results add to the emerging evidence that meaningful and familiar stimuli can enhance visual working memory processes and also show that remembering meaningful stimuli and simple features share core active cognitive processes.

Additionally, the meaningfulness benefits showed up even in the first five trials of the task in our results, showing how robust the effect is in visual working memory.

We also find that the amount of working memory increase individuals get from using meaningful stimuli also correlates with fluid intelligence scores, suggesting a link between meaningfulness benefit and fluid intelligence abilities.

In this paper we show that working memory performances for both real-world objects and colored circles reliably correlate with individual differences in fluid intelligence but not with crystallized intelligence measures.