CCN 2026 | Learning & Memory (CT)

CCN 2026 | Learning & Memory (CT)

by Cognitive Computational Neuroscience

Life & Natural Systems Neuroscience PSANNeurosciencesPSAN5Cognitive and behavioural neuroscience
🎙 Cognitive Computational Neuroscience 👥 4K 📅 August 12, 2026 ⏱ 60 min 👁 104 📄 conference talk 🧭 2026-08-15
Available in: English (current) Français

Keywords

memorabilityconceptual diversityspatial structureepisodic memoryeye tracking

Summary

This video is a recording of the Learning & Memory contributed talk session at the 9th Annual Conference on Cognitive Computational Neuroscience (CCN 2026). It features four talks. The first talk by Dyllan Simpson investigates what makes categories memorable, finding that within-category conceptual diversity is a stronger predictor of memorability than perceptual diversity or deep neural network embeddings. The second talk by Michelle B. Hefner explores how spatial structure facilitates category learning, showing that learning objects in a simple 1D layout leads to better categorization performance than a more complex interdigitated layout, and that this effect is reproduced by a neural network model with rich learning. The third talk by Jonathan Nicholas examines eye movements during deliberation, showing that people look at the encoding locations of items relevant to their decisions, and proposes a resource-rational model where looking at nothing reflects optimal sampling from episodic memory. The fourth talk by Abe Leite integrates vision and language in long-term memory metamers, and the fifth by Gal Vishne decomposes iconic memory decay with a joint-feature model. The session also includes a talk by Sze Chai Kwok on memory subspaces in the monkey precuneus. The talks are technical and aimed at a specialized audience.

201 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talks present novel empirical findings and computational models. Simpson’s talk provides strong evidence for the role of conceptual diversity in memorability, using a large dataset and careful control of perceptual factors. Hefner’s talk offers a clear behavioral effect and a computational model that partially explains it, though the model lacks the response congruency effect. Nicholas’s talk presents a compelling normative model of eye movements during memory-based decisions, supported by eye-tracking data. The argumentation is generally rigorous, with appropriate caveats and ongoing work acknowledged. However, as conference talks, the depth is limited and some details are omitted.

Scientific Rigor, Source Quality, Title Accuracy

The talks are based on research that is presumably peer-reviewed, but no specific sources are cited in the video. The description provides a link to the conference page for the session, which likely contains abstracts and possibly papers. The title accurately reflects the content. The video is a recording of a scientific session, so the quality is high, but the lack of explicit citations within the video limits the ability to verify claims directly.

186 words

Title / Content Match

The title accurately reflects the content: a session on learning and memory at CCN 2026, featuring contributed talks.

Quality & Reliability

8/10

The video presents peer-reviewed research from a reputable conference (CCN). The speakers are researchers from recognized institutions. The content is technical and based on experimental data and computational models. However, as a conference recording, it lacks the depth of a full paper and the claims are not independently verified in this format.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

Concurring Sources

Contribution & Novelties

The session presents several novel contributions: Simpson’s work introduces a new measure of category memorability based on conceptual diversity, showing its superiority over perceptual diversity and DNN embeddings. Hefner’s study provides evidence that spatial structure influences category learning through structured representations, with a computational model that partially reproduces the effect. Nicholas’s research offers a normative account of eye movements during memory-based decisions, linking looking at nothing to optimal sampling from episodic memory. These findings advance our understanding of memory and learning processes.

Pour aller plus loin :

  • Representational similarity analysis — A method used to compare neural representations across conditions and models.
  • Episodic memory — The memory system for specific events, central to Nicholas’s talk.
  • Resource-rational analysis — A framework for understanding cognition as optimal under constraints, relevant to Nicholas’s model.
  • Category learning — The process of acquiring categories, relevant to Hefner’s talk.
  • Deep neural networks in vision — Models like VGG and CLIP used in Simpson’s study.

158 words

Radar Profile

The radar profile shows high scores across all dimensions, indicating a technically rigorous and informative session. The lowest score is in 'fiabilite_globale' due to the lack of explicit citations, but overall the content is reliable and well-presented.

Reliability 8/10