Netherlands

This paper introduces a novel perspective to significantly mitigate catastrophic forgetting in continuous learning (CL), which emphasizes models&#39; capacity to preserve existing knowledge and assimilate new information. Current replay-based methods treat every task and data sample equally and thus can not fully exploit the potential of the replay buffer.
In response, we propose COgnitive REplay (CORE), which draws inspiration from human cognitive review processes. CORE includes two key strategies: Adaptive Quantity Allocation and Quality-Focused Data Selection. The former adaptively modulates the replay buffer allocation for each task based on its forgetting rate, while the latter guarantees the inclusion of representative data that best encapsulates the characteristics of each task within the buffer. Our approach achieves an average accuracy of 37.95\% on split-CIFAR10, surpassing the best baseline method by 6.52\%. Additionally, it significantly enhances the accuracy of the poorest-performing task by 6.30\% compared to the top baseline. Code is available at https://github.com/sterzhang/CORE.

**Authors:**

Jianshu Zhang: School of Cyber Science and Engineering; Yankai Fu: School of Cyber Science and Engineering; Ziheng Peng: School of Cyber Science and Engineering; Dongyu Yao: Wuhan University; Kun He: Wuhan University

CogSci 2024

CORE: Mitigating Catastrophic Forgetting in Continual Learning through Cognitive Replay

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. <br>
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) <br><br><br>
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>
<br><br><br><br>

To access this event page, you need to log in with the **email address you registered with**. <br>Access credentials will be sent to your email from Underline -  subject line "Welcome to ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

**Online Registration Form**: https://acl.swoogo.com/acl2026

Registration Required

ACL 2026

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

This paper introduces a novel perspective to significantly mitigate catastrophic forgetting in continuous learning (CL), which emphasizes models' capacity to preserve existing knowledge and assimilate new information. Current replay-based methods treat every task and data sample equally and thus can not fully exploit the potential of the replay buffer.
In response, we propose COgnitive REplay (CORE), which draws inspiration from human cognitive review processes. CORE includes two key strategies: Adaptive Quantity Allocation and Quality-Focused Data Selection. The former adaptively modulates the replay buffer allocation for each task based on its forgetting rate, while the latter guarantees the inclusion of representative data that best encapsulates the characteristics of each task within the buffer. Our approach achieves an average accuracy of 37.95\% on split-CIFAR10, surpassing the best baseline method by 6.52\%. Additionally, it significantly enhances the accuracy of the poorest-performing task by 6.30\% compared to the top baseline. Code is available at https://github.com/sterzhang/CORE.

**Authors:**

Jianshu Zhang: School of Cyber Science and Engineering; Yankai Fu: School of Cyber Science and Engineering; Ziheng Peng: School of Cyber Science and Engineering; Dongyu Yao: Wuhan University; Kun He: Wuhan University

technical paper

The 46th Annual Meeting of the Cognitive Science Society was an in-person meeting held in Rotterdam, The Netherlands at the Postillion Hotel & Conference Centre.

**ON-DEMAND PROGRAM ACCESS AFTER THE CONFERENCE**<br>
Recordings of the invited program are now available. You can access them by clicking on the 'Schedule' icon on the left. Select view by week and navigate to the day and time slot of the recording you wish to view. Click on the time slot to access the recording.

Recordings available (click on each recording to view):

**Thursday, July 25**<br>
0900: [Keynote Speaker Morgan Barense on Enhancing real-world event memory](https://underline.io/events/465/sessions/17997/lecture/99084-dynamics-between-minds-and-the-environment)<br>
1000: [Gleitman Award Winner Isabelle Dautriche on Language Foundations: Insights from acquisition, communication, cognition, and more](https://underline.io/events/465/sessions/18000/lecture/99085-gleitman-talk)<br>
1415: [Invited Symposium: Dynamics between minds and the environment](https://underline.io/events/465/sessions?eventSessionId=18016&searchGroup=lecture)<br>
1700: [Rumelhart Prize Presentation Speaker Alison Gopnik on Exploit, explore, empower: Three ages and three intelligences](https://underline.io/events/465/sessions/18028/lecture/99161-rp1-rumelhart-prize-presentation)<br>

**Friday, July 26**<br>
0900: [C.L. de Carvalho-Heineken Prize Keynote Speaker Kia Nobre on Focusing in memory](https://underline.io/events/465/sessions/18035/lecture/99162-hp1-c-l-de-carvalho-heineken-prize-keynote-address)<br>
1030: [Rumelhart Symposium: Childhood as exploration](https://underline.io/events/465/sessions/18038/lecture/100346-childhood-as-exploration)<br>
1415: [Elman Prize Symposium](https://underline.io/events/465/sessions?eventSessionId=18053&searchGroup=lecture)<br>
1600: [Invited Symposium: Dynamics between minds](https://underline.io/events/465/sessions?eventSessionId=18065&searchGroup=lecture)<br>
1745: [Keynote Speaker Andrea E. Martin on Neural dynamics encode the structure and statistics of language](https://underline.io/events/465/sessions/18077/lecture/99271-neural-dynamics-encode-the-structure-and-statistics-of-language)<br>

**Saturday, July 27**<br>
0900: [Keynote Speaker Gregor Schöner on How higher cognition emerges from the dynamics of strongly interacting neural populations](https://underline.io/events/465/sessions/18083/lecture/99273-dynamics-within-the-mind)<br>
1000: [Glushko Talks](https://underline.io/events/465/sessions?eventSessionId=18086&searchGroup=lecture)<br>
1500: [Invited Symposium: Dynamics within minds](https://underline.io/events/465/sessions?eventSessionId=18100&searchGroup=lecture)<br>


#### **CogSci 2024 Program Book**
<div style="position:relative;padding-top:0;width:900px;height:500px;"><iframe style="position:absolute;border:none;width:100%;height:100%;left:0;top:0;" src="https://online.fliphtml5.com/ebtyf/cble/"  seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe></div>


**The Program Book can be downloaded [here](https://drive.google.com/file/d/1yRVp1cBqRnAbdYi4GXWV_pZeyWNsRquv/view?usp=sharing).**

#### **Letter from the President**

Welcome to the 2024 Meeting of the Cognitive Science Society in Rotterdam! CogSci 2024 brings together a large community of cognitive scientists who have traveled here from around the world, as well as a smaller group of remote presenters. I would like to extend an especially warm welcome to our first-time CogSci meeting attendees, and hope that 
their experience inspires them to join our future meetings for many years to come.

I want to thank this year’s conference co-chairs, Larissa K Samuelson, Stefan Frank, Mariya Toneva, Allyson Mackey, and Eliot Hazeltine, for putting together an amazing series of invited talks and symposia around the theme of Dynamics of Cognition. The co-chairs also deserve credit for handling a record-breaking number of submissions and producing a truly exciting conference program 

Beyond the invited talks and symposia dedicated to this year’s conference theme, I would like to highlight the presentations and events honoring the 
Rumelhart, Elman, Gleitman and C .L . de Carvalho Heineken prize winners, including the Rumelhart reception on Thursday. As you may know, we are recording all of the invited talks and symposia, and all prize-winners’ talks, and will be making this content available after the conference. 

The Cognitive Science Society relies on the volunteer efforts of our Governing Board members who work on matters related to our membership; conference policies; diversity and inclusion, international, and outreach initiatives; prizes and much more. We welcome interest from Society members who would like to become more involved in what we do. The 
Society is deeply grateful to Rick Dale and Andrea Bender, who serve as editors of the two Society journals, Cognitive Science and Topics in Cognitive Science respectively. We especially acknowledge our editors’ generosity in donating their stipends each year to D&I and outreach efforts. 

This year’s Annual Meeting depends critically on the support of Marischal De Armond and his team at Podium Conference Specialists: Sharon Zwack, Cendrine De Vis, Sarah-Kate Burke, and Rachel Waller worked tirelessly to secure the venue and handled a multitude of logistical and organizational issues, from arranging coffee breaks and receptions to making sure that a huge number of posters fits comfortably into the available space. Finally, and importantly, I would like to acknowledge the Cognitive Science Society Executive Officer, Erica Wojcik, for managing the complex and ever expanding sphere of our Society’s activities.

I hope you enjoy this year’s meeting and the many different opportunities to engage with the vibrant community of cognitive scientists gathered here. I also hope you get a chance to explore the great city of Rotterdam with its many attractions, museums, restaurants and scenic views. If you’d like to talk more about our Society, or simply want to say hello, please stop by!

![](https://assets.underline.io/markdown_image/1/image/3c66d0e21fff30e204df1138ffd15173.jpeg)

**Anna Papafragou**  
President, Cognitive Science 
Society 2023-2024

**CODE OF CONDUCT**<br>
By attending the CogSci 2024 Conference, you are required to adhere to the society’s [Code of Conduct](https://drive.google.com/file/d/16-6KkptF0Gn3ZYGJDlpqpwTdImw45Ng0/view?usp=drive_link).

**ABOUT THE COGNITIVE SCIENCE SOCIETY**<br>
The Cognitive Science Society brings together researchers from around the world who hold a common goal: understanding the nature of the human mind. The mission of the Society is to promote Cognitive Science as a discipline, and to foster scientific interchange among researchers in various areas of study, including Artificial Intelligence, Linguistics, Anthropology, Psychology, Neuroscience, Philosophy, and Education.

The Society is a non-profit professional organization and its activities include sponsoring an annual conference and publishing the journals Cognitive Science and TopiCS.
<br><br>

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

It looks like you are not registered for this event. 

To access the site please register [**here**](https://cognitivesciencesociety.org/registration/). 

Please register!

The 46th Annual Meeting of the Cognitive Science Society presents the latest research across cognitive science and highlights the theme of Cognition in Context.

Human learning is sensitive to rule-like structure and the curriculum of examples used for training. In tasks governed by succinct rules, learning is more robust when related examples are blocked across trials, but in the absence of such rules, interleaving is more effective. To date, no neural model has simultaneously captured these seemingly contradictory effects. Here we show that this same tradeoff spontaneously emerges with “in-context learning” (ICL) both in neural networks trained with metalearning and in large language models (LLMs). ICL is the ability to learn new tasks “in context” — without weight changes — via an inner-loop algorithm implemented in activation dynamics. Experiments with pretrained LLMs and metalearning transformers show that ICL exhibits the blocking advantage demonstrated in humans on a task involving rule-like structure, and conversely, that concurrent in-weight learning reproduces the interleaving advantage observed in humans on tasks lacking such structure.

**Authors:**

Jacob Russin: Brown; Ellie Pavlick: Brown University; Michael J. Frank: Brown University

Human Curriculum Effects Emerge with In-Context Learning in Neural Networks

This study investigates cue-based memory retrieval during sentence processing. Cue-based retrieval theories argue that the parser uses lexical and structural information as retrieval cues to retrieve items from memory. Evidence for cue-based memory retrieval comes from research showing that non-target representations matching retrieval cues interfere with target retrieval. However, the degree of susceptibility to this similarity-based interference has been debated, having led to the development of different computational models. This study focuses on two cue-based models and tests their predictions in two experiments. The results suggested similarity-based interference, but its patterns were not fully compatible with these models. To reconcile these findings within the framework of cue-based memory retrieval, this paper presents a model that assigns substantial weight to the structure-based cue and incorporates the notions of initial retrieval and revision. Results from simulations indicate that the model incorporating these assumptions provides a better fit to the observed data.

**Authors:**

Hiroki Fujita: University of Potsdam

Memory Retrieval Processes during Real-Time Language Comprehension: Empirical Evidence and Computational Modelling

One hallmark of human language is its combinatoriality---reusing a relatively small inventory of building blocks to create a far larger inventory of increasingly complex structures. 
In this paper, we explore the idea that combinatoriality in language reflects a human inductive bias toward representational efficiency in symbol systems. We develop a computational framework for discovering structure in a writing system. Built on top of state-of-the-art library learning and program synthesis techniques, our computational framework discovers known linguistic structures in the Chinese writing system and reveals how the system evolves towards simplification under pressures for representational efficiency. We demonstrate how a library learning approach, utilizing learned abstractions and compression, may help reveal the fundamental computational principles that underlie the creation of combinatorial structures in human cognition, and offer broader insights into the evolution of efficient communication systems.

**Authors:**

Guangyuan Jiang: Massachusetts Institute of Technology; Matthias Hofer: MIT; Jiayuan Mao: Massachusetts Institute of Technology; Lionel Wong: Massachusetts Institute of Technology; Josh Tenenbaum: MIT; Roger Levy: Massachusetts Institute of Technology

Finding structure in logographic writing with library learning

We examine the relation between event segmentation in language and cognition in the domain of motion events, focusing on Turkish, a verb-framed language that segments motion paths in separate linguistic units (verb clauses). We compare motion events that have a path change to those that did not have a path change. In the linguistic task, participants were more likely to use multiple verb phrases when describing events that had a path change compared to those that did not have a path change. In the non-linguistic Dwell Time task, participants viewed self-paced slideshows of still images sampled from the motion event videos in the linguistic task. Dwell times for slides corresponding to path changes were not significantly longer than those for temporally similar slides in the events without a path change. These findings suggest that event units in language may not have strong and stable influences on event segmentation in cognition.

**Authors:**

Bilge Tınaz: Özyeğin University; Ercenur Ünal: Özyeğin University

Event Segmentation in Language and Cognition

Eating onion ice cream is improbable, and levitating ice cream is impossible. But scooping ice cream using sadness is not just impossible: it is inconceivable. While prior work has examined the distinction between improbable and impossible events, there has been little empirical research on inconceivability. Here, we report a behavioral and computational study of inconceivability in three parts. First, we find that humans reliably categorize events as inconceivable, separate from probable, improbable, and impossible. Second, we find that we can decode the modal category of a sentence using language-model-derived estimates of subjective event probabilities. Third, we reproduce a recent finding that improbable events yield slowest response times in a possibility judgment task, and show that inconceivable events are faster to judge than impossible and improbable events. Overall, our results suggest that people distinguish the impossible from the inconceivable, and such distinctions may be based on graded rather than discrete judgments.

**Authors:**

Jennifer Hu: Harvard University; Felix Anthony Sosa: Harvard University; Tomer D. Ullman: Harvard University

Shades of Zero: Distinguishing impossibility from inconceivability

Telicity is an important semantic feature pointing to event construal: telic verb phrases denote bounded events with an inherent endpoint while atelic verb phrases denote unbounded events without such an endpoint. Languages encode telicity in different ways. Unlike English, Mandarin lacks an overt count-mass distinction and allows bare noun objects to form verb phrases. Would this cross-linguistic difference influence event perception? Experiment 1 elicited descriptions of bounded vs. unbounded events from English and Mandarin native speakers. A clear cross-linguistic difference was found: English speakers mostly used telic predicates for bounded events and atelic predicates for unbounded events while Mandarin speakers gave atelic predicates with bare noun objects for both event types. Experiment 2 explored how English and Mandarin speakers tracked the temporal structure of bounded vs. unbounded events. The two language groups performed similarly. The way people describe events may not affect the way they track event temporal profiles.

**Authors:**

Yue Ji: Beijing Institute of Technology; Anna Papafragou: Unversity of Pennsylvania

Do Cross-Linguistic Differences Influence Event Perception?

Daily experiences are conceptualized as events involving multiple participants and their relations (i.e., thematic roles). When describing events, speakers often do not include all event participants involved. Here, we explore how underlying conceptual requirements and language-specific encoding options influence the content of event descriptions in speech and gesture in two typologically different languages (English, Turkish). Focusing on conceptually peripheral instruments whose status is highly debated, we manipulated the conceptual status of event participants by including events that ‘require’ or ‘allow’ otherwise syntactically optional instruments. Results showed that the require-allow distinction did not manifest uniformly in Turkish and English in speech, gesture, or when both modalities were considered. However, mention of highly optional event participants (e.g., allowed instruments) was affected by language-specific syntactic encoding options. We conclude that, under more naturalistic elicitation conditions, planning descriptions of instrument events is more heavily affected by language-specific encoding than conceptual prominence of the roles.

**Authors:**

Christiana Moser: University of Toronto; Bahar Tarakçı: Özyeğin University; Ercenur Ünal: Özyeğin University; Myrto Grigoroglou: University of Toronto

Multimodal Description of Instrument Events in Turkish and English

As AI systems like language models are increasingly integrated into decision-making processes affecting people's lives, it's critical to ensure that these systems have sound moral reasoning. To test whether they do, we need to develop systematic evaluations. We provide a framework that uses a language model to translate causal graphs that capture key aspects of moral dilemmas into prompt templates. With this framework, we procedurally generated a large and diverse set of moral dilemmas---the OffTheRails benchmark---consisting of 50 scenarios and 400 unique test items. We collected moral permissibility and intention judgments from human participants for a subset of our items and compared these judgments to those from two language models (GPT-4 and Claude-2) across eight conditions. We find that moral dilemmas in which the harm is a necessary means (as compared to a side effect) resulted in lower permissibility and higher intention ratings for both participants and language models. The same pattern was observed for evitable versus inevitable harmful outcomes. However, there was no clear effect of whether the harm resulted from an agent's action versus from having omitted to act. We discuss limitations of our prompt generation pipeline and opportunities for improving scenarios to increase the strength of experimental effects.

**Authors:**

Jan-Philipp Fränken: Stanford University; Kanishk Gandhi: Stanford University; Tori Qiu: Stanford University; Ayesha Khawaja: Stanford University; Noah Goodman: Stanford University; Tobias Gerstenberg: Stanford University

Procedural Dilemma Generation for Moral Reasoning in Humans and Language Models

Moral inference is an emerging topic of critical importance in artificial intelligence. The contemporary approach often relies on language modelling to infer moral relevance or moral properties of a concept such as "smoking". This approach  demands complex parameterisation and costly computation, and it tends to disconnect with psychological accounts of moralization. We present a simple cognitive model for moral inference grounded in theories of moralization. Our model  builds on word association network known to  capture  human semantics and draws on rich psychological data. We demonstrate that our moral association graph model performs competitively to state-of-the-art language models, where we evaluate them against a comprehensive set of data for  automated inference of moral norms and moral judgment of concepts, and in-context moral inference. Moreover, we show that our model discovers intuitive concepts underlying  moral judgment and is applicable to informing short-term temporal changes in moral perception.

**Authors:**

Aida Ramezani: University of Toronto; Yang Xu: University of Toronto

Moral association graph: A cognitive model for moral inference

Models based on the Transformer neural network architecture have seen success on a wide variety of tasks that appear to require complex ``cognitive branching''-- or the ability to maintain pursuit of one goal while accomplishing others. In cognitive neuroscience, success on such tasks is thought to rely on sophisticated frontostriatal mechanisms for selective gating, which enable role-addressable updating-- and later readout-- of information to and from distinct ``addresses'' of memory, in the form of clusters of neurons. However, Transformer models have no such mechanisms intentionally built-in. It is thus an open question how Transformers solve such tasks, and whether the mechanisms that emerge to help them to do so bear any resemblance to the gating mechanisms in the human brain. In this work, we analyze the mechanisms that emerge within a vanilla attention-only Transformer trained on a simple sequence modeling task inspired by a task explicitly designed to study working memory gating in computational cognitive neuroscience. We find that, as a result of training, the self-attention mechanism within the Transformer specializes in a way that mirrors the input and output gating mechanisms which were explicitly incorporated into earlier, more biologically-inspired architectures. These results suggest opportunities for future research on computational similarities between modern AI architectures and models of the human brain.

**Authors:**

Aaron Traylor: Brown University; Jack Merullo: Brown University; Michael J. Frank: Brown University; Ellie Pavlick: Brown University

Premium content

Downloads

Next from CogSci 2024

Human Curriculum Effects Emerge with In-Context Learning in Neural Networks

Similar lecture

FOREVER: Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES