United States

Chunks allow us to use long-term knowledge to efficiently represent the world in working memory. Most views of chunking assume that when we use chunks, this results in the loss of specific perceptual details, since it is presumed the contents of chunks are decoded from long-term memory rather than reflecting the exact details of the item that was presented. However, in two experiments, we find that in situations where participants make use of chunks to improve visual working memory, access to instance-specific perceptual detail (that cannot be retrieved from long-term memory) increased, rather than decreased. This supports an alternative view: that chunks facilitate the encoding and retention into memory of perceptual details as part of structured, hierarchical memories, rather than serving as mere “content-free” pointers. It also provides a strong contrast to accounts in which working memory capacity is assumed to be exhaustively described by the number of chunks remembered.

CogSci 2021

Chunks are not “Content-Free”: Hierarchical Representations Preserve Perceptual Detail within Chunks

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair:<br>
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for  Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025.  ACL2025continues our field’s 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.*<br>
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)<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 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

ACL 2025

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

technical paper

### Looking for Cogsci 2023? Please [click here](https://underline.io/events/405/reception) to switch to 2023 edition.


**Dear Colleagues,**

Welcome to the 43rd Annual Conference of the Cognitive Science Society! 

This year’s conference brings together the latest research from all Cognitive Science and also highlights this year’s theme **Comparative Cognition: Animal Minds.** 

The program features four plenary speakers: Friederike Range (University of Veterinary Medicine, Vienna), Onur Güntürkün (Ruhr University Bochum), Julia Fischer (University of Göttingen), and Erich Jarvis (The Rockefeller University). The keynotes and invited symposia showcase the broad spectrum of ideas that encompass the conference theme of comparative cognition. 

Further, the program includes the Rumelhart Prize keynote presentation by Susan Goldin-Meadow (University of Chicago), and symposia to celebrate the winner of the Elman Prize: Wei Ji Ma (New York University), the Heineken Award winner Robert Zatorre (McGill University), and the Robert J. Glushko Dissertation Prizes session.

The program committee for CogSci 2021 received 1,025 submissions, including 710 full papers, 275 member abstracts, 8 publication-based talks, as well as 15 proposals for symposia, and 17 for workshops and tutorials. After a rigorous review process, the committee selected 152 papers for oral presentation and inclusion in the conference proceedings (14.8%), 332 papers for poster presentation and inclusion in the proceedings (32.4%), and 175 papers for poster presentation with inclusion of abstracts in the proceedings (17.1%) We also selected 275 submitted member abstracts. In addition, we accepted 6 publication-based talks, 12 symposia, and 9 workshops and tutorials to make for an exciting and inclusive program.

We hope that you enjoy the conference this year!<br><br>

**Your Program Chairs,**
![](https://assets.underline.io/uploads/markdown_image/1/image/99984908caba5e6453e5fc5ba34f3507.jpg)

**Learn more about the Vienna Cognitive Science Hub at the University of Vienna, hosts of CogSci 2021**

<iframe width="560" height="315" src="https://www.youtube.com/embed/464h1m3iU68" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

To review the proceedings for the conference or review the searchable program, please visit the [website](https://cognitivesciencesociety.org/program/).


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.

Human cognition can tackle a wide range of problems, producing scalable results within a manageable timeframe. Many cognitive models can predict human behavior but lack such scalability, resulting in rapidly increasing processing times for more complex inputs. We present a task where participants mentally sort sequences of rectangles by size while we measure reaction times (RTs) and accuracy. By manipulating the size of the input and the presence of latent structure in the sequences, we investigate i) how mental sorting scales with input complexity, ii) how latent structure influences scaling, and iii) how mental computations can be captured by plausible cognitive models. Our results reveal RTs scale linearly with sequence length, and participants can learn and actively use latent structure to sort faster. This behavior is in line with a noisy sorting algorithm, which sequentially rules out potential hypotheses about the latent structure, thus reducing complexity while retaining accuracy.

How does mental sorting scale?

Planning and organization of one's actions are crucial for successfully performing everyday activities such as setting the table. While existing research has addressed planning for well-defined tasks and control of already established sequences, little is known about how such sequences are planned in ill-defined tasks such as everyday activities. Initial attempts suggest that planning may be opportunistic, based on a number of environmental factors to minimize cognitive and physical effort. We address two questions arising from the existing work: First, to what extent is variation in human everyday activity behavior captured by the proposed opportunistic consideration of environmental factors? We address this questions by employing machine learning baselines to gauge the proposed models explanatory scope. Second, to what extent are existing models of sequence control consistent with opportunistic action organization? We address this by investigating and discussing the implications opportunistic planning has for the mechanisms currently assumed for sequence control.

Planning and Action Organization in Ill-Defined Tasks: The Case of Everyday Activities

Knowing that something is unknown is an important part of human cognition. While Bayesian models of cognition have been successful in explaining many aspects of human learning, current explanations of how humans realise that they need to introduce a new concept are problematic. Bayesian models lack a principled way to describe ignorance, as doing so requires comparing the probabilities of concepts in the model with the probabilities of concepts not present in the model, which is by definition impossible. Formal definitions of uncertainty (e.g. Shannon-entropy) are commonly used as a substitute for ignorance, but we will show that these concepts are fundamentally distinct, and thus that something more is needed. Enhancing probability theory to allow Bayesian agents to conclude that they are ignorant would be an important advance for both cognitive engineering and cognitive science. In this research project, we formally analyse this challenge.

Certainly Strange: A Probabilistic Perspective on Ignorance

Reading comprehension research has identified sources of children’s difficulty with inference-making: lack of semantic/content knowledge and logical reasoning difficulty. NLP tools modeling semantic knowledge (e.g. BERT) can predict adult inference-making, but it is unclear whether they can predict children’s inference-making difficulty. In our ongoing study, we will examine whether our new inference difficulty metric can predict kindergarten students’ inference-making, using empirical data from a classroom intervention (ELCII). Students were given verbal information on a topic and multiple-choice questions, which require students to draw an inference from two given scaffolds. To develop this metric, we will train BERT on children’s books and ELCII content to compute an additive inference vector, the sum of the two vectorized scaffolds. The cosine distance between the additive and correct inferences may indicate inference difficulty. Results will indicate whether a probabilistic semantic space can model children’s inferences or if other components (e.g. logic) should be considered.

A metric of children’s inference-making difficulty during language comprehension

Co-speech gestures can enhance a listener’s understanding of a spoken message, yet children show a greater benefit of gesture than adults (Hostetter, 2011). We explore whether this effect may be driven by language proficiency, and in turn, by differences in how children visually attend to gesture when they are processing narratives in their stronger vs. weaker language. Bilingual children were shown narratives with scripted gesture in their stronger and weaker languages, while their visual attention was monitored. Memory for narratives was then assessed. Our findings suggest language proficiency does affect the degree to which children benefit from co-speech gesture – children showed a greater boost from co-speech gesture when processing narrative in their weaker language. Results also suggest that greater attention to gesture when processing ones’ weaker language may underlie this effect.

Language Proficiency Impacts the Benefits of Co-Speech Gesture for Narrative Understanding Through a Visual Attention Mechanism

Symbolic representations of number are instrumental to mathematical reasoning and many aspects of social organization. What explains their emergence in human cultures? To understand how functional and cognitive constraints impact people’s communication about number, we used a drawing-based reference game to investigate how human dyads coordinated to form novel number systems. We found a systematic bias towards symbols exploiting 1-to-1 correspondence to objects in visual arrays, and that this strategy was contingent on the communicative relevance of number. Moreover, the meaning of these symbols was transparent to third party observers not present during their production. Finally, model-based analyses of these symbols' visual properties suggest that the ability to decode exact quantity from them may rely on perceptual processing mechanisms beyond those sufficient for object recognition. These findings contribute to our understanding of how both communicative need and capacity for visual abstraction constrain the emergence of iconic representations of exact number.

Improvised Numerals Rely on 1-to-1 Correspondence

Categories can be learned through different strategies. Sometimes we may use abstract rules to categorize objects, and other times we may rely on the perceptual similarity among stimuli. The ability to categorize objects based on a common pattern develops since early childhood and exhibits systematic age differences. Numerous studies demonstrated that younger children rely on similarity-based processes, while older children employ rule-based categorization strategies (Miles & Minda, 2009; Rabi & Minda, 2014; Deng & Sloutsky, 2016). We used a model-based approach to investigate individual differences in category learning in pre-school children (6 years old) and primary school children (6-8 and 10-11 years old). Our results suggest that older children were more likely to employ a rule-based categorization strategy and demonstrated better learning outcomes. Lastly, we employed several computational models of categorization to uncover the properties of the process that may best account for the obtained results.

Category learning in preschool and primary school children: The use of rule-based and similarity-based strategies

Early vocabulary development is heavily influenced by children’s language environment. For example, higher SES children who have frequent and rich interactions with caregivers tend to have larger vocabulary size than lower SES children who have less stable home environment (e.g., Hoff & Naigles, 2002). Because of the recent pandemic, most Canadian children have experienced dramatic changes to their day-to-day life. In the current study, we ask whether the pandemic environment affects children’s vocabulary growth. We compare the MacArthur-Bates Communicative Development Inventory (CDI) scores of 11- to 34-month-olds collected before (N=1365) and after (N=301) the onset of COVID-19. Preliminary results show that CDI scores collected for toddlers after the pandemic onset are significantly lower than those collected before. We will continue to collect additional CDI data, and examine how factors such as screen time, reading time, SES, and number of people the child has regular interaction with mediate children’s vocabulary size.

Vocabulary Growth in Infancy and Toddlerhood: The Impact of COVID-19

Data interpretation is crucial in modern society. One common data structure that people frequently encounter is 2 x 2 tables. Past work suggests that the nature of the variables affects how people interpret 2 x 2 tables. Specifically, people interpret tables with symmetric variables (present/present; e.g., treatment A vs. treatment B) more accurately than tables with asymmetric variables (present/absent; e.g., treatment vs. no treatment). This study tested whether interpreting tables with symmetric variables could scaffold later interpretation of tables with asymmetric variables. Undergraduates interpreted tables and rated the importance of each cell to their interpretations. Some participants interpreted tables with symmetric variables before tables with asymmetric variables; others interpreted only tables with asymmetric variables. Participants who first interpreted tables with symmetric variables later judged cells in the bottom row of asymmetric tables to be more important. Thus, experience with symmetric variables shifted participants’ views of tables with asymmetric variables.

Interpreting Data Tables: Can Variable Symmetry Scaffold Performance?

Compositionality is a core feature of human cognition and behavior. People readily decompose visual objects into parts and complex procedures into subtasks. Here we investigate how these two abilities interact to support learning in a block-tower assembly experiment. We measured the way participants segmented these towers based on shape information alone, and asked how well the resulting parts explained the procedures other participants used to build them. We found that people decomposed these shapes in consistent ways and the most common parts appeared especially frequently as subroutines in the assembly experiment. Moreover, we found that the subroutines participants used converged over time, reflecting shared biases toward certain ways of reconstructing each tower. More broadly, our findings suggest important similarities between the perceptual and procedural abstractions humans use to perceive and interact with objects in their environment.

Downloads

Next from CogSci 2021

How does mental sorting scale?

Similar lecture

Dynamic Chunking and Selection for Reading Comprehension of Ultra-Long Context in Large Language Models

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from CogSci 2021

How does mental sorting scale?

Similar lecture

Dynamic Chunking and Selection for Reading Comprehension of Ultra-Long Context in Large Language Models

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads