United States

This is the first kinematic investigation of articulator motion in Austrian Sign Language, which connects kinesiology of sign production and linguistic markers of Aktionsart in the native language of the Deaf community in Austria. Our work used a 3D motion capture approach to sign language analysis to investigate the relationship between the semantics (event structure) of signed verbs, and kinematics of hand articulator movement. The data indicates that the underlying semantics of events in verb signs is reflected in sign duration and acceleration of the dominant hand during sign production. The finding that articulator dynamics (acceleration and deceleration of hand motion) characterizes the event structure in verb signs has significance for linguistic theory of visual communication, and understanding of the relationship between iconicity in sign language, and perceptual biases in meaning construction based on visual input.

CogSci 2021

Event visibility in sign language motion: Evidence from Austrian Sign Language (ÖGS)

### Welcome to IJCNLP-AACL 2025! 
<br>It is a great honor to host this joint conference in Mumbai, India, from December 20 to 24, 2025. The joint conferences of IJCNLP and AACL are organized with alternating leadership in the Asia-Pacific region. The event is run by the Asian Federation of Natural Language Processing (AFNLP) in odd years, and by AACL in even years, while it is organized solely by ACL when the annual ACL meeting is held in the region. This year, the conference is primarily organized by AFNLP.<br>
*Kentaro Inui
MBZUAI, UAE
General Chair, IJCNLP-AACL 2025*<br>
Read full message and download the Conference Handbook [**here**](https://drive.google.com/file/d/1UTwxkAqSqI-GAoJC3wE1zZt5VP1Y8GX0/view?usp=sharing).
<br><br><br><br>

IJCNLP-AACL 2025

The 14th IJCNLP & 4th AACL will be held in Mumbai, India from December 20th to December 24th, 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.

A simple memory component is amended to local (“Pursuit”; Stevens,  Gleitman,  Trueswell, and Yang (2017)) and  global (e.g., Yu  and  Smith  (2007);  Fazly,  Alishahi,  and  Stevenson(2010)) models of cross-situational word learning. Only a finite (and small) number of words can be concurrently learned; successfully  learned  words  are  removed  from  the  memory buffer  and  stored  in  the  lexicon. The  memory  buffer  improves the empirical coverage for both local and global learn-ing models.  However, the complex task of homophone learn-ing (Yurovsky & Yu, 2008) proves a more decisive advantage for the local model (dubbed Memory Bound Pursuit;  MBP). Implications and limitations of these results are discussed.

Memory Constraints on Cross Situational Word Learning

After learning that one stimulus predicts an outcome (e.g., an aqua-colored rectangle leads to shock) and a very similar stimulus predicts no outcome (e.g., a slightly greener rectangle leads to no shock), some participants generalize the predictive relationship on the basis of physical similarity to the predictive stimulus, while others generalize on the basis of the relational difference between the two stimuli (e.g., “higher likelihood of shock for bluer stimuli”). To date, these individual differences in generalization rules have remained unexplored in associative learning. Here, we present evidence that a given individual simultaneously entertains belief in both “similarity” and “relational” rules, and generalizes using a mixture of these strategies. Using a “mixture of experts” modelling framework constrained by participants self-reported rule beliefs, we show that considering multiple rules predicts generalization gradients better than a single rule, and that generalization behavior is better described as switching between, rather than averaging over, different rules.

A Mixture of Experts in Associative Generalization

Many studies have demonstrated that testing students on to-be-learned materials can be an effective learning activity. However, past studies have also shown that some practice test formats are more effective than others. Open-ended recall or short answer practice tests may be effective because the questions prompt deeper processing as students must generate an answer. With closed-ended testing formats such as multiple-choice or true-false tests, there are concerns that they may prompt only superficial processing, and that any benefits will not extend to non-practiced information or over time. They also may not be effective for improving comprehension from text as measured by how-and-why questions. The present study explored the utility of practice tests with closed-ended questions to improve learning from text. Results showed closed-ended practice testing can lead to benefits even when the learning outcome was comprehension of text.

Can Closed-ended Practice Tests Promote Understanding from Text?

Across the world's languages, children reliably learn nouns more easily than verbs. Attempts to understand the difficulty of verb learning have focused on determining whether the challenge stems from differences in the linguistic usage of nouns and verbs, or instead conceptual differences in the categories that they label. We introduce a novel metric to quantify the contributions of both sources of difficulty using unsupervised learning models trained on corpora of language and images. We find that there is less alignment between the linguistic usage of verbs and their categories than for nouns and their categories. However, this difference is driven almost entirely by differences in the structure of their visual categories: Relative to nouns, events described by the same verb are more variable and events described by two different verbs are more similar. We conclude that differences between noun and verb learning need not be due to fundamental differences in learning processes, but may instead be driven by the difficulty of one-shot generalization from verbs' visual categories.

A common framework for quantifying the learnability of nouns and verbs

People learn by both decomposing and combining concepts; most accounts of combination are either compositional or conjunctive. We augment the DORA model of representation learning to build new predicate representation by combining (or compressing) existing predicate representations (e.g., building a predicate a_b by combining predicates a and b). The resulting model learns structured relational representations from experience and then combines these relational concepts to form more complex, compressed concepts. We show that the resulting model provides an account of a category learning experiment in which categories are defined as novel combinations of relational concepts.

Compression: A Lossless Mechanism for Learning Complex Structured Relational Representations

Despite being designed for performance rather than cognitive plausibility, transformer language models have been found to be better at predicting metrics used to assess human language comprehension than language models with other architectures, such as recurrent neural networks. Based on how well they predict the N400, a neural signal associated with processing difficulty, we propose and provide evidence for one possible explanation—their predictions are affected by the preceding context in a way analogous to the effect of semantic facilitation in humans.

Different kinds of cognitive plausibility: why are transformers better than RNNs at predicting N400 amplitude?

Relational schema induction involves a series of learning tasks conforming to a common (group-like) structure. The paradigm contrasts associative versus relational aspects of learning for cognitive, developmental and comparative psychology. Yet, a theory accounting for the relationship between these forms of learning has not been fully developed. We use (mathematical) category theory methods to redress this situation: both forms of learning involve a (universal) construction that differs in terms of ``dimensionality'', i.e. one-dimensional (associative) versus two-dimensional (relational). Accordingly, the development of relational learning pertains to changes in the dimensionality of the underlying relational schemas induced.

A formal comparison/contrast of associative and relational learning: a case study of relational schema induction

Audio‐visual (AV) associative learning is central to many aspects of cognitive development and is key in reading acquisition. Most studies thus far have examined AV associative learning involving linguistic stimuli. Yet it is of importance to examine cross-modal learning free of familiarity confounds. We, therefore, designed an AV learning paradigm relying on novel, non-linguistic auditory and visual stimuli, which were both unfamiliar to participants. On top of AV learning, we collected performance in reading-related abilities, as well as in more domain-general skills, in a population of healthy Italian-speaking adults (N=57). By fitting trial-by-trial performance in our novel learning task, we demonstrate the expected variability in speed of learning (learning rate) across participants. We then show that speed of learning in our novel learning task is positively associated with working memory and replicate this result in a set of French-speaking participants (N=32), showing that it holds in another language.

A novel non-linguistic audio-visual learning paradigm to test the cognitive correlates of learning rate

Current models of cognitive control address selection among tasks in terms of a cost-benefit tradeoff. Importantly, they usually assume a fixed level of competence for each candidate task when estimating its value. However, performing a task can improve competence through learning, which should be factored into estimates of future value. Here, we consider an extension of the Expected Value of Control (EVC) theory that includes such estimates. We demonstrate that control allocation is a function of task learnability, and show the use of this model by generating novel predictions in cognitive effort discounting tasks. We argue that the value of learning in control allocation may account for the seemingly paradoxical finding that sometimes participants prefer more difficult (i.e., costly) tasks, and discuss how the model can be leveraged to further our understanding of human decision making and cognitive impairments.

The Value of Learning and Cognitive Control Allocation

Does knowing certain words help children learn other words? We hypothesized that knowledge of more general (more superordinate) words at time1 would lead to faster vocabulary growth as measured through vocabulary checklists administered at later timepoints. We find that this is indeed the case. Children who have similar vocabularies at time1 , but differ in their productive knowledge of more general words such as “animal,” “picture,” and “get” go on to have different rates of word learning. Knowledge of more general words is associated with faster vocabulary growth, particularly of words semantically related to the superordinate terms they are reported to produce. This positive relationship between knowledge of more general words and word learning remains even when controlling for measures of verbal and nonverbal intelligence.

Downloads

Next from CogSci 2021

Memory Constraints on Cross Situational Word Learning

Similar lecture

Pose-Based Sign Language Spotting via an End-to-End Encoder Architecture

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

Memory Constraints on Cross Situational Word Learning

Similar lecture

Pose-Based Sign Language Spotting via an End-to-End Encoder Architecture

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads