United States

CogSci 2025

Do Young Children Learn Words from the Company they Keep?

statistical learning

corpus studies

psychology

language acquisition

natural language processing

# Welcome everyone to ACL 2022!

The 60th Annual Meeting of the Association for Computational Linguistics is taking place May 22-27, 2022 as a hybrid event, in Dublin and online. We are happy to welcome all of you to this anniversary edition with an almost 50-50 in-person and virtual participation. 
The main conference program features oral presentations, in-person and virtual posters and demo sessions, a plenary session for our best paper presentations and awards, three amazing keynote events and two new initiatives of invited talks: Spotlight Talks for Young Rising Stars (STIRS) and The Next Big Idea Talks. Posters (including Findings of ACL 2022) and demos are grouped by areas for both the in-person and the virtual sessions. For the virtual component, the talks will be on Zoom and the posters and the demos will be in GatherTown. The Student Research Workshop will have an oral session and a poster session as part of Poster Session 1. The program also features eight Tutorials and 28 Workshops. 

 
We wish you a wonderful conference! 
[**The ACL 2022 Organizing Committee**](https://www.2022.aclweb.org/organisers)
 
[**Conference Handbook**](https://drive.google.com/file/d/1_BUCMfhMVrjG9E2e71aHdHeE28KSje0l/view?usp=sharing) 
[**Mini Handbook**](https://drive.google.com/file/d/1qlBKl0wzmlVF1oCeMQl3BahLd9nLP5Ce/view?usp=sharing) 
[**Posters and Demo guides**](https://drive.google.com/file/d/1UucMAoCNncIOaH1rMMDa0owuG9qgvJTG/view?usp=sharing)

ACL 2022

The Association for Computational Linguistics (ACL) is the premier international scientific and professional society for people working on computational problems involving human language, a field often referred to as either computational linguistics or natural language processing (NLP). 

poster

### Welcome to CogSci Conference 2025!

The 47th Annual Meeting of the Cognitive Science Society was a hybrid meeting held in San Francisco. 

<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/amvr/" seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe></div>

#### About

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.

#### Our History 

* **Society Creation** 
The Society was incorporated as a 501(c)(3) non-profit professional organization in Massachusetts in 1979. The organizing committee included Roger Schank, Allan Collins, Donald Norman, and a number of other scholars from psychology, linguistics, computer science, and philosophy. 
 
* **Conference Creation** 
The first conference on cognitive science was held at La Jolla, California in August, 1979, and has occurred annually since then. The proceedings of each conference are published, and those from most years are available through Lawrence Erlbaum Associates, Inc. The annual proceedings of the Cognitive Science Conference represent a major source of information on new work and new ideas in the scientific study of thinking. In 1990, the Society, with help from an anonymous donor, established the David Marr Prize for the best student paper at each annual meeting.
 
* **Journal Creation** 
The Journal, Cognitive Science, began publication in 1976, and is now published by Wiley-Blackwell. The Executive Editor is currently Richard P. Cooper of Birkbeck, University of London, and there are 18 Associate Editors and a 30-member editorial board. It serves as the premier outlet for research reports that intersect two or more disciplines. Copyrights for articles published in the journal are held by the Society. The Governing Board of the Cognitive Science Society voted in late 2006 to found a new journal, Topics in Cognitive Science (topiCS). The Editor in Chief is Wayne Gray, Cognitive Science Department, Rensselaer Polytechnic Institute. The journal seeks to fill a niche not occupied by Cognitive Science Journal or other cognitive science journals. Membership in the Society includes a subscription to Cognitive Science and TopiCS. Copyrights for articles published in the journal are held by the Society.
 

#### Code of Conduct

By attending the CogSci 2025 Conference, you are required to adhere to the society’s **[Code of Conduct](https://drive.google.com/file/d/1ChPuihLy6jE_BWqfO7J2KKgX35JW2zsM/view?usp=sharing)**.

You need to log in with the email address you registered with. 

Login credentials were sent to you from Underline -  subject line "Welcome to the CogSci 2025 Conference". Please be sure to check your spam/promotional inbox  if you do not see an email confirmation right away.





Please log in to join this event.

To access the site, please register [**here**](https://cognitivesciencesociety.org/registration/).

If you are registered and feel like you are seeing this message by mistake, please make sure you are logged in with the same email that you registered with. 

Please register!

The 47th Annual Meeting of the Cognitive Science Society presents the latest research across cognitive science and highlights the theme of Cognition in Context.

We adapted Irving’s (2014) Image Control and Recognition Task (ICRT) to explore a phenomenon we term outsight. The ICRT is a visual synthesis task: Participants construct a mental image of an object following stepwise instructions. They are then asked to name and subsequently draw the imagined object. We focus on trials when participants after having failed to name their mental image, could do so after having drawn it. In this exploratory study, such outsight recognition occurred on 29% of the ICRT trials. In addition, outsight recognition was accompanied by some of the phenomenological markers associated with aha! experiences. We offer some reflections on the importance of reified imagery for creativity.

Externalizing Imagery: Exploring the Phenomenology of Outsight

Traditional education is instructor-centered. Providing exploratory learning activities before instruction typically engages students and improves learning. However, the design of exploratory learning activities can impact learning processes. This study investigated whether using tables or graphs in a statistics activity impacted exploratory learning processes and outcomes. Undergraduate students (N=252) in classroom and lab settings were taught about standard deviation. In instruct-first conditions, students received instruction, then an activity including either graphs or tables. In explore-first conditions, students explored either activity before instruction. After exploring, participants in the explore-first condition reported greater knowledge gaps and curiosity than the instruct-first condition. Graphical materials reduced cognitive load compared to tabular. However, instructional order and activity design did not impact learning outcomes (procedural knowledge, conceptual knowledge, representational transfer). Conceptual understanding was highest if students attempted multiple solutions while exploring graphical materials. Depth of exploration may affect conceptual benefits, especially when using graphical materials.

Exploring a Problem Before Instruction Using Graphs versus Tables

Understanding how experts and novices allocate visual attention during mathematical problem-solving can reveal novel insights into cognitive processing. This study investigates eye-movement patterns in linear algebra tasks using entropy and recurrence quantification analysis (RQA). Eye-tracking data were collected from participants of varying expertise levels, analyzing fixation duration, scan paths, and pupillometry between key Areas of Interest (AOIs). Results indicate that experts exhibit lower entropy, suggesting a more systematic, targeted approach, whereas novices display higher entropy, reflecting exploratory and less efficient search strategies. A Welch Two-Sample t-test confirmed significant differences in entropy scores, with experts showing greater attentional focus in key areas. These findings highlight the role of visual attention in mathematical cognition. Our research underscores the potential of entropy-based metrics for assessing problem-solving strategies, with implications for content sequencing and designing instructional tools that scaffold visual attention in STEM education.

Through the Eyes of Expertise: Decoding Mathematical Cognition with Eye-Tracking and Entropy

Studies of category learning have revealed individual differences in decision-making, such that the same stimulus may be categorized differently across individuals. Modeling accounts have explained these differences in terms of how attention weights are distributed across stimulus dimensions that distinguish between category responses. These weights are typically assumed to reflect an individual’s beliefs about which dimensions are most relevant to their goals. The current work investigates the possibility that instead of being purely strategic, attention weights are constrained by what was encoded into memory during learning. Participants (N=120, age 18-25) completed a category learning task while gaze was recorded as an exogenous measure of attention. Model-based analyses using gaze to predict behavior revealed that accounting for partially-encoded representations was necessary for predicting individual differences in feature memory and categorization.

Gaze Insights into Partially-Encoded Representations of Objects and Categories

Is virtual reality (VR) necessary to conduct complex immersive eye-tracking studies or can similar data be recorded and analyzed in the real world? 

We adapted a spatial navigation paradigm from VR to the real city Limassol. Specifically, we combined a free exploration of 130 min with two pointing tasks while recording eye-tracking, head-tracking and GPS data. We labeled the eye-tracking data with a new classification pipeline and found a similar gaze distribution over object categories in both the real world and VR. Furthermore, we hand-labeled fixations on buildings to apply a graph-theoretical analysis. When comparing our results with VR we found some differences (e.g. graph density, diameter), but also many similarities in viewing behavior (e.g. hierarchy index, gaze-graph-defined landmarks).  

Overall, our work showcases the feasibility of complex eye-tracking experiments in the real world and highlights the similarity of viewing behavior in both the real world and virtual reality.

Eye tracking in the real world: a graph-theoretical analysis and comparison to virtual realty

Walking, traditionally considered an automated process, can become cognitively demanding during dual-task (DT). Up to now, language-motor interactions, such as walking and listening to speech remain underexplored in DT studies, despite the frequent co-occurrence of these activities in daily life. In addition, research on embodied semantics points at the potential of certain words’ meaning interacting with actual body movements. Yet no study so far has addressed this issue in relation to gait. This ongoing experiment examines a) the potential motor-cognitive interference of concurrent walking and speech processing and b) the potential semantic effects when action verbs are actively processed during walking. We tested 20 adults using motion capture with concurrent optical imaging (fNIRS) to assess gait variation along with frontal and motor cortex activation. Preliminary findings suggest that gait patterns remain consistent with and without speech processing during walking. However, processing action-related verbs while walking is associated with reduced motor cortex activation.

Cognitive and motor dynamics of speech processing during walking

Computer-based assessments enable the collection of fine-grained response process data through log files. We propose a novel method for extracting latent dimensions from such multidimensional response timing data, based on applying the Weighted MDS (WMDS) model. In our method, dissimilarities among examinees in their response timing vectors are computed, one such matrix for each test item, then WMDS is applied to this collection of matrices. The resulting latent dimensions represent variation among examinees in their patterns of response timing variables, with the dimension weights of the WMDS model reflecting differences across items in the importance of the latent dimensions. Latent dimensions are interpreted via permutation-based importance, correlation analysis and network analysis. Our method is demonstrated using response data from the PISA 2018 Reading and Mathematics assessments. Results show that the extracted latent dimensions are statistically reliable, educationally interpretable, and boost predictive accuracy when used in conjunction with item scores.

Extracting Latent Dimensions from Multidimensional Response Timing Data

Compounding is a common word formation process in many
languages around the world. Previous semantic analyses of
compounding suggest that analogy and composition are cru-
cial cognitive processes that underlie the formation of new
compounds, but these processes are typically considered sep-
arately. Here, we formulate a computational model of com-
pounding that integrates both analogy and composition. Com-
pared to simpler baselines, we show that the model combining
both processes achieves the best performance in predicting the
constituents of attested compounds in English, Chinese, and
German. Our work extends previous semantic-based accounts
of compounding via a computational approach that can be eval-
uated using large-scale crosslinguistic data

Modelling compounding across languages with analogy and composition

Can children detect a lie when the liar unintentionally reveals essential information (displaying so-called semantic leakage)? Furthermore, because previous research found that theory of mind (ToM) is a factor in children’s lie production ability, what role does ToM play in their lie detection ability? An experiment was carried out with 128 Dutch-speaking children (4 - 14 years old). Children’s lie detection ability was assessed using a story in which one of the characters produced a lie signalled by semantic leakage. A false-belief task was used to test children’s first-order ToM (Kevin thinks that…) and second-order ToM (Marieke thinks that Kevin thinks that…). Children were 73% accurate in identifying the liar by referring to semantic leakage. Their performance improved with age, with 12-year-olds showing ceiling performance. Finaly, first-order ToM was a significant predictor, even after controlling for age, suggesting that lie production and lie detection partly rely on shared cognitive functions.

It takes one to know one: Theory of mind helps children to detect lies that are revealed by semantic leakage

Research shows that shape perception is sensitive to both roundedness or angularity (e.g., Bar&Neta, 2006) and symmetry (e.g., Dehaene et al., 2006), and that these features also affect the perception of similarity (Tversky, 1977). Roundness is the measure of how closely the shape of an object approaches that of a circle. In an online quasi-experiment (n=74), we tested combination pairs of 19 geometric figures (created according to symmetry properties and roundedness) to answer the question whether the symmetry would contribute more than the roundedness of the figure corners to the roundness perception. Participants did a forced choice task on figure pairs presented in a random order. The results show that for regular polygons, roundedness determines the similarity assessment even when the symmetry is up to three orders higher. For pairs containing regular figures, choices were made faster compared to pairs with asymmetrical and non-regular figures.

Downloads

Next from CogSci 2025

Externalizing Imagery: Exploring the Phenomenology of Outsight

Similar lecture

Word Acquisition in Neural Language Models