United States

CogSci 2025

Disagreement as a catalyst for early learning

### Welcome to the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL) <br><br>
Continuing its mission of expanding and involving the science community of all European countries, EACL had selected the Ukrainian community for the 16th EACL, which took place online due to the COVID pandemic. Unfortunately, the ongoing war made the organisation in Kyiv impossible. Considering the importance of physical interaction among researchers, especially after the restrictions imposed by the COVID pandemic, in addition to an online mode, the conference will be held in Dubrovnik, Croatia, from 2 to 6 of May, 2023. The original aim of strengthening the connection with the Ukrainian community will not change as our program will feature a dedicated session and a workshop to highlight work on Ukrainian language technologies. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.<br><br><br><br>

You need to register for the conference in order to access this site. Please visit https://2023.eacl.org/registration for more information.

EACL 2023

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

invited talk

### 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**<br>
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. 
<br><br>
* **Conference Creation**<br>
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.
<br><br>
* **Journal Creation**<br>
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.
<br><br>

#### 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)**.
<br><br>


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.

Social routine words – e.g. yes, no, hi, bye, okay, and thank you – are among the first words children learn across different languages and they help constitute a foundational set of actions for conducting social interactions. Despite this, we know little about how these words are acquired. In this paper we begin by showing that social routine words are systematically acquired earlier than statistical models of word acquisition predict. Furthermore, we argue this gap is due to a selective focus of word learning science on reference – how words are mapped onto concepts and events – a relationship which is absent for social routine words. Rather than looking at the properties of words per se, we address this gap by instead looking at how children become conversational partners. Enroute to becoming a conversational partner, a child must orient themselves to others' expectations about the position & composition of their turns relative to their social partner’s. We hypothesize that second pair parts (turns which respond, e.g. agreeing, acknowledging, reciprocating a greeting, etc.) afford more scaffolding from caregivers than first pair parts, and therefore words used to compose such responses are more learnable. To support this hypothesis, we sampled 1,442 conversational turns from 5 mother-child dyads (12mo-28mo) and manually labeled the position of these turns within adjacency pairs. We found that 12-month-old’s talk is made mainly in response to caregivers who initiate adjacency pairs, but this initiation asymmetry in conversation disappears by 28-months. Because social routine words either stereotypically or frequently are used to compose second position turns, this pattern of initiation asymmetry could explain their early acquisition. More generally, our observation likely reflects a transition from scaffolding to children’s active learning.

Initiation Asymmetry in the Ontogenesis of Social Routines: In Conversation, Caregivers Scaffold 1-Year Olds to Respond, but 2-year Olds Initiate

Qualitative comparison provides learning opportunities that support a range of cognitive abilities. The present research clarifies how the contrasting goals of qualitative comparison, processing similarities versus differences, contribute to the overall impact that comparison has on learning. We argue that attending to similarities produces memory traces that are dissociable from those left by attending to differences, and that these memory traces reflect a tradeoff between generalization and individuation. Across two preregistered experiments, participants either attended to similarities or differences between stimuli. We show that whereas attending to similarities yielded better performance on a category recognition task (Experiment 1), attending to differences yielded better performance on an old/new recognition task (Experiment 2). Comparison thus serves as a learning opportunity that enables agents to manage the generalization-individuation tradeoff by shifting their attention toward similarities or differences.

Comparisons promote a tradeoff between generalization and individuation

Do children understand that people vary in how attentive they are and recognize that people prefer attentive social partners? Across six experiments, we showed participants one agent engaging attentively with a child puppet and another who was distracted throughout the interaction. Across four experiments, four and five-year-olds (total N= 132; overall mean: 4.85; range: 4.0-5.9 years) failed to distinguish the agents. Six and seven-year-olds (total N=131; overall mean: 7.01; range: 6.0-7.9 years) succeeded given repeated interactions but not robustly: fewer than half the children consistently chose the attentive agent. By contrast, adults succeeded given a single demonstration. Children’s difficulty was not due to task demands; four and five-year-olds readily distinguished agents who did and did not satisfy the puppet’s desires. Thus, although children understand attention as a discrete mental state very early in development, and react negatively when adults are not responsive, children may be relatively insensitive to cues to attention as a continuous mental state.

Not seeing it: What young children don't understand about attention

Human learners’ ability to detect rule-governed elements plays an important role in cognitive functions. While extensive research on the acquisition of regularities among adjacent items has provided robust and reliable evidence, learning structured patterns among non-adjacent components-known as non-adjacent dependencies (NADs)-remains far more tentative and only occurs under specific conditions. A past study by Lu and Mintz (2023) found that human learners need more training exposure to detect NADs in visual sequences of object transformations compared to sequences of human actions, but is unclear why. Building on this work, we present a series of three experiments to investigate whether learning NADs from visual dynamic sequences can be enhanced by maintaining the identifiability of the object throughout its transformations. In addition, we explore the effect of providing auditory information-speech or pure tones-along with the visual object transformation sequences. Our findings demonstrate that (a) NAD learning succeeded when speech cues co-occurred and matched with NAD-type frames but failed in the absence of auditory cues (Experiment 1); (b) pure tones presented contingently with the visual sequences also facilitated NAD learning (Experiment 2); and (c) regardless of whether speech or tones were used as additional cues, adult learners were unable to detect NADs when the relationship between specific auditory stimuli and specific visual object transformation sequences was disrupted (Experiment 3).

Acoustic Cues Facilitate the Acquisition of Non-adjacent Dependencies in Sequences of Dynamic Object Transformation

Many tasks feel like chores, while others are fun. Why? Here we leverage a popular puzzle game, Sokoban, to explore potential sources of variation in how enjoyable different levels of this game are to solve. In Sokoban, players navigate a grid world, pushing boxes onto goal locations while avoiding getting stuck. We first analyzed natural game play statistics (n = 442 puzzles) and found that some variation in enjoyment ratings could be jointly predicted by surface-level features (e.g., puzzle area) and solution complexity. Next, we measured how much participants reported enjoying a puzzle immediately after attempting it (N= 250 participants). We found that on successful attempts, participants enjoyed it more when they took fewer moves, whereas when unsuccessful, having made more moves was associated with greater enjoyment. Together, these studies advance understanding of how both features of the task environment and the dynamics of exploration make some activities more fun than others.

What makes people think a puzzle is fun to solve?

With the surge in cognitive gaming data, understanding players' learning patterns and cognitive growth has become increasingly important. These data offer valuable opportunities to study individual cognitive development during learning.
However, the diversity of player profiles and the complexity of gaming tasks pose significant challenges for accurate skill prediction. Specifically, the heterogeneity of player profiles leads to diverse and complex learning trajectories; data sparsity and temporal dynamics further exacerbate these challenges.
To address these challenges, we propose the MFCGT (Multi-feature Forget Cognitive Game Tracing) model.
First, we perform multi-feature selection to extract key features from player behavior data to reduce noise and improve prediction accuracy.
Second, we introduce a time-aware decay mechanism that simulates skill degradation using an exponential decay function, ensuring the model captures the impact of forgetting on learning trajectories.
Finally, we incorporate an attention mechanism to dynamically identify the most relevant historical performance for the current task, thereby enhancing the model's predictive capability.
The experiment results show that MFCGT significantly outperforms traditional models in skill prediction tasks. Additionally, MFCGT effectively captures players' learning dynamics and forgetting effects, providing more accurate learning predictions.

Enhancing Cognitive Game Tracing via Diverse Information and Time-aware Modeling

For decades, social psychologists have wondered about the cognitive foundations of social stereotype use. Arguments have generally centred either resource constraints, framing stereotypes as `energy-saving devices', or `fit', framing stereotypes as tools to represent real structure in the social environment that sometimes go awry. These resource-based and fit-based accounts have typically been presented as being in opposition to one another. In this paper, we seek to show that both are compatible under an information bottleneck model of agent representation. Through a simple simulation experiment, we demonstrate how stereotype use emerges in resource-rational representations as a function of both capacity constraints and the structure of the social environment. We then use the same framework to consider a possible explanation for the outgroup homogeneity bias in terms of limited cognitive capacity.

Downloads

Next from CogSci 2025

Initiation Asymmetry in the Ontogenesis of Social Routines: In Conversation, Caregivers Scaffold 1-Year Olds to Respond, but 2-year Olds Initiate

Similar lecture

Why Don't You Do It Right? Analysing Annotators' Disagreement in Subjective Tasks

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 2025

Initiation Asymmetry in the Ontogenesis of Social Routines: In Conversation, Caregivers Scaffold 1-Year Olds to Respond, but 2-year Olds Initiate

Similar lecture

Why Don't You Do It Right? Analysing Annotators' Disagreement in Subjective Tasks

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads