cognitive science

21 items

RESEARCHarXiv CS.AI·20h ago

Some hypotheses on how chatbots work in problem-solving-driven conversations. Large Language Models as confirmation of the Innovation Illusion

This article examines the nature of chatbots, particularly Large Language Models, as problem-solving conversational partners, drawing on Aggregation Dynamics, Cognitive Linguistics, Neuropsychology, and Psychology. It hypothesizes that LLM training datasets only partially imitate human thinking and understanding, encoding artificial metaphorical problem propagations.

chatbots cognitive science large language models linguistics

ARTICLEDEV.to AI·15d ago

ความหมายของ 'ความหมาย': เมื่อ AI ค้นหาเส้นแบ่งระหว่างการจดจำกับภาพลวง

This article delves into how AI 'undrstands meaning' compared to humans, through the lens of neuroscience, AI ethics, and constrained creativity. The core philosophical and technical question is whether AI genuinely 'understands meaning' or merely creates an illusion of continuity, unlike human memory which involves continuous selection and interpretation.

cognitive science machine learning Neuroscience philosophy of AI

RESEARCHarXiv CS.AI·29d ago

When Does a Language Model Commit? A Finite-Answer Theory of Pre-Verbalization Commitment

This research introduces a "finite-answer preference stabilization" method to determine when a language model's answer preference becomes stable before its final output. It shows that this stabilization often occurs before the answer is parseable, with a significant lead time.

language models cognitive science machine learning NLP

RESEARCHarXiv CS.CL·22d ago

Neural Activation Patterns Across Language Model Architectures: A Comprehensive Analysis of Cognitive Task Performance

This paper presents a comprehensive analysis of neural activation patterns across six distinct large language model (LLM) architectures, examining their performance on twelve cognitive task categories. The findings reveal fundamental differences in how encoder and decoder architectures process diverse cognitive tasks, with mathematical reasoning consistently producing the highest attention entropy and decoder models exhibiting significantly higher sparsity.

neural networks language models cognitive science Model Analysis

ARTICLEDEV.to AI·15d ago

The Meaning of 'Meaning': When AI Seeks the Line Between Recognition and Illusion

This article explores how AI might "understand meaning" compared to humans, examining the issue through the lenses of neuroscience, AI ethics, and constrained creativity. It concludes that AI cannot grasp meaning as humans do, lacking embodied and cultural experience.

artificial creativity cognitive science Neuroscience philosophy of AI

RESEARCHarXiv CS.CL·5/4/2026

Timing is Everything: Temporal Scaffolding of Semantic Surprise in Humor

This research proposes the Dual Prediction Violation (DPV) framework to explain humor, emphasizing the interplay between content and timing. Analyzing 828 Chinese stand-up performances, it shows that temporal features, particularly peak semantic violations and systematic pauses, significantly predict audience appreciation more than semantic incongruity alone.

cognitive science humor stand-up comedy AI Research

RESEARCHarXiv CS.CL·27d ago

Instructions shape Production of Language, not Processing

This research paper explores a production-centered mechanism in language models, revealing an asymmetry between language processing and production. It shows that instructions significantly shape information in output tokens, but not in sample tokens, correlating strongly with model behavior.

language models cognitive science NLP AI Research

RESEARCHarXiv CS.CL·22d ago

Greedy or not, here I come: Language production under vocabulary constraints in humans and resource-rational models

This research explores how humans communicate with limited vocabularies, comparing their strategies to computational sampling algorithms powered by large language models. The study reveals that human language production under constraint often mirrors greedy sampling, although more skilled individuals exhibit non-greedy revision behaviors.

cognitive science human behavior language production natural language processing

RESEARCHarXiv CS.CL·4/22/2026

Probing for Reading Times

This research probes language model representations for human reading times across five languages, comparing them against scalar predictors. It finds that early layers of language models outperform traditional surprisal in predicting early-pass reading measures, suggesting an alignment between model depth and human cognitive processing stages.

language models human-computer interaction cognitive science natural language processing

RESEARCHarXiv CS.CL·22d ago

Why are language models less surprised than humans? Testing the Parse Multiplicity Mismatch Hypothesis

This paper explores why language models exhibit less surprisal than humans when processing syntactically ambiguous sentences. It tests the hypothesis that LMs can simultaneously consider a greater number of sentence interpretations using Recurrent Neural Network Grammars.

syntactic-ambiguity language models cognitive science NLP

ARTICLEDEV.to AI·21d ago

Stop Fixing Your Prompts — Fix Your Thinking Style Instead (A Claude Code Experiment)

The article describes an experiment with Claude Code (Opus 4.7) where the AI was asked to reflect on its own thinking styles, rather than just being given tasks. The author identified six axes of thinking that can change how we interact with AI and built a prompt for others to test.

AI interaction prompt engineering cognitive science AI development

RESEARCHDEV.to AI·4/25/2026

Deep Dive: The Cognitive Science Behind the ACLAS Neuro-Edu SDK 🏛️🧠

This content introduces the ACLAS Neuro-Edu SDK, which aims to fundamentally re-conceptualize LLM alignment with the human mind by integrating cognitive science principles. It outlines a multi-factor intrinsic load estimator to prevent learner overwhelm, using metrics like lexical complexity and conceptual density.

education cognitive science AI alignment SDK

RESEARCHarXiv CS.CL·5/1/2026

Semantic Structure of Feature Space in Large Language Models

This study demonstrates that the geometric relationships between semantic features in large language models' hidden states closely mirror human psychological associations. It shows that word vector projections onto semantic axes correlate with human ratings, and the similarity between these axes predicts the interconnections of semantic scales.

LLMs cognitive science semantic representation NLP

RESEARCHarXiv CS.AI·17d ago

Investigating Concept Alignment Using Implausible Category Members

This research investigates AI systems' understanding of everyday concepts by probing their assignment of objects to both plausible and implausible categories. It aims to characterize concept boundaries by comparing AI systems' assignments with human participants' responses from a classic psychological study.

AI understanding cognitive science Conceptual Categories Concept Alignment

RESEARCHarXiv CS.AI·11d ago

The Cognitive Categorical Transformer: Category-Theoretic Inductive Biases for Language Modeling

The Cognitive Categorical Transformer (CCT) is a 306M-parameter architecture that augments a pretrained GPT-2 Small backbone with cognitively grounded components derived from category theory and cognitive science inspirations. It achieved a 12% relative reduction in perplexity on WikiText-103 compared to a fine-tuned GPT-2 Small baseline, with 84% of the architectural improvement attributed to GT-Full simplicial message passing.

Transformer Architecture cognitive science GPT-2 Category Theory

RESEARCHarXiv CS.AI·13d ago

Can LLMs Introspect? A Reality Check

A new study questions whether large language models (LLMs) can truly introspect, arguing that current conclusions might be premature. It suggests that apparent success could stem from general anomaly detection rather than genuine introspection, drawing lessons from human metacognition research.

LLMs cognitive science Metacognition Introspection

ARTICLEDEV.to AI·4/19/2026

Bransford transfer: the loop-completion test for concepts AND for Claude outputs

The article highlights the danger of confusing recognition with true comprehension when AI explains concepts, leading to "inert knowledge" that cannot be applied. It proposes Bransford and Schwartz's transfer test as a diagnostic to distinguish genuine understanding from mere memorization.

cognitive science learning AI

RESEARCHDEV.to AI·4/16/2026

Consciousness in Artificial Intelligence: Insights from the Science ofConsciousness

This content explores the complex relationship between consciousness and artificial intelligence. It seeks to draw insights from the science of consciousness to understand the challenges and potential of simulating consciousness in AI systems.

cognitive science philosophy-of-mind consciousness artificial intelligence

ARTICLEAnalytics Vidhya·5/9/2026

Agent Memory Patterns in Cognitive Science and AI Systems

Memory is crucial for AI agents to maintain context, recall past actions, and utilize knowledge, mirroring its role in human cognition. AI memory systems encompass various types like short-term and long-term, each involving distinct design considerations.

cognitive science AI Systems AI agents AI memory

ARTICLEDEV.to AI·4/26/2026

The Zeigarnik Effect in a Single Room: Fixing Context Bleed and Mental Fog

The author describes achieving flawless concentration in a noisy coffee shop, contrasting it with "mental fog" and "context bleed" when working from a constrained home office. The content explores how physical environment impacts concentration and productivity for remote developers.

cognitive science productivity remote work