This study investigates how AI companion chatbots influence emotional attachment patterns among users, focusing on linguistic markers of dependency in negative emotion conversations. Using mixed methods, we analyzed 6,030 English conversations from an AI chatbot dataset.

This study investigates how AI companion chatbots influence emotional attachment patterns among users, focusing on linguistic markers of dependency in negative emotion conversations.

We introduced the novel problem of predicting evoked emotions in conversations (PEC) and pro-posed two novel neural network models to address it – BiLSTM-PEC and DGCN-PEC.

To achieve this purpose, we draw on the effectiveness of Conditional Random Field (CRF) for modeling sequential dependencies to explicitly model emotional interactions in the conversation.

As a branch of sentiment analysis tasks, emotion recognition in conversation (ERC) aims to explore the hidden emotions of a speaker by analyzing the sentiments in utterance.

Emotion recognition in conversations (ERC) is a pivotal component of affective computing, involving a common two-stage paradigm where pre-trained language models first extract context-independent features, followed by the encoding of contextual information and the modeling of emotional dependencies.

As a branch of sentiment analysis tasks, emotion recognition in conversation (ERC) aims to explore the hidden emotions of a speaker by analyzing the sentiments in utterance.

Conversational AI (CAI) has become widely embedded in daily life, raising concerns about excessive reliance on these systems. This study developed and validated the Conversational AI Dependence Scale (CAID), a novel two-dimensional measure assessing instrumental dependence and emotional dependence.

Based on the structural equation modeling, this paper analyzes the potential pathways for artificial intelligence chatbots to induce the user’s emotional dependence.

This study extends dual-process models of media dependency to the context of AI-powered conversational agents, addressing a theoretical gap in existing literature that has primarily focused on traditional media or internet technologies.