When AI companions become witty: Can human brain recognize AI-generated irony?

2025-10-20

AI Summary

This study investigates whether humans perceive AI-generated irony as intentional social behavior. Method: Using event-related potentials (ERP), with emphasis on the P200 (early attentional capture) and P600 (semantic reanalysis) components, alongside behavioral measures, we compared neural and cognitive processing of irony produced by AI versus human agents. Results: Participants were significantly more likely to attribute semantic incongruity in AI-generated utterances to computational error rather than intentional irony; correspondingly, both P200 and P600 amplitudes were markedly attenuated for AI sources. Intent attribution strength correlated positively with perceived sincerity. This work provides the first empirical evidence that AI agency systematically diminishes irony intentionality attributions and identifies the underlying neural mechanisms—revealing that socio-pragmatic language comprehension critically depends on pre-existing mental-model assumptions about the interlocutor. Achieving genuine social intelligence thus requires not only surface-level linguistic competence but also the construction of human-trustworthy intentional representations.

Technology Category

• Humans and AI: Emotional Intelligence
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Philosophy and Ethics of AI: Artificial General Intelligence

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

As Large Language Models (LLMs) are increasingly deployed as social agents and trained to produce humor and irony, a question emerges: when encountering witty AI remarks, do people interpret these as intentional communication or mere computational output? This study investigates whether people adopt the intentional stance, attributing mental states to explain behavior,toward AI during irony comprehension. Irony provides an ideal paradigm because it requires distinguishing intentional contradictions from unintended errors through effortful semantic reanalysis. We compared behavioral and neural responses to ironic statements from AI versus human sources using established ERP components: P200 reflecting early incongruity detection and P600 indexing cognitive efforts in reinterpreting incongruity as deliberate irony. Results demonstrate that people do not fully adopt the intentional stance toward AI-generated irony. Behaviorally, participants attributed incongruity to deliberate communication for both sources, though significantly less for AI than human, showing greater tendency to interpret AI incongruities as computational errors. Neural data revealed attenuated P200 and P600 effects for AI-generated irony, suggesting reduced effortful detection and reanalysis consistent with diminished attribution of communicative intent. Notably, people who perceived AI as more sincere showed larger P200 and P600 effects for AI-generated irony, suggesting that intentional stance adoption is calibrated by specific mental models of artificial agents. These findings reveal that source attribution shapes neural processing of social-communicative phenomena. Despite current LLMs' linguistic sophistication, achieving genuine social agency requires more than linguistic competence, it necessitates a shift in how humans perceive and attribute intentionality to artificial agents.

Problem

• Investigating whether humans attribute intentionality to AI-generated irony
• Comparing neural responses to irony from AI versus human sources
• Examining how source attribution affects irony comprehension mechanisms

Innovation

• ERP components measure neural irony responses
• Compare human and AI intentional stance adoption
• P200 and P600 reveal reduced AI intent attribution