Speaking or Writing: Do Response Times Influence Anthropomorphism Differently for ADHD and Neurotypical Users in a Mental Health Chatbot?

Abstract

The emergence of large language models (LLMs) has made it easier than ever to create chatbots capable of generating human-like responses to user inputs. Moreover, improvements in text-to-speech and speech-to-text make it possible to converse with these systems, not just in text but also through speech. These improvements have led to an increase in chatbot applications across various contexts, such as customer service and healthcare. This study examined the tendency to anthropomorphize chatbots in a mental health assessment context. Participants were randomized to interact with the chatbot via a text-based or voice-based interface. Neurotypical individuals (n = 62) and individuals with ADHD (n = 45) were recruited. The analysis revealed a significant Modality × Neurotype interaction, indicating that modality affected the groups differently. Follow-up simple-effects analyses suggested that ADHD participants tended to anthropomorphize less in the voice condition than the text condition, whereas neurotypical participants showed no reliable difference between modalities. Because response latency differed across conditions, causal attributions to modality alone cannot be made. We discuss response timing as a likely driver and implications for inclusive chatbot design.