This paper introduces the Word Synchronization Challenge, a novel benchmark to evaluate large language models (LLMs) in Human-Computer Interaction (HCI). This benchmark utilizes a dynamic game-like framework to test LLMs’ ability to mimic human cognitive processes through word associations. By simulating complex human interactions, it assesses how LLMs interpret and align with human thought patterns during conversational exchanges, essential for effective social partnerships in HCI. Initial findings highlight the influence of model sophistication on performance, offering insights into the models’ capabilities to engage in meaningful social interactions and adapt behaviors in human-like manners. This research advances understanding of LLMs’ potential to replicate or diverge from human cognitive functions, paving the way for more nuanced and empathetic human-machine collaborations.