Literary Theory of Mind and LLM Fictions: Implications for Awareness

We are delighted to welcome N. Katherine Hayles to the University of Copenhagen for a guest lecture. N. Katherine Hayles is a leading voice in the study of literature, media, and technoscience, known internationally for her contributions to posthumanist theory, digital humanities, and the cultural implications of computation. 

Theory of Mind (ToM) tests try to determine if an entity can imagine what another entity is thinking. Simple “false belief” tests have been used with chimpanzees, very young children, and neurodivergent adults such as autistics and schizophrenics. They have also been used with LLMs, and the results generally indicate LLMs such as GPT-5 do have ToM.  Forty years ago, G. G. Gallup proposed that theory of mind strongly correlates with self-awareness, and in an even stronger version, that self-awareness is necessary for ToM to emerge. Empirical evidence for conscious animals such as chimpanzees has confirmed the first hypothesis, with mixed results for the second.

This talk will discuss new evidence using the innovative approach of applying literary ToM techniques to ten fictions written by GPT-5, which to our knowledge has not been tried before. Literary ToM (LToM) differs from traditional ToM tests, which typically focus on behavior rather than language (necessary for nonhuman animals and young children), because it uses techniques developed through centuries of literary close reading. Devised to interrogate complex states of mind in literary fictions, LToM techniques have an even stronger correlation with self-awareness than does traditional ToM. The results of our analyses indicate a strong correlation with forms of awareness, but a consideration of LLMs’ cognitive capacities indicates they lack a persistent sense of self, emotions, and introspective knowledge of their instantiations. Rather, they infer these qualities from the human-authored texts they ingest, thus deriving them second-hand. The fictions we analyzed blended a nuanced sense of human selfhood with LLM functions, indicating that although the LLM did not have an interior sense of self, it had extensive knowledge of human selfhood. We call this kind of simulated or derivative awareness (awareness of human awareness) “inverse awareness,” because it inverts what the LLM  knows about human selfhood and attributes it to its own functions, introducing innovative modifications appropriate to the LLM umwelt. The talk will present excerpts from the stories, along with our analyses and conclusions.

Bio

Katherine Hayles is the Distinguished Research Professor at the University of California, Los Angeles, and the James B. Duke Professor Emerita from Duke University. Her research focuses on the relations of literature, science and technology in the 20th and 21st centuries. Her twelve print books include Postprint: Books and Becoming Computational (Columbia, 2021), Unthought: The Power of the Cognitive Nonconscious (Univ. of Chicago Press, 2017) and How We Think: Digital Media and Contemporary Technogenesis (Univ. of Chicago Press 2015), in addition to over 100 peer-reviewed articles.  Her books have won several prizes, including  The Rene Wellek Award for the Best Book in Literary Theory for How We Became Posthuman: Virtual Bodies in Literature, Cybernetics and Informatics, and the Suzanne Langer Award for Writing Machines.  She has been recognized by many fellowships and awards, including two NEH Fellowships, a Guggenheim, a Rockefeller Residential Fellowship at Bellagio, and two University of California Presidential Research Fellowships.  She is a member of the American Academy of Arts and Sciences.  Hew latest 2025 book is Bacteria to AI:  Human Futures with our Nonhuman Symbionts.