I consider poetry composition from both the “inside” view of a poet and the “outside” view of a cognitive psychologist. From the perspective of a psychologist, I review behavioral and neural studies of the reception and generation of poetry, with emphasis on metaphor and symbolism. Taking the perspective of a poet, I discuss how the seeds for a poem may arise. Finally, I consider the prospects for future developments in a field of computational neurocognitive poetics.

In this paper we examine the effect of two cognitive variables, Semantic Neighborhood Density and Interpretive Diversity, in first, distinguishing between literary (poetic) and nonliterary metaphor, and second, in determining what makes for a good metaphor. Analyses of items taken from a widely used set ofmetaphor norms indicated that while literary and nonliterary metaphor did not differ in many ways, the poetic items tended to 1) contain concepts that came from a more dense semantic space, 2) contain topic and vehicles (...) that came from equally dense semantic space, 3) suggest a greater number of possible interpretations as the topic and vehicle became more semantically dissimilar, and 4) evoke more emergent interpretations (i.e., less likely to be a characteristic of the topic or vehicle when considered separately). In addition, we found one way that the two variables were related to metaphor goodness: better metaphors were those with vehicles that came from increasingly less dense semantic space. This correlation was only reliable for literary, poetic items, presumably because these items were taken from a richer semantic environment suggesting many more alternative possibilities. (shrink)

Despite the exceptional performance of large language models (LLMs) on a wide range of tasks involving natural language processing and reasoning, there has been sharp disagreement as to whether their abilities extend to more creative human abilities. A core example is the interpretation of novel metaphors. Here we assessed the ability of GPT-4, a state-of-the-art large language model, to provide natural-language interpretations of a recent AI benchmark (Fig-QA dataset), novel literary metaphors drawn from Serbian poetry and translated into English, and (...) entire novel English poems. GPT-4 outperformed previous AI models on the Fig-QA dataset. For metaphors drawn from Serbian poetry, human judges – blind to the fact that an AI model was involved – rated metaphor interpretations generated by GPT-4 as superior to those provided by a group of college students. In interpreting reversed metaphors, GPT-4, as well as humans, exhibited signs of sensitivity to the Gricean cooperative principle. In addition, for several novel English poems GPT-4 produced interpretations that were rated as excellent or good by a human literary critic. These results indicate that LLMs such as GPT-4 have acquired an emergent ability to interpret literary metaphors, including those embedded in novel poems. (shrink)