Scientific Study of Literature - Volume 5, Issue 2, 2015

In this state-of-the-art review, I start with an illustrative example of behavioral data collected during the reading of a love poem reflecting one of many aspects that form the object of the Scientific Study of Literature, i.e., literary experience. A further section discusses key ingredients of literary experience, i.e., immersive and aesthetic processes. The paper’s core part analyses four recent representative empirical studies covering perspectives ranging from phenomenology to cognitive neuroscience. In the final sections, a number of critical theoretical and methodological issues are considered. Systematic research on the ontogeny of literary experiences is identified as a main desideratum of the field and further recommendations for the future scientific study of literature are proposed.

Jacobs (2016) raises a number of insightful and provocative points about the study of literary experience, including the importance of development, the promise of process models, and the role of quantitative methods. In the present comment, we first elaborate on one aspect of the literary experience that seems to be neglected by his introductory comments, namely, that that experience is not limited to the act of reading but can easily extend to long after the reading is completed. Based on this insight, we then offer an analysis of the types of measurements that might be used in the empirical study of literary processing. While this analysis is not necessarily incompatible with Jacob’s discussion, we believe that it offers several new insights that are not readily apparent.

The process of reading and writing poetry is increasingly conducted by non-professionals. The current study utilized a series of regression models to explore the mechanism through which beauty judgements of non-professional poetry are made. The analysis addressed the relationships among the decision that a poem was written by a published poet (authorial attribution), a perception of the quality of the writing, the emotional response to the poem and a beauty judgement of the poem. 54 participants from two graduate applied linguistics programs rated 5 non-professional poems for their beauty, emotive response, quality of writing and semi-professional status of the writer. Analyses were conducted on averaged ratings across all five poems. The results indicate the beauty judgements, emotive response and quality of writing judgements were closely related. The decision that a poem is written by a published poet predicted the quality of writing and emotional response to the poem. An inconsistent mediation model was determined, in which increases in the semi-professional status of the writer increased the self-reported emotive response and quality of writing which in turn increased the beauty judgement of the poem. The results suggest a mechanism by which convergence of aesthetic judgement with novice reviewers is directed by the social sanctioning of the authority and quality of the writer.

The digital era has brought with it a shift in the field of literary editing in terms of the amount and kind of textual variation that can reasonably be annotated by editors. However, questions remain about how far readers engage with textual variants, especially minor ones such as small-scale changes to punctuation. In this study we present an eye-tracking experiment investigating reader sensitivity to variations in surface textual features of prose fiction. We monitored eye movements while participants read textual variants from Dickens and James, hypothesising that readers may pay more attention to lexical rather than punctuation changes. We found longer reading times for both types, but only lexical changes also increased reading times for the rest of the sentence. In addition, eye-movement behaviour and conscious ability to report changes were highly correlated. We discuss the implications for how such methods might be applied to questions of “literary” significance and textual processing.

In literary theory, there are two common proposals about the nature of literature: that literariness is a distinctive characteristic of literary texts and that literariness is mostly characterized by highly foregrounded textual features. Our studies focused on the effects of genre expectations. In two experiments, we examined whether particular rhetorical figures (features, such as oxymora, synesthesia, and personification) are processed differently when readers think they are embedded in literary sentences. Participants were first induced to think that the sentences they were reading were taken either from literary texts (literary group) or from newspaper articles (news group). They then read sentences containing rhetorical figures and control sentences that did not contain figurative language. Results of Experiment 1 were consistent with prior research indicating that genre expectations influence how a text is processed (Zwaan, 1991, 1994). Moreover, specifically in the news group, genre expectations similarly affect reading times for sentences with rhetorical figures and sentences containing low frequency and long words, suggesting that foregrounding presents a lexical challenge elicited by micro-level linguistic and rhetorical “obstacles.” However, Experiment 2 did not replicate the findings of Experiment 1. These conflicting results prompt consideration of the limitations of an exclusive focus on foregrounding; in literary texts there is always an interchange between backgrounding and foregrounding elements.