Response to “What Poetry Reveals About the Sciences of the Mind”

This response addresses what we consider misinterpretations of our empirical studies in the article “What Poetry Reveals About the Sciences of the Mind.” To start with, the studies are about the aesthetic dimensions of poetry reading, not about the creation of art, let alone about “creativity” in general. The article time and again conflates the terms creativity and aesthetics. Our study tracking eye movements during poetry reading reveals correlational, not causal, insights into aesthetic experiences, contrary to the determinism suggested by the author of the commentary. In this response, we also discuss the complementary nature of empirical research and philosophical approaches in their respective capabilities to address the complexities of art reception, including individual differences among readers. Ultimately, we advocate for an integrative approach that values both subjective interpretations and empirical insights into the variety of ways in which individuals engage with art.A recent programmatic article (Brown 2024) opts for a feedback loop between two fields that are diagnosed to be almost completely separate: research in “creativity” (production in design and in the arts) and “aesthetics,” or research on the subjective perception of products of creativity. In contrast, “What Poetry Reveals” pursues this valuable goal in a way that can barely live up to it. It exclusively addresses two studies that involved no focus at all on the creation of art, but were exclusively devoted to select dimensions of art reception. One is an empirical study on what eye movements reveal about (readings of) poetry (Menninghaus and Wallot 2021), the other a programmatic theoretical article on how empirical studies might best capture aesthetic experiences (Wassiliwizky and Menninghaus 2021).Given the focus on these two reference studies, it is not surprising that the author of “What Poetry Reveals” actually invokes the concept of creativity only twice throughout his entire article, whereas the concept of aesthetic(s) is employed far more frequently. We therefore feel entitled to frame our response exclusively within the field of research in aesthetics, while leaving the issue of creativity aside.The two studies under scrutiny in “What Poetry Reveals” investigated empirically and/or discussed theoretically cognitive, affective, and aesthetic dimensions of (reading) poetry.1 The empirical study (Menninghaus and Wallot, 2021) focused on the tracking of eye movements during poetry reading. The author of “What Poetry Reveals” states early in his essay that this focus on eye movements — which are in parts involuntary and hence not under our conscious cognitive control — is indicative of “an aesthetics based on materialism and determinism” (269).We do not share this understanding. To be sure, any effort aimed at analyzing reliable empirical relationships or correlations between two (or more) variables is faced with the problem that some deterministic component needs to be assumed. Accordingly, statistical models used to analyze eye-tracking data use deterministic terms — that is, numbers or functions that are held to represent truthful estimates regarding the objects, or phenomena, under scrutiny. Importantly, however, the interpretation of the results obtained in this way goes beyond the deterministic properties required by the statistical models.Our study did not in any way aim to circumvent the importance of subjective perception and reflection in encounters with artworks. Rather, the involuntary eye movements were analyzed for potential insights they might yield into subjective ratings for cognitive, affective, and aesthetic response dimensions. Results show that eye movements, in fact, explain relevant variance in these ratings. The author of “What Poetry Reveals” actually acknowledges this finding and ultimately describes the overall result picture as one of “expanding complexity rather than of contracting simplicity” (269) and as an “indeed true and informative” example of valuable science (268). Yet he still misinterprets these results as a deterministic explanation of the ratings through the involuntary eye movements.Importantly, we neither claim nor implicitly suggest that the objectively measurable eye movements are the cause of the ratings. Rather, the evidence we present is of a merely correlational nature (specific ratings and specific eye movements tend to cooccur). No causal priority of the objectively measurable eye movements is claimed in the article.According to the author of “What Poetry Reveals,” our study does not provide an integrative interpretation of all the meanings and thoughts the poems presented might evoke in their readers. We fully agree. Our study was not designed to live up to this goal. At the same time, this by no means implies that empirical studies are generally incapable of capturing the richness of feelings and meanings individual poems are likely to elicit in readers. Rather, investigating (a) free self-reports of readers regarding their encounters with poems along with (b) subjective ratings on a broad battery of experimenter-selected scales and data on “objectively” measurable responses during reading (e.g., heart rate, eye movements, and neural-activation patterns in the brain) has great potential to reveal more about the richness and variance of individual experiences with poems than any expert reading of a poem can possibly reveal. After all, the latter is invariably limited by the reading experiences and preferences of a single reader (N = 1).In any event, the inexhaustible richness and complexity of art can by its very definition not possibly be addressed by a single study — be it humanist or empirical in nature. Empirical approaches to this ambitious issue require extended research programs involving multimethod studies that would, for instance, also measure the emotional and cognitive response dimensions related to aesthetic experiences. The eye-tracking study discussed by the author does not and cannot reasonably claim to cover all these meaningful aspects — it is criticized for failing to gather data which by definition lies beyond its scope.In our study, we presented a set of poems in original as well as modified versions. In the altered versions, rhyming words were replaced with non-rhyming synonyms, and/or meter was made irregular by either changing the order of the words or using synonyms. As a potential technical weakness of this approach, the author of “What Poetry Reveals” notes that — given that we replaced some original words through experimenter-selected synonyms that depart from the requirements of rhyme and meter — the modified versions of the poems are likely to also contain some semantic differences.We fully agree: even subtle differences in word form tend to invoke differences in meaning. Importantly, however, this by no means invalidates the experimental approach. After all, the minor semantic differences between individual original words and their synonyms can go in any direction and are only gradual. In contrast, in our study, the differences of the target words regarding the formal features of rhyme and meter are categorical and even binary. In all likelihood, therefore, the rhyme and meter effects of the experimental modification will be far more powerful across all poems in all versions than are the concomitant nonsystematic semantic differences that are likely to result from replacing some words in the original poems with synonyms.We readily acknowledge that other studies of poetry may be more meaningful for the goals the author of “What Poetry Reveals” prefers to pursue than the eye-tracking study discussed here. In fact, the first author of this response has published far more close readings than empirical studies of poems (see Menninghaus 2005, 2022). Seeking to understand potential intentions of the authors of the poems under scrutiny is a common aim of close readings, in line with the remarks by the author of “What Poetry Reveals” (286–87, from “In one important sense” to “Brattico”).Moreover, the first author of the present response has in other contexts also elaborated on what the author of “What Poetry Reveals” refers to as important guidance for poetry readings: namely, Kant's notion of “aesthetic ideas” (cf. Menninghaus 1999: 15 – 31). Thus, we are far from rejecting the philosophical preferences of the reviewer. Yet we strongly believe that the author of “What Poetry Reveals” stages as alternatives what in the end are not alternatives, but rather complementary efforts to understand poetry in all its richness.The author of “What Poetry Reveals” proposes that a comprehensive view of art reception should involve “a more expansive view of persons” who engage in art-reception (273). Again, we agree. However, if individual readers were to give extensive accounts of all ideas, emotions, and other responses they experienced while reading poems (be these self-selected or experimenter-selected), this would only scratch the surface of investigating the person variable of art reception. Analyses of such self-reports become scientifically meaningful only if the readers undergo a variety of general psychological and art-related tests. These could pave the way for a systematic understanding of how individual differences of readers inform their self-motivated encounters with artworks and the meanings these have for them. To date, fairly little is known regarding person-traits of poetry-likers. (For a recent study from our group that provides first insights specifically into person-dependent effects of specific features of poetic diction, see Menninghaus et al. 2023.)Close readings of individual poems by individual humanist readers typically aim to reveal aspects of the poems under scrutiny that had previously escaped the attention of other professional readers. This striving for novelty and originality mimics, or at least continues, the ambitions of the actual authors of poems. By virtue of their erudition, such close readings invite readers to feel and think about individual poems the way the respective scholars do. Studies of this type can greatly enrich the processes of reception in other readers. However, they are not by themselves informative about how individuals differ in the ways they perceive particular artworks.