Abstract
With the growing prevalence of health and illness narratives on digital platforms, research examining the social psychological processes involved in these storytelling environments remains scarce. This paper addresses this research gap by conducting a mixed-methods study of digital storytelling within the UK’s healthcare context, focusing on online consumer reviews of the medical memoir, Do no harm: stories of life, death and brain surgery (Marsh, 2014). Utilising computer-assisted text analysis methods of LIWC-22 and the Sketch Engine, linguistic cues for cognitive, affective, social and perceptual processes are identified in a corpus of online consumer reviews. A subsequent qualitative analysis, based on ‘narrative modulation’ (Huang, 2024, 2020), investigates the role of these processes in constructing and developing storylines across the user reviews. Finally, the study explores how consumer reviews in the form of ‘small stories’ challenge canonical narratives in the UK’s healthcare services. This research advances the field of narrative studies by emphasising the role of social psychological processes (Chung and Pennebaker, 2019) in modulating emerging, evolving and counter narratives in digital storytelling. The findings reveal an instrumental role of social psychological processes, as signalled by linguistic cues, in shaping narrative threads in online user reviews. This study not only develops narrative modulation as a valuable concept for narrative analysis, but also underscores its effectiveness when combined with computer-assisted text analysis tools for in-depth examinations of narrative data. Furthermore, it provides critical insights into digital storytelling in healthcare contexts, promoting knowledge transfer across narrative studies, stylistics, social psychology and medical humanities.
Keywords
1. Introduction
Digital narratives on online platforms have become increasingly embedded in the lives of individuals and communities, constantly shaping and reshaping the ways people perceive and discuss health and wellbeing. This transformation has led to a significant diversification in the landscape of health and illness narratives, including private and public discussions about health and wellbeing on social media, online user feedback on healthcare services and consumer reviews on medical and health-related products (Douglas et al., 2020; Gubrium et al., 2014; Levett-Jones et al., 2015; Montgomery et al., 2019; Park et al., 2021; Rice, 2020). Despite notable progress in health narrative research, comprehensive understanding of the social psychological processes integral to digital storytelling is still limited. With the exception of a few existing studies that have discussed emotional and affective responses in health narratives (e.g., Eichstaedt et al., 2015; Stage 2017), there remains a knowledge gap about how these and other processes across social psychological domains are reflected through linguistic cues in digital health and healthcare narratives. Additionally, while online user reviews have been explored in digital narrative research across various contexts (Allington, 2016; Vásquez, 2015, 2019), online consumer book reviews on health and healthcare topics have yet to receive extensive scrutiny. However, investigations into these reviews warrant research attention because they provide unique insights into the narrative strategies and social psychological processes employed by reviewers in discussing health topics. Analysis of this narrative data can uncover the ways in which online user reviews of medical-themed literature influence and reflect personal experiences and perceptions of health and wellbeing.
This study addresses these theoretical gaps by conducting a case study that aims to explore the role of social psychological processes in constructing digital storytelling in UK’s healthcare contexts. The types of social psychological processes examined in this study include cognitive, affective, social and perceptual processes (Boyd et al., 2022; Rude et al., 2004) involved in digital storytelling, particularly in the context of health narratives (For further information see Sections 2.2 and 3.2). In this paper, the term ‘social psychological processes’ is used in an inclusive manner to describe these processes. As will be illustrated in subsequent sections (see Sections 4.1 and 4.2), these processes, identified through linguistic cues in online consumer reviews, play an instrumental role in modulating storylines and narratives within the healthcare domain.
The narrative data examined in the case study is selected from online consumer reviews of a medical memoir Do no Harm: Stories of Life, Death and Brain Surgery by Henry Marsh (2014). Online user reviews of Marsh’s memoir were chosen for this study because the memoir offers a rare glimpse into the life of a neurosurgeon working in Britain’s National Health Service (NHS). The memoir is not only an important contribution to medical literature but also a profound exploration of the human condition through the lens of neurosurgery (Arnold-Forster 2019). The memoir’s candid and compelling portrayal of the emotional and psychological turmoil experienced by surgeons, patients and their families appeals to a broad audience (Stephen, 2017), leading to rich interpretations and emotional responses in online book reviews. Furthermore, the memoir presents a poignant narrative that counters the institutional ideology of the NHS, resonating with many readers in their online reviews. For further details, see Section 4.2, including an analysis of an excerpt from the NHS long term plan (2019). These online narratives provide a wealth of data for examining the social psychological processes at play. By analysing these reviews, this study offers a unique perspective on the interaction between readers’ experiences and the narrative of the memoir, illuminating the role of storytelling in digital spaces in shaping and mediating public understanding and attitudes towards healthcare and wellbeing. To ensure the relevance of the analysis to the UK context, this study specifically focuses on consumer reviews originating from the United Kingdom, as indicated by the reviewers’ locations and verified purchase status on digital platforms (See Section 3.1 for details).
With the support of the narrative data, the primary objective of the case study is to investigate the role of social psychological processes in shaping digital storytelling in the UK’s healthcare contexts. This objective can be elaborated into three specific parts. Firstly, the study aims to identify relevant linguistic cues that indicate cognitive, affective, social and perceptual processes in the narrative data. Secondly, the study seeks to categorise and quantify these linguistic cues in order to reveal patterns and insights into how these processes manifest in the reviews. Thirdly, the study investigates how these social psychological processes influence and modulate storytelling within the user reviews.
To achieve this, a mixed-methods approach is developed to integrate both quantitative and qualitative analyses, specifically tailored to illuminate the interplay between linguistic cues, social psychological processes and their contributions to narrative construction. In response to Parts 1 and 2 of the research objective, the first part of the methodological approach employs computer-assisted text analysis using LIWC-22 (Boyd et al., 2022) and the Sketch Engine (Kilgarriff et al., 2014). This phase is directed towards identifying linguistic cues that reflect cognitive, affective, social and perceptual processes in the text of online reviews of the memoir. Using the text analysis tools, these identified cues are further investigated for their linguistic frequencies, features and patterns in the narrative data. As will be presented in Sections 3.2 and 4.1 of the paper, the quantitative part of the analysis provides important initial insights into the social cognitive processes that help construct online reviewers’ personal and collective experiences and perceptions of healthcare and wellbeing.
Following the initial findings from the quantitative analysis, the study conducts a subsequent qualitative analysis using the concepts of ‘narrative modulation’ (Huang, 2020, 2024) and the ‘small stories’ approach (Georgakopoulou, 2007; Vásquez, 2014) to examine the ways the identified linguistic cues shape and develop storylines within the reviews. In response to Part 3 of the research objective, the qualitative approach focuses on how relevant social psychological processes dynamically navigate and influence the trajectory of storytelling (See Sections 2.3, 3.2 and 4.2 for details). By extending the initial quantitative insights into a deeper contextual understanding, the qualitative narrative analysis further elucidates how online consumer reviews, acting as ‘small stories’, modulate, counter and challenge some of the prevailing narratives within the UK’s healthcare system, offering fresh perspectives on digital narratives and their sociocultural impact.
By examining online consumer reviews of Marsh’s medical memoir with the analytical approaches outlined above, this study addresses the aforementioned research gaps in digital narratives in the UK’s healthcare context. The study produces new empirical findings and engages with complex problems of narrative modulation and public perception of healthcare. As will be demonstrated in the following sections, the case study illustrates how different analytical frameworks—namely LIWC-22, the Sketch Engine, narrative modulation and the small stories approach—can be synthesised to achieve the research objectives. Furthermore, this integration extends and advances the framework of narrative modulation as a middle-range analytical concept for narrative studies. The research offers critical insights into how social psychological processes shape public discourse on health and wellbeing, thereby advancing theoretical understanding and providing a foundation for practical implications in the healthcare sector.
In the following sections, Section 2 outlines the theoretical background, discussing the key concepts and frameworks that underpin the design and implementation of the study. Section 3 details the data collection and analytical methods for both the quantitative and qualitative parts of the study. Section 4 presents and discusses the findings from the analyses, illustrating how social psychological processes shape and modulate digital health narratives via linguistic cues. Finally, Section 5 considers the key contributions and implications of the study.
2. Theoretical background
This section presents a review of the relevant theoretical background, focusing on studies of digital storytelling in healthcare contexts with the approach of ‘small stories’, research on psychological processes in computer-assisted text analysis, and an overview of the concept of ‘narrative modulation'.
2.1. A small stories approach to digital storytelling in healthcare contexts
Digital storytelling (DST) involves storytelling with digital components, such as blogs, vlogs, podcasts, social media posts, online reviews, videos and pictures (De Fina and Perrino, 2019; Dunford and Jenkins, 2017). In health narrative research, DST is recognised as a legitimate approach for communicating the experiences of healthcare clients and practitioners, facilitating knowledge transfer in healthcare, and impacting clinical practice and health policies (Hardy, 2016). DST has contributed to studies in medical humanities (Matthews and Sunderland, 2017), but challenges remain in integrating differing epistemological and methodological approaches between the disciplines of humanities and biomedical sciences (Ahlzén R, 2019; Roest et al., 2021). Linguistics research can support this integration by focusing on linguistic features, styles, patterns and structures in the storytelling process (Demjén, 2020). One suitable linguistic method for health narrative research is the ‘small stories’ approach (Georgakopoulou, 2007; Georgakopoulou et al., 2023). Small stories are typically emergent, ongoing and personal narratives, contrasting with big stories, also referred to as ‘canonical narratives’ (Georgakopoulou, 2007:17). Canonical narratives are typically produced by institutions or organisations wielding significant power. These narratives serve not only as prototypical, polished forms of storytelling but also as vehicles for conveying institutional ideologies and identities, reflecting dominant cultural or societal norms (Bamberg, 2014; Georgakopoulou and De Fina, 2015).
Adopting the small stories approach, this study is further informed by Vásquez’s research on online consumer reviews (2012, 2014). Vásquez (2014) conceptualises online reviews as part of a ‘cline of narrativity’, a continuum where reviews vary from less to more narrative in content and structure. This spectrum acknowledges the diversity in narrative depth within reviews, ranging from accounts of personal experiences and description to evaluations, emotional responses, and so forth. This perspective allows for the examination of online reviews of differing forms, content and lengths as small stories, including ‘micro-narratives’ that are concise and low on the traditional narrativity spectrum (Vásquez, 2014: 137). Aligned with this perspective, this study views online reviews of Do no harm: stories of life, death and brain surgery (Marsh, 2014) as small stories that vary along the continuum of narrativity. These reviews are seen as situated and emergent stories, each with varying structures and content, contributing to a collective understanding and dialogue about the reviewers’ experiences and evaluations of the book in healthcare contexts.
In addition to the cline of narrativity, Vásquez’s work adapts the concept of ‘narrative dimension’ originally developed by Ochs and Capps (2001), illustrating that online reviews manifest narrative dimensions 1 such as moral stance, linearity, tellability, tellership and embeddedness. This understanding challenges the conventional view of narratives as well-structured and monologic. It allows us to see each review as an individual, yet interconnected, contribution to a larger tapestry woven through the corpus (See Vásquez, 2014 for a detailed discussion of narrative dimensions; see also Sections 2.3 and 4.2 for interconnections between narrative dimensions and narrative modulation).
Accordingly, this study treats each review both as a standalone story and as a part of a collaborative storytelling process. Each review is considered a standalone story because it exists in its own right, representing an individual’s unique perspective and experience. Simultaneously, these reviews are collaborative, as they are interconnected within a broader network and context. This includes the specific book being reviewed, the interwoven storylines and themes, the reviewers’ shared sentiments and values, and sometimes connections to previous reviews and relevant resources. This duality acknowledges the individuality of each contribution while situating it within a collective narrative landscape. The theoretical underpinnings of these aspects, particularly how the social psychological processes are involved in modulating and constructing these narratives, will be discussed in the subsequent two sections.
2.2. Social psychological processes in digital storytelling about health and wellbeing
Adopting the small stories and living narrative approaches, narratives are seen as both cognitive constructs and discursive actions (Gergen and Gergen, 2006; Page, 2010; Ryan, 2004). While previous research has examined cognitive and affective aspects of digital storytelling in health and wellbeing contexts (Banerjee and Green, 2012; Giaxoglou, 2022; Stage 2019), a deeper exploration of social psychological processes remains pertinent. Such exploration should consider how these processes mobilise linguistic and discursive resources within these contexts.
To address the above aspects, linguistic cues can serve as reliable indicators of speakers’ psychological processes and states (Toma and D'Angelo, 2015; Consedine et al., 2012; Rude et al., 2004; Stirman and Pennebaker, 2001). LIWC (Linguistic Inquiry and Word Count), a computerised text analysis tool, is designed to analyse linguistic expressions of psychological processes. This tool has been applied in social psychology and health communication research to identify social psychological content in language samples from online media in the contexts of illness, health and wellbeing (Eichstaedt et al., 2015; Ireland et al., 2015; Tausczik et al., 2012). LIWC comprises a dictionary component and a text processing program. The most recent version, LIWC-22 (Boyd et al., 2022), includes over 12,000 linguistic tokens to assess a range of social and psychological constructs across 80 categories. LIWC-22 uses a text processing program to produce a word count of the target text against the dictionary tokens, generating a frequency list for each category of all matched words. Recognising linguistic features and patterns in computer-mediated communication (CMC), LIWC-22 provides a systematic and transparent approach to surveying linguistic data representing social psychological phenomena (Boyd et al., 2022).
To achieve the best results, LIWC-22 should be used in conjunction with other methods such as corpus-analytical methods (Chung and Pennebaker, 2019; Demjén, 2020). In this paper, LIWC-22 joins force with the Sketch Engine, a web-based corpus analysis software, to identify relevant linguistic cues for social psychological processes involved in consumer online reviews of Do no harm: stories of life, death and brain surgery (Marsh, 2014). The Sketch Engine enables detailed corpus analyses by identifying keywords, concordances and collocates, which help to reveal patterns and contexts of linguistic usage in the narrative data (Kilgarriff et al., 2014). These aspects of the narrative data identified by the Sketch Engine are further analysed using LIWC-22 to categorise them into relevant social psychological processes (Boyd et al., 2022). This integration of LIWC-22 and the Sketch Engine forms the quantitative part of the case study. These tools work collaboratively to identify and quantify linguistic tokens that correspond to cognitive, affective, social and perceptual processes in the online reviews. The initial results obtained from the quantitative analysis are subsequently analysed in the qualitative narrative analysis with the aid of the concept of narrative modulation, reviewed in Section 2.3.
2.3. Narrative modulations in digital storytelling and health narratives
This section transitions from quantitative, computer-assisted analytical methods to qualitative approaches, presenting an analytical concept of ‘narrative modulation’ that is well-suited to explore the social psychological processes that contribute to digital storytelling in healthcare contexts. ‘Narrative modulation' is a middle-range analytical concept that comprises a set of dynamic processes, typically fulfilled by social psychological processes, that serve to navigate the trajectory of storytelling, mobilise narrative resources, and configure storylines and themes (Huang, 2024, 2020). Social psychological processes that contribute to the modulation process are referred to as ‘narrative modulators’. In this study, ‘themes’ are generally based on topical matters, and ‘storylines’ can be seen as ‘narrative threads’ that run longitudinally through a text, like currents of a river that convey meaning and contextual detail (Thomas et al., 2009: 791. See Section 4.2 for further discussion of storylines).
Narrative modulation has been applied in previous studies (Huang, 2024, 2020) to analyse written and spoken narratives in healthcare contexts. The studies found that social-cognitive processes such as conceptual frames, metaphors, attunement and misattunement, acted as ‘narrative modulators’ that navigated and channelled the development of multiple storylines across participants’ written and spoken narratives about their living experiences with health, illness and the healthcare system. Drawing on cognitive narratology and narrative inquiry (Gergen and Gergen, 2006; Herman, 2007), narrative modulation aligns with key concepts of the ‘small stories’ approach discussed in Section 2.1. This makes narrative modulation a suitable concept for analysing online consumer reviews that take the form of ‘living’ narratives (Ochs and Capps, 2001) and are amenable to analysis via the ‘small stories’ approach (Vásquez, 2014).
The impact of narrative modulation on the storytelling process depends on three factors: stance, scope and strength (Huang, 2020). ‘Stance’ encompasses the values, beliefs and opinions that drive modulation, akin to the ‘moral stance’ (Ochs and Capps, 2001; Vásquez, 2014), which conveys the tellers’ and listeners’ viewpoints on events and socio-cultural moral perspectives. The ‘scope’ of narrative modulation covers the extent to which the process influences various narrative resources. This includes the depth and range of influence on storylines, themes and narrative elements such as characters/actors, events, conditions, thoughts and feelings (see Kleres, 2010, for a detailed discussion of narrative elements). The concept of ‘strength’ in narrative modulation, as informed by the ‘force’ image-schema from cognitive linguistics (Johnson, 2005; see also Huang, 2020), refers to the intensity or impact that a modulation process has in storytelling. The ‘strength’ of modulations can be exemplified through the use of ‘Extreme Case Formulations’ (ECFs), which are expressions invoking the utmost degree of experiences and events (Sidnell, 2004; Whitehead, 2015). As will be elaborated in Section 4.2, the use of ECFs such as ‘absolutely’, ‘desperately’ and ‘astonishingly’ not only reflects but also amplifies social psychological processes in online review narratives. These ECFs demonstrate a strong ‘force’ image-schema within the modulation process, indicating a high level of narrative impact.
Narrative modulation not only influences the trajectory of storytelling but can also shape relevant narrative dimensions as discussed in Section 2.1 (Vásquez, 2014). In the context of online review narratives, such as the one presented in this study, the modulation process may alter narrative linearity within a review, highlight certain moral stances of the reviewer, develop tellership and tellability through social psychological processes functioning as narrative modulators, as well as embedding the review within broader, interconnecting narratives and discourses.
In summary, the application of narrative modulation as an analytical concept is particularly pertinent to this study. It enables a nuanced exploration of how social psychological processes shape and influence the construction of digital storytelling in healthcare contexts. This approach can unpack the layers of meaning within online consumer narratives, offering deeper insights into the interplay between individual experiences and broader, collective discourses in healthcare contexts. Following this, narrative modulation is employed to form the qualitative component of the analysis of online consumer reviews of Marsh’s medical memoir Do no harm: stories of life, death and brain surgery (2014). Details of the research data and methodology will be elaborated in the subsequent section.
The integrated framework adopted in this study—combining the ‘small stories’ approach, LIWC-22, the Sketch Engine and the concept of ‘narrative modulation’—provides a comprehensive and robust theoretical foundation for analysing digital storytelling in healthcare contexts. This framework tackles the research gaps identified in Section 1. Specifically, it addresses the scarcity of research examining the social psychological processes involved in digital health narratives, the limited exploration of online consumer book reviews on health and healthcare topics, and the potential for implementing suitable methodologies to form a mixed-methods approach. By offering a case study that employs quantitative and qualitative approaches well suited for examining social psychological processes in digital narratives, this paper presents an original application and integration of these analytical tools and concepts to produce new empirical findings and insights. In doing so, the study contributes to the advancement of knowledge in narrative studies, social psychology and healthcare communication by enhancing the understanding of digital health narratives and informing healthcare practices and policies.
3. The current study: data and analytic methods
3.1 Data
The current study investigates – via linguistic cues – the ways social psychological processes contribute to digital storytelling in health and healthcare contexts. The type of digital storytelling discussed is online consumer reviews of a medical memoir situated in the UK’s healthcare services. The book selected for this study is Do no harm: stories of life, death and brain surgery (Marsh, 2014; hereafter referred to as Do no harm). The book was selected for its relevance to the UK’s healthcare services and its substantial number of online consumer reviews. Do no harm is a memoir by Henry Marsh, a retiring neurosurgeon in the UK’s National Health Service (NHS), providing a first-person account of the author’s experiences in a busy neurosurgical department. As of 2022, when this study was conducted, the memoir had garnered over 4000 user ratings and 2000 verified reviews on Amazon.co.uk.
This study chose consumer reviews of the memoir from Amazon.co.uk as its data source for several reasons. Firstly, Amazon’s book reviewing platform is well regarded for enabling ordinary readers to share their personal experiences with a broad online community (Allington, 2016). Secondly, Amazon is a prominent global e-commerce platform for printed and digital books, with memoirs and biographies being its top-selling genres (McLoughlin, 2022). Book reviews on Amazon.co.uk likely represent broad readerships with diverse opinions and backgrounds, as supported by a study on the varied and comprehensive nature of opinions presented in Amazon book reviews by diverse online communities (Danescu-Niculescu-Mizil et al., 2009). Thirdly, in addition to being a discursive and vernacular literacy practice (Vásquez, 2015), online book reviews also represent a ‘naturalistic’ form of reader response data as opposed to ‘experimental’ data generated from laboratory settings (Hall, 2008; Mansworth, 2022; Swann and Allington, 2009). Thus, online book reviews from Amazon.co.uk can serve as suitable research data for studying storytelling and storyline constructions, enabled by social psychological processes that occur in the reviewers’ own social and personal environments. As observed by Loseke (2019), narrative meaning-making is contingent upon these environments, encompassing cognitive, emotional and moral dimensions. In this light, the reviewers’ personal and social environments contextualise and make sense of their perceptions, interpretations and evaluations of the books, contributing to the meaning, authenticity and relatability of the reviews.
Consistent with the study’s focus on digital storytelling within the UK’s healthcare context, only reviews on Amazon.co.uk that satisfied the criteria of ‘verified purchase only’ and ‘originating from United Kingdom’ were included in this study. The contents of all selected reviews were carefully scrutinised to ensure authentic consumer authorship (Mansworth, 2022; Townsend and Wallace, 2016). Ethical approvals from the researcher’s institution were granted for the study, and personal information was removed for data anonymisation. Upon applying these criteria, a total of 1595 reviews consisting of 70,535 words were collected, forming a corpus of UK-based online consumer reviews of the memoir Do no Harm (Marsh, 2014). The corpus will henceforth be referred to as the DNH-R corpus in the following sections.
3.2. Analytical approach
The analysis of the DNH-R corpus consists of two parts. The first part utilises a top-down, quantitative method, deploying computer-assisted text analysis tools of LIWC-22 and the Sketch Engine to identify linguistic tokens connected to social psychological processes. The second part of the study conducts a bottom-up, qualitative analysis, focusing on how these social psychological processes support and facilitate the modulation and construction of storytelling within the reviews.
Part One of the study (Section 4.1) begins with an analysis of keywords in the DNH-R corpus, followed by a categorisation of the identified keywords into their corresponding types of social psychological processes, as guided by the psychometrics in the LIWC programme. In corpus linguistics research, keywords are words that occur significantly more frequently in a target corpus than in a comparable reference corpus. Keyword analysis is a potent method to identify words that are characteristic of a type of discourse, and provides insights into the main theme or ‘aboutness’ of the discourse domain (Egbert et al., 2020:29). The Sketch Engine was selected to perform the keyword analysis due to its apt reference corpus, the ‘English Web corpus 2020 (enTenTen20)’. Encompassing 36 billion words from various topics and sources, including health and healthcare, it serves as a fitting reference for the DNH-R corpus.
With the target corpus’s focus on health and illness, an emergence of health-related keywords is anticipated. As McIntyre and Walker (2019) and Lugea (2022) articulate, while statistical significance and effect size indicate saliency, they do not necessarily imply interpretative significance in keywords. Thus, this study extends beyond mere keyness scores, incorporating further analyses in Section 4.1 such as concordance and collocates to decipher meanings and interpretations within the corpus. In addition, the multifaceted nature of keywords is considered. As observed in Section 4.1, keywords such as ‘surgeon’ and ‘neuro-surgeon’ belong to both the categories of ‘health’ and ‘social processes’ in the LIWC-22 dictionary, thus enriching the narrative and thematic complexity in the reviews. Finally, as emphasised in the Introduction, an important aspect of the study is exploring how online reviews of medical-themed literature respond to the institutional ideology of healthcare systems like the NHS, which will be analysed in Section 4.2. These rationales ensure that health-related keywords, with their contributions to meaning construction and storytelling in the reviews, are included in the analysis of this study. The study’s multi-layered and systematic analytical approach strives to uncover the intricate layers of meaning shaping the narratives within the DNH-R corpus.
After keywords in the DNH-R corpus are identified, they are categorised into relevant social psychological processes using the LIWC-22 programme, including cognition, affect, perception and social processes (Boyd et al., 2022). Selected keywords are then analysed by the Key Words in Context (KWIC) function in Sketch Engine for a deeper understanding of their usage and meaning within the reviews. The KWIC tool, which provides contextual information surrounding the search words in a corpus, has proven its utility in analysing computer mediated communication (CMC) in healthcare contexts (Brookes, 2020).
Based upon these initial findings, Section 4.2 reports Part Two of the study with a qualitative analysis that scrutinises narrative modulations observed in the user reviews. Each review is treated as an individual text and forms a unit of analysis. As discussed in Section 2.1, each user review is considered a unique narrative in its own right, while also contributing to a collective body of discourse within the corpus. This dual nature of the reviews enables a detailed narrative analysis within individual reviews and maintains a holistic understanding of the shared and overarching narratives across the entire corpus.
The storylines, or narrative threads, intertwining across the reviews, represent a form of collaborative storytelling. In this study, collaborative storytelling does not focus on direct responses between reviews, especially since Amazon’s current review interface does not support direct interactions between reviewers. Nonetheless, narrative collaborations between reviews can be effectively examined through the shared storylines and narrative threads, demonstrating a collective contribution to the discourse surrounding the memoir, thereby forming a unique narrative fabric.
In the qualitative analysis of the study, each user review included in the DNH-R corpus was manually scrutinised for the emergence of storylines and sub-storylines in relation to relevant narrative elements (e.g. actors, events, conditions, thoughts and feelings; see Kleres, 2010). The analytical process was facilitated by the NVivo software for the coding and organisation of the narrative data until no new storylines emerged in the reviews. This analytical approach acknowledges the narrative complexity within each review, capable of encompassing multiple storylines. This process ensures a thorough and nuanced understanding of the narrative threads that run longitudinally through the reviews, creating a rich tapestry of interconnected storylines.
Following the analysis, four overarching storylines are identified within the reviews, spanning from the reviewers’ experiences of reading the book to their reflections on matters of illness, wellbeing and healthcare (see Section 4.2). The study investigates narrative modulations within and between storylines in relation to the social psychological processes identified in Part One of the analysis. Analysis of selected examples in Section 4.2 will reveal the dynamic interaction between these storylines as they emerge and evolve across a variety of narrative dimensions. The section will also discuss how the reviews, in the form of ‘small stories’, challenge some of the canonical narratives in the UK’s healthcare services.
4. The case study: narrative modulations of storytelling in user reviews with social psychological processes
4.1. Linguistic cues for the social psychological processes in the DNH-R corpus
In Part One of the case study, the keywords of the DNH-R corpus are first identified using the keyword analysis function in the Sketch Engine programme. The reference corpus used for calculating the keyness scores is the ‘English Web corpus 2020 (enTenTen20)’ discussed in the previous section. Following this, a selection of top-ranking keywords is arranged into five relevant psychometric categories according to the LIWC-22 dictionary. The term ‘psychometrics’ used in this study refers to the field within psychology that is concerned with the theory and technique of measuring psychological variables. It emphasises the systematic measurement of latent constructs, which are essential but not directly observable attributes or processes, such as the social psychological processes captured by the LIWC-22 tool. For a full discussion of the psychometric categories and the developments of the LIWC’s internal dictionary, see Boyd et al. (2022). In LIWC-22, words can belong to more than one category in the dictionary. The five categories selected from the dictionary are: cognition, affect and social processes, as well as the categories of perception and health. Below is a brief description of the five categories selected for LIWC analyses in this study.
The first three of these categories (i.e., cognition, social and affective processes), with linguistic tokens relating to social and psychological constructs, constitute the main parts of the LIWC internal dictionary. The category of ‘cognition’ represents an overarching dimension that includes sub-categories of all-or-none thinking (e.g., ‘all’, ‘none’, ‘never’), cognitive processes (e.g., ‘think’, ‘know’, ‘understand’) and memory-related processes (e.g., ‘remember’, ‘forget’). Social processes include words that indicate social behaviours (e.g., ‘care’, ‘love’, ‘help’), moralisation (e.g., ‘right’, ‘wrong’, ‘judge’) and social references (e.g., words relating to identities such as ‘we’ and ‘you’, ‘surgeons’ and ‘patients’). Affective processes encompass both positive (e.g., ‘good’, ‘happy’) and negative (e.g., ‘bad’, ‘sad’) sentiments and emotions, which are further organised under four sub-categories of ‘positive tone’, ‘positive emotion’, ‘negative tone’ and ‘negative emotion’. In accordance with LIWC-22, this paper assigns an overall ‘positive valence’ to words found in the ‘positive tone’ and ‘positive emotion’ sub-categories and a ‘negative valence’ to those in the sub-categories of ‘negative tone’ and ‘negative emotion’. For more details of affective processes and their associated valences, see Barrett (2006) and Colombetti (2005).
Beyond these main categories, the LIWC-22 dictionary includes a set of expanded categories. The two expanded categories considered in this study are ‘perception’ and ‘health’. ‘Perception’ includes words that express attention, motion, feeling, as well as sensory experiences (e.g., ‘look’, ‘come’, ‘feel’, ‘cool’, ‘in’, ‘out’). This study includes words in the category of ‘perception’ in the discussions because this category supports embodiment across social psychological processes in storytelling (Ryu and Price, 2021). For example, the word ‘fascinating’ represents both affective and perceptual processes in relation to one’s reading experience. Finally, the category of ‘health’ contains words relating to illness, wellness and mental health. Words in this category are also included in the examination of the DNH-R corpus because they are relevant to the topics and contexts of health and healthcare, as reflected in the reviews.
Top 20 keywords in the DNH-R corpus and their corresponding categories.
Sample concordance lines of the keyword ‘fascinating’.
According to the LIWC-22 psychometrics, the word ‘insight’ is a linguistic cue for cognitive processes. When ‘insight’ appears in online consumer reviews, it would be reasonable to assume that the word is used in its generally accepted sense, meaning ‘(the ability to have) a clear, deep and sometimes sudden understanding of a complicated problem or situation’ (Cambridge Dictionary, 2022). This sense in everyday contexts is largely in line with views in social psychology, where insights are ‘shifts in our understanding, actions, perceptions, goals, and emotions’ (Klein, 2014: 162). These ‘fascinating insights' therefore indicate a transformative aspect in the reviewers’ reading experiences where they develop new perceptions and understanding. These shifts are examples of narrative modulations, capable of reconfiguring the reviewers’ telling of their personal experiences regarding health and healthcare matters. This modulation process fosters the emergence of new, personalised narratives, resonating with the notion in Sections 2.1 and 3.2 where online reviews are seen as stories themselves. Moreover, the sharing of these insights on public platforms like Amazon not only exemplifies narrative modulations in the reviewers’ individual storytelling but also invites and inspires other readers to engage with and contribute to these evolving narratives. These aspects will be further explored in the next section.
The analysis in this section offers initial findings of how social psychological processes shape the storytelling in narrative contexts and form a holistic view of the readers’ experiences and perceptions of health and healthcare services. Building upon these initial findings, the subsequent section (Section 4.2) will apply the concept of ‘narrative modulation’ to analyse the social psychological processes observed in the DNH-R corpus, with the aim of gaining a deeper understanding of the dynamics of storytelling in user reviews.
4.2. Modulations of storytelling with psychological processes
As discussed in Section 2.3, ‘narrative modulation' entails the dynamic configuration of storylines, themes and narrative elements driven by social psychological processes. Storylines, akin to narrative threads, imbue a narrative with depth and contextual richness. Compatible with this view, Wolf (2017) proposes an understanding of overarching storylines as ‘narrative braids’ that are made up of finer narrative threads (i.e., sub-storylines), which interweave and crisscross each other in storytelling, forming a ‘narrative fabric’ (i.e., a story). Adopting this view of storylines, this paper examines the modulating functions fulfilled by social psychological processes in weaving narrative threads in the online user reviews of the memoir.
Main storylines in the user reviews.
These main storylines are each supported by a series of sub-storylines (i.e., finer ‘narrative threads’) in the storytelling. For example, Storyline 1 is made up of sub-storylines including ‘personal backgrounds’, ‘personal opinions of the book and author’, ‘anecdotes during reading’, and so on. For the purpose of this paper, the discussion of narrative modulations mainly focuses on the four overarching storylines outlined in Table 3.
These storylines identified within the user reviews are integral to the narrative dimensions and modulations that constitute collaborative storytelling. These storylines—encompassing personal reading experiences, thoughts about health and wellbeing, perceptions of medical professions, and recommendations —demonstrate dynamic interplay of narrative elements. Each review, often containing multiple storylines, contributes to a larger narrative tapestry, showcasing narrative dimensions such as moral stance, tellership, tellability and embeddedness in individual as well as collective reader experience. This aspect of multi-threaded storytelling will be presented in the following analyses.
Storyline 1, ‘Sharing personal reading experiences’, focuses on reviewers’ reading experiences of the memoir. Given that the main purpose of online reviews is for the users to share their experiences with the product with other consumers, it is not surprising that Storyline 1 is a main component across the reviews in the DNH-R corpus. Closely connected to Storyline 1 are the other three main storylines, which recur across the user reviews. Collectively, these four storylines offer a multi-dimensional view of the reviewers’ storytelling about their lived and living experiences in relation to the book under review. The following two excerpts (Excerpt 1 and 2) illustrate the construction of storylines in relation to narrative modulations. Applying the same colour-coding scheme used in Table 2, relevant words in the excerpts are colour-coded according to their associated social psychological processes, with multicoloured words representing processes from applicable categories under the LIWC-22 psychometrics.
Excerpt 1:
The storytelling in Excerpt 1 initiates with the reviewer’s personal engagement in reading the memoir, referencing Storyline 1. The colour-coded words in this excerpt signify how reading experiences are shaped by cognitive, affective, perceptual and social processes. As delineated in Section 2.3, narrative modulations navigate storytelling based on their strength, scope and stance. This strength is evident in Excerpt 1 through ‘Extreme Case Formulations' (ECFs), linguistic markers that intensify storytelling and reflect deep emotional involvement (Sidnell, 2004; Whitehead, 2015). Words like ‘absolutely’, ‘desperately’ and ‘astonishingly’, alongside phrases such as ‘dreaded picking (the book) up’ and ‘couldn’t put (the book) down’, are examples of ECFs in Excerpt 1 that not only bolster narrative force but also vividly convey the reviewer’s immersive reading experience. These linguistic cues, embodying significant narrative strength, underpin the social psychological processes that craft a compelling narrative of the reviewer’s experience.
As Storyline 1 continues to develop in the review, it intertwines with the other storylines. Social, cognitive and affective processes (e.g., ‘surgeons’; ‘humans’; ‘make mistakes’; ‘doom’; ‘successes’ etc.) collectively mobilise available narrative resources in the co-text and context (e.g., the non-highlighted texts in the excerpt) for the construction of Storylines 2 and 3 (i.e., ‘thinking about health and wellbeing’ and ‘perceiving medical professionals and healthcare services’). These processes help extend the scope of narrative modulations from an initial focus on the reviewer’s personal experiences to issues relating to life, health and healthcare in a much broader social context. Furthering the modulation process, Storyline 4 (‘making recommendations’) emerges when the reviewer addresses potential readers directly with social references ‘you’ and ‘your’, and employs affective responses relating to ‘triumph’ and ‘tragedy’ to reinforce the significance of the storytelling. The perceptual processes in the concluding remark (e.g., ‘stay in’; ‘long’) emphasise the lasting impact of the book on the readers. This correlates with the narrative dimension of ‘tellability’ in small stories research (Ochs and Capps, 2001; adapted in Vásquez, 2014), where the teller justifies the significance of their personal experiences in consumer reviews. This, in turn, enhances the strength of narrative modulations in the reviewer’s storytelling. ‘Making recommendations’ is a recurring storyline found in the DNH-R corpus. Manifested as a speech act, this storyline highlights the interactivity and collaboration in the reviews, creating a reader-directed narrative through personal, engaging communication (Vásquez, 2014). This performative aspect aligns with the small stories approach, which emphasises narratives — including online user reviews — as dynamic, interactive processes that actively mediate experiences and understandings in digital spaces (Georgakopoulou et al., 2023).
In addition to scope and strength, the social psychological processes that participate in modulating the storylines in Excerpt 1 also contribute to an overall positive stance. This is supported by the five-star rating given in the review, as well as the affective processes with a positive valence (e.g., ‘well written’; ‘astonishingly honest’). The positive stance is also established in Excerpt 2, which displays a progressive modulation process throughout its storytelling.
Excerpt 2:
The positive stance in Excerpt 2 is established at the start of the review by processes associated with a positive valence (i.e., ‘absolutely brilliant’). This initial stance strengthens as the storytelling develops, where cognitive, social, perceptual and affective processes join force to give a positive appraisal of Storyline 1 (e.g., ‘thought provoking’; ‘honest’; ‘important’). Similar to Excerpt 1, the scope of storytelling in Excerpt 2 grows broader alongside the development of storyline 2 and 3 with the aid of cognitive and social processes (e.g., ‘questions’; ‘life’; ‘difficult decisions’, etc.). These processes further navigate the development of Storyline 4 (i.e., ‘making recommendations’) with the statement that the book is ‘something everybody should read’. Throughout the review, the modulating strength of the social psychological processes develop progressively. At the end of the review, the social psychological processes deliver a compelling force of modulation (e.g.: ‘Go buy (this book)’; ‘(will) not regret’). These processes extend the storytelling from the reviewer’s ‘lived’ narrative about reading the memoir to a ‘living’ narrative in which past experiences lead to new experiences by having ongoing impacts on the reviewers and potential readers. For instance, potential readers are encouraged to ‘buy the book’, and the reviewer ’cannot wait to read the author’s second book’. In doing so, both the ‘lived’ and ‘living’ aspects are foregrounded in the reviewer’s narrative, illuminating past, ongoing and future experiences (Ochs and Capps, 2001). This dynamic of bridging the personal narrative to broader relevance and readership in online reviews is supported by Vásquez’s research, which notes that online reviews often conclude with reader-directed speech acts, functioning as codas that connect personal experiences to their significance for the reader (Vásquez, 2014: 148). For further discussions of the ways codas contribute to rhetorical relations and intentionality in online consumer reviews, see Chik and Taboada (2020) and Virtanen and Stormbom (2024).
The storylines discussed so far within the DNH-R corpus are modulated by social psychological processes with predominantly positive valences. Being the primary ‘narrative threads’ that weave the fabric of storytelling, these storylines strengthen an overall positive tone and sentiment across the reviews. Despite this, contents with negative or critical attitudes can also be observed in the reviews. As explained in Section 3.2, Part 2 of the study conducted a qualitative analysis of the online reviews with the aid of the NVivo software, where storylines and relevant narrative elements were identified and analysed within the reviews in the DNH-R corpus. This close examination of the narrative data reveals that most of the contents containing reviewers’ negative attitudes relate to a part of Storyline 3, namely ‘perceiving healthcare services’. More specifically, the reviewers’ negative sentiments are directed at the ‘bureaucracy’ and ‘(mis)management’ within the NHS, as Figure 1 shows. This differentiation between the general positive reception of the book and the specific critiques of the healthcare system illustrates the complexity and multi-dimensionality of the narrative discourses within the reviews, supporting an intricate configuration of the intertwined narrative threads. Word Sketch visualisation of the NHS in the DNH-R corpus.
Figure 1 is a ‘Word Sketch’ visualisation generated by the Sketch Engine software. It displays the collocates of the search word ‘NHS’ in the DNH-R corpus. Words such as ‘bureaucracy’, ‘management’ and ‘managers’ are displayed as top collocates of the search word ‘NHS’, indicating bureaucracy and management as the main aspects when the reviewers mention the NHS in their reviews. To illustrate the reviewers’ perceptions of the NHS, Excerpts 3, 4 and 5 provide examples of the construction and modulations of Storyline 3 surrounding the word ‘NHS’ and its collocates with the aid of social psychological processes:
Excerpt 3:
Excerpt 4:
Excerpt 5:
Contrary to the previous excerpts that convey predominately positive valences, Excerpts 3-5 contain affective processes with a negative tone (e.g., ‘frustration’, ‘ridiculous’, ‘spineless’, ‘greasy’ and ‘heartache’). Whilst these negative sentiments are aimed at bureaucracy and mismanagement within the NHS, affective processes associated with a positive tone, for instance, ‘admire’, ‘enthusiasm’, ‘appreciate’ and ‘laughed’, are used in support of the dedications and commitments of medical professionals who strive to deliver patient care within the NHS. Social psychological processes not only motivate this divide of positive and negative sentiments within Storyline 3 (i.e., ‘perceiving the medical professions and healthcare services’), they also justify the reviewers’ opinions by constructing a counter-narrative (Bamberg and Wipff, 2020). This counter-narrative challenges the canonical narrative of the NHS that promotes its efficiency and patient care. Excerpt 6 below contains texts selected from the NHS long term plan (2019), which is a key publication for stakeholders including the general public in the UK 3 . This example illustrates the NHS’s mainstream narrative that promotes its service efficiencies and successful leadership and management.
Excerpt 6: Examples from the NHS long term plan (2019: 9-11)
Applying the same colour-coding procedure used in the previous excerpts, linguistic tokens in the examples are highlighted according to their corresponding categories in the LIWC-22 dictionary. Social psychological processes with a positive valence – for example, ‘high’ ‘satisfaction’, ‘generally improving outcomes’, ‘strong overall’ (efficiency) – are found in the examples, putting the NHS in an overall positive light. When addressing areas needing improvement in the NHS infrastructure, there are social psychological processes that convey a negative sentiment (e.g., ‘frustration’; ‘missed opportunity’). However, the storytelling in Excerpt 6, as well as throughout the NHS long term plan, is sustained by a dominant storyline where the NHS Leadership has been consistently delivering ‘high levels of patient satisfaction’ and ‘will continue to drive efficiencies’ (cited from Excerpt 6). This finding accords to a recent study of the NHS’s online communications to the general public (Chałupnik and Brookes, 2021), which revealed that NHS organisations have seen an increased use of bureaucratic jargon in online communications as part of the ongoing marketisation of the healthcare system. The narrative in Excerpt 6 forms a direct contrast with the storytelling in the user reviews, which criticise ‘the depressing charade that we can be more “efficient” with fewer resources’, hindering medical professionals’ caregiving with ‘more and more layers of management’ (cited from Excerpts 4 and 5). Importantly, although the same linguistic token ‘we’ is used in both sets of excerpts, it indicates different social processes in each context. In the examples from the NHS long term plan, the word ‘we’ is used as a social process to position the NHS leadership as the authoritative teller who controls the dominant narratives of UK’s healthcare services. In the user reviews that have been discussed in this study, ‘we’ is used by the reviewers as a social process to represent themselves as members of a community. In relation to the healthcare context in the UK, ‘we’ stand for the general public and healthcare clients who appreciate the services and dedications of frontline healthcare professionals; yet detest the bureaucracy and mismanagement within the NHS. The analysis of the social psychological processes in their narrative context illustrates the dynamic construction of storytelling. On one hand, the storytelling challenges the NHS’s canonical narratives that prioritise cost efficiency over patient care. On the other hand, it offers sympathy, appreciation and support for medical professionals’ resistance and resilience to the bureaucracy and mismanagement in the UK’s healthcare services.
This juxtaposition of opposing sentiments within the review narratives exemplifies the reviews’ fluidity in the ‘contextual embeddedness’, enabling meaning and identity being regularly constituted and continually under construction (Bamberg and Wipff, 2020: 72). This is evident in the above excerpts, where the narrative modulation processes with mixed valences facilitate a dynamic oscillation between the reviewers’ frustration with administrative bureaucracy and inefficiencies in the NHS and appreciation for the dedication of medical staff.
This section has discussed the construction of the four main storylines in the user reviews in the DNH-R corpus. Based on the linguistic cues for social psychological processes and guided by the concept of ‘narrative modulation’, the analysis of the narrative data examines the stance, scope and strength of the relevant social psychological processes that serve to modulate storytelling across the user reviews. The analysis reveals that the modulation process supports emergent aspects of a storyline (e.g., constructing new insights and creating new experiences), sustains and advances existing storylines (e.g., developing one’s view on issues relating to health, wellbeing and the quality of life), as well as to challenge canonical narratives (e.g., constructing counter-narratives about the bureaucracy and mismanagement in the NHS). Informed by the small stories approach, the discussion of the user reviews also addresses the dynamic and interactive aspects of storytelling in the narrative modulation process. These narratives, reflecting a spectrum of emotions and judgments, vividly illustrate the complex interplay between individual experiences and broader societal perspectives on healthcare. From the blended appreciation and critique of the NHS to personal reflections on health and medical professionalism, the user reviews weave together diverse storylines. This rich narrative tapestry underscores the power of small stories in shaping and modulating public discourse, demonstrating how each unique narrative contributes to a collective understanding of the healthcare system.
5. Conclusion
Consumer reviews have become an increasingly important type of digital storytelling, wherein reviewers share personal experiences in a narrative form with online audiences (Vásquez, 2012). As these digital narratives of personal experiences continue to grow as a vital form of human communication, there is an increasing need for research to widen our understanding of their evolving forms, contents and contexts. This study addresses significant research gaps identified in the existing literature: the underexplored areas of the social psychological processes in digital health narratives, the lack of examination of online consumer book reviews related to healthcare and wellbeing, and the opportunity for methodological advancements that enable thorough examinations of the ways social psychological processes contribute to digital storytelling in healthcare contexts.
To bridge these gaps, this research develops a mixed-methods approach, integrating quantitative, computer-assisted text analysis tools, namely LIWC-22 and the Sketch Engine, and qualitative narrative analysis guided by narrative modulations and small stories research. Using this integrated approach, this paper systematically examines the online user reviews of the medical memoir Do no harm: stories of life, death and brain surgery by Marsh (2014). The study underscores the ways in which social psychological processes serve as ‘narrative modulators’ to configure and navigate the construction of multiple storylines across the consumer reviews. It also reveals how these online reviews reflect and challenge prevailing institutional ideologies within the UK’s healthcare system.
This study makes the following original contributions to advance the fields of narrative studies, social psychology and healthcare communication. Firstly, it develops and validates a mixed-methods framework that synthesises computer-assisted text analysis with qualitative narrative inquiry, providing a comprehensive tool for examining digital storytelling in healthcare contexts. This approach overcomes the limitation of the context-free, dictionary-based approach of LIWC-22 (Boyd, 2017) by combining LIWC analyses with corpus-linguistic methods, as well as narrative and discursive studies of the texts in their embedded contexts, guided by the principle of narrative modulations. Secondly, this research offers empirical findings that highlight the dynamic role of social psychological processes in shaping online consumer reviews, revealing how these processes configure emerging, evolving and countering storylines. Thirdly, the research provides critical insights into how digital health narratives reflect and contest institutional ideologies, informing healthcare practices and policies.
The research findings contribute to healthcare practices and policies in several ways. By highlighting the lived experiences and perceptions of healthcare consumers, this study offers valuable insights into patient concerns, expectations and experiences with the healthcare system and healthcare professions. These insights can guide healthcare providers in improving patient communication, tailoring care to meet patient needs and addressing systemic issues that affect patient satisfaction. Additionally, by uncovering how digital narratives challenge institutional ideologies, the study suggests areas where healthcare policies may need to be re-evaluated or reformed to better align with patient experiences and expectations. This includes addressing bureaucratic inefficiencies and improving overall management within the NHS. Such reforms can lead to more patient-centred policies and practices that foster trust and engagement between patients and healthcare providers. Furthermore, understanding the nuanced views on health and illness presented in these narratives can help in developing more empathetic and responsive healthcare environments.
Overall, the study enhances the understanding of digital health narratives among academics, practitioners and policymakers by elucidating the interplay between individual experiences and broader socio-cultural discourses. It not only bridges interdisciplinary gaps between narrative studies, social psychology and medical humanities, but also sets the stage for future research on the lived and living aspects of digital storytelling in healthcare contexts and beyond. These findings have practical implications, enriching methodologies and analytical techniques for exploring the complexities of digital narratives and their impact on healthcare communication and public perception.
Footnotes
Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
