Abstract: Engaging audiences is widely regarded as essential for science communicators aiming to bridge the gap between the expert knowledge they convey and their target audience’s presumed lack of expertise. Drawing on the concept of discursive interpersonality (Suau et al., 2021), this study explores a corpus of 30 digital feature articles from the SciDis Corpus (Pascual and Sancho-Ortiz, 2024) in order to identify and analyse pragmatic strategies and their associated (meta)discursive features used by science communicators to make expert knowledge accessible and engaging for lay audiences. Results show that writers deploy a range of strategies and resources, including not only interactional metadiscourse features, as expected, but also strategic resources adapted from journalistic and narrative discourses. In sum, the study shows that audience-oriented pragmatic strategies (Lorés, 2024b) extend beyond traditional metadiscourse (Hyland, 2005b). Further conclusions point to editorial and disciplinary differences in the use of pragmatic and (meta) discursive engaging strategies.
Keywords: Engagement, Digital discourse analysis, Science communication, Popularisation, Metadiscourse.
Summary:: 1. Introduction: the (digital) popularisation of scientific knowledge. 2. Engagement as an essential aspect of popularised scientific knowledge. 3. Discursive interpersonality and the exploration of popularised scientific discourse. 4. Corpus and methods. 5. Results: pragmatic strategies and (meta)discursive resources. 5.1. Hooking the audience’s attention. 5.2. Creating an emotional bond with the audience. 5.3. Enhancing direct interaction with the audience. 5.4. Establishing a common territory of experience. 6. A quantitative analysis: discussion of data. 7. Conclusion. Acknowledgment. References. Appendix.
How to cite: Lorés, R. (2025). Discursive Interpersonality: Engaging Audiences in Digital Feature Articles. Círculo de Lingüística Aplicada a la Comunicación 103 (2025): 199-274. https://dx.doi.org/https://doi.org/10.5209/clac.101048
1.
Introduction: the (digital) popularisation of scientific knowledge
The traditional divide between scientists and the public has been
questioned for decades now (Jones et al., 2015; Pilkington, 2018;
Freddi, 2020; Pascual, Plo-Alastrué and Corona, 2023), leading to the
recognition that popular science is designed for diverse audiences,
including experts, displaying varying levels of knowledge, needs, and
expectations (Myers, 2003; Hyland, 2010; Banks and Martino, 2019).
Readers are no longer seen as passive consumers but as active
participants in the broader societal discourse surrounding science
(Pilkington, 2018: 14). In this view, the priority lies not just in
explaining scientific events but also in interpreting their societal
significance, thereby promoting the democratisation of scientific
discourse (Moirand, 2003; Motta-Roth and Scherer, 2016). The
democratisation of science has been enhanced by the role played by
digital technology in reshaping communication and knowledge-sharing. The
tools afforded by Web 2.0 and social media have significantly enhanced
the sharing of textual, visual, and audio content while enabling new
forms of interaction and changing “power dynamics” (Sterk and Van Goch,
2023: 20), thus altering traditional dynamics between scientists and the
public. This trend has sparked considerable academic interest (e.g.,
Luzón, 2013; Bondi et al., 2015; Luzón and Pérez Llantada, 2019; Bondi
and Cacchiani, 2021; Facchinetti, 2021; Labinaz and Sbisà, 2021; Lorés,
2023, 2024b; Pontrandolfo and Piccioni, 2022; Plo-Alastrué and Corona,
2023; Mur-Dueñas 2024; Pascual, 2024, 2025; Sancho-Ortiz, 2025). Thus,
new genres emerging from digital innovation (e.g., blogs, social media
posts, video abstracts, podcasts) have been explored, as well as
traditional genres migrating and adapting to online formats with new
forms and functions (e.g. feature articles, research digests, breaking
news, among others).
At the basis of the democratisation of science is the idea of
popularisation, a concept that goes beyond mere simplification of
scientific knowledge. The act of popularising and its resulting product
(popularized discourse) can be tackled from many theoretical
perspectives (discourse analysis, pragmatics, communication studies,
media studies, sociology, history, science), as it involves language in
use and the interaction between different actors, in various contexts,
on different media, and about different disciplinary fields.
The idea of a continuum is pervasive in science communication
studies, with popular discourse at one end and academic discourse at the
other (Hilgartner, 1990; Myers, 2003; Hyland, 2010; Luzón, 2013; Sterk
and Van Goch, 2023), and scientific findings travelling along a
“communicative path” (Bucchi, 2008). This “continuum view” (Giannoni,
2008; Sterk and Van Goch, 2023), in which popularisation is seen as a
matter of degree, somehow opposes the view of the “deficit model”, which
assumes that popularisation discourse involves a one-way process from
experts to lay audiences, and which understands that the main function
of popularisation is pedagogical, as the audience is always in lack of
knowledge (Myers, 2003; Motta-Roth and Scherer, 2016).
But why focus on popular discourse? As Sterk and Van Goch (2023: 4)
indicate, “[a]lthough the research fields of science communication and
science journalism have existed for several decades, controversy
surrounding research findings is one of the reasons why it remains
important to discuss the discourse from an academic stance”. I would
also add that deepening into our knowledge of how science is
communicated and “ popularised” may help us distinguish between good and
bad practices of science communication, at a time where the concept of a
“reliable voice” seems to have more to do with the number of followers
than with the quality and the reliability of the science transmitted.
Thus, in line with Myers (2003: 267), when analysing popular science
texts, “[w]e need to question who the actors are, how the various
discourses interact, what modes are involved, and what is communicated -
and we need to consider what these questions imply for text analysts”,
more so when these popular texts take advantage of the affordances of
the digital mode. Popularisation is not only about information: it is
also about interaction (Moirand, 2003; Myers, 2003).
In the following sections, the concept of engagement and discursive
interpersonality will be discussed, and the object of study will be
introduced, followed by the presentation and discussion of qualitative
and some quantitative results. Some final remarks by way of conclusion
will also be offered.
2.
Engagement as an essential aspect of popularised scientific
knowledge
Engagement is inextricably linked to the popularisation discourse of
science. It signals writers’ awareness of the audience’s presence,
expectations, and needs. The intent to link up with the audience and
engage them in some kind of interaction (a question of degree depending
on the media and technical affordances associated with it) is at the
heart of the general purpose of any sample of popularised discourse, and
the factor that explains many discursive choices made by the authors,
whether scientists or mediators.
Among others, three studies on popularisation have singled out
engagement as a major theme in scientific popularised discourse and deal
explicitly with it: Hyland (2010), Luzón (2013), and Sterk and Van Goch
(2023), who build their proposal on the previous two. By formulating the
concept of “proximity”, which refers to the writer’s control of
rhetorical features, Hyland (2010) identifies five “facets” of proximity
in a corpus of 120 research articles and 120 popular science articles:
Organisation, Argument Structures, Credibility, Stance, and Reader
Engagement. The facet of Reader Engagement is drawn from the
metadiscourse model (Hyland, 2005b) and refers to:
an alignment dimension of interaction where writers acknowledge and
connect to others, recognising the presence of their readers, pulling
them along with their argument, focusing their attention, acknowledging
their uncertainties, including them as discourse participants, and
guiding them to interpretations (Hyland 2010: 125).
Sticking close to the metadiscourse model, two resources of reader
engagement are explored: reader pronouns and questions.
In her study of science blog posts, Luzón (2013) offers a wide view,
exploring the discursive strategies used by science bloggers to
communicate and recontextualise scientific discourse and to engage their
diverse audiences with scientific issues. Luzón (2013) identifies two
rhetorical strategies to recontextualise scientific information:
strategies to tailor information and strategies to engage the reader.
Focusing on the latter, the author (2013: 446) offers a list of “devices
to signal awareness of their audience, engage the readers and guide them
to particular positioning”. This list of resources offers a variety of
features which combine more structural devices (e.g. titles) with
typical engagement metadiscursive resources at the lexicogrammatical
level (e.g. questions, inclusive pronouns, references to reader) and
with pragmatic resources (e.g. humour, expression of feelings or
emotional reactions).
With the purpose of designing an overarching model which is usable in
any subgenre of popularisation discourse, among other aims, and based on
Hyland (2010) and Luzón (2013), Sterk and Van Goch (2023) propose an
analytical framework of 34 strategies captured under five themes:
Subject Matter, Tailoring Information to the Reader, Credibility,
Stance, and Engagement. As far as Engagement is concerned, the authors
(2023: 56-57) include eight different strategies, namely:
titles/subheadings, references to popular lore, beliefs and popular
culture, self-disclosure of the authors’ public or personal life,
inclusive pronouns, features of conversational discourse, questions,
humour, and explicit self-reference, thus situating at the same level
pragmatic strategies (e.g. “self-disclosure of the authors’ public or
personal life” or “humor”) and (meta)discursive features (e.g.
questions, inclusive pronouns, explicit self-reference, etc.). In my
view, however, the use of these (meta)discursive resources in
popularised practices obeys to a series of pragmatic strategies, enacted
by the writers to foster an engaged attitude on their audience. Thus,
the focus of the present study as part of
the exploration of popularising discourse is the identification and
discussion of pragmatic strategies used by mediators (journalists and
scriptwriters) to communicate scientific findings in an engaging way.
These pragmatic strategies are instantiated by means of (meta)discursive
resources which include lexicogrammatical but also multi-sentence
features, as well as some multimodal resources which exploit the
technical affordances of the digital media.
3.
Discursive interpersonality and the exploration of popularised
scientific discourse
Discursive interpersonality (Suau et al., 2021) stems from two
concepts: discursive turn (Jaworski and Pritchard, 2005), which
sees discourse as a departure point for analysis, and
interpersonality (Lorés-Sanz et al., 2010; Mur-Dueñas et al.,
2010), considered as an umbrella term, broader than interpersonal
metadiscourse, which offers coverage for diverse discursive and textual
phenomena that seem to be excluded from the metadiscourse model.
Following Mur-Dueñas et al. (2010: 83), interpersonality is defined as
an encompassing term that “can integrate all the linguistic resources
that writers can use to bring about the intended interpersonal effects
and make their text successful”.
Based on studies of different academic and professional genres, Suau
et al. (2021) coin the concept of discursive interpersonality
to problematise the notion of interpersonal metadiscourse. Interpersonal
metadiscourse has proved to be conditioned by contextual variables, such
as genre, discipline, language, and corpus, as has been shown in a large
number of studies (Dahl, 2004; Yakhontova, 2006; Lorés-Sanz, 2011;
Ivorra Pérez, 2015; Suau-Jiménez, 2016; Herrando-Rodrigo, 2019), thus
making it very difficult to explain certain linguistic and discursive
phenomena from the rigid view offered by the more conventional
metadiscourse model.
The notion of discursive interpersonality offers a more flexible
alternative to the conventional application of the metadiscourse model,
as conceived by Mauranen (1993), Crismore et al., (1993), Hyland and Tse
(2004), and Hyland (2005b), among others, incorporating propositional
features which also contribute to the interpersonal play between writers
and readers.
The present study takes one step further in the exploration of
discursive interpersonality by applying this concept to the analysis of
popularised scientific discourse published online. While the discourse
of science popularisation has been extensively investigated from various
linguistic, discursive and rhetorical perspectives (Myers, 1991;
Calsamiglia and Van Dijk, 2004; Giannoni, 2008; Hyland, 2010; Luzón,
2013; Gotti 2014), a significant gap remains in our understanding of the
interplay between pragmatic strategies and (meta) discursive resources
used in online scientific popularisation. Adopting the perspective of
discursive interpersonality enables us to examine how expert writers
engage their audiences through discursive resources rooted in pragmatic
intents and extending beyond interpersonal metadiscourse.
This study will focus on the digital feature article. Traditionally
an offline genre, feature articles have migrated to online platforms,
retaining their function in the ecosystem of popularising scientific
practices while potentially leveraging the affordances of digital media.
Their purpose is to disseminate research within a specific subject area
to heterogeneous audiences, including both specialists and the general
public interested in the topic, for whom complex research must be made
accessible. Digital feature articles are basically characterised by
three key aspects. Firstly, they are writer-mediated discourse, usually
written by a science journalist or scriptwriter. Secondly, they
summarise multiple research findings, rather than focusing on a single
scientific paper; that is, they represent an instance of digital
research dissemination genres, a mediated form of providing public
access to scientific knowledge (Lorés, 2024b), in contrast with
digital summary genres (Luzón, 2023), written by the scientists
themselves to summarise and promote their published research. Finally,
they are published in scientific magazines (as opposed to academic
journals), often organised around disciplinary topics.
Thus, focusing on the scientific digital feature article as an object
of study, in this contribution I seek to answer the following research
questions:
What audience-oriented pragmatic strategies are used in digital
feature articles on scientific topics to foster engagement?
Which (meta) discursive resources realise these audience-oriented
pragmatic strategies in digital feature articles on scientific topics?
Do multimodal resources play any role in the pragmatic strategies
identified?
How does the concept of discursive interpersonality
contribute to understanding the discursive action taken in digital
feature articles on scientific topics?
4. Corpus and methods
The corpus collected for the present study includes a dataset of
feature articles written by professional science journalists and
published in two widespread online publications: The Smithsonian
Magazine (www.smithsonianmag.com/)
(Smith) and Popular Science (www.popsci.com/) (PopSci).
The dataset consists of a total of 30 texts (15 from each publication)
on the topics of mental and physical health (see Appendix). Medical and
health issues are topics of social interest which greatly attract the
general public’s attention, as some informed sources state (see, for
instance, the “Encuesta de percepción social de la ciencia y la
teconología”, published by FECYT, in the Spanish national context, or
the “European citizen’s knowledge and attitudes towards science and
technology”, published by the Eurobarometer of the European Commission,
that attests the relevance given by European citizens to research on
healthcare
issues). Moreover, it is not risky to state that health questions are
some of the most Googled topics and a recurrent subject of online
searches for the general public (see the 1000 most asked questions in
2025 meetglimpse.com/top-searched/most-searched-questions/),
especially after a specific health crisis, like Covid 19.
The 30 texts selected were authored by a diversity of writers (10
different authors in Popular Science and 14 in The
Smithsonian Magazine), which ensured a variety of writing styles.
The dataset, which makes up a total of 48,537 words, is considered to be
representative of its type, taking into account that “providing indepth,
nuanced, and contextualized insights into an interaction, a particular
social phenomenon, or a specific discursive practice, often involves
carrying out highly detailed, fine-grained qualitative analysis, which
necessitates smaller rather than larger datasets” (Vásquez, 2022: 7). As
there was no section entitled “feature articles” in any of the
publications, they were selected attending to their completion of
criteria to be considered as such: (i) they were writer-mediated
(written by professional journalists); (ii) they summarised and
presented scientific findings from various sources on a single topic;
(iii) they were not presented as news, and, (iv) they were not
transcriptions of podcasts.
The distribution of words per publication is detailed in Table 1. As shown, there was a
significant difference in the length of the publications, which, in the
case of the quantitative data, was normalised per 1000 words.
Table 1. Number of words of digital feature articles by publication (SciDis
corpus)
No. of texts
Average number of
words
Range of words in each
text
Total word number
The Smithsonian Magazine
15
2031,73
1415-3053
30,476
Popular Science
15
1204,07
576-2256
18,061
48,537
To investigate digital feature articles from the point of view of
discursive interpersonality and operationalise the study, I here
followed Pascual (2024: 71) and understood pragmatic strategy as “the
visible textual proof of speakers’ utmost intentions, in varying degrees
of explicitness, for the completion of a communicative purpose”.
Pragmatic strategies are then instantiated as a range of various
(meta)discursive features which are identified and realised through a
variety of pragmatic and discursive roles that mediators play in
popularising specialised knowledge.
A primarily qualitative discursive approach was taken here to explore
pragmatic strategies and their instantiations in engaging audiences in
digitally-mediated scientific discourse. This approach allowed the
identification and classification of the cases as categories which were
then formulated as a taxonomy. Categories were developed in an inductive
(data-driven) way. Quantification was complementary in this study, whose
aim and design focused mainly in the identification of a representative
range of pragmatic strategies used in on- line writer-mediated
popularised scientific texts and their associated (meta)discursive
resources. As Brunner and Diemer (2022: 214) state: “Any analysis of
digital discourse needs to take into account its complexity. Although it
is useful to quantify key features and to limit the dataset to allow
some quantitative interpretation, the key to understanding digital
discourse is careful qualitative analysis”. Thus, quantification is
present at various specific points where realisations are well-delimited
and devoid of any formal or functional fuzziness or ambiguity.
The pragmatic strategies included in the present study illustrate a
wide range of the possible intents used in mediated popularised
scientific discourse. The identification of pragmatic strategies stopped
adding types when a point of saturation was reached (Holton, 2007),
meaning that each new strategy and their associated resources was less
representative than the ones already identified, and suggesting that
those listed provided good coverage. Thus, some data-driven yet less
representative types were discarded and not included in the taxonomy.
This was the case, for instance, of a strategy detected in one of the
textual instances in which the audience is taken to act as scientists,
and attributions of scientific knowledge are made through the use of the
inclusive personal pronouns we (“When it comes to lab experiments on
parent-child attachment, we may know everything we need to know— and
have for more than 60 years” (PopSci1))
To carry out the identification of pragmatic strategies, the texts
were first analysed manually, by means of close reading. Once a
preliminary taxonomy had been designed, the qualitative data analysis
software NVivo was used. The software allowed me first to codify the
strategies in the texts in a systematic way and also to refine the
initial taxonomy by including new cases which had been overlooked in the
manual analysis. To complement the qualitative findings offered by
NVivo, the corpus analysis toolkit AntConc was used for more specific
lexicogrammatical realisations, such as the use of personal
pronouns.
5.
Results: pragmatic strategies and (meta) discursive resources
The present study explores the pragmatic and (meta)discursive
resources that scientific communicators and, more specifically, those
who act as mediators (journalists and scriptwriters) employ in their
online texts to engage their audiences. The pragmatic strategies
oriented towards engaging audiences are taken to be
“audience-oriented” strategies (Lorés, 2024b: 9), which “help readers
grasp complex information and enable them to engage in a meaningful
dialogue with experts”. The salient strategies identified in the corpus
under analysis, which do not exhaust the list of potential strategies
that mediators may use, are geared towards establishing a horizontal
relation between writer and reader, an interaction among equals in which
writers address readers as members of a community of interested
peers.
The exploration of audience-oriented pragmatic strategies led to the
identification of four different strategies used by mediators to present
potentially complex issues to lay audiences by introducing those issues
in a sufficiently engaging and motivating way.
The pragmatic strategies identified are:
– Hooking the audience’s attention
– Creating an emotional bond with the audience
– Enhancing direct interaction with the audience
– Establishing a common territory of experience
These strategies will be presented together with the (meta)discursive
resources that typically realise them, ordered by their frequency.
5.1. Hooking the audience’s
attention
Writers make use of resources taken from various discursive
conventions, thus leading to the adoption of a hybrid type of discourse,
partly scientific, partly journalistic. The following features
illustrate this point:
i) Popular press titles
Using resources typical of the popular press seems common in
scientific research dissemination discourse. The similarities between
the titles chosen for the feature articles under study and those
appearing in publications as popular as The Huffington Post are
evident. Following Scott (2023) and Finkbeiner (2024), Escandell (2025:
185) claims that, although in the past the function of press titles was
to provide informative summaries of news articles, in recent years, the
aim of promoting readers’ curiosity to attract their attention is
gaining ground. This engagement purpose in the use of titles is
transferred to scientific popularisation discourse.
Three different formulations for titles were identified:
– Type 1. Titles which create expectations on the reader by
announcing solutions to health problems and concerns:
(1) How to Deal With Work Stress and Recover From
Burnout (Smith1)
(2) Burnout is Real. Here’s How to Spot It — and Recover
(PopSci2)
(3) Is Saliva the Next Frontier in Cancer Detection?
(Smith6)
The realisation of these titles is typically the “how to” discursive
type. Others make use of evaluating words such as truth, real,
improve or new, thus highlighting the validity and
reliability of the information provided:
(4) A New Tool Could Help Detect Breast Cancer Earlier
(Smith 7)
– Type 2. Titles that advance research findings by summarising them,
following a deductive approach (Moirand, 2003; Luzón, 2013; Hyland and
Zou, 2020; Lorés, 2024a), contrary to the inductive approach which
characterises the research articles from which these feature articles
emerge. The fact that findings are already presented in the title ends
with the suspense created by Type 1 but triggers the audience’s interest
in other ways, by focusing on the path followed to achieve those results
and on their justification.
(5) This Dissolvable Impact Could Revolutionize Pain
Management (Smith11)
(6) 4-Day Work Week Shows Big Benefits for Both Workers and
Employers in UK (PopSci5)
– Type 3. Titles that problematise common beliefs (sometimes
formulated as warnings), thus raising the interest of the reader:
(7) Muscle Memory Is Real, but It’s Probably Not What You
Think (PopSci3)
(8) Millions of Americans Take Aspirin to Prevent Heart
Disease — but Should They? (PopSci7)
Questions in titles, which appeal directly to readers, serve various
communicative purposes, such as problematising common beliefs, as shown
in (8). Other communicative aims include creating expectations about
solutions for medical conditions, as seen in (3) above, or expressing
doubts that might be in people’s minds, as shown in (9):
(9) Should Parents Worry About New Research Linking Kind’s
Mental Health and Individual Sports? (Smith14)
Questions address the reader directly and represent the most obvious
strategy for dialogic involvement (Hyland, 2002). Their use in feature
article titles underscores their primary “hooking function,” projecting
a question/answer structure onto the text and, consequently, suggesting
that the text will provide solutions for the problems posed in the
title.
The predominance of the first type of title over the others is shown
in Table 2:
Table 2. Discursive types of titles. Raw numbers and percentages.
1. Create expectations by
announcing solutions or information about problems or
worries
2. Advance
findings
3. Problematise common beliefs
(formulated as warnings)
Popular Science
6 (18%)
6 (18%)
3 (10%)
The Smithsonian Magazine
12 (36%)
2 (6%)
1 (3%)
Total
18 (54%)
8 (24%)
4 (12%)
As shown by the data, more than half of the titles (54%) obey to type
1 and strategically create expectations that contribute to enticing the
readers’ attention so that they end up reading the texts, sometimes even
fringing the strategy of clickbaiting, a technique by means of which
headlines exaggerate claims with the purpose of enhancing traffic to a
particular webpage, as in the following title:
(10) How Dangerous Is Myocarditis? The Truth About the
Scary-sounding Condition (PopSci14)
Assuming the limitation of the corpus, the data seem to point towards
preferences hold by each publication as to the discursive type of title
favoured. Whereas Popular Science formulates titles of type 1
(create expectations) and type 2 (advance findings) to the same extent,
The Smithsonian Magazine clearly favours type 1. These
divergences might point towards editorial and writer’s preferences,
which do not change the fact that mediators make use of a range of
different discursive types to formulate their titles as open gates to
the text.
ii) Narrative opening
There is a significant number of feature articles that start with the
narration of a story, which is usually a common citizen’s or a
scientist’s:
(11) In the late 1950s, dentist and U.S. Navy Captain Kirk C.
Hoerman, then a young man in his 30s, attempted to answer a bold
question: Might the saliva of prostate cancer patients have different
characteristics from that of healthy people? Could it contain traces of
a disease that’s so far away from the mouth? Without wasting more of
their ownsalivaon elaborate discussion, Hoerman and his colleagues from the
department of dental research at the Naval Training Center in Great
Lakes, Illinois, got down to work. (Smith6)
This strategy has also been identified in other popularised texts,
such as newspaper articles (Moirand, 2003), and research digests (Lorés,
2024a). The writer creates a scenario at the beginning of the text,
intending to arouse the audience’s interest and providing a specific
contextual setting for the scientific information presented in the
article The strategy of opening the text with a narrative also
“humanises” the scientist to the point of aligning them with the
audience.
iii) Use of appealing visuals to contextualise the text
and appeal to the audience’s interest on the topic
Although feature articles do not appear to exploit multimodality and
digital affordances as extensively as other digital forms (e.g., social
media), they still benefit from and utilise the chance to incorporate
visually appealing elements enhanced by technical improvements. All the
digital feature articles analysed include a picture at the top, just
below the title, and some of them include other visuals inside the text,
mostly pictures, but also, in a few cases, videos, graphs and
infographics.
Pictures at the top of the page and the preceding titles give way to
a communicative ensemble which fosters understanding, reinforces meaning
and, mainly, catches the readers’ attention. In their book Reading
Images. The Grammar of Visual Design, Kress and van Leeuwen (2021)
establish the difference between narrative and conceptual
representations. Whereas narrative representations picture participants
in some kind of (inter)action, conceptual images represent a static
concept and have a component of timelessness. The images represented at
the top and interacting with the titles in the corpus of digital feature
articles are mainly of a narrative representative type. The examples
below illustrate both types:
(12) Narrative (Smith9)
(13) Narrative (PopSci3)
(14) Conceptual (Smith11)
(15) Conceptual (PopSci7)
The overall use of narrative vs conceptual representations at the top
of the text is clear: 22 narrative pictures out of 30, which represents
73.3%. This is in line with Mur-Dueñas’s (2024) findings in her study of
non-verbal explanatory strategies in three digital practices (30 texts
from The Conversation, feature articles and research digests
from three different fields, Health, Economy and Natural Sciences), in
which only two textual instantiations of research digests included
conceptual visuals. However, it seems relevant to indicate that the
preference for conceptual or narrative representations does not seem to
be linked to the genre or practice itself. The number of analysed texts
is too small to make any claims about the distribution of types of
picture; however, in the sample studied, this distribution does not seem
to be homogeneous: whereas only 1 picture is conceptual in The
Smithsonian Magazine feature articles (6.7%), 7 out of 15 (46.7%)
include a conceptual visual in Popular Science texts (not used
as a source in Mur-Dueñas, 2024), with no apparent connection between
the type of picture and the type of title, as described in point ii)
above.
Visuals are also present in the body of the text, namely, pictures,
videos, graphs, and infographics, as well as drawings, sketches. The use
of mid-text pictures is not very frequent in the digital feature
articles analysed. Only 9 texts (out of 30) include them, with a total
of 12 pictures, all of them narrative representations, which points
towards the preference of a “telling a story” type of visual. Apart from
pictures, other visuals are also found in mid-text position in the
corpus under study, although to a much lesser extent. Only 4 feature
articles include a video. One of the videos (PopSci1) is a recording of
an experiment carried out with monkeys by a scientist in the 50’s. The
second one (PopSci4) is an extract from a TED Talk on the topic of the
feature article (collective grief); the third one (Smith7) features the
promotion of a device to self-detect breast cancer which has been
awarded a prize, and, finally, a fourth video (Smith11) shows, in the
microscope, how an implant dissolves to relieve pain. These four videos
aim to illustrate technical aspects of the issues discussed, with more
hands-on demonstrations for advancements in physical health.
Regarding graphs and infographics, there was a noticeable imbalance
between the two publications: while only 1 feature article included a
graph in Popular Science, The Smithsonian Magazine
featured graphs in 5 texts, totalling 6 graphs and infographics. These
visuals represented survey results and other quantitative data. Finally,
7 visuals such as drawings, not real pictures, were also included across
4 different feature articles from the same publication (The
Smithsonian Magazine). These visuals, found exclusively in articles
on physical health, depicted physical processes. Popular
Science did not use this type of illustration, perhaps pointing at
in-house preferences.
Although the data are very limited, they may still point at certain
trends in the use of visuals: 15 articles (50%) did not contain any
visual apart from the header illustration, with 11 of these in
Popular Science, whose articles were shorter (averaging 1,204
words per text versus 2,031 words in The Smithsonian Magazine).
This difference may indicate that visuals not only contribute to
comprehension but also to maintaining reader engagement. Thus, it can be
opularizati that longer articles have a greater need for visuals to
capture reader attention and encourage engagement. Among the remaining
texts, the inclusion of visuals varied: 4 contained one (13.3%), 7
contained two (23.3%), 3 contained three (10%), and 1 contained four
(3.3%).
As shown, the strategy of “hooking the audience’s attention” in
mediated scientific opularization relies strongly on conventional
journalistic resources, such as opening the text with a narration or
using “catchy” titles. The multimodal dimension is also part of the
resources used in this strategy, with visuals of different format
(pictures, graphs, infographics, videos) that attract the audience’s
attention and help to maintain it along the text. On a few occasions,
multimodality, already present in offline feature articles, is exploited
through affordances of the digital platform, such as incorporating
videos, which serve the double purpose of hooking the audience’s
attention and enhancing their comprehensibility of the more technical
issues.
5.2. Creating an emotional bond with the audience
Appealing to the audience’s emotions is an expected strategy that
science journalists may use to engage them in rather technical issues.
To fulfill this strategy, a combination of discourse and metadiscourse
features has been identified.
i) Attitude markers to express writer’s
emotions
Interactional metadiscourse plays a significant role in establishing
an emotional bond with the reader. More specifically, attitude markers,
used to express the writer’s attitudes to the propositional material
they present (Hyland, 2005a: 32) are well represented in the digital
scientific feature articles discussed:
(16) You have to remove the cause of your stress, and that
often requires structural changes in the work- place. That was
disappointing for me to hear. (PopSci2)
(17) And unfortunately, our brains were
not built for this (Smith3)
By expressing their attitude, the journalists empathise with the
possible reaction that the research findings might have on the audience;
that is, they act as part of the audience, which helps construct an
emotional bond with their readers.
ii) Conversational style
As the literature on popularised discourse has already shown (Luzón,
2013; Gotti, 2014; Garzone, 2020; Sterk and van Goch, 2023; inter
alia), popularisations are typified by features of conversational
discourse, for instance, informal language, as the following examples
show:
(18) Applied in the context of the family unit, this research
seemed to suggest that forceful detachment on the part of ma and
pa were essential ingredients in creating a strong, independent
future adult. (PopSci1)
(19) We can all use a chance to unplug and unwind, but
here’s the rub: Recovery from work tends to be the most
difficult and elusive for those who need it most. (Smith1)
A conversational style fosters proximity with the audience and,
although unacceptable in more academic registers, it seems to be
characteristic of the digital feature articles analysed, probably
enhanced by digital media and consistent with the less formal style
prevalent in popular science.
iii) Presentation of scientists as individuals
One way journalists have to engage audiences is to make them
visualise scientists not (or not only) as experts in possession of the
almost unreachable good of knowledge but also, and mainly, as
individuals, offering a human dimension that may allow the audience to
see them as members of the same community. This is clearly the case of
the scientists who are not only introduced by name and affiliation but
also through a link to their personal page, which usually offers a
picture, taking advantage of the technical affordances facilitated by
the digital platforms, on which scientific feature articles are nowadays
disseminated. Presenting the scientist as a “real persona” fosters both
their credibility as an expert in the field and a bond with the
audience. The use of this strategy is illustrated in the following
examples, where the names of the scientists are hyperlinked to their
pages:
(20) “They struggled with traditional forms of medical and
therapeutical intervention,” says Girija Kaimal (drexel.edu/cnhp/faculty/profiles/KaimalGirija/),
an art therapist at Drexel University and the president of the American
Art Therapy Association (AATA). (Smith2)
Blurring the identity barriers between scientists, journalists and
audience allows them all to enact other roles, and transit in both
directions the bridges which connect expert sources, mediators and
readers.
iv) Recount of personal or other’s problems and
experiences
As part of the engagement strategy of creating a bond with the
audience, journalists and mediators may seek the empathy of their
audience by sharing personal stories and hardships related to the topic
of the feature articles, sometimes their own stories (see examples 22
and 23 below). This resource, which might be called “personalisation”,
has been identified in previous analyses of popularised science (Myers,
1991; Giannoni, 2008). Unlike the strategy of using a narrative style at
the beginning of the article (see section 5.1 above), whose purpose is
to offer a specific contextual setting for the scientific information
presented in the text, the purpose of the strategy discussed here is to
foster the audience empathy towards the problem, situation, or
experience narrated. The different discursive positions of these
strategies within the text also highlight their distinct communicative
purposes. Given the personal nature of the topic of the feature articles
included in the corpus (i.e. physical and mental conditions) this
strategy may contribute to building trust and rapport between the writer
and their audience:
(22) I’ll repeat it again and again until my friends get
tired of hearing it. But in July of 2021, these responses started to
feel hollow. I was burned out. “I feel like I sprained my
brain,” I told my friend over the phone. When I wasn’t working, I felt
fine; when I tried to use my head, it felt like putting weight on a bum
ankle. (PopSci2)
(23) In 2017, Hennie Thomson checked herself into a hospital
for six weeks of in-patient treatment for anorexia nervosa. She was
compulsively over-exercising — running, spinning or cross-training three
to four hours daily. She ate only one meal each day of the same four
foods. And she felt she had hit the bottom of a deep depression.
(Smith4)
v) Humour and irony
The role played by humour and irony in human interaction has been
widely explored from a linguistic and pragmatic perspective (Raskin,
1985; Attardo, 1994; Yus, 2016). The scientific feature articles, as
instances of popularised discourse, are not devoid of it. Examples 24
and 25 illustrate some of the cases in the corpus:
(25) A once-simple decision can morph into the kind of thorny
equation that honors algebra didn’t prepare me for.
(Smith3)
In the texts analysed, humour emerges as the result of unexpected
comparisons or exaggerations of ordinary actions and experiences. The
examples show that writers use humour to grab the audience’s attention
and make complex topics more understandable. Also, humour helps create a
bond by sharing common experiences, thus enhancing the accessibility
and, therefore, impact of the scientific information presented.
The strategy of “creating an emotional bond with the audience” is, as
shown, a productive resource which exploits linguistic registers
(combining conversational styles with a standard written style),
metadiscourse features (attitude markers) and pragmatic resources
(humour and irony as a way of establishing a bond with the audience).
Moreover, this strategy leverages the affordances of digital media
through hyperlinks, enabling scientists to adopt a more personal
identity as individuals and citizens, thus fostering audience
rapport.
5.3. Enhancing
direct interaction with the audience
Audiences are directly addressed by writers as a way to make them
feel part of the same group of interested people in the scientific issue
under discussion. In this interaction, readers are perceived as
individuals with whom the writer interacts at a one-to-one level. This
direct appeal of writers to readers is a form of engagement that points
to how writers recognise their readers’ active participation, capture
their attention, and address their concerns (Hyland, 2005a). In the
texts analysed, this is done through three metadiscursive features,
categorised in Hyland’s (2005a) model as markers of engagement:
questions, second-person pronouns and directives.
i) Use of mid-text questions
Although questions were previously addressed in section 5.1 regarding
their role in hooking readers’ attention as a recurrent structural
element in titles, they are revisited here as strategic ways of
fostering direct interaction with the audience when they appear within
the text, suggesting potentially divergent or additional communicative
purposes based on their discursive position. Examples 26 and 27
illustrate these uses:
(26) Simple questions can help identify whether an older
adult needs to be evaluated for anxiety, he and other experts suggested:
Do you have recurrent worries that are hard to control? Are you having
trouble sleeping? Have you been feeling more irritable, stressed, or
nervous? Are you having trouble with concentration or thinking? Are you
avoiding things you normally like to do because you’re wrapped up in
your worries? (PopSci12)
ii) Direct address to the audience through
second-person pronouns
Hyland (2005b: 133) states that second-person pronouns “function to
elicit reader involvement and promote group solidarity”. However, he
notes that in academic writing in English, authors are generally advised
to avoid these pronouns, as they are considered “inappropriately
informal and conversational”.
But what may be deemed inappropriate in one communicative context,
such as academic discourse, can be entirely suitable in another (i.e.
popularised discourse). Thus, the use of second person forms (you,
your) is very frequent in the corpus under study, as example 28
shows:
(28) You need 13 vitamins in order to live,
and though you can actually get most of them from
eating a variety of meats, you’re going to miss out on
some crucial ones if you totally forego flora.
(PopSci15)
The normalised frequency of second-person pronouns in the corpus is
4.43 per thousand words. However, when examining each publication
individually, this frequency varies significantly, with 8.58 in
Popular Science but only 1.97 in The Smithsonian
Magazine. Given that the texts were authored by diverse writers,
which ensures variability of writing styles, one of the explanations
points at the existence of distinct in-house styles, each favoring and
employing different engagement and (meta)discourse resources.
iii) Use of directives to invite the audience to
perform some kind of action
Directives “instruct the reader to perform an action or to see things
in a way determined by the writer” (Hyland, 2005a: 184). These direct
appeals to the reader seem to be a popularising feature identified also
in other texts, such as journal editorials (Giannoni, 2008), and
research digests (Lorés, 2023, 2024a, 2024b), their main function being
“to link up with the readership and audience and to get it involved
dialogically” (Giannoni, 2008: 225). These actions might be physical
(29) or cognitive (30):
(29) Definitely don’t consume your caffeine
with other drugs like cocaine, which can magnify the worst heart effects
of both substances. (PopSci10)
(30) Imagine you’re running late and tempted
to speed. (Smith3)
Enhancing direct interaction with the audience seems to be a rather
“metadiscursive” pragmatic strategy, due to its heavy reliance on
engagement markers as described in the metadiscourse model.
5.4. Establishing
a common territory of experience
To create a common ground of experience seems to be another pragmatic
strategy used by scientific mediators to encourage their audience to
step into a territory of new knowledge. This strategy has been
identified in the metadiscourse framework as “appeals to shared
knowledge”, that is, markers that explicitly indicate that the reader
should “recognize something as familiar or accepted” (Hyland 2005a:
184). In the metadiscourse model, typical engagement markers indicating
appeals to shared knowledge are the use of the inclusive pronoun
we and polyphonic discourse markers such as of course.
This metadiscoursal strategy aligns with Brown and Levinson’s (1987)
conceptualisation of linguistic politeness, particularly its
instantiation through politeness strategies and their linguistic and
discursive markers. Thus, both the metadiscourse category of appealing
to shared knowledge and the positive politeness strategy of claiming
common ground can encompass and explain the use of certain discursive
resources identified in these texts, to make the audience participate in
a shared territory.
i) Research applications to real-life
situation
Apart from the use of the lexicogrammar, writers resort to other
discursive devices which also contribute to the pragmatic strategy of
establishing a common territory of experience, such as the explicit
presentation of applications to real-life problems or situations, as in
example 31:
(31) Perhaps, for instance, someone would accept a free
indoor dinner at a favorite restaurant this weekend instead of opting
for a free dinner-and-dessert combo in a month or two, when case rates
might fall. Our minds tend to either dismiss or overinflate(chicagounbound.uchicago.edu/cgi/viewcontent.cgi?article=12421&context=journal_articles)
small risks, with no middle ground. (Smith3)
Thus, the appeal and comprehension of specialised knowledge are
enhanced by relating it to everyday experience, shown in the
following:
(32) That could mean signing up for a virtual painting class,
“sharing” a fancy bottle of wine during a Zoom happy hour, or investing
in outdoor recreation (PopSci4)
ii) Inclusive we to appeal to shared
experience, beliefs and knowledge
As Hyland (2005a: 182) indicates, the use of the inclusive pronoun
we “sends a clear signal of membership by textually
constructing both the writer and the reader as participants with similar
understanding and goals”. First-person plural forms (we, us,
our) are widely found in the texts under analysis with an inclusive
function, as illustrated in examples 33 and 34:
(33) Emerging evidence suggests that interactions between
saliva and food may even help to shape which foodswelike to eat.(Smith10)
(34) Cortisol helps the body run from whatever is threatening
it; it raisesourheart rate and helpsourbody harvest energy from glucose.
(PopSci2)
Some quantitative data of the use of inclusive we in the
corpus under study are the following (Table 3):
Table 3. Use of inclusive first person plural pronouns. Raw numbers and
frequency per 1,000 words.
We
Us
Our
Total
Popular Science
7 (0.39)
16 (0.88)
34 (1.88)
57 (3.16)
The Smithsonian Magazine
32 (1.05)
4 (0.13)
26 (0.85)
62 (2.03)
Total
39 (0.8)
20 (0.41)
60 (1.24)
119 (2.45)
Apart from the general use of inclusive we as an engagement
device, the interesting point here seems to be the different use of this
device in one publication and the other, with Popular Science
(3.16) making a slightly more frequent use of it than The
Smithsonian Magazine (2.03). Moreover, each publication shows
preferences for distinct forms: Popular Science for the
possessive adjective our and The Smithsonian Magazine
for the inclusive pronoun we as subject. These preferences for
distinct grammatical categories of the pronoun in each publication
reveal a more nominalised function of the inclusive we in the
former publication and a more agentive role in the latter.
iii) Appeal to popular culture or everyday knowledge and habits
The final discursive resource to be discussed is the journalists’
appeal to popular culture, as a way to create a common ground with their
audience. This strategy, also used in other popularising practices such
as the research digest (Lorés, 2023, 2024a, 2024b), is a form of
intertextuality. As defined by Bazerman (2004:86), intertextuality
points at “the explicit and implicit relations that a text or utterance
has to prior, contemporary and potential future texts”, weaving other
voices into the discourse (Kristeva, 1986; Fairclough, 1992). However,
the effectiveness of this intertextual strategy lies on the writers’
careful assessment of the audience’s familiarity with the popular
culture references. A lack of recognition can be counterproductive,
potentially widening the gap it is intended to bridge. Examples 35 and
36 below illustrate the use of this resource in the digital practices
under study:
(35) In 1946, Dr.
Benjamin Spock (no relation to Dr. Spock of Star Trek) authored
Baby and Child Care, the international bestseller, which sold 50
million copies in Spock’s lifetime. (PopSci1)
(36) The process starts with the blue bubble of a texted
invitation or a date flagged on the calendar—a party Saturday, a
sibling’s high-school basketball game, a second cousin’s middle-school
Quidditch match, a cross-country flight for a grandparent’s 90th
birthday. (Smith3)
The analysis of engagement strategies that foster a common ground of
experience in popularised science writing reveals the importance of
establishing a common territory of experience to connect with readers.
This approach, exemplified by the use of inclusive pronouns, real-life
applications, and references to popular culture, plays a significant
role in bridging the gap between scientific knowledge and everyday
understanding.
6. A quantitative
analysis: discussion of data
Although this is primarily a qualitative analysis and the limited
data prevent robust generalisations, a quantitative analysis has also
been conducted, as it provides an overview of the distribution of the
four major audience-oriented pragmatic strategies identified in the
corpus, as well as their associated (meta) discursive resources.
The four pragmatic strategies identified were present in all the
texts included in the study. However, a more fine-grained analysis of
the (meta) discursive resources yields more nuanced results:
Table 4. Presence of (meta)discursive features in texts. Number of texts and
percentages.
Pragmatic strategies
(Meta)Discursive resources
Raw numbers
Percentage of use
1. Hooking the audience’s
attention
Popular press titles
30
100%
Narrative opening
15
50%
Appealing visuals
15
50%
2. Creating an emotional bond with the
audience
Attitude markers
21
70%
Conversational style
8
26.6%
Scientists as individuals
8
26.6%
Recount of personal problems
7
23.3%
Humor/Irony
6
20%
3. Enhancing direct interaction with the
audience
Mid-text questions
13
43.3%
Second-person pronouns
10
33.3%
Directives
4
13.3%
4. Establishing common territory of
experience
Research applications to real life
9
30%
Inclusive we
8
26.2%
Appeal to popular culture
5
16.5%
Table 4 offers an overview of
the data presented in section 5. As observed, a pervasive feature was
the use of titles emulating those found in the popular press, designed
to captivate the readers’ attention. Attitude markers, a metadiscursive
resource, were highly prevalent, appearing in 70% of the texts. Equally
noteworthy was the presence of a narrative style (especially at the
beginning of the text) and the use of engaging visuals, both of which
were present in 50% of the texts. The strategic insertion of mid-text
questions, fostering a question-and-answer rhetorical structure, was
also significant, occurring in 43.3% of the feature articles. The
remaining (meta)discursive resources were also employed to different
degrees, serving engaging purposes. These ranged from purely
metadiscursive interactional features, such as second-person pronouns
and the inclusive we, to more pragmatic devices like humour,
irony and references to popular culture. The affordances of digital
media were leveraged through the inclusion of hyperlinks to researchers’
webpages, emphasising scientists as individuals.
When placing the lense on the distribution of features, we observe
that some articles display a high number and range of engagement
resources (between 9 and 13 resources used out of 14) and others, very
few (1 or 2), show that popularising, at least as far as engagement is
concerned, is a matter of degree.
Differences are also identified between the two publications in some
of the features, whereas the use of others is very similar, as Table 5 shows:
Table 5. Presence of (meta)discursive features in each publication. Number of
texts and percentages between brackets.
Pragmatic strategies
(Meta)Discursive resources
Popular
Science
The Smithsonian
Magazine
1. Hooking the audience’s
attention
Popular press titles
15 (100%)
15 (100%)
Narrative opening
6 (40%)
9 (60%)
Appealing visuals
4 (26.6%)
11 (73%)
2. Creating an emotional bond with the
audience
Attitude markers
11 (73.3%)
10 (66.6%)
Conversational style
4 (26.6%)
4 (26.6%)
Scientists as individuals
3 (20%)
5 (33.3%)
Recount of personal problems
1 (6.6%)
6 (40%)
Humor/Irony
3 (20%)
3 (20%)
3. Enhancing direct interaction with the
audience
Mid-text questions
6 (40%)
7 (46.4%)
Second-person pronouns
7 (46.4%)
3 (20%)
Directives
3 (20%)
1 (6.6%)
4. Establishing common territory of
experience
Research applications to real life
5 (33.3%)
4 (26.6%)
Inclusive we
5 (33.3%)
3 (20%)
Appeal to popular culture
3 (20%)
2 (13.3%)
Publications display notable differences in the use of certain
elements: visual content, second-person pronouns, and the discursive
feature of sharing personal problems. In contrast, the two publications
display striking similarities in numerous (meta)discursive features
identified as realisations of audience-oriented
pragmatic strategies within the corpus: the adoption of a
conversational style, the use of humour and irony, of attitude markers
and inclusive we, the inclusion of research applications to
real-life scenarios, the references to popular culture, and the
inclusion of mid-text questions. Therefore, the similarities in the
presence and utilisation of these (meta)discursive resources, which
typify the popularised discourse of digital feature articles, outweigh
the differences between the two publications. This commonality suggests
a broader trend in the genre, which goes beyond individual publication
styles.
Finally, another comparison of quantitative data has been carried out
on the basis of the discipline, as shown in Table 6:
Table 6. Presence of (meta)discursive features in each discipline (mental and
physical health). Number of texts and percentages between brackets.
Pragmatic strategies
(Meta)Discursive resources
Mental Health
Physical Health
1. Hooking the audience’s
attention
Popular press titles
15 (100%)
15 (100%)
Narrative opening
4 (26.6%)
10 (66,6%)
Appealing visuals
6 (40%)
9 (60%)
2. Creating an emotional bond with the
audience
Attitude markers
10 (66.6%)
11 (73.3%)
Conversational style
5 (33.3%)
3 (20%)
Scientists as individuals
4 (26.6%)
2 (13.3%)
Recount of personal problems
6 (40%)
4 (26.6%)
Humor/Irony
4 (26.6%)
1 (6.6%)
3. Enhancing direct interaction with the
audience
Mid-text questions
6 (40%)
7 (46.4%)
Second-person pronouns
6 (40%)
4 (26.6%)
Directives
2 (13.3%)
2 (13.3%)
4. Establishing common territory of
experience
Research applications to real life
5 (33.3%)
3 (20%)
Inclusive we
6 (40%)
2 (13.3%)
Appeal to popular culture
2 (13.3%)
3 (20%)
The contrastive analysis of quantitative data across health
disciplines also revealed interesting differences in four main features:
the use of a narrative opening and visuals, more prevalent in physical
health articles, and the use of inclusive we and the discursive
resource of sharing problems, more frequent in mental health content,
thus pointing at the specificities of the disciplines, with physical
health dealing with tangible aspects of the body, diseases and
treatment, and mental health dealing with internal, more sensible
experiences which may demand a more subtle, more supportive and
empathetic way of presenting and approaching the problems. The remaining
(meta)discursive and textual devices appeared with similar frequency
across both health disciplines, suggesting the existence of a consistent
realization of engagement strategies characterising the popularised
discourse of digital feature articles in this field.
Finally, by combining quantitative results per publication and
discipline, we can identify a set of (meta) discursive features that
define the engaging style in feature articles published online. These
key elements include titles mimicking popular press headlines, a
conversational style, attitude markers, strategically placed mid-text
questions, references to real-life situations, hyperlinks to scientists’
webpages, and mentions of popular culture. This core set of features is
likely to appear in feature articles in other fields, a hypothesis that
future studies may further confirm.
7. Conclusions
The present study has explored the pragmatic strategies and
associated (meta)discursive resources used in digital feature articles
to mediate scientific information and engage diversified audiences. The
analysis shows a purposeful use of audience-oriented strategies that
work to create a horizontal relationship between writer and reader as
peers within a field community. With regard to the first research
question posed (what audience-oriented pragmatic strategies are used in
digital feature articles on scientific topics to foster engagement),
four major pragmatic strategies have been identified: hooking the
audience’s attention, creating an emotional bond with the audience,
enhancing direct interaction with the audience and establishing common
territory of experience. All these strategies aim to foster engagement,
thereby bridging knowledge gaps between expert sources and general
audiences.
Following the identification of audience-oriented pragmatic
strategies, the second research question was meant to explore the (meta)
discursive resources that realise these strategies in scientific digital
feature articles. The results suggest that popularised scientific
discourse leverages journalistic strategies to attract and
retain the audience. Titles similar to those of popular media were
consistently used to draw readers into the text. Moreover, attitude
markers and narrative styles were prevalent, particularly at the
beginnings of articles, serving the purpose of humanising scientific
subjects, and enhancing their accessibility. Multimodal components were
not exploited to quite the degree seen in other digital practices such
as social media; however, in spite of their ancillary role in this
digital genre, they still contributed to audience engagement with
visuals serving to both attention-getting and comprehensibility
purposes.
The quantitative analysis suggests partially consistent patterns
across publications and disciplines. While both Popular Science
and The Smithsonian Magazine shared a core set of strategies
for audience engagement, some differences were observed in the frequency
and variety of resources used, perhaps reflecting editorial preferences.
The comparison of mental and physical health articles indicated that,
while both contexts deployed similar engagement strategies, there were
variations in narrative voice, use of visuals, and personal
storytelling, pointing to some specific communicative necessities for
each discipline. Overall, these findings suggest that these forms of
engagement in scientific communication share an underlying rationale and
a high degree of adaptability shaped around the genre, but also around
what audiences are likely to expect.
Finally, the third research question aimed to determine how the
concept of discursive interpersonality could enhance our
understanding of the discursive action taken in scientific digital
feature articles. This study has intended to show the role of
propositionality and other discursive resources in fostering audience
engagement and interaction. Thus, since the metadiscourse model falls
short in fully explaining the discursive action taken by mediators to
engage their audiences in specialised knowledge, the more encompassing
concept of discursive interpersonality (Suau et al., 2021) may account
for the wide variety of features that combine pragmatic, discursive,
multimodal and lexicogrammatical levels to motivate and facilitate
audiences’ access to knowledge.
The present study has certain limitations. The dataset, comprising a
small selection of texts, offers limited information about the
quantitative distribution of strategies and their associated resources.
Moreover, the disciplines selected, although of significant public
interest, may also introduce a bias to the results. However, this close
analysis has provided interesting insights into the diversity of
pragmatic strategies used by mediators to foster public understanding of
complex issues and the variety of (meta)discoursal resources that
instantiate these audience-oriented strategies. Further research could
explore the effectiveness of these strategies across different online
contexts and across diverse audiences. Understanding the interplay
between pragmatic strategies, (meta)discursive resources, and audience
reception is crucial for advancing effective science communication in
the digital age.
Acknowledgement
This research is a contribution to the InterGEDI research group (intergedi.unizar.es/). Grant
PID2021- 122303NB-100, funded by MCIN/AEI/ 10.13039/501100011033 and by
“ERDF A way of making Europe”, as well as by Gobierno de Aragón
(H16_23R), Spain.
Banks, David, & di Martino, Emilia. (2019). Introduction: Linguistic and discourse issues in contemporary scientific communication. Aspects of communicating science to a variety of audiences. Journal of Pragmatics, 139, 185-189.https://doi.org/10.1016/j.pragma.2018.10.011
Bazerman, Charles. (2004). Intertextuality: How texts rely on other texts. In Charles Bazerman, & Paul Prior (Eds.), What writing does and how it does it: An introduction to analyzing texts and textual practices, (pp. 83-96). New Lawrence Erlbaum Associates.
Bondi, Marina, & Cacchiani, Silvia. (2021). Editorial. Knowledge communication and knowledge dissemination in a digital world. Journal of Pragmatics, 186, 117-123. https://doi.org/10.1016/j.pragma.2021.10.003
Bondi, Marina, Cacchiani, Silvia, & Mazzi, Davide. (2015). Discourse in and through the Media. Cambridge Scholars Publishing.https://doi.org/10.1177/0741088313493610
Brown, Penelope, & Levinson, Stephen C. (1987). Politeness: Some universals in language use. Cambridge University Press.
Brunner, Marie-Louise, & Diemer, Stefan. (2022). Cross-platform analysis. In Camilla Vásquez (Ed.), Research methods for digital discourse (pp.197-218). Bloomsbury Academic.
Bucchi, Massimiano. (2008). Of deficits, deviations and dialogues: Theories of public communication in science. In Massimiano Bucchi, & Brian Trench (Eds.), Handbook of public communication of science and technology (pp. 57-76). Routledge. https://doi.org.10.4324/9780203928240-11
Calsamiglia, Helena, & Van Dijk, Teun. (2004). Popularization discourse and knowledge about the genome. Discourse & Society, 15(4), 369-389. https://doi.org/10.1177/0957926504043705
Crismore, Avon, Markannen, Raija, & Steffensen, Margaret. (1993). Metadiscourse in persuasive writing: A study of texts written by American and Finnish university students, Written Communication, 10(1), 39-71.
Dahl, Trine. (2004). Textual metadiscourse in research articles: A marker of national culture or of academic discipline? Journal of Pragmatics, 36, 1807-1825. https://doi.org/10.1016/j.pragma.2004.05.0041808
Escandell-Vidal, Victoria. (2025). Las comillas en los titulares de prensa española: evidencialidad y perspectivización emocional. Círculo de Lingüística Aplicada a la Comunicación, 101, 183-199. https://doi.org/10.5209/clac.100079
Facchinetti, Roberta. (2021). News discourse and the dissemination of knowledge and perspective: from print and monomodal to digital and multi semiotic. Journal of Pragmatics, 186, 195-206. https://doi.org/10.1016/j.pragma.2021.01.015
Fairclough, Norman. (1992). Discourse and social change. Polity Press.
Finkbeiner, Rita. (2024). The pragmatics of headlines. Central issues and future research avenues. Journal of Pragmatics, 222, 17-22. https://doi.org/10.1016/j.pragma.2024.01.006
Freddi, Maria. (2020). Blurring the lines between genres and audiences: Interaction in science blogs. Discourse and Interaction, 13(2), 9-35.https://doi.org/10.5817/DI2020-2-9
Garzone, Giuliana. (2020). Specialized communication and popularization in English. Carocci Editore.
Giannoni, Davide Simone. (2008). Popularizing features in English journal editorials. English for Specific Purposes, 27, 212-232.https://doi.org/10.1016/j.esp.2006.12.001
Gotti, Maurizio. (2014). Reformulation and recontextualization in popularization discourse. Ibérica, 27, 15-34.
Herrando-Rodrigo, Isabel. (2019). Raising awareness around writers’ voice in academic discourse. An analysis of writers’ (in)visibility. Brno Studies in English, 45(2), 53-76. https://doi.org/10.5817/BSE2019-2-3
Hilgartner, Stephen. (1990). The dominant view of popularization: Conceptual problems, political uses. Social Studies of Science, 20(3), 519-539.https://doi.org/10.1177/030631290020003006
Holton, Judith A. (2007). The coding process and its challenges. In Anthony Bryant, & Kathy Charmaz (Eds.), The Sage handbook of grounded theory (Part III), pp. 265-89. Sage. https://dx.doi.org/10.4135/9781848607941.nl3
Hyland, Ken. (2005a). Stance and engagement: A model of interaction in academic discourse. Discourse Studies, 7(2), 173-192. https://doi.org/10.1177/146144560505036
Hyland, Ken. (2005b). Metadiscourse: Exploring interaction in writing. Continuum.
Hyland, Ken. (2010). Constructing proximity: relating to readers in popular and professional science. Journal of English for Academic Purposes, 9(2), 116-127. https://doi.org/10.1016/j.jeap.2010.02.00
Hyland, Ken, & Tse, Polly (2004). Metadiscourse in academic writing: a reappraisal. Applied Linguistics, 25(2), 156-177.
Hyland, Ken, & Zou, Hang Joanna. (2020). In the frame: signalling structure in academic articles and blogs. Journal of Pragmatics, 165, 31-44.https://doi.org/10.1016/j.pragma.2020.05.002Luz
Ivorra Pérez, Francisco. (2015). The impact of cultural dimensions and politeness on the engagement markers of Spanish, British and US business websites. In Maria de la O Hernández-López, & Lucía Fernández-Amaya (Eds.), A multidisciplinary approach to service encounters. Studies in pragmatics 14 (pp. 139-163). Brill.
Jaworski, Adam, & Pritchard, Annette. (2005). Discourse, communication and tourism. Channel View Publications / Multilingual Matters.https://doi.org/10.21832/9781845410216
Jones, Rodney J., Chick, Alice, & Hafner, Christopher. (2015). Introduction: discourse analysis and digital practices. In Rodney H Jones, Alice Chik, & Christopher Hafner (Eds.), Discourse and digital practices: Doing discourse analysis in the digital age (pp. 1-17). Routledge. https://doi.org/10.4324/9781315726465
Kress, Gunther, & van Leeuwen, Theo. (2021). Reading images. The grammar of visual design. 3rd edition. Taylor and Francis. https://doi.org/10.4324/9781003099857
Kristeva, Julia. (1986). Word, dialogue and novel. In Toril Moi. (Ed.), The Kristeva reader (pp. 35-61). Blackwell.
Labinaz, Pablo, & Sbisà, Marina. (2021). The problem of knowledge dissemination in social network discussions. Journal of Pragmatics, 175, 67-80.https://doi.org/10.1016/j.pragma.2021.01.009
Lorés-Sanz, Rosa. (2011). The construction of the author’s voice in academic writing: The interplay of cultural and disciplinary factors. Text & Talk, 31(2), 173-193.https://doi.org/10.1515/text.2011.008
Lorés, Rosa. (2023). Dual voices, hybrid identities: the recontextualization of research in the digital dissemination of scientific discourse. Círculo de Lingüísticas Aplicada a la Comunicación, 93, 69-84. https://dx.doi.org/10.5209/clac.85566
Lorés, Rosa. (2024a). Digesting psychology: metadiscourse as a recontextualising tool in the digital communication of disciplinary research. Chinese Journal of Applied Linguistics, 47(2), 178-195. https://doi.org/10.1515/CJAL-2024-0202
Lorés, Rosa. (2024b). Mediating expert knowledge: The use of pragmatic strategies in digital research digests. Journal of Pragmatics, 232, 26-40. https://doi.org/10.1016/j.pragma.2024.08.007
Luzón, María José. (2013). Public communication of science in blogs: recontextualizing scientific discourse for a diversified audience. Written Communication, 30(4), 428-457. https://doi.org/10.1177/0741088313493610
Luzón, María José. (2023). ‘Excited to see our latest work published’: recontextualizing research results in biomedical tweetorials. Journal of English for Academic Purposes, 66, Article101308. https://doi.org/10.1016/j.jeap.2023.101308
Luzón, María José, & Pérez-Llantada, Carmen. (2019). Science communication on the Internet: old genres meet new genres. John Benjamins.
Mauranen, Anna. (1993). Cultural differences in academic rhetoric. Peter Lang.
Moirand, Sophie. (2003). Communicative and cognitive dimensions of discourse on science in the French mass media. Discourse Studies, 5(2), 175-206.https://doi.org/10.1177/1461445603005002309
Motta-Roth, Désirée, & Scotti-Scherer, Anelise. (2016). Science popularization: Interdiscursivity among science, pedagogy and journalism. Bakhtiniana, 11(2), 171-194. http://dx.doi.org/10.1590/2176-457323671
Mur-Dueñas, Pilar, Lorés-Sanz, Rosa, & Lafuente-Millán, Enrique. (2010). Editorial. Journal of English for Academic Purposes, 9(2), 83-85.
Mur-Dueñas, Pilar. (2024). Digital dissemination practices: An analysis of explanatory strategies in the process of recontextualising specialised knowledge. Discourse & Interaction, 17(1), 94-114. https://doi.org/10.5817/DI2024-1-94.
Myers, Greg. (1991). Lexical cohesion and specialized knowledge in science and popular science texts. Discourse Processes, 14(1), 1-26. https://doi.org/10.1080/01638539109544772
Myers, Greg. (2003). Discourse studies of scientific popularization: questioning the boundaries. Discourse Studies, 5(2), 265-279. https://doi.org/10.1177/1461445603005002006
Pascual, Daniel. (2024). Pragmatic macrocategories and microstrategies in research project homepages: meaning-making verbal and visual devices. In Sanna-Kaisa Tanskanen, Lotta Lehti, Kristin Vold Lexander, Ulla Tuomarla, Mikko Virtanen, & Chaoqun Xie (Eds.), Explorations in Internet pragmatics: identity, intentionality and interpersonal interaction (pp. 67-91). Brill. https://doi.org/10.1163/9789004694453_005
Pascual, Daniel. (2025). Dialogic markers in Ask and Expert webpages on environmental discourse. Language & Dialogue, 15(1), 156-181.https://doi.org/10.1075/ld.00191.pas
Pascual, Daniel, Plo-Alastrué, Ramón, & Corona, Isabel. (2023) Digital scholarly practices in scientific communication: Paths and goals in research dissemination. In: Ramón Plo-Alastrué, & Isabel Corona (Eds.), Digital scientific communication: identity and visibility in research dissemination (pp. 3-30). Palgrave Macmillan.https://doi.org/10.1007/978-3-031-38207- 9_1
Pascual, Daniel, & Sancho-Ortiz, Ana Eugenia. (2024). Investigating recontextualisation processes in scientific digital practices: The SciDis database. Revista Electrónica de Lingüística Aplicada, 23, 101-118. https://doi.org/10.58859/rael.v23i1.649
Pilkington, Olga. (2018). Presented discourse in popular science: professional voices in books for lay audiences. Brill.
Plo-Alastrué, Ramón, & Corona, Isabel. (Eds.). (2023). Digital scientific communication: identity and visibility in research dissemination. Palgrave Macmillan.
Pontrandolfo, Guanluca, & Piccioni, Sara. (2022). Comunicación especializada y divulgación en la red. Aproximaciones basadas en corpus. Routledge.
Sancho-Ortiz, Ana Eugenia. (2025). Environmental discourse on social media: Exploring engagement in X/Twitter for environmental purposes. Language & Dialogue, 15(1), 105-133. https://doi.org/10.1075/ld.00189.san
Sterk, Florentine Marnel, & Van Goch, Merel M. (2023). Re-presenting research. A guide to analyzing popularization strategies in science journalism and science communication. Palgrave MacMillan. https://doi.org/10.1007/978-3-031-28174-7
Suau-Jiménez, Francisca. (2016). What can the discursive construction of stance and engagement voices in traveler forums and tourism promotional websites bring to a cultural, cross-generic and disciplinary view of interpersonality?, Ibérica 31, 199-220.
Suau-Jiménez, Francisca, Lorés-Sanz, Rosa, Mapelli, Giovanna, & Herrando-Rodrigo, Isabel. (2021). La interpersonalidad discursiva como alternativa al metadiscurso interpersonal. Onomázein, 54, 113-141. https://doi.org/10.7764/onomazein.54.07
Vásquez, Camilla. (Ed.). (2022). Research methods for digital analysis. Bloomsbury Academic.
Yakhontova, Tatyana. (2006). Cultural and disciplinary variation in academic discourse: the issue of influencing factors. Journal of English for Academic Purposes, 5, 153-167.
Yus, Francisco. (2016). Humour and relevance. Benjamins.