Reading Aloud Practices: Providing Joint Accessibility to Texts within an Unfamiliar Interface-Mediated Game Activity

1 Reading aloud in interaction

Meeting the practical challenge of making written text¹ audibly accessible² for co-present interlocutors in interaction, reading aloud has had its place in our society for thousands of years and, as a relevant social practice, has been a means of communication ever since: from the public reading of(f) a cultures’ sacred scriptures to the most private reading aloud to a toddler before bedtime. Coming from that, reading aloud has been the object of i.a., physiological, linguistic, didactic and historical queries through different epochs (c.f., Busch 2002). In particular, the science-historical discourse in the context of research on reading (aloud) has been aiming at reconstructing the socio-cultural significance of reading aloud through the ages. For example, in Antiquity, both literate education and text documents were rare and of special value. So, being able to read and to read aloud to others was reserved for the elites. Whereas the illiterate majority of the population, on the other side, was dependent on that they were being read aloud (e.g. for the Middle Ages, see Glauch & Green 2010; Wendehorst 1986; for the Antiquity, see Heilmann 2021; von Lovenberg 2018; Busch 2002).

Today, literacy is the norm, and text can be found to be mediated over all kinds of surfaces, e.g., printed on paper or displayed on digital interfaces, and even without the medium of a tangible surface, e.g., by holograms. This is why in everyday life, reading aloud can be encountered in a multitude of multi-party, face-to-face situations, e.g., when reading aloud a smartphone text message to friends, a book passage in the classroom, a bedtime story to children. Often enough, reading aloud is encountered in those very social situations where people are working together, sharing a text’s content with another.

The following work about the practice of reading aloud is my dissertation which is affiliated with the ‘ORBIT’³ research project at the University of Luxembourg. In this project, a series of participant user studies was conducted. It was recorded how uninstructed groups of adult participants engage in a collaborative game activity mediated by an interactive tabletop together (henceforth ITT⁴, see section 3.2.5 for more details on its features), both unfamiliar to them. When investigating the video data corpus, I could observe specific collaborative strategies of participants while working for accomplishing the game activity. One of these participants’ practices observable in all video study trials is reading aloud. Interestingly, in close examination of the video data, it became apparent that reading aloud is not simply conducted by the participants as a unilateral practice but that it is mobilised as a practice in the interactional context of the group’s co-organisation⁵ towards accomplishing the collaborative game activity together. So, in the process of familiarising with the unfamiliar game activity, participants read aloud chunks of game texts that we – as the game’s designers – had implemented all over the interface screen. From a game design perspective, we had provided this text as a resource for the participants’ support and guidance through the game activity when needed. In fact, what I could observe was that participants in all study trials – regularly and repeatedly – turned to game text as a game-relevant resource to read it aloud. Interestingly, in doing so, the reading aloud participant makes that game text accessible for her/his co-present participants at the ITT even though they could all often visually access the text in question for reading it (individually and silently). Coming from that initial observation, I decided to direct my research focus onto that very practice of reading aloud in interaction and how it is conducted not as an individual practice but as a joint one.

In this process I encountered that although reading aloud is frequently dealt with in scientific studies, there is quite a gap in research (see section 2.5 for the detailed description of the desideratum). Namely, until now, reading aloud has mainly been investigated on the basis of instructed aloud readings of individual participants or on the basis of audio data only. As will be shown in more detail in chapter 2, e.g., in psychological and linguistic studies, reading aloud is often instructed and recorded for building homogenous speech data corpora in order to investigate cognitive processes (c.f., Gårding 1981) or prosodic features of read-aloud speech (c.f., Thorsen 1985; Sluijter & Terken 1992). Whereas in, e.g., educational studies, reading aloud is often utilised and examined as an instrument for children’s language promotion (c.f., Lehmden et al. 2015; Becker & Müller 2015; Schönauer-Schneider 2012; Belgrad et al. 2011).

So far, it has rarely been investigated how reading aloud is conducted in uninstructed interactions, considering not only the participants’ verbal speech but also the holistic multimodal accomplishment in its sequential ‘unfolding’ and embedment in interaction. That is the reason why in this work, I will tackle the above-mentioned desideratum in read-aloud research, conducting qualitative case analyses by means of Conversation Analysis (henceforth CA, c.f., Sidnell & Stivers 2012) in its multimodal understanding (c.f., Mondada 2014; see chapter 4 for more details). This approach allows me to empirically investigate participants’ interactional multimodal conduct, making observable how and under which interactional conditions participants mutually and multimodally co-organise reading aloud in game interactions and which interactional challenge participants address with it in particular. So, apart from the study approaches listed above, I will consider reading aloud as a multimodal practice in its interactional occurrence (instead of examining, e.g., only the verbal level) made observable through the investigation of video recordings (instead of working only on field observations or audio recordings) of participants’ unscripted reading aloud in multi-party face-to-face interaction (instead of, e.g., monologue or instructed reading situations) in a human-interface interaction context.

In the analyses that follow in this work, the CA approach onto video data makes analytically available how participants in the process of jointly working out how to play an unfamiliar game activity, reading chunks of game text from the interface aloud, both to and with each other.

For this purpose, all investigations in this work have been conducted and presented on the basis of empirical data examples. The data basis for this work’s qualitative investigation is formed by a sub-corpus⁶ of over one hour (74 min.) of recorded video data. This sub-corpus entails video recordings from two different participant study trials with English-speaking participants. In total, I could identify 45 cases of participants’ reading aloud in this sub-corpus, showing that reading aloud is an interactionally relevant and robust practice for different participants throughout different study trials alike (see table 7, chapter 4, section 4.4). As mentioned earlier, the video recordings show participants engaged in an unfamiliar game activity at an unfamiliar ITT (see chapter 3). As a group of three participants, they are to work together as a space mining crew in a planetary environment in order to accomplish an overall game task: steering a shared mining rover over the planetary surface (resp. the ITT screen they are standing at) for collecting and transporting valuable minerals and at the same time preventing the rover from running out of energy and preventing damage to its hull. The group of researchers from the ORBIT project as the game’s designers have provided for a text-mediated guidance through that game activity. In particular, this game-relevant text is mediated via multiple text chunks all over the horizontal interface screen to the group of participants. Engaging in the game activity named ‘Orbitia’, it is observable that participants regularly and recurrently turn to those chunks of game text displayed on the interface screen to read it aloud.

Hence, in this work, there will be four empirical cases of reading aloud analysed and presented (from a case collection of 45 cases of reading aloud examined and analysed in total). In chapter 5, by means of an exemplary reading-aloud case (case 1), I will stepwise fan out a read aloud’s constituent multimodal components that enable the identification of reading-aloud cases in the video data in the first place. I will particularly show how participants are doing reading aloud as an interactional practice of meeting the practical challenge of accomplishing a game activity (research question, henceforth RQ).

It is observable that meeting the interactional challenge of making certain game text audibly accessible for others in order to jointly progress in the game activity, reading aloud is commonly conducted by one participant at a time (c.f., Sacks et al. 1974) in a distinct and loud (enough) manner for other co-present interlocutors to hear who might not have visual access to the game text in question.

Building on the empirical observation and micro-analytic reconstruction of how reading aloud is commonly conducted in interaction (see case 1), I will present two further case analyses (case 2 and 3) pursuing the question of how participants respond to faltering reading aloud that does not provide that audible accessibility any longer (see sections 6.1 and 6.2). For that, I will focus on situations where participants with co-visual access onto the text in question treat the audibility of an ongoing reading aloud by another participant as problematic and a reason to intervene. This way, they manage to restore the audibility of the ongoing reading aloud again. So far, it has not been investigated which communicative function such collaborative interventions (e.g., ‘turn-sharing’, c.f., Lerner 2002) take in the context of ongoing read-aloud accounts.

In a fourth and last case analyses (case 4) I will investigate under which interactional conditions participants (different from the cases before) do not treat the audibility of a read aloud as relevant to progress with the game (see section 6.3). For that, I will focus on a situation where all three participants have visual access onto the text in question, engage in a partly overlapping mumbled reading engagement. I will show that the content of the game text turned to is not made audibly accessible in a way that a participant’s reading aloud typically does in the video data examined. Instead, the group of participants mobilises those displays of a reading engagement which is – even though mumbled – still monitorable for their co-participants and still accomplishes that they are all mutually assured that they all have read that one text for progressing with informed joint next actions in the game.

Thereafter, in the last chapter (chapter 7), I will bring together and discuss my examination of reading-aloud practices mobilised by participants engaged in the game activity ‘Orbitia’. Furthermore, I will outline the research and practitioner contexts to which this work can contribute relevant insights before I list potential paths for further research.

1 In the following work, I will refer to written text as ‘text’, and in the specific context of the game activity, I will refer to written game text as ‘textual interface informings’ (short: tii’s, see details in chapter 3, section 3.3).