“What is the Use of a Book Without Pictures?” An Exploration of the Impact of Illustrations on Reading Experience in A Monster Calls

This article examines the effect of Jim Kay’s illustrations on the experience of reading A Monster Calls by Patrick Ness. The author compares the responses of six Key Stage Three children (11–14 years old), three of whom were given an illustrated version of the text, and three a non-illustrated version. The children with an illustrated copy engaged with the text more deeply and critically than the others. They were also more likely to relate the story to their own lives. The illustrations were found to work alongside the participants’ own visualisations rather than replacing them, and opened up further possible interpretations rather than limiting them. The illustrations did not appear to have influenced the participants’ overall enjoyment of the book, nor did they significantly alter the readers’ views on key themes and ideas of the text.

Monster Calls was released in a non-illustrated edition. With two different versions on the market, several questions arose: what purpose do the illustrations serve, and what impact does their presence or absence have on the experience of reading the novel? While there is considerable research on the role of visuals in picture books, very little scholarship exists on the impact of illustrations in novels. Several scholars, such as Margaret Marshall (1988), Sharon Goodman (2009) and Donnarae MacCann and Olga Richard (1973), suggest that children find illustrations in novels to be distracting, but these views are based on anecdotal, not empirical, evidence. However, as Evelyn Arizpe and Morag Styles (2003) demonstrate, listening to children's voices when dealing with children's texts can prove highly illuminating and lead to results that may not be anticipated by a theoretical approach alone. These findings motivated me to place readers' voices at the forefront of my study, acknowledging children's personal experiences and views.
As well as filling a gap in scholarship, gaining an understanding of how children use illustrations in novels could have implications for the classroom. If illustrations do, as Nodelman (1988, pp. 220-221) suggests, force the reader to consider the text more deeply, then they could be used to develop skills in critical analysis. For instance, Judith Graham (1990, p. 26) concludes that Charles Keeping's illustrations of Alfred Noyes's narrative poem The Highwayman help support and engage those students with weaker visualisation skills. If this observation also holds true for novels, then illustrated texts could be used to engage more students with reading. To gain insight into these questions, I spoke with six participants of 11-14 years old about their experiences of reading illustrated or non-illustrated copies of A Monster Calls. I focused my examination on the key research areas of engagement, interpretation, picturing and narrative rhythm.

Theoretical Perspectives
To date, no theory on the role of illustrations in novels exists. I have therefore examined perspectives which draw on literary theory, art theory, illustration theory and picture-book theory, and will briefly discuss how these must be adapted when dealing with illustrated novels. Maria Nikolajeva and Carole Scott (2000) argue that a book with visuals is an illustrated novel, rather than a picture book, when ''the text is not dependent on illustrations to convey its essential message' ' (p. 227). This is the form that A Monster Calls takes, as evidenced by its publication in both illustrated and non-illustrated form.
To understand how children respond to texts with illustrations, it is vital to grasp the process they undergo when reading a text. Reader-response theory, founded by Louise Rosenblatt (1938) and further developed by Roland Barthes (1977) and Wolfgang Iser (1978), asserts that readers' personal experiences have a big impact on their interpretation of texts. Margaret Meek (1988) and Gunther Kress (2003) see the reading process as a combination of textual information and personal experience. Kress (2003, pp. 143-144) suggests that writing provides ''reading paths''-structures set out in a clear order which readers then fill with their own meaning. This approach to reading has significant implications for my research. If children are actively making meaning from the words they read rather than simply absorbing information, then the presence of illustrations could provide additional and possibly conflicting ideas, which they must marry with the meaning they create from the words. Michael Benton and Geoff Fox (1985, pp. 5-7) and Donald Fry (1985, p. 66) argue that so-called ''picturing''-building images inside the reader's mind-is a vital part of the reading process, and that the presence of illustrations may interfere with it. However, recent research by Tessa Dekker et al. (2014) has shown that automatic picturing skills are actually developed over time, and are much stronger in adults than in children. Therefore, for those without strong automatic picturing skills, illustrations could potentially support, rather than interfere with, their picturing process while reading.
Since illustrated books contain both text and images, it is also important to consider the process of reading pictures. This is an active process, with no single authoritative interpretation (Grigg, 2003;Arnheim, 1992;Mitchell, 1986). David Perkins (1994) draws a distinction between perceiving the evidence of an image and reflecting upon its meaning. To read an image, one must look closely and purposefully. When looking at children's responses to illustration, it is therefore important to consider how much they are engaging with the images. If the participants of the study interrogate the illustrations carefully and critically, it will affect the overall meaning they take from the book. If they simply skim past the images, the impact will be more limited. In addition to looking carefully, John Berger discusses the importance of drawing connections between what we see and what we know when creating meaning (Berger, 1972;Berger and Mohr, 1989). This theory has profound implications for the ways in which children approach illustrations. As children do not have the same level of knowledge and experience as adults, their ability to interpret illustrations and draw comparisons may differ from that of adults. However, Kress argues that this meaning making is not as reliant on individual interpretation as Berger suggests; rather, it is guided by the illustrations. Not only words, but also images can create ''reading paths,'' by ''producing an order in the arrangement of elements which are already filled with meaning'' (Kress, 2003, pp. 143-144). If we assume that Kress is correct, then it is possible that the participants' interpretations may be partially directed, and therefore rely less on their individual experiences.
As distinct modes of communication, text and illustrations have their own expressive possibilities and limitations (Kress and Van Leeuwen, 1996). As a result, Kress claims that when used together, each mode will carry only a part of the meaning or ''informational load' ' (2003, p. 141). In the context of an illustrated novel, the potential of illustrations to carry additional information might have a significant impact on the participants' responses to the text as a whole. Other scholars have suggested that words and pictures not only provide separate meanings working towards a whole, but can also influence the way each mode is read and alter each other's meaning. However, these critics differ in their assessment of this process's effect. Peter Hunt (2009, p. 10) argues that the juxtaposition opens up a multitude of interpretive possibilities. According to Nikolajeva and Scott (2000, p. 232), it is only when the words and pictures provide alternative information or contradict each other that they have this effect. When they express similar meanings, the reader's role becomes more passive, as there is little left to interpret.
Conversely, Nodelman (1988, pp. 220-221) claims that pictures limit the text and vice versa, shutting down rather than opening up possibilities, as the words and illustrations inform the reader how to interpret each other. Nevertheless, illustrations can, he argues, influence readers to consider more deeply an idea presented by a text (Nodelman, 1988, p. 248). It is important to note that Hunt and Nikolajeva and Scott write about picture books, rather than illustrated novels. When the words and images are both necessary for conveying meaning, their symbiotic relationship can be used to great effect. It does not necessarily follow that their impact will be the same when the text can stand alone without images. In this article, I will therefore explore whether the illustrations in A Monster Calls limit or open up possible interpretations, and how they affect readers' critical engagement with the text.
One potentially problematic factor of including illustrations in a novel is the contrasting rhythms of the visual and verbal modes. In Words About Pictures, Nodelman (1988, p. 242) points out that it is not possible to look at a picture and read words at the same time, and that each requires a different kind of looking. In addition, he describes narrative rhythm as climactic, with one event leading to the next, fostering a desire to turn the page quickly and find out what happens next in the story. Pictures, alternatively, are individual beats, which interrupt this climactic flow. According to Nodelman (1988, p. 247), a picture book addresses this conflict because the visuals can replace descriptive passages in a novel. These descriptive passages encourage the reader to slow down and imagine the surroundings or characters. But what happens when the text contains those descriptive passages as well as a visual representation? To explore this issue I have discussed with the participants whether the illustrations forced them to pause before the story could continue, and if so, whether they became intrusive to the narrative flow. If they did, they may have affected the readers' engagement with the written text.
From my literature review, the following questions have emerged as being particularly significant to developing an understanding of the role that illustrations play in a novel: • Do illustrations encourage children to engage with text in novels or distract them from it? • Do illustrations affect children's interpretation of a text? • Are illustrations helpful or counterproductive to the picturing process?

The Text
A Monster Calls is a young adult novel which tells the story of Conor, a thirteenyear-old boy whose mother is dying of cancer. One night the yew tree next to his house comes alive and turns into a monster which demands one thing from Conor: the truth. Over the course of the book, the monster tells Conor three stories, each with a difficult or unexpected truth at its heart, and each bearing relevance to the difficulties Conor is facing-friendship issues, bullying, and his difficult relationships with his grandmother and father. At the end of the book, the monster demands that Conor tell him the truth of his real feelings about his mother's illness and the guilt and pain he is wrestling with. In facing up to this truth, Conor is finally able to let go of his mother and imagine a future without her. Throughout the book, Patrick Ness interweaves fantasy and realism, creating multiple layers within the text. Readers must constantly work to construct their own interpretation of the presumed reality of the book, and decide what is ''true'' and what is not. Jim Kay's illustrations are complex and dark, with multiple levels of detail that need to be actively sought out by the reader. The visuals tend to take two different forms: the double page spread, where specific moments in the narrative are depicted, and partial-page illustrations, which bleed into the text and serve to create atmosphere. As with the layers of fiction within the story, the illustrations are not quite fixed into these two different forms. They sometimes spread across multiple pages, with the reader having to turn the page to get the full view. The illustrations thus convey a sense of time and movement as they carry on from one page turn to another.

The Research
This research uses a qualitative methodology to analyse children's reading experiences and perceptions, since the detailed discussions that become possible in qualitative settings tend to yield a richness of data that enables children's voices to be heard more fully (Robson, 2002, p. 25). The six participants were given a copy of the book by their teacher, who asked them to read it at home over the space of three weeks. Half of the participants were given an illustrated copy, the other half a non-illustrated copy. With the book came an explanation of the project and a request for the children to provide an initial response to the text. After three weeks, I conducted individual interviews with them, followed by a group discussion. These were taped on a digital voice recorder, transcribed and analysed.
The initial response was a valuable research tool as it allowed me to gain a fuller picture of the children's opinions (Drever, 2003, p. 8). Participants were given no specific parameters in their responses, so that they would feel comfortable discussing anything that seemed particularly important to them, including aspects which I might not have anticipated. They could draw or write their responses, or both, choosing the mode in which they felt most comfortable (Gardner, 1980, p. 90). The initial responses were discussed at the beginning of the interviews, not only to ensure that I was not misinterpreting them (Rabey, 2003, p. 118), but also to ease the participants into the interview process. I then continued with raising key questions (mostly open) and probes (Drever, 2003, p. 24). The individual interviews helped to establish differences between the two groups, as participants would be more likely to express their own views rather than being tempted to follow the majority view of the group (Woods, 1981, p. 18). The group discussion then gave the children the opportunity to debate their ideas. I also scheduled a more general discussion of illustrations in books, so that any overarching conceptions about the form could be examined and discussed. The research was carried out in compliance with the BERA revised guidelines for educational research (BERA, 2011). 1 Six children from a rural secondary school in Cambridgeshire were chosen to participate in the study. The children I worked with were from across Key Stage 3 (11-14 years old). I chose to work with this age group due to the content and complexity of the book. As A Monster Calls deals with sensitive emotional issues, the participants needed to be old enough to engage with the material effectively. The text also requires a reasonable level of reading ability due to its complex vocabulary and structure. The participants were divided equally by gender and age to minimise bias from these factors. They were chosen by the school's Head of English, who selected them based on their competence with reading and the likelihood that they would enjoy and gain something from the project. All of the participants were confident readers who read for pleasure in their own time. Due to the sensitive nature of the book, in which the protagonist's mother is dying from cancer, the Head of English read the book before selecting the children. I asked her to take their social and family situations into account to ensure that the interviews would not be traumatic for any of the participants. The Head of English was also in the room during the interviews, so that she would be able to follow up with any pastoral care that might be needed as a result of discussing potentially sensitive issues. The school has a very large student body, which enabled me to select participants who did not share friendship groups and had not discussed the book with each other before the interviews. I have changed the names of the participants to protect their identity. This leads to the following list of participants: I coded the interview transcripts to examine responses to the following areas of interest which arose from the literature: engagement, interpretation, picturing, illustrations as limiting or opening up possible interpretations and narrative rhythm. I linked the interview findings with the participants' initial responses (with the exception of Oscar and Nathan, who did not complete the initial response) to check for inconsistencies that might have arisen as a result of the interview process and to gain further insight into the participants' views. I then compared the findings from the two different groups to establish whether there were any significant differences. The results of the analysis are presented below, organised by parameter.

Engagement
The participants with the non-illustrated copy of the book all stated that they enjoyed the book and that they felt engaged by the story line and the character of Conor. Nathan and Bonnie both enjoyed the mix of fantasy and everyday reality, and the complex picture it presented. All three children discussed the author's use of suspense positively. They had all wanted to read on to find out what was going to happen next, and Nathan in particular seemed to view the book as an almost sentient force, saying ''I just thought the book kind of wanted you to read it.'' The narrative drive in this book was therefore highly effective in creating and sustaining engagement. Bonnie and Oscar both commented on feeling very emotionally engaged with the character of Conor. When questioned, Nathan asserted that this was also the case for him, but he did not elaborate on this. The responses in this group suggest that, to a degree at least, the writing is powerful enough on its own to engage a reader emotionally. Bonnie discussed her engagement with the character in terms of her own personal experience of losing her mother to cancer. She said that the book's presentation was a highly accurate portrayal of how she had thought and felt, and that this accuracy helped her to relate to the story. Her response demonstrates the importance of personal context when reading a text, as suggested by Rosenblatt (1938) and Iser (1978).
Like the group with the non-illustrated copies, the participants in the illustrated group all enjoyed the book and felt engaged by it. Fern said: ''I thought it was quite interesting because there was, like, kind of the secret about the truth, and you didn't find out until the end, and you wanted to find out what it was.'' As with the readers of the non-illustrated editions, the suspense built into the storytelling was very important for this group's engagement with the text.
When asked whether they felt emotionally engaged by the character of Conor, all three participants in the illustrated group related his experience to their own lives. Like Bonnie, Fern also had a relative who died from cancer. Neither Clara nor Mark had comparable personal experiences, yet both went beyond simple sympathetic responses to Conor's situation. They discussed the gap in experience between themselves and Conor, reflected on how this distance made them think about both Conor's feelings and speculated how they might feel in his situation. Aside from Bonnie, who had familial experience with cancer, the participants with nonillustrated copies did not attempt to relate the events of the book to their own lives. While this may have been a coincidence, it is possible that the presence of the illustrations made the story feel more real, and therefore encouraged the children to draw parallels with their own lives, though none of them explicitly mentioned this.
When asked for their overall views on the book and how engaged they felt, none of the participants with the illustrated copy mentioned the visuals spontaneously. When asked directly how the illustrations affected their engagement, their views were somewhat mixed. Fern was very positive about their impact, stating that ''the pictures just seem to make it come alive.'' This supports Nodelman's (1988, p. 69) suggestion that illustrations in novels can create energy which enhances the enjoyment of the reader. Mark found the construction of the illustrations especially interesting, and had taken the time to examine them in some depth. Throughout both the individual and group interviews he consistently referred to the illustrations in a very positive light, indicating that they supported his enjoyment of the book. Clara, however, had mixed views: I think that if they have illustrations, you want them to be completely engaged and by them not having to work so hard it takes a bit of less engagement by having something in front of you that will just give you what you need to know without you making any effort for it. So that's why I thought I didn't really like the illustrations that much, but then again, you can see how it can be interpreted in different ways so I enjoyed that kind of side.
For Clara, the active process of picturing was an important part of her engagement with a book. Interestingly, the illustrations in particular scenes, such as Conor's nightmare, do not appear to have interfered with her engagement, but rather encouraged her to engage further by considering multiple possibilities. This suggests an apparent contradiction between her perception of the general reading experience and her detailed discussion of particular moments. Her reflection on the role of illustrations indicated that they were not important, but her discussion of specific moments suggested that they were.

Interpretation and Comprehension
The participants who read a non-illustrated edition demonstrated that they had critically engaged with the text by considering different possible interpretations. Much of the discussion revolved around the ''real'' or fantastical status of the monster and the events in the story. Bonnie's responses indicated that she not only recognised the need to construct her own view of the incident, but also that she actively used clues from the text, such as the lack of other characters' responses to certain incidents, to make judgements and build meaning. All three participants described having weighed several possibilities. The interviews demonstrated this level of engagement quite clearly, but it was not represented in the initial responses. Neither Oscar nor Nathan completed an initial response, so it is difficult to know what their views on these issues were before they were influenced by questioning. Bonnie's initial response described the book as ''thought provoking and compelling,'' which shows some appreciation of having to actively engage with the text, but she did not provide any further details. It is not clear to what extent this minimal response can be attributed to a disinclination to complete written work; brevity notwithstanding, however, her response does suggest that the questioning may have played a significant role in drawing out thoughtful responses from the participants.
One of the key research questions was whether the illustrations would, as Nodelman (1988, pp. 220-221) argued, change and limit interpretive possibilities. Consequently, I examined one key illustrated moment, the monster's first appearance, in close detail. This scene has two major illustrations, the first of the monster outside Conor's window and the second depicting Conor waking up the next day to find yew leaves on the floor of his room.
When the three non-illustrated participants discussed this episode, they all commented on the ambiguity of the scene. The text describes that yew leaves from Conor's dream are lying on his bedroom floor after he wakes up. Their presence made them uncertain about what was real and what was a dream. This ambiguity was also reflected in the responses of the illustrated group. Both Mark and Clara commented on the leaves and the impact they had on their construction of the monster as real or not. Fern noted the ambiguities, but seemed to feel that the leaves indicated that the monster must be real, commenting that the leaves ''made me think as though, like, it made me know it wasn't a dream, and that he's actually seen him.'' Her response showed a far higher level of certainty than the other participants' views. It is possible that seeing the image of Conor's feet standing on the yew leaves made the presence of the monster more concrete for Fern, but she did not reference the illustration as an influence on her ideas. Mark and Clara did not share that interpretation and their meaning making was not directed along the same lines as Fern's. This observation challenges the idea that illustrations restrict the possibility of multiple interpretations (Nodelman, 1988, pp. 220-221).
We repeated the exercise of comparing responses to a particular episode for two further illustrated scenes. The results again showed that there was no single clear interpretation held by all three participants. Nor did the participants have noticeably different views from those with non-illustrated copies, which suggests that the words had a more significant impact on their interpretation than the illustrations. When discussing the visuals more generally, however, all three participants noted how the illustrations affected their interpretations of particular moments. They all discussed returning to illustrated moments in the book to re-examine the scenes there, whereas none of the participants with non-illustrated versions discussed revisiting or reviewing any part of the book. Mark and Clara both explicitly stated that the illustrations had made them reconsider their initial ideas. Furthermore, none of the participants felt that the pictures narrowed down possible interpretations. Both Mark and Fern argued that the illustrations added something to the text, as opposed to taking something away, while Clara was very explicit about the way they opened up possibilities: Uh, well you kind of read it, create your own thoughts, the next page kind of shows you the illustrator's version and how you could take it in a completely different, in different ways, and I thought, if you arranged it like this it could look like that. I thought it was good how different interpretations would come out in different ways, so I kind of went back and thought, no, that could actually look like that, it was interesting.
Clara's comments echo Nikolajeva and Scott's (2000, p. 232) argument that illustrations open up further possible interpretations and encourage revisiting moments within a text. This process may occur with the juxtaposition of image and text within a novel, as well as in a picturebook, where the two modes are more equally balanced.
In examining the initial responses I found further evidence which suggested that the participants with the illustrated copy had considered the book in greater depth than those with the non-illustrated copy. All three participants in the former group completed their initial response, unlike those in the latter. Fern and Clara chose to both draw and write a response, while Mark only used writing. Fern's written response was a simplistic discussion of what she enjoyed about the book, but when explaining her drawing, she talked about how complex the novel's illustrations were, and commented on how they had complemented the story by providing additional detail. Her responses showed that she had thought carefully about the interaction between the pictures and the text and about the role and composition of the illustrations. Clara chose to draw a moment from the end of the book, where Conor has to let go of his mother and allow her to die. For her, she explained, it summed up the message of the book. She discussed the importance of Conor being able to move on and not be held down by his grief. This showed that she had thought deeply about the themes of the book, and was able to imagine how Conor would develop after the end of the story. Clara's written response showed a level of critical engagement which was not found in Bonnie's, who had a non-illustrated copy. Clara analysed the techniques of author and illustrator and acknowledged the role of the reader in constructing meaning, going well beyond a surface level discussion of what she did or did not enjoy about the book.
Mark's initial response similarly showed a level of critical engagement not found in Bonnie's. Like Clara, he discussed the techniques used by both the author and illustrator, and showed admiration for the difficulty involved in creating the pictures successfully. He too addressed the reader's need to take an active role in constructing meaning, and identified what he saw as the book's central message of how to deal with the pain of letting go of a loved one. Taken together, the initial responses suggest that those with the illustrated copy engaged more deeply and critically with the book than the others.

Picturing
The ability of the participants with a non-illustrated copy to visualise what they were reading varied considerably. Oscar had a very strong picture of the characters in his mind, building his own visualisations with both clues from the text and his own ideas about what bullies and intelligent boys look like. By contrast, Bonnie found visualising the monster extremely difficult. For her initial response she tried to draw the monster to help her develop a clear picture of it in her mind. This attempt initially seemed extremely significant, as it appeared that being able to clearly picture the character might help her make meaning, as suggested by Benton and Fox (1985) and Fry (1985). During the course of the group interview, though, Bonnie rejected this idea: Me: I know you were saying that you found it quite hard to visualise the tree. Bonnie: Yeah, but I didn't mind that though, because I liked how it meant that you could kind of decide what it looked like in your head.
As Bonnie stated in her individual interview that she never gained a fully clear image of the tree, this response suggests that mental pictures do not have to be fixed and clear in order to create meaning, challenging Fry's conclusion that picturing is fundamental to reading (1985, p. 66). Bonnie engaged with the text, and demonstrated that she was able to make meaning despite not having a clear mental image of the characters. She also felt that the story was alive to her, commenting at various times during her interview that she found it ''gripping,'' and that ''you are still thinking about it afterwards.'' This challenges Benton and Fox's assertion that picturing is the most important way a reader brings a story to life (1985, pp. 5-6). Despite their differing picturing abilities, all three of the participants with nonillustrated copies displayed a negative attitude towards illustrations. When asked in the group interview if they would have preferred to read an illustrated copy of the book, all three participants felt very strongly that illustrations would have prevented their own ability to picture.
Bruno Bettelheim (1976, pp. 59-60) has argued that illustrations can restrict a reader's ability to create their own pictures, but the participants with an illustrated edition of A Monster Calls did not appear to have this difficulty. Fern described the pictures as complementing the writing by providing more detail. For her, they were an addition to her own images, rather than a substitute. Mark had less developed visualisation skills than some of the other participants, and believed the illustrations supported the picturing process. Throughout both his individual interview and the group discussion Mark referred several times to the illustrations as actively helping him make meaning from the text by enabling him to visualise the events of the story.
Clara had a somewhat different approach to the picturing process. She had very clear ideas of her own pictures, but also enjoyed seeing the illustrator's ideas: I think they stayed close to the book, but you can always have a different view, you know, you can have a different image in your mind, it's just, but I think they stayed close to it by, how, if you look back, like I looked at it, and then you read it again, and you could actually see it as if it was like that, I kind of saw it a little bit differently, but it's possible it could be like that, it's interesting to see.
Clara's response shows that the illustrations did not replace her own visualisations, but rather ran parallel to them. For her, having more than one visualisation ''option'' opened up the possibilities of the book, rather than competing with each other. She did not feel that there was a ''correct'' visualisation of a moment, nor did Kay's illustrations take precedence over her own pictures. However, in the group discussion, all three participants expressed being glad that the illustrations only showed Conor's silhouette and never his features, as they felt this would have limited their imaginations. It is possible that the lack of detail in the illustrations made it easier for the participants to retain their own visualisations alongside the pictures. Since they did not engage with more detailed illustrations during the study (which might have provided evidence for comparison), it is not possible to assert this with any confidence. Considering that all three stated in their interviews that the illustrations did not replace their own visualisations, yet expressed reservations about the role of illustrations when discussing them in general situations, there are some contradictions in their comments which seem to be related to talking about illustrations in the abstract and reflecting upon specific examples. As such, it is difficult to assess the impact that the style of illustrations had upon the reading experiences of the participants.

Narrative Rhythm
A key question for this study was whether the illustrations would have an impact on the climactic narrative rhythm of the book by causing the reader to stop frequently. All three participants with illustrated copies talked about wanting to read quickly to find out what would happen next, suggesting that climactic narrative rhythm was still dominant in the illustrated version of the book. When asked whether the illustrations interfered with that rhythm, Fern expressed a preference for the doublepage spreads over those which contained both illustrations and words. She felt that having to switch between reading modes on one page was so disruptive that she simply chose not to look at the partial-page illustrations. While Clara agreed with Fern, Mark felt differently. He acknowledged that having two different modes on the same page produced a contrast, but this inconsistency does not appear to have caused the same level of disruption for him as it did for Fern and Clara. Although all of the participants stopped to examine the full-page illustrations closely, none of them felt that these pauses interrupted the climactic rhythm, but rather allowed them to explore a particular moment in more depth and consider further possibilities. Their responses suggest that when the two modes are kept separate the impact upon rhythm is minimised, but when the reader is asked to switch between modes on the same page, there is a level of disruption to that rhythm which can have a negative impact upon the reader's experience of the text.

Conclusion
Overall, the illustrations in A Monster Calls did not have a big impact on engagement. When initially asked whether or not they liked the book and felt engaged by the story, none of the children with an illustrated copy mentioned the visuals. While Fern, Mark and Clara all had positive views of the illustrations, their impact was strongly subordinate to that of the writing. The reader's ability to empathize with Conor did, however, correlate with the presence of illustrations. Out of those with the non-illustrated copy, only Bonnie, who had personal experience of familial cancer, talked about the book in relation to her own life. In contrast, all of the children in the illustrated group related the situation to their own lives. The sample size for this study is too small to assert with confidence that the characters felt more real to them because of the illustrations, but it raises an interesting correlation that is worthy of further exploration.
It is also worth noting that the participants in this study all regularly read for pleasure and were confident readers. However, they rarely read illustrated books and may therefore have automatically assumed that the illustrations were less important than the story, an assumption that may have been influenced by the relatively low status of images in the average classroom (Millard and Marsh, 2001). It would be worth exploring whether illustrations have a greater impact upon the engagement of reluctant readers, who may not find the written word to be engaging by itself. Since there is some evidence that illustrations in comic books and graphic novels can help to engage reluctant readers (Dorrell et al., 2004;Norton, 2003), it is possible that illustrations in novels may have a similar impact. However, there is a very different ratio of text to image in illustrated novels and in graphic novels, so further research would need to be conducted to ascertain whether illustrations in novels can encourage engagement.
The interviews make clear that the illustrations did not have significant impact on the children's overall interpretation of the larger issues within the text, such as whether or not the monster was real. Certainly there was no evidence for Nodelman's suggestion that illustrations narrow down options (Nodelman 1988, pp. 220-221). Although some participants made similar interpretations, no two were quite the same, and there was no distinct gap between the two groups. However, the illustrations did affect the way the participants constructed their ideas, with the readers of the illustrated edition re-examining moments in the text and challenging their initial assumptions in a way that was not reported by the non-illustrated group. Despite a lack of evidence from the participants' responses to suggest that the illustrations narrowed down interpretive possibilities, there was definitely an underlying assumption within both groups that illustrations limited interpretations, even among participants who described how the illustrations had presented them with additional interpretations. This apparent contradiction might be a result of prejudices about illustrations, and this could raise a potential barrier to the effective use of pictures when reading. Although all the participants with an illustrated copy in this study used the images very effectively to engage more deeply with the book, a negative perception of illustrations may deter children from choosing an illustrated text and thus deprive them of the richer experience of re-exploring ideas that illustrations appear to provide.

Compliance with Ethical Standards
Conflict of interests The author declares that she has no conflict of interests.
Ethical approval All participants in this study received consent forms which described the nature of the research and gave explicit consent that responses could be used in publications and conference presentations. As the participants were all under 18 years of age, these consent forms were all signed by a parent or legal guardian. These are available to view on request.In addition, all participants gave their verbal consent to take part in the research and were informed that they could cease participation at any time they chose.The Head of English of the school at which the research was carried out also signed a consent form as a 'gatekeeper', (available to view on request) and was fully informed of the ethical considerations surrounding the research project. She was also present in the room whilst the interviews were conducted and prepared to provide any necessary pastoral support following the research.The research was carried out to BERA revised guidelines for educational research (BERA, 2004).
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.