Kindergartners' visual attention during storybook reading and subsequent vocabulary learning

Kindergartners’ visual attention during storybook reading and subsequent vocabulary learning

Roxette M. van den Bosch

Leiden University

Research Master Thesis (June the 26th, 2013)

Developmental Psychopathology in Education and Child Studies

Roxette M. van den Bosch, S0827843

r.m.van.den.bosch@umail.leidenuniv.nl

First reader: Z. K. Takacs, MSc

Abstract

Children’s visual attention during storybook reading and their resulting vocabulary learning were

examined. Participants were 21 Dutch kindergartners with an average to large general receptive

vocabulary. Eye-tracking was used to assess visual attention within the illustrations during storybook

readings in which text and illustrations were presented either simultaneously or successively.

Irrespective of mode of presentation, the children appeared to follow the oral text when visually

inspecting the illustration. During simultaneous readings participants, especially the ones with lower

inhibitory control skills, looked longer at the parts of the illustration that were highlighted by the text

than at the rest of the illustration. In contrast, during non-simultaneous readings they looked as long at

the text-relevant as at the text-irrelevant parts or even longer at the text-irrelevant parts. Children’s

visual attention was positively related to their vocabulary learning when the illustrations and the oral

text were available at the same time. A positive effect of the book readings was found on word

learning from the books as compared to a control condition. However, no differences in vocabulary

learning between simultaneous and non-simultaneous readings were found. In terms of attention,

displaying storybook illustrations on the whiteboard, visible to all children, during classroom reading

activities might be beneficial for all children, but especially for those with lower inhibitory control

skills. For them, displaying the illustrations in a large format while they listen to the narration will

guide them in focusing their attention and resisting distracters within the storybook illustrations.

Kindergartners’ visual attention during storybook reading and subsequent vocabulary learning

Reading storybooks to young preschool children is a common activity in most western

families (Bus, 2002). Reading aloud does not only take place in children’s homes, it is also a valued

and ever-recurring activity in kindergartens; 90% of Dutch kindergarten classroom teachers read

storybooks with their pupils at least three times a week or even on a daily basis (Ghonem-Woets,

2009). Shared reading is often considered a source of entertainment and pleasure (Sonnenschein et al.,

1997) but an extensive literature base (see meta-analyses by Bus, van IJzendoorn, & Pellegrini, 1995;

Mol & Bus, 2011; Mol, Bus, de Jong & Smeets, 2008) indicates that shared reading is also a

meaningful and important source of literacy and language learning. It contributes to children’s

emergent literacy, their reading performance and their language development (Bus et al., 1995).

Shared storybook reading stimulates the development of the prerequisites for reading and

writing prior to formal literacy instruction (Bus, 2005). Examples of such prerequisites or emergent

literacy skills are interest in books, acquaintance with books and written language, print awareness and

phonological awareness (Justice & Kaderavek, 2002). As such, being read aloud to from an early age

lays the foundation for learning how to read and to appreciate reading. At the start of formal reading

instruction and learning to read in school, children who have been exposed to shared storybook

reading have an advantage compared to children who were not exposed to storybooks, in terms of

larger vocabularies and more developed comprehension skills (Bus et al., 1995; Mol & Bus, 2011).

Being read aloud to remains important throughout the primary school years; shared reading explained

8% of the variance in school-aged children’s reading abilities (Bus et al., 1995).

The effects of shared storybook reading are even stronger for children’s language

development than for their emergent literacy and reading skills (Bus et al., 1995). Children’s age at

which their parents start reading aloud to them turns out to be a strong predictor of children’s language

skills (DeBaryshe, 1993), indicating that initiating shared storybook reading early on – even before

children start talking – is important in promoting children’s language development and skills.

Exposure to print – which for young children comes down to being read aloud to – explains 12% of

the variance in kindergartners’ oral language skills and continues to be influential later on; in primary

school, middle school, high school, and university print exposure explains respectively 13, 19, 30, and

34% of the variance in students’ language skills (Mol & Bus, 2011).

Given the nature of the language used in storybooks, it is not surprising that shared storybook

reading promotes language skills. Written language – such as displayed in storybooks – is richer, more

varied, and more complex than spoken language as occurring in ordinary conversations or in

prime-time television shows (Hayes & Ahrens, 1988; Sulzby, 1985). This implies that shared storybook

reading provides young children with high quality linguistic input – containing more sophisticated

syntax and low-frequency words – which stimulates their language development (Sénéchal, LeFevre,

Hudson, & Lawson, 1996). Shared storybook reading seems to be especially helpful for vocabulary

learning (Sénéchal & LeFevre, 2001). As storybooks contain a more varied vocabulary and include

words which are used less often in daily conversations (Sulzby, 1985), listening to storybooks – read

aloud by adults – appears to be an effective activity to promote incidental word learning (Ewers &

Brownson, 1999).

The acquisition of new words is a complex process (Carey, 1978). When children encounter a

novel word for the first time they form an initial idea about the words’ meaning which is entered in

their mental lexicon (fast mapping). After encountering the same word multiple times their knowledge

about the word becomes more complete and its meaning and characteristics are encoded in memory

(slow mapping). This implies that several encounters with an unknown word are required before a

particular word is learned (Jenkins & Dixon, 1983). Although some studies (e.g., Ewers & Brownson,

1999; McLeod and McDade, 2011) revealed that children are able to learn novel words after brief

single storybook exposure, most studies (e.g., Biemiller & Boote, 2006; Horst, Parsons, & Bryan,

2011) stress the importance of repeated readings. In general, children enjoy hearing the same story

several times (Biemiller & Boote, 2006) and reading a storybook two or more times to children

facilitates vocabulary learning and increases the probability that new words will be learned from the

story (Sénéchal & LeFevre, 2001). Moreover, children do not only benefit from repeated reading of

the same stories, they also profit from hearing new or unknown words repetitively in the same story

(Robbins & Ehri, 1994). According to Robbins and Ehri (1994) the same storybooks should be read at

least two times to kindergartners and the unknown words should be included several times within the

story in order to extend children’s receptive vocabulary (i.e., understanding of words).

Vocabulary is a strong predictor of reading comprehension, reading achievement, and

academic success (Biemiller & Boote, 2006; Torgesen, 2000). To comprehend written text children

must identify and understand the individual words in the text in order to construct and represent the

meaning of the text. If a words’ meaning is unknown that particular word will not be understood

(Nation, 2006) and therefore a low vocabulary – implying that many words in a text will be unknown

– will lead to reading comprehension problems (Biemiller & Boote, 2006), which in turn will interfere

with academic success (Cain & Oakhill, 2007). This negative causal spiral illustrates the importance of

paying attention to children’s vocabulary development and examining opportunities to stimulate their

vocabulary acquisition from early on.

There is evidence that (repetitive) shared storybook reading has a direct positive influence on

preschool children’s vocabulary development (Bus et al., 1995; Mol & Bus, 2011; Mol et al., 2008;

Sénéchal & LeFevre, 2001). However, it is not yet known exactly how children learn word meanings

from storybook exposure (Horst et al., 2011). An additional, and even more important, motive for a

closer examination of vocabulary learning through storybook reading is that vocabulary gains obtained

by shared storybook reading are modest (e.g., Justice, Meier, & Walpole, 2005; Robbins & Ehri,

1994). Further research is necessary to establish under which conditions storybook reading is most

beneficial in terms of vocabulary learning. The present study provides more insight into vocabulary

learning during shared storybook reading and strategies to enhance children’s vocabulary learning

through shared storybook reading.

Storybooks do not only provide children with rich and varied linguistic information (Sulzby,

1985), they also offer detailed and appealing illustrations which give a pictorial description of the text

(Carney & Levin, 2002; Verhallen & Bus, 2011). In fact, storybooks tell their story in two ways;

verbally by the text and non-verbally by the illustrations (Carney & Levin, 2002). The illustrations

seem to play an important role in reinforcing the informative value of textual information and a

facilitating role in understanding, remembering, and learning from text (David, 1998; Glenberg &

Langston, 1992). As such, illustrations appear to be helpful in identifying word meanings and

vocabulary learning as well (Sadoski, 2005). Although it seems likely that the illustrations play a

meaningful role in vocabulary learning through shared book reading, it is currently unknown how

exactly children use the additional information offered by storybook illustrations (Verhallen & Bus,

2011) while being read aloud to. Do children treat the text and the illustrations as isolated sources of

information, or do they integrate both sources of information to establish a coherent unity?

According to Sipe (1998) readers and listeners construct an integrated meaning of a story by

constantly switching their attention between the words in the text and the illustrations. In fact, the

verbal information provided by the text and the visual information provided by the illustrations in

storybooks should not be interpreted separately. On the contrary, the text and the illustrations are often

inseparable in the sense that both sources of information complete and reinforce each other (Sipe,

1998). The text might be more precise than the illustrations and the illustrations may go beyond the

text (Carney & Levin, 2002; Verhallen & Bus, 2011). Indeed, the combination of text and illustrations

has been found to be more beneficial for story comprehension than either the text or the illustrations

alone (Sipe, 1998). The same line of reasoning seems applicable for vocabulary learning. The

combination of text and illustrations provides a more complete understanding of the word’s meaning

than either the (spoken or written) text or the illustrations on its own (Mayer, 2003; Sadoski, 2005).

The power of the combination of text and illustrations can be explained by Sadoski and

Paivio’s (2004) dual coding theory and Mayer’s (2003) cognitive theory of multimedia learning. The

dual coding theory assumes that text and illustrations are represented and processed in two different

mental systems; linguistic information is encoded in the verbal system, while visual information is

encoded in the non-verbal system (Sadoski & Paivio, 2004). These two different mental systems

operate independently in information processing, but they can also stimulate each other to optimize

information processing. Applying this theory to shared storybook reading, the spoken narration will be

encoded in the verbal system, whereas the information obtained by looking at the illustrations will be

encoded in the non-verbal system. Information which has been encoded in two different ways appears

to be more exhaustive and more detailed, with the result that dual-encoded information is more

comprehensive and easier to retrieve than information which has been encoded in only one way

(Sadoski & Paivio, 2004).

The combination of verbal and non-verbal (visual) information does not only lead to better

comprehension (Sadoski & Paivio, 2004). According to Mayer’s (2003) cognitive multimedia learning

theory dual coding also facilitates learning. As compared to either words or illustrations, combining

both fosters deeper learning. The combination of text and illustration strengthens children’s memory

traces for word meanings by establishing mental connections between the spoken words and their

visual referents, which enhances the chance that particular words will be learned (Acha, 2009;

Sadoski, 2005). Multimedia learning, however, is not effective under all circumstances (Mayer, 2003).

For example, to maximize learning from multimedia material corresponding verbal and visual

information should be presented at the same time, which is referred to as the temporal contiguity

principle (Mayer, 1999). Presenting verbal and visual information simultaneously facilitates dual

coding because it is easier to make connections between verbal and visual representations when both

information types are in the working memory at the same time.

The results of eye-tracking studies (Eberhard et al., 1995; Evans & Saint-Aubin, 2005;

Verhallen & Bus, 2011) provide preliminary support for Sipe’s (1998) assumption that readers or

listeners integrate verbal and visual information into a coherent whole, and interpret the words in

terms of the illustrations and the illustrations in terms of the words. That is, individuals’

eye-movement patterns – or visual attention – seem to coincide with the spoken text. The study by

Eberhard and colleagues (1995) for example, showed that eye movements and fixations are strongly

related to the accompanying spoken instructions. When participants listened to spoken instructions,

they looked at the objects that were mentioned in the instructions. According to Evans and

Saint-Aubin (2005) the same principle applies to shared book reading: Pre-schoolers’ visual attention while

listening to the narration was to some degree dependent on the spoken text. When the text highlighted

a small detail of the illustration, children fixated more on that particular detail than when the text did

not mentioned that particular detail. In addition, Verhallen and Bus (2011) showed that five-year-old

learners of Dutch as a second language fixated more often and for longer on the visual elements that

were highlighted by the text compared to not highlighted elements of the illustration during storybook

reading.

Taking into account the preliminary results of eye-tracking studies (Eberhard,

Spivey-Knowlton, Sedivy, & Tanenhaus, 1995; Evans & Saint-Aubin, 2005; Verhallen & Bus, 2011) that

children link the storybook illustrations to the spoken text and applying the temporal contiguity

hypothesis (Mayer, 1999) to shared storybook reading implies that the oral text and the accompanying

illustrations of the storybook should be presented at the same time to children. This seems logical, but

during shared reading activities in kindergarten classrooms text and illustrations are often not

presented simultaneously to the children. That is, the majority of shared book reading activities in

kindergarten classrooms (72%) are class activities (Ghonem-Woets, 2009), during which all children

sit around the teacher who reads the storybook. Mostly, teachers read a page and show the

hear the text before they see the illustration, which does not facilitate the integration of two sources of

information and may even hinder children’s text comprehension and vocabulary learning.

The extent to which children benefit from storybook exposure – and thus from the

combination of verbal and visual information – is likely to be affected by child characteristics as well

(Davidse, de Jong, Bus, Huijbregts, & Swaab, 2011; Plass, Chun, & Mayer, 1998). In terms of

vocabulary learning, children’s a priori vocabulary knowledge is suggested to be an important factor

in individual differences in profiting from listening to storybooks (Ewers & Brownson, 1999). Several

studies (e.g., Justice et al., 2005; Penno, Wilkinson, & Moore, 2002; Sénéchal, Thomas, & Monker,

1995) have revealed that children with higher language and vocabulary skills profit more from

storybook reading than children with lower vocabulary skills, as indicated by greater gains in

vocabulary learning.

It seems also plausible that children’s individual differences in regulatory skills – like

attention, memory, and inhibitory control – influence the extent to which children benefit from

literacy input (Davidse et al., 2011; Kegel, van der Kooy-Hofland, & Bus, 2009). In support of this

assumption, the study of McClelland and colleagues (2007) revealed that behavioral control predicted

preschoolers’ emergent literacy skills, vocabulary, and math skills. Behavioral control or inhibitory

control refers to the ability to focus attention, to ignore distracters, and to suppress automatic

behavioral responses (Blair, 2002). In a different study (Bierman, Nix, Greenberg, Blair, &

Domitrovich, 2008), inhibitory control was related to the acquisition of language and emergent

literacy skills. However, the study by Davidse and colleagues (2011) did not find support for a

moderating role of inhibitory control in profiting from storybook exposure. The present study

considers both children’s a priori vocabulary and inhibitory control skills.

Aim, Research Questions, Hypotheses, and Relevance of the Study

The main aim of the present study is to investigate whether children’s visual attention within

the illustrations is different when the oral text and the illustrations of the storybooks are presented

simultaneously or successively. As a secondary aim, it will be investigated whether children’s visual

attention within the storybook illustrations is related to vocabulary learning and whether the children

learned words as a result of inspection of the illustrations during the book readings. In addition, the

role of children’s inhibitory control skills will be examined. The research questions are:

1.

Does the way of presenting the storybook – oral text and illustration at the same time

(simultaneous presentation) or narration first and illustration thereafter (non-simultaneous

presentation) have an effect on children’s visual attention within the illustrations?

2.

Is children’s visual attention related to their vocabulary learning?

3.

Did the children learn words from the storybook exposures? Furthermore, did children learn

more words from the simultaneous as compared to the non-simultaneous condition?

Firstly, based on the dual coding theory of Sadoski and Paivio (2004) and previous eye-tracking

studies (Eberhard et al., 1995; Evans & Saint-Aubin, 2005; Verhallen & Bus, 2011) it is hypothesized

that children spend more time looking at the elements of the illustration that are highlighted by the text

than at the elements which are not mentioned in the text. According to Mayer’s (1999) principle of

temporal contiguity, this will be especially true when text and illustrations are presented

simultaneously, rather than successively. Secondly, based on Mayer’s (2003) multimedia learning

theory, it is hypothesized that simultaneous presentation of oral text and illustration facilitates

vocabulary learning, as indicated by a positive relationship between the time children spent looking at

the text-relevant parts of the illustration and their vocabulary growth. Thirdly, it is expected that the

book readings, and the simultaneous readings in particular, will result in vocabulary gains.

Method

Participants

Participants were recruited by inviting schools to partake in the study via e-mail and phone. In

the participating school – a school in a small village in the western part of the Netherlands – consent

forms were handed out to the parents of all children from two kindergarten classes. Consent was

obtained for 29 children and those children participated in the three test sessions prior to the book

reading intervention. After those three sessions eight children were excluded from further

participation: Five six-year-old children were excluded to narrow down the age range, one child had

outlying scores for some of the pretests, one child refused to participate any longer and one child went

on holiday. The final sample consisted of 21 children (12 boys, 9 girls) of four and five years old (M =

63.76 months, SD = 7.07).

Design

The present study was an experimental study using a within-subject design. It included background

measures, a pretest measure, an intervention period, and posttest measures. The intervention consisted

of two book reading sessions in which storybooks were read to the children. Their eye

movements were registered with an eye-tracker while they listened to the stories and the

accompanying illustrations were shown to them on a screen. Children read two books per session: one

book in the simultaneous condition and another book in the non-simultaneous condition. In the

simultaneous condition the text and the accompanying illustration were presented at the same time,

whereas in the non-simultaneous condition the text was presented first and the accompanying

illustration was presented thereafter. A third book – which the children did not read – served as a

control condition.

The first and second reading sessions were identical: During the second session children

received the same two books as during the first reading session and the books were presented in the

same condition. In other words, one story was presented to them in a simultaneous fashion and the

other in a non-simultaneous fashion and the children heard each story twice. To avoid any effect of

books or order of presentation, as shown in Table 1, the order of the books and which book each child

received in which condition – simultaneous, non-simultaneous, or control – were counterbalanced.

Table 1

Intervention reading groups

Book reading session 1

Book reading session 2

Book

Condition

Book

Condition

Group 1

N = 4

Rokko Krokodil

Na-apers

Beer is op Vlinder

Sim

Non-sim

Control

Na-apers

Rokko Krokodil

Beer is op Vlinder

Non-sim

Sim

Control

Group 2

N = 4

Beer is op Vlinder

Rokko Krokodil

Na-apers

Non-sim

Sim

Control

Rokko Krokodil

Beer is op Vlinder

Na-apers

Sim

Non-sim

Control

Group 3

N = 4

Na-apers

Rokko Krokodil

Beer is op Vlinder

Sim

Non-sim

Control

Rokko Krokodil

Na-apers

Beer is op Vlinder

Non-sim

Sim

Control

Group 4

N = 3

Beer is op Vlinder

Na-apers

Rokko Krokodil

Non-sim

Sim

Control

Na-apers

Beer is op Vlinder

Rokko Krokodil

Sim

Non-sim

Control

Group 5

N = 3

Beer is op Vlinder

Na-apers

Rokko Krokodil

Sim

Non-sim

Control

Na-apers

Beer is op Vlinder

Rokko Krokodil

Non-sim

Sim

Control

Group 6

N = 3

Rokko Krokodil

Beer is op Vlinder

Na-apers

Non-sim

Sim

Control

Beer is op Vlinder

Rokko Krokodil

Na-apers

Sim

Non-sim

Control

Note. Sim = Simultaneous condition; Non-sim = Non-simultaneous condition.

Procedure

Preceding the book reading sessions the Peabody Picture Vocabulary Test and the Peg

Tapping Task were administered to assess general vocabulary and inhibitory skills. A self-constructed

Meaning Recognition Task testing the vocabulary knowledge of the target words of the books was

administered before and after the book reading intervention to assess vocabulary learning. Some other

background measures and self-constructed vocabulary posttests were administered as well, but are not

reported in the present paper.

The tests and the readings were spread over seven sessions of approximately 20-25 minutes.

The behavioral testing was carried out by trained Bachelor’s and Master’s students and the two book

reading sessions on the screen of the eye-tracking device were conducted by a trained Master’s student

and a PhD student. All sessions took place in a separate room in the school and were recorded on a

videotape. Children were collected from the classroom and brought back by one of the experimenters.

Two experimenters were present during the testing and intervention sessions. In the testing

sessions one of them tested the child and the other was responsible for handling the video camera and

writing down the responses of the child. In the reading sessions one experimenter instructed the child

and the other handled the eye-tracker. Eye movements of the children were recorded during both book

reading sessions. The readings were preceded by a five-point calibration procedure in which the

eye-tracker measures the characteristics of the eyes of the participant, which is necessary for calculating

the direction of the eye movements (Tobii Technology, 2010). The children were rewarded with

stickers after each test which they collected on a ‘diploma’.

Afterwards, all tests were scored from the videos by the first scorer – which was the same

person for all tests – and a second scorer. In the rare case of missing or unclear video recordings the

notes taken during testing were used. Each disagreement between scorers was resolved through

discussion and interrater-reliabilities were calculated for the Peg Tapping task and the vocabulary test.

Materials

In the intervention three storybooks were used: Beer is op Vlinder [Bear is in love with

Butterfly] (van Haeringen, 2004); Na-apers [Mimics] (Veldkamp & de Boer, 2006); and Rokko

Krokodil [Rokko the Crocodile] (de Wijs & van den Hurk, 2001). These three books are

age-appropriate, include short text chunks per page, and are quite similar to each other. The children were

not familiar with these three books.

For each of the three books the illustrations were scanned and divided over 19 pages. The

original text of the books was recorded by professional narrators and the written text was removed

from the illustrations. The scanned illustrations and the oral text were presented to the children on the

screen of the eye-tracker. In both conditions – simultaneous and non-simultaneous – children heard the

same text and saw the same illustrations for the same amount of time. The only difference between the

two conditions was that in the simultaneous condition the oral text of each page was played while the

children were looking at the illustration of that page, whereas in the non-simultaneous condition the

children heard the text of the page first (while looking at a black screen) and saw the illustration of that

page only thereafter. Because children were presented the illustrations for exactly the same amount of

time in both conditions, books took twice as long to play in the non-simultaneous as compared to the

simultaneous presentation.

For each of the three stories one detailed illustration was selected for the analyses of visual

attention. These detailed illustrations (Figure 1) included both elements which were and were not

highlighted by the text. The parts of the illustration highlighted by the text are the text-relevant parts

and the rest of the visual elements are the text-irrelevant parts. The text-relevant parts of the

illustration were defined as the areas of interest (AOIs), as shown in Figure 1, and children’s visual

attention for these areas was compared to their visual attention for the rest of the illustration.

Figure 1. The three selected illustrations for the analyses of the eye-tracking data (a) and the areas of

interest (AOIs) in these illustrations (b). From top to bottom, the illustrations of Beer is op Vlinder

[Bear is in love with Butterfly], Na-apers [Mimics], and Rokko Krokodil [Rokko the crocodile] are

shown.

Twenty-seven target words – nine words per book – were selected in order to investigate

vocabulary learning. These words (see Appendix) were the most difficult words of the books and they

were not mentioned more than four times in the story.

Apparatus

The Tobii T120 eye-tracking device – a remote eye-tracker system – was used to register the

eye movements of the children while they were looking at the storybook illustrations on the screen of

the Tobii. The distance between the children’s eyes and the screen was approximately 60 centimeters.

The Tobii uses cornea reflection and infrared to determine the position of the participants’ eyes on the

screen (Tobii Technology, 2010). The reflection patterns and other visual information of the

participant are used to calculate the three-dimensional position of each eye in order to define at which

part of the screen the participant is looking. The Tobii T120 eye-tracker collects 120 gaze data points

per second for each eye and has an accuracy of 0.5º, which corresponds to an average error of 0.5 cm

between the measured gaze direction and the actual gaze direction.

Eye movements serve as a measure of visual attention (Henderson, 2003). In the present study

the total duration of the fixations at the text-relevant and text-irrelevant parts of the illustration was

used as an indicator of visual attention. Total duration of fixations refers to the time children spent

looking at those parts of the illustrations over the whole time the picture was presented. Tobii Studio

2.2.6 was used to analyze the eye-tracking data.

Measures

Peabody Picture Vocabulary Test (PPVT-III-NL). The Dutch version of the PPVT

(Schlichting, 2005) is a standardized measure of children’s general receptive vocabulary. Receptive

knowledge of spoken words is measured by a multiple-choice task consisting of 17 sets of 12 items,

which were presented on a computer screen for the present study. Each item includes four pictures and

children are asked to pick the correct picture in accordance with the word presented orally by the

experimenter. This task results in a raw score and a standard score (with a mean of 100 and a standard

deviation of 15), which is an age-related indicator of children’s general language skills (Schlichting,

2005). In the current sample children’s PPVT scores ranged from 96 to 135 (M = 111.10, SD = 10.69),

suggesting that these children possessed an average to large receptive vocabulary.

Peg Tapping Task (PT). In this task of inhibitory control (Diamond & Taylor, 1996) children

are asked to tap twice when the experimenter taps once and vice versa. In order to react appropriately

children have to hold the two rules in mind and they have to inhibit the natural response to imitate the

experimenter. The task consisted of 16 items and the resulting score is the percentage of correct

reactions from all items. Inter-rater reliability for this task was high (r = .948, p = .01).

Meaning Recognition Task (MR). The MR task is a self-constructed measure of receptive

vocabulary knowledge of the target words of the three books and includes two language-based items –

one true, one false – for the 27 target words, resulting in 54 items. The experimenter provides the

children with sentences which include the target words (e.g., “You shake hands when you meet new

people.” and “Shaking hands is the same as waving.”) and children have to decide whether this

sentence is true or not. The resulting score is the percentage of items answered correctly per condition.

The MR task was both pre- and post-tested in order to establish vocabulary learning. Cohen’s Kappa

inter-rater reliability was .953 (p < .001) for the pretest and .942 (p < .001) for the posttest scores.

Data Analyses

due to equipment failure. These missing values were replaced by the average scores of the rest of the

participants.

For the purpose of follow-up analyses, children with a relatively low Peg Tapping score (the

lowest 50%) were assigned to the low group (N = 11, M = .81, SD = .05) and children with a relatively

high score (the highest 50%) were assigned to the high group (N = 10, M = .98, SD = .03). One

outlying Peg Tapping score (Z = -2.55) was winsorized and for the low Peg Tapping group one

outlying value for the total fixation duration at the text-irrelevant parts of the illustration for the first

simultaneous reading (Z = -2.52) was winsorized. This winsorized eye-tracker variable was used in the

separate analyses for the low and the high Peg Tapping groups.

Results

Data Inspection

The assumption of normality was met for children’s total fixation duration at the text-relevant

and text-irrelevant parts of the illustration, for both the simultaneous and non-simultaneous readings.

Children’s scores on the Peg Tapping Task and the Meaning Recognition Pretest and Posttest were

normally distributed as well.

For both the low and the high Peg Tapping group, the eye-movement variables – including the

winsorized variable of total duration of fixations at the text-irrelevant parts in the first simultaneous

reading – and the winsorized Peg Tapping variable had a normal distribution.

Eye-gaze Patterns

In order to answer the first research question – whether children’s looking behavior at the

illustrations differed between the simultaneous and non-simultaneous presentation of oral text and

illustrations – a 2 x 2 x 2 repeated measures analysis of covariance (ANCOVA) was conducted. Total

fixation duration within the illustrations was entered as the dependent variable, condition

(simultaneous and non-simultaneous), relevancy (parts highlighted by the text and text-irrelevant

parts), and session (first and second reading sessions) as the within-subject factors, and Peg Tapping as

the covariate. The results revealed a significant main effect of relevancy (F(1, 19) = 5.67, p = .028, ηp²

= .23) and significant interaction effects between relevancy and condition (F(1, 19) = 15.92, p = .001,

η

relevancy, condition, and Peg Tapping (F(1, 19) = 14.68, p = .001, η

² = .44).

The main effect of relevancy implied that children looked longer at the parts of the

illustrations that were highlighted by the text (M = 6.12, SE = .46) than at the text-irrelevant parts (M =

6.09, SE = .28). According to the interaction effect between relevancy and condition this was only true

for the simultaneous readings (F(1, 19) = 12.94, p = .002, ηp² = .41). When text and illustrations were

presented simultaneously, children looked longer at the parts highlighted by the text (M = 6.55, SE =

.77) than at the text-irrelevant parts (M = 5.89, SE = .35). In contrast, there was no difference in

children’s looking times for the text-relevant parts (M = 5.69, SE = .49) and the text-irrelevant parts

(M = 6.30, SE = .43) when the illustrations were only shown after the text (F(1, 19) = .10, p = .754).

In order to interpret the direction of the interaction effects including the Peg Tapping score

follow-up analyses were conducted separately for the low and the high Peg Tapping groups. The

gender distribution was equal for both groups (χ² (1) = .06, p = .801) and the groups did not differ on

age (t(19) = -.95, p = .355), general receptive vocabulary (t(19) = -.20, p = .843), or pretest knowledge

of the target words either (t(19) = -.81, p = .427).

These follow-up analyses were 2 x 2 x 2 repeated measures analyses of variance (ANOVAs)

with total duration of fixations within the illustration as the dependent variable and with condition,

relevancy, and session as the within-subject factors. The results did not reveal a main effect of

relevancy for either the low (F (1, 10) = 1.31, p = .280) or the high Peg Tapping group (F (1, 9) =

1.58, p = .241). The interaction between relevancy

and condition, on the other hand, was significant

for the low group (F (1, 10) = 5.70, p = .038, ηp² = .36), but not for the high group (F (1, 9) = 3.46, p =

.096), implying that the looking behavior of children with a high score on the Peg Tapping Task was

not affected by whether the text and illustration were presented to them simultaneously or one by one.

In contrast, the looking behavior of children with a low score on the Peg Tapping Task was affected by

the way of presenting text and illustrations.

Table 2

Effect of text-relevancy of the parts of the illustrations on total fixation duration per session and

condition for the Low Peg Tapping Group

(N = 11)

Text-relevant

Text-irrelevant

Session and

condition

M (SE)

M (SE)

F

p

η

Session 1,

simultaneous

8.61 (1.19)

4.79 (.22)

9.84

.011*

.50

Session 1,

non-simultaneous

5.99 (.78)

5.68 (.60)

.11

.746

-

Session 2,

simultaneous

7.11 (1.39)

5.55 (.60)

.74

.410

-

Session 2,

non-simultaneous

4.91 (.77)

6.80 (.84)

5.39

.043*

.35

Note. Text-relevant and text-irrelevant refer to the total fixation durations at the parts of the

illustrations highlighted by the text and the text-irrelevant parts.

* p < .05.

shown in Table 2, but not for condition. The effects of relevancy implied that when illustration and

text were presented simultaneously, children with a low score on the Peg Tapping Task looked longer

at the parts of the illustration that were highlighted by the text than at the text-irrelevant parts.

Interestingly, this was only the case for the first reading and not for the second reading session, as was

illustrated by the interaction effect between relevancy and session (F (1, 10) = 8.56, p = .015, ηp² =

.46). On the other hand, when children with lower inhibitory control skills heard the text first and saw

the illustration only thereafter, they looked longer at the text-irrelevant than at the text-relevant parts

of the illustration, but this was true only during the second reading.

Eye-gaze Patterns and Vocabulary Learning

A mediation model was used in order to answer the second research question and thus to

determine whether the time children spent looking at the parts of the illustration that were highlighted

by the text mediated the relationship between their pre- and posttest vocabulary scores. This mediation

model was analyzed using Preacher and Hayes’ (2008) path analysis of indirect effect and

bootstrapping estimates, and was tested separately for the simultaneous and the non-simultaneous

conditions. The variables were entered in Hayes’ (2011) SPSS INDIRECT macro (Version 4.2) for

testing indirect effects in mediation models. Percentage of items correct on the vocabulary pretest was

entered as the independent variable, percentage of items correct on the vocabulary posttest as the

dependent variable, the mean total fixation duration at the text-relevant parts of the illustration of the

first and second reading as the mediator, and the Peg Tapping score as the covariate. The bootstrap

estimates which are presented hereafter are based on 1,000 samples and path values represent

unstandardized regression coefficients.

The results of the analyses of direct and indirect effects of looking behavior during

simultaneous storybook readings on word learning and the corresponding bootstrapping estimates are

shown in Figure 2. The model indicates that the time children spent looking at the parts of the

illustration highlighted by the text fully mediated the relationship between their pretest and posttest

vocabulary scores when the text and illustrations were presented simultaneously. That is because the

relationship between the pretest and the posttest vocabulary scores (c path = .63, p = .032) became

non-significant when the mediator – total fixation duration at the parts of the illustration highlighted

by the text – was included in the model (c’ path = .25, p = .373). The indirect effect through the

mediator (i.e. the difference between the total and the direct effect) was significant with a

point-estimate of .37 (p = .015) and a 95% bias-corrected bootstrap confidence interval of .0834 to 1.0327.

From the direction of the a and b paths it can be concluded that – when the Peg Tapping score

was included in the model – greater pretest vocabulary knowledge lead to longer looking times at the

text-relevant parts of the illustration; which in turn lead to more word learning shown by higher

posttest vocabulary scores. This mediation model was only applicable for the simultaneous condition,

in which the children heard the text and saw the illustration at the same time. The results of the

bootstrapping analyses for the non-simultaneous condition – in which children heard the text first and

saw the illustration thereafter – revealed a non-significant model including a non-significant indirect

effect with a point-estimate of -.0197 (p = .139) and a 95% bias-corrected bootstrap confidence

interval of -.3297 to .1175.

a) Direct pathway

Percentage of items

correct at the MR pretest

c path = .63 (p = .032*)

Percentage of items

correct at the MR posttest

b) Indirect or mediated pathway

Mediator

Total duration of fixations at

text-relevant parts of the

illustration

a path = 15.44

b path = .02

(p = .017*)

(p = .021*)

Percentage of items

correct at the MR pretest

c’ path = .25 (p = .373)

Percentage of items

correct at the MR posttest

Figure 2. Mediation model of the effect of children’s eye-movement patterns on vocabulary learning

for the simultaneous readings. Peg Tapping scores were included as a covariate. Path values represent

unstandardized regression coefficients.

* p < .05.

Vocabulary Learning

In order to answer the question whether children learned words during the two book reading

sessions it was tested whether children’s vocabulary knowledge improved from pretest to the posttest.

A 2 x 3 repeated-measures ANOVA was performed, with the percentage of items correct at the MR

task as the dependent variable, and measurement time (pretest and posttest) and condition

(simultaneous, non-simultaneous and control) as within-subject factors. The Peg Tapping scores were

not included as a covariate in this model because children’s Peg Tapping scores had no significant

effect on their vocabulary learning (F(1,19) = .12, p = .773).

Mauchly’s test indicated that the assumption of sphericity was met for the main effect of

condition (χ² (2) < .001, p = .861) and for the interaction effect between measurement time and

condition (χ² (2) = 2.35, p = .309). A Helmert contrast was chosen to compare vocabulary learning in

the two book readings conditions with vocabulary learning in the control condition to check the effect

of book reading. A Simple contrast was chosen to compare vocabulary learning in the simultaneous

and non-simultaneous readings to check the effect of mode of presentation.

The results revealed a marginally significant main effect of measurement time (F(1, 20) =

4.03, p = .058, ηp² = .17) in favor of the posttest. Children’s mean vocabulary score improved from

62% (SE = 1.70) of the items answered correctly at the pretest to 67% (SE = 2.50) of the items

answered correctly at the posttest. Non-significant results were revealed for the main effect of

condition (F(2, 19) = .18, p = .838) and for the interaction effect between measurement time and

condition (F(2, 19) = 1.79, p = .194).

Analyzing the effect of the book readings on vocabulary learning by applying a Helmert

contrast revealed a marginally significant interaction effect between measurement time and condition

(F(1) = 3.21, p = .088, η

² = .14), indicating that at the posttest children’s vocabulary scores improved

from 63% to 67% in the simultaneous condition and from 59% to 68% in the non-simultaneous

condition, whereas their scores only improved 1% in the control condition (from 64% at pretest to

65% at posttest). These results suggest a positive effect of book reading on vocabulary learning. In

contrast, analyzing the effect of mode of presentation – simultaneous versus non-simultaneous

readings – by applying a simple contrast revealed that the difference in vocabulary learning between

the simultaneous and the non-simultaneous readings was not significant (F(1) = .37, p = .550).

Discussion

The present study was designed to investigate whether Dutch kindergartners’ visual attention

within storybook illustrations differs when the oral text and the illustrations are presented

simultaneously or successively. In addition, it was examined whether their visual attention within the

illustrations is related to vocabulary learning from the stories. Furthermore, it was assessed whether

storybook readings and in particular simultaneous readings resulted in the acquisition of new words.

The results support the first hypothesis that the way of presenting storybooks – simultaneously

or non-simultaneously – affects kindergartners’ visual attention within the illustrations. During the

simultaneous but not during the non-simultaneous readings, children looked longer at the text-relevant

parts than at the text-irrelevant parts of the illustrations. This was especially true for children with

lower inhibitory skills. The second hypothesis, that children’s visual attention and vocabulary learning

are related, at least in the simultaneous condition, is fully supported by the results. The results of the

present study replicated the positive effect of storybook reading on children’s vocabulary learning

found in numerous studies before (e.g., Ewers & Brownson, 1999; Sénéchal & LeFevre, 2001;

Verhallen & Bus, 2010). However, no support was found for the third hypothesis that children learn

more new words from the simultaneous than from non-simultaneous storybook readings.

The first important finding of the present study is that, regardless of condition, children looked

longer at the parts of the illustrations that were highlighted by the text than at the text-irrelevant parts

of the illustrations. This result is in line with Sipe’s (1998) idea of integrating text and illustrations and

with Sadoski and Paivio’s (2004) dual coding theory. It suggests that children follow the oral text

when visually inspecting the illustrations during storybook readings, as was also found in previous

eye-tracking studies (Evans & Saint-Aubin, 2005; Verhallen & Bus, 2011). The finding that children

appear to follow the oral text and thus integrate text and illustrations during storybook readings

justifies the hypothesis that the mode of presenting text and illustrations – simultaneously or

non-simultaneously – affects children’s visual attention within the illustrations.

Indeed, as expected, mode of presentation resulted in differences in children’s looking times

for text-relevant and text-irrelevant parts of the illustrations. When the oral text and the accompanying

illustrations were presented simultaneously, children looked longer at the parts of the illustrations that

were highlighted by the text than at the text-irrelevant parts. In contrast, such a difference in looking

times was not found for the situation in which the oral text was presented first and the accompanying

illustration was only shown thereafter. During the non-simultaneous readings the children looked as

long at the text-relevant as at the text-irrelevant parts of the illustrations. These results are in

accordance with the principle of temporal contiguity (Mayer, 1999); it is easier to integrate words and

illustrations when both types of information are presented at the same time as opposed to successively.

The mode of presenting text and illustrations during storybook readings especially appeared to

affect the visual attention patterns of children with lower inhibitory control skills. When the text and

the accompanying illustrations were presented simultaneously to these children, they looked longer at

the text-relevant than and the text-irrelevant parts of the illustrations, but only during the first reading

session, and not during the second reading session. Interestingly, the opposite pattern was found for

their non-simultaneous readings. During the first non-simultaneous reading children with lower

inhibitory control skills looked as long at the text-relevant as at the text-irrelevant parts, whereas they

looked longer at the text-irrelevant parts than at the text-relevant parts during the second

non-simultaneous reading. An explanation might be that the children with lower inhibitory control skills

did not listen to the narration as well the second time. Because they had heard the story and seen the

illustrations before, the impulse to look at the distracters might have been stronger and more difficult

to resist than the first time the storybook was presented to them.

Similar results were not found when solely analyzing the eye-tracking data from children with

higher inhibitory skills. For these children mode of presentation did not result in explicit differences in

looking times for text-relevant and text-irrelevant parts of the illustrations for either of the conditions.

The more obvious effects of mode of presentation of text and illustrations on the visual attention of

children with lower inhibitory control skills as compared to children with higher inhibitory control

skills might be explained in terms of profiting from literacy input (Davidse et al., 2011; Kegel et al.,

that children’s cognitive control skills such as inhibitory control skills might not play a role in

profiting from literacy input because of the highly structured nature of parent-child storybook reading

activities. However, the nature of parent-child storybook readings differs from shared storybook

readings in the classroom. That is, during parent-child readings children often sit on their parent’s lap

or beside the parent and they see the illustrations while the parent is reading the story, whereas the text

and the accompanying illustrations are presented successively in kindergarten classrooms.

Non-simultaneous readings might therefore be perceived as less structured than Non-simultaneous readings,

which can clarify why inhibitory control skills appear to play a role in the storybook readings in the

present study.

As it is more difficult to establish mental connections between words and illustrations when

corresponding words and visual elements are not in working memory at the same time (Mayer &

Moreno, 2002), it is plausible that successively presenting the text and the illustrations appeals more to

children’s inhibitory control skills. When text and illustrations are presented successively, children can

not directly integrate the two sources of information (Sadoski & Paivio, 2004) and as a result the

guidance of the spoken text may be less efficient in directing children’s visual attention to elements

which are highlighted by the text. For that reason, selecting the text-relevant visual details and

resisting the visual distracters will be a more demanding task than when text and illustrations are

presented together. Since children with higher inhibitory control skills are less distracted by irrelevant

information and process and retain information more efficiently (Blair & Razza, 2007; Ray & Smith,

2010), simultaneous presentation of text and illustrations might be less important for them than it is for

children with lower inhibitory skills. This assumption is in accordance with the results of McClelland

and colleagues (2007) who found support for a positive relationship between inhibitory control skills

and emergent literacy and vocabulary skills.

Eye-gaze Patterns and Vocabulary Learning

As expected, with regard to Mayer’s (2003) multimedia learning theory and the principle of

temporal contiguity (Mayer, 1999), a relationship between children’s visual attention and their

vocabulary learning was found when the oral text and the illustrations were presented at the same time.

During the simultaneous readings, children’s vocabulary learning was fully mediated by their visual

attention for the text-relevant elements of the illustrations. That is, children who looked longer at the

elements of the illustrations that were highlighted by the spoken text, ended up with higher posttest

vocabulary scores. This result indicates that if the visual referent of the spoken word is located the

chance of learning that particular word increases (Acha, 2009). A similar relationship between visual

attention and vocabulary learning was not present for the non-simultaneous readings, providing

support for the power of the simultaneously presented spoken words and visual referents in terms of

vocabulary learning (Mayer, 2003; Sadoski, 2005).

During the present study children’s knowledge of the target words increased. The effect of the

book readings – the simultaneous and the non-simultaneous readings together – on vocabulary

learning was marginally significant, though its effect size (ηp² = .14) represented a medium to large

effect. The magnitude of the effect of the book readings on vocabulary learning accentuates the

practical importance of the effect. Contrary to the expectations – and despite the positive relationship

between visual attention and vocabulary learning as was only found for the simultaneous readings in

the present study – no significant differences in vocabulary gains between the simultaneous and the

non-simultaneous storybook readings were found. This result implies that presenting the text and the

accompanying illustrations simultaneously or non-simultaneously to the children in the current sample

did not make a difference in terms of their vocabulary learning.

The absence of differences in vocabulary learning between the two types of readings can be

explained by children’s general language skills. As was proposed by Mol and colleagues (2008) the

quality of book readings might be more influential for children with weaker language skills than for

children with stronger language skills. The mode of presentation – simultaneous or non-simultaneous

presentation of text and illustration – can be interpreted as an element of book reading quality. The

children in the current sample all have average to good general language skills and therefore looking at

the illustration either while listening to the spoken text or after the spoken text might not make a

difference for these children in terms of vocabulary learning.

A related explanation and an additional argument for this line of reasoning is that, for the

children in the current study, the verbal information as presented by the oral text might have been

sufficient to understand the story (Plass et al., 1998), as they had substantial knowledge of part of the

target words. The children could already answer around 60% of the vocabulary items correctly at the

pretest, which indicates that part of the target words’ meanings were already known before the

storybook reading intervention was implemented. Meaningful learning requires paying attention to the

presented information, organizing the incoming information into a consistent whole, and integrating it

with prior knowledge (Mayer & Moreno, 2003). In case of storybook readings children are provided

with verbal information (words) and visual information (illustrations), which should be integrated in

order to achieve story comprehension and vocabulary learning (Sipe, 1998). The illustrations in

storybooks provide visual definitions of words (Kim & Gilman, 2008) and the combination of verbal

and visual information strengthens children’s memory traces for word meanings by establishing

mental connections between the spoken words and their visual referents (Mayer, 1999; Sadoski &

Paivio, 2004), which increases the chance that particular words will be learned. Children who are more

experienced with the target words and who have more knowledge about their meanings might already

be able to establish mental connections between the target words and their prior knowledge about

these words (Mayer & Andersson, 1992), making the visual definitions of the words as displayed by

the illustrations redundant.

An alternative explanation for the absence of the difference in vocabulary learning between

the simultaneous and the non-simultaneous readings can be found in the types of target words used in

the current study. The words used as target words were nouns, verbs, adverbs, and adjectives. Nouns

usually refer to concrete objects (e.g., director, fire), whereas verbs – and adverbs and adjectives – are

often more abstract (e.g. to mind, jealous) as they refer to actions, events, or activities (Ard & Beverly,

2004). As such verbs, adverbs, and adjectives are often less visible and therefore they are more

difficult to display in static illustrations (Ard & Beverly, 2004; Verhallen & Bus, 2006). Therefore, it

seems plausible that it is easier to establish mental connections between nouns and their visual

referents than between verbs, adverbs, adjectives and their visual referents. This implies that for nouns

the simultaneous condition might be more beneficial than for verbs, adverbs, and adjectives. For that

reason, a difference in vocabulary learning between simultaneous and non-simultaneous readings

might have been found if all target words in the current study would have been nouns.

Limitations and Future Directions

The main limitation of the present study is that the storybook readings do not fully mirror

storybook readings as they occur in kindergarten classrooms. In the first place, teacher-child

interaction was not included in the readings. Children only listened to the story in the experiment,

while, in contrast, reading activities in schools are usually interactive. Kindergarten teachers

encourage children’s active participation during storybook readings (Sipe, 2002). Preceding, during,

and after storybook readings teachers ask children questions about the book cover, the title, the

illustrations, and the theme of the storybook. Furthermore, teachers stimulate children to predict what

will happen in the story, to ask questions about the story, and to relate the story to children’s

experiences. Moreover, 90% of the teachers provide children with explanations as they encounter

unfamiliar words (Biemiller & Boote, 2006; Ghonem-Woets, 2009), which is important because

explaining the meanings of new words during shared reading accelerates children’s vocabulary

learning (Justice et al., 2005). Even brief and direct explanations can enhance children’s vocabulary

gains by offering opportunities to link words with their meanings (Biemiller & Boote, 2006). Active

participation, labeling unknown words, and providing word elaborations focus children’s attention on

new words and enable them to practice retrieving new words, resulting in stronger memory traces for

these words, which in turn facilitates learning (Sénéchal et al., 1995; Ewers & Brownson, 1999).

Moreover, during the intervention readings children were being read aloud to individually,

whereas storybook readings in the classroom are typically large group activities (Ghonem-Woets,

2009). The study of Morrow and Smith (1990) revealed that being read aloud to in small groups of

three children resulted in greater story comprehension than either one-to-one readings or whole-class

readings, due to social interaction during shared reading. This might be true for vocabulary learning as

well, as children’s vocabulary learning can be enhanced by active participation and interaction (Horst

et al., 2011; Sénéchal & LeFevre, 2001). Although including social teacher-child and child-child

interactions seem to be a beneficial element of learning from storybooks, whole-class readings also

imply more sources of distraction, which makes it more difficult for children to pay attention to the

story and the illustrations (Morrow & Smith, 1990). For that reason, presenting text and illustrations at

the same time (temporal contiguity; Mayer, 1999) might be even more important – in terms of

profiting from storybook exposure – in group reading settings than in individual reading settings.

Other limitations of the present study include the homogeneity of the sample with regard to

children’s general receptive vocabulary level and including eye-tracking data for only one illustration

per storybook. In addition, the eye-tracking data only allows drawing conclusions about how long

children look at particular elements of the illustrations, not about the order of their eye-gaze patterns,

implying that it is not known how closely the children followed the text. Furthermore, children might

have learned more words if the stories were repeated more than once, as reading the same story twice

is considered as a minimum for word learning (e.g., Biemiller & Boote, 2006; Robbins & Ehri, 1994).

Despite the limitations the present study also has its strengths. Its major strengths is the use of

the eye-tracking method to investigate how kindergartners use illustrations during storybook readings,

which is an online measure of their visual attention (Henderson, 2003). Other strengths include: using

a within-subject design instead of a between-subject design, using commercially available storybooks,

including several storybooks written by different authors and illustrated by different illustrators and

taking into account child characteristics (a priori general vocabulary and inhibitory control skills).

Future research should reveal whether the results of the present study will be replicated in a

larger sample and with more difficult storybooks, as the storybooks used in the intervention might not

have been challenging enough for the children in the present study. It is also interesting to investigate

whether the same results will be obtained for kindergartners with lower vocabulary levels. Other

recommendations for future research are: including teacher-child interaction and word meaning

elaborations, repeating each story more than once; and considering whole-class instead of individual

readings. Moreover, further investigation on the role of number of target words known at pretest and

the role of inhibition and attention in learning vocabulary through storybook readings is warranted.

Practical Implications

If it is indeed true that children integrate the text and the illustrations into a coherent whole to

construct meaning while being read aloud to, it will be important to make storybook illustrations

visible to all children during shared storybook reading activities in the classroom. With the

introduction of whiteboards and smart boards in primary schools (Smith, Higgins, Wall, & Miller,

2005) it has become possible to show the storybook illustrations on the board during shared reading

activities (Gill & Islam, 2011), such that text and illustrations are presented simultaneously to the

children. Listening to the story and seeing the accompanying illustrations at the same time, will

facilitate the integration of verbal (text) and visual information (illustrations), which in turn might

eventually stimulate children’s vocabulary learning from storybook readings.

inhibitory control skills. Especially for children with lower inhibitory control skills, their visual

attention differs for readings in which text and illustration are presented simultaneously or

successively, whereas mode of presentation does not influence the visual attention of children with

higher inhibitory control skills to the same extent. As inhibitory control skills are related to attention

skills and ignoring distracters (Blair, 2002) and as focusing attention during whole-class reading aloud

activities is more demanding than during individual readings (Morrow & Smith, 1990) kindergarten

teachers are highly recommended to invest in and to concentrate on children’s self-regulated behavior

and their engagement during storybook readings (Kegel & Bus, 2012).

For children that are easily distracted due to lower inhibitory control skills, presenting the

storybook illustrations on the whiteboard or smart board screen during whole-class storybook readings

might be particularly helpful in directing and focusing their attention on the storybook illustrations.

Moreover, it seems likely that strategies to enhance children’s vocabulary learning from shared

storybook reading – such as encouraging active participation, labeling the visual referents of unknown

words, and explaining difficult words and elaborating on their meanings (e.g., Biemiller & Boote,

2006; Ghonem-Woets, 2009; Sénéchal et al., 1995) – can be put into practice more easily, when the

illustrations are presented in a larger format and when all children can look at the illustration together.