Contingent Responsivity in E-Books Modeled From Quality Adult-Child Interactions: Effects on Children’s Learning and Attention

Abstract

Experiences of contingent responsivity during shared book reading predict better learning outcomes. However, it is unclear whether contingent responsivity from a digital book could provide similar support for children. The effects on story recall and engagement interacting with a digital book that responded contingently on children’s vocalizations (contingent book) were investigated, with a focus on the role of individual differences in attention. The study used a within-subject design with 3 experiments from 90 3- to 5-year-old children. Children were presented with a contingent book and 3 noncontingent control conditions: a board book (Experiment 1), a static digital book (Experiment 2), and an animated book (Experiment 3). The use of the contingent book significantly increased children’s story recall and was found to be especially useful for children with less developed attention regulation. (PsycINFO Database Record (c) 2019 APA, all rights reserved)

Full text

1
The official citation that should be used for this material is:
Eng, C. M., Tomasic, A. S., & Thiessen, E. D. (2019). Contingent responsivity in E-books
modeled from quality adult-child interactions: Effects on children’s learning and attention
Developmental Psychology (in press). https://doi.org/10.1037/dev0000869
© 2019, American Psychological Association. This paper is not the copy of record and may not
exactly replicate the final, authoritative version of the article, published in Developmental
Psychology. The final article will be available, upon publication, via its DOI:
10.1037/dev0000869
Contingent Responsivity in E-Books Modeled from Quality Adult-Child Interactions:
Effects on Children’s Learning and Attention
Cassondra M. Eng
Anthony S. Tomasic
Erik D. Thiessen
Carnegie Mellon University
Author Note
Cassondra Eng, Department of Psychology; Anthony Tomasic (tomasic@cs.cmu.edu),
Language Technologies Institute, Erik Thiessen (thiessen@andrew.cmu.edu), Department of
Psychology. Correspondence concerning this article should be addressed to Cassondra Eng,
Department of Psychology, Carnegie Mellon University, 346 Baker Hall, 5000 Forbes Avenue,
Pittsburgh, PA 15213. cassonde@andrew.cmu.edu. Cassondra Eng declares that she has no
conflict of interest. Anthony Tomasic has equity ownership in View Update Media, LLC, which
is developing products related to the research being reported. Erik Thiessen declares that he has
no conflict of interest.

CONTINGENT RESPONSIVITY IN E-BOOKS 2
Acknowledgements
This work was supported in part by the Institute of Education Sciences, U.S. Department
of Education, through grant R305B150008 to Carnegie Mellon University. The opinions
expressed are those of the authors and do not represent the views of the Institute or the U.S.
Department of Education. We thank Kristen Boyle, Tori Iatarola, Amru Palaniyappan, Ermina
Lee, Hyunji Do, and Clara Lee for assistance in data collection, Xavier Artache and Marie Shaw
for coding the attention task in MATLAB, Thacher Hurd for permission to use Zoom City and
Cat’s Pajamas, and Phapimol Yoovidhya and Marisa Lu for programming the contingent books.
We also thank Catarina Vales for helpful feedback on a previous version of this paper and the
children and educators at the Carnegie Mellon University Children’s School who made this
project possible.
Data Transparency and Openness
The data collected in this study and the scripts used to analyze these data are available in
KiltHub–Carnegie Mellon’s comprehensive institutional repository, which is part of the Figshare
repository platform—at the following link: https://doi.org/10.1184/R1/7740170.v1.

CONTINGENT RESPONSIVITY IN E-BOOKS 3
Abstract
Experiences of contingent responsivity during shared book reading predict better learning
outcomes. However, it is unclear whether contingent responsivity from a digital book could
provide similar support for children. The effects on story recall and engagement interacting with
a digital book that responded contingently on children’s vocalizations (contingent book) were
investigated, with a focus on the role of individual differences in attention. The study used a
within-subject design with 3 experiments from ninety 3- to 5-year-old children. Children were
presented with a contingent book and three noncontingent control conditions: a board book
(Experiment 1), a static digital book (Experiment 2), and an animated book (Experiment 3). The
use of the contingent book significantly increased children’s story recall, and was also found to
be especially useful for children with less developed attention regulation.
Keywords: attention; individual differences; educational technology; preschool-age children;
contingent interactions
Supplemental materials: http://dx.doi.org/10.1037/dev0000869.supp

CONTINGENT RESPONSIVITY IN E-BOOKS 4
Contingent Responsivity in E-books Modeled from Quality Adult-Child Interactions: Effects on
Children’s Learning and Attention
Introduction
A critical feature of effective communication between an adult and child is the presence
of contingent responsivity; that is, that the feedback that the child receives is dependent upon
their behavior (Bornstein, Tamis-LeMonda, Hahn, & Haynes, 2008). In the context of spoken
language, a contingent response is when an adult responds reliably, promptly, and accurately to a
child’s vocalizations (Roseberry, Hirsh-Pasek, & Golinkoff, 2014). Contingent responsivity
enables children to feel in control, maintains their focus, permits self-pacing, and encourages
children to continue the interaction when each of their vocalizations is met with an appropriate
response (Hirsh-Pasek et al., 2015). This interaction is in contrast to adult-directed interactions of
one-sided control in which the parent leads and the child passively complies, or resists and
disengages. Contingent caregiver responsivity extends children’s capacities, and is positively
associated with emerging cognitive and language competence (Tamis‐LeMonda, Bornstein, &
Baumwell, 2001).
Shared book reading is a common activity that promotes contingent responsivity.
Experiences of contingent responsivity during shared book reading predict better reading and
language outcomes (Dickinson, Griffith, Golinkoff, & Hirsh-Pasek, 2012). The role of
contingency is especially important during shared book reading because children are given the
opportunity for conversational turns that lay the groundwork for literacy skills. Adults facilitate
children’s learning during shared book reading when they add more information, ask follow-up
questions, praise, or point to relevant story content such as pictures that match the pronounced
words (Hargrave & Sénéchal, 2000). Adults’ contingent responsivity through pauses and

CONTINGENT RESPONSIVITY IN E-BOOKS 5
prompts direct children’s attention to relevant story content that engage and help children
interpret and understand the story narrative (Strouse, O'Doherty, & Troseth, 2013). When
children vocalize and attempt to read along, adults facilitate children’s understanding of the story
with reinforced responsiveness by pointing to relevant story content such as pictures that match
the pronounced words (Flack, Field, & Horst, 2018). These nonverbal cues or gestures can serve
as forms of contingent support, such as adults pointing instead of using verbal cues. Contingent
responsivity encourages the child’s communicative efforts and facilitates children's ability to
learn features, such as congruent pictures associated with the pronounced words, and enables
them to later reproduce an understanding about the words in relation to the story (Bus, 2001).
Prior studies have found that children learn significantly more when adults ask children
to repeat target words from the text through questions and comments (Blewitt, Rump, Shealy, &
Cook, 2009; Sénéchal, 1997). Echo reading—a common pedagogical practice at school and at
home—is a shared reading book strategy that occurs when children repeat a phrase or a sentence
immediately after it is read to them (Stahl & Heubach, 2005). The repeated reading component is
intended to provide practice so that children develop fluent and automatic reading (Jennings,
Caldwell, and Lerner, 2013). However, it is unknown whether children receiving a nonverbal
contingent response upon repeating the text in an echo reading paradigm would have similar
benefits as when adults contingently respond to children through questions and comments.
Children’s earliest experiences with books are no longer limited to paper, and books are
now accessible in the form of electronic books (e-books) through computers, smartphones, and
tablets (Rideout, 2017). Many e-books have interactive features such as embedded animations,
games, and sound effects that are activated by touching a spot on the screen (i.e., hotspots;
Piotrowski & Krcmar, 2017). However, these kinds of interactive features have often been found

CONTINGENT RESPONSIVITY IN E-BOOKS 6
to negatively affect learning in young children, in stark contrast to the contingent responsivity
provided by adult co-readers (Bus, Takacs, & Kegel, 2015; Parish-Morris et al., 2013). While the
contingent responsivity adults provide synchronizes with children’s story-related vocalizations,
interactive features common in digital books—puzzles, games, hotspots, erroneous visuals and
sound effects—may draw children’s attention away from the story-related elements relevant to
the narrative (De Jong & Bus, 2002).
There is evidence that specific types of contingent interactions using e-books promote
different emergent literacy skills. Children’s word reading and phonological awareness improved
when the e-book interaction included adult responses that focused children’s attention on the
sounds of words in the story, compared to reading printed books with contingent adult responses,
reading e-books alone, or receiving the regular kindergarten program (Segal-Drori, Korat,
Shamir, & Klein 2010). Children’s expressive vocabulary was enhanced when an e-book posed
extratextual vocabulary questions, compared to e-books with hotspots, e-books without
questions, or independently reading the e-book (Smeets & Bus, 2012). Tactile contingency such
as requiring children to touch a relevant image improved 2- to 3-year old children’s word
learning compared to touching anywhere on the screen or passive interaction with the screen
(Kirkorian, Choi, & Pempek, 2016). Conversely, it has been reported that children with low self-
control are prone to excessive tapping of touchscreens, which may lead to more arbitrary
tapping, frustration, and less learning (Troseth, Russo, & Strouse, 2016). One of the goals of this
study was to design a paradigm in which an e-book responds contingently to children’s
vocalizations, reducing the need for tactile responsivity from children.
Other studies show that when comparing language exposure through contingent
interactions with a person, children demonstrate quantifiably smaller amounts of learning from

CONTINGENT RESPONSIVITY IN E-BOOKS 7
exposure to the same material presented noncontingently through digital media (Dore et al.,
2018; Roseberry et al., 2014). Research on children’s verbal behaviors during shared book
reading have found that children’s repetition of parents’ vocalizations is uniquely and positively
related to children’s overall story retelling abilities (Kang, Kim, & Pan, 2009); yet, little is
known about whether children receiving a contingent response from an e-book upon repeating
the text would have similar benefits on children’s story recall. There is evidence that contingent
responsivity in adult-child interactions produce large gains in learning outcomes and contingent
responsivity in e-book interactions are capable of improving vocabulary knowledge and
phonemic awareness in young children. It is unknown whether contingent responsivity in a
digital book could provide similar support for children’s ability to recall, retell, and describe key
story events.
The current experiments examined whether a digital book that responds contingently to
child vocalizations (contingent book) would increase story recall and engagement compared to
three noncontingent control conditions. Children were read two commercially available stories
marketed for prekindergarten children matched in age-appropriate content, length, and
readability. In Experiment 1, we investigated the effects of a contingent book on children’s story
recall and engagement compared to a noncontingent board book control. Second, we examined
whether the contingent book might be especially useful for children with less developed attention
regulation. In Experiment 2, we replicated Experiment 1 with a noncontingent digital book so
children were presented with both stories on a digital platform to be certain that the effects found
in Experiment 1 were not due to the novel effect of technology. In Experiment 3, we replicated
Experiments 1 and 2 with a noncontingent animated book to ensure that the effects found in the
first two experiments were not solely driven by the illumination and movement from animations.

CONTINGENT RESPONSIVITY IN E-BOOKS 8
Participants’ engagement to each presentation was assessed by coding eye gaze duration towards
each book condition, and children were asked questions about each narrative to assess story
recall. Independent assessments of attention regulation and verbal ability were also administered.
Based on previous work on contingent responsivity in adult-child interactions increasing
children’s learning and engagement, we hypothesized that the contingent book would increase
children’s story recall and looking duration towards the book during the reading session
compared to the noncontingent books. Previous studies have found that contingent responsivity
supports children’s limited attention skills and facilitates learning by directing attention to
relevant content (Kirkorian, Choi, & Pempek, 2016; Nussenbaum & Amso, 2016). Therefore, we
hypothesized that the addition of story-related features that activate contingently on children’s
vocalizations would be especially beneficial for children who are easily distracted and score
lower on the attention task. That is, because of these children’s limited ability to focus on
relevant material while suppressing extraneous details, they are the ones who might benefit the
most from the guidance of self-paced story features that match their vocalizations.
Experiment 1
Method
Participants All participants were recruited from the same pre-primary school on the campus of
a private university in a Mid-Atlantic city in the United States. The school environment
represents local racial and economic diversity with children being 74% White, 10% Asian or
Pacific Islander, 7% Black or African American, 6% Middle-Eastern, and 3% Hispanic, with
only 15% from university-affiliated households, and 33% of children attending on a partial or
full scholarship. We based our target sample size on prior published work assessing the effect of
e-books on child learning. The sample sizes in each experiment are comparable to prior studies

CONTINGENT RESPONSIVITY IN E-BOOKS 9
which examined the effects of e-books on learning in preschool-aged children (Moody, Justice,
& Cabell, 2010; Smeets & Bus, 2012; Strouse, O'Doherty, & Troseth, 2013). Experiment 1 used
a within-subject design with data from 35 children (16 males, 19 females) ages 3 to 5-years-old
(M = 55.54 months, SD = 9.64 months). An additional child was tested but excluded due to
equipment failure. The experimental protocol was approved by the Carnegie Mellon University
Institutional Review Board (Study title: Learning and Development from Infancy to Adulthood,
IRBSTUDY2015_00000471). Signed consent was obtained from the parents of participants.
Children were tested individually by hypothesis-blind research assistants and given stickers for
participation.
Materials and Procedure Full descriptions of the materials, instructions, and procedure are
provided in supplemental materials. Here we provide summaries of each.
Book Conditions To maintain a high level of ecological validity, children were read two
commercially available stories marketed for emergent readers written and illustrated by the same
author (Thacher Hurd), matched in artistic style, page length, and readability: Cat’s Pajamas and
Zoom City. Children were read one of the stories in the presentation of a noncontingent board
book control condition, and the other story in the presentation of the experimental contingent
book condition. Condition and story order were counterbalanced. With permission from the
author, both stories were converted into digital versions with the addition of contingent
responsivity. The contingent responsivity was story-related animations that activated
contingently on the child’s vocalizations. When children said a word from the story aloud, the
contingent book responded with story-related animations (e.g., child says, “car” and a picture of
a car animates by popping off the page). The contingent book was presented to children on an
Apple iPad (9.4 in x 6.6 in). When a word from the story was vocalized by the child, a congruent

CONTINGENT RESPONSIVITY IN E-BOOKS 10
picture that represented the word grew in size with a short (500 millisecond) animation, and then
shrank back to its original size. For example, when children vocalized the text from Zoom City
“fix the headlight,” the animation of a wrench turning generated when the word “fix,” was
vocalized, and the animation of the car’s headlight flashing on and off generated when the word
“headlight” was vocalized (see Figure 1, for static image of book page). Animations represented
the meaning of verbs and nouns in the text (see Table S1 and Table S2, in the online
supplemental material, for vocalized words that produced animations in each story).
Figure 1. Adapted with permission from “Zoom City,” by T. Hurd, p. 7-8. Copyright 1998 by
Thacher Hurd.
The contingent book responded only to the child’s vocalizations, not the reader’s. To make this
feature possible, the digital versions of Cat’s Pajamas and Zoom City were converted into Apps
using Framer Software, a design tool engineered for interaction work. Framer Software permitted
the iPad to partner with another device so that the visual display on the iPad was also mirrored
on a MacBook Pro 15-inch laptop. When the reader-child dyad swiped to the next page on the
iPad, this was also displayed on the laptop. As the reader in the reader-child dyad read the
contingent book to the participant on the iPad, a trained experimenter in an operations control
room activated the content-related animations contingently on the child’s vocalizations using the
laptop. If a child vocalized the word from the story “car,” the experimenter using the laptop
activated the animation of the car. The testing room with the reader-child dyad was connected to
the operations control room, but the rooms were separated with a curtain (see Figure S1 in the

CONTINGENT RESPONSIVITY IN E-BOOKS 11
online supplemental material for layout of how the experimenter activated animations
contingently on the child’s vocalizations). This setup made the pronounced vocalizations of the
child audible to the experimenter, which signaled when to generate the animations.
The reader in the reader-child dyads was a trained hypothesis-blind research assistant
who was instructed to read the books aloud to participants. For both book conditions, the reader
was the same and started each session by reading the first line of the story, and then said “Now
it’s your turn to read!” to the child and then paused. This prompt was stated for the first page of
each story and children responded by repeating the first line segment. This prompt encouraged
children to vocalize similar to how adult readers in dialogic shared reading interactions pause
and encourage children to repeat the words in the story during the shared reading sessions
(Strouse, O'Doherty, & Troseth, 2013). For both conditions, the reader followed the protocol of
pausing after each line in the story for five seconds, until the child vocalized the words to model
prior research paradigms finding a positive association between the duration of time adults pause
for children to respond and the information children retain from the story (Read, Macauley &
Furay, 2014; see Table 1, for total pages, lines, and animations, by story). The reader continued
reading each line in the book like traditional shared book reading and if the child did not practice
reading the line after 5 seconds, the reader moved on and read the next line. Because the amount
of feedback is highly variable from adults during shared book reading and the quantity and
quality of feedback influences children’s learning outcomes, these instructions were
implemented to ensure the effects of the contingent book and noncontingent control book could
be examined with minimal influence from extra-textual talk from the reader. On average, each
story had approximately three words per line because short segments suited for beginning
readers have been found to be optimal length in echo reading interactions (Jennings, Caldwell,

CONTINGENT RESPONSIVITY IN E-BOOKS 12
and Lerner, 2013). In both reading conditions, the reader listened attentively to children’s
vocalizations and as in traditional shared book reading if children made any extratextual queries
or comments—if any—were answered with the prompt: “That’s interesting, what do you think?”
so children’s extratextual comments and queries were neither encouraged or discouraged
(Sénéchal, 1997). Vocalizations were coded as the percentage of words the participant said aloud
out of the total words in the story. Across all 3 experiments, there were no significant differences
in vocalizations and there was a ceiling effect: children were inclined to vocalize all of the words
in the stories after the reader across conditions. Therefore, vocalization data were not included in
subsequent analyses (see Table 1, for mean vocalizations, by story; see the online supplemental
material for vocalization coding and reader protocol).
Table 1
Pages, lines, animations, and vocalizations, by story
Zoom City
Cat’s Pajamas
Total Pages
14
14
Total Lines
29
31
Total Animations
32
30
Mean Vocalizations (SD)
96.74% (5.67%)
96.80% (6.16%)
Procedure All sessions took place in the same testing room that permitted detailed audio and
video recordings. Reader-child dyads were taped with two digital cameras: a Logitech C920 HD
Pro Webcam and a Panasonic HDC-HS80 Camcorder. Each session was recorded using a Talent
USB-1 Studio Condenser Microphone to obtain high-quality recordings of children’s responses
to the recall questions. Audacity 2.1.3 software was used to record and create an audio file
(sample rate 44100 Hz, sample format 16-bit, bit rate 96 kbps) of each session. Once produced,
the audio file was exported to MP3 format for analysis. Engagement and responses to the recall
questions were coded offline based on video and audio recordings of the testing sessions.

CONTINGENT RESPONSIVITY IN E-BOOKS 13
Measures
Story Recall Measure Story recall is considered one of the most appropriate assessments for
children, and is assessed through narrative reconstruction of retelling events that play a central
role in the structure of stories (Kendeou et al., 2009). Narrative reconstruction consists of
children recalling the characters, settings, character goals, and solutions from the stories
(Gibbons et al., 1986). At the end of each book condition, children were asked questions that
probed their memory for details about the story that fit the narrative reconstruction criteria,
which were pilot tested on twenty 3 and 5-year-olds who were read both stories in the lab.
Questions were adjusted so assessments were equally challenging between stories, with neither
presenting floor or ceiling effects. The final assessment included 10 questions, scored out of a
total of 14 points, for each story about the setting, plot, theme, resolution, and character
descriptions, goals, and actions (see the online supplemental material for recall assessments, by
story). There were seven 1-point questions, two 2-point questions and one 3-point question. For
example, in Zoom City the main character’s actions were fixing the bumper, headlight, and
engine on a car. For the 3-point question, children were asked to recall which parts on the car the
character fixed. Children could receive full credit if in their response they identified the 3 car
parts that were fixed, 2 points if they identified 2 parts, 1 point if they identified 1 part, and 0
points if they failed to recall the parts that were fixed or provided an incorrect response.
Similarly for scoring, in Cat’s Pajamas the main character’s actions were making music using
drums, cans, and a horn. For the 3-point question, children were asked to recall which
instruments the character played. Children could receive full credit if in their response they
identified the 3 instruments that were played, 2 points if they identified 2 instruments, 1 point if
they identified 1 instrument, and 0 points if they failed to recall the instruments that were played

CONTINGENT RESPONSIVITY IN E-BOOKS 14
or provided an incorrect response. Story recall was measured as the percentage of correct
responses (out of 14 possible points). Hypothesis- and condition-blind research assistants who
received extensive training on using the audio recordings of each session listened to session
recordings and coded story recall performance. Inter-rater reliability using Cohen's kappa
(Cohen, 1960) was .86, indicating substantial coder consistency.
Engagement Measure Time on task (i.e., attending to the book while being read to) was
measured via gaze fixation duration, which is a common measure of engagement in a variety of
settings and is a particularly appropriate measure in the context of reading. Hypothesis-blind
research assistants reviewed the video recordings of the testing sessions to calculate the child’s
fixation duration to each book condition from the direction of the participant’s gaze. A “look” to
the contingent book or control book was coded each time the child's gaze was directed at the
book presentation. When the participant’s gaze shifted (i.e., to the reader or off-task), a look to
the new direction was coded. Each eye shift was judged as either towards or away from the book,
and the duration of the resulting looks was analyzed to calculate total looking time. Total reading
time was calculated as the time period from the moment the first word of the book (the title) was
read aloud by the reader and continued until the book was finished. Engagement was measured
as the percentage of time children spent looking at the book condition out of total reading time.
For eye gaze durations towards each book condition, inter-rater reliability (Cohen's kappa = .93)
was established for at least 20% of the entire sample.
Attention Measure Between the reading sessions, children participated in a modified attention
subtest from the Developmental NEuroPSYchological Assessments (NEPSY; Cuevas & Bell,
2014; Korkman, Kirk, & Kemp, 1998). The attention subtest is a visual cancellation task in
which participants are asked to maintain selective attention and focus on targets with speed and

CONTINGENT RESPONSIVITY IN E-BOOKS 15
accuracy. Children actively scanned a visual environment and pointed only to items that matched
that target stimuli (i.e., bowling pins) on a page containing both distractors and targets as quickly
as possible in 180 seconds. Performance on the task was calculated using the total number of
attention task errors and the total amount of search time to complete the task. Accuracy
(distractor hits) and speed (search time) from the attention task were standardized using Z-scores
and averaged together to create the composite variable: Distractibility. This composite variable
measured children’s ability to complete the task accurately without getting distracted, and
fluently with speed.
Verbal Ability Measure The Peabody Picture Vocabulary Test (PPVT-IV; Dunn & Dunn,
2007) was administered to children at subsequent laboratory visits within 3 weeks of the initial
lab visit. A trained researcher presented a series of pictures to the child (four pictures per page),
and verbally said a word that matched one of the pictures. Children were asked to point to the
picture that the word described. The PPVT is a nationally standardized instrument, and the
measure of interest was participants’ age-based standardized scores.
Data Analytic Approach To investigate our primary hypothesis—that contingent interactive
features would improve story recall—we assessed how well children could answer questions
related to the content of the story they had heard, administered after each story was finished.
Story recall was measured as the percentage of correct responses out of 14 possible points (see
Table 2, for raw and scaled scores). To assess possible order effects and sex differences, we
conducted a mixed factorial analysis of variance (ANOVA) on story recall, factoring order and
sex as between-subjects variables and book condition as the within-subject variable.
Next, we investigated whether the use of the contingent book might be especially useful
for children with less developed attention regulation. For this analysis, a recall difference score

CONTINGENT RESPONSIVITY IN E-BOOKS 16
for each child was calculated by subtracting the noncontingent control book recall score from the
contingent book recall score. Difference scores estimated changes in story recall performance
from using the contingent book, such that higher and positive scores indexed greater gains in
story recall. To examine the association between changes in recall from the contingent condition
and attention regulation, attention task distractors and time to complete the attention task were
standardized using Z-scores and averaged together to create the composite variable:
Distractibility. Prior research has found that children’s attentional control in preschool is related
to later reading skills (Conners, 2009; Franceschini et al., 2012). To control for the potential role
of verbal ability in the association between attention and recall difference scores, participants
returned to the laboratory within 3 weeks of the initial lab visit and were administered the PPVT
to ensure that findings would not be entirely due to variance shared with verbal ability (see
Tables S3 and S4 in the online supplemental material for correlation coefficients between
measures of attention, recall, engagement and verbal ability). To examine the extent to which
Distractibility uniquely predicted how much children’s story recall changed from the contingent
condition, a multiple regression analysis was conducted that included Distractibility and verbal
ability as predictors of children’s recall difference scores, and age in months as a covariate. Post-
hoc power analyses using G*Power (Faul, Erdfelder, Buchner, & Lang, 2009) indicated that the
samples in each experiment were adequate for the mixed factorial ANOVAs (Power .94 - .99)
and multiple regression analyses (Power .95 - .99) to detect the effect sizes observed.
Results
Story Recall There was a main effect of book condition; children’s recall scores were
significantly higher in the contingent book condition (M = 60.20%, SE = 3.13%) than the
noncontingent board book condition (M = 47.36%, SE = 2.94%), F(1, 30) = 39.57; p < .0005; ηp2

CONTINGENT RESPONSIVITY IN E-BOOKS 17
= .57. There was no main effect of order, F(3, 30) = 1.04, p = .39, or sex, F(1, 30) = .12, p = .74,
or significant interactions between any of these factors and story recall (all ps > .19). The
outcome for each condition of the recall measure followed a normal distribution and there were
no outliers. Follow-up pairwise comparisons after Bonferroni corrections revealed that on
average, children scored 12.65% (SE = 2.01%) higher on the recall assessment in contingent
book condition compared to the noncontingent book condition, 95% CI [8.55%, 16.76%], p <
.0005. All but six participants exhibited higher recall scores using the contingent book compared
to using the noncontingent board book, two of which exhibited identical scores across
conditions. Taken together, these results indicate that children’s mean story recall scores after
being read to from the contingent book were higher compared to being read to from the
noncontingent board book, regardless of the story or order in which the books were presented
(see Figure 2, for paired box plot).
Figure 2. Paired box plot of recall scores in the noncontingent book and contingent book
conditions. Data points were jittered in R by .02 to prevent overplotting (Team, 2017).
Engagement The percentage of time children spent looking at the book condition out of total
reading time did not significantly differ between the contingent book condition and the
noncontingent board book condition (see Table 2, for mean reading time and mean looking time

CONTINGENT RESPONSIVITY IN E-BOOKS 18
towards book conditions). The results indicated a ceiling effect: children attended to the book
throughout the entire reading session for both conditions (see the online supplemental material
for the analyses on the effect of book condition on engagement).
Table 2
Recall and Engagement measures, by condition
Measure M (SD)
Noncontingent Board
Book
Contingent Book
Story Recall Raw Scores
6.63 (2.44)
8.43 (2.59)
Story Recall (% of 14)
47.35% (17.42%)
60.20% (18.52%)
Total Reading Time (ms)
141364 (37795)
142909 (34787)
Total Looking Time at Book (ms)
137303 (31290)
141182 (32214)
Engagement (Looking/Reading Time %)
97.89% (4.52%)
99.01% (2.87%)
The Role of Individual Differences in Attention Recall difference scores ranged from -14.29%
to 42.86%, with a mean of 12.86% (SD = 12.23%). Standardized z-scores of attention task
distractors (M = 2.83; SD = 3.82) and time to complete the attention task (M = 125.40 s; SD =
55.61 s) were combined to create the composite variable of Distractibility (M = -.22; SD = .71).
Twenty-six participants returned to the laboratory and were administered the PPVT (M=117.65,
SD =12.83). Higher Distractibility scores, r(35) = .57, 95% CI [4.82, 14.79], p < .0005 (see
Figure 3A), and lower verbal ability scores r(26) = -.42, 95% CI [-.85, -.04], p < .04 (see Figure
3B), were associated with how much children’s recall changed from the contingent condition.
Figure 3. Scatterplots of correlations between recall difference scores and performance on the
attention task and PPVT. (A) Greater Distractibility and (B) Lower verbal ability were associated
with higher recall difference scores. There were no outliers.

CONTINGENT RESPONSIVITY IN E-BOOKS 19
There was not a statistically significant interaction (Fchange = 1.36, df = 6, 19, p = .29) between
Distractibility, age, and verbal ability and their effects on recall difference scores. Therefore, the
final multiple regression analysis performed excluded the interaction terms as they were not
significant. The additive model (Fchange = 6.03, df = 3, 22, p = .004) revealed that Distractibility
accounted for unique variance in changes in recall using the contingent book (β = 9.76, t = 3.33,
p = .003, 95% CI [3.67, 15.85]), but verbal ability and age did not (ps > .21; see Table 3). About
45.12% of the variability in recall difference scores is accounted for by taking the values of
Distractibility, verbal ability, and age into account. Thus, when accounting for other types of
variables that may affect how much children’s recall changes from the contingent book condition
such as age and verbal ability, children’s Distractibility was the only unique predictor.
Table 3
Multiple Regression Analysis Predicting Changes in Recall from Contingent Condition
β
SE
t
p
95% CI
F
df
R2
Model
0.004
6.02
3, 22
.45
Distractibility
9.76
2.94
3.33
0.003
[3.67, 15.85]
Verbal Ability
-0.23
0.18
-1.28
0.213
[-0.61, 0.14]
Age
1.04
2.97
-0.35
0.730
[-5.13, 7.21]
Note. N = 26 for children who completed the PPVT
Discussion
Results from Experiment 1 indicate that the use of the contingent book resulted in higher
mean story recall compared to the use of the noncontingent board book. It was also found that
children with less developed attention regulation exhibited the greatest gains in recall from the
contingent book. However, it is an open question as to whether story recall was enhanced from
the contingent book because of the contingent responsivity, or because presenting a story on a
digital platform is superior to a traditional board book, perhaps due to a novelty effect. A novelty
effect occurs when a new technology is instituted and performance improves simply because

CONTINGENT RESPONSIVITY IN E-BOOKS 20
participants are exposed to a new device, not necessarily because participants are exposed to a
more effective one (Clark, 1985). Experiment 2 begins to explore this possibility by replicating
Experiment 1 with the control condition of a noncontingent static digital book, eliminating the
possibility of the effects being driven by the novel effect of technology. Experiment 2 also
attempts to replicate the results of investigating whether the contingent book might be especially
useful for children with less developed attention regulation.
Experiment 2
The goal of this experiment was to assess the possibility that the results on children’s
story recall in Experiment 1 were solely because exposure to a book on a digital platform is
superior to a board book, perhaps due to a novelty effect (that is, children may have been more
attentive to the iPad than the board book simply because they were less familiar with iPads). To
assess this possibility, in this experiment both a contingent and a noncontingent story are
presented to children via an iPad. If the effects on recall observed in Experiment 1 are due to the
contingent responsivity of the story (rather than the presence of an iPad), the story recall
advantage for the contingent book should be reproduced in this experiment.
Method
Participants All participants were recruited from the same pre-primary school and none of these
children participated in Experiment 1. The study used a within-subject design with data from 33
children (20 males, 13 females) ages 3 to 5-years-old (M = 53.32 months, SD = 7.15 months).
An additional child was tested but excluded due to speaking English as a second language with
low proficiency; this child could not understand the stories or the recall questions.
Materials and Procedure Procedure and apparatus (recording equipment, reader protocol) were
identical to those described in Experiment 1. The materials were nearly identical to those of

CONTINGENT RESPONSIVITY IN E-BOOKS 21
Experiment 1. The one difference was that the noncontingent control condition was a static
digital book presented on an Apple iPad (9.4 in x 6.6 in). The contingent and static books were
identical in platforms, but the contingent book responded contingently on children’s
vocalizations while the static book remained motionless. As in Experiment 1, children were read
Cat’s Pajamas and Zoom City. Each story was read on an iPad, presented either with
noncontingent static images or with animations that contingently responded to the child’s
utterance. Condition and story order were randomly assigned and counterbalanced.
Measures Measures of story recall, engagement, attention, and verbal ability were identical to the
measures described in Experiment 1.
Results
The same data analytic approach as in Experiment 1 was used.
Story Recall There was a main effect of book condition, in that children’s recall scores were
significantly higher in the contingent book condition (M = 64.72%, SE = 2.47%) compared to the
noncontingent static book condition (M = 45.89%, SE = 1.89%), F(1, 28) = 42.34; p < .0005; ηp2
= .60 (see Table 4, for raw and scaled scores). There was no main effect of order, F(3, 28) = .19,
p = .91, or sex, F(1, 28) = 2.18, p = .15. There were also no significant interactions between any
of these factors and story recall (all ps > .37). Follow-up pairwise comparisons after Bonferroni
corrections revealed that on average, children scored 18.17% (SE = 2.79%) higher on the recall
assessment in contingent book condition compared to the noncontingent book condition, 95% CI
[12.45%, 23.89%], p < .0005. All but five participants exhibited higher recall scores using the
contingent book compared to using the noncontingent static book, all five of whom exhibited
identical scores across conditions. The outcome for each condition of the recall measure
followed a normal distribution and there was one outlier with a recall score of 28.57% in the

CONTINGENT RESPONSIVITY IN E-BOOKS 22
contingent book condition. With the removal of this outlier, there was still evidence of a main
effect of book condition on recall, F(1, 27) = 48.00; p < .0005; ηp2 = .64. Taken together, these
results indicate that children’s mean story recall scores after being read to from the contingent
book were higher compared to being read to from the noncontingent static book, regardless of
the story or order in which the books were presented (see Figure 4, for paired box plot).
Figure 4. Paired box plot of recall scores in the noncontingent book and contingent book
conditions. Data points were jittered in R by .02 to prevent overplotting (Team, 2017).
Engagement Consistent with Experiment 1, there was a ceiling effect: children attended to the
book throughout the entire reading session for both conditions (see Table 4, for mean reading
time and mean looking time towards book conditions and the online supplemental material for
the analyses on the effect of book condition on engagement).
Table 4
Recall and Engagement measures, by condition
Measure M (SD)
Noncontingent Static
Book
Contingent Book
Story Recall Raw Scores
6.42 (1.52)
9.06 (1.98)
Story Recall (% of 14)
45.89% (10.87%)
64.72% (14.17%)
Total Reading Time (ms)
140433 (35567)
145900 (34098)
Total Looking Time at Book (ms)
136567 (35437)
143167 (33213)
Engagement (Looking/Reading Time %)
97.10% (3.65%)
98.18% (2.13%)

CONTINGENT RESPONSIVITY IN E-BOOKS 23
The Role of Individual Differences in Attention With the goal of replicating results from
Experiment 1, we examined whether the use of the contingent book might be especially useful
for participants with less developed attention regulation. Recall difference scores ranged from
0.00% to 57.14%, with a mean of 18.83% (SD = 15.45%). Standardized z-scores of attention task
distractors (M = 5.94; SD = 6.09) and time to complete the attention task (M = 169.29 s; SD =
33.08 s) created the composite variable of Distractibility (M =.50; SD =.75). Thirty-two
participants returned to the laboratory and were administered the PPVT (M = 116.91, SD =
16.37). Higher Distractibility scores, r(33) = .71, 95% CI [9.16, 19.76], p < .0005 (see Figure
5A) were associated with how much children’s story recall changed from using the contingent
book condition, but verbal ability scores were not, r(32) = -.04, 95% CI [-.32, .39], p = .83 (see
Figure 5B). There were three outliers with PPVT scores of 85, 83, and 65. With the removal of
these outliers, there was still not a significant association between verbal ability scores and
children’s recall difference scores, r(29) = .23, 95% CI [-.23, .94], p = .23.
Figure 5. Scatterplots of correlations between recall difference scores and performance on the
attention task and PPVT. (A) Greater Distractibility was associated with higher recall difference
scores. (B) Verbal ability was not statistically associated with recall difference scores.
There were no statistically significant interactions (Fchange = .39, df = 6, 25, p = .77) between
Distractibility, age, and verbal ability and their effects on recall difference scores. Therefore, the
final multiple regression analysis performed excluded the interaction terms as they were not

CONTINGENT RESPONSIVITY IN E-BOOKS 24
significant. The additive model (Fchange = 12.19, df = 3, 28, p < .0005) revealed that
Distractibility (β = 17.57, t = 5.94, p < .0005, 95% CI [11.51, 23.64]) accounted for unique
variance in changes in story recall using the contingent book, but verbal ability and age did not
(all ps > .05; see Table 5). About 56.64% of the variability in recall difference scores is
accounted for by taking the values of Distractibility, verbal ability, and age into account.
Table 5
Multiple Regression Analysis Predicting Changes in Recall from Contingent Condition
β
SE
t
p
95% CI
F
df
R2
Model
< 0.0005
12.19
3, 28
.566
Distractibility
17.57
2.96
5.94
< 0.0005
[11.51, 23.64]
Verbal Ability
0.10
0.12
0.83
0.41
[-0.15, 0.35]
Age
6.68
3.36
2.05
0.05
[-0.01, 13.37]
Note. N = 32 for children who completed the PPVT
Discussion
Results from Experiment 1 revealed that children’s recall scores were significantly higher
in the contingent book condition compared to a noncontingent board book condition. These
findings were replicated in Experiment 2 using a noncontingent electronic book condition,
discarding the possibility that exposure to the contingent book was superior for children’s story
recall compared to the noncontingent board book due to the novel effect of being presented with
an iPad. Findings from Experiment 2 also replicated the result from Experiment 1 that the
contingent book was especially useful for children with less developed attention regulation.
Although verbal ability was not found to be a significant predictor of how much
children’s recall changed from the contingent book condition, this finding does not indicate that
vocabulary ability is completely unrelated to story recall. The books utilized in this study
contained vocabulary that was designed to be comprehensible to all children in this age range. A
plausible reason that individual differences in vocabulary knowledge were not predictive of how
much children’s recall benefited from the contingent book is because children’s verbal ability

CONTINGENT RESPONSIVITY IN E-BOOKS 25
skills were above the vocabulary level of the books and therefore, the vocabulary gains would
not be as evident as the recall gains.
Children use salience cues like illumination and movement when learning content, and
children’s selective attention to salient features congruent with content predicts better learning
(Moore, Angelopoulos, & Bennett, 1999). It is plausible that animations congruent to the text
enhanced children’s story recall by orienting their attention to nonverbal information that
matched the story narrative. Experiment 3 begins to explore this possibility by replicating
Experiment 1 and 2 with the control condition of an animated book presentation in which
congruent animations are deployed, but they do not respond contingently on the vocalizations of
the child.
Experiment 3
The goal of this experiment was to assess the possibility that the results on children’s
story recall in Experiments 1 and 2 were solely because exposure to animations is superior to
motionless images, perhaps due to salience cues. To do so, in this experiment participants were
presented with both a contingent book condition and a noncontingent animated book control. In
the animated book condition, the story-related animations are deployed for each page, but they
do not respond contingently to the vocalizations of the child. If the effects observed in
Experiment 1 and 2 are due to the contingent responsivity of the story (rather than the mere
presence of animations), the recall advantage for the contingent condition should be reproduced
in this experiment.
Method
Participants All participants were recruited from the same pre-primary school as in Experiments
1 and 2, and none of these children participated in either Experiment 1 or Experiment 2. The

CONTINGENT RESPONSIVITY IN E-BOOKS 26
study used a within-subject design with data from 22 children (11 males, 11 females) ages 3 to 5-
years-old (M = 54.29 months, SD = 7.33 months). Four additional children were tested but
excluded due to equipment failure. The sample size is smaller than the sample sizes in
Experiments 1 and 2 because there were no children left in the pre-primary school to test and
recruiting participants from different schools might introduce potential confounds.
Materials and Procedure Procedure and apparatus were identical to those described in
Experiments 1 and 2. The materials were nearly identical to the two previous experiments. The
one difference was that the noncontingent control condition was an animated book. As in
Experiments 1 and 2, children were read Cat’s Pajamas and Zoom City. Children were read each
story in the presentation of a noncontingent animated book or in the presentation of a contingent
book. Condition and story order were randomly assigned and counterbalanced. The contingent
and animated books were identical in pictures, text, platforms (Apple iPad 9.4 in x 6.6 in), and
animations; however, while the contingent book’s animations deployed contingently on
children’s vocalizations, the animated book’s animations were deployed at the start of each page.
Identical to Experiments 1 and 2, for the contingent book, the reader started each session by
reading the first line of the story, and then said, “Now it’s your turn to read!” to the child. For the
animated book, the animations were deployed, the reader read the first line of the story and then
said, “Now it’s your turn to read!” to the child (see the online supplemental material for reader
protocol).
We explored the control conditions of having the animations deploy at the end of each
page’s narration and in response to the adult’s vocalizations rather than the child’s, but pragmatic
considerations and pilot testing favored page-initial animations. Deploying the animations
contingently on the adult reader’s vocalizations and having the animations occur after the page

CONTINGENT RESPONSIVITY IN E-BOOKS 27
was read were control conditions too similar to the contingent book condition because several of
the animations appeared to respond contingently on children’s vocalizations by chance. By
having the animations deploy at the start of each page, it was possible to have a control book
condition in which the relevant story information was highlighted without the animations
appearing to respond contingently on children’s vocalizations.
Measures Measures of story recall and attention were identical to the measures described in
Experiment 1 and Experiment 2. Due to the time-intensive and laborious nature of coding
children’s eye gaze fixations and recruiting children to return to the laboratory for an additional
testing session, the engagement and vocabulary measures were not included in Experiment 3
because there were no effects of book condition on engagement and the multiple regression
analyses revealed that PPVT performance lacked predictive power in Experiments 1 and 2.
Results
The same data analytic approach as in Experiment 1 and Experiment 2 was used with the one
difference that the multiple regression analysis conducted did not include verbal ability as a
predictor of recall difference scores, only Distractibility and age.
Story Recall There was a main effect of book condition, in that children’s recall scores were
significantly higher in the contingent book condition (M = 59.42%, SE = 3.06%) compared to the
noncontingent animated book condition (M = 45.13%, SE = 2.46%), F(1, 17) = 39.97; p < .0005;
ηp2 = .70. There was no main effect of order, F(3, 17) = 3.05, p = .06, or sex, F(1, 17) = .23, p =
.64. There were also no significant interactions between any of these factors and story recall (all
ps > .22). Follow-up pairwise comparisons after Bonferroni corrections revealed that on average,
children scored 14.33% (SE = 2.27%) higher on the recall assessment in contingent book
condition compared to the noncontingent book condition, 95% CI [9.55%, 19.11%], p < .0005.

CONTINGENT RESPONSIVITY IN E-BOOKS 28
All but three participants exhibited higher recall scores using the contingent book compared to
using the noncontingent animated book, two of which exhibited identical scores across
conditions. The outcome for each condition of the recall measure followed a normal distribution
and there was one outlier with a recall score of 100.00% in the contingent book condition. With
the removal of this outlier, there was still evidence of a main effect of book condition on
children’s story recall, F(1, 16) = 35.90; p < .0005; ηp2 = .69). Taken together, these results
indicate that children’s mean story recall scores using the contingent book were higher compared
to the noncontingent animated book, regardless of the story or order in which the books were
presented (see Figure 6, for paired box plot).
Figure 6. Paired box plot of recall scores in the noncontingent book and contingent book
conditions. Data points were jittered by .02 in R to prevent overplotting (Team, 2017).
The Role of Individual Differences in Attention Recall difference scores ranged from -7.14%
to 35.71%, with a mean of 14.29% (SD = 10.80%). Standardized z-scores of attention task
distractors (M = 2.36; SD = 3.26) and time to complete the attention task (M = 108.66 s; SD =
62.94 s) created the composite variable of Distractibility (M =-.41; SD =.79). The multiple
regression analysis showed no significant interactions (Fchange = 0.00003, df = 3, 18, p = 1.0)
between Distractibility and age and their effects on recall difference scores (β = -.02, t = -.01, p =

CONTINGENT RESPONSIVITY IN E-BOOKS 29
1.0). Therefore, the final multiple regression analysis performed excluded the interaction term as
it was not significant. The additive model (Fchange = 8.46, df = 2, 19, R2 = .47, p = 0.002) revealed
that Distractibility (β = 9.02, t = 3.95, p = .0009, 95% CI [4.24, 13.81]) accounted for unique
variance in changes in recall using the contingent book, but age did not (β = -2.86, t = -.97, p =
.35, 95% CI [-9.04, 3.32]). There were no outliers. About 47.10% of the variability in recall
difference scores is accounted for by taking the values of Distractibility and age into account.
The contingent book was especially helpful for children with less developed attentional control:
as children’s measure of Distractibility increased, they showed more benefit in recall from using
the contingent book (see Figure 7).
Figure 7. Scatterplot of correlation between recall difference scores and performance on the
attention task: greater Distractibility was associated with higher recall difference scores.
A possible explanation for the fact that age was never a significant predictor of how
much children’s story recall gained from the contingent book condition is because age is
correlated with Distractibility, so age in and of itself is not a strong predictor (see Figure S2, in
the online supplemental material, for scatterplot of correlation between age in months and
attention regulation across all three experiments). Animations that respond contingently may
improve story recall beyond animations alone because even when animations direct attention to
relevant story information, the salient cues are not effective unless they are in sync with the
child. Contingent responsivity in adult-child interactions enhances learning because following a

CONTINGENT RESPONSIVITY IN E-BOOKS 30
child’s vocalizations with relevant responses adapts to the child’s current focus of attention and
encourages children to continue engaging throughout the reading experience; whereas
animations that come beforehand do not give children immediate and appropriate feedback that
match the child’s communicative attempts that would foster understanding of the story content.
General Discussion
These data provide the first, to our knowledge, systematic analysis of whether contingent
responsivity on children’s utterances from a digital book could support children’s learning. There
were significant differences in story recall, with nearly all children exhibiting higher recall scores
from the contingent book compared to a noncontingent board book. Our findings were
strengthened when these results were replicated in a second experiment comparing the use of the
contingent book with a noncontingent static book, discarding the possibility that exposure to the
contingent book was superior to the board book due to the novel effect of a digital device.
Furthermore, it was shown that the animations were not driving the effect in a follow-up control
experiment comparing the use of the contingent book with a noncontingent animated book.
Similar to the effects contingent responsivity from adults have on children during shared book
reading, when each of the children’s vocalizations was returned with reinforced guidance to
appropriate story content, this enabled children to later reproduce a better understanding about
the story. One limitation is we were unable to conduct accurate a priori power analyses because
of the novel technical aspects of the study methodology. Although conservative interpretations
should be drawn from post-hoc power analyses and the small sample sizes of each experiment,
the present findings extend the literature in several ways. First, this is one of the first within-
subject studies that investigated the effects of contingent responsivity from a digital book on
story recall with a focus on individual differences in attention regulation. Second, the present

CONTINGENT RESPONSIVITY IN E-BOOKS 31
study highlighted the specific nature of the contingent responses that is helpful for learning:
children receiving a contingent response upon repeating the text. Third, the findings expand our
knowledge of how preschool-aged children learn from contingent responsivity instilled in an e-
book compared to how preschool children learn from other book contexts.
We hypothesized that children who attend to more distractors and take longer to complete
the attention task (i.e., children with less developed attention regulation skills) would exhibit
greater recall gains from the contingent book. Our findings support this hypothesis: the
contingent responsivity was especially useful for children with the highest Distractibility scores.
Results also showed that the associations between Distractibility and changes in recall from the
contingent book were largely not due to variance shared with verbal ability. Children’s attention
regulation is a significant predictor of academic achievement not only when they enter formal
schooling, but continue to predict academic success until several years later in development
(Franceschini et al., 2012). While prior research has found that interactive features are distracting
and detract from learning, caregiver behavior characterized by appropriately high levels of
responsiveness has been found to buffer poor attention regulation in children (Graziano, Calkins,
& Keane, 2011). The interesting finding in the differential impact of attention regulation for the
contingent book advances current theories of the beneficial effects from contingent responsivity
on young children’s learning from a digital device. The novelty of these findings is that this
study investigated individual differences in attention regulation in predicting story recall in the
context of the interaction with various designs of e-books, a topic that is understudied. Although
this study clearly exhibits a contingent responsivity advantage on story recall for children with
less developed attention regulation, it does not address specific hypotheses regarding this

CONTINGENT RESPONSIVITY IN E-BOOKS 32
advantage. Two plausible overlapping hypotheses that support this advantage are the roles of
positive reinforcement learning and associative learning with attention regulation.
Contingent use of positive reinforcement reduces attention regulation problems
(Lunkenheimer, et al., 2008). The comprehension of word-referent relations increases when
adults provide actions considered to be positive to children (i.e., attention, praise, following their
lead) contingent upon characteristics of the child's vocalizations (Whitehurst, 1988). The
contingent responsivity advantage in e-books on story recall for children with less developed
attention regulation could potentially be elicited engagement through positive reinforcement:
children’s responses are required for progress through the story, e-books uniquely respond to
children’s responses, and acknowledgement and rewards through a contingent response from the
book occur when children answer correctly (Troseth et al., 2016). Contingent responsivity also
encourages children to be active rather than passive participants, and children’s learning is
optimized when they are engaged rather than distracted (Dore et al., 2018). Following a child’s
vocalizations with relevant responses adapts to the child’s current focus of attention and
encourages children to continue engaging throughout the reading experience; whereas typical e-
books and interacting with books alone do not give children immediate and appropriate feedback
that match the child’s communicative attempts that would foster understanding of the story
content. The contingent responses upon repeating the story text may provide children with
feelings of accomplishment and may serve as positive reinforcement, which in turn enhance
learning (Troseth et al., 2016).
Children scored lower on the attention task because these children’s ability to selectively
attend to relevant information while suppressing irrelevant, extraneous information is less
efficient. Many storybook designs for young children integrate colorful visuals and decorational

CONTINGENT RESPONSIVITY IN E-BOOKS 33
illustrations. Although the inclusion of entertaining visuals in children’s reading materials have
enormous potential to engage children—if children are not given the appropriate guidance to the
content related to the story text—these additional visuals might be counterproductive if they
distract children from processing the narrative (e.g., exploring pictures of cats when they should
be focusing on pictures of cars). Children verbalizing a word with contingent feedback to the
matching referent guides children's attention to story-relevant features that may improve
encoding, storage, and retention of material, and thereby facilitate subsequent retrieval and use
(Schunk, 1986). The contingent animations that are synced with children’s vocalizations signify
to children the relevant material related to the story they should attend to, and help them develop
a better understanding of the story because the animations in the contingent book match the
simultaneously pronounced story text. Children may encode the vocalized words and associated
animations to form a unitary representation of the story content (Baker, Olson, & Behrmann,
2004). The actions of the contingent book and the children are coordinated and in sync with one
another: the children vocalize, and the contingent book responds with referents that match those
vocalizations. Because the animations in the contingent book match the simultaneously
pronounced story text, the children are not forced to constantly switch between exploring the
entertaining visuals in the storybooks and processing the story narrative. Instead, the visuals
helped integrate nonverbal information and language. This contingent responsivity may be
especially useful for children with less developed attention regulation because it encourages
children to focus on relevant story content that matches their words, leading to better attention to
the main story elements and therefore higher levels of story recall.
Children performed better in the contingent book condition across all ages in our
experiments. The developmental trajectory for the effectiveness of contingent responsivity in

CONTINGENT RESPONSIVITY IN E-BOOKS 34
digital books might change with a variety of age ranges. For younger children ages 1- to 2-years
old who are developing their language production skills, the effects of contingent responsivity on
story recall might not be as effective because developmentally, this age group is still learning to
produce vocalizations. Attention regulation skills display protracted development and are still
developing during the time when children begin formal schooling (Fisher & Kloos, 2016).
Contingent responsivity in e-books may continue to be helpful for beginning readers ages 6- to
8-years old, who are in the process of learning how to read. A recent detailed analysis of 100 of
the most popular books for beginning readers indicated that books targeting this age group
commonly contain design features that increase attentional competition for young children: on
average, 86.56% of a book’s pages contained extraneous illustrations irrelevant to the story
narrative, and attention allocation towards extraneous details were found to be negatively
associated with children’s ability to recall key story details (Godwin, Eng, Murray, & Fisher,
2019). Therefore, contingent responsivity in e-books may be helpful for beginning readers ages
6- to 8-years old, who are also commonly exposed to extensively embellished storybooks and
whose attention regulation skills are still developing. The contingent responsivity may become
less useful for older children ages 9- to 11-years old, who are starting to transition to chapter
books without illustrations and whose attention regulation and reading skills are more developed.
A limitation to this study is that the recall assessment primarily focused on questions that
require children to recall story information through identification and description. Although the
main outcome measure in this study mainly focused on the recall of key story events, recent
research has found that early childhood teachers use recall questions as the primary instructional
strategy for comprehension in school settings and tend to ask lower-order literal questions that
elicit one word responses from children (Walsh & Hodge, 2018), increasing the generalizability

CONTINGENT RESPONSIVITY IN E-BOOKS 35
of these findings. Story recall competency is not only associated with global reading skills, but it
is also a strong predictor of later reading comprehension and indicates a complex mix of
children’s understanding of stories, mastery of pragmatics, syntax, and semantics (Suggate,
Schaughency, McAnally, & Reese, 2018). Furthermore, the most significant difference between
children with reading disabilities and children with typical reading skills has been found to be
performance on story retelling, showing that the recall of story elements is a useful context for
identifying strengths and weaknesses in children’s competency in understanding story narratives
(Westerveld & Gillon, 2010). While the current study cannot make direct conclusions on the
effect of contingent responsivity from an e-book on children’s global story comprehension skills,
the contingent responsivity did improve children’s ability to recall key story details, a predictor
of overall comprehension.
Future studies conducted with children from low socioeconomic (SES) households,
second language learners, and children with attention deficits may validate our findings and give
us a more thorough understanding of the effects of the contingent book for a diverse range of
children. Due to decreasing costs, marketing strategies, and subsidies by providers mobile media
such as smartphones, iPads, and tablets are more accessible—even for low SES children. In
2011, 73% of higher income families owned a mobile device compared to the 34% of lower
income families; in 2017, the percentages rose to 99% for higher income families and 96% for
lower income families (Rideout, 2017). While the digital divide is closing, the amount of mobile
media use and parental co-engagement with media still differs between low SES and high SES
households. Children from lower SES households almost spend twice as much time in front of a
screen compared to children from higher SES households, and this exposure is also more likely
to be unsupervised by a parent (Radesky & Christakis, 2016; Rideout, 2017). Children’s use of

CONTINGENT RESPONSIVITY IN E-BOOKS 36
the target words interacting with the book alone or without appropriate adult guidance may not
be as consistent unless the book interacts in the manner adults were instructed to in the current
study; but, given that children may interact with digital devices alone, contingent responsivity
functionality in e-books may be practical for times in which adults are unavailable and when
children are reading alone, and might even be more beneficial than hardcopy, static, and
animated books as evidenced by the results from this study. Because experimenters were trained
to do the bare minimum of simply reading the text aloud to children with minimal interaction and
the contingent book led to increased story recall, we hypothesize similar results would be
reproduced if prompts and narration were coming from the contingent book while children
interact with the book alone or without the presence of an adult.
Media use by preschool children may not be by itself the critical concern; however,
poorly designed educational devices might be. If a caregiver were reading a book to a child, it
would seem almost obvious that stopping the child in the middle of the page to play a game or
make an irrelevant noise would interrupt the flow of the story and distract the child from
understanding the narrative. Yet, this structure is how many interactive digital books are
designed: with tactile puzzles, memory tasks, or entertaining sound effects and animations
activated spontaneously on the story pages in ways that are not central to the narrative (Vaala,
Ly, & Levine, 2015). When well-deployed and designed, features in technology have the
potential to enrich, not hinder learning experiences for children.

CONTINGENT RESPONSIVITY IN E-BOOKS 37
References
Baker, C. I., Olson, C. R., & Behrmann, M. (2004). Role of attention and perceptual grouping in
visual statistical learning. Psychological Science, 15(7), 460-466.
http://dx.doi.org/10.1111/j.0956-7976.2004.00702.x
Blewitt, P., Rump, K. M., Shealy, S. E., & Cook, S. A. (2009). Shared book reading: When and
how questions affect young children's word learning. Journal of Educational
Psychology, 101(2), 294. http://dx.doi.org/10.1037/a0013844
Bornstein, M. H., Tamis-LeMonda, C. S., Hahn, C. S., & Haynes, O. M. (2008). Maternal
responsiveness to young children at three ages: Longitudinal analysis of a
multidimensional, modular, and specific parenting construct. Developmental
psychology, 44(3), 867. http://dx.doi.org/10.1037/0012-1649.44.3.867
Bus, A. G. (2001). Joint caregiver-child storybook reading: A route to literacy development. In
S. B. Neuman and D. K. Dickinson (Eds.), Handbook of early literacy research (Vol. 1,
pp. 179 –191). http://dx.doi.org/10.5860/choice.39-2923
Bus, A. G., Takacs, Z. K., & Kegel, C. A. (2015). Affordances and limitations of electronic
storybooks for young children's emergent literacy. Developmental Review, 35, 79-97.
http://dx.doi.org/10.1016/j.dr.2014.12.004
Clark, R. E. (1985). Confounding in educational computing research. Journal of Educational
Computing Research, 1, 137–148. http://dx.doi.org/10.2190/hc3l-g6yd-bak9-eqb5
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological
Measurement, 20, 37–46. http://dx.doi.org/10.1177/001316446002000104
Conners, F. A. (2009). Attentional control and the simple view of reading. Reading and Writing:
An Interdisciplinary Journal, 22, 591– 613. http://dx.doi.org/10.1007/s11145-008-9126-x

CONTINGENT RESPONSIVITY IN E-BOOKS 38
Cuevas, K., & Bell, M. A. (2014). Infant attention and early childhood executive function. Child
Development, 85(2), 397-404. http://dx.doi.org/10.1111/cdev.12126
De Jong, M. T., & Bus, A. G. (2002). Quality of book-reading matters for emergent readers: An
experiment with the same book in a regular or electronic format. Journal of Educational
Psychology, 94(1), 145. http://dx.doi.org/10.1037/0022-0663.94.1.145
Dickinson, D. K., Griffith, J. A., Golinkoff, R. M., & Hirsh-Pasek, K. (2012). How reading
books fosters language development around the world. Child development
research, 2012. 1–15. http://dx.doi.org/10.1155/2012/602807
Dore, R. A., Hassinger-Das, B., Brezack, N., Valladares, T. L., Paller, A., Vu, L., ... & Hirsh-
Pasek, K. (2018). The parent advantage in fostering children's e-book
comprehension. Early Childhood Research Quarterly, 44, 24-33.
http://dx.doi.org/10.1016/j.ecresq.2018.02.002
Dunn, L. M., & Dunn, D. M. (2007). Peabody Picture Vocabulary Test, (PPVT-IV): A Measure
of Receptive Vocabulary for Standard American English.
http://dx.doi.org/10.1002/9780470373699.speced1554
Eng, C., Thiessen, E., & Tomasic, A. (2019). Contingent Responsivity in Technology Modeled
from Quality Adult-Child Interactions: Effects on Young Children’s Learning and
Attention (Version 1). figshare. Dataset. https://doi.org/10.1184/R1/7740170.v1
Faul, F., Erdfelder, E., Buchner, A., & Lang, A. G. (2009). Statistical power analyses using G*
Power 3.1: Tests for correlation and regression analyses. Behavior research
methods, 41(4), 1149-1160. http://dx.doi.org/10.3758/BRM.41.4.1149
Fisher, A., & Kloos, H. (2016). Development of selective sustained attention: The role of
executive functions. In J. A. Griffin, P. McCardle, & L. S. Freund (Eds.), Executive

CONTINGENT RESPONSIVITY IN E-BOOKS 39
function in preschool-age children: Integrating measurement, neurodevelopment, and
translational research, 215-237. http://dx.doi.org/0.1037/14797-010
Flack, Z. M., Field, A. P., & Horst, J. S. (2018). The effects of shared storybook reading on word
learning: A meta-analysis. Developmental Psychology, 54, 1334 –1346.
http://dx.doi.org/10.1037/dev0000512
Franceschini, S., Gori, S., Ruffino, M., Pedrolli, K., & Facoetti, A. (2012). A causal link between
visual spatial attention and reading acquisition. Current Biology, 22(9), 814-819.
http://dx.doi.org/10.1016/j.cub.2012.03.013
Gibbons, J., Anderson, D. R., Smith, R., Field, D. E., & Fischer, C. (1986). Young children's
recall and reconstruction of audio and audiovisual narratives. Child Development, 1014-
1023. http://dx.doi.org/10.1111/j.1467-8624.1986.tb00262.x
Godwin, K., Eng, C., Murray, G., & Fisher, A. (2019). Book design, attention, and reading
performance: Current practices and opportunities for optimization. Proceedings of the
Annual Meeting of the Cognitive Science Society, 41, 1851–1857. Retrieved from
https://mindmodeling.org/cogsci2019/papers/0326/0326.pdf
Graziano, P. A., Calkins, S. D., & Keane, S. P. (2011). Sustained attention development during
the toddlerhood to preschool period: Associations with toddlers' emotion regulation
strategies and maternal behaviour. Infant and Child Development, 20(6), 389-408.
http://dx.doi.org/10.1002/icd.731
Hargrave, A. C., & Sénéchal, M. (2000). A book reading intervention with preschool children
who have limited vocabularies: The benefits of regular reading and dialogic
reading. Early Childhood Research Quarterly, 15(1), 75-90.
http://dx.doi.org/10.1016/S0885-2006(99)00038-1

CONTINGENT RESPONSIVITY IN E-BOOKS 40
Hirsh-Pasek, K., Zosh, J. M., Golinkoff, R. M., Gray, J. H., Robb, M. B., & Kaufman, J. (2015).
Putting education in “educational” apps: lessons from the science of
learning. Psychological Science in the Public Interest, 16(1), 3-34.
http://dx.doi.org/10.1177/1529100615569721
Jennings, J. H., Caldwell, J. S., & Lerner, J. W. (2014). Reading problems: Assessment and
teaching strategies. Pearson Allyn and Bacon.
Kang, J. Y., Kim, Y. S., & Pan, B. A. (2009). Five-year-olds' book talk and story retelling:
Contributions of mother—child joint book reading. First Language, 29(3), 243-265.
http://dx.doi.org/10.1177/0142723708101680
Kendeou, P., Van den Broek, P., White, M. J., & Lynch, J. S. (2009). Predicting reading
comprehension in early elementary school: The independent contributions of oral
language and decoding skills. Journal of educational psychology, 101(4), 765.
http://dx.doi.org/10.1037/a0015956
Kirkorian, H. L., Choi, K., & Pempek, T. A. (2016). Toddlers’ word learning from contingent
and noncontingent video on touch screens. Child development, 87(2), 405-413.
http://dx.doi.org/10.1111/cdev.12508
Korkman, M., Kirk, U., & Kemp, S. (1998). NEPSY: A developmental neuropsychological
assessment. San Antonio, TX: Psychological Corporation.
Lunkenheimer, E. S., Dishion, T. J., Shaw, D. S., Connell, A. M., Gardner, F., Wilson, M. N., &
Skuban, E. M. (2008). Collateral benefits of the family check-up on early childhood
school readiness: Indirect effects of parents' positive behavior support. Developmental
psychology, 44(6), 1737. http://dx.doi.org/10.1037/a0013858

CONTINGENT RESPONSIVITY IN E-BOOKS 41
Moody, A. K., Justice, L. M., & Cabell, S. Q. (2010). Electronic versus traditional storybooks:
Relative influence on preschool children’s engagement and communication. Journal of
Early Childhood Literacy, 10(3), 294-313. http://dx.doi.org/10.1177/1468798410372162
Moore, C., Angelopoulos, M., & Bennett, P. (1999). Word learning in the context of referential
and salience cues. Developmental psychology, 35(1), 60. http://dx.doi.org/10.1037/0012-
1649.35.1.60
Nussenbaum, K., & Amso, D. (2016). An attentional Goldilocks effect: An optimal amount of
social interactivity promotes word learning from video. Journal of Cognition and
Development, 17(1), 30-40. http://dx.doi.org/10.1080/15248372.2015.1034316
Parish‐Morris, J., Mahajan, N., Hirsh‐Pasek, K., Golinkoff, R. M., & Collins, M. F. (2013).
Once upon a time: parent–child dialogue and storybook reading in the electronic
era. Mind, Brain, and Education, 7(3), 200-211. http://dx.doi.org/10.1111/mbe.12028
Piotrowski, J. T., & Krcmar, M. (2017). Reading with hotspots: Young children's responses to
touchscreen stories. Computers in Human Behavior, 70, 328-334.
http://dx.doi.org/10.1016/j.chb.2017.01.010.
Radesky, J. S., & Christakis, D. A. (2016). Keeping children’s attention: The problem with bells
and whistles. JAMA pediatrics, 170(2), 112-113.
http://dx.doi.org/10.1001/jamapediatrics.2015.3877
Read, K., Macauley, M., & Furay, E. (2014). The Seuss boost: Rhyme helps children retain
words from shared storybook reading. First Language, 34(4), 354-371.
http://dx.doi.org/10.1177/0142723714544410
Rideout, V. (2017). The common sense census: Media use by kids age zero to eight. San
Francisco, CA: Common Sense Media, 263-283.

CONTINGENT RESPONSIVITY IN E-BOOKS 42
Roseberry, S., Hirsh‐Pasek, K., & Golinkoff, R. M. (2014). Skype me! Socially contingent
interactions help toddlers learn language. Child development, 85(3), 956-970.
http://dx.doi.org/10.1111/cdev.12166
Rueda, M. R., Posner, M. I., & Rothbart, M. K. (2005). The development of executive
attention: Contributions to the emergence of self-regulation. Developmental
neuropsychology, 28(2), 573-594. http://dx.doi.org/10.1207/s15326942dn2802_2
Schunk, D. H. (1986). Verbalization and children's self-regulated learning. Contemporary
Educational Psychology, 11(4), 347-369. http://dx.doi.org/10.1016/0361-
476X(86)90030-5
Segal-Drori, O., Korat, O., Shamir, A., & Klein, P. S. (2010). Reading electronic and printed
books with and without adult instruction: Effects on emergent reading. Reading and
Writing: An Interdisciplinary Journal, 23, 913–930. http://dx.doi.org/10.1007/s11145-
009-9182-x
Sénéchal, M. (1997). The differential effect of storybook reading on preschoolers' acquisition of
expressive and receptive vocabulary. Journal of child language, 24(1), 123-138.
http://dx.doi.org/10.1017/s0305000996003005
Smeets, D. J., & Bus, A. G. (2012). Interactive electronic storybooks for kindergartners to
promote vocabulary growth. Journal of experimental child psychology, 112(1), 36-55.
http://dx.doi.org/10.1016/j.jecp.2011.12.003
Stahl, S. A., & Heubach, K. M. (2005). Fluency-oriented reading instruction. Journal of Literacy
Research, 37(1), 25-60. http://dx.doi.org/10.1207/s15548430jlr3701_2
Strouse, G. A., O'Doherty, K., & Troseth, G. L. (2013). Effective coviewing: Preschoolers’
learning from video after a dialogic questioning intervention. Developmental

CONTINGENT RESPONSIVITY IN E-BOOKS 43
psychology, 49(12), 2368. http://dx.doi.org/10.1037/a0032463
Suggate, S., Schaughency, E., McAnally, H., & Reese, E. (2018). From infancy to adolescence:
The longitudinal links between vocabulary, early literacy skills, oral narrative, and
reading comprehension. Cognitive Development, 47, 82-95.
http://dx.doi.org/10.1016/j.cogdev.2018.04.005
Tamis‐LeMonda, C. S., Bornstein, M. H., & Baumwell, L. (2001). Maternal responsiveness and
children's achievement of language milestones. Child Development, 72(3), 748-767.
http://dx.doi.org/10.1111/1467-8624.00313
R Core Team. (2017). R: A language and environment for statistical computing (Version 3.4.2)
[Statistical computing and graphics software]. Vienna, Austria: R Foundation for
Statistical Computing. Retrieved from https://www.r-project.org/
Troseth, G. L., Russo, C. E., & Strouse, G. A. (2016). What’s next for research on young
children’s interactive media?. Journal of Children and Media, 10(1), 54-62.
https://doi.org/10.1080/17482798.2015.1123166
Walsh, R. L., & Hodge, K. A. (2018). Are we asking the right questions? An analysis of research
on the effect of teachers’ questioning on children’s language during shared book reading
with young children. Journal of Early Childhood Literacy, 18(2), 264-294.
http://dx.doi.org/10.1177/1468798416659124
Westerveld, M. F., & Gillon, G. T. (2010). Oral narrative context effects on poor readers' spoken
language performance: Story retelling, story generation, and personal
narratives. International Journal of Speech-Language Pathology, 12(2), 132-141.
http://dx.doi.org/ 10.3109/17549500903414440
Whitehurst, G. J., & Valdez-Menchaca, M. C. (1988). What is the role of reinforcement in early

CONTINGENT RESPONSIVITY IN E-BOOKS 44
language acquisition? Child Development, 430-440. http://dx.doi.org/10.1111/j.1467-
8624.1988.tb01478.x
Vaala, S., Ly, A., & Levine, M. H. (2015). Getting a read on the app stores: A market scan and
analysis of children’s literacy apps. Retrieved from the Joan Ganz Cooney Center at
Sesame Workshop website http://www.joanganzcooneycenter.org/wp-
content/uploads/2015/12/jgcc_gettingaread.pdf