Young children’s demand for reciprocal engagement: evidence from typically-developing and autistic preschoolers

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: RECIPROCAL ENGAGEMENT 1

Young Children’s Demand for Reciprocal Engagement: Evidence from Typically-
Developing and Autistic Preschoolers

Allie Khalulyan

A Dissertation Presented to the
Faculty of the Graduate School
University of Southern California (USC)
In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy (Ph.D.)
Psychology
August 2018

RECIPROCAL ENGAGEMENT 2
Table of Contents
Acknowledgements .......................................................................................................... 4
Abstract ............................................................................................................................ 7
Chapter 1: General Introduction ....................................................................................... 9
Chapter 2: Study 1 - The Mutuality Demand Goes Beyond Gaze ................................. 18
Introduction .............................................................................................................. 18
Method ..................................................................................................................... 20
Results ..................................................................................................................... 26
Discussion ............................................................................................................... 30
Chapter 3: Study 2 - A Follow-Up on Alternative Explanations ...................................... 35
Introduction .............................................................................................................. 35
Method ..................................................................................................................... 36
Results ..................................................................................................................... 40
Discussion ............................................................................................................... 42
Chapter 4: Study 3 - A Comparison of Autistic Children and Controls ........................... 44
Introduction .............................................................................................................. 44
Method ..................................................................................................................... 47
Results ..................................................................................................................... 54
Discussion ............................................................................................................... 57
Chapter 5: Overview, Theoretical Implications, and Future Recommendations ............ 65
References ..................................................................................................................... 72
Appendix A ..................................................................................................................... 82
Appendix B ..................................................................................................................... 83

RECIPROCAL ENGAGEMENT 3
List of Figures
Study 1
Figure 1: Tasks & Conditions 1 ............................................................................... 19
Figure 2: Tasks & Conditions 2 ............................................................................... 20
Figure 3: Stimuli ....................................................................................................... 21
Figure 4: Aerial View ............................................................................................... 22
Figure 5: Graph of Results ...................................................................................... 29
Study 2
Figure 6: Tasks & Conditions 1 ............................................................................... 37
Figure 7: Tasks & Conditions 2 ............................................................................... 38
Figure 8: Aerial View ............................................................................................... 39
Figure 9: Graph of Results ...................................................................................... 41
Study 3
Figure 10: Tasks, Conditions, & Stimuli ................................................................... 49
Figure 11: Negation Test Stimuli ............................................................................. 49
Figure 12: Aerial View ............................................................................................. 51
Figure 13: Graph of Results .................................................................................... 55

List of Tables
Study 1
Table 1: Table of Results ......................................................................................... 28
Study 3
Table 2: Sample Characteristics ............................................................................. 55

RECIPROCAL ENGAGEMENT 4

Acknowledgements
I am grateful to Dr. Henrike Moll for her guidance, and to the children and families who
participated in this research through USC’s Minds in Development Lab. Thank you to
Kidspace Children’s Museum for kindly allowing me the space to conduct testing, and to
Dr. Jonathan Tarbox as well as the FirstSteps for Kids clinic for their recruitment of
preschoolers with autism. I am equally indebted to the research assistants who helped
me complete data collection. Special thanks to Corey Pettit, Jun Min, Alison Wood,
Maddy Suennen, and Kristina Fenske. Thank you to my committee members, Drs.
Frank Manis, Toby Mintz, and Janet Levin, for helping me navigate this process. Last
but not least, I am thankful to my partner, Matthew Stone, as well as my friends and
family for their continued support.

RECIPROCAL ENGAGEMENT 5

Funding
The work presented in this thesis was supported in part by the Organization for Autism
Research and USC’s Gold Family Graduate Fellowship.

RECIPROCAL ENGAGEMENT 6
This thesis is primarily informed by the following work:

Moll, H., & Khalulyan, A. (2016). “Not see, not hear, not speak”: Preschoolers
think they cannot perceive or address others without reciprocity. Journal of
Cognition and Development, 18(1), 152-162.
Khalulyan, A., Moll, H., & Tarbox, J. (2017). Young children’s judgments of
others’ visibility: A comparison of typically-developing and autistic
preschoolers. Manuscript in preparation.
Also by the author:
Gillespie-Lynch, K., Khalulyan, A., del Rosario, M., McCarthy, B., Gomez, L.,
Sigman, M., & Human, T. (2013). Is early joint attention associated with
school-age pragmatic language? Autism, 19(2), 168-177.
Moll, H., Khalulyan, A., & Moffett, L. (2016). 2.5-year-olds express suspense
when others approach reality with false expectations. Child Development,
88(1), 114-122.

RECIPROCAL ENGAGEMENT 7
Abstract

How does mutual gaze, or eye contact, play a role in shaping young children’s
notion of reciprocal engagement? Prior research shows that typically-developing
children under the age of five years often negate seeing others in the absence of mutual
gaze (e.g., Moll, Arellano, Guzman, Cordova, & Madrigal, 2015). Children’s negations
have been interpreted to reflect a mutualistic or reciprocal notion of person-to-person
relation, a ‘mutuality demand’ of sorts, i.e., “I can only see you if you also see me (and
vice versa).” After around five years of age, children begin to adopt a more ‘objective’
perspective and affirm seeing others in the absence of mutual gaze.
The focus of this thesis is to identify other possible conditions under which
children’s mutuality demand manifests. First, little to no attention has been paid to
whether young children hold a similar mutuality criterion for other modes of interaction,
namely vocal communication. Much like they say they cannot see another person
whose eyes are covered, will children also deny that they can hear or speak to another
person whose ears or mouth area, respectively, is covered? Given the primacy of gaze,
it is possible that children will only demand mutuality for gaze and no other modality. It
is also possible that children will demand mutuality for any modality that allows for
reciprocal engagement (i.e., gaze and vocal communication). Second, it is unknown if
children with autism conceptualize persons outside of mutual gaze differently than their
typically-developing peers. Given core deficits in gaze and reciprocal social interaction,
it is expected that children with autism will affirm seeing others in the absence of eye
contact. This work has significant implications for our understanding of the boundaries
governing young children’s mutuality demand during the early formative years.
RECIPROCAL ENGAGEMENT 8
Keywords: mutual gaze, reciprocity, autism spectrum disorder, person perception,
bidirectional engagement

RECIPROCAL ENGAGEMENT 9
Chapter 1: General Introduction
Background
Mutual gaze represents a potent form of bidirectional social interaction. Early on,
mutual gaze helps to facilitate nonverbal communication between infants and their
caregivers, and later establishes shared attention between children and their social
partners toward objects or events (Moll, Carpenter, & Tomasello, 2007). The
importance of shared gaze also plays out in children’s judgments of seeing others.
Most notably, prior work has identified that typically developing children between two
and four years of age often deny that they can see another person whose eyes are
covered (Moll et al., 2015; Russell, Gee, & Bullard, 2012) or even whose gaze is
averted (Russell et al., 2012). Therefore, children seem to endorse a ‘mutualistic
conception of seeing’ (see Moll et al., 2015), which can be summarized as: “I can see
you only if you can see me at the same time [and vice versa].” The interpretation that
follows is that typically developing children are, especially in the early formative years,
inclined to relate to others reciprocally and are inherently aware of the self in relation to
others. Between four and five years of age, children begin to relax this mutuality
demand and acknowledge that they can see a person with covered eyes.
Literature Review
Flavell, Shipstead, & Croft (1980) were among the first to conclude that young
children report being unable to see others in the absence of mutual gaze. That is, young
children report that they are unable to see others when the other party’s eyes are
covered. Children’s mutuality demand for seeing other people suggests that they
engage in some level of perspective-taking, albeit faulty: i.e., “You cannot see me, so I
RECIPROCAL ENGAGEMENT 10
cannot see you.” In some sense, children’s conceptions of mutuality are constrained by
second persons whereas older children and adults rely on objective or “correct”
personal accounts, i.e., “I can see you irrespective of your visual experience”.
The investigators also tested the commonly-held notion that children make
egocentric conclusions about their perception of other people. They found that children
whose eyes were covered provided negative responses when another person facing
them asked, “Do I (adult) see you (child)?” However, they also responded negatively
when the situation was flipped, i.e., when another person’s eyes were covered,
suggesting that what is at play is an error with respect to perspective-taking and not
egocentric perceptual judgments. They further claimed that children gradually ‘phase
out’ of such erroneous negations around five years of age, at which time they report
being able to see others irrespective of eye contact.
Later, McGuigan and Doherty (2006) shed light on several important exceptions
to children’s faulty perceptual judgments. Instead of occluding just the eyes, they
covered the entire head of presented agents. The investigators assessed whether a
figure’s orientation, whether upright or upside down, influences children’s judgments
about seeing others. Regardless of orientation, children judged an agent could be seen
only if the head was in plain sight. McGuigan (2009) further delineated the conditions
under which children demand mutual gaze by manipulating the orientation of a figure’s
head relative to an observer. Interestingly, the mere presence of a head did not satisfy
children’s mutuality demand—instead, only forward-facing heads and uncovered eyes
satisfied children’s mutuality demand. When the figure was facing away from the
RECIPROCAL ENGAGEMENT 11
observer either partially (e.g., 90 degree turn) or completely (180 degree turn), children
reported that they could not see the figure.
Thereafter, Russell et al. (2012) replicated prior findings that children consider
themselves and others unable to be seen when either party cannot reciprocate gaze.
Moreover, children affirmed that their head or the head of another could be seen when
the eyes were covered—suggesting their ability to differentiate between the eyes as a
proxy for personhood and other parts of the body. In another experiment, the
investigators confirmed that the visibility of the eyes, and not visual access, sets the
groundwork for children’s demand of mutuality. Despite children’s prior training and
knowledge that visual properties were afforded by a pair of outwardly opaque, but
inwardly transparent glasses, they nevertheless still denied being able to see the other
wearing these glasses. Although another agent could still see them, and this visual
experience was seemingly bidirectional, eye contact that was ‘out in the open’ was what
mattered to children.
More recently, Moll et al. (2015) investigated children’s negations of seeing
others in the absence of mutual gaze as a function of the verb employed in the
perceptual question. The investigators tested the speculation (see Ryle, 1954/2002)
that ‘see’ is best characterized as a detection or mental verb whereas ‘look at’ is an
inspection or behavioral verb. They found that children’s negations were strongest
when the verb see versus look at was employed. The investigators interpreted their
finding by suggesting that ‘see’ fits within the figurative realm of speech and implies
some sort of successful mutual engagement (e.g., “I would like to see you”) whereas
RECIPROCAL ENGAGEMENT 12
‘look at’ carries a literal connotation and implies directing one’s visual attention to
another person or thing.
Moreover, the children with the firmest grasp of the personal pronoun ‘each
other’ were more likely to negate seeing another person in the absence of mutual gaze.
This finding suggests that a pragmatic understanding of reciprocal interaction is related
to children’s demand for mutual regard. As a control to rule out haphazard responding,
Moll et al. (2015) also presented children with artifacts that had a particular salient
characteristic occluded (e.g., the headlights of a toy truck). There was no generalization
of negative responses from agents to artifacts, that is, children affirmed seeing partially
occluded artifacts, demonstrating that they reserve the mutuality demand for agents.
A phenomenon related to children’s mutuality demand concerns their poor hiding
strategies. Children are characterized as largely egocentric because they fail to hide
themselves properly from others, often only considering their own viewpoint. In games
of hide-and-seek, children adopt a peculiar strategy. They hide themselves by simply
covering their eyes or head instead of their entire body. This seemingly ineffective
hiding strategy might come off as egocentric on the surface, but it actually underscores
how central the eyes are to young children in matters of personhood. Russell et al.
(2012) found that children explicitly judged the faulty hiding strategies (e.g., covering
only the eyes or head) of other agents as appropriate and effective. In a similar vein,
these same children denied that they themselves and others could be seen in the
context of averted gaze.
Despite children’s mutuality demand for gaze being understudied, the above-
mentioned body of work provides converging evidence that children under five years of
RECIPROCAL ENGAGEMENT 13
age find themselves unable to see others when gaze is not reciprocated. Such a
developmental ‘growth error’ (see Russell et al., 2012) is important because it may
reference an important marker of healthy social-cognitive development. According to
Russell et al. (2012), children’s stringent criteria for mutual regard might index a natural
outgrowth of an earlier-developing drive for shared attention, or the intersubjective
sharing of experience between the child and her social partner. In other words, seeing
becomes a mutualistic act because children’s engagement is largely in the context of
bidirectional sharing of attention. Thus, those children who are more typical, but not
those that display challenges, with respect to shared attention should present with a
more robust mutuality demand. This speculation should be investigated further with
empirical studies to help us trace the developmental progression of the mutuality
demand, and to help us understand how it is situated within a broader social-cognitive
context. The mechanism of shared attention, and children’s intersubjective experiences
with social agents in general, is so pervasive in the early years that it might set the
groundwork for how children conceptualize bidirectional social interactions. In this way,
children’s insistence on mutual gaze could be viewed as a necessary byproduct of prior
forms of relating to and connecting with other people, and ought to be examined in
close relation to other social-cognitive phenomena (i.e., shared attention) to better
understand the processes that underpin children’s insistence on mutual gaze.
But what does it really mean when children say they cannot see others, or when
they themselves cannot be seen, outside of mutual gaze? It is simply not the case that
their negations reference a complete incapacity to perceive an agent with covered eyes.
The language used in prior research (e.g., Flavell et al., 1980; McGuigan & Doherty,
RECIPROCAL ENGAGEMENT 14
2006; Russell et al., 2012) seems to indicate that “children negate the visibility of
themselves and others” outside mutual gaze. Rather, children’s nay-saying has more to
do with the verb in-question, and their understanding of “see” as a success or detection
verb. Therefore, children do not think that another person with covered eyes is
conceptually imperceptible, but rather that the verb “see” is applied felicitously to cases
where gaze is mutual or bidirectional. Indeed, young children are able to, in a unilateral
fashion, visually apprehend or perceive of agents who do not make eye contact with
them. This is evidenced by Moll et al.’s (2015) finding that children say they can “look
at,” but not “see,” another person whose eyes are covered. Both verbs index visual
perception, however, the deeper linguistic meaning of “see” denotes some form of
upshot or success, wherein an act cannot be completed in the absence of mutuality or
bidirectionality. In light of this, children might be more conservative in their use of the
verb “see” and under-extend it to cases that allow for mutual or bidirectional
engagement.
Research Questions and Statement of the Problem
1. Is the mutuality demand confined to gaze, or does it operate under other domains?
A fuller understanding of why children negate seeing others outside mutuality
warrants that we examine whether their negations stem from a broader insistence on
bidirectional engagement. In this vein, do children’s negations index something deeper
about their disposition toward relating to other in reciprocal ways, or is it the case that
their negations are specific to acts of seeing? Studies on the phenomenon of the
mutuality demand have focused on gaze alone, so it is completely unknown if young
children also demand reciprocity for vocal communication (i.e., hearing and speech).
RECIPROCAL ENGAGEMENT 15
This narrow focus on gaze is understandable given the primacy and exceptional status
of gaze in ontogeny. Gaze is a uniquely potent social signal that allows agents to
simultaneously and instantly establish common ground unlike vocal communication,
which hinges on the pragmatic rules of conversational turn-taking. While it is true that
gaze has an uncontested special quality to it, other modalities, like vocal
communication, also open up a space for bidirectional engagement. Therefore, do
children likewise deny that they can hear/speak to another person whose ears/mouth
area is covered, respectively? If children also negate hearing and speaking to another
person in the absence of mutual audibility/addressability, this would suggest that young
children’s mutuality demand is not reserved for gaze.
2. Do children with autism, who show core deficits in reciprocal interaction, affirm seeing
others in the absence of mutual gaze?
The final question concerns whether the mutuality demand plays out differently in
autism spectrum disorder (ASD or autism). How will children with autism respond
compared to their typically developing peers when asked if they can see others outside
of mutual gaze? Given well-established deficits in eye contact, dyadic interaction, and
reciprocal engagement, is it likely that children with autism will respond in a more
‘objective’ or adult-like manner to questions about seeing others? Children with autism
in contrast to typically-developing peers, show marked challenges in “interpersonal
connectedness” (Hobson, 1993, 2002). More specifically, they display core deficits and
challenges in reciprocal interaction and communication (American Psychiatric
Association, 2013). These deficits are often quite pronounced in early gaze behaviors:
Compared to typically developing peers, children with autism are slower to detect direct
RECIPROCAL ENGAGEMENT 16
gaze (Senju, Yaguchi, Tojo, & Hasegawa, 2003) and do not preferentially look to the
eyes of approaching adults (Jones, Carr, & Klin, 2008).
Overall, these missed opportunities for eye contact can negatively impact major
areas of social cognition, as children with autism have a relatively difficult time using
gaze cues to pick up the mental states (Baron-Cohen, 1995) or social intentions
(Loveland & Landry, 1986) of other people. Outside of gaze, challenges with reciprocal
engagement are evident in other behaviors, such as, but not limited to, diminished
social smiling (Ozonoff, Iosif, Baguiro, Cook, Moore, Hill, Hutman et al., 2010),
impairments in interactive social play (Wolfberg & Schuler, 1999), and reduced initiation
of shared attention (Mundy, Sigman, & Kasari, 1994). Given the implication that mutual
gaze helps children to learn through others, imitate them, and connect with them
socially, it is important that early interactions in autism foster joint engagement via
mutual gaze between children and caregivers.
Framework
To address the abovementioned research questions, this thesis will focus on
findings from three key experiments. In Study 1, it was explored whether children’s
insistence on reciprocity permeates other forms of connecting with others—i.e., whether
they can not only see, but also hear and speak, respectively, to an agent whose
corresponding facial area was covered (Chapter 2). In Study 2, additional control
conditions were run to rule out alternative accounts of children’s denials that they are
able to perceive or address others outside mutual engagement (Chapter 3). In Study 3,
it was investigated whether children with ‘high-functioning’ autism, in contrast to their
typically-developing peers, demand mutual gaze given core deficits in reciprocal social
RECIPROCAL ENGAGEMENT 17
interaction (Chapter 4). The final section will conclude by summarizing the findings
across all three studies and offering directions for future research (Chapter 5).

RECIPROCAL ENGAGEMENT 18
Chapter 2: Study 1 - The Mutuality Demand Goes Beyond Gaze

Introduction

It is not known if young children’s ‘growth error’ (see Russell et al., 2012) of an
overly strict demand of reciprocity is limited to gaze or if it reflects a broader
phenomenon that permeates other forms of bidirectional engagement. It is possible that
children assign a special status to gaze over other modalities because gaze allows for a
special kind of synergy between two people—one that allows for simultaneous and
instant recognition. Due to a uniquely human white sclera, against which the
movements of the iris can be easily detected, (Kobayashi & Koshima, 1997), human
gaze is a powerful and salient social signal (Kleinke, 1986). This structural difference
facilitates an efficient way for conspecifics to signal and detect gaze (Kobayashi &
Koshima, 2001). A readiness for gaze is further evident shortly after birth, when
newborns prefer open over closed eyes (Batki, Baron-Cohen, Wheelwright, Connellan,
& Ahluwalia, 2000) and detect direct over averted gaze (Farroni, Csibra, Simion, &
Johnson, 2002). Moreover, the eyes are special because, unlike any other modality,
they serve a dual function of perception and communication (Gobel, Kim, & Richardson,
2015).
Another area that children might be inclined toward reciprocity is vocal
communication (i.e., hearing and speech). In order to make space for any meaningful
flow of information, one person speaks while the other listens, and vice versa. These
dynamics of vocal communication break down without a reciprocal, yet sequential
structure. However, gaze permits something more. It represents an unparalleled case
of ‘meeting of the minds’ between two people. The pragmatic rules of turn-taking that
RECIPROCAL ENGAGEMENT 19
apply to the domain of vocal communication do not necessarily apply in the domain of
gaze, which has the powerful position of instantly and simultaneously binding agents in
a shared social context.
This study will be the first to explore whether children extend the mutuality
demand across broader domains of reciprocal engagement. In light of the above
arguments, both opposing views are plausible: (1) That children’s demand of mutuality
and reciprocal engagement holds only in the domain of gaze due to its special status as
the only modality which allows for instant and simultaneous reciprocity and mutual
recognition; (2) That children have a flexible or broad notion of reciprocity, and that this
plays out in vocal communication.
The above-outlined positions were investigated by asking preschoolers (3- and 4-
years-old) whether they can see (Seeing Task), hear (Hearing Task), or speak to
(Speaking Task) an adult whose eyes, ears, or mouth were/was occluded (Occlusion
Condition), respectively—as shown in Figure 1 and Figure 2. Children’s answers were
compared to a control condition in which the facial area was not occluded (No Occlusion
Condition).

Figure 1. The adult in the Seeing (left), Hearing (center), and Speaking (right) Task. Occlusion by hand.

RECIPROCAL ENGAGEMENT 20

Figure 2. The adult in the Seeing (left), Hearing (center), and Speaking (right) Task. Occlusion by device.

Method
Participants
Participants were 24 (12 female) 35-52-month-olds (M = 45;06 months; range =
36;05 – 52;29 months). They were recruited from a local children’s museum or from a
database of parents whose children had previously participated in developmental
studies. Pre-established criteria for participation were that children spoke English, had
no history of physical or mental disabilities, and were born full-term (≥ 37 weeks
gestation). Over half (15) of the participants were White, eight were ‘Other,’ and one
was Asian. Participants were tested on an individual basis in a testing room at the local
children’s museum (n = 21) or at the university’s child research laboratory (n = 3)
located in a large American city.
Materials
A soft soccer ball (20 cm in circumference) was used as a warm-up toy. In the
experimental procedure, one of two experimenters (E1 or E2) used either her hands or
a device to cover a specific facial area. For the Eyes Task and Mouth Task, a black fold
made of felt (23 cm long, 7 cm wide) was used. For the Ears Task, a pair of black
protective earmuffs (13 cm in diameter) was used. Figure 3 shows these occluding
devices.
RECIPROCAL ENGAGEMENT 21

Figure 3. Devices used to occlude the eyes or mouth (left) and ears (right).
Design
A within-subjects design was employed. There were two conditions for all three
(Eyes, Ears, Mouth) tasks: An Experimental and a Control Condition, with each child
receiving both conditions of each task. There were four trials per condition: two with E1
as the agent who occluded her facial area and two with E2 as the agent who occluded
her facial area. In one of the trials, E1 or E2 used her hands—and, in the other trial,
she used the designated device to cover her facial area. Children thus received a total
of 24 trials (three tasks across two conditions, two occluder types, and two
experimenters). The order of experimenters was held constant for each of the three
tasks, with E1 always going first. The order of occluders and conditions was quasi-
randomized, with each child receiving one of six pre-established orders.
Procedure
Parents were positioned behind the child or observed the procedure through a
one-way mirror from an adjacent room. The child and experimenters sat in a triangular
formation, with the child approximately 85 cm across from both experimenters. E1 and
E2 sat next to one another approximately 60 cm apart. Figure 4 illustrates the seating
RECIPROCAL ENGAGEMENT 22
layout during the experimental session. Prior to the actual experiment, the
experimenters and child played with a small ball for 90 seconds (or until the child was
acclimated). During this brief warm-up period, the experimenters refrained from using
perceptual terms such as ‘see’ or ‘look at.’ E1 then withdrew the ball and started the
experiment with a demonstration phase in which it was made apparent to the child that
when the experimenter occluded a facial area (e.g., eyes, ears, mouth), her perception
or speech was impaired. The demonstration phase was important in particular for the
Ear Task, in which it was less clear that the experimenter was perceptually impaired.
The order of the demonstration phase varied depending on the counterbalancing
schedule.

Figure 4. Aerial view of the seating arrangement during the experimental session.
Demonstration Phase
Eyes Demonstration Phase
In this demonstration phase, E1 positioned the blindfold around her neck and
exclaimed, “[Name of child], let us show you something! I’ll place my hands/this
[blindfold] here, like this” (E1 placed her hands/pulled the blindfold over her eyes). E2
Door
Camera 1
Camera 2
E2
C
E1
E2
RECIPROCAL ENGAGEMENT 23
then held up 3 fingers and asked E1, “Say how many fingers!” to which E1 replied “I
don’t know.” In order to establish E1’s current state of visual impairment, E2 stated to
the child, “She can’t—her eyes are covered.” E1 uncovered her eyes by removing her
hands or pulling the blindfold down around her neck. This demonstration was
presented twice to account for the different occluder types (hands or blindfold) used
during the task.
Ears Demonstration Phase
In this demonstration phase, E1 positioned the earmuffs around her neck and
exclaimed, “[Name of child], let us show you something! I’ll place my hands/these
[earmuffs] here, like this” (E1 used her hands/the earmuffs to cover her ears). E2 called
out to E1 and said, “Repeat after me—I like turtles!” to which E1 replied, “Huh, what?”
In order to establish E1’s current state of auditory impairment, E2 stated to the child,
“She can’t—her ears are covered.” E1 uncovered her ears by removing her hands or
pulling the earmuffs down around her neck. This demonstration was presented twice to
account for the different occluder types (hands or earmuffs) used during the task
Mouth Demonstration Phase
In this demonstration phase, E1 positioned the mouthfold around her neck and
exclaimed, “[Name of child], let us show you something! I’ll place my hands/this
[mouthfold] here, like this” (E1 used her hands/the mouthfold to cover her mouth). E2
called out to E1 and said, “Repeat after me—I like ice cream!” E1 replied with
incoherent mumbling. In order to establish E1’s current state of communicative
impairment, E2 stated to the child, “She can’t—her mouth is covered.” E1 uncovered
her ears by removing her hands or pulling the earmuffs down around her neck. This
RECIPROCAL ENGAGEMENT 24
demonstration was presented twice to account for the different occluder types (hands or
earmuffs) used during the task.
Experimental Tasks
Eyes Task
Transitioning from the demonstration phase to the task, E1 looked at the child
and exclaimed, “And, now, like this!” as she covered her eyes with her hands/the
blindfold. E2 then asked the child, “And, right now, [name of child], can you see [name
of E1]?” During the control trials, the child was asked the same test question, except E1
sat silently with her eyes in plain view. For the remaining trials, the test question was
shortened to the phrase: “How about now—can you see [name of E1]?” The procedure
was repeated with E1 and E2 switching roles (i.e., E2 performed the eye occlusions and
E1 asked the test questions).
Ears Task
Transitioning from the demonstration phase to the task, E1 looked at the child
and exclaimed, “And, now, like this!” as she covered her ears with her hands/ the
earmuffs. E2 then asked the child, “And, right now, [name of child], can you hear [name
of E1]?” During the control trials, the child was asked the same test question as E1 sat
silently with her ears in plain view. For the remaining trials, the test question was
shortened to the phrase: “How about now—can you hear [name of E1]?” The procedure
was repeated with E1 and E2 switching roles (i.e., E2 performed the ear occlusions and
E1 asked the test questions).
Mouth Task
RECIPROCAL ENGAGEMENT 25
Transitioning from the demonstration phase to the task, E1 looked at the child
and exclaimed, “And, now, like this!” while covering her mouth with her hands/the
mouthfold. E2 then asked the child, “And, right now, [name of child], can you speak to
[name of E1]?” During the control trials, the child was asked the same test question as
E1 sat silently with her mouth in plain view. For the remaining trials, the test question
was shortened to the phrase: “How about now—can you speak to [name of E1]?” The
procedure was repeated with E1 and E2 switching roles (i.e., E2 performed the mouth
occlusions and E1 asked the test questions).
Scoring and Reliability
Children’s responses during the session were live-scored, with E1 and E2 taking
turns (i.e., if E1 was presenting a particular task, E2 asked test questions and scored
responses, and vice versa). A positive response (1) was scored whenever the child
responded with a “yes” or nodded her head. And a negative response (0) was scored
whenever the child responded with a “no” or shook her head. If the child’s response
was not clear, E1 asked for clarification by saying, “Is that a yes or a no?” If the child
mumbled or was inaudible, E1 asked, “[Child’s name], can you please speak more
loudly?” The manner in which children responded, whether verbally or gesturally, was
also scored. In order to assess inter-rater reliability, a research assistant blind to study
hypotheses coded a randomly selected sample of 6 (25%) of the children’s recorded
responses. The rater agreed with the experimenters’ judgment on all trials but one,
leading to a Cohen’s Kappa of .99. The raters agreed on all trials except one. Out of
576 trials, six could not be coded because a child responded by addressing the
experimenter (calling her name or asking her questions, three trials), gave no
RECIPROCAL ENGAGEMENT 26
discernable (one trial) or no (two trials) answer. These trials were excluded. For every
child, the sum of negative responses in a given combination of task and condition was
calculated and divided by the number of valid trials (four, unless a trial was omitted),
thus leading to proportional scores (for example, if a child negated three out of four
questions, her score was .75).
Results
Analyses were run using a logistic regression model with Huber-White sandwich
estimators of variance to account for repeated measurement clustering within subjects.
These analyses yielded no effects of occlusion type (hands vs. device), order of tasks or
condition, p’s > .39. Average differences between the negative responses in the
Occlusion and No Occlusion conditions were computed for each task via marginal
estimates from a task-by-condition interaction. Standard errors were computed using
the delta method (Oehlert, 1992). These average differences were used to assess how
the occlusion affected the children’s sense of being able to perceive or address the
other while removing the degree to which children have the same sense when there is
no occlusion. Table 1 shows how many children received a given difference score, with
the mean difference scores in the final column. These were .56 (SE = .09) for the
Seeing Task, .34 (SE = .08) for the Hearing Task, and .22 (SE = .08) for the Speaking
Task, respectively. These mean difference scores were compared to zero using the
marginal estimates from the logistic regression model. To counter alpha-inflation, p-
values were adjusted using the Holm-Bonferroni procedure (Holm, 1979). In the Seeing
Task, children negated the other’s visibility significantly more often when the adult’s
eyes were occluded than when they were unoccluded, p < .001, d = 1.27. In the Hearing
RECIPROCAL ENGAGEMENT 27
Task, the same children negated the adult’s audibility significantly more often when the
adult’s ears were occluded than they were unoccluded, p < .001, d = .83. In the
Speaking Task, they negated the adult’s addressability via speech significantly more
often when the adult’s mouth was covered than when it was not covered, p < .01, d =
.65. As Table 1 shows, the differences in means reflect a shift to negative answers in
Occlusion compared to No Occlusion trials, thus showing that the effect is not driven by
just a few, perhaps confused, children. It must be noted, however, that a subset of the
children did not vary their response across Occlusion and No Occlusion trials (i.e.,
proportional difference score = .00). One third of the sample (n = 8) did not differentiate
between the Occlusion and No Occlusion Condition of the Seeing Task. Moreover, a
little over one half of the sample also did not differentiate between the Occlusion and No
Occlusion Condition of the Hearing (n = 13) and Speaking Task (n = 13), respectively.
For the one child who received a proportional difference score of .33, this was because
this child only received 3 out of 4 trials for the Occlusion and No Occlusion Condition,
respectively. This child responded with a “no” to 1 out of 3 occlusion trials and with a
“yes” to all 3 no occlusions trials, thus resulting in a mean difference score of .33.
Figure 5 shows mean proportions of children’s negative responses broken down by task
and condition.
To test if children discriminated to different degrees between occlusions and non-
occlusions depending on the task, the mean difference scores for the tasks were
compared to one another using the aforementioned logistic regression with task-by-
condition interaction. These results showed that children distinguished significantly
more strongly between occlusion and non-occlusion of the eyes than of the ears, p =
RECIPROCAL ENGAGEMENT 28
.029, d = .78, and of the mouth, p = .001, d = .89. Children also showed significantly
greater discrimination between occlusion and non-occlusion of the ears compared to the
mouth, p = .034, d = .37. These findings suggest that children’s subjective sense of
being able to perceive or address another is most severely disrupted when the other’s
eyes are obstructed and, furthermore, that this disruption is stronger for the senses
(seeing, hearing) than for speech production.
Table 1. Number of children who received a given difference score for the Seeing,
Hearing, and Speaking Task.

Proportional Difference Score

Task -.25 .00 .25 .33 .50 .75 1 Total Mean Difference
Seeing - 8 - - 4 2 10 24 .56 (SE = .09)
Hearing - 13 1 - 2 4 4 24 .34 (SE = .08)
Speaking 1 13 3 1 1 3 2 24 .22 (SE = .08)
Note: Difference scores were derived from the marginal estimates from the task-by-condition interaction
in a logistic regression model of the negative responses. The child with a score of .33 had three instead of
four total valid trials.

RECIPROCAL ENGAGEMENT 29

Figure 5. The mean proportion of negative responses in the Eyes, Ears, and Mouth Task across the
Experimental and Control Condition. Error bars represent standard errors of the mean. The asterisks
denote statistical significance at a level of alpha = 0.05.

It was also tested if the younger (M = 40;30) and older (M = 50;07) half of the
sample (median split at 44;18 months) differed in the extent to which they discriminated
between occlusions and non-occlusions. The two groups were compared by examining
the interaction terms of age x condition in logistic regression models for each of the
three tasks. Unadjusted p-values are presented. In none of the tasks did the 3- and 4-
year-olds produce significantly different response patterns (p = .09, d = .03 for the
Seeing, p = .35, d = .56 for Hearing, and p = .57, d = .33 for the Speaking Task). Given
the small group sizes to detect a statistically significant group difference, it was only
descriptively apparent that the 4-year-olds had higher difference scores than the 3-year-
RECIPROCAL ENGAGEMENT 30
olds throughout the tasks, thus revealing that they have not yet outgrown the mutuality
constraint.
Further, we also analyzed the extent to which the three tasks, despite their
apparent differences, share a common thread that enables children to respond
consistently across the three modalities. To test this, we ran correlations between
difference scores for each of the three tasks. The results reveal that children’s
responses were highly correlated, with Spearmans’ s r = .82 for Seeing and Hearing, r =
.60 for Seeing and Speaking, and r = .72 for Hearing and Speaking (all ps < .001).
Children responded very consistently across the three different modalities, which
suggests that some common underlying ingredient can explain their negations: We
interpret this as evidence that children's general demand of reciprocity is what drives
such consistent responding across the tasks.
In conclusion, the data support the view that young children demand reciprocal
relatedness most robustly, but not only, for gaze. They not only deny another’s visibility
when her eyes are covered, but they also tend to sense that they cannot hear or
verbally address another whose ears or mouth are/is covered, respectively. This
suggests that children’s insistence on reciprocity is not confined to seeing, but spans
across different modes of interpersonal engagement.
Discussion
The findings from the current study suggest that children’s sense of reciprocal
relatedness pervades multiple modalities. These findings are partially consistent with
each of the above-postulated views: (1) That children’s demand of reciprocal
engagement is uniquely applied in the case of shared gaze, (2) That children do not
RECIPROCAL ENGAGEMENT 31
limit this demand to gaze, but rather extend it to other modalities. Whereas prior work
focused entirely on vision, this study demonstrates that children not only demand
mutual gaze, but they also demand mutual addressability. They denied being able to
hear or speak to someone whose ears or mouth area was covered, respectively.
Moreover, children in the current study responded consistently across all three
modalities, as evidenced by the correlations between our seeing, hearing, and speaking
tasks.
This broader insistence on mutuality by way of bidirectional vocal communication
can be summed up as: “I cannot hear you unless you can hear me” and “I cannot speak
to you unless you can speak to me.” This shows that young children do not view
communication in one direction. Instead, they conceptualize communication as a
bidirectional act between two individuals. Nevertheless, gaze is the primary modality for
children’s insistence on reciprocity. Children’s denials of mutual engagement were
strongest when mutual gaze was interrupted. This finding highlights the unparalleled
potency of gaze as a social signal. It is only in the context of mutual gaze that two
persons can experience an instant and simultaneous self-other synergy. Of course, it
should also be noted that are semantic differences between the verbs “see” and “hear”
or “speak” that might, at least partially, account for the greater number of negations we
observe in the Seeing Task. As Moll et al. (2015) suggests, children’s negations
differentiate between mental terms like “see” and behavioral terms like “look.” Extending
this logic to other modalities that make possible bidirectional engagement, it is plausible
that children might provide more negations for the terms “hear” and “speak” as opposed
to their more behavioral counterparts “listen to” and “talk to.” Thus, children’s mutuality
RECIPROCAL ENGAGEMENT 32
demand might discriminate more specifically at the semantic level, and not just at the
broader level of perceptual domain. Along these lines, success or mental verbs that are
embedded with some sort of mutualistic connotation are expected to drive children’s
mutuality demand more so than behavioral verbs.
To conclude, the data suggest that reciprocal engagement is a key ingredient in
children’s sense of being able to perceive or relate to others. For young children,
relating to others implies that perception or communication travels in both directions.
Children’s negations observed in our study are evidence of their disposition toward
reciprocal engagement. They often judge themselves unable to relate to someone who
is not poised to simultaneously relate back. This in turn leads children to under-extend
words such as see, hear, and speak, when used in reference to another agent who is
not in a position to reciprocate. These instances of under-extension suggest that
children’s perspective-taking abilities may be inextricably linked with their knowledge of
words that convey mutualistic meaning.
Limitations and Contributions
A weakness of the current study reflects the need for more stringent control
procedures. These control procedures are necessary to establish that children’s
negations reflect their mutualistic stance and not a misunderstanding of what was asked
of them. Possibilities hinting at task confusion have been addressed by previous
experiments, but not to the extent that covers modal correspondence. For example,
McGuigan (2009) asked children whether they can see an agent whose mouth was
covered, and Flavell et al. (1980) asked children whether they can see an agent whose
body parts (e.g., back or arm) were covered. In both studies, children affirmed the
RECIPROCAL ENGAGEMENT 33
visibility of agents whose eyes are exposed and whose other facial or bodily areas are
covered—illustrating that they do not indiscriminately provide negative responses when
just any area of the body is covered.
Notwithstanding the above weakness, there are several notable strengths of the
current study. The first is that it effectively replicated prior findings on gaze and
significantly enhanced our understanding of children’s application of the mutuality
demand to other sense modalities. The sample size (N = 24) was adequately large for
the purposes of studying typical development and reflected the overall rigorous methods
of the study. Furthermore, the within-subjects design allowed each participant to serve
as her own baseline and thus reduced errors associated with individual differences.
Future Directions
Future studies should incorporate procedures to rule out alternative
interpretations wherein children’s negations are not reflective of a demand for reciprocal
engagement. The following two reductive accounts seem plausible. The first is that
children may have confused the subject and object in the test question and believed
that they were asked if the adult (whose facial area was covered) could see/hear/speak
to them. Given the saliency of the covered area, it seems pragmatically sound and
highly plausible that this is in fact what the experimenter meant to ask. After all, it is the
other’s (i.e., the adult’s) perception that is impaired, and so it is possible that children
misunderstand the question accordingly. This reversal simply involves asking “Can X
see you?” instead of “Can you see X?” If children provide negative responses to these
questions, investigators can reasonably rule out subject-object confusion.
RECIPROCAL ENGAGEMENT 34
The second is that children might not be mapping the verb in–question (e.g., see)
to the appropriate facial area (e.g., eyes). It is therefore necessary to develop cross-
modality tasks in which children are asked, for instance, whether they can ‘see’ another
person whose mouth is covered, hear a person whose eyes are covered, or speak to a
person whose nose is shut. If this hypothesis is correct, then children should affirm
these questions—showing that they engage in a one-to-one mapping of the relevant
facial areas and demand mutual engagement along the same ‘communication
channels.’ However, some ‘spill-over’ of negations between different modalities might
be expected without this necessarily disconfirming a modal correspondence hypothesis.
For example, children might find themselves unable to speak to someone whose ears
are covered, or even to someone whose eyes are covered. Under those conditions,
important elements of a fully engaged, face-to-face interaction are missing and so
children negate the question because the other cannot take up the message or refuses
to engage.

RECIPROCAL ENGAGEMENT 35
Chapter 3: Study 2 - A Follow-Up on Alternative Explanations
Introduction
The current study is a follow-up control experiment (hereafter called ‘Study 2’ or
‘current study’) to the main experiment (hereafter called ‘Study 1’). Study 2 addresses
possible alternative explanations for the data in Study 1. The first alternative
interpretation is that maybe the children thought they were asked about the adult’s
ability to perceive or address them. In this way, were the children confused about the
directionality of the question being asked – or did they confuse the subject and objects
parts of speech? This seems like a pragmatically reasonable interpretation. The adult’s
perception/speech was saliently impaired, so it might seem strange to children that they
were asked about their own instead of the adult’s current state. However, this is
unlikely because prior research shows that children respond differentially, depending on
the verb deployed in the question. Specifically, they deny that they can see, but affirm
that they can look at another whose eyes are occluded (Moll et al., 2015). These kinds
of distinctions that children draw go against the view that their negations stem from a
confusion of the subject and object parts of speech.
The second alternative interpretation concerns the issue of modal
correspondence. Is it the case that children provide negative answers in the event that
any facial area is covered, regardless of the modality specified by the verb (i.e., ‘see’
related to eyes). Nevertheless, some degree of cross-modal transmission supports the
reciprocal engagement hypothesis. One notable case would be if children might judge
that they cannot speak to someone who cannot see or hear them because their greeting
does not get through to the other. However, it is expected that children should largely
RECIPROCAL ENGAGEMENT 36
attend to modal mapping. For example, children grant that they can see an agent
whose mouth is covered (McGuigan & Doherty, 2006).
In the current study, these issues are addressed with two control conditions,
each of which was compared to the critical ‘Occlusion Condition’ in Study 1 (in which
children were asked if they can see (Seeing Task), hear (Hearing Task), or speak to
(Speaking Task) an agent whose eyes, ears, and mouth were/was covered,
respectively). In the Subject Reversal Condition, children were asked if the adult could
see/hear/speak to them. The rationale was that if children in Study 1 believed they
were asked about the other’s ability to perceive or speak, their answers should closely
resemble those when actually asked about the other in the current study. In the
Modality Mismatch Condition, children were asked if they could see/hear/speak to an
adult whose ears/eyes/nose were/was covered or shut, respectively.
It is predicted that children will show more negations in the Subject Reversal than
the Occlusion Condition of Study 1, but fewer negations in the Modality Mismatch than
the Occlusion Condition of Study 1.
Method
Participants
Twenty-four (12 female) preschoolers (M = 45;26; range = 25;48 – 55;13), none
of whom took part in Study 1, participated. Ten were tested at a children’s museum and
14 at the University. Thirteen were multiracial, two Hispanic, seven White, one was
Asian, and one was African-American. No child had any known disabilities. Their
socioeconomic backgrounds varied greatly, with annual incomes ranging from <
RECIPROCAL ENGAGEMENT 37
$20,000 to > $120,000. Three further children were tested but excluded because they
were unresponsive (one), or inattentive (two).
Materials
In addition to the materials used in the critical experiment (see Figure 3), a long
black hairclip (9 x 1 x 1 cm) was used by E1 to pinch her nose shut in the Modality
Mismatch Condition of the Speaking Task.
Design
The Subject Reversal Condition always preceded the Modality Mismatch
Condition to avoid starting with strange (mismatch) questions, which might make
children question the seriousness or reliability of the adult. Both conditions were
composed of twelve trials, four for each task: two in which the organ was covered by
hand, and two in which it was covered by the device. There were 24 trials in total.
Figure 6 and Figure 7 shows the manner in which the experimenter occluded her facial
areas by hand and device.

Figure 6. The adult in the Seeing (left), Hearing (center), and Speaking (right) Task of the Modality
Mismatch Condition. Occlusion by hand.

RECIPROCAL ENGAGEMENT 38

Figure 7. The adult in the Seeing (left), Hearing (center), and Speaking (right) Task of the Modality
Mismatch Condition. Occlusion by device.

Procedure
Subject Reversal Condition
A brief demonstration phase, in which children were shown the disabling effects
of occluding a given facial area, preceded each task. E1 then covered her eyes in the
Seeing Task, her ears in the Hearing Task, and her mouth in the Speaking Task—as in
Study 1. E2, who sat opposite from E1 and next to the child, asked the child, “Can
[name of E1] see/hear/speak to you?” followed by, “Can [name of E1] see/hear/speak to
me?” (The perceiver was thus always E1, with E2 and the child as objects of
perception.) E1 then used the alternative means (e.g., her hands) to cover the same
facial area, and E2 repeated the same questions.
Modality Mismatch Condition
Consistent with the other tasks, it was briefly demonstrated to the child that E1
could not smell (a fragrance) while wearing the nose clip. E1 covered her ears in the
Seeing Task, her eyes in the Hearing Task, and pinched her nose in the Speaking Task.
After the Subject Reversal Condition, E2 moved positions and seated herself directly
beside E1 in the Modality Mismatch Condition. E2 then asked, “Can you see/hear/speak
RECIPROCAL ENGAGEMENT 39
to [name of E1]?” E2 then covered her own facial area in the same way, and E1 asked
the child the same questions about E2. Figure 8 illustrates an aerial view of the seating
arrangement during the Subject Reversal and Modality Mismatch Condition.

Figure 8. Aerial view of the seating arrangement. E2 was seated directly beside C and both were across
from E1 in the Subject Reversal Condition. E2 then moved positions and sat directly beside E1 in the
Modality Mismatch Condition.

Scoring and Reliability
Children’s responses were scored as positive or negative. Positive responses
(scored as ‘0’) included ‘yes’ and nodding; negative responses (scored as ‘1’) included
‘no’ and head-shaking. Data were scored live by the experimenters and by an
independent rater who scored the recorded responses unaware of condition. For every
child, the sum of negative responses in a given combination of task and condition was
calculated and divided by the number of valid trials (four, unless a trial was omitted),
thus leading to proportional scores (for example, if a child negated three out of four
questions, her score was .75). A total of nine out of 576 trials could not be coded
RECIPROCAL ENGAGEMENT 40
because the child said, “I don’t know” (four), gave no answer (four) or said “Huh?” (one).
Proportional scores were calculated in the same manner as in the critical experiment.
Inter-rater reliability was 100% (κ = 1).
Results
Analyses were conducted using logistic regression models with Huber-White
sandwich estimators of variance to account for repeated measurement clustering within
subjects. There were no effects of occlusion type or task order (p’s > .14). Figure 9
shows the mean proportions of children’s negative responses as a function of task and
condition. Comparisons of the Subject Reversal to the Occlusion Condition, and the
Modality Mismatch to the Occlusion Condition for each task were conducted in the
logistic regression model via a condition-by-task interaction. The marginal estimate for
the average difference between conditions was computed for each task. The Holm-
Bonferroni procedure for alpha adjustment was used for comparisons between tasks
within each hypothesis.
Seeing Task
As predicted, children produced more negations in the Subject Reversal than in
the Occlusion Condition of Study 1 (marginal significance, p = .053, d = .55). Also in
line with our prediction, they produced fewer negations in the Modality Mismatch than in
the Occlusion Condition, p = .003, d = .92.
Hearing Task
As predicted, children produced more negations in the Subject Reversal than in
the Occlusion Condition, p = .004, d = .93, and fewer negations in the Modality
Mismatch than in the Occlusion Condition, p = .009, d = .82.
RECIPROCAL ENGAGEMENT 41
Speaking Task
As predicted, children gave more negative answers in the Subject Reversal than
in the Occlusion Condition, p = .009, d = .80. Counter to our prediction, they negated
the questions equally in the Modality Mismatch and the Occlusion Condition, p = .90, d
= .01.

Figure 9. Mean proportions of children’s negative responses in the Current Experiment (with standard
errors of the mean as error bars) broken down by task and condition. Note. The Occlusion Condition
stems from Study 1. § denotes that the question was the same as in the Occlusion Condition. ** denotes
significance at p < .01; m.s. denotes marginal significance (p = .053).

Discussion
The results of the current study refute that children in the Occlusion Condition of
Study 1 thought they were asked about the adult’s ability to perceive or speak to them.
When the children were actually asked about the adult’s ability to perceive or address
them, their negations were almost at ceiling, thus demonstrating that children were
RECIPROCAL ENGAGEMENT 42
aware of the adult’s impaired state. The different levels of negation (with only marginal
significance for the Seeing Task, which may be attributed to low power due to the high
scores in the Occlusion Condition) demonstrate that the children discriminated between
questions about their own perception/speech and that of the other.
The results of the current study also rule out that children indiscriminately give
negative answers whenever the other’s face is partially occluded. The children judged
that they could see the other when her ears, and hear her when her eyes were covered,
providing evidence that modal correspondence was important to them. However, they
equally denied being able to speak to the adult when her mouth was covered, and her
nose was pinched. One can only speculate about the cause of this null result, but it is
possible that structural asymmetries between speech and the senses might explain it.
Speech is neither entirely disabled when the mouth is covered (E1 made sounds in the
demonstration phase) nor entirely intact when the nose is pinched (E1 spoke with a
nasal voice in the demonstration phase). Some of the negations might also result from
children having misunderstood the question whether they can speak to the adult as a
request to talk to her. For instance, one child began to speak to the experimenter in two
omitted trials in Study 1. Their negations might convey that they do not wish to speak to
the adult.
Overall, the findings from this control experiment strongly confirm that children’s
negations in Study 1 reflect a demand for reciprocal engagement. The data provide
compelling evidence that children’s negative responses cannot be reduced to a
pragmatic misunderstanding of the questions or arbitrary responding. The data suggest
that children do not just negate the questions because 1) they misunderstand the
RECIPROCAL ENGAGEMENT 43
question (reverse subject and object), or 2) any facial area is covered. Future work
should further delineate the boundaries of children’s insistence on reciprocal
engagement and explore whether children’s tendency to demand mutual hearing and
addressability is mediated by direct (as opposed to averted) gaze.

RECIPROCAL ENGAGEMENT 44
Chapter 4: Study 3 - A Comparison of Autistic Children and Controls
Introduction

Children’s insistence on mutual gaze highlights the importance of reciprocal
interactions during the early formative years. Second-personal, face-to-face
engagement is “intrinsically rewarding” for infants (Kinsbourne & Helt, 2011: p. 5).
Shortly after birth, infants prefer open over closed eyes, (Batki et al., 2000) and direct
over averted gaze (Farroni et al., 2002). They also show a preference for contingent
over less contingent social exchanges (Gergely & Watson, 1996, 1999; Watson, 1972)
and engage in “interactional synchrony” with others by mimicking and anticipating their
behaviors (Condon & Sander, 1974; Meltzoff & Moore, 1977). Later in development,
children spontaneously engage in turn-taking and role-reversal (Tomasello, 1999;
Tomasello, Carpenter, & Hobson, 2005).
Unlike their typically-developing (TD) peers, children with autism do not enter so
naturally into second-personal relations with others. Instead, they exhibit deficits in
reciprocal social contexts and are diminished in their sense of ‘interpersonal
connectedness’ (Hobson, 1993, 2002). The core feature of autism is impairments in
reciprocal interaction and communication (DSM-V; American Psychiatric Association,
2013). This deficit often first manifests in gaze behavior. Children with autism have
difficulties detecting another person’s direct gaze (Senju et al., 2003) and do not
preferentially look to others’ eyes (Jones et al., 2008). Their ability to recognize faces
tends to be impaired (Klin, Sparrow, De Bildt, Cicchetti, Cohen, & Volkmar, 1999), an
effect that is correlated with a hypoactivation of the “facial fusiform area” (see Schultz,
2005, for an overview) and other atypical activation patterns (Bailey, Braeutigam,
RECIPROCAL ENGAGEMENT 45
Jousmäki, & Swithenby, 2005; Dalton, Nacewicz, Johnstone, Schaefer, Gernsbacher,
Goldsmith, … & Davidson, 2005).
Individuals with autism look disproportionately longer toward the mouth (Klin,
Jones, Schultz, Volkmar, & Cohen, 2002) or other areas of the face (Pelphrey, Sasson,
Reznick, Paul, Goldman, & Piven, 2002). Further, they exhibit difficulties using others’
eyes as indicators of their intentions (Loveland and Landry, 1986) or mental states
(Baron-Cohen, 1995). Eye contact can cause an abnormally high skin conductance in
individuals with autism (Kylliäinen & Hietanen, 2006) —an effect that is correlated with
the severity of social impairment (Kaartinen, Puura, Mäkelä, Rannisto, Lemponen,
Helminen, … & Hietanen, 2012). This enhanced arousal to eye contact can lead to a
series of negative effects, causing them to miss out on key learning experiences that is
only facilitated via face-to-face interaction.
Gaze is not the only area where reciprocity breaks down in individuals with
autism. Other markers include reduced social smiling and communicative babbling
(Ozonoff, Iosif, Baguio, Cook, Hill, Hutman, … & Steinfeld, 2010), and diminished social
play (Jordan, 2003; Wolfberg, 1999). While children with autism are able to gesture to
recruit others’ attention, they tend to deploy these gestures to request desired objects
and not to initiate interaction (Charman, Swettenham, Baron-Cohen, Cox, Baird, &
Drew, 1997; Mundy, Kasari, & Sigman, 1992; Mundy, Sigman, & Kasari, 1994).
Moreover, children with autism make pronoun reversal errors, often failing to address
others with ‘you’ (see Lee et al.,1994; Hobson, Lee, & Hobson, 2010). Autistic children
also tend not to adhere to norms of conversational turn-taking (Prizant & Rydell, 1993)
and struggle to keep track of the conversational topic between self and other (Hale &
RECIPROCAL ENGAGEMENT 46
Tager-Flusberg, 2005; Tager-Flusberg & Anderson, 1991). In games of role reversal,
they often fail to alternate between complementary roles like offering and taking
(Carpenter, Tomasello, & Striano, 2005) and do not apprehend the equivalence of self
and other in these roles (Hobson and Meyer, 2005).
The abovementioned body of evidence suggests that children with autism are
inclined to relate to others in an objective third-personal sense. If this is the case, and if
typically-developing children’s negations of another’s visibility outside of eye contact
manifest their preference for second-personal engagement, then it is expected that
autistic children will affirm that they can see others who cannot see them. Children with
autism should therefore ‘bypass’ the developmental phase in which typically-developing
children express a strong inclination toward bidirectional engagement.
The Current Study
This experiment compared autistic children’s and typically-developing children’s
judgments about perceiving other people in the absence of mutual gaze. In the critical
experimental condition, a human’s eyes were occluded (Occlusion Condition). In a
control condition, her eyes were unoccluded (No Occlusion Condition). An analogous
task with objects served to control for the possibility that children with autism sense they
cannot see any partially covered object—whether human or otherwise. An object
control task is important because autism tends to be characterized by a perceptual style
that favors parts over wholes (“weak central coherence”, see Happé, 2005; Happé and
Frith, 2006) and lets their attention “stick” to particular features or details (Landry and
Bryson, 2004; Zwaigenbaum, Bryson, Rogers, Roberts, Brian, & Szatmari, 2005). As a
consequence, these children might deny seeing partially covered objects of all sorts. To
RECIPROCAL ENGAGEMENT 47
control for the alternate possibility that children with autism might haphazardly affirm all
questions, a ‘negation test’ was included to verify that children are able to deny
questions that call for a negative response.
A group of children with autism was compared to a group of typically-developing
children. Every child with autism was matched on the basis of gender and chronological
age with three typically-developing children. To ensure that the groups do not differ in
their general cognitive and language skills, each child completed the visual reception
and (expressive and receptive) language subscales of the Mullen Scales of Early
Learning (MSEL; Mullen, 1995). Additionally, each child’s parent completed the Social
Responsiveness Survey-2 (SRS-2; Constantino, 2012), which was used to confirm that
the groups markedly differ in social responsiveness (with lower levels of sociality in
autistic than control children).
Method

Participants

Participants were 36 (12 female) typically-developing 3- to 4-year-olds (M =
52;04, range = 34;07 – 59;24) and 12 (4 female) 3- to 4-year-olds with ASD (M = 52;23,
range = 36;22 – 60;15). All children spoke and understood English. Typically-
developing children were recruited from a database of children whose parents
volunteered to participate in studies of child development (n = 21), from a nearby
children’s museum (n = 14), or a local preschool (n = 1). Nine were White, 8 Asian, 7
Hispanic, 2 African-American, and 10 multiracial. Their socioeconomic backgrounds
varied widely, with annual family incomes from $20,000 to above $120,000. No
typically-developing children were excluded.
RECIPROCAL ENGAGEMENT 48
Of the children with autism, nine were recruited from a local intervention clinic for
autism and three from a local parent support group. Six were tested at home, four at
the intervention clinic, and two at the University’s child laboratory. Five children were
White, 3 African-American, 1 Asian, 1 Hispanic, and 2 multiracial. Their socioeconomic
background varied, with annual family incomes from $40,000 to above $120,000. Two
additional autistic children were tested but excluded because they failed the negation
test (1) or because a parent retracted consent following the test (1). All children in this
group met the criteria for autism spectrum disorder as specified by the Diagnostic and
Statistical Manual-V and the Autism Diagnostic Observation Schedule-Generic (DSM-V:
APA, 2013; ADOS-G: Lord et al., 2000) and were independently diagnosed by
clinicians. All had received therapy for at least 6 months (M = 16 months) prior to
participating in the study. These therapeutic interventions served to improve children’s
skills in joint attention, perspective-taking, and communication.
Materials

In the Human Task, two experimenters (E1 and E2) adopted the role of the
human which sat facing the child. In the Object Task, stimuli for children to perceive
were a house (23 x 15 x 15 cm) with a door and two windows, and a toy clock (23 cm
diameter, 8 cm high) with two hands and removable numbers. An occlusion strip made
of black felt (18 x 6 cm) was used to occlude the agents’ eyes and objects’ parts, as
shown in Figure 10.
RECIPROCAL ENGAGEMENT 49
Figure 10. One of experimenters (upper panels) and one of the objects (bottom panels) in the Occlusion
(left panels) and No Occlusion (right panels) Condition, respectively.

Negation Test. A picture book with four white pages (15 x 21 cm) was used.
Each page showed a single image (9 x 6 cm) of a dog, a chair, a fire truck, and a
banana, respectively. The page order was variable (see Design). A penguin puppet
named “Leo” (26 x 16 x 13 cm), operated by E1, asked children about the objects’
identity. Figure 11 shows the puppet and two of the book pages.

Figure 11. Sample of the stimuli used in the Negation Test. The puppet (top) posed questions about the
identity of objects (bottom) shown in the picture book.
RECIPROCAL ENGAGEMENT 50
Design
There were two tasks: the Human Task and the Object Task. For 50% of the
children in each group, the Human Task was presented before the Object Task and vice
versa for the other 50%. Each task included eight trials: four Occlusion trials and four
No Occlusion trials. In two Occlusion trials in each task, the experimenter used a felt
strip to cover her eyes or the object’s parts; in the other two, she used her hands.
Occlusion trials directly preceded or followed the corresponding No Occlusion trial, in
which the human or object was not occluded. Conditions alternated in an order that
was counterbalanced between tasks.
The Negation Test was delivered between the Human and the Object Task and
was comprised of four questions (one per page). Page order was quasi-randomized.
Questions posed on the first two pages and the last page called for a negative answer
(because the puppet misidentified the objects on these pages). The question asked on
the third page called for a positive answer (because the puppet correctly identified the
object on this page). This was interspersed to break up the sequence of questions
calling for a negative response and thereby to demand of children that they pay
attention to every question.
Procedure
After the parent filled out the required forms, the child was brought to the test
room. Parents were in the test room only if the child or parent insisted that she, the
parent, be present. In this case, she was seated outside of the child’s view and was
asked to remain silent. The child was seated in a chair, with E1 to her left and E2 to her
RECIPROCAL ENGAGEMENT 51
right. Then the experiment began with the Human or Object Task. Figure 12 depicts an
aerial view of the seating arrangement.

Figure 12. Aerial view of the seating arrangement.

Human Task
Depending on the counterbalancing schedule, either E1 or E2 took on the role as
agent and sat down facing the child at a distance of about 75 cm. What happened next
varied by condition. In the Occlusion Condition, the agent announced, “Now I’ll place
this/my hands here, like this!” and covered her eyes with the occlusion strip or her
hands. Next, the other experimenter asked, “Right now, [name of child], can you see
[name of E1]?” In the corresponding No Occlusion trial, which directly preceded or
succeeded the Occlusion trial, E2 asked the same question with the sole difference that
E1’s eyes were unoccluded. The same was repeated with the other means of occlusion
(e.g., hands if occluding strip was used earlier). This procedure was then repeated with
E1 and E2 in reversed roles of agent and questioner.
RECIPROCAL ENGAGEMENT 52
Object Task
This task took place at a table, with E1 sitting to the child’s left and E2 to her
right. E1 brought out the first object. For the house, she said, “This is a house. It has a
door – here, you can go inside!” and made finger movements toward the door. For the
clock, she said, “This is a clock. It has numbers – here, you can take them out!” and
removed some of the numbers. After the child briefly played with the object (about 45s),
E1 held the object approximately 75 cm in front of the child, roughly at eye level. In the
Occlusion Condition, E1 stated “Now I’ll place this (occlusion strip)/my hands here, like
this!”, covered the house’s windows/clock’s hands, and asked the child: “Right now,
[name of child], can you see the [name of object, e.g., house]?” In the corresponding No
Occlusion trial, which either directly preceded or succeeded the Occlusion trial, E1
asked the same question with the sole difference that the object was unoccluded. The
procedure was then repeated with the other object.
Negation Test
E1 brought out the picture book and introduced Leo, the puppet. Leo looked at
the object on the first page and asked, “Is this a [mislabel, e.g., bird]?” After the child
answered, Leo replied “Ok”. E1 turned the page and Leo asked the analogous question
about the next object etc. Leo mislabeled the first two and the final object (‘bird’ for dog,
‘bed’ for chair, ‘bike’ for firetruck, ‘ice-cream’ for banana), but correctly labeled the third
object. To be included in the final sample, children had to negate at least two of the
three questions containing mislabels. One child from the ASD group did not meet this
criterion and was excluded for this reason.
Additional Measures
RECIPROCAL ENGAGEMENT 53
Mullen Scales of Early Learning
These scales assess young children’s cognitive skills in a semi-structured
setting. The Visual Reception, Receptive Language, and Expressive Language scales
were selected because our dependent measure requires verbal or gestural answers to
questions about visual perception. Completion of the scales took approximately 30
minutes. Answers were coded live and were subsequently scored by the first author.
Raw scores were transformed into standardized equivalents for each scale.
Social Responsiveness Scale-2
This parent-report survey contains 65 items on a 4-point Likert scale and aims to
identify the presence and severity of symptoms indicative of autism. Parents took about
10 minutes to complete the survey, which was subsequently scored by the first author.
Scores lower than 59T are within the normal range; scores at or higher than 59T
indicate social impairment.
Scoring and Reliability
Responses were coded live during the session. For the perceptual questions
(Human and Object Task), positive answers included “yes” and head nodding; negative
ones included “no” and head shaking. Because the primary measure of interest was
children’s negations of these questions, negative responses were scored as “1” and
positive ones as “0”. For the negation test, correct responses (the child correctly
affirmed/denied the question asked by the puppet) were scored as “1” and incorrect
ones (the child erroneously affirmed/denied the question asked by the puppet) were
scored as “0”. To assess inter-rater reliability for these two measures, a research
assistant, who was unaware of group membership and the hypothesis, coded the
RECIPROCAL ENGAGEMENT 54
recorded answers of a randomly selected subset of 25% of the children in each group (4
autistic children, 9 typically-developing children). Inter-rater reliability was perfect, both
for the perceptual questions and for the negation test (Kappa = 1 for each).
Results
Independent samples t-tests were conducted to compare the groups’
chronological age, their performance on cognitive and language tests and their social
responsiveness measured via parental report (see Table 2 for averaged scores and
Appendix A for individual scores). The groups were similar in chronological age, p =
.76, and they showed similar performances in language skills, p = .08, and visual
reception, p = .24. As expected, they differed in social responsiveness, with autistic
children being less socially responsive than their control counterparts, p = .01. The
majority (75%) of autistic children (M = 74T) scored at or above the threshold of 59T,
whereas none of the typically-developing children (M = 45T) passed this threshold.
To compare the groups’ answers in the seeing-task, generalized linear mixed
models (GLMMs) for repeated measures were run. This model featured a binomial
error structure and a logit link function. The full model included group (autism vs.
control) as between-subjects factor, task (Human vs. Object) and condition (Occluded
vs. Unoccluded) as repeated-measurement variables, and children’s negations as
dependent variable. Figure 13 shows the mean proportion of negations broke down by
group, task, and condition.
Table 2. Sample characteristics broken down by group.

Autism (n = 12; 8 male)

Control (n = 36; 32 male)

Mean SD Range

Mean SD Range
RECIPROCAL ENGAGEMENT 55
Age in Months 52 6.1 36-60

52 5.4 34-60
Cognitive (MSEL
*
) 131.4 27 94-172

152.8 26.6 81-204
Language 43.6 9.6 30-65

51.7 10.7 30-73
Visual Rec. 44.3 12.9 20-65

49.3 10.6 21-69
Social (SRS
**
) 69 26.4 49-145

45 5.7 36-58
*MSEL denotes the Mullen Scales of Early Learning and is represented here as a composite score of the
Language (Receptive Language, Expressive Language) and Visual Reception subscales. **SRS denotes
the Social Responsiveness Scale; scores above 59T indicate social impairment.

Figure 13. Mean proportion of children’s negative responses (with standard errors of the mean as error
bars) broken down by group, task, and condition. Note. * and ** indicate statistical significance at p < .05
and p < .01, respectively

An initial analysis was run to test for effects of gender, cognitive measures
(language and visual reception scores), and parental report of social responsiveness.
No such effects were found, all p’s > .2, so these variables were removed from further
analysis. As expected, there was a strong effect of condition, p < .01, OR = 144.13
[95% CI = 38.30 – 542.33], with children giving many more negative answers when
humans and objects were occluded than when they were not occluded. The largeness
RECIPROCAL ENGAGEMENT 56
of this effect results from the low variability of answers (2% negations) in the No
Occlusion Condition. See Appendix B for a breakdown of children’s responses. As
predicted, there was also a task effect, with children giving more negative answers to
questions about humans compared to objects, p = .01, OR = 2.34 (95% CI = 1.25 –
4.39). Counter our prediction, there was no group effect, p = .87, OR = 1.11 (95% CI =
.31 – 3.95): ASD and TD children were overall equally likely to negate the questions.
A subsequent analysis with interaction terms (group, task, condition) revealed a
significant 3-way interaction for group × task × condition, p < .01. The analysis was
then stratified by group to locate the source of the task × condition interaction. For the
ASD group, there was a main effect of condition, p < .01, OR = 49.05 (95% CI = 4.96 –
484.64) and of task, p = .03, OR = 3.40 (95% CI = 1.14 – 10.16), and a condition × task
interaction, p < .01, OR = 5.32 (95% CI = 1.41 – 2.01). Children in this group gave
more negative responses when a human’s eyes were occluded than both, when her
eyes were not occluded, and when an object’s features were occluded. For the TD
group, there was an effect of condition, p < .01, OR = 256.16 (95% CI = 56.28 –
1166.03), but the effect of task failed to reach significance, p = .07. There was no
interaction between these factors, p = .40. TD children gave more negative answers in
response to occluded humans or objects compared to unoccluded humans or objects,
but they did not show a greater number of negations in response to occluded humans
than occluded objects. The absence of a significant difference between children’s
negations in response to occluded humans compared to objects is inconsistent with our
prediction and with previous findings (e.g., Moll et al., 2015). This null result will be
discussed below.
RECIPROCAL ENGAGEMENT 57
Discussion
The current study lends further support to the phenomenon of young children’s
mutuality demand for gaze. In line with previous work (e.g., Flavell et al., 1980; Moll et
al., 2015; Russell et al., 2012), a large number of 3- and 4-year-olds said that they
cannot see a person whose eyes were covered. Surprisingly, TD children almost
equally denied that they could see an object whose characteristic features were
covered. Thus, they failed to discriminate between humans and objects. The most
likely reason for this null result is the conservative nature of the object task. Perhaps
covering central parts of objects obscures both the function and identity of these
objects. After all, a clock is not really a clock without hands, and a house is just a box
without windows. This task was adapted from Moll et al.’s (2015), who found that most
children affirm seeing the objects with the parts covered. One possible explanation for
the discrepancy between Moll et al.’s (2015) and the present study might be in their
inclusion of a third object (a toy truck whose headlights were occluded). Children may
have been less inclined to negate seeing a truck with occluded headlights. Relative to
the hands of a clock or the windows of a house, the headlights of a truck are not
centrally located.
Importantly, the current study assessed how preschoolers with autism respond to
questions about seeing others in the absence of mutual gaze. It was predicted that they
will affirm seeing a person whose eyes are covered. This prediction was informed by
well-documented findings suggesting that autistic children have a limited capacity to
relate to others as second persons, as shown by their deficits in imitation (Dapretto,
Davies, Pfeifer, Scott, Sigman, Bookheimer, & Iacoboni, 2006; Williams, Whiten,
RECIPROCAL ENGAGEMENT 58
Suddendorf, & Perrett, 2001), role reversal (Carpenter et al., 2005; Lee, Hobson, &
Chiat, 1994), participation in social games (Jordan, 2003), discourse (Capps, Kehres, &
Sigman, 1998; Tager-Flusberg, 1999), and other reciprocal activities. On the whole,
their behavior suggests that they tend to view others as third, not second, persons.
However, against this view and against our prediction, the majority of ASD children in
our study denied that they could see someone whose eyes were covered. Their
negative answers descriptively outnumbered those of TD peers. This lack of apparent
difference between our groups prompted important alternative explanations of our data.
In what follows, three different interpretations of this unexpected finding will be offered,
and ultimately the final interpretation will be supported.
First, let us now consider the possibility that our task was not necessarily second
personal. As such, it is entirely possible that our seeing task is not an index of
bidirectional engagement. Perhaps children do not really ‘demand’ mutual gaze at all,
but rather tend to negate the visibility of any object of perception (human or not) that is
partially covered. The presentation of partially covered persons or objects is a
perceptually salient act, especially when it is preceded or followed by the presentation
of persons or objects in plain view. Thus, it is possible that children are not necessarily
attending to the question at-hand, but rather taking an “all-or-nothing” approach in their
judgments (i.e., if something is covered, even a little bit, then I cannot see it). These
possibilities can be addressed by looking to prior experiments (e.g., Moll et al., 2015;
Moll & Khalulyan, 2016) that discredit conservative interpretations. Moll et al. (2015)
found that three-year-olds were more likely to deny an agent’s visibility when asked if
they can see but not look at agents whose eyes were covered, indicating that children
RECIPROCAL ENGAGEMENT 59
are sensitive to the semantics of the question. Importantly, children equally affirmed
seeing and looking at partially covered objects (e.g., a truck with its headlights covered).
Moreover, children’s negations of another in the absence of eye contact was shown not
to stem from a pragmatic misunderstanding of the question “Can you see X?” (Moll &
Khalulyan, 2016; ‘Study 2’). As such, the subject/object parts of speech were flipped
(e.g., Can X see you?) and children were asked to report whether the agent whose,
e.g., eyes were covered could see them. Children correctly negated (near ceiling) that
another person with compromised ability could perceive them. Collectively, these
above-mentioned studies uphold our current mutual gaze task as second personal.
A second plausible explanation that might account for our findings concerns the
‘authenticity’ of our autism sample. Along these lines, it is likely that the children in the
ASD sample were ‘false positives.’ In other words, they were somehow misdiagnosed
with autism when they were just typically-developing or even developmentally delayed.
Such claims might be valid, but they can be concretely addressed with multiple sources
of diagnostic verification. Among these sources include clinical (ADOS, DSM-V) and
non-clinical measures (parental report on social skills, as measured by the SRS-2). A
more credible account is that because our sample was on the higher functioning end of
the spectrum, their social communication skills and intelligence presented with levels
higher than those with more severe forms of autism (e.g., Kanner’s “classic” autism).
Our rationale to recruit high functioning children and not those with more severe forms
of autism was to ensure overall cooperation, a willingness to respond to experimenters’
queries, and sustained attention during test. Thus, our findings should be interpreted as
specific to a particular subset of autism, namely those in the high functioning bracket.
RECIPROCAL ENGAGEMENT 60
For these reasons, the third and final interpretation accepts the seeing task as a
measure of children’s inclination for second-personal relations with others. What it
instead rejects is the idea that ASD children are categorically incapable of entering into
such relations. Instead, it is argued here that ASD children have the necessary
resources for second personhood under specific social conditions (see Kinsbourne &
Helt, 2011, pp. 45-47 for “intact, but underused” attentional resources). Although ASD
children are deficient in spontaneously directing their attention to or sharing things about
the world with others (e.g., Dawson, Meltzoff, Osterling, Rinaldi, & Brown, 1998; Mundy
& Newell, 2007), children with high-functioning ASD have the capacity to respond to
others’ bids for engagement (Leekam & Moore, 2001). Specifically, it is argued here
that when their attention is called upon that children with ASD can engage with others in
second-personal ways. Another factor concerns ASD individuals’ incongruous
performance across explicit and implicit measures of perspective taking. Those with
high-functioning ASD are more adept at verbally judging than at behaviorally registering
a perspective different than their own (Senju, 2012). Thus, whether ASD children will
relate to others as second persons depends on their level of functioning and,
importantly, on the degree to which the interaction is appropriately structured.
Despite well-documented joint attention deficits in autism (e.g., Mundy, Sigman,
& Kasari, 1990), there is evidence that children with ASD are able to respond to bids for
joint attention (i.e., ability to follow the direction of gaze and gestures of others) during
interaction with others (Naber, Bakermans-Kranenburg, Van IJzendoorn, Dietz, Van
Daalen, Swinkels, … & Van Engeland, 2008). On the other hand, children with autism
exhibit marked challenges with initiating episodes of joint attention (i.e., ability to use
RECIPROCAL ENGAGEMENT 61
gaze direction and gestures to direct another’s attention toward interests). On this view,
our seeing task captured ASD children’s capacity for responding to bids for joint
attention (i.e., gaze or point following), a skill that becomes more typical with
development (Mundy & Jarrold, 2010). Specifically, our seeing task was a face-to-face
interaction, and the experimenter used ostensive, explicit cues (e.g., gaze, pointing,
emphasizing personal pronouns such as ‘you’) to get the child to attend to her. If she
happened to look away or provided an ambiguous response, the experimenter
redirected the child’s attention toward the adult with occluded eyes. It is important to
note that all ASD children provided responses on all trials of our task, an outcome that
may have been attenuated if the task called for children’s spontaneous bids for
interaction.
Along these lines, Hurwitz and Watson (2016) found that a subgroup of children
with ASD have the capacity for appropriate response to joint attention bids. When the
group was broken into high and low nonverbal ability (as measured by the Visual
Reception subscale of the MSEL), it was observed that children with higher nonverbal
ability all demonstrated an appropriate response to another’s bid for joint attention.
However, compared to developmentally delayed children, ASD children showed lower
rates of joint attention. The authors argue that the reduced levels of joint attention might
result from autistic children’s ‘sticky attention’ or a difficulty to disengage and shift
attention from one object or feature to another (Landry & Bryson, 2004). One critical
strategy for recruiting the children’s attention is to structure the interaction such that the
adult offers, if necessary, bids for joint attention. Although children with ASD produced
lower rates of response to joint attention than a group of developmentally delayed
RECIPROCAL ENGAGEMENT 62
children, they nonetheless were able to respond to bids for joint attention once
engaged.
Another important feature of successful reciprocal interaction is the ability of
individuals to represent another’s view of the world as distinct from their own. There
appears to be a dichotomy in terms of explicit and implicit perspective taking in autism,
the first of which is impaired, but later-developing and the second of which remains
absent throughout development. Despite claims that autistic persons suffer a general
impairment of theory of mind (Baron-Cohen, 1995), it has been shown that individuals
with autism with high verbal ability perform well on explicit measures of theory of mind—
i.e., on measures that ask participants to predict or at least explicitly formulate what an
agent will do or what his beliefs or expectations are (Happé, 1995; Senju, 2012;
Callenmark, Kjellin, Rönnqvist, & Bölte, 2014). Those with ASD exhibit challenges in
more nuanced social settings (e.g., inferring the mental states of others from pictures of
eyes; Baron-‐Cohen, Jolliffe, Mortimore, & Roberston, 1997), and show a startling
absence of spontaneous false belief attribution, whereby they are unable to produce
anticipatory behavior on the basis of another’s false view of reality (Senju, Southgate,
White, & Frith, 2009; Senju, 2012).
In line with these findings, the ASD children in our seeing task explicitly judged
that they could not see the other—presumably on the grounds that the other was not
able to see them. Even though this judgment is false by adult standards, it testifies that
the child considered the other person’s point of view. Moreover, all but one ASD child
passed our negation test, which required children to correctly negate a mislabeled
picture. They exhibited the ability to explicitly correct another agent’s (i.e., a
RECIPROCAL ENGAGEMENT 63
misinformed puppet’s) incorrect knowledge about the world, oftentimes accompanying
their firm negation with the correct label (“No, That’s not X, that’s a(n) Y”). This would
likely not be the case had our task measured an implicit form of perspective taking, such
as anticipatory eye or head movements.
A limitation of the current study concerns the small sample size of the autism
group. However, given the narrowly confined eligibility criteria (i.e., 3- to 4-years of age,
‘high’ functioning,’ verbal, English-speaking) for participation, it was a major challenge to
identify, recruit, and test a larger number of children for the autism group. Additionally,
the autistic children in this study are virtually ‘typical’ with respect to intelligence and
verbal ability. They make up the higher functioning end of the autism spectrum and are
thus not representative of autism as a whole. The findings stemming from this study
can therefore only be made in reference to high-functioning autistic children. The
rationale for precluding more severe cases of autism was to ensure an understanding of
and a general willingness to respond to test questions. On the other hand, the robust
one-to-one matching procedures of the current study is a major strength. For each child
with ASD, three TD children were selectively matched on the basis of age and gender.
The groups were then checked to be similar on overall cognitive skill, and to differ on
social responsiveness.
To conclude, our study suggests that young children with “high-functioning” ASD
are capable of entering into second personal relations with others. Although they are
less likely than typically-developing children to initiate second personal engagement and
are impaired in their ability to spontaneously take into account others’ viewpoints, they
nevertheless are able to relate to others as second persons. Our findings might inform
RECIPROCAL ENGAGEMENT 64
early intervention programs that target autistic children’s sense of interpersonal
connectedness.

RECIPROCAL ENGAGEMENT 65
Chapter 5: Overview, Theoretical Implications, and Future Recommendations
The overarching aim of this thesis was to present a cohesive account of
children’s mutuality demand for reciprocal engagement. Prior findings from a small
cohort of studies converge on a single phenomenon, which is that preschoolers find
themselves unable to see another person when eye contact is interrupted, either by
means of eye occlusion (Moll et al., 2015; Russell et al., 2012) or by means of change
in gaze direction from direct to averted (Russell et al., 2012) Importantly, children were
able to distinguish persons and bodies, as they often denied seeing another person
whose eyes were covered, but affirmed her when her, e.g., legs were covered
(McGuigan & Doherty, 2006). This suggested that for young children seeing is a
bidirectional act. Much like as adults, we feel we cannot get through to another person
wearing opaque sunglasses, children express by way of their negations a limited sense
of their capacity for second-person engagement when sensory and communication
channels do not flow both ways.
The studies in this thesis made a significant contribution to our understanding of
the scope of children’s mutuality demand. The extant literature had exclusively focused
on young children’s insistence on mutual gaze (see Flavell et al., 1980; Russell et al.,
2012; Moll et al., 2015; McGuigan, 2009; McGuigan & Doherty, 2006; Bridges &
Rowles, 1985). Study 1 revealed that children’s strong inclination toward bidirectional
interaction is not unique to gaze, but also holds for vocal communication—even though
gaze is the primary sense modality for which children insist on mutuality. In this sense,
children adopt the perspective of the other despite the faultiness in their judgments that
they cannot see others whose eyes are covered. They bypass their own first-person
RECIPROCAL ENGAGEMENT 66
perspective, basing their judgments on whether the person before them can or cannot
perceive and/or address the world. Study 2 took the main findings a step further by
outlining, testing, and ruling out alternative hypotheses that could have possibly
accounted for the data. This study provided evidence against the notion that children’s
negations reflect some sort of confusion or misunderstanding. The data suggest that
children do not confuse the subject and object parts of speech when another person
with covered eyes asks them “Can you see me?”, and that they do not respond
indiscriminately, but rather correctly map the verb (e.g., see) to the respective sense
modality (e.g., eyes). Further, Study 3 found that the mutuality demand is not a
phenomenon that is specific to typically-developing preschoolers. We now know that
high-functioning autistic children also judge that they cannot see others when mutual
gaze is breached, and take this as an indication that they can engage with others in
second-personal ways. Although both typically-developing and autistic children say they
are unable to see others in the absence of mutual gaze, it is plausible that they each
took a different route to arrive at similar negations. Perhaps autistic children’s
responses reflect something specific about our seeing-task that facilitates ‘normal’
performance.
Relations Between Children’s Hiding and Mutuality Demand
Future investigations should unpack children’s hiding strategies for agents, e.g.,
dolls and humans, alongside their demand for mutual gaze. Is there a common thread
that unites children’s hiding behavior and their tendency to say that they cannot see
others in the absence of mutual gaze? Young children tend to hide from other people in
a curious manner: they cover their eyes or head, as opposed to their entire body. On
RECIPROCAL ENGAGEMENT 67
the surface, these rather ineffective and poor hiding strategies might be interpreted in
favor of egocentrism (Piaget & Inhelder, 1956), or the notion that young children are
unable to distinguish their own viewpoint from that of another person. What is
considered egocentric in terms of their hiding behavior is that, by merely covering their
eyes, children conflate their own lack of vision with that of the person in front of them,
i.e., “If I cannot see, then you cannot see; therefore, I am hiding.” However, children’s
mutuality demand for gaze suggests that they are not egocentric at all, as they are able
to consider the other’s visual perspective. If they were truly egocentric, they would only
say that when their own eyes are covered that others cannot see them. But, they negate
the inverse as well: they say that they cannot see others whose eyes are covered. Of
course, the kind of perspective-taking that plays out in children’s mutuality demand for
gaze is faulty, as children should not need to go through the other to situate their own
first-personal perspective. What is important is that they are attempting to integrate their
own perspective with that of the other, even when it is completely unnecessary to do so.
Given children’s tendency to hide by covering their eyes, or deem that it is
acceptable for agents to hide by covering their eyes (e.g., Russell et al., 2012), very
little, if anything, is known about the possible relationship between children’s hiding and
their mutuality demand. It is likely that those children with a stronger demand for mutual
gaze will also be the ones to hide themselves and others by covering their eyes. Other
important areas of inquiry concern unpacking children’s negations. How do they
respond when directly asked to explain why it is that they cannot see the other person?
Children’s verbal explanations will help us to better understand how they rationalize or
arrive at their seemingly erroneous perceptual judgments. Importantly, their curious
RECIPROCAL ENGAGEMENT 68
hiding behavior might be better explained by a desire for mutuality, and not by
egocentrism, if their explanations reference something in the way of their limited
capacity to engage with the other collectively. For example, when asked why she can’t
see the other whose eyes are covered, the child might say, “Because our eyes don’t
meet,” or “Because we can’t see each other” as opposed to “Because her eyes are
covered / I can’t see her eyes.” The first two types of responses, but not the last two,
would provide a reasonable basis for interpreting their hiding behavior as being driven
by mutuality—if children’s explanations hint at an understanding that seeing is a
bidirectional act.
Does Shared Attention Account for Children’s Insistence on Mutuality?
Children’s denials of seeing others in the absence of mutual regard have been
deemed a developmental ‘growth error’ (Russell et al., 2012). It is an error because
children erroneously bypass their own first-personal perspective in their judgments
regarding others’ visibility. In doing so, they unnecessarily adopt the second-personal
perspective as their own, i.e., “I cannot see you because you cannot see me.”
Furthermore, it is a growth error because of its temporal confinement to the preschool
years. By the age of five years or so, children correct their mistake and begin to provide
adult-like answers, acknowledging that others who do not simultaneously see them can
be seen. In order to understand which features of the social environment are
responsible for faulty perspective-taking, it is important to consider the larger social-
cognitive landscape, and specific mechanisms that contribute to the development of the
above-mentioned growth error.
RECIPROCAL ENGAGEMENT 69
Moreover, Russell et al. (2012) postulated that children’s insistence on reciprocal
gaze might reflect an “over application” of the principles of shared attention, a social-
cognitive milestone that allows children to coordinate their gaze with that of another
toward an external object or event. The phenomenon of shared attention is so
immersive in the early years (Trevarthen, 1979) that children’s conceptions of the self
and other become intertwined (Macmurray, 1961). This “self in relation to the other”
undergoes a rather prolonged development before children are able to grasp that there
can be a divergence of minds, which includes visual perspectives as well as epistemic
and affective states. In establishing a basis for shared attention, there must be a
synchronicity of attention between two people. At times, this attention is directed toward
an external entity in the world. Without this shared attentional space, there cannot be a
“meeting of minds”—or an awareness between two agents that something in the world
can be co-perceived. Under this interpretation, we can see how children’s experience
with shared attention can influence their conceptualization of reciprocal engagement.
Shared attention via mutual gaze is so influential in allowing children to connect and
engage with others in socially meaningful ways. Thus, children’s mutuality demand can
be viewed as a normal offshoot of earlier milestones in the context of typical
development. This theoretical relation between children’s experience with shared
attention and their subsequent demand for mutual engagement ought to be empirically
tested. The prediction is that there will be a positive correlation between children’s
shared attention and their adherence to the mutuality demand.
How and Why Do Children Phase Out of the Mutuality Demand?
RECIPROCAL ENGAGEMENT 70
Lastly, future work should focus on delineating the conditions under which the
mutuality demand phases out around five years of age. Does this process coincide with
other social-cognitive developments such as passing the standard false belief task?
Perhaps children adopt a more cognitively mature or adult-like stance for seeing others
outside mutual gaze right around the time they start to fully represent other minds.
To better understand why children undergo a developmental growth error with respect
to conceptualizing reciprocal interactions, we must also consider the larger context of
perspective-taking during this window of time. Interestingly, the gradual phasing out of
the mutuality demand roughly coincides with the development of a full-fledged theory of
mind around four or five years of age. In contrast to claims that, prior to having a theory
of mind, children are largely egocentric (e.g., Fishbein, Lewis, & Keiffer, 1972), the
mutuality demand evidences that this is not the case. Despite its erroneous logic,
children’s demand for reciprocal engagement represents their attempt to adopt the
perspective of another person. This approach is erroneous because it comes at the cost
of children’s own perspective, which is the source of accurate and veridical first-person
judgments. In other paradigms that require perspective-taking, children’s “default mode”
of reporting their own perspective suggests that perhaps they are considering the “ideal”
perspective in any given situation. With standard false belief tasks, children who have
not yet reached a mature theory of mind respond through their own perspective not
because of an egocentric bias—but rather because their own perspective in a given
situation is representative of reality and thus optimal or ideal (e.g., “You should look
here because that’s where your X is currently”). Given the concurrent emergence
(around 4.5-5 years of age) of adult-like responding and theory of mind tasks, it is
RECIPROCAL ENGAGEMENT 71
reasonable to postulate that a shared social-cognitive mechanism gives rise to the
development of perspective-taking. This honing of perspective-taking progresses
gradually, before children can acquire a fuller understanding of other minds.

RECIPROCAL ENGAGEMENT 72
References
American Psychiatric Association. (2013). Diagnostic and statistical manual on mental
disorders (DSM-V). 5
th
ed. Washington, DC: American Psychiatric Publishing.
Bailey, A. J., Braeutigam, S., Jousmäki, V., & Swithenby, S. J. (2005). Abnormal
activation of face processing systems at early and intermediate latency in
individuals with autism spectrum disorder: A magnetoencephalographic
study. European Journal of Neuroscience, 21(9), 2575-2585.
Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind.
Cambridge, MA: MIT Press.
Baron-‐Cohen, S., Jolliffe, T., Mortimore, C., & Roberston, M. (1997). Another advanced
test of theory of mind: Evidence from very high functioning adults with autism or
Asperger syndrome. Journal of Child Psychology and Psychiatry, 38(7), 813-822.
Batki, A., Baron-Cohen, S., Wheelwright, S., Connellan, J., & Ahluwalia, J. (2000). Is
there an innate gaze module? Evidence from human neonates. Infant Behavior
and Development, 23(2), 223-229.
Bridges, A., & Rowles, J. (1985). Young children's projective abilities: What can a
monster see? Educational Psychology, 5(3-4), 251-266.
Callenmark, B., Kjellin, L., Rönnqvist, L., & Bölte, S. (2014). Explicit versus implicit
social cognition testing in autism spectrum disorder. Autism, 18(6), 684-693.
Capps, L., Kehres, J., & Sigman, M. (1998). Conversational abilities among children
with autism and children with developmental delays. Autism, 2(4), 325-344.
RECIPROCAL ENGAGEMENT 73
Carpenter, M., Tomasello, M., & Striano, T. (2005). Role reversal imitation and
language in typically developing infants and children with autism. Infancy, 8(3),
253-278.
Charman, T., Swettenham, J., Baron-Cohen, S., Cox, A., Baird, G., & Drew, A. (1997).
Infants with autism: An investigation of empathy, pretend play, joint attention, and
imitation. Developmental Psychology, 33(5), 781-789.
Condon, W. S., & Sander, L. W. (1974). Neonate movement is synchronized with adult
speech: Interactional participation and language acquisition. Science, 183(4120),
99-101.
Constantino, J. N. & Gruber, C. P. (2007). Social responsiveness scale (SRS). Los
Angeles, CA: Western Psychological Services.
Dalton, K. M., Nacewicz, B. M., Johnstone, T., Schaefer, H. S., Gernsbacher, M. A.,
Goldsmith, H. H., … & Davidson, R. J. (2005). Gaze fixation and the neural
circuitry of face processing in autism. Nature Neuroscience, 8(4), 519-526.
Dapretto, M., Davies, M. S., Pfeifer, J. H., Scott, A. A, Sigman, M., Bookheimer, S. Y., &
Iacoboni, M. (2006). Understanding emotions in others: Mirror neuron
dysfunction in children with autism spectrum disorders. Nature Neuroscience,
9(1), 28-30.
Dawson, G., Meltzoff, A. N, Osterling, J., Rinaldi, J., & Brown, E. (1998). Children with
autism fail to orient to naturally occurring social stimuli. Journal of Autism and
Developmental Disorders, 28(6), 479-485.
RECIPROCAL ENGAGEMENT 74
Farroni, T., Csibra, G., Simion, F., & Johnson, M. H. (2002). Eye contact detection in
humans from birth. Proceedings of the National Academy of Sciences, 99(14),
9602-9605.
Fishbein, H. D., Lewis, S., & Keiffer, K. (1972). Children’s understanding of spatial
relations: Coordinating of perspectives. Developmental Psychology, 7, 21-33.
Flavell, J. H., Shipstead, S. G., & Croft, K. (1980). What young children think they see
when their eyes are closed. Cognition, 8(4), 369-387.
Frith, U. (2001). Mindblindness and the brain in autism. Neuron, 32(6), 969-979.

Gergely, G., & Watson, J. S. (1996). The social biofeedback model of parental affect
mirroring. The International Journal of Psycho-Analysis, 77, 1181-1212.
Gergely, G., & Watson, J. S. (1999). Early socio-emotional development: Contingency
perception and the social-biofeedback model. Early Social Cognition:
Understanding Others in the First Months of Life, 60, 101-136.
Gobel, M. S., Kim, H. S., & Richardson, D. C. (2015). The dual function of social
gaze. Cognition, 136, 359-364.
Hale, C. M., & Tager-Flusberg, H. (2005). Social communication in children with autism:
The relationship between theory of mind and discourse development.
Autism, 9(2), 157-178.
Happé, F. (1995). The role of age and verbal ability in the theory of mind task
performance of subjects with autism. Child Development, 66(3), 843-855.
Happé, F. (2005). The weak central coherence account of autism. Handbook of Autism
and Pervasive Developmental Disorders, 1(3), 640-649.
RECIPROCAL ENGAGEMENT 75
Happé, F., & Frith, U. (2006). The weak coherence account: detail-focused cognitive
style in autism spectrum disorders. Journal of Autism and Developmental
Disorders, 36(1), 5-25.
Hobson, R. P. (1993). Autism and the development of mind. Hove: Erlbaum.
Hobson, R. P. (2002). The cradle of thought. London: Macmillan.
Hobson, R. P., Lee, A., & Hobson, J. A. (2010). Personal pronouns and communicative
engagement in autism. Journal of Autism and Developmental Disorders, 40(6),
653-664.
Hobson, R. P., & Meyer, J. A. (2005). Foundations for self and other: A study in
autism. Developmental Science, 8(6), 481-491.
Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian
Journal of Statistics, 65-70.
Hurwitz, S., & Watson, L. R. (2016). Joint attention revisited: Finding strengths among
children with autism. Autism, 20(5), 538-550.
Jones, W., Carr, K., & Klin, A. (2008). Absence of preferential looking to the eyes of
approaching adults predicts level of social disability in 2-year-old toddlers with
autism spectrum disorder. Archives of General Psychiatry, 65(8), 946-954.
Jordan, R. (2003). Social play and autistic spectrum disorders: A perspective on theory,
implications and educational approaches. Autism, 7(4), 347-360.
Kaartinen, M., Puura, K., Mäkelä, T., Rannisto, M., Lemponen, R., Helminen, M., …
Hietanen, J. K. (2012). Autonomic arousal to direct gaze correlates with social
impairments among children with ASD. Journal of Autism and Developmental
Disorders, 42(9), 1917-1927.
RECIPROCAL ENGAGEMENT 76
Kinsbourne, M., & Helt, M. (2011). Social entrainment of typically developing and
autistic children. In D. Fein (Ed.), The neuropsychology of autism (pp.339-365).
Oxford: Oxford University Press.
Kleinke, C. L. (1986). Gaze and eye contact: A research review. Psychological Bulletin,
100(1), 78-100.
Klin, A., Jones, W., Schultz, R., Volkmar, F., & Cohen, D. (2002). Visual fixation
patterns during viewing of naturalistic social situations as predictors of social
competence in individuals with autism. Archives of General Psychiatry, 59(9),
809-816.
Klin, A., Sparrow, S. S., De Bildt, A., Cicchetti, D. V., Cohen, D. J., & Volkmar, F.
(1999). A normed study of face recognition in autism and related
disorders. Journal of Autism and Developmental Disorders, 29(6), 499-508.
Kobayashi, H., & Kohshima, S. (1997). Unique morphology of the human
eye. Nature, 387(6635), 767.
Kobayashi, H., & Koshima, S. (2001). Unique morphology of the human eye and its
adaptive meaning. Comparative studies on external morphology of the
primate eye. Journal of Human Evolution, 40(5), 419-435.
Kylliäinen, A., & Hietanen, J. K. (2006). Skin conductance responses to another
person’s gaze in children with autism. Journal of Autism and Developmental
Disorders, 36(4), 517-525.
Landry, R., & Bryson, S. E. (2004). Impaired disengagement of attention in young
children with autism. Journal of Child Psychology and Psychiatry, 45(6), 1115-
1122.
RECIPROCAL ENGAGEMENT 77
Lee, A., Hobson, R. P. & Chiat, S. (1994). I, you, me, and autism: An experimental
study. Journal of Autism and Developmental Disorders, 24(2), 155-176.
Leekam, S. R., Moore, C. (2001). The development of attention and joint attention in
children with autism. In T. Charman & J. A. Burack (Eds.), The development of
autism: Perspectives from theory and research (pp. 105–129), Mahwah, NJ:
Lawrence Erlbaum Associates.
Loveland, K. A. & Landry, S. H. (1986). Joint attention and language in autism and
developmental language delay. Journal of Autism and Developmental
Disorders, 16(3), 335-349.
Macmurray, J. (1961). Persons in relation. Amherst, NY: Humanity Books.
McGuigan, N. (2009). Does the direction in which a figure is looking influence whether it
is visible? Journal of Genetic Psychology, 170(3), 227-233.
McGuigan, N. & Doherty, M. J. (2006). Head and shoulders, knees and toes: Which
parts of the body are necessary to be seen? British Journal of Developmental
Psychology, 24(4), 727 732.
McMyler, B. (2011). Testimony, trust, and authority. New York: Oxford University Press.
Meltzoff, A. N. & Moore, M. K. (1977). Imitation of facial and manual gestures by human
neonates. Science, 198(4312), 75-78.
Moll, H., Arellano, D., Guzman, A., Cordova, X., & Madrigal, J. A. (2015). Preschoolers’
mutualistic conception of seeing is related to their knowledge of the pronoun
‘each other.’ Journal of Experimental Child Psychology, 131, 170-185.
Moll, H., Carpenter, M., & Tomasello, M. (2007). Fourteen-‐month-‐olds know what others
experience only in joint engagement. Developmental Science, 10(6), 826-835.
RECIPROCAL ENGAGEMENT 78
Moll, H., & Khalulyan, A. (2016). “Not see, not hear, not speak”: Preschoolers think they
cannot perceive or address others without reciprocity. Journal of Cognition and
Development, 18(1), 152-162.
Mullen, E. M. (1995). Mullen scales of early learning (MSEL). Circle Pines: American
Guidance Service.
Mundy, P., & Jarrold, W. (2010). Infant joint attention, neural networks and social
cognition. Neural Networks, 23(8-9), 985-997.
Mundy, P., & Newell, L. (2007). Attention, joint attention, and social cognition. Current
Directions in Psychological Science, 16(5), 269-274.
Mundy, P., Kasari, C., & Sigman, M. (1992). Nonverbal communication, affective
sharing, and intersubjectivity. Infant Behavior and Development, 15(3), 377-381.
Mundy, P., Sigman, M., & Kasari, C. (1994). Joint attention, developmental level, and
symptom presentation in autism. Development and Psychopathology, 6(3), 389-
401.
Mundy, P., Sigman, M., & Kasari, C. (1990). A longitudinal study of joint attention and
language development in autistic children. Journal of Autism and Developmental
Disorders, 20(1), 115-128.
Naber, F. B., Bakermans-Kranenburg, M. J., Van IJzendoorn, M. H., Dietz, C., Van
Daalen, E., Swinkels, S. H., … & Van Engeland, H. (2008). Joint attention
development in toddlers with autism. European Child & Adolescent Psychiatry,
17(3), 143-152.
Oehlert, G. W. (1992). A note on the delta method. The American Statistician, 46(1), 27-
29.
RECIPROCAL ENGAGEMENT 79
Ozonoff, S., Iosif, A. M., Baguio, F., Cook, I. C., Hill, M. M., Hutman, T., … & Steinfeld,
M. B. (2010). A prospective study of the emergence of early behavioral signs of
autism. Journal of the American Academy of Child and Adolescent
Psychiatry, 49(3), 256-266.
Pelphrey, K. A., Sasson, N. J., Reznick, J. S., Paul, G., Goldman, B. D., & Piven, J.
(2002) Visual scanning of faces in autism. Journal of Autism and Developmental
Disorders, 32(4), 249-261.
Piaget, J., & Inhelder, B. (1956). The child’s conception of space. London: Routledge &
Kegan Paul.
Prizant, B. M. & Rydell, P. J. (1993). Assessment and intervention considerations for
unconventional verbal behavior. Communicative Alternatives to Challenging
Behavior: Integrating Functional Assessment and Intervention Strategies, 3, 263-
297.
Russell, J., Gee, B., & Bullard, C. (2012). Why do young children hide by closing their
eyes? Self-visibility and the developing concept of self. Journal of Cognition and
Development, 13(4), 550-576.
Ryle, G. (1954/2002). Dilemmas. New York: Cambridge University Press.
Schultz, R. T. (2005). Developmental deficits in social perception in autism: The role of
the amygdala and fusiform face area. International Journal of Developmental
Neuroscience, 23(2), 125-141.
Senju, A. (2012). Spontaneous theory of mind and its absence in autism spectrum
disorders. The Neuroscientist, 18(2), 108-113.
RECIPROCAL ENGAGEMENT 80
Senju, A., Maeda, M., Kikuchi, Y., Hasegawa, T., Tojo, Y., & Osanai, H. (2007).
Absence of contagious yawning in children with autism spectrum
disorder. Biology Letters, 3(6), 706-708.
Senju, A., Southgate, V., White, S., & Frith, U. (2009). Mindblind eyes: An absence of
spontaneous theory of mind in Asperger syndrome. Science, 325(5942), 883-
885.
Senju, A., Yaguchi, K., Tojo, Y., Hasegawa, T. (2003). Eye contact does not facilitate
detection in children with autism. Cognition, 89(1), B43-B51.
Small, W. (2014). Teaching and telling. Philosophical Explorations, 17(3), 372-387.
Tager-Flusberg, H. (1999). A psychological approach to understanding the social and
language impairments in autism. International Review of Psychiatry, 11(4), 325-
334.
Tager-‐Flusberg, H., & Anderson, M. (1991). The development of contingent discourse
ability in autistic children. Journal of Child Psychology and Psychiatry, 32(7),
1123-1134.
Tomasello, M. (1999). Having intentions, understanding intentions, and understanding
communicative intentions. In P. D. Zelazo, J. W. Astington, & D. R. Olson (Eds.),
Developing theories of intention (pp. 63–75). Mahwah: Erlbaum.
Tomasello, M., Carpenter, M., Hobson, R. P. (2005). The emergence of social cognition
in three young chimpanzees. Monographs of the Society for Research in Child
Development, 70(1), vii–132.
Trevarthen, C. (1979). Communication and cooperation in early infancy: A description of
RECIPROCAL ENGAGEMENT 81
primary intersubjectivity. In M. Bullowa (Ed.), Before speech: The beginning of
interpersonal communication (pp. 321-347). Cambridge: Cambridge University
Press.
Watson, J. S. (1972). Smiling, cooing, and “the game." Merrill-Palmer Quarterly of
Behavior and Development, 18(4), 323-339.
Williams, J. H., Whiten, A., Suddendorf, T., & Perrett, D. I. (2001). Imitation, mirror
neurons and autism. Neuroscience & Biobehavioral Reviews, 25(4), 287-295.
Wolfberg, P. J. & Schuler, A. L. (1999). Fostering peer interaction, imaginative play and
spontaneous language in children with autism. Child Language Teaching and
Therapy, 15(1), 41-52.
Zwaigenbaum, L., Bryson, S., Rogers, T., Roberts, W., Brian, J., & Szatmari, P. (2005).
Behavioral manifestations of autism in the first year of life. International
Journal of Developmental Neuroscience, 23(2), 143-152.

RECIPROCAL ENGAGEMENT 82
Appendix A
Mullen Scales
Participants
Gender Age (mos.) VisRec RecLang ExpLang SRS
ASD 1 F 54 52 45 65 59
TD Controls F 55 55 34 41 45

F 55 55 57 37 76

F 52 45 43 37 52
ASD 2 F 36 52 64 56 54
TD Controls F 34 67 64 56 48

F 39 50 46 42 37

F 39 42 43 46 42
ASD 3 M 56 40 39 33 64
TD Controls M 54 40 57 52 38

M 55 60 67 66 40

M 53 48 45 41 48
ASD 4 F 50 34 36 38 57
TD Controls F 52 33 41 44 49

F 53 69 70 65 42

F 50 44 56 51 53
ASD 5 M 60 35 47 48 64
TD Controls M 59 43 63 68 44

M 57 39 54 45 49

M 57 47 59 40 49
ASD 6 M 58 61 42 44 145
TD Controls M 56 34 62 52 45

M 56 60 48 39 49

M 55 58 53 54 40
ASD 7 M 52 55 43 40 68
TD Controls M 53 48 50 42 52

M 51 50 45 41 54

M 50 59 74 65 46
ASD 8 F 50 20 36 38 64
TD Controls F 52 40 70 55 49

F 51 39 40 47 39

F 47 43 47 55 41
ASD 9 M 54 42 39 38 56
TD Controls M 53 55 65 61 36

M 53 42 41 40 45

M 51 50 61 51 58
ASD 10 M 55 40 52 40 83
TD Controls M 54 59 70 65 40

M 57 55 51 43 55

M 56 48 51 52 38
ASD 11 M 52 35 32 28 49
TD Controls M 55 21 26 34 48

M 48 59 78 60 55

M 49 59 53 57 43
ASD 12 M 56 65 54 49 65
TD controls M 54 69 55 44 41

M 53 52 57 48 43

M 59 40 58 35 44

RECIPROCAL ENGAGEMENT 83
Appendix B
Number (and percentage) of children’s affirmations and negations broken down by task,
group, occluder, and condition.
Children’s Response
Yes

No

Count %

Count %
Task

Object 289 75.3

95 24.7
Human 262 68.2

122 31.8
Group

TD 414 71.9

162 28.1
ASD 137 71.4

55 28.6
Occluder

Device 271 70.6

113 29.4
Hand 280 72.9

104 27.1
Condition

No Occlusion 375 97.7

9 2.3
Occlusion 176 45.8

208 54.2

Abstract (if available)

Abstract How does mutual gaze, or eye contact, play a role in shaping young children’s notion of reciprocal engagement? Prior research shows that typically-developing children under the age of five years often negate seeing others in the absence of mutual gaze (e.g., Moll, Arellano, Guzman, Cordova, & Madrigal, 2015). Children’s negations have been interpreted to reflect a mutualistic or reciprocal notion of person-to-person relation, a ‘mutuality demand’ of sorts, i.e., “I can only see you if you also see me (and vice versa).” After around five years of age, children begin to adopt a more ‘objective’ perspective and affirm seeing others in the absence of mutual gaze. ❧ The focus of this thesis is to identify other possible conditions under which children’s mutuality demand manifests. First, little to no attention has been paid to whether young children hold a similar mutuality criterion for other modes of interaction, namely vocal communication. Much like they say they cannot see another person whose eyes are covered, will children also deny that they can hear or speak to another person whose ears or mouth area, respectively, is covered? Given the primacy of gaze, it is possible that children will only demand mutuality for gaze and no other modality. It is also possible that children will demand mutuality for any modality that allows for reciprocal engagement (i.e., gaze and vocal communication). Second, it is unknown if children with autism conceptualize persons outside of mutual gaze differently than their typically-developing peers. Given core deficits in gaze and reciprocal social interaction, it is expected that children with autism will affirm seeing others in the absence of eye contact. This work has significant implications for our understanding of the boundaries governing young children’s mutuality demand during the early formative years.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Children's ""mutuality demand"" in seeing, hearing, and speech

PDF

Considering the effects of disfluent speech on children’s sentence processing capabilities and language development

Asset Metadata

Creator Khalulyan, Allie (author)

Core Title Young children’s demand for reciprocal engagement: evidence from typically-developing and autistic preschoolers

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 08/03/2018

Defense Date 05/08/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag autism spectrum disorder,bidirectional engagement,mutual gaze,OAI-PMH Harvest,person perception,reciprocity

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Moll, Henrike (committee chair), Levin, Janet (committee member), Manis, Franklin (committee member), Mintz, Tobin (committee member)

Creator Email Khalulya@usc.edu,khalulyan89@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-49376

Unique identifier UC11669021

Identifier etd-KhalulyanA-6624.pdf (filename),usctheses-c89-49376 (legacy record id)

Legacy Identifier etd-KhalulyanA-6624.pdf

Dmrecord 49376

Document Type Dissertation

Format application/pdf (imt)

Rights Khalulyan, Allie

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA