A longitudinal study of pupillary light reflex in 6- to 24-month children

Abstract

Pupillary light reflex (PLR) is an involuntary response where the pupil size changes with luminance. Studies have shown that PLR response was altered in children with autism spectrum disorders (ASDs) and other neurological disorders. However, PLR in infants and toddlers is still understudied. We conducted a longitudinal study to investigate PLR in children of 6–24 months using a remote pupillography device. The participants are categorized into two groups. The ‘high risk’ (HR) group includes children with one or more siblings diagnosed with ASDs; whereas the ‘low risk’ (LR) group includes children without an ASD diagnosis in the family history. The participants’ PLR was measured every six months until the age of 24 months. The results indicated a significant age effect in multiple PLR parameters including resting pupil radius, minimal pupil radius, relative constriction, latency, and response time. In addition, the HR group had a significantly larger resting and minimal pupil size than the LR group. The experimental data acquired in this study revealed not only general age-related PLR changes in infants and toddlers, but also different PLRs in children with a higher risk of ASD.

Full text

1
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
A longitudinal study of pupillary
light reex in 6- to 24-month
children
Clare Kercher1, Leila Azinfar1, Dinalankara M. R. Dinalankara1,2, T. Nicole Takahashi3,
Judith H. Miles3 & Gang Yao1*
Pupillary light reex (PLR) is an involuntary response where the pupil size changes with luminance.
Studies have shown that PLR response was altered in children with autism spectrum disorders
(ASDs) and other neurological disorders. However, PLR in infants and toddlers is still understudied.
We conducted a longitudinal study to investigate PLR in children of 6–24 months using a remote
pupillography device. The participants are categorized into two groups. The ‘high risk’ (HR) group
includes children with one or more siblings diagnosed with ASDs; whereas the ‘low risk’ (LR) group
includes children without an ASD diagnosis in the family history. The participants’ PLR was measured
every six months until the age of 24 months. The results indicated a signicant age eect in multiple
PLR parameters including resting pupil radius, minimal pupil radius, relative constriction, latency, and
response time. In addition, the HR group had a signicantly larger resting and minimal pupil size than
the LR group. The experimental data acquired in this study revealed not only general age-related PLR
changes in infants and toddlers, but also dierent PLRs in children with a higher risk of ASD.
Autism Spectrum Disorders (ASDs) are complicated disorders that are marked by persistent decits in social
communication and interactions and by restricted, repetitive patterns of behavior, interests or activities1. Initially
chronicled 75 years ago2, ASDs now aect about 2.47% children and adolescents in USA alone3. Although the
etiology of ASD is still not fully understood, our understanding of this disorder has since been signicantly
improved owing to a large amount of physiological, psychological, and neurological studies. Evidence sup-
ports that the outcome in children with ASDs can be greatly improved by using early behavioral intervention4,5.
Unfortunately, most children do not receive an ASD diagnosis until aer the age of four6, although early signs
may appear as young as 12 months of age7. erefore, there is a great interest in nding eective biological mark-
ers for early screening of risk of autism and assessing responses to interventions.
Pupillary light reex (PLR) is the involuntary and nearly instantaneous pupil size change that occurs as a
response to the luminous intensity of light that falls on the retina. e pupil size is controlled by the dilator
and sphincter muscles innervated primarily by the sympathetic and parasympathetic branches of the autonomic
nervous system (ANS), respectively8. In 1961, Rubin observed that the pupils in 7 to 12 years old children with
ASD constricted slower in responses to light adaption compared to typically developing children9. Using a com-
puterized pupillography system, Fan et al. discovered that pupils of children with ASD took a greater amount
of time to respond to short (0.1 s) light stimuli and constricted less and more slowly than those with typical
development10. Similar atypical PLR responses were also reported in subsequent studies in children with ASD of
dierent ages using pupillography and eye-tracking devices11–13. In addition, studies have shown that quantitative
PLR responses were associated with sensory behaviors and autism traits14,15. e PLR’s potential for early identi-
cation of risk of autism was recently demonstrated by Nyström et al.16. ey reported that the pupil constricted
more in 9- to 10-month old infants who later received an ASD diagnosis and the amount of PLR constriction was
correlated with the severity of ASD symptoms.
Resting pupil size and PLR parameters are known to change with age. Existing literature indicates that resting
pupil size increases from infants to teenagers17–19 and then decreases with age thereaer20–22. In comparison with
resting or static pupil size, there are limited studies on age eect on PLR. Still, current evidence indicates that
1Department of Biomedical, Biological & Chemical Engineering, University of Missouri, Columbia, MO, 65211, USA.
2Department of Computer Engineering, University of Sri Jayewardenepura, Nugegoda, Sri Lanka. 3Thompson Center
for Autism and Neurodevelopmental Disorders, University of Missouri, Columbia, MO, 65211, USA. *email: YaoG@
missouri.edu
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved

2
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
PLR parameters can change with age19. Studies suggested that the age trend might be altered in association with
ASD. For example, PLR latency (the delay between stimulation onset and the beginning of pupil constriction)
decreased from 6 to 8 years in children of typical development; this trend was not apparent in age-matched chil-
dren with ASD11, suggesting that the PLR dierences between individuals with and without ASD may change
with age. Interestingly, Nyström et al. later reported that the PLR latency was shorter in young children with
high-risk of ASD23, in contrast with reports that older children with ASD had longer latency than typically devel-
oping children10,11. Exiting experimental evidence12 suggested that dierent age trends may explain apparent
inconsistencies in ASD associated atypical resting pupil sizes reported in the literature24–26.
Despite the importance of age eect on PLR, no age-dependent longitudinal study was reported in literature.
In particular, age-dependent PLR data in infants and toddlers are scarce due to the challenges in measuring PLR
in young children. is study used the recently developed remote PLR (rPLR) system to investigate the PLR
changes in children from 6 to 24 months old. is novel rPLR system is capable of imaging pupil size changes at a
high spatial resolution without the need of any restrain during the test, which makes it ideal to test PLR in young
children27. e participants were categorized into two groups based on their susceptibility to ASD. e risk of
younger siblings developing an ASD is signicantly higher if an older sibling has an ASD diagnosis28. erefore,
the ‘high risk’ (HR) group includes children with one or more siblings diagnosed with ASD. On the other hand,
the ‘low risk’ (LR) group includes children not associated with ASD in the family history. We intended to answer
the following questions: (1) whether the PLR parameters are age-dependent in the 6–24 months of age range, and
(2) whether any atypical parameters exist in the ‘high risk’ group of children.
Results
e Pervasive Developmental Disorders Screening Test-II (PDDST-II) scores were recorded as a simple screen-
ing for neurodevelopmental disorders in the participants from 12- to 24-month old. Figure1 shows the distri-
bution of the PDDST-II scores at 12-, 18-, and 24-month. In the HR group, the PDDST-II score changed from
2.05 ± 2.22 at 12-month, to 2.00 ± 2.60 at 18-month, and 1.86 ± 2.80 at 24-month. A few children (4 at 12-month,
3 at 18-month, 2 at 24-month) in the HR group had a score of 5 or above, suggesting potential developmen-
tal disorders29. In the LR group, the PDDST-II score appeared to increase slightly with age from 0.43 ± 0.65 at
12-month, to 0.68 ± 0.95 at 18-month, and 0.81 ± 0.83 at 24-month. However, none of the participants in the LR
group scored more than three.
Figure2 illustrates example pupilograms obtained from two participants in the LR and HR groups who com-
pleted all 4 tests at dierent ages from 6-month to 24-month. e pupilogram curves shown were averaged results
Figure 1. Distribution of the PDDST-II scores obtained at 12-, 18-, and 24-month in both LR and HR groups.
Figure 2. Example mean PLR curves obtained from a subject in the LR group and a subject in the HR group at
dierent ages. e errors bars indicate the standard error. e dashed lines indicated the time of the stimulation
ashes.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

3
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
from all successful trials obtained during a single PLR test. e PLR followed the typical PLR curve as those
observed in the older children. e pupil size was relatively stable before the stimulation (marked as dashed lines).
e pupil then started to constrict aer a delay (the latency period), reached a minimum, and then started to
recover back to the baseline. Overall, the amount of constriction in the HR group appeared to be slightly smaller
than the LR constriction.
Figure3 shows all extracted PLR parameters at different ages in both the HR and LR groups (see also
Supplementary Fig.S1). As a group, the base pupil radius, minimal pupil radius, and relative constriction all
increased with age; whereas the latency, response time, and constriction time showed a decreasing trend with age.
However, there were considerable variations amongsubjects, which justied the use of a random intercept in the
linear mixed-eects model(LMM) analysis. In addition, the HR group appeared to have a larger pupil than the
LR group. Both the base and minimal pupil radii appeared to be larger in the males. No clear group or sex eect
was observed in the other four parameters. e data from the two HR participants who received diagnoses at
the end of this study were labeled using symbols in Fig.3. e triangle symbol represented the one with an ASD
diagnosis and the circle represented the other with a diagnosis of global developmental delay.
e above observations were examined using the LMM analysis. Table1 shows the estimations of xed eects
and the corresponding 95% condence intervals (CI). e group (HR vs LR) had a signicant eect in three PLR
parameters: base radius (F = 10.02, p = 0.003), minimal radius (F = 11.62, p = 0.001), and relative constriction
(F = 5.82, p = 0.020). In comparison with the LR group, the pupils in the HR group were bigger before stimu-
lation (t = 3.17, p = 0.003), remained bigger at the maximal constriction (t = 3.41, p = 0.001); but the relative
Figure 3. e extracted PLR parameters (base radius, minimal radius, relative constriction, latency,
constriction time, and response time) in all participants at dierent age groups. e data were separated into
the high-risk (HR) and low-risk (LR) groups and males (M) and females (F) in each group. e symbols
indicated data from the two HR participants who received diagnoses at the end of this study with the triangles
representing the one diagnosed with an ASD and the circles representing the other with a diagnosis of global
developmental delay.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

4
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
constriction was smaller (t = −2.41, p = 0.020). All timing parameters (latency, response time, and constriction
time) were similar in the HR and LR groups.
e LMM analysis revealed that the sex eect was signicant only in the two pupil-size related parameters:
base radius (F = 8.41, p = 0.006) and minimal radius (F = 6.01, p = 0.019). As shown in Table1, in comparison
with the boys, the girls had smaller based pupil radius (t = −2.90, p = 0.006) and minimal pupil radius (t = −2.45,
p = 0.019).
e LMM analysis indicated that age had a signicant eect on base radius (F = 14.00, p < 0.001), minimal
radius (F = 10.52, p < 0.001), relative constriction (F = 3.74, p = 0.014), latency (F = 7.13, p < 0.001), and response
time (F = 6.34, p = 0.001). In constriction time, the LMM showed a marginal age eect (F = 2.626, p = 0.055).
e follow-up pairwise comparisons conrmed that the base radius increased signicantly with age (p < 0.05)
between any two age-group pairs except between 12-mo and 18-mo. Similarly, the minimal pupil radius increased
with age (p < 0.05) between any two age-group pairs except between 12- and 18-mo and 6- and 12-mo. e rela-
tive constriction showed an overall increasing trend with age. However, the dierence reached signicance only
between 6- and 12-mo, and between 6-mo and 24-mo. e decreasing tend in latency was signicant (p < 0.05)
between 6- and 24-mo, 12- and 24-mo, and between 18- and 24-mo. e pairwise comparison revealed that the
decreasing trend in response time was signicant between 12- and 18-mo, and between 12- and 24-mo.
Discussion
is study revealed signicant age trends in the pupil size, constriction, latency, and response time in 6- to
24-month children. e trend seen in the base pupil radius was consistent with previous reports that pupil size
increased from birth until teenage years in typically developing children17,18. e observation that male children
had slightly bigger pupil size was also in agreement with previous studies17. A close examination of the corre-
lations among the six PLR parameters (Fig.4) indicated that base pupil radius and minimal pupil radius were
highly correlated (Pearson correlation r = 0.945). erefore, the similar eects of age, sex, and group on base
radius and minimal radius can be expected.
e increase in relative constriction with age, in particular from 6-mo to 12-mo and to 24-mo, appeared to be
consistent with a previously observed trend in 2-year to 3-year old children12. e age trends in base and minimal
pupil radii were opposite to that of the constriction. Such opposite trends cannot be simply explained based on
correlation. As shown in Fig.4, the relative constriction only had a week negative correlation with base radius
(r = −0.265) and a moderate negative correlation with minimal radius (r = 0.523).
e observation of latency decreasing with age was consistent with previous results reported in 2- to 6-year
and 6- to 18-year old children of typical development11,12. Taken together, these data suggested that PLR latency
Parameter Estimate Std. Error df t Sig. 95% CI [Lower, Upper]
Based radius (mm)
Intercept 2.11 0.06 53.05 33.74 0 [1.98, 2.23]
[Group = HR]a0.21 0.07 41.36 3.17 0.003 [0.08, 0.35]
[Sex = F]b−0.19 0.07 41.30 −2.90 0.006 [−0.33, −0.06]
[Age = 6]c−0.26 0.05 90.54 −5.46 0 [−0.36, −0.17]
[Age = 12] −0.13 0.03 84.50 −3.75 0 [−0.20, −0.06]
[Age = 18] −0.02 0.03 83.80 −0.46 0.649 [−0.08, 0.05]
Min radius (mm)
Intercept 1.77 0.06 50.26 27.90 0 [1.65, 1.90]
[Group = HR]a0.23 0.07 40.48 3.41 0.001 [0.10, 0.37]
[Sex = F]b−0.17 0.07 40.52 −2.45 0.019 [−0.31, −0.03]
[Age = 6]c−0.21 0.05 87.43 −4.48 0 [−0.30, −0.12]
[Age = 12]c−0.09 0.03 82.66 −2.90 0.005 [−0.16, −0.03]
[Age = 18]c0.01 0.03 81.84 0.18 0.858 [−0.06, 0.07]
Constriction (%)
Intercept 29.32 1.69 51.43 17.40 0.000 [25.94, 32.70]
[Group = HR]a−4.38 1.81 41.40 −2.41 0.020 [−8.04, −0.71]
[Age = 6]c−4.10 1.24 88.39 −3.32 0.001 [−6.55, −1.64]
[Age = 12]c−0.91 0.87 83.61 −1.05 0.299 [−2.65, 0.82]
[Age = 18]c−1.12 0.87 82.80 −1.30 0.199 [−2.85, 0.60]
Latency (ms)
Intercept 240.62 5.17 66.50 46.51 0 [230.29, 250.95]
[Age = 6]c20.68 5.36 99.11 3.86 0 [10.05, 31.31]
[Age = 12]c15.59 3.92 86.23 3.97 0 [7.79, 23.39]
[Age = 18]c11.51 3.90 85.36 2.95 0.004 [3.75, 19.27]
Response time (ms)
Intercept 618.36 13.36 56.24 46.30 0 [591.60, 645.11]
[Age = 6]c28.06 12.09 89.35 2.32 0.023 [4.03, 52.09]
[Age = 12]c36.28 8.74 80.33 4.15 0 [18.89, 53.66]
[Age = 18]c13.58 8.60 79.33 1.58 0.118 [−3.53, 30.68]
Table 1. Estimates of xed eects obtained using the linear mixed-eects model (LMM). ae LMM model
used results from LR group as the reference. be LMM model used results from the male group as the
reference. ce LMM model used results from the 24-month group as the reference.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

5
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
decreases from 6 months old until 9~10 years old in the typically developing children. A similar decreasing trend
was observed in the PLR response time, although the response time had only moderate correlation with latency
(Pearson correlation r = 0.446). e PLR constriction time was not correlated with latency and showed no sig-
nicant age trend.
e observation that base pupil size was larger in the HR group than in the LR group appeared similar to
the dierence previously reported between 2-year old children with ASD and those of typical development12.
e observation of a smaller relative constriction in the HR group was similar to that observed in older children
aected by ASD using a desktop PLR device10,11. Following previous studies11,23, the relative or normalized con-
striction was used as a way to compensate the variations caused by dierent baseline pupil sizes. Interestingly,
a careful examination indicated such a dierence in relative constriction was due to the larger resting pupil
radius which was the denominator in calculating the relative constriction C% = (Ro2 − Rm2)/Ro2. No signicant
group eect was observed when the simple pupil size change Ro − Rm was analyzed using the LMManalysis.
Nevertheless, this observation was inconsistent with a previous study by Nyström et al. who reported that PLR
constriction was larger in 9–10-month-old with high risk of ASD16,23. Such inconsistency may be attributed to dif-
ferent methodology and testing conditions used. ere were signicant variations in the room lighting conditions
and optical stimulations among previously reported studies due to dierent testing systemsused. Changes in
lightadaptation and optical stimulation can greatly aect the PLR response and may lead to altered age trends22.
e PLR dierences observed between HR and LR children are similar to those reported in older children
between those aected by autism and those of typical development. Presumably, only very few in the HR group
may be eventually diagnosed with an ASD. Such observation could be attributed to genetic or possible environ-
mental factors; but further studies are necessary to understand this. e lack of a strong correlation between based
radius, relative constriction, latency, and constriction time may suggest these parameters are modulated under
dierent neurological mechanisms. A bigger pupil size and a smaller constriction may be consistent undera
stronger sympathetic modulation8. e constriction speed is controlled by iris muscle contraction and thus is
more under the inuence of parasympathetic modulation. On the other hand, the latency represents essentially
the neural signal transduction and processing speed, which could be aected by synaptic function, white matter
maturation, or network connectivity, all implicated in ASD30–33.
Two children in the HR groups received diagnoses at the end of this study: one was diagnosed with ASD and
the other with global developmental delay. e one diagnosed with ASD had PDDST-II score of 7 at both 12- and
Figure 4. Correlations among the six PLR parameters. e numbers in plots indicate Pearson’s correlation
coecients. **Correlation is signicant at the 0.01 level (2-tailed). * Correlation is signicant at the 0.05 level
(2-tailed).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

6
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
18-mo; whereas the other had a score of 4 at 12-mo and 6 at 18-mo. When examining the PLR results from these
two participants against the entire data set, no obvious distinct patterns were observed for the child diagnosed
with global developmental delay. However, the one who received an ASD diagnosis showed some interesting pat-
terns (Fig.3). First, this participant had a latency of 206.6 ms at 12-mo, the smallest among all participants, which
increased to 233.3 ms at 18-mo. Meanwhile, the constriction time decreased greatly from 455.6 ms at 12-mo
to 284.7 ms at 18-mo, which was the largest reduction among all participants. It is interesting to note that the
observation of a small PLR latency at 12-mo in the one diagnosed with an ASD appeared to be consistent with
previous speculation that ASD may be associated with shortened latency in infants, but longer latency in older
children11,12,23. Such age dependent dierence suggested the possible use of PLR as an indicator of atypical devel-
opmental trajectory in children.
In summary, we conducted a longitudinal study of PLR behaviors in 6-mo to 24-mo children with and without
high risk of developing ASD. e results indicated signicant age trends in base pupil radius, minimal pupil size,
and latency. Specically, this study showed that the base and minimal pupil size increased with age signicantly,
while the latency decreased signicantly during this period. Furthermore, atypical PLR parameters seen in previ-
ous studies of older children with ASD were also observed in younger children age 6–24-months. We have found
that the pupils of the children with higher risk of ASD were, on average, larger at both resting state and the time
of maximal constriction. e one participant who was diagnosed with ASD at the end of the study showed some
distinct patterns in PLR latency and constriction speed. Additional studies in a large population are necessary to
further evaluate these observations. In future studies, it will be valuable to also assess the eect of developmental
age in addition to chronological age. More advanced data analysis methodology such as Bayesian factor analysis
may also be employed to explore further the interactions among dierent factors.
Methods
Participants. Forty-two children participated in this study. All participants were recruited through the
ompson Center for Autism and Neurodevelopmental Disorders at the University of Missouri (MU). Twenty-
three participants made up the high-risk group (HR), which consists of children who have at least one sibling
diagnosed with ASD. e low-risk (LR) group had 19 children who have no family history of autism or other neu-
rodevelopmental disorders. is study was approved by the Institutional Review Board (IRB) of the University of
Missouri. All methods were performed in accordance with the relevant IRB guidelines and regulations. Written
informed consents were obtained from the parents/guardians prior to the PLR test.
Table2 shows the number of participants at each of the four nominal testing ages of 6-, 12-, 18-, and 24-month.
e actual age distributions were also shown in the table. e test data from one girl at 6-month and one girl at
12-month, both from the HR group, were not successful because they either could not look at the screen or their
excessive movement did not allow clear pupil images to be recorded. e nal dataset consists of PLR measures
from nine children (6 in HR and 3 in LR) who successfully completed test at all four ages, 24 children (9 in HR
and 15 in LR) who successfully completed tests at three ages, and nine children (8 in HR and 1 in LR) who only
completed tests at two or one age.
One participant in the HR group reported vision problems due to Usher syndrome34. is child’s PLR results
were included in the overall data analysis because they did not show any obvious dierences from other children’s
data in the group. All other participants reported neither vision problems nor any family history of eye disorders.
Participants were requested to withhold medications 48 hours prior to the test, unless it was necessary. Two sub-
jects received vaccinations within 24 hours of testing of their 6- and 12-month tests. ree subjects reported to
have taken antibiotics before their tests (two at 6-months and one at 24-months). Children in the LR group are
typically developing during the study period based on the family’s report of their most recent well-baby checkups.
In addition, the Pervasive Developmental Disorders Screening Test-II (PDDST-II, Pearson Clinical Assessment)29
scores were recorded as a simple screening for neurodevelopmental disorders in the participants before 24 month
of age. By the end of this study, two HR participants received diagnoses at the MU ompson Center: one with
ASD and one with non-ASD global development delay. Another participant was reported to have minor speech
delay. No developmental problems were reported by the parents of other participants on their recent well-baby
checkups.
Test procedure. All participants were tested using a remote PLR (rPLR) instrument. e details of the system
and testing arrangement have been described in detail previously27. e rPLR utilizes a tracking system to follow
the position of the subject’s right eye. e data of the eye’s position is then used to focus and change the direction
of the PLR imaging camera to the subject’s right pupil. is design enables PLR measurements in children with-
out the need of a strict physical restraint. e PLR tests were conducted in a bright room with illuminance level
measured at ~ 120 lux. e participants were placed in front of the rPLR system in a car seat or seated on a parent’s
6-mo 12-mo 18-mo 24-mo
Low-risk Number 2 F/3 M 9 F/10 M 9 F/10 M 8 F/8 M
Actua l age 6.0 ± 0.7 mo 11.3 ± 1.3 mo 17.4 ± 0.5 mo 23.7 ± 0.5
High-risk Number 5 F/6 M 9 F/13 M 8 F/12 M 6 F/8 M mo
Actua l age 6.3 ± 0.8 mo 12.2 ± 1.1 mo 17.6 ± 0.6 mo 23.4 ± 0.5 mo
Table 2. Number of participants and their age distributions at the four nominal testing ages (F: female; M:
male).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

7
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
lap. Cartoon videos were shown on a projection screen on the wall ~ 200 cm from the participants to maintain
their attention during the test. e videos that displayed on the projection screen had a size of 81.3 cm × 55.9 cm
(width × height). PLR was elicited by ashing the projection screen using a ceiling-mount green LED (530 nm
wavelength). We focused on studying the transient constriction phase of the PLR induced by a brief 100 ms ash
as in previous studies10–12. e transient PLR responses are mainly mediated by the cone and rod photoreceptors
under photopic conditions35–37. e stimulation wavelength used in this study is approximately midway between
the wavelengths at the peak sensitivities (V-lambda) of rods and cones. e stimulus light intensity at the position
of the eye was calibrated as 4.1 μW/cm2 (13.0 log photons cm−2 s−1). e rPLR system used 850-nm near infrared
(NIR) LED array to illuminate the subject’s right pupil for imaging.
Each PLR test lasted less than 10 minutes with about 20–25 PLR trials recorded from the participants as they
watched the cartoons. ere was a minimum of a 20 s interval between two consecutive trials. To begin with, the
participant was given a few minutes to get comfortable and acclimate to the testing room environment. Within
each PLR trial, the pupil images were recorded for 2 s starting 0.25 s prior to the 100 ms optical stimulation.
e PLR data recorded from all trials in each test were processed o-line to create the pupilogram curve to
quantify the change of pupil size with time. Similar to previous studies10,11, the following six PLR parameters were
extracted from the resulting pupilograms to characterize the constriction phase of the pupil responses (Fig.5).
1. e baseline pupil radius Ro was calculated as the average pupil radius prior to the light stimulation onset.
2. e minimal pupil radius Rm was calculated as the smallest pupil radius during constriction.
3. e relative constriction C% of the pupil was calculated as C% = (Ro2 − Rm2)/Ro2. is PLR measure nor-
malized changes in pupil area against the baseline pupil area.
4. e PLR latency tL was calculated as the time interval between the beginning of the stimulation and the
onset of the pupillary constriction. e constriction onset was determined as the rst deection data point
when pupil started to constrict consistently.
5. e constriction time tC was calculated as the time interval between the onset of the constriction and the
minimum pupil radius size.
6. e response time tR was calculated as the time interval between the stimulation onset and when pupil
reaches the minimal size, which was equivalent to tL + tC.
Multiple PLR trials were acquired during a single test. All PLR trials that could not be used to construct the
pupilogram were discarded. ese failed trials were generally caused by excessive eye/head movement. e PLR
parameters were calculated from all remaining successful PLR trials. Not all PLR parameters could be obtained
from all trials due to eye closing, movement, or blinking within the 2 s acquisition window. For each PLR param-
eter, trial results were considered as outliers if the values were more than 3 times of the Scaled Median Absolute
Deviation (MAD) away from the median of all successful trial results measured in the same test. Aer removing
all outliers, the mean PLR parameters from all remaining trials were calculated and used in the nal data analysis.
Figure 5. An illustration of the quantitative PLR parameters extracted from a measured pupilogram.
RoRmC%tLtRtC
Outliers mean ± std 0.3 ± 0.7 0.3 ± 0.7 0.4 ± 0.7 0.3 ± 0.6 0.4 ± 0.7 0.3 ± 0.7
% ≤ 1 93.7% 92.1% 89.7% 92.1% 88.9% 91.3%
Good trials mean ± std 16.5 ± 5.3 12.8 ± 6.0 12.4 ± 5.9 12.2 ± 5.6 12.1 ± 5.9 11.0 ± 5.5
% ≥ 5 99.2% 89.7% 88.1% 89.7% 84.1% 82.5%
Table 3. e distributions of number of outliers removed and the remaining good PLR trials for each PLR
parameter.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

8
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
e average and standard variation of numbers of good PLR trails for each parameter are shown below in Table3.
e majority of tests yielded at least ve good PLR trials that were included in the nal data analysis.
Statistical analysis. A linear mixed-eects model (LMM) was applied with maximum likelihood method to
determine the main eects of participant group (HR or LR), age, and sex on PLR parameters. A random intercept
model was applied. e eect of age as represented in four age groups (6-, 12-, 18-, and 24-month) was treated as
repeated measure with “Scaled Identity” as the repeated covariance type. Neither the group × age interaction, nor
the sex × age interaction, nor the group × sex interaction was found signicant during the model selection pro-
cess. erefore, no interaction term was included in the nal LMM analysis. Follow-up pairwise comparison with
Bonferroni condence interval adjustment was used to compare mean PLR parameters between dierent groups.
Alpha was set at 0.05 for all statistical tests. All statistical analysis was conducted in IBM SPSS Statistics V25.
Data availability
e datasets generated and/or analyzed during the current study are available from the corresponding author on
reasonable request.
Received: 7 August 2019; Accepted: 13 January 2020;
Published: xx xx xxxx
References
1. American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders (DSM-5®). (American Psychiatric
Publisher, 2013).
2. anner, L. Autistic disturbances of aective contact. Nerv. Child. 2, 217–250 (1943).
3. Xu, G., Strathearn, L., Liu, B. & Bao, W. Prevalence of autism spectrum disorder among US children and adolescents, 2014–2016.
Jama 319, 81–82 (2018).
4. Dawson, G. et al. andomized, controlled trial of an intervention for toddlers with autism: the early start Denver model. Pediatrics
125, e17–e23 (2010).
5. Warren, Z. et al. A systematic review of early intensive intervention for autism spectrum disorders. Pediatrics 127, e1303–e1311
(2011).
6. Baio, J. et al. Prevalence of autism spectrum disorder among children aged 8 years—autism and developmental disabilities
monitoring networ, 11 sites, United States, 2014. Morb. Mortal. Wkly. Rep. 67, 1 (2014).
7. Macari, S. L. et al. Predicting developmental status from 12 to 24 months in infants at ris for autism spectrum disorder: A
preliminary report. J. Autism Dev. Disord. 42, 2636–2647 (2012).
8. McDougal, D. H. & Gamlin, P. D. Autonomic control of the eye. Compr. Physiol. 5, 439–473 (2011).
9. ubin, L. S. Patterns of pupillary dilatation and constriction in psychotic adults and autistic children. J. Nerv. Ment. Dis. 133,
130–142 (1961).
10. Fan, X., Miles, J. H., Taahashi, N. & Yao, G. Abnormal transient pupillary light reex in individuals with autism spectrum disorders.
J. Autism Dev. Disord. 39, 1499–1508 (2009).
11. Daluwatte, C. et al. Atypical pupillary light reex and heart rate variability in children with autism spectrum disorder. J. Autism Dev.
Disord. 43, 1910–1925, https://doi.org/10.1007/s10803-012-1741-3 (2013).
12. Dinalanara, D. M., Miles, J. H., Nicole Taahashi, T. & Yao, G. Atypical pupillary light reex in 2–6–year–old children with autism
spectrum disorders. Autism Res. 10, 829–838 (2017).
13. Lynch, G. T., James, S. M. & VanDam, M. Pupillary response and phenotype in ASD: Latency to constriction discriminates ASD
from typically developing adolescents. Autism Res. 11, 364–375 (2018).
14. DiCriscio, A. S. & Troiani, V. Pupil adaptation corresponds to quantitative measures of autism traits in children. Sci. Rep. 7,
6476–6476, https://doi.org/10.1038/s41598-017-06829-1 (2017).
15. Daluwatte, C., Miles, J. H., Sun, J. & Yao, G. Association between pupillary light reex and sensory behaviors in children with autism
spectrum disorders. Res. Dev. Disabil. 37, 209–215, https://doi.org/10.1016/j.ridd.2014.11.019 (2015).
16. Nyström, P. et al. Enhanced pupillary light reex in infancy is associated with autism diagnosis in toddlerhood. Nat. Comm. 9, 1678
(2018).
17. MacLachlan, C. & Howland, H. C. Normal values and standard deviations for pupil diameter and interpupillary distance in subjects
aged 1 month to 19 years. Ophthalmic Physiol. Opt. 22, 175–182 (2002).
18. Silbert, J. F., Matta, N. S., Tian, J., Singman, E. L. & Silbert, D. I. Normative data on pupil size and anisocoria in children. J. AAPOS
17, e28–e29 (2013).
19. Daluwatte, C. et al. Age-dependent pupillary light reex parameters in children. in 2012 Annual International Conference of the
IEEE Engineering in Medicine and Biology Society, 3776–3779 (IEEE, San Diego, CA, USA 2012).
20. Winn, B., Whitaer, D., Elliott, D. B. & Phillips, N. J. Factors aecting light-adapted pupil size in normal human subjects. Investig.
Ophthalmol. Vis. Sci. 35, 1132–1137 (1994).
21. Bitsios, P., Prettyman, . & Szabadi, E. Changes in autonomic function with age: a study of pupillary inetics in healthy young and
old people. Age Ageing 25, 432–438 (1996).
22. Lobato-incón, L.-L. et al. Pupillary behavior in relation to wavelength and age. Front. Hum. Neurosci. 8, 221 (2014).
23. Nyström, P., Gredebäc, G., Bölte, S. & Falc-Ytter, T. Hypersensitive pupillary light reex in infants at ris for autism. Mol. Autism
6, 10 (2015).
24. Anderson, C. J. & Colombo, J. Larger tonic pupil size in young children with autism spectrum disorder. Dev. Psychobiol. 51, 207–211
(2009).
25. Nuse, H. J., Vivanti, G. & Dissanayae, C. Brief report: evidence for normative resting-state physiology in autism. J. Autism Dev.
Disord. 44, 2057–2063 (2014).
26. Martineau , J. et al. Can pupil size and pupil responses during visual scanning contribute to the diagnosis of autism spectrum
disorder in children? J. Psychiatr. Res. 45, 1077–1082 (2011).
27. Dinalanara, D. M., Miles, J. H. & Yao, G. rPL: an imaging system for measuring pupillary light reex at a distance. App Opt. 53,
7787–7795 (2014).
28. Ozono, S. et al. ecurrence ris for autism spectrum disorders: a Baby Siblings esearch Consortium study. Pediatrics 128,
e488–e495 (2011).
29. Siegel, B. Pervasive developmental disorders screening test (PDDST). (Springer 2013).
30. B ashat, D. B. et al. Accelerated maturation of white matter in young children with autism: a high b value DWI study. Neuroimage 37,
40–47 (2007).
31. Weinstein, M. et al. Abnormal white matter integrity in young children with autism. Hum. Brain Mapp. 32, 534–543 (2011).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9
SCIENTIFIC REPORTS | (2020) 10:1205 | https://doi.org/10.1038/s41598-020-58254-6
www.nature.com/scientificreports
www.nature.com/scientificreports/
32. Hull, J. V. et al. esting-state functional connectivity in autism spectrum disorders: a review. Front. Psychiatry 7, 205 (2017).
33. Bourgeron, T. From the genetic architecture to synaptic plasticity in autism spectrum disorder. Nat. Rev. Neurosci. 16, 551–563,
https://doi.org/10.1038/nrn3992 (2015).
34. Yan, D. & Liu, X. Z. Genetics and pathological mechanisms of Usher syndrome. J. Hum. Genet. 55, 327–335, https://doi.org/10.1038/
jhg.2010.29 (2010).
35. McDougal, D. H. & Gamlin, P. D. e inuence of intrinsically-photosensitive retinal ganglion cells on the spectral sensitivity and
response dynamics of the human pupillary light reex. Vis. Res. 50, 72–87, https://doi.org/10.1016/j.visres.2009.10.012 (2010).
36. Tsujimura, S. I. & Touda, Y. Delayed response of human melanopsin retinal ganglion cells on the pupillary light reex. Ophthalmic
Physiol. Opt. 31, 469–479 (2011).
37. eenan, W. T. et al. A visual circuit uses complementary mechanisms to support transient and sustained pupil constriction. elife 5,
e15392, https://doi.org/10.7554/eLife.15392 (2016).
Acknowledgements
is study was supported by a grant from the National Science Foundation (CBET-1507066). We thank all the
participating families for helping this project. We also thank Melissa Mahurin and Becky Gerdes for recruiting
participants.
Author contributions
J.H.M. and G.Y. designed the study. T.N.T. and J.H.M. identified, recruited the participants, collected and
examined medical diagnoses. L.A., D.M.R.D., C.K., and G.Y. conducted the PLR tests. C.K. and G.Y. processed,
analyzed the data, and wrote the main manuscript. All authors reviewed and revised the manuscript.
Competing interests
D.M.R.D., J.H.M. and G.Y. hold two US patents (US9050035B2, US9314157B2) “Device to measure pupillary
light reex in infants and toddlers” related to the remote PLR device used in this study. All other authors declare
no competing interests.
Additional information
Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-58254-6.
Correspondence and requests for materials should be addressed to G.Y.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2020
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com