Owner-rated hyperactivity/impulsivity is associated with sleep efficiency in family dogs. A non-invasive EEG study

Abstract

Subjective sleep disturbances are reported by humans with attention-deficit/hyperactivity disorder (ADHD). However, no consistent objective findings related to sleep disturbances led to sleep problems removal from ADHD diagnostic criteria. As a model for human ADHD with questionnaires validated for this purpose, dogs have been used also because their sleep physiology can be measured by non-invasive methods similarly to humans. We recorded spontaneous sleep EEG in family dogs during a laboratory session. We analyzed the association of sleep macrostructure and deep sleep (NREM) slow-wave activity (SWA) with a validated owner-rated ADHD questionnaire, assessing inattention (IA), hyperactivity/impulsivity (H/I) and total (T) scores. Higher H/I and T were associated with lower sleep efficiency and longer time awake after initial drowsiness and NREM. IA showed no associations with sleep variables. Further, no association was found between ADHD scores and SWA. Our results are in line with human studies in which poor sleep quality reported by ADHD subjects is associated with some objective EEG macrostructural parameters. This suggests that dogs’ H/I natural variation is useful to gain a deeper insight of ADHD neural mechanisms.

Full text

1
Vol.:(0123456789)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports
Owner‑rated hyperactivity/
impulsivity is associated with sleep
eciency in family dogs:
a non‑invasive EEG study
Cecília Carreiro
1,2,5*, Vivien Reicher
1,3,5, Anna Kis
2,4 & Márta Gácsi
2,3
Subjective sleep disturbances are reported by humans with attention‑decit/hyperactivity disorder
(ADHD). However, no consistent objective ndings related to sleep disturbances led to the removal
of sleep problems from ADHD diagnostic criteria. Dogs have been used as a model for human ADHD
with questionnaires validated for this purpose. Also, their sleep physiology can be measured by non‑
invasive methods similarly to humans. In the current study, we recorded spontaneous sleep EEG in
family dogs during a laboratory session. We analyzed the association of sleep macrostructure and
deep sleep (NREM) slow‑wave activity (SWA) with a validated owner‑rated ADHD questionnaire,
assessing inattention (IA), hyperactivity/impulsivity (H/I) and total (T) scores. Higher H/I and T were
associated with lower sleep eciency and longer time awake after initial drowsiness and NREM. IA
showed no associations with sleep variables. Further, no association was found between ADHD scores
and SWA. Our results are in line with human studies in which poor sleep quality reported by ADHD
subjects is associated with some objective EEG macrostructural parameters. This suggests that natural
variation in dogs’ H/I is useful to gain a deeper insight of ADHD neural mechanisms.
Attention-decit/hyperactivity disorder (ADHD) is one of the most prevalent psychiatric conditions1, being
characterized by symptoms of inattention (IA) and/or hyperactivity/impulsivity (H/I)2. Patients with ADHD
might also show impairments in the academic and social domains3. Although sleep disturbances were removed
from the diagnostic criteria for ADHD due to the lack of empirical evidence4,5, several studies indicated that
sleep disturbances are more common in ADHD patients of all ages than in the general population6–8. In addi-
tion to subjective measures (e.g., parent questionnaire reports9), objective sleep parameters (e.g., actigraphy10;
polysomnography11) were also applied to explore the associations between sleep and ADHD.
Regarding electroencephalography (EEG) sleep macrostructure, previous ndings indicated that ADHD
children had lower sleep eciency than the control; meanwhile, in sleep parameters of NREM (stage 1, 2 and
3), REM and REM sleep latency, no dierences were observed6. However, a more recent meta-analysis found
signicant dierences, but only in NREM stage 1; specically, ADHD children spent more time in stage 1, indi-
cating a lighter sleep compared to controls12. Interestingly, no ndings were replicated in adults with ADHD7.
Discrepancies between the meta-analyses might be explained by the fact that Cortese etal.6 included studies in
which children had primary sleep disorders. In such cases, it is problematic to conclude if sleep disturbances in
the ADHD group were actually due to ADHD itself or if the results were biased by the primary sleep problems.
Furthermore, most of the studies did not dierentiate ADHD subjects into subtypes. In the few studies when
individuals are separated by their ADHD subtype diagnosis, sleep disturbances were dierent depending on
symptoms, comorbidities and medications (e.g.13,14).
With respect to spectrum power dierences, most studies focus on slow wave activity (SWA; EEG power in
the 0.75–4.5Hz band) in NREM sleep, as this parameter has a crucial role in synaptic homeostasis15, plasticity16
and memory consolidation17. Previous studies—using high density EEG—observed higher levels of SWA in
NREM sleep in ADHD children in the centro-posterial scalp derivation18–20. Yet, another study reported reduced
SWA across the scalp in children and adolescents with ADHD compared to controls11. A longitudinal sleep
OPEN
1Doctoral School of Biology, Institute of Biology, ELTE Eötvös Loránd University, Budapest, Hungary. 2Department
of Ethology, Institute of Biology, ELTE Eötvös Loránd University, Budapest, Hungary. 3MTA-ELTE Comparative
Ethology Research Group, Budapest, Hungary. 4Institute of Cognitive Neuroscience and Psychology, Research
Centre for Natural Sciences, Budapest, Hungary. 5These authors contributed equally: Cecília Carreiro and Vivien
Reicher. *email: ceciliacarreiro@gmail.com

2
Vol:.(1234567890)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
study reported no group dierences between ADHD and the control, yet the percent of SWA in the rst NREM
period was lower in the ADHD group21. According to a recent meta-analysis, the observed dierences might
reect developmental alterations, as spectral dierences in ADHD patients seem to be age-related; increased
SWA in early childhood and decreased SWA in late childhood/adolescence, with the transition point at around
10years old22. In adults, sleep spectral parameters were less studied. One study found increased SWA in ADHD
individuals compared to non-ADHD controls23, while another study reported no dierences between control
and ADHD groups24.
us, it is clear that even more robust and sophisticated investigations are required to better understand the
associations between ADHD and sleep. Incorporating relevant animal models into the investigation oers a
promising opportunity to clarify these divergent ndings and to expand our knowledge in how ADHD aects
sleep. ese models can provide insight into specic aspects of disorders, which cannot be gained from research
with humans due to practical and ethical reasons25,26.
Several studies have demonstrated that the family dog is a promising animal model for human socio-cog-
nitive behaviors as well as their neuro-cognitive background. is includes attachment (behavior27; EEG28;
neuroimaging29), voice processing (behavior30; ERP31; neuroimaging32) and learning (behavior33; EEG34,35). Fur-
thermore, the family dog is increasingly recognized as a model for human neuropsychiatric conditions, such as
obsessive–compulsive disorder36, autism37 and ADHD-like characteristics38. Dierent questionnaires, originally
designed to measure human behavior, have been successfully applied to measure dog behavior (personality39;
impulsivity40). e rst questionnaire validated for dogs to evaluate ADHD-like traits41 was developed on the
basis of a human parental questionnaire, the DuPaul ADHD Rating Scale-IV42. Its reliability was validated and
its factor structure was replicated by an independent study43.
Behavioral tests have also been applied to assess ADHD-like traits in dogs. Higher owner-rated H/I scores
were associated with intolerance on delayed reward44; dogs with higher H/I scores preferred immediate reward
even though the food amount was smaller. Likewise, dogs that displayed higher social impulsivity (faster
approach towards friendly strangers) had a positive association with dopamine receptor gene polymorphisms,
which are known to be related to ADHD45. Recent ndings on canine self-inhibition using a behavioral Go/
No-Go paradigm showed that dogs’ higher H/I scores were associated with a greater number of mistakes due to
poorer self-inhibition, and dogs’ higher IA scores were associated with a shorter time to make these mistakes38.
ese ndings are in line with previous results on children with ADHD (e.g.46). Importantly, none of these
questionnaires or the behavioral tests have been dedicated to diagnosing ADHD in dogs. In our study, the ques-
tionnaire was also applied to assess characteristics of the ADHD-like spectrum.
In this study, we present the rst step to assess associations between ADHD scores and sleep EEG parameters
in the dog. e increasing interest in canine sleep research stems from its advantages to investigate the sleep of
a domesticated species, which in many respects is comparable to human sleep (for review, see47). Due to dogs’
natural cooperativeness, untrained family dogs have already been measured in a number of dierent non-invasive
sleep EEG studies (e.g.28,48). We assumed that dogs’ ADHD factors (IA, H/I and total scores based on41) would
show similar associations with sleep EEG variables to that of humans with ADHD. Specically, dogs with higher
scores on ADHD were expected to sleep less (lower sleep eciency), spend more time awake aer falling asleep
(more wakefulness aer sleep onset, WASO) and spend more time in supercial sleep (drowsiness: transitional
stage between quiet wakefulness and sleep49). Regarding SWA (1–4Hz), we did not formulate a clear hypothesis
due to the controversial results reported in the human literature. In the present study, we explored any potential
associations between dogs’ ADHD scores and SWA during NREM sleep.
Method
This study was approved by the Scientific Ethics Committee for Animal Experimentation (Állatkísérleti
Tudományos Etikai Tanács) of Budapest, Hungary, by categorizing it as a non-invasive study (number of ethi-
cal permission: PE/EA/853-2/2016). e experimental protocols were conducted according to the guidelines
of the Declaration of Helsinki and to ARRIVE guidelines, as outlined by the Association for the Study Animal
Behaviour (ASAB). Owners participated with their dogs in this study without any monetary compensation and
gave written consent.
e location of the sleep measurements was in one of the two fully equipped laboratories for canine EEG
measurements, depending on the laboratory availability. One laboratory is located at the Eötvös Loránd Uni-
versity (ELTE) and the other one is located at the Research Centre for Natural Sciences, Institute of Cognitive
Neuroscience and Psychology.
A more detailed description of the protocols below can be found in Supplementary Information.
Subjects. A total of 86 family dogs were recruited from the Family Dog Project (ELTE, Department of Ethol-
ogy) database. ey participated in a sleep EEG session during daytime in the presence of the owners and had
an ADHD questionnaire rated by their owners. In this study, we analyzed dogs between 6months and 14years
of age (Mage in months ± SD: 70.45 ± 52.10), including 36 males and 50 females of various breeds (54 purebred
dogs and 32 mixed breed dogs).
Attention‑decit/hyperactivity disorder questionnaire. Dogs’ owners were asked to answer the
ADHD questionnaire validated for dogs41, which assesses individual dierences of inattention (IA; 6 items of
attentional levels), hyperactivity/impulsivity (H/I; 7 items of motor activity/impulsivity levels) and the combi-
nation of total scores (T; 13 items). e questionnaire was developed based on a validated version used for a
parent-report rating scale of ADHD and related problems in children (ADHD-RS-IV42). Higher scores indicate
greater diculties with IA and H/I. In addition to IA and H/I scores, total (T = IA + H/I) score was also evaluated.

3
Vol.:(0123456789)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
Statements related to attention and activity were presented in the same mixed order. e owners had to choose
among the given answers, representing how frequent the statement was true for their dogs on a Likert scale.
Initially, the questionnaire was given to the owners in a paper form and, later, it was converted to an online ver-
sion. During to this transition from the paper to the digital questionnaire, a fault happened related to the answer
options and the 4-point Likert scale (N = 34) was changed to a 5-point Likert scale (N = 52). To use all the 86
answers as a unied database, we transformed the scales into a uniform 5-point Likert scale, using normalized
scores. e details of the transformations can be found in the Supplementary Information (Fig.S1, TableS1).
Electroencephalography. Prior the sleep EEG measurements to have dogs’ sleep as standardized as pos-
sible, owners were instructed to keep the dogs in a typical routine. is means that dogs slept normally the night
before, woke up as usual in the morning and had no extra activity or stress, as variations in dogs’ routine are
known to increase sleep pressure and aect sleep variables (e.g.49).
Dogs were assessed in a non-invasive sleep EEG recording with the minimum duration of 1.5h and a maxi-
mum duration of 3h during daytime (depending on owners’ availability; 70% started in the aernoon (12–18h),
25% in the evening (18–20h) and 5% before noon). e variance in the record duration occurred due to dier-
ences in subject compliance. Some dogs aer waking up became very active, thus, the recording had to be nished
regardless of elapsed time. Other dogs, even if they woke up, continued lying and relaxing next to their owner
and/or fell back asleep (thus, recording could be continued for the duration of 3h). A detailed description of the
most recent polysomnographic method and EEG electrode placement (Fig.1) can be found in Reicher etal.50.
We followed the validated polysomnography (PSG) method on dogs49. Dogs were measured between 2015 and
2020. During this period, electrode placement and recording methods were improved (for detailed description
of the most recent method, see50). In the case of 72 dogs, the old setup was used, thus, one active EEG channel
(frontal: Fz) and an eye movement channel were recorded. While in the case of 14 dogs, the current setup was
used, specically, four active EEG channels and an eye movement channel were recorded. In this case, Fz and
Cz channels were placed over, respectively, the anteroposterior midline of the skull; F7 and F8 channels were
placed on the right and le zygomatic arch next to the eyes, all referred to G2. e ground electrode (G1) was
attached to the le musculus temporalis. Also, an additional channel, labeled EOG, was visualized as the bipolarly
referenced F7-F8 electrodes. In all dogs, at least the frontal electrode (Fz) was active, thus, in this study only
data from this electrode were used for spectral analyses. See Fig.1 for electrode placement. Gold-coated Ag/
AgCl electrodes were used, secured by Signa Spray Electrode Solution (Parker, United States) and EC2 Grass
Figure1. Positions of the electrodes relative to the three-dimensional model and endocranial cast of the skull
of a pointer dog: (A) lateral, (B) anterior and (C) superior views, image by Kálmán Czeibert. Placement of
the electrodes (Fz–Cz: frontal and central midline; F8–F7: right and le electrodes placed on the zygomatic
arch; Ref: reference electrode or G2; Gnd: ground electrode or G1). In this study, the statistical analysis was
performed with the Fz data only. All other electrodes were merely used to aid sleep stage scoring.

4
Vol:.(1234567890)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
Electrode Cream (Grass Technologies, United States). Impedance values of the EEG electrodes were kept under
20 kΩ during the recordings.
Recordings were obtained with one of the following technical arrangements:
(1) In case of 14 dogs, the signal was collected, amplied and digitized at a sampling rate of 1000Hz/channel
using the 40-channels NuAmps amplier (© 2018 Compumedics Neuroscan) and DC-recording, later saved
in .cnt format with the Scan 4.3 Acquire soware (© 2018 Compumedics Neuroscan) then converted to
.edf format using MatLab EEG Toolbox.
(2) In case of 72 dogs, the signals were collected, pre-ltered, amplied and digitized at a sampling rate of
1024Hz/channel, using the 25 channel SAM 25R EEG System (Micromed, Mogliano Veneto, Italy), and
the System Plus Evolution soware with second-order lters at 0.016Hz (high pass) and 70Hz (low pass).
To correct for dierences in EEG lter characteristics across recording devices, a standard calibration process
(dog51,52; human53,54) was implemented on devices (1) and (2). For more details, see Supplementary Information.
Data analysis. e sleep EEG assessment followed the protocol described by Reicher etal.50. Both sleep
macrostructure and spectral data were analyzed. Sleep macrostructure variables were examined in 86 dogs (Mage
in months ± SD: 70.45 ± 52.10) and the spectral variable in 70 dogs (Mage in months ± SD: 69.91 ± 51.40) as 16
dogs did not reach NREM sleep during their sleep recording.
Sleep recordings were visually scored in accordance with standard criteria55 adapted for dogs, previously
shown to reliably identify stages of wake, drowsiness, NREM and REM in dogs49,56. A program developed by
Ferenc Gombos (Fercio’s EEG Plus, 2009–2022) was used to analyze and export data. e recordings were manu-
ally scored, artifacts on the EEG channels were excluded and the program provided data of macrostructure and
spectrum variables from the dierent sleep stages.
e macrostructural variables of interest were sleep eciency, relative time spent awake aer the rst epoch
scored as drowsiness (WASO 1) and aer NREM sleep (WASO 2), relative time spent in drowsiness, NREM and
REM sleep. Due to the variance of total record time, relative values were used in the analyses for all sleep macro-
structure variables to control for any potential biases. Regarding spectral analysis, due to the relevance of SWA
in deep sleep for the restorative function of sleep and ADHD (e.g.21,22), we analyzed only this frequency range.
Average power spectral densities were calculated by a Fast Fourier Transformation (FFT) algorithm applied to
the 50% overlapping Hanning-tapered 4-s windows of the EEG signal of the Fz-G2 derivation. Dogs are known
to show notable individual-level variation in morphological features regarding head musculature and skull shape
and thickness57 that might inuence the EEG data. To circumvent measurement error that might arise from these
dierences, absolute power was normalized by computing the relative power spectra of the Fz data for SWA in
the frequency range of 1–4Hz. is means that the absolute spectrum values for the 1–4Hz frequency range on
the Fz channel were divided by the sum of the absolute Fz values for the 1–30Hz range.
e macrostructure and spectrum variables of interest are summarized in Table1.
Statistical analysis. e statistical analyses were performed in R (version 3.6.3: RCoreTeam, 2014). ADHD
variables (IA, H/I and T scores) as well as most sleep parameters (except from WASO 1and WASO 2) showed
non-normal distribution based on Shapiro-Wilkson normality test. Prior further analysis, age-related dier-
ences in our variables were also checked with Kendall’s partial rank correlations as age had non-normal distri-
bution. Only NREM and the SWA were aected by age. us, in the case of the nal analysis of ADHD-related
dierences in NREM sleep and in SWA, partial correlations were performed, controlling for age. Kendall’s rank
correlations were conducted between ADHD-related variables and sleep parameters (sleep eciency, WASO 1,
WASO 2, drowsiness, REM, NREM and SWA). To control for multiple comparisons, Benjamini–Hochberg cor-
rection was conducted.
Table 1. Variables of interest regarding dogs’ ADHD questionnaire and sleep EEG data. (N)REM: (non)rapid
eye movements sleep stage.
Assessment Variable Measure
Dog ADHD questionnaire
T score total sum of all 13 items (1–65)
IA score sum of 6 items related to inattention (1–30)
H/I score sum of 7 items related to hyperactivity/impulsivity (1–35)
EEG data
Sleep eciency time spent sleeping/record duration (%)
WASO 1 time spent awake aer the rst epoch scored as drowsiness/record duration (%)
WASO 2 time spent awake aer the rst epoch scored as NREM/record duration (%)
Drowsiness time spent in drowsiness/record duration (%)
NREM time spent in NREM/record duration (%)
REM time spent in REM/record duration (%)
SWA slow-wave activity relative power spectrum (frequency range: 1–4Hz)

5
Vol.:(0123456789)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
Additionally, since the dogs’ sleep was measured at dierent hours of the day (i.e. dierent circadian rhythms)
due to practical reasons (availability of the owners), we ran Pearson correlations to check whether the EEG
recording start time was related to the ADHD scores used for correlation analyses and to the EEG variables.
Results
See Table2 for a summary of statistical results.
ere was a negative association between ADHD T scores and sleep eciency, and a positive association
between T scores and WASO 1 and WASO 2 (Fig.2). Actual power analysis showed that these associations had
a medium eect size (respectively, Cohen’s d = 0.573; 0.569; 0.668; provided by G*Power version 3.1.9.758). T
scores were not associated with drowsiness, NREM and REM sleep duration, and SWA.
Aer controlling for false discovery rate due to multiple comparisons, the associations between IA and the
sleep variables were no longer signicant (Fig.3).
H/I was negatively associated with sleep eciency and positively associated with WASO 1 and WASO 2
(Fig.4). Actual power analysis showed that these associations had a medium eect size (respectively, Cohen’s
Table 2. Results of Kendall tau’s B correlations between ADHD and sleep related variables (* statistically
signicant).
Independent variable Dependent variable Kendall tau’s B p value corr. p value Condence intervals
Tot a l
Sleep eciency −0.197 0.008* 0.031* −0.342, −0.053
WASO 1 0.196 0.009* 0.031* 0.051, 0.341
WASO 2 0.223 0.003* 0.031* 0.070, 0.376
Drowsiness 0.109 0.146 0.245 −0.039, 0.257
NREM −0.065 0.377 0.465 −0.210, 0.099
REM −0.118 0.188 0.225 −0.260, 0.023
SWA 0.055 0.501 0.554 −0.092, 0.236
Inattention
Sleep eciency −0.172 0.024 0.062 −0.317, −0.028
WASO 1 0.164 0.031 0.073 0.017, 0.311
WASO 2 0.181 0.018 0.053 0.026, 0.336
Drowsiness 0.109 0.152 0.245 −0.037, 0.256
NREM −0.073 0.323 0.424 −0.202, 0.100
REM −0.083 0.282 0.394 −0.233, 0.067
SWA 0.024 0.765 0.765 −0.068, 0.258
Hyperactivity/Impulsivity
Sleep eciency −0.208 0.006* 0.031* −0.352, −0.065
WASO 1 0.198 0.009* 0.031* 0.056, 0.339
WASO 2 0.203 0.007* 0.031* 0.052, 0.354
Drowsiness 0.093 0.221 0.331 −0.055, 0.240
NREM −0.051 0.489 0.554 −0.212, 0.108
REM −0.133 0.082 0.172 −0.265, −0.002
SWA 0.042 0.608 0.638 −0.160, 0.194
Figure2. Visualization of Kendall’s rank correlations that showed signicant associations between ADHD
total score and (A) sleep eciency (τb = −0.197,corr. p = 0.031), (B) WASO 1 (τb = 0.196,corr. p = 0.031) and (C)
WASO 2 (τb = 0.223,corr. p = 0.031) aer controlling for multiple correlations (Benjamini-Hochberg). Shaded
areas indicate condence intervals.

6
Vol:.(1234567890)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
d = 0.573; 0.569; 0.668; provided by G*Power version 3.1.9.758). Furthermore, H/I was not associated with drowsi-
ness, NREM and REM sleep duration, and SWA.
ere was no association between the ADHD scores (T: r = −0.059, p = 0.590; IA: r = −0.064, p = 0.560; H/I:
r = −0.037, p = 0.737) and the EEG recording start time, as well as no association between EEG variables (Sleep
eciency: r = 0.053, p = 0.624; WASO 1: r = −0.003, p = 0.973; WASO 2: r = −0.011, p = 0.916) and the EEG record-
ing start time.
Discussion
We examined whether family dogs’ owner-rated ADHD scores are related to their sleep structure and NREM
EEG spectrum. Our results supported our hypothesis regarding sleep eciency, but did not in case of drowsiness
and we did not nd associations between ADHD scores and slow wave activity.
Dogs with higher T and H/I scores showed lower sleep eciency and a similar tendency was detected in
case of IA. Similarly, children6 and adults59 diagnosed with ADHD were reported to have lower sleep eciency.
is parallel holds true despite the fact that the current dog study used the ADHD score as a continuous vari-
able, while previous human studies compared ADHD versus control individuals using a set cut-o. However,
according to a more recent PSG-based meta-analysis, no dierences in sleep eciency were detected between
control and ADHD groups in either children12 or adults7. It is noteworthy that most human studies, focusing on
the associations between ADHD and sleep, measured individuals with ADHD as a universal group and did not
dierentiate between subtypes (for review, see7). One of the exemptions, somewhat similarly to our ndings,
found that the H/I subtype had lower sleep eciency, while IA and combined subtypes had no signicant dif-
ferences in their sleep60. Interestingly, one study, based on parental rating, found that children of IA subtype had
the fewest reported sleep problems compared to H/I and combined subtypes13. Other studies reported dierent
associations; compared to the combined subtype, IA subtype was associated with poorer sleep quality in adults61.
e diverse results suggest dierences according to specic ADHD subtypes (for review, see62).
In the current study, the time dogs spent awake aer drowsiness (WASO 1) and aer NREM sleep (WASO 2)
onset also showed correlations with ADHD scores. Higher T and H/I were positively associated with WASO 1 and
Figure3. Visualization of Kendall’s rank correlations that showed non-signicant tendency between ADHD
inattention score and (A) sleep eciency (τb = −0.172, corr. p = 0.062), (B) WASO 1 (τb = 0.164, corr. p = 0.073)
and (C) WASO 2 (τb = 0.181, corr. p = 0.053) aer controlling for multiple correlations (Benjamini–Hochberg).
Shaded areas indicate condence intervals.
Figure4. Visualization of Kendall’s rank correlations that showed signicant associations between
ADHD hyperactivity/impulsivity score and (A) sleep eciency (τb = −0.208,corr. p = 0.031), (B) WASO 1
(τb = 0.198,corr. p = 0.031) and (C) WASO 2 (τb = 0.203,corr. p = 0.031) aer controlling for multiple correlations
(Benjamini–Hochberg). Shaded areas indicate condence intervals.

7
Vol.:(0123456789)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
2. ese results are in line with our ndings about sleep eciency (i.e. if dogs sleep less, they spend more time
awake either at the beginning of the recording or aer sleep onset). Regarding awakenings and specic ADHD
subtypes, fewer data is available in human literature. A study using EEG found that ADHD H/I children had
greater sleep fragmentation60. Another study analyzing abnormal movements in ADHD children during the night
suggested higher sleep fragmentation compared to control63, which is in line with increased WASO64,65. Most of
these studies used dierent indices to quantify movements in sleep and the data cannot be directly compared,
but they all are indicative of more awakenings and sleep fragmentation in ADHD.
We did not nd associations between ADHD scores and drowsiness, NREM and REM sleep duration. It is
dicult to parallel our results with previous human ndings as the human literature shows inconsistent results.
Cortese etal.6, in a meta-analysis, observed no associations between ADHD and sleep macrostructure variables
in children. A more recent meta-analysis found that children with ADHD spent more time in stage 1 NREM,
compared to control12. In adults, however, a replication study did not nd dierences between ADHD and
control7. We have to note that, despite all our eorts to have pre-sleep activity uniform across subjects, there
might be still a consistent bias between high versus low ADHD dogs in this respect (e.g. dogs scoring high on
the H/I factor are expected to spontaneously engage in more activity during their typical routine). It is known
that increased pre-sleep activity aects subsequent sleep structure, causing, e.g., increased sleep eciency49,66.
In contrast, our current ndings show, in accordance with the human literature on ADHD sleep6,59, that dogs
with higher T scores (which includes the activity factor) had lower sleep eciency. Moreover, in our dogs, the
dierent hours the EEG started (i.e. dierent stages of the circadian rhythms) had no association with the ADHD
scores nor with the EEG variables. us, it is unlikely that the result reported here would be due to dierences
in pre-sleep activity since that would predict a dierence in the opposite direction. However, it is possible that
our results are the sum of eects from dierences inherent in the ADHD factors and dierences in pre-sleep
activity; this would mean that, e.g., the dierence in sleep eciency is actually greater than reported here, but it
was partly oset by the dogs’ physical activity. Following the same line of thought, it is possible that other sleep
macrostructure dierences could have been entirely masked, showing non-signicant ndings. Disentangling
these separate eects is obviously near to impossible, not only in dogs, but also in humans, since interfering
with the usual routine of subjects in order to unify pre-sleep activity would inherently result in stress for those
implicated and, as a result, it would bias the sleep data yet again.
With respect to SWA, aer controlling for age, we found no associations with ADHD scores. is is contrary
to the human literature; ADHD delayed SWA decline during brain maturation seems to be more evident in ado-
lescents (12–16.5years old21). us, it has been suggested that SWA dynamics and sleep homeostatic recovery
may occur more slowly, specically, in ADHD adolescents than in other age groups. In our study, although there
was a large age variability, we did not separate our subjects in dierent age groups due to the sparse literature on
clear age ranges from puppies to senior dogs (e.g.51). Furthermore, in dogs, it is not clear at what age the onset
of SWA decrease occurs and what other factors aect it. For instance, in our previous study, we found that SWA
was not only associated with age, but with dogs’ weight as well; larger-sized dogs had more SWA than smaller-
sized ones (for more details, see51). In addition, most recent studies mapped dierent topographical features
in ADHD children, using high-resolution EEG method18,20. In dogs’ sleep EEG studies, four EEG channels are
applied (e.g.67,68) and, in the current research, only the frontal EEG channel was analyzed. us, alternations in
cortical topography of dog sleep could not be investigated.
Besides these physiological factors, psychosocial factors have also been reported to be involved in the multiple
underlying pathways of ADHD69. us, ADHD has been seen as a socio-emotional dysfunction. ese would be
important to consider in future canine research as, e.g., the dog-owner social bond is in association with altered
SWA dynamics during sleep (alpha-delta anticorrelation28).
It is important to emphasize that dogs that participated in this study did not receive the diagnosis of ADHD,
thus they represent the typical family dog population and their levels of IA and H/I. is implies a relevant dif-
ference from human studies that are always composed of subjects diagnosed for ADHD. A sample focused on
dogs with the highest ADHD scores would benet comparative analyses, approaching to more similar behavioral
disorders in human ADHD patients. Also, it is well documented that ADHD symptoms tend to decrease from
childhood to adulthood and studies in humans analyze individuals in separated age groups24,70. Due to practi-
cal reasons, however, the age range in our study was wider than usual human ADHD research. us, it would
be fruitful to analyze potential dierences within and between specic age groups in future studies with dogs.
Our study considered EEG data only from the rst time the dogs were assessed in an unfamiliar place during
daytime, constituting a dierent routine during their sleep time compared to sleeping at home in nighttime66. In
humans, sleep studies are mainly conducted during nighttime. is dierence between human nighttime and
canine daytime sleep measurements further challenges the comparisons. To handle this dierence of daytime
sleep, as part of our EEG protocol, owners were instructed to keep the dog on a typical day, e.g., no extra activity
carried out or sleep deprivation, which could increase sleep pressure and aect sleep variables as shown in previ-
ous sleep studies in dogs49,66. However, we did not have information on the exact duration of time spent awake
and this specic control (e.g. dierent length of time dogs are awake before the sleep EEG) would be interesting
for further studies on sleep pressure and SWA. When analyzing the eect of other factors such as daytime versus
nighttime sleep, pre-sleep activity and location, Bunford etal.66 observed that dogs had, e.g., a longer WASO
1 in daytime sleep, indicating that the time of the sleep related to ADHD needs further analysis. However, it
is important to highlight that our ndings are similar to studies on the eect of ADHD in WASO 1 in humans
(e.g.60). e ndings of pre-sleep activity and location aecting dogs’ sleep variables previously observed (e.g.
more REM sleep in dogs sleeping at home versus at the laboratory)66, in our study, had no signicant associations
and were unlikely aected by ADHDtraits in dogs. erefore, although all dogs were assessed under the same
condition and our ndings of within-individual relations between age and sleep parameters may not be aected

8
Vol:.(1234567890)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
by these factors, it is recommended in future studies the assessment of home and/or nighttime recordings to
evaluate these relationships under dierent conditions.
Also, it is possible that our results related to sleep EEG variables were aected by a rst/second-night eect at
least to a certain extent (dogs50; humans71). Further, dogs’ sleep habits (frequency of sleeping away from home)
also aect their sleep pattern, especially on the rst sleep occasion in a new environment50. In the current sam-
ple, we had no information on most dogs’ sleep habits, as their sleep assessment had been conducted before the
paper on dogs was published50. us, in future dog sleep studies, it is suggested to consider dogs’ sleep habits in
order to address this eect in dogs.
Finally, objective behavioral evaluations of executive functions could provide complementary information
about some dimensions of ADHD and their relationship with sleep, such as performance observed in a Go/
No-Go paradigm in dogs38,72. Many studies have indicated that disagreements on ADHD diagnostic and symp-
toms of the subtypes might be one of the fundamental biases in analyses trying to establish objective measures
of sleep disturbances in ADHD patients25,26. us, behavioral tests and an improved version of dog ADHD
questionnaire (e.g.43) can be helpful in this direction.
Conclusion
In this large-scale non-invasive sleep study on owner-rated ADHD questionnaire and sleep EEG parameters,
we found that some objective indices of sleep are associated with higher ADHD scores in dogs. is suggests
that similar neural features might underlie dogs’ and humans’ natural variation of hyperactivity/impulsivity.
Data availability
e datasets used and/or analyzed during the current study are available from the corresponding author on
reasonable request.
Received: 27 June 2022; Accepted: 16 January 2023
References
1. Polanczyk, G. V., Willcutt, E. G., Salum, G. A., Kieling, C. & Rohde, L. A. ADHD prevalence estimates across three decades: An
updated systematic review and meta-regression analysis. Int. J. Epidemiol. 43, 434–442. https:// doi. org/ 10. 1093/ ije/ dyt261 (2014).
2 . Evans, S. W., Owens, J. S. & Bunford, N. Evidence-based psychosocial treatments for children and adolescents with attention-decit/
hyperactivity disorder. J. Clin. Child Adolesc. Psychol. 43, 527–551. https:// doi. org/ 10. 1080/ 15374 416. 2013. 850700 (2014).
3. Gaub, M. & Carlson, C. L. Behavioral characteristics of DSM-IV ADHD subtypes in a school-based population. J. Abnorm. Child
Psychol. 25, 103–111. https:// doi. org/ 10. 1023/A: 10257 75311 259 (1997).
4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. 4th edn. 866. https:// doi. org/
10. 1176/ ajp. 152.8. 1228 (American Psychiatric Association, 1994).
5. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-V. 5th edn. 947:59–66. https:// doi.
org/ 10. 1176/ appi. books. 97808 90425 596 (American Psychiatric Association, 2013).
6. Cortese, S., Faraone, S. V., Konofal, E. & Lecendreux, M. Sleep in children with attention-decit/hyperactivity disorder: Meta-
analysis of subjective and objective studies. J. Am. Acad. Child Adolesc. Psychiatry 48, 894–908. https:// doi. o rg/ 10. 1097/ CHI. 0b013
e3181 ac09c9 (2009).
7. Diaz-Roman, A., Mitchell, R. & Cortese, S. Sleep in adults with ADHD: Systematic review and meta-analysis of subjective and
objective studies. Neurosci. Biobehav. Rev. 89, 61–71. https:// doi. org/ 10. 1016/j. neubi orev. 2018. 02. 014 (2018).
8. Becker, S. P. ADHD and sleep: Recent advances and future directions. Curr. Opin. Psychol. 34, 50–56. https:// doi. org/ 10. 1016/j.
copsyc. 2019. 09. 006 (2020).
9. O wens, J. A., Maxim, R., Nobile, C., McGuinn, M. & Msall, M. Parental and self-report of sleep in children with attention-decit/
hyperactivity disorder. Arch. Pediatr. Adolesc. Med. 154, 549–555. https:// doi. org/ 10. 1001/ archp edi. 154.6. 549 (2000).
10. Moreau, V., Rouleau, N. & Morin, C. M. Sleep of children with attention decit hyperactivity disorder: Actigraphic and parental
reports. Behav. Sleep Med. 12, 69–83. https:// doi. org/ 10. 1080/ 15402 002. 2013. 764526 (2014).
11. Furrer, M. et al. Sleep EEG slow-wave activity in medicated and unmedicated children and adolescents with attention-decit/
hyperactivity disorder. Transl. Psychiatry 9, 1–8. https:// doi. org/ 10. 1038/ s41398- 019- 0659-3 (2019).
12. Díaz-Román, A., Hita-Yáñez, E. & Buela-Casal, G. Sleep characteristics in children with attention decit hyperactivity disorder:
Systematic review and meta-analyses. J. Clin. Sleep Med. 12, 747–756. https:// doi. org/ 10. 5664/ jcsm. 5810 (2016).
13. Mayes, S. D. et al. ADHD subtypes and comorbid anxiety, depression, and oppositional-deant disorder: Dierences in sleep
problems. J. Pediatr. Psychol. 34, 328–337. https:// doi. org/ 10. 1093/ jpepsy/ jsn083 (2009).
14. Wagner, J. & Schlarb, A. A. Subtypes of ADHD and their association with sleep disturbances in children. Somnol. Schlaorschung
Schlafmed. 16, 118–124. https:// doi. org/ 10. 1007/ s11818- 012- 0569-6 (2012).
15. Tononi, G. & Cirelli, C. Sleep and synaptic homeostasis: A hypothesis. Brain Res. Bull. 62, 143–150. https:// doi. org/ 10. 1016/j. bra in
resbu ll. 2003. 09. 004 (2003).
16. Rasch, B. & Born, J. About sleep’s role in memory. Physiol Rev. https:// doi. org/ 10. 1152/ physr ev. 00032. 2012 (2013).
17. Tononi, G. & Cirelli, C. Sleep and the price of plasticity: From synaptic and cellular homeostasis to memory consolidation and
integration. Neuron 81, 12–34. https:// doi. org/ 10. 1016/j. neuron. 2013. 12. 025 (2014).
18. Castelnovo, A. et al. Sleep power topography in children with attention decit hyperactivity disorder (ADHD). Children 9, 197.
https:// doi. org/ 10. 3390/ child ren90 20197 (2022).
19. Miano, S. et al. Shooting a high-density electroencephalographic picture on sleep in children with attention-decit/hyperactivity
disorder. Sleep 42, zsz167. https:// doi. org/ 10. 1093/ sleep/ zsz167 (2019).
20. Ringli, M. et al. Topography of sleep slow wave activity in children with attention-decit/hyperactivity disorder. Cortex 49, 340–347.
https:// doi. org/ 10. 1016/j. cortex. 2012. 07. 007 (2013).
21. Darchia, N. et al. Longitudinal assessment of NREM sleep EEG in typically developing and medication-free ADHD adolescents:
First year results. Sleep Med. 80, 171–175. https:// doi. org/ 10. 1016/j. sleep. 2021. 01. 052 (2021).
22. Biancardi, C., Sesso, G., Masi, G., Faraguna, U. & Sicca, F. Sleep EEG microstructure in children and adolescents with attention
decit hyperactivity disorder: A systematic review and meta-analysis. Sleep 44, zsab006. https:// doi. org/ 10. 1093/ sleep/ zsab0 06
(2021).
23. Lindberg, N. et al. ADHD and sleep in homicidal men with antisocial personality disorder. Neuropsychobiology 50, 41–47. https://
doi. org/ 10. 1159/ 00007 7940 (2004).

9
Vol.:(0123456789)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
24. Philipsen, A. et al. Sleep in adults with attention-decit/hyperactivity disorder: A controlled polysomnographic study including
spectral analysis of the sleep EEG. Sleep 28, 877–884. https:// doi. org/ 10. 1093/ sleep/ 28.7. 877 (2005).
25. Majdak, P. et al. A new mouse model of ADHD for medication development. Sci. Rep. 6, 1–18. https:// doi. org/ 10. 1038/ srep3 9472
(2016).
26. Russell, V. A. Overview of animal models of attention decit hyperactivity disorder (ADHD). Curr. Protoc. Neurosci. 54, 9–35.
https:// doi. org/ 10. 1002/ 04711 42301. ns093 5s54 (2011).
27. Topál, J. & Gácsi, M. Lessons we should learn from our unique relationship with dogs: An ethological approach. in Crossing
Boundaries (eds. Birke, L. & Hockenhull, J.). 161–186. https:// doi. org/ 10. 1163/ 97890 04233 041_ 010 (Brill Academic Press, 2012).
28. Carreiro, C., Reicher, V., Kis, A. & Gácsi, M. Attachment towards the owner is associated with spontaneous sleep EEG parameters
in family dogs. Animals 12, 895. https:// doi. org/ 10. 3390/ ani12 070895 (2022).
29. Gábor, A. et al. Social relationship-dependent neural response to speech in dogs. Neuroimage 243, 118480. https:// doi. org/ 10.
1016/j. neuro image. 2021. 118480 (2021).
30. Salamon, A., Száraz, J., Miklósi, Á. & Gácsi, M. Movement and vocal intonation together evoke social referencing in companion
dogs when confronted with a suspicious stranger. Anim. Cogn. 23, 913–924. https:// doi. org/ 10. 1007/ s10071- 020- 01401-3 (2020).
31. Bálint, A., Eleőd, H., Magyari, L., Kis, A. & Gácsi, M. Dierences in dogs’ event-related potentials in response to human and dog
vocal stimuli; a non-invasive study. R. Soc. Open Sci. 9, 211769. https:// doi. org/ 10. 1098/ rsos. 211769 (2022).
32. Andics, A. et al. Neural mechanisms for lexical processing in dogs. Science 1979(353), 1030–1032. https:// doi. org/ 10. 1126/ scien
ce. aaf37 77 (2016).
33. Fugazza, C., Moesta, A., Pogány, Á. & Miklósi, Á. Social learning from conspecics and humans in dog puppies. Sci. Rep. 8, 1–10.
https:// doi. org/ 10. 1038/ s41598- 018- 27654-0 (2018).
34. Kis, A. et al. e interrelated eect of sleep and learning in dogs (Canis familiaris); An EEG and behavioural study. Sci. Rep. 7,
1–6. https:// doi. org/ 10. 1038/ srep4 1873 (2017).
35. Iotchev, I. B. et al. Averaging sleep spindle occurrence in dogs predicts learning performance better than single measures. Sci. Rep.
10, 1–6. https:// doi. org/ 10. 1038/ s41598- 020- 80417-8 (2020).
36. Ledford, H. Dog DNA probed for clues to human psychiatric ills: Project will compare gene data to owners’ assessments of how
their companions behave. Nature 529, 446–448. link.gale.com/apps/doc/A441882776/AONE?u=anon~491f828a&sid=googleSc
holar&xid=18507466 (2016).
37. Topál, J., Román, V. & Turcsán, B. e dog (Canis familiaris) as a translational model of autism: It is high time we move from
promise to reality. Wiley Interdiscip. Rev. Cognit. Sci. 10, e1495. https:// doi. org/ 10. 1002/ wcs. 1495 (2019).
38. Bunford, N. et al. Associations among behavioral inhibition and owner-rated attention, hyperactivity/impulsivity, and personality
in the domestic dog (Canis familiaris). J. Comp. Psychol. 133, 233. https:// doi. org/ 10. 1037/ com00 00151 (2019).
39. Gosling, S. D., Kwan, V. S. Y. & John, O. P. A dog’s got personality: A cross-species comparative approach to personality judgments
in dogs and humans. J. Pers. Soc. Psychol. 85, 1161. https:// doi. org/ 10. 1037/ 0022- 3514. 85.6. 1161 (2003).
40. Wright, H. F., Mills, D. S. & Pollux, P. M. Development and validation of a psychometric tool for assessing impulsivity in the
domestic dog (Canis familiaris). Int. J. Comp. Psychol. 24. https:// escho larsh ip. org/ uc/ item/ 7pb1j 56q (2011).
41. Vas, J., Topál, J., Péch, É. & Miklósi, Á. Measuring attention decit and activity in dogs: A new application and validation of a
human ADHD questionnaire. Appl. Anim. Behav. Sci. 103, 105–117. https:// doi. org/ 10. 1016/j. appla nim. 2006. 03. 017 (2007).
42. DuPaul, G. J., Power, T. J., Anastopoulos, A. D. & Reid, R. ADHD Rating Scale-IV: Checklist, Norms and Clinical Interpretations.
https:// psycn et. apa. org/ record/ 1998- 06605- 000 (Guilford Press, 1998).
43. Csibra, B., Bunford, N. & Gácsi, M. Evaluating ADHD assessment for dogs: A replication study. Animals 12, 807. https:// doi. org/
10. 3390/ ani12 070807 (2022).
44. Wright, H. F., Mills, D. S. & Pollux, P. M. J. Behavioural and physiological correlates of impulsivity in the domestic dog (Canis
familiaris). Physiol. Behav. 105, 676–682. https:// doi. org/ 10. 1016/j. physb eh. 2011. 09. 019 (2012).
45. Hejjas, K. et al. Molecular and behavioral analysis of the intron 2 repeat polymorphism in the canine dopamine D4 receptor gene.
Genes Brain Behav. 8, 330–336. https:// doi. org/ 10. 1111/j. 1601- 183X. 2008. 00475.x (2009).
46. Jiang, X., Liu, L., Ji, H. & Zhu, Y. Association of aected neurocircuitry with decit of response inhibition and delayed gratication
in attention decit hyperactivity disorder: A narrative review. Front. Hum. Neurosci. https:// doi. org/ 10. 3389/ fnhum. 2018. 00506
(2018).
47. Bódizs, R., Kis, A., Gácsi, M. & Topál, J. Sleep in the dog: Comparative, behavioral and translational relevance. Curr. Opin. Behav.
Sci. 33, 25–33. https:// doi. org/ 10. 1016/j. cobeha. 2019. 12. 006 (2020).
48. Bálint, A. et al. Potential physiological parameters to indicate inner states in dogs: e analysis of ECG, and respiratory signal
during dierent sleep phases. Front. Behav. Neurosci. https:// doi. org/ 10. 3389/ fnbeh. 2019. 00207 (2019).
49. Kis, A. et al. Development of a non-invasive polysomnography technique for dogs (Canis familiar is). Physiol. Behav. 130, 149–156.
https:// doi. org/ 10. 1016/j. physb eh. 2014. 04. 004 (2014).
50. Reicher, V. et al. Repeated aernoon sleep recordings indicate rst-night-eect-like adaptation process in family dogs. J. Sleep Res.
https:// doi. org/ 10. 1111/ jsr. 12998 (2020).
51. Reicher, V. et al. Developmental features of sleep electrophysiology in family dogs. Sci. Rep. 11, 22760. https:// doi. org/ 10. 1038/
s41598- 021- 02117-1 (2021).
52. Iotchev, I. B. et al. Age-related dierences and sexual dimorphism in canine sleep spindles. Sci. Rep. 9, 1–11. https:// doi. org/ 10.
1038/ s41598- 019- 46434-y (2019).
53. Pótári, A. et al. Age-related changes in sleep EEG are attenuated in highly intelligent individuals. Neuroimage 146, 554–560. https://
doi. org/ 10. 1016/j. neuro image. 2016. 09. 039 (2017).
54. Ujma, P. P. et al. Sleep spindles and intelligence: Evidence for a sexual dimorphism. J. Neurosci. 34, 16358–16368. https:// doi. org/
10. 1523/ JNEUR OSCI. 1857- 14. 2014 (2014).
55. Berry, R. B. et al. e AASM manual for the scoring of sleep and associated events. Rules terminology and technical specications,
Darien, Illinois. Am. Acad. Sleep Med. 176, 2012 (2012).
56. G ergely, A. et al. Reliability of family dogs’ sleep structure scoring based on manual and automated sleep stage identication.
Animals 10, 927. https:// doi. org/ 10. 3390/ ani10 060927 (2020).
57. Czeibert, K., Sommese, A., Petneházy, Ö., Csörgő, T. & Kubinyi, E. Digital endocasting in comparative canine brain morphology.
Front. Vet. Sci. https:// doi. org/ 10. 3389/ fvets. 2020. 565315 (2020).
58. Faul, F., Erdfelder, E., Buchner, A. & Lang, A.-G. Statistical power analyses using G* Power 3.1: Tests for correlation and regression
analyses. Behav. Res. Methods 41, 1149–1160. https:// doi. org/ 10. 3758/ BRM. 41.4. 1149 (2009).
59. Sobanski, E., Schredl, M., Kettler, N. & Alm, B. Sleep in adults with attention decit hyperactivity disorder (ADHD) before and
during treatment with methylphenidate: A controlled polysomnographic study. Sleep 31, 375–381. https:// doi. org/ 10. 1093/ sleep/
31.3. 375 (2008).
60. Platon, M. J. R., Bueno, A. V., Sierra, J. E. & Kales, S. Hypnopolygraphic alterations in attention decit disorder (ADD) children.
Int. J. Neurosci. 53, 87–101. https:// doi. org/ 10. 3109/ 00207 45900 89865 91 (1990).
61. Yoon, S. Y. R., Jain, U. R. & Shapiro, C. M. Sleep and daytime function in adults with attention-decit/hyperactivity disorder:
Subtype dierences. Sleep Med. 14, 648–655. https:// doi. org/ 10. 1016/j. sleep. 2013. 03. 003 (2013).
62. Kirov, R. & Brand, S. Sleep problems and their eect in ADHD. Exp. Rev. Neurother. 14, 287–299. https:// doi. org/ 10. 1586/ 14737
175. 2014. 885382 (2014).

10
Vol:.(1234567890)
Scientic Reports | (2023) 13:1291 | https://doi.org/10.1038/s41598-023-28263-2
www.nature.com/scientificreports/
63. Konofal, E., Lecendreux, M., Bouvard, M. P. & Mouren-Simeoni, M.-C. High levels of nocturnal activity in children with attention-
decit hyperactivity disorder: A video analysis. Psychiatry Clin. Neurosci. 55, 97–103. https:// doi. org/ 10. 1046/j. 1440- 1819. 2001.
00808.x (2001).
64. Lee, H. K. et al. Sleep and cognitive problems in patients with & attention-decit hyperactivity disorder. Neuropsychiatr. Dis. Treat.
https:// doi. org/ 10. 2147/ NDT. S69562 (2014).
65. Haba-Rubio, J., Ibanez, V. & Sforza, E. An alternative measure of sleep fragmentation in clinical practice: e sleep fragmentation
index. Sleep Med. 5, 577–581. https:// doi. org/ 10. 1016/j. sleep. 2004. 06. 007 (2004).
66. Bunford, N. et al. Dierences in pre-sleep activity and sleep location are associated with variability in daytime/nighttime sleep
electrophysiology in the domestic dog. Sci. Rep. 8, 1–10. https:// doi. org/ 10. 1038/ s41598- 018- 25546-x (2018).
67. Reicher, V., Kis, A., Simor, P., Bódizs, R. & Gácsi, M. Interhemispheric asymmetry during NREM sleep in the dog. Sci. Rep. 11,
18817. https:// doi. org/ 10. 1038/ s41598- 021- 98178-3 (2021).
68. Bolló, H. et al. REM versus Non-REM sleep disturbance specically aects inter-specic emotion processing in family dogs (Canis
familiaris). Sci. Rep. 10, 1–8. https:// doi. org/ 10. 1038/ s41598- 020- 67092-5 (2020).
69. Salari, R., Bohlin, G., Rydell, A.-M. & orell, L. B. Neuropsychological functioning and attachment representations in early
school age as predictors of ADHD symptoms in late adolescence. Child Psychiatry Hum. Dev. 48, 370–384. https:// doi. org/ 10.
1007/ s10578- 016- 0664-1 (2017).
70. Gregory, A. M., Agnew-Blais, J. C., Matthews, T., Mott, T. E. & Arseneault, L. ADHD and sleep quality: Longitudinal analyses
from childhood to early adulthood in a twin cohort. J. Clin. Child Adolesc. Psychol. 46, 284–294. https:// doi. org/ 10. 1080/ 15374
416. 2016. 11834 99 (2017).
71. Le Bon, O. et al. e rst-night eect may last more than one night. J. Psychiatric Res. 35, 165–172. htt ps:// doi. org/ 10. 1016/ S0022-
3956(01) 00019-X (2001).
72. Bunford, N., Csibra, B. & Gácsi, M. Individual dierences in response to ambiguous stimuli in a modied Go/No-Go paradigm
are associated with personality in family dogs. Sci. Rep. 9, 11067. https:// doi. org/ 10. 1038/ s41598- 019- 47510-z (2019).
Acknowledgements
We are very grateful to Anna Egerer, Anna Gergely, Barbara Csibra, Dóra Szabó, Enikő Kovács, Ivaylo B. Iotchev,
Katinka Tóth, Sára Szakadát and Zsóa Bognár for assisting with recruiting subjects, recordings (behavior and/
or EEG) and technical questions. Borbála Turcsán provided invaluable help in organizing the PSG database.
We thank all dog owners participating in our study. is research has been implemented with the support
provided by the Ministry of Innovation and Technology of Hungary from the National Research, Development
and Innovation Fund (FK 128242, K 132372, ÚNKP-21-3 & ÚNKP-21-5 New National Excellence Program);
the MTA-ELTE Comparative Ethology Research Group (01031); the János Bolyai Research Scholarship of the
Hungarian Academy of Sciences; the Stipendium Hungaricum Program; the Eötvös Hungarian State Scholarship.
Author contributions
Conceptualization: C.C. and M.G; data curation: C.C, V.R., A.K. and M.G; formal analysis: C.C. and V.R.; funding
acquisition: M.G.; investigation: C.C., V.R. and A.K.; methodology: C.C. and M.G.; project administration: C.C.
and M.G.; resources: C.C., V.R., A.K. and M.G; supervision: M.G.; validation: C.C. and V.R.; visualization: C.C.;
writing - original dra: C.C., V.R. and M.G.; writing - review & editing: C.C., V.R., A.K. and M.G. All authors
have read and agreed to the published version of the manuscript.
Funding
Open access funding provided by Eötvös Loránd University.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary Information e online version contains supplementary material available at https:// doi. org/
10. 1038/ s41598- 023- 28263-2.
Correspondence and requests for materials should be addressed to C.C.
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 is 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
Creative Commons licence, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted 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 licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
© e Author(s) 2023