Effects of human–dog interactions on brain activation

Abstract

Human-dog interaction brings meaningful emotional support and health advantages, highlighting its positive effects on human well-being. However, little is known about the brain activity during these interactions. This study investigated the effect of interacting with a dog on brain activity using electroencephalography. Results revealed that petting a dog and making eye contact stimulates higher brain activity in the beta and gamma frequency bands compared to touching a stuffed toy dog. The findings of this study indicate that interactions with a dog can improve attention and concentration. Our study is noteworthy for providing fundamental data to investigate human brain activity during interaction with dogs. Human-dog interactions can serve as beneficial interventions that lead to various physiological responses and emotional benefits. Based on these results, it would be meaningful to develop more structured animal-assisted interventions customized to improve attention.

Introduction

Humans are social beings who engage in ongoing interpersonal interactions and validate their existence through socializing with one another¹. Engagement extends beyond relationships among humans, and many studies have shown that interaction with animals also brings meaningful support and positive effects to humans^2,3. Dogs are one of the most frequently discussed animals in human-animal interaction research. Dogs are able to communicate with humans^4,5 and empathize with human emotions⁶, leading to the development of lasting bonds as faithful companions beyond mere practical usefulness. Despite certain inconsistencies and the need for further research, positive interactions between humans and dogs can lead to mutual physiological changes^7,8, resulting in a variety of health benefits for both physical and mental well-being.

Considerable attention has been paid to the health effects of interactions with dogs, yet there remains a gap in the literature regarding the study of brain activity in such interactions, with very limited research⁹. Studies have mainly focused on vital signs, endocrinological parameters⁹, and subjective evaluations of emotional changes^10,11. This creates the need for in-depth research on the correlation between brain activity and human-dog interactions. The thoughts, emotions, and behavioral responses of an individual are caused by communication between nerve cells in the brain. Mental processes influence physiological responses, which in turn provide feedback to influence thoughts and feelings¹², and this is reflected in the signature brainwave patterns¹³. Brainwaves are analyzable signals in the form of waves created by the brain’s activity. Electroencephalography is a frequently used noninvasive technique that provides a direct and accurate measure of electrical brain activity¹⁴, enabling the investigation of psychological mechanisms underlying perception and behavior through the analysis of real-time data¹⁵. It has been reported that stroking dogs may result in a reduction of stress levels¹⁶, an increase in neural activity in the prefrontal cortex¹⁷, and higher brain activity in the relative low beta power spectrum in the prefrontal, frontal, parietal, and occipital lobes¹⁸.

Therefore, we conducted an experiment to examine changes in brainwave activity during interactions involving physical contact with a dog, specifically petting the dog with or without eye contact, compared to petting a stuffed toy dog. The overall objective of this study is to investigate whether engaging in deep interactions, such as petting a dog while making eye contact, leads to an increase in brain activity in specific frequencies.

Materials and methods

Participants

A recruitment notice containing a brief description of the study and participation criteria such as age range, fear of dogs, and health status was posted in pet grooming salons and dog grooming schools in Seongnam, South Korea. Participants were recruited in March 2025, and the study was conducted that same month. A total of 32 eligible adult participants were initially recruited for the study, but one canceled the appointment and another withdrew before the experiment began, resulting in a final sample of 30 participants. The study included healthy adult participants in the age range of 20 to 60 years old without allergies or fear of dogs. Participants with a medical history of cardiovascular diseases (such as hypertension, unstable angina, heart attack, or cardiac surgery), those with mental illnesses such as depressive or anxiety disorders currently using relevant medications, and pregnant or lactating women were excluded according to the criteria outlined in the recruitment notice. Participants were requested to refrain from smoking or consuming caffeinated beverages within three hours prior to the activity in order to prevent any potential stimulating effects¹⁹. Before participating in the experiment, participants were provided with a thorough explanation of the study procedures and precautions, and informed consent forms were obtained. Informed consent to publish identifying images in an online open-access publication was obtained from participants. Demographic data, including age, gender, height, and weight, were recorded after obtaining informed consent. Collected data were recorded and maintained using numerical identifiers to protect participant anonymity. As an incentive for participants, a product valued at KRW 15,000 was provided. The protocol was designed in accordance with the Declaration of Helsinki, the Bioethics and Safety Act of South Korea, and the standard operating procedures of the Institutional Review Board of Konkuk University. The protocol and experimental guidelines were reviewed and approved by the Konkuk University Institutional Review Board, with reference number KKUIRB-202,502-HR-041. All methods were performed in accordance with the relevant guidelines and regulations.

Dog

A six-year-old female standard poodle was involved in the study. The dog had an appropriate temperament for this type of task, with a calm attitude and enjoyed interacting with unfamiliar people. She was fully trained in basic obedience, manners, aggression control, and sociability, and qualified as an animal-assisted activity dog after passing the Korean Kennel Club certification evaluation. The dog received vaccinations and underwent routine veterinary examinations for a health assessment. The dog’s health was carefully monitored to ensure there were no diseases or parasites, and it was not provided with raw meat or any other sources of uncooked protein. She was regularly bathed and groomed both before and during the research period to maintain cleanliness. Following consultation with a professional dog handler, a schedule of approximately four exercise sessions per day was determined to be suitable in order to prevent overworking the dog, taking into account its daily average exercise requirements and current health status. Accordingly, the session was restricted to a maximum of four times daily, with each session lasting approximately 60 min, including six minutes of activities involving dogs. An experienced dog handler was on standby to be prepared for any unexpected situation, and to control and protect the dog. There were no instances of discomfort experienced by either the participants or the dogs throughout the duration of the experiment. The protocol was developed in adherence to the guidelines established by the International Association for Human-Animal Interaction Organizations (IAHAIO)²⁰. All methods were performed in accordance with the relevant guidelines and regulations. We have confirmed that this study does not require approval from the Konkuk University Institutional Animal Care and Use Committee, as it did not involve experiments with a participating dog, invasive procedures, or medicinal treatments. The dog is owned by the lead researcher, and consent was obtained for all research activities involving the dog.

Experimental conditions

The experiment took place in a room measuring 9.7 m x 3.7 m, simply designed with white walls and a ceiling. Our priority was to select the experimental setting to be familiar and reassuring for the participating dog. This place was where the dog spent time on a daily basis. The place provided sufficient space for the activity, and the dog had a separate place to rest when not participating. It was silent and blocked outside noise to minimize any possible influence. The room conditions were consistently regulated, with an average temperature of 21.4 ± 0.7 °C, humidity at 34.4 ± 5.6%, and illuminance measuring 581.9 ± 66.6 lx.

Experimental procedure

This study had a within-subject design with repeated measurements. Each participant engaged in four types of contact activities during a single session visit (Fig. 1). To prevent the order of activities from influencing the results, the sequence was randomized. The activities include petting a dog while making eye contact (DOG_EO), petting a dog without visual interaction with eyes closed (DOG_EC), petting a stuffed toy dog with eyes open (SDOG_EO), or petting a stuffed toy dog with eyes closed (SDOG_EC) (Fig. 2). The commonly recognized design of a stuffed toy dog was used in this study (IKEA, GOSIG GOLDEN Soft Toy, 70 cm). A detailed explanation of each activity was presented before the experiment.

Fig. 1.

Experimental procedure. EEG, electroencephalogram; Resting_EO, resting with eyes open; Resting_EC, resting with eyes closed; DOG_EO, pet a dog while making eye contact; DOG_EC, pet a dog with eyes closed; SDOG_EO, pet a stuffed toy dog with eyes open; SDOG_EC, pet a stuffed toy dog with eyes closed.

Fig. 2.

The appearance of activities performed by the participants. (a) DOG_EO, pet a dog while making eye contact. (b) DOG_EC, pet a dog with eyes closed. (c) SDOG_EO, pet a stuffed toy dog with eyes open. (d) SDOG_EC, pet a stuffed toy dog with eyes closed.

Prior to commencement of the activity, the participants rested by sitting in a chair to minimize stimulation. They stared at a wall for three minutes, and then closed their eyes for three minutes. Participants performed each activity once for three minutes in a random order and were instructed to refrain from speaking or making sudden movements while their brainwaves were being measured. Participants were asked to fill out a subjective emotions assessment questionnaire immediately following each activity to report their emotional states. Before performing the next activity, participants rested in a chair for five minutes. The entire experimental procedure took approximately 60 min for each participant. There were no occurrences of unintended or unfavorable incidents throughout the process of data collection.

Electroencephalogram measurements

EEG data were collected using a wireless dry device (Quick-20; Cognionics, San Diego, CA) to measure the brainwave patterns of each participant during the activities (Fig. 3A). This device is equipped with a dry electrode system that reduces the risk of electric shock compared to wet electrode systems that use an electrolyte gel, and allows for easy removal from the scalp if participants encounter any discomfort. This device is primarily used in the field of neuroscience and is certified as safe by the European Commission and the Federal Communications Commission. Potential differences are determined by amplifying the measured electric signals and collecting the processed data from the dry electrode positioned in contact with the scalp. EEG signals were collected from eight channels: the left prefrontal lobe (Fp1), right prefrontal lobe (Fp2), left frontal lobe (F3), right frontal lobe (F4), left parietal lobe (P3), right parietal lobe (P4), left occipital lobe (O1), and right occipital lobe (O2), with a reference electrode (A1) placed on the left earlobe according to the International 10–20 system²¹(Fig. 3B).

Fig. 3.

Electroencephalogram measurements. (a) Wearing the wireless dry electroencephalography device (Quick-20; Cognionics, San Diego, CA). (b) International 10–20 system of electrode placement. Highlighted sites (Fp1, Fp2, F3, F4, P3, P4, O1, and O2) indicate those measured and analyzed in this study. Fp1, left prefrontal lobe; Fp2, right prefrontal lobe; F3, left frontal lobe; F4, right frontal lobe; P3, left parietal lobe; P4, right parietal lobe; O1, left occipital lobe; O2, right occipital lobe; A1, reference electrode on the left earlobe.

Subjective emotions assessment

The Semantic Differential Method (SDM) was assessed after each activity stimulation. The SDM is a questionnaire for examining the impact of environmental variables on participants’ emotional states, developed by Osgood in 1952²². This questionnaire comprises three adjective questions in which respondents choose between “comfortable-uncomfortable,” “natural-artificial,” and “relaxed-awakening.” These questions are assessed using a 13-point Likert scale, ranging from − 6 to 6. The higher result value corresponds to a more positive emotional state.

Data processing and analysis

The analysis of EEG measurements was performed using Bioteck Analysis Software (Bioteck; Daejeon, South Korea). EEG data were frequency-filtered from 0.5 Hz high-pass to 100 Hz low-pass using a built-in amplifier and brain mapping program, and captured at a sampling rate of 1 kHz. The EEG waveform records changes in neural electrical activity within the cortex and appears in various frequency bands that represent different states of the brain. Depending on the frequency band, brainwaves are classified into delta, theta, alpha, beta, and gamma waves²³. Generally, lower frequency bands are active during states of stability and relaxation, whereas higher frequency bands are active during states of increased activity or arousal. The amount of the frequency band is expressed as power, reflecting the increase and decrease in EEG power values within the cortex²⁴. EEG spectral analysis is one of the primary methods used in neuroscience to quantify the level of oscillatory activity across different frequencies within EEG recordings²⁵. Spectral analysis was used to identify the relative power within various frequency bands from the collected raw EEG data: alpha (8–13 Hz / total 4–50 Hz), slow alpha (8–11 Hz / total 4–50 Hz), fast alpha (11–13 Hz / total 4–50 Hz), beta (13–30 Hz / total 4–50 Hz), low beta (12–15 Hz / total 4–50 Hz), mid beta (15–20 Hz / total 4–50 Hz), high beta (20–30 Hz / total 4–50 Hz), and gamma (30–50 Hz / total 4–50 Hz).

The analysis was conducted to examine changes in brainwaves and emotional states during the activity of petting a dog with or without eye contact, compared to petting a stuffed toy dog. Descriptive statistical analyses of the EEG data and subjective emotional state data were performed using SPSS (version 28 for Windows; IBM Corp., Armonk, NY, USA) through a one-way repeated measures analysis of variance (ANOVA). Multiple comparisons of the main effects across activities were conducted using Bonferroni correction; the activities are listed in order of significance (a > b > c > d). A p-value of < 0.05 was considered statistically significant. Demographic information such as sex, age, height, and weight was analyzed using Microsoft Excel (Microsoft Office 365 ProPlus; Microsoft, Redmond, WA, USA) to generate descriptive statistics, including the mean and standard deviation (SD).

Results

Demographic characteristics

A total of 32 healthy adult participants who met the eligibility criteria were initially recruited for the study, with 30 participants completing the experiment. The average age was 26.0 ± 8.4 years, with an equal distribution of 15 men and 15 women. Their average height and weight were 169.3 ± 9.0 cm and 65.1 ± 13.5 kg, respectively.

Electroencephalogram

The electroencephalogram (EEG) results of relative alpha (RA) and relative slow alpha (RSA) power spectrum analysis indicated significantly higher activity in the prefrontal lobes, frontal lobes, parietal lobes, and right occipital lobe while participants were petting a stuffed toy dog with their eyes closed (p < 0.001). Additionally, petting a dog with eyes closed, as well as petting a stuffed toy dog with eyes closed, is associated with increased activation in the left side of the occipital lobe (p < 0.001). The relative fast alpha (RFA) power spectrum analysis indicated a statistically significant increase in activity in the prefrontal and occipital lobes during the act of either petting a dog or a stuffed toy dog under conditions where the eyes are closed (p < 0.001). However, in the frontal and parietal lobes, it was higher while petting the stuffed toy dog with eyes closed (p < 0.001) (Table 1).

Table 1.

Results of the relative alpha (RA), relative slow alpha (RSA), and relative fast alpha (RFA) power spectral analyses via EEG based on the activities. *** = significant at p < 0.001 via One-way Repeated Measures ANOVA. Multiple comparisons a > b > c > d via Bonferroni correction. SD = standard deviation; df = degree of freedom; DOG_EO, pet a dog while making eye contact; DOG_EC, pet a dog with eyes closed; SDOG_EO, pet a stuffed toy dog with eyes open; SDOG_EC, pet a stuffed toy dog with eyes closed; Fp1, left prefrontal lobe; Fp2, right prefrontal lobe; F3, left frontal lobe; F4, right frontal lobe; P3, left parietal lobe; P4, right parietal lobe; O1, left occipital lobe; O2, right occipital lobe.

Activity	Mean ± SD
	Fp1	Fp2	F3	F4	P3	P4	O1	O2
RA
DOG_EO	0.137 ± 0.029 c	0.144 ± 0.032 c	0.146 ± 0.020 d	0.148 ± 0.020 d	0.148 ± 0.021 d	0.156 ± 0.021 d	0.146 ± 0.028 c	0.150 ± 0.024 c
DOG_EC	0.203 ± 0.051 b	0.199 ± 0.045 b	0.212 ± 0.051 b	0.213 ± 0.052 b	0.221 ± 0.065 b	0.236 ± 0.064 b	0.230 ± 0.070 a	0.229 ± 0.069 b
SDOG_EO	0.149 ± 0.025 c	0.154 ± 0.025 c	0.160 ± 0.023 c	0.162 ± 0.019 c	0.161 ± 0.025 c	0.168 ± 0.026 c	0.162 ± 0.034 b	0.154 ± 0.028 c
SDOG_EC	0.234 ± 0.056 a	0.238 ± 0.053 a	0.254 ± 0.059 a	0.255 ± 0.059 a	0.267 ± 0.063 a	0.283 ± 0.058 a	0.238 ± 0.070 a	0.266 ± 0.071 a
F (df)	48.314 (1.871)	46.993 (1.961)	61.922 (2.134)	55.688 (1.879)	63.276 (1.785)	76.422 (1.926)	30.643 (2.116)	57.757 (2.036)
P	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***
RSA
DOG_EO	0.092 ± 0.023 d	0.097 ± 0.024 c	0.094 ± 0.016 d	0.096 ± 0.016 d	0.093 ± 0.015 d	0.098 ± 0.016 d	0.091 ± 0.021 c	0.092 ± 0.018 c
DOG_EC	0.146 ± 0.048 b	0.143 ± 0.045 b	0.153 ± 0.051 b	0.154 ± 0.050 b	0.150 ± 0.061 b	0.158 ± 0.060 b	0.160 ± 0.065 a	0.156 ± 0.062 b
SDOG_EO	0.102 ± 0.020 c	0.106 ± 0.020 c	0.107 ± 0.018 c	0.108 ± 0.015 c	0.103 ± 0.018 c	0.108 ± 0.019 c	0.103 ± 0.024 b	0.096 ± 0.020 c
SDOG_EC	0.176 ± 0.057 a	0.179 ± 0.054 a	0.191 ± 0.059 a	0.190 ± 0.059 a	0.184 ± 0.065 a	0.194 ± 0.064 a	0.166 ± 0.066 a	0.187 ± 0.066 a
F (df)	38.896 (1.605)	38.236 (1.741)	50.898 (1.845)	47.247 (1.703)	38.599 (1.483)	44.188 (1.669)	24.118 (1.990)	44.773 (1.799)
P	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***
RFA
DOG_EO	0.045 ± 0.009 b	0.048 ± 0.010 b	0.051 ± 0.006 c	0.052 ± 0.007 c	0.055 ± 0.007 c	0.058 ± 0.007 c	0.055 ± 0.007 b	0.057 ± 0.007 b
DOG_EC	0.057 ± 0.015 a	0.056 ± 0.009 a	0.059 ± 0.012 ab	0.059 ± 0.012 ab	0.071 ± 0.016 b	0.078 ± 0.022 b	0.070 ± 0.019 a	0.073 ± 0.026 a
SDOG_EO	0.047 ± 0.006 b	0.049 ± 0.007 b	0.054 ± 0.005 bc	0.054 ± 0.006 bc	0.058 ± 0.008 c	0.060 ± 0.009 c	0.059 ± 0.011 b	0.058 ± 0.010 b
SDOG_EC	0.057 ± 0.011 a	0.059 ± 0.010 a	0.064 ± 0.014 a	0.064 ± 0.014 a	0.083 ± 0.025 a	0.089 ± 0.025 a	0.072 ± 0.023 a	0.079 ± 0.027 a
F (df)	15.241 (2.330)	16.148 (2.302)	12.065 (2.113)	14.653 (2.160)	29.262 (1.705)	33.180 (1.672)	13.730 (1.758)	15.679 (1.636)
P	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***

The EEG results of the relative beta (RB) power spectrum analysis showed that brain activity in the prefrontal lobes is high when petting a dog under conditions where the eyes are closed (p < 0.001). No significant differences were found in the frontal lobes. Activity in the right parietal lobe (p < 0.01) and right occipital lobe (p < 0.001) was higher when petting a stuffed toy dog with eyes open. However, the left parietal lobe showed higher activity both when petting a dog while making eye contact and when petting a stuffed toy dog with open eyes (p < 0.01), and the left occipital lobe showed higher activity when petting a dog while making eye contact (p < 0.001). The relative low beta (RLB) power spectrum analysis showed that brain activity was high in both sides of the prefrontal lobes when petting a dog with eyes closed (p < 0.01, p < 0.05, left and right, respectively). No notable variations in the RLB power spectrum were detected in the other regions of the brain. The relative mid beta (RMB) power spectrum analysis showed a significant increase in the prefrontal (p < 0.001) and frontal (p < 0.01, p < 0.001, left and right, respectively) lobes when participants closed their eyes and petted either a real dog or a stuffed toy dog. However, no significant differences in the RMB power spectrum were observed in the parietal and occipital lobes (Table 2).

Table 2.

Results of the relative beta (RB), relative low beta (RLB), and relative mid beta (RMB) power spectral analyses via EEG based on the activities. NS = non-significant; *, **, *** = significant at p < 0.05, 0.01, and 0.001 via One-way Repeated Measures ANOVA. Multiple comparisons a > b > c via Bonferroni correction. SD = standard deviation; df = degree of freedom; DOG_EO, pet a dog while making eye contact; DOG_EC, pet a dog with eyes closed; SDOG_EO, pet a stuffed toy dog with eyes open; SDOG_EC, pet a stuffed toy dog with eyes closed; Fp1, left prefrontal lobe; Fp2, right prefrontal lobe; F3, left frontal lobe; F4, right frontal lobe; P3, left parietal lobe; P4, right parietal lobe; O1, left occipital lobe; O2, right occipital lobe.

Activity	Mean ± SD
	Fp1	Fp2	F3	F4	P3	P4	O1	O2
RB
DOG_EO	0.300 ± 0.063 bc	0.299 ± 0.063 bc	0.335 ± 0.042	0.337 ± 0.037	0.360 ± 0.025 a	0.358 ± 0.026 ab	0.365 ± 0.033 a	0.366 ± 0.029 ab
DOG_EC	0.337 ± 0.037 a	0.335 ± 0.036 a	0.340 ± 0.034	0.342 ± 0.037	0.351 ± 0.033 ab	0.349 ± 0.033 ab	0.343 ± 0.031 b	0.351 ± 0.031 b
SDOG_EO	0.293 ± 0.050 c	0.290 ± 0.048 c	0.337 ± 0.026	0.332 ± 0.028	0.361 ± 0.021 a	0.360 ± 0.024 a	0.349 ± 0.028 ab	0.370 ± 0.020 a
SDOG_EC	0.326 ± 0.042 ab	0.325 ± 0.036 ab	0.328 ± 0.035	0.333 ± 0.032	0.342 ± 0.032 b	0.340 ± 0.035 b	0.338 ± 0.035 b	0.340 ± 0.030 c
F (df)	9.775 (1.937)	9.612 (1.998)	0.903 (2.257)	0.745 (3)	5.812 (2.211)	5.721 (3)	6.516 (3)	15.824 (2.196)
P	0.000 ***	0.000 ***	0.421 NS	0.528 NS	0.004 **	0.001 **	0.000 ***	0.000 ***
RLB
DOG_EO	0.061 ± 0.011 b	0.065 ± 0.013 ab	0.071 ± 0.007	0.071 ± 0.009	0.076 ± 0.008 a	0.080 ± 0.009	0.077 ± 0.008	0.080 ± 0.008
DOG_EC	0.071 ± 0.016 a	0.070 ± 0.010 a	0.072 ± 0.009	0.073 ± 0.010	0.080 ± 0.011 a	0.083 ± 0.013	0.079 ± 0.010	0.080 ± 0.012
SDOG_EO	0.063 ± 0.007 ab	0.064 ± 0.007 b	0.073 ± 0.006	0.072 ± 0.007	0.079 ± 0.009 a	0.082 ± 0.009	0.079 ± 0.010	0.079 ± 0.010
SDOG_EC	0.067 ± 0.008 ab	0.069 ± 0.008 ab	0.071 ± 0.009	0.073 ± 0.011	0.083 ± 0.016 a	0.086 ± 0.015	0.077 ± 0.012	0.078 ± 0.012
F (df)	5.786 (2.361)	4.047 (2.082)	0.426 (3)	0.409 (3)	3.616 (1.839)	2.666 (1.708)	0.897 (3)	0.463 (3)
P	0.003 **	0.021 *	0.735 NS	0.747 NS	0.037 NS	0.087 NS	0.446 NS	0.709 NS
RMB
DOG_EO	0.091 ± 0.018 b	0.091 ± 0.018 b	0.103 ± 0.013 b	0.105 ± 0.012 b	0.111 ± 0.010	0.112 ± 0.010	0.115 ± 0.016	0.115 ± 0.016
DOG_EC	0.108 ± 0.016 a	0.108 ± 0.014 a	0.112 ± 0.013 a	0.113 ± 0.014 a	0.114 ± 0.012	0.115 ± 0.013	0.111 ± 0.010	0.111 ± 0.010
SDOG_EO	0.091 ± 0.013 b	0.091 ± 0.012 b	0.108 ± 0.009 ab	0.106 ± 0.010 b	0.114 ± 0.010	0.115 ± 0.010	0.111 ± 0.007	0.114 ± 0.008
SDOG_EC	0.110 ± 0.016 a	0.109 ± 0.015 a	0.113 ± 0.015 a	0.115 ± 0.013 a	0.114 ± 0.013	0.115 ± 0.015	0.109 ± 0.013	0.110 ± 0.011
F (df)	23.522 (3)	21.332 (3)	5.721 (3)	7.856 (3)	1.412 (3)	0.742 (2.227)	1.760 (3)	2.077 (2.305)
P	0.000 ***	0.000 ***	0.001 **	0.000 ***	0.245 NS	0.494 NS	0.161 NS	0.126 NS

The analysis of the relative high beta (RHB) power spectrum indicated higher brain activity in the prefrontal lobes when participants closed their eyes and petted a dog (p < 0.05). No significant differences were found in the frontal lobes. Activity in the parietal and occipital lobes was higher when participants were in the eyes open condition. The left parietal lobe (p < 0.001) and the left occipital lobe (p < 0.01) were mostly activated when petting a dog while making eye contact. The right parietal lobe was most activated when petting either a dog or a stuffed toy dog with eyes open (p < 0.001), whereas the right occipital lobe was most activated when petting a stuffed toy dog with eyes open (p < 0.001) (Table 3).

Table 3.

Results of the relative high beta (RHB) and relative gamma (RG) power spectral analyses via EEG based on the activities. NS = non-significant; *, **, *** = significant at p < 0.05, 0.01, and 0.001 via One-way Repeated Measures ANOVA. Multiple comparisons a > b > c via Bonferroni correction. SD = standard deviation; df = degree of freedom; DOG_EO, pet a dog while making eye contact; DOG_EC, pet a dog with eyes closed; SDOG_EO, pet a stuffed toy dog with eyes open; SDOG_EC, pet a stuffed toy dog with eyes closed; Fp1, left prefrontal lobe; Fp2, right prefrontal lobe; F3, left frontal lobe; F4, right frontal lobe; P3, left parietal lobe; P4, right parietal lobe; O1, left occipital lobe; O2, right occipital lobe.

Activity	Mean ± SD
	Fp1	Fp2	F3	F4	P3	P4	O1	O2
RHB
DOG_EO	0.170 ± 0.044 ab	0.166 ± 0.044 ab	0.185 ± 0.027 a	0.186 ± 0.024	0.200 ± 0.016 a	0.194 ± 0.018 a	0.200 ± 0.025 a	0.199 ± 0.020 ab
DOG_EC	0.183 ± 0.024 a	0.182 ± 0.024 a	0.182 ± 0.023 a	0.181 ± 0.024	0.188 ± 0.024 bc	0.184 ± 0.023 ab	0.183 ± 0.026 b	0.190 ± 0.025 bc
SDOG_EO	0.162 ± 0.039 b	0.158 ± 0.037 b	0.182 ± 0.018 a	0.179 ± 0.018	0.195 ± 0.015 ab	0.191 ± 0.014 a	0.188 ± 0.027 ab	0.205 ± 0.016 a
SDOG_EC	0.175 ± 0.028 ab	0.171 ± 0.023 ab	0.170 ± 0.025 a	0.172 ± 0.024	0.180 ± 0.023 c	0.175 ± 0.024 b	0.182 ± 0.028 b	0.183 ± 0.026 c
F (df)	3.424 (1.834)	4.204 (1.985)	3.569 (2.403)	2.356 (3)	9.901 (2.257)	8.932 (3)	4.833 (3)	12.465 (2.151)
P	0.044 *	0.020 *	0.026 NS	0.077 NS	0.000 ***	0.000 ***	0.004 **	0.000 ***
RG
DOG_EO	0.288 ± 0.078 a	0.280 ± 0.074 ab	0.306 ± 0.057 a	0.299 ± 0.056 a	0.334 ± 0.046 a	0.315 ± 0.051 a	0.330 ± 0.060 a	0.333 ± 0.052 a
DOG_EC	0.284 ± 0.058 ab	0.285 ± 0.050 a	0.271 ± 0.056 b	0.265 ± 0.054 ab	0.282 ± 0.058 c	0.268 ± 0.057 b	0.281 ± 0.068 b	0.298 ± 0.062 b
SDOG_EO	0.258 ± 0.062 b	0.247 ± 0.062 c	0.277 ± 0.043 b	0.267 ± 0.045 b	0.315 ± 0.039 b	0.299 ± 0.039 a	0.307 ± 0.085 a	0.344 ± 0.041 a
SDOG_EC	0.261 ± 0.042 ab	0.249 ± 0.038 bc	0.231 ± 0.045 c	0.227 ± 0.045 c	0.252 ± 0.051 d	0.235 ± 0.047 c	0.283 ± 0.073 b	0.280 ± 0.063 b
F (df)	3.122 (2.276)	5.963 (2.023)	18.177 (3)	14.128 (2.383)	27.668 (2.156)	27.061 (2.076)	6.609 (3)	22.623 (2.184)
P	0.044 *	0.004 **	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***	0.000 ***

The EEG results of the relative gamma (RG) frequency band power spectrum analysis showed higher activity in the prefrontal lobes when petting a dog. The left prefrontal lobe was more active when participants pet a dog while making eye contact, and the right prefrontal lobe was more active when participants petted a dog with their eyes closed (p < 0.05, p < 0.01, left and right, respectively). Significantly increased brain activity was observed in the frontal lobes (p < 0.001) and the left parietal lobe (p < 0.001) when petting a dog while maintaining eye contact. However, the right parietal lobe and both sides of the occipital lobes were activated when participants were petting either a dog or a stuffed toy dog under eyes open conditions (p < 0.001) (Table 3).

Semantic differential method (SDM)

The findings regarding the emotional responses associated with each activity indicated notable disparities across all variables pertaining to comfort, naturalness, and relaxation during tasks that incorporated an actual dog. Participants felt more comfortable (p < 0.001) and natural (p < 0.001) in both conditions when they petted the dog, whether by closing their eyes or making eye contact, compared to engaging in activities with a stuffed toy dog. Participants responded that they were most relaxed when petting a dog and making eye contact (p < 0.001) (Fig. 4).

Fig. 4.

Comparisons of the semantic differential method (SDM) results for each activity. (a) comfortable, (b) natural, and (c) relaxed. *** = significant at p < 0.001 via Repeated Measures ANOVA. Multiple comparisons: a > b > c via Bonferroni correction. Error bars represent 95% confidence intervals. DOG_EO, pet a dog while making eye contact; DOG_EC, pet a dog with eyes closed; SDOG_EO, pet a stuffed toy dog with eyes open; SDOG_EC, pet a stuffed toy dog with eyes closed.

Discussion

This study investigated brain activities and emotional responses according to the type of physical contact with dogs. The EEG records of healthy adults were analyzed while engaging in activities such as petting a dog with their eyes closed or making eye contact, as well as petting a stuffed toy dog with their eyes closed or open. The findings suggested that the presence of a dog or a stuffed toy dog, in conjunction with the conditions of eyes closed or making eye contact, elicited distinct brainwave patterns.

The brain activity in the RA and RSA power spectrum indices was high while petting a stuffed toy dog with eyes closed. Alpha waves are related to brain stability and relaxation and are commonly activated when the eyes are closed or during rest²⁶. The slow alpha wave occurs prior to the transition to a theta wave and generally represents a relaxed body state²⁷, reflecting meditative states and unresponsiveness to external stimuli²⁸. The increased RA and RSA power spectrum observed during the task in which participants closed their eyes and petted the stuffed toy dog can be interpreted as reflecting a state of deep relaxation and low cognitive demand.

The brain activity in the RFA power spectrum index showed high activity not only when petting a stuffed toy dog but also when petting a dog, in both conditions with eyes closed. The RFA power spectrum precedes the shift to beta waves, indicating increased levels of attention and focus during a state of relaxation²⁷. It can refer to the brain’s stable state for mental rotation tasks, such as cognitive judgment, learning, and creative thinking²⁹. This can be explained as a stage of relaxation accompanied by an increasingly immersive level of attention, where the reactions to touching the stuffed toy dog and the real dog are all reflected.

The beta band refers to fast wave activities that typically occur during states of enhanced consciousness, indicating increased attention and focus³⁰. Overall, beta band brain activity, including RLB and RHB power spectrum indices, was higher in the prefrontal lobes while petting a real dog with eyes closed. The RLB power spectrum appears during concentration and is associated with minimized unfocused attention³¹. The RLB power spectrum, also known as the sensorimotor rhythm (SMR)³², is associated with reaction time, psychomotor skills, and spatial abilities^33,34. However, the RMB power spectrum exhibited increased brain activity in both the prefrontal and frontal lobes when petting a real dog with eyes closed, as well as when petting a stuffed toy dog. The RMB power spectrum is associated with concentration³¹, as well as with problem-solving, interest in external objects and logical thinking³⁵. Higher brain activity observed in the RMB power spectrum when petting a real dog, as well as a stuffed toy dog, may be related to logical thinking, but further investigation is needed.

Despite some differences in the beta band frequencies and the brain regions activated, overall, the activity involved in interacting with a dog is noticeable. The frontal lobe plays a vital role in regulating physical movements³⁶, moral behavior, and various cognitive functions³⁷. The prefrontal cortex is a key functional region within the frontal lobe³⁸. The prefrontal lobe integrates input from various cortical regions to coordinate movement, cognitive processes, emotional responses, and social interactions, as well as to efficiently regulate thoughts, actions, and feelings³⁹. Interestingly, the power spectrum of RB and RHB in the parietal and occipital lobes was prominent during activities with eyes open. This result corresponds with a previous study that also indicated that the beta band power of EEG recorded at occipital sites is associated with visual attention⁴⁰. Moreover, our result is consistent with previous research that suggests visual stimuli not only activate the entire cortex but also specific brain regions involved in visual processing⁴¹. The parietal lobe is responsible for the integration and interpretation of various sensory inputs from the body⁴². The occipital lobe serves as the primary neural region for processing visual information, usually becoming active in response to visual stimuli^40,43.

In comparison to the outcomes of RA and RSA, engaging in the activity of petting an actual dog has been observed to affect the RFA and beta waves. This phenomenon could be attributed to heightened levels of attention while interacting with the dog. Additionally, the higher alpha and beta activity observed in the eyes closed condition is correlated with a previous study that reported a decrease in delta, alpha, and beta activities in the open eyes condition compared to the closed eyes condition⁴¹.

The RG power spectrum index exhibited increased activity in the prefrontal and frontal lobes while petting an actual dog, especially when making eye contact. The occurrence of gamma-band activity is reported to be closely related to visual stimulus processing⁴⁴. Gamma oscillations are widely regarded as functionally relevant signals in the brain⁴⁵. Functional mechanisms of binding, attention, and memory are all linked to brain activity within the gamma frequency range, which is a key cognitive process involved in accurately perceiving the external environment and guiding subsequent actions⁴⁶. Our results correlate with the participants’ interpretation of external information regarding a dog and their subsequent actions in approaching and engaging with the dog. Furthermore, previous studies have reported that increases in the high-beta and low-gamma frequency ranges are associated with positive feedback that activates the core areas of the reward network^47,48. This is related to our findings derived from our subjective assessments of emotional states. The results of the SDM showed that participants favored the activity with real dogs, experiencing more comfortable, natural, and relaxed feelings. Regarding interacting with an actual dog as a positive experience, our results suggest that the high level of RG activity observed during interactions with a dog not only indicates an increase in attention but also implies that participants find the experience rewarding and enjoyable.

This study investigated the impact of interactions with a dog on changes in brainwave activity, beyond the effects of tactile and visual sensory stimulation, by comparing these interactions to those with a stuffed toy dog. Focusing on the brainwave changes in the frontal and prefrontal lobes, which are involved in cognitive processes, emotional responses, and social interactions³⁹, our findings can be summarized as follows. The RA and RSA power spectrum indices were most activated when petting a stuffed toy dog with eyes closed, while the RFA and RMB were observed when petting both a stuffed toy dog and a real dog with eyes closed. The RB, RLB, and RHB power spectrum indices were higher when petting a real dog with eyes closed, and the RG power spectrum index showed increased brain activity in association with interacting with a real dog, especially when engaging in petting accompanied by eye contact. Overall, findings from the present study indicate that healthy adults exhibit increased brain activity associated with relaxation, similar to meditation, when petting a toy dog with their eyes closed, and heightened brain activity related to attention and focus when interacting with a real dog. Although certain frequencies of RFA and RMB exhibited increased brain activity in both the dog and stuffed toy dog conditions, a more comprehensive interpretation indicated that frequency ranges associated with attentional focus were all influenced by interaction with the dog. These findings, which reveal significant variations in brainwave patterns across frequencies, suggest that interaction with a dog, beyond mere tactile or visual stimulation, effectively stimulates brain activity.

Our results corresponded with previous studies, showing that children had faster beta brain activity after therapy dog sessions post-surgery⁴⁹, increased RFA power spectrum while horseback riding compared to mechanical exercises⁵⁰, higher frontopolar activity during stroking a miniature horse compared to a plush animal⁵¹, and exhibited more prefrontal brain activity when interacting with a dog than with a stuffed animal¹⁷. Furthermore, in our previous study comparing brainwaves during different activities with a dog, the RLB power spectrum was most active during massage and grooming the dog¹⁸. Our study showed a similar result, showing that the brainwave frequency ranges associated with attention and concentration were more closely correlated with activity involving a real dog. It is noticeable that gamma wave activation is closely connected not just with touching a dog, but also with making eye contact. This finding can also be related to another study that reported mutual gaze exchanges between dogs and humans increased oxytocin concentrations, fostering a stronger bond and enhancing positive interactions⁷.

Participants were able to immediately distinguish between the stuffed toy dog and a real dog, whether their eyes were open or closed during the task. Increased brainwave activity in the RA and RSA power spectra, which reflect a relaxed state with low cognitive demands during petting a stuffed toy dog, may be attributed to the lower physiological arousal required compared to petting a real dog, potentially leading to reduced feelings of tension. In this respect, a real dog represents a more complex stimulus than a stuffed toy dog, which may have played a role in eliciting increased brainwave activity associated with attention. Another possibility is that petting a real dog may require more complex motor adaptations from the participants, which could potentially contribute to the observed effects. Subjective assessments of emotional responses were significantly elevated during interactions with real dogs compared to those involving toy dogs. Thus, interacting with dogs may enhance participants’ emotional engagement more than interacting with stuffed animals. This heightened emotional engagement is associated with elevated prefrontal activity and may lead to enhanced attention¹⁷.

Ultimately, it’s becoming clearer that what increases attention and stimulates brain activity comes from interactions with living animals, not just tactile or visual stimulation. Interventions that increase activity in the gamma and beta bands have been used to enhance individual performance in cognitive control and memory modulation⁵². Based on these findings, it will be possible to develop an animal-assisted intervention program aimed at enhancing positive health outcomes by improving attention and concentration. Our findings suggest that interaction with a dog is associated with positive subjective experiences and brain activation related to attention.

Conclusion

Research investigating the relationship between brain activity mechanisms and the impact of human-dog interactions is currently limited, and the data is insufficient. The present study only includes healthy adult subjects, which limits the generalizability of our findings to a diverse set of populations. Another possible limitation of this study is that it primarily involved individuals who do not fear dogs, which may have introduced bias into the results. Nonetheless, our study is important as a new attempt to investigate the impact of interacting with dogs on human brain activity and represents a preliminary advance in this exploration.

In conclusion, the present study highlights that interactions with dogs can increase brain activity in the beta and gamma power spectra in healthy participants, thereby activating attention and concentration. Additionally, the subjective evaluation of emotions indicated that engagements with dogs elicited positive emotional responses. The interaction with dogs can serve as a beneficial intervention that leads to various physiological responses and emotional advantages. The study findings provide foundational evidence for developing more structured, customized animal-assisted interventions customized to improve attention. In future studies, verifying the validity of these findings and measuring the EEG correlation between brain regions to assess brain connectivity and function will be necessary to better understand the brain activating mechanisms behind the effects of human-animal interaction.

Author contributions

O.Y. conceived the study. O.Y., J.H., and S.P. developed the methodology. O.Y. conducted the investigation, data curation, formal analysis, and wrote the original draft. O.Y., J.H., and S.P. reviewed and edited the manuscript. J.H., and S.P. supervised the project. All authors reviewed and approved the final manuscript.

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (project no.: RS-2023-00217567).

Data availability

All data generated or analyzed during this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.