How Early Digital Experience Shapes Young Brains During 0-12 Years: A Scoping Review

ABSTRACT

Research Findings: Early digital experience (e.g. screen time and digital use) is believed to impact children’s brain development, functionally and structurally, but this impact has not been systematically reviewed. In this scoping review, we synthesized and evaluated 33 collected studies on children’s digital use (ages 0–12) and their associated brain development published between January 2000 and April 2023. The synthesis of the evidence revealed that (1) digital experience does have positive and negative impacts on children’s brains, structurally and functionally; (2) it could cause structural and functional changes in children’s frontal, parietal, temporal, and occipital lobes, brain connectivity, and brain networks; and the most vulnerable area is the prefrontal cortex and its associated executive function, and (3) early digital experience has both positive and negative impacts on children’s brain structure longitudinally. Practice or Policy: Educators and parents should be aware of the potential effects of digital experience on children’s brain development and provide appropriate guidance, mediation, and support for children’s digital use. Policymakers should establish and implement evidence-based policies and regulations to protect children’s digital well-being.

Digital devices are ubiquitous in young children’s daily lives and play a significant role in their early learning and development (Cao & Li, 2023; Dong et al., 2020, 2022; Wang et al., 2023). However, the impact of early digital experience on brain and behavioral development is controversial and has raised public concerns (World Health Organization, 2019) and research interests (Browne et al., 2020; Chassiakos et al., 2016; Kucirkova et al., 2018). The existing evidence has reported mixed and sometimes contradictory findings about its impact on executive function (EF), which refers to a set of cognitive processes that support goal-oriented behavior and adaptive responses to novel situations (Bustamante et al., 2023; Li et al., 2023). Some (Huber et al., 2018) suggested that active screen time (e.g., using interactive apps) could enhance EF, whereas some (McHarg et al., 2020) found that passive screen time (e.g., watching TV) could impair it. Furthermore, a recent meta-analysis of the existing behavioral evidence by Bustamante et al. (2023) found no significant results about its impact on EF, which challenges common sense and anecdotal evidence. These contradictory findings have jointly demonstrated a critical research gap: most existing studies focused on EF with behavioral approaches, lacking the impact on brain structure and brain function with neuroimaging evidence. This research gap has left a significant knowledge gap, preventing us from understanding how digital use shapes young brains and informing researchers, policymakers, and educators in their professional practice. Therefore, a scoping review is needed to synthesize the neuroimaging evidence of the relation between digital use and brain development in early childhood, taking into account the possible moderating effects of relevant characteristics. This review will address these fundamental questions and contribute to theoretical development, practical improvement, and policymaking in this field.

Digital Experience in Early Childhood

Digital devices are pieces of electronic hardware that can communicate, display, generate, receive, send, share, store, or process information with binary code. The family of digital devices is very comprehensive and inclusive, including (but not limited to) (1) cameras, web cameras, and other video devices; (2) microphones, earphones, and other audio devices; (3) phones, smartphones, and other telecommunication devices; (4) laptops, PCs, pads, tablets, and other computer devices; (5) scanners, printers, keyboards, and other PC peripherals; (6) smart watch, smart glasses, VR and other wearable devices; (7) hard disks, USB drives, and other storage devices; (8) thermometer, sphygmomanometer, and other digital measurements; and (9) robots, ChatGPT, and other educative and generative AI devices. This unexhausted list is extending drastically today, and so is its usage and significance in human learning and life, influencing children’s studies, entertainment, and social interactions. They must learn to employ various digital devices, such as cameras, pads, laptops, scanners, and printers, to read, observe, learn, and understand the world. This kind of experience of interacting with or engaging with digital devices could be labeled “digital experience,” an essential part of young children’s early learning and development experience (Harley et al., 2018; Kabali et al., 2015). In this study, all the interactions/using experiences with any digital devices are regarded as “digital experiences”.

Digital devices enable children to access various digital activities, such as games, internet browsing, and video watching, regardless of time and place (Ding & Li, 2023). This digital experience differs from the traditional learning experience that does not involve digital devices and may have various implications for child development. Some researchers have identified the following potential benefits of digital experience for children: (1) it facilitates children’s autonomous learning; (2) it creates online communities with diverse social interactions; (3) it develops children’s digital literacy skills that are essential for future success; (4) it promotes children’s mental well-being when used in moderation; and (5) it enhances children’s creativity and self-expression (Limone & Toto, 2021; Yadav & Chakraborty, 2021). In contrast, others have also pointed out the possible drawbacks of digital experience in children’s daily life (Huda et al., 2017), such as the risk of “digital addiction,” as young children (ages 0–6) are especially vulnerable to the negative effects of inappropriate digital experience (Ding & Li, 2023). For example, Hermawati et al. (2018) found that infants who spent more than 3 hours per day on screen-based activities had attention problems and hyperactivity symptoms. Undheim (2021) conducted a literature review of the empirical studies published in the last decade to examine the current trends and issues related to digital technology in early childhood education. Based on this review, they recommended that digital technology should be integrated into pedagogical practice in national early childhood curricula and teacher education programs. However, this literature review concentrated on the decade spanning 2010 to 2020 and overlooked the seminal studies from this millennium’s inaugural decade (2000–2010). There has been a lack of studies on this topic, thus necessitating the exploration of evidence beyond the confines of the current decade to the initial era when children encountered various digital devices. To bridge this knowledge gap, this scoping review endeavors to broaden the temporal parameters to encompass 2000 to 2023, synthesizing the neuroimaging findings surrounding neuroplasticity during the formative periods.

Neuroplasticity in the Early Years

Neuroplasticity is well-known as the property and capacity of the neural networks in human brains to change due to growth, experience, and environmental factors (Costandi, 2016). These changes may involve functional and structural ones ranging from the micro level, such as the individual neuron and neuronal circuit, to the macro level, such as the cortical region and neural network (see Ng et al., 2023). For instance, functional plasticity, which refers to the brain’s ability to alter and adapt the functional properties of neurons, can occur in four ways: homologous area adaptation, map expansion, cross-model reassignment, and compensatory masquerade (Grafman, 2000). Structural plasticity refers to the fact that the brain is malleable and can be changed in both circuit and network through learning activities, environmental influences, and training (Davidson & McEwen, 2012; Fuchs & Flügge, 2014; Goh & Park, 2009; Li et al., 2014; McEwen, 2018; Shaffer, 2016; Zhao & Li, 2010; Zimmerman et al., 2020). Both functional and structural neuroplasticity could be regarded as activity-dependent plasticity, which arises from cognitive functions and personal experience, including early digital experience.

Corresponding to “structural plasticity” and “functional plasticity,” early experience can shape brains structurally and functionally (Grafman, 2000). For instance, Li et al. (2014) systematically reviewed the existing studies on second language learning and bilingual experience and found structural brain changes in young children, including increased gray matter (GM) density and white matter (WM) integrity. Moreover, even short-term language learning or training could cause structural changes in young children’s brains (Li et al., 2014). These empirical studies jointly implied that daily experience could shape the brain in a certain way, which drove us to rethink the impact of digital experience in early childhood. Recently, Bustamante et al. (2023) conducted the first meta-analytic synthesis of existing evidence on the relation between overall screen time and EF in young children. However, they found no significant association between them or any moderators. These null results on the impact of digital experience on brain functions might not reflect the truth; instead, it might be caused by a small sample size (as only 15 articles were found even with no time limitation of search) and limited effect size (Bustamante et al., 2023). In addition, they only focused on the impact on EF with primarily behavioral data, leaving the brain structures and functions unexplored. Accordingly, there is a need to conduct a scoping review of the overall impact of digital experience on brain development.

The Current Study

The rapid development and widespread use of digital technology have raised global concerns about the potential effects of digital experience on young children’s cognitive and brain development. Some scholars have argued that digital experience changes the way children process information and alters their brain structures and functions (Helsper & Eynon, 2010; Prensky, 2001). Previous studies using behavioral measures have reported both positive and negative impacts of digital experience on children’s cognitive outcomes and behaviors, such as EF, attention-deficit symptoms, emotional and social intelligence, and technology addiction (Bustamante et al., 2023; Korte, 2020; Li et al., 2021; Small et al., 2020). However, there is a lack of systematic review of the neuroimaging evidence on how early digital experience affects the brain development of children, especially those aged 0–12 years, a sensitive period of high neural plasticity and rapid growth of executive attention (Best et al., 2009; Conway & Stifter, 2012; Horowitz-Kraus et al., 2016; Zelazo & Carlson, 2012). This sensitive period for neuroplasticity is a window of time when the brain is more responsive to environmental experiences and can undergo significant changes in its structure and function. For example, the sensitive period for visual development is 0–8 years, while that for language acquisition is 0–12 years (Gabard-Durnam & McLaughlin, 2020). Therefore, this study aimed to synthesize the existing neuroimaging evidence on the impact of early digital experience on young brains during the 0–12 years to fully understand the brain’s plasticity. Moreover, the “early” digital exposures, generally pertaining to those obtained during the preschool years (ages 0–6), may have a prolonged effect that could span a few years until early adolescence (Aged 10–12). As a result, in this research, we have expanded the age criterion to include ages 0–12.

Accordingly, we conducted a scoping review, which is a suitable method to explore the breadth, extent, and nature of the available research on this emerging and controversial topic and identify the knowledge gaps and research needs in this field. This scoping review will not only systematically categorize and summarize the findings according to the types of digital activities, outcomes of interest, methods used, and populations studied but also provide new insights and implications for parents, educators, and other stakeholders who are concerned about children’s digital use and suggest future directions for integrating neuroimaging technology into early childhood studies. The following research questions guided this study:

1. Whether or not there is an impact of digital experience on children’s brain structures and functions? If yes, is it a positive or a negative effect?

2. What brain areas are affected by the digital experience?

3. Is the impact of digital experience short-term or long-term?

Method

This study has adopted the framework of a scoping review proposed by Arksey and O’Malley (2005): 1) identifying the research question, 2) identifying the relevant studies, 3) studying the selection, 4) charting the data; and 5) collating, summarizing and reporting results. Through these five stages, this study was expected to identify the types of neural technologies being used in the investigation of the relevant studies, the categories of the impact of digital use on children’s brains (structural vs. functional; positive vs. negative), the affected brain areas being focused among the existing literature; as well as the length of time of which the specific kind of impact lasts.

Identifying Relevant Studies

Two search strategies were used to identify the relevant studies: a database search and a hand search. First, the database search was conducted with Google Scholar, Scopus, Web of Science, ProQuest, PubMed, and EBSCO Host to search and identify electronic resources. This search covered the period between January 2000 to April 2023. Next, the first author conducted a hand search to find the references not found through the database searches. This manual process inspects and identifies more relevant studies by checking the reference lists of identified articles and documents. This 2-step search aimed to thoroughly identify all the research articles on “digital use/experience on children’s brain development” published in the first two decades of this millennium.

Literature Search Strategy

The following search terms with two Boolean operators (“AND” and “OR”) were used in the database search: (“digital devices” OR “screen use” OR”screen time” OR “television” OR “computer” OR “laptop” OR “tablet” OR “smartphone” OR “video games” OR “online games” OR “E-reader” OR “artificial intelligent devices” OR “virtual reality devices” OR “technology”) AND (“children” OR “pre-schooler” OR “preschool-aged” OR “infant” OR “toddler” OR “kindergartner” OR “early childhood”) AND (a set of brain functional and structural terms such as “executive function” OR “inhibitory control” OR “working memory” OR “white matter area” OR “gray matter area”).

During the database search, 2621 records with related topics were identified. After removing 1078 duplicated items, 1422 articles remained after this search stage (see Figure 1: PRISMA Flow Diagram). The first author and the team screened all the records to assess eligibility. Accordingly, 26 articles were confirmed to meet the criteria. And the follow-up hand search identified another seven related journal articles. Finally, 33 potential articles were identified for this scoping review.

Figure 1. PRISMA flow diagram of literature search and data charting.

Selecting Studies

For this scoping review, we used a set of criteria to ensure that only full-text, peer-reviewed journal articles were included. Table 1 presents the inclusion and exclusion criteria adopted in this study:

Table 1. Inclusion/Exclusion criteria.

Criterion	Inclusion	Exclusion
Scope of research	Empirical studies	Not empirical studies Secondary data analysis Literature reviews, editorials, position statements, perspectives
Type of documents	Peer-reviewed scholarly journal articles	Not Peer-reviewed scholarly journal articles
Language	English	Other languages other than English
Educational level	Children age from 0–12 years old	No age mentioned in the article Children age over 12 years old Children with special need
Period	January, 2000-March, 2023	Prior to 2000 and after 2022
Research topic	The impact of digital use on children’s brain	Articles that did not focus on the impact on children’s brain Results only present behavior results

Based on the inclusion and exclusion criteria, the first and last authors screened the defined 1422 papers, respectively. Title and Abstract screening led to the exclusion of 57 articles, with 1307 remaining. Age criteria led to the exclusion of 110 articles, Children with SEN criteria led to 692 exclusions of findings, no digital use criteria led to 45 exclusion of results, and no impact on brain criteria led to 460 exclusions, with 26 remaining. By hand-searching the reference list, seven additional articles were included, resulting in 33 articles included at the end. The two screeners agreed on 33 out of 33 of the paper screenings, and no disagreement was found. Therefore, the reliability of this paper screening was 100%, which means perfect agreement on the screening result.

After this step, the 33 remaining articles were assessed for eligibility. Full-text article screening was conducted independently by each of the four authors, and disagreements were resolved through Zoom meeting discussions among the four authors. As a result, 33 articles were finalized for inclusion in the review. The entire process of selecting the relevant studies is detailed in the Prisma Flow diagram in Figure 1.

Charting the Data

Two processes drove the action of charting the selected data. Firstly, for each selected study, the following information was extracted: author(s), title, year of publication, journal title, country, participants’ characteristics (gender, age, sample size), school settings, research design, data collection method(s), type of neuroimaging technology, type of digital experience investigated in the study, the measure used to assess the construct(s), the specific task (s) (applied in functional studies), the studied outcomes of brain development, and a summary of the results. Two synthesized summaries of all this information are presented in Tables 2 and 3. Secondly, the authors thematically analyze and summarize the articles on the impact of digital experience on children’s brain development with a focus on functional and structural changes. The second and the third authors initially extracted the information listed above and passed it on to the two other authors. The four authors had regular online discussion meetings to share their opinions on the data extraction, resolve conflicts, and agree on the results.

Figure 2. Country and sample size.

Table 2. Demographic information of the 33 studies reviewed.

Author/year (Country)	Tech	Sample size	Age (Months)	Participants’ Details
				% Male	Race	Normal	SES
Matsuda and Hiraki (2006) (Japan)	fNIRS	13	84–168	77%	100% Japanese	100% right-handed	N/A
Baumgartner et al. (2006) (German)	EEG	12	100.8–128.4	42%	100% German	100% right-handed; No neurological or psychiatric disease, prior head trauma, sensory impairment, or subjective cognitive impairment	N/A
Shimada and Hiraki (2006) (Japan)	NIRS	13	6–7	38%	100% Japanese	69% right-handed; 23% both hands nearly equally; 8% left-handed	N/A
Han et al. (2007) (China)	fMRI	10	120	50%	100% Japanese	100% right-handed; No neurological or psychiatric history participated, color blind; Normal or corrected-to-normal vision	N/A
Murray et al. (2006) (USA)	fMRI	8	108–156	63%	N/A	No known behavioral disturbances	N/A
Cantlon et al. (2013) (USA)	fMRI	27	51.6–129.6	41%	N/A	N/A	N/A
Takeuchi et al. (2013) (Japan)	MRI	276 (CS) 216 (Long)	67.2–220.8	48% (CS) 51%(Long)	100% Japanese	100% right-handed; No history of malignant tumors or head traumas involving loss of consciousness	N/A
Mondéjar et al. (2016) (Spain)	EEG	12	96–144	50%	N/A	No cognitive issue	N/A
Takeuchi et al. (2016) (Japan)	MRI	240 (CS) 189 (Long)	68.4–220.8(CS) 100.8–255.6 (Long)	48% (CS) 50% (Long)	100% Japanese	N/A	N/A
Pujol et al. (2016) (Spain)	MRI	260	84–123.6	52%	100% Spanish	N/A	N/A
Bergen et al. (2017) (USA)	EEG	22 (S1) 22(S2)	84–144	50% (S1) 68% (S2)	N/A	N/A	High- and middle SES with college educated parents (S1); from families with more diverse backgrounds in communities surrounding the university (S2).
Li et al. (2017) (China)	fNIRS	72	48–72	47%	100% Chinese	N/A	From middle SES families with most parents working at a university and annual income equivalent to US $20,000 to $40,000.
Horowitz-Kraus and Hutton (2017) (USA)	MRI	19	96–144	58%	100% Caucasian	Within the normal range of nonverbal intelligence with no history of neurological, emotional or attention disorders	From average SES households with a median income of $42 000
Baker et al. (2018) (USA)	fMRI	10	88.8–102	N/A	The majority of students were Caucasian	10% received special education services	30% qualified for free/reduced lunch at school
Hutton et al. (2018) (USA)	fMRI	27	50–66	56%	N/A	No prematurity before 38 weeks, developmental delay, head trauma, bilingual/non-English speaking household, kindergarten attendance, and standard contra-indications to MRI	N/A
Takeuchi et al. (2018) (Japan)	MRI	284 (CS) 223 (Long)	68.4–220.8 (CS) 100.8–255.6 (Long)	49% (CS) 52% (Long)	N/A	Normal intelligence value, including IQ < 80	N/A
Paulus et al. (2019) (USA)	MRI	4277	108–120	53%	60% White, 9% Black, 19% Hispanic, 2% Asian, 10% Other	N/A	Diversified parental educational levels
Twait et al. (2019) (Israel)	EEG	32	48–72	60%	100% Hebrew	100% right-handed; Normal or corrected-to-normal vision in both eyes, normal hearing, and had a normal neurological and developmental history	Middle-class
Kostyrka-Allchorne et al. (2019) (England)	EEG	36	84.6	69%	N/A	N/A	From two primary schools located in a semi-rural county of England
Li et al. (2019) (China)	fNIRS	16 (child) 19 (adult)	71.14–85.94 (child) 228.59–286.47 (adult)	63% (child) 47% (adult)	N/A	N/A	N/A
Hutton et al. (2020) (USA)	DTI	47	46.8–61.8	43%	N/A	Living in a household of native English speakers, born at least 36 weeks’ gestation without a history of neurodevelopmental disorder conferring risk of language delays, no contraindications to MRI, such as metal implants	Annual household income: 17% 0-25K,6% 25K-50K, 13% 50K-100K, 23% 100K-150K, 19% over 150K Maternal educational level: 7% Below High School, 15% Some college/associate degree, 40% College graduate, 38% Above college
Zivan et al. (2019) (Israel)	EEG	30	57.04–70.48	53%	N/A	Normal developmental history without a history of attention difficulties	Middle-class
Rodriguez-Ayllon et al. (2019) (Netherlands)	DTI	2352	9.54–10.70	50%	67% Dutch, 8% Other Western, 25% Non-Western	N/A	Maternal education: 8% Low, 27% Medium, 65% High
Li et al. (2020) (China)	fNIRS	20	48–72	45%	100% Chinese	N/A	From an urban Children’s Learning Centernormal
Wetzel et al. (2021) (German)	EEG	37	81.6–109.2	54%	100% German	97% right-handed, 1% left-handed; No medication affecting the central nervous system, neurological disorders, hearing disorders, or attention deficit hyperactivity disorder	N/A
Li et al. (2021) (China)	fNIRS	38	48–75.6	55%	100% Chinese	100% right-handed	N/A
Chaarani et al. (2022) (USA)	fMRI	2217	108–120	37%	N/A	SST sample: 90% right-handed, 10% left-handed n-Back sample: 89% right-handed, 11% left-handed	N/A
Hutton et al. (2022) (USA)	MRI	52	37–63	44%	N/A	No prior or current kindergarten attendance; No documented history of head trauma with loss of consciousness or neurodevelopmental condition likely to confer cognitive delay	Annual household income: 13% 0-25K,17% 25K-50K, 29% 50K-100K, 21% 100K-150K, 9% Above 150K Maternal educational level: 8% High School or Less, 17% Some College, 42% College graduate, 33% More than college
Zhao et al. (2022) (USA)	MRI	9738	108–120	52%	67% White, 14% Black, 17% Other/Mixed, 2% Asian	N/A	Family income: 28% 0-50K, 29% 50K-100K, 43% Above 100K Maternal educational level: 4% Below High Diploma, 8% HSDiploma/GED, 25% Some College, 27% Bachelor, 43% Post Graduate Degree
Lewin et al. (2023) (USA)	EEG	20	45.6–121.2	65%	N/A	A monolingual speaker of English; no history of stuttering, stutterno, and neurological disease or injury, speech, language, or hearing disorder had normal or corrected-to-normal vision and hearing	Average parental education: 4-year college degree Average Household income:70K-80K
Law et al. (2023) (Singapore)	EEG	157	12–108	54%	50% Chinese, 34% Malay,15% Indian	N/A	Maternal education: 32% High school and below, 39% Diplomas/certificates, 29% University and above Monthly household income: 23% Below 2K, 25% 2K-4K, 26% 4K-6K, 27% over 6K
Chen et al. (2023) (USA)	fMRI	3891	108–131	47%	71% White, 11% African American 2% Asian, 4% Others, 11% Multiracial population	N/A	Total conbiner family income: 8% 25K, 11% 25K-50K, 26% 50K-100K, 34% 100K-200K, 13% Above 20K, 7% unknown Parental Education: 4% Below High school diploma, 9% High school diploma, 29% Some college or associate degree, 32% Bachelor’s degree, 26% Post graduate degree
Zhao et al. (2023) (USA)	MRI	5166	108–120	54%	70% White, 12% Black, 2% Asian, 16% Other/Mixed	N/A	Parental highest education level: 11% Up to HS graduation/GED, 26% Some College, 27% Bachelor’s Degree, 35% Post-Graduate Degree Household income: 25% 0K-50K, 28% 50K-100K, 40% Above 100K

Note. CS = Cross-sectional; DTI = Diffusion Tensor Imaging; EEG = Electroencephalogram; fMRI = Functional Magnetic Resonance Imaging; fNIRS = Functional Near-Infrared Spectroscopy; Long = Longitudinal; MRI = Magnetic Resonance Imaging; NIRS = Near-Infrared Spectroscopy; = Study 1; S2 = Study 2.

Table 3. Technological information of the 33 studies reviewed.

	Research Design			Research Results
Author/year (Country)	Collection Method of Data	Type of Digital Experience	Assessment(measure/specific task)	Type of Impact	Outcomes	Limitations
Matsuda & Hiraki, 2006 (Japan)	Experiment	Games	two commercially available video games, namely Super Smash Bros. Melee (Nintendo Co. Ltd., Kyoto, Japan) and Tetris (Success Co. Ltd., Tokyo, Japan)	functional	Negative impact: Video game playing commonly leads to a reduction in DPFC activity compared to the resting state, regardless of age, encompassing both adults and children beyond infancy stages.	Face data exclusions due to motion artifacts; Lack of comprehension of experimental instructions
Baumgartner et al. (2006) (German)	Experiment	Screen-Based Media Use	five different roller coaster simulations	functional	Mixed Impact: Strong spatial presence experiences in an arousing and noninteractive VR world are associated with markedly increased activity in parietal/occipital areas of the brain together with decreased activity in frontal structures involved in the executive system of the brain; Autonomic somatic reactions are increased along with activation in brain structures known to be involved in the somatic and visceral representation of the body state and emotion processing.	Unanswered neurophysiological questions on spatial presence
Shimada and Hiraki (2006) (Japan)	Experiment	Screen-Based Media Use	a female demonstrator manipulating an attractive noncommercial toy (when the demonstrator tapped lateral parts of a box, a gadget installed inside the box sprung out) presented either through a TV monitor or were live	functional	Negative Impact: The sensorimotor activity was more pronounced in response to an action of others than to object movement in the live setting, while it was not observed in the TV setting.	NIRS data acquisition limitation; uncertainty statistical analyses; did not use better NIRS wavelengths; exist contamination of the motor activity
Han et al. (2007) (China)	Experiment	Virtual Visual Scenes	movie clips depicted human beings in real visual scenes; cartoon clips depicted artificial characters in virtual visual scenes	functional	Positive Impact: Cartoon clips elicited activation in the bilateral MT and posterior STS, bilateral superior parietal lobule, and MPFC.	Lack of functional role of MPFC activation in children watching cartoon clips
Murray et al. (2006) (USA)	Experiment	Screen-Based Media Use	the televised violent videotape segments	functional		a small sample focusing only on one type of violence; lack of a matched nonviolent comparison task; did not consider additional factors influencing the brain response to TV violence
Cantlon et al. (2013) (USA)	Experiment	Screen-Based Media Use	a single 20.3-min montage of clips from children’s educational television shows Sesame Street video	functional	Mixed Impact: The right IPS appears to be more mature than the left IPS in children. In early childhood the IPS responds in a content-specific manner to numerical information presented naturalistically and that both the amplitude and temporal pattern of the neural response are related to children’s school-based math performance.
Takeuchi et al. (2013) (Japan)	Survey	Video Viewing and Editing	a self-report questionnaire of the duration of TV or video viewing	structural	Mixed Impact: Cross-sectional and longitudinal analyses indicate increased rGMV in the frontopolar and medial prefrontal areas; Cross-sectional analyses show increased rGMV and rWMV in areas of the visual cortex. Longitudinal analyses suggest increased rGMV in the hypothalamus/septum and sensorimotor areas. Both cross-sectional and longitudinal analyses indicate a decrease in VIQ.	inconclusive functional relevance of certain areas
Mondéjar et al. (2016) (Spain)	Experiment	Games	an adaptation of the Trail Making Test (TMT); the interference test. An adaptation of the well-known test of Stroop(1935); the Washers test, an adaptation of the Tower of Hanoi (TOH).	functional	Positive Impact: High activation in the frontal area in the videogame phase of evaluation. Greater significance is given to the activation of theta waves when the player is demanded to give a new answer in the frontal activation and in a generalized way in most users. The activation in terms of beta waves when remembering a necessary answer to continue in the game in the frontal area and in some point on an occipital level.	lack of the long-term effects of videogames on executive function (EF) evaluation; did not focus on utilizing supervised learning techniques and expanding the participant population
Takeuchi et al. (2016) (Japan)	Survey	Games	a self-report questionnaire of the duration of VGP during weekdays	structural	Negative Impact: Longer VGP is associated with greater MD in extensive regions and lower verbal intelligence, both cross-sectionally and longitudinally. A higher PIQ was associated with lower MD in extensive regions in the brain, and higher FSIQ and VIQ were both associated with lower MD in the left thalamus, left hippocampus, left putamen, left insula, left Heschl gyrus and associated white matter bundles.	not an intervention study; did not gather data on socioemotional measures and specific effects of certain types of videogames
Pujol et al. (2016) (Spain)	Both	Games	a parent-reported questionnaire of indicating which games were most frequently played in the previous year; Difficulties Questionnaire (SDQ) of child behavioral problems; A neighbor- hood socioeconomic status vulnerability index; child version of the Attentional Network Test(ANT)	both	Mixed Impact: In the gaming group, the weekly time spent gaming was steadily associated with higher scores. Video gaming was associated with higher functional connectivity in the putamen and caudate nucleus maps. In the cognitive domain, video game use was associated with faster motor response to visual stimulation. Video gamers did not demonstrate more problematic behavior than nongamers. At a neural level, structural and functional brain changes associated with gaming use were most evident with respect to basal ganglia circuitry.	limited coverage of all domains; did not assess the formal reliability and accuracy of parent-reported digital game time; lack of the long-term effect and analyses stratified by game content or type
Bergen et al. (2017) (USA)	Experiment	Games	a simplified computer program that required the performance of two different types of responses for the “space visitors” (correct color responses or imaginative choice responses)	functional	Positive Impact: During the choice condition compared to the color condition, there were larger N2 and P3 amplitudes in the parietal lobe (Pz), indicating increased neural activity related to decision-making. The age of the child played a significant role, with younger children showing more frontal lobe activity during mixed choice conditions, suggesting higher cognitive load and difficulty with set shifting. Older children exhibited a more positive inflection in the parietal lobe, indicating greater allocation of mental resources to decision-making. There were developmental differences in how children responded to the two educational game conditions, highlighting variations in cognitive processes and strategies employed during gameplay.	lack of the effects of video game play on brain wave activation
Li et al. (2017) (China)	Experiment	Games & Video Viewing and Editing	study materials included a fantastical game Dr. Panda in Space, which was played on an iPad; Go-No-Go task	functional	Positive Impact: Viewing fantastical events consumed more cognitive resources and resulted in disruption of inhibitory control. In contrast, there was no such disruptive effect after participants interacted and touched fantastical events via a touch screen device. Viewing real events did not have a negative effect on their inhibitory control compared to interacting with them. The posttest increase in children’s Coxy-Hb for viewing but not interacting suggests that there may be a compensation mechanism when interacting with fantastical events. Both the form and content of mobile devices can produce effects on children’s cognitive skills.	did not yoke participants in video and game conditions to control for differences; lack of the role of presence in touch screen and smaller mobile devices; lack of the long-term effect
Horowitz-Kraus and Hutton (2017) (USA)	Both	Screen-Based Media Use	a parent-reported screen time activities questionnaire; the Test of Nonverbal Intelligence; a vocabulary task from the Peabody Picture Vocabulary Test(PPVT); the Woodcock-Johnson Tests of Cognitive Abilities	functional	Negative Impact: Time spent reading was positively correlated with higher functional connectivity between the seed area and left-sided language, visual and cognitive control regions. Reported screen-time hours were negatively correlated with functional connectivity between the visual word form area and the regions related to cognitive control such as the right Brodmann areas. Higher reported screen time was correlated with decreased functional connectivity between the visual word form area and the regions related to language and cognitive control.	did not explore the effects of screen time and reading environments, and interrelationships with the neurodevelopment
Baker et al. (2018) (USA)	Experiment	Screen-Based Media Use	three digital math apps on iPads: Montessori Numbers: Base-10 Blocks (100–999), Motion Math: Zoom (levels 2+), and LetsTans Kids.	functional	Negative Impact: Increases in cortical activation would be identified in regions of the brain corresponding to the left and right intraparietal sulci. This pattern of activity was observed for all three digital math apps, but it was not present during rest. The left and right regions of the parietal lobe, which were consistently active during math play, were not active during rest. Regions of the PFC would be active during each digital math app while maintaining minimal activity during rest. fNIRS is capable of identifying reductions in PFC activity when children stop playing with the apps.	a small sample size; lack of generalizability
Hutton et al. (2018) (USA)	Experiment	Screen-Based Media Use	three different children’s storybooks were presented without interruption in three formats	functional	Negative Impact: For illustration relative to audio, FC was decreased within the language network and increased between visual, DMN, and cerebellar networks, suggesting decreased strain on the language network afforded by pictures and visual imagery. Between-network connectivity was decreased for all networks for animation relative to the other formats, particularly illustration, suggesting a bias toward visual perception at the expense of network integration.	may not apply to other styles stories; exist MRI technological constraints; limited to five functional networks
Takeuchi et al. (2018) (Japan)	Survey	Internet and Pad Use	WAIS-III; WISC-III; a self-report questionnaire of the frequency of internet use	structural	Negative Impact: Cross-sectional rGMV/rWMV analysis: The frequency of internet use did not significantly correlate with rGMV or rWMV of any of the brain areas. Longitudinal rGMV/rWMV analysis: The frequency of internet use in the pre-experiment showed a significant negative correlation with the change in rGMV/rWMV of a widespread anatomical cluster. A higher frequency of internet use was associated with decreased verbal intelligence and smaller volume increases in widespread brain areas after a few years.	not an intervention study
Paulus et al. (2019) (USA)	Survey	Screen-Based Media Use	a youth-report screen media activity questionnaire; a parent-reported CBCL; a youth performance NIH Toolbox	structural	Mixed Impact: Those individuals with higher SMA-related GFA 1 and GFA 4 scores had significantly higher externalization scores; With higher SMA-related GFA 4 scored lower on fluid abilities; With higher SMA-related GFA 1 and higher SMA-related GFA 4 had lower crystallized abilities. Those individuals with higher SMA-related GFA 2 had higher fluid abilities and showed better performance on crystallized abilities.	did not allow drawing causal inferences; did not use a higher resolution approach; lack of better control of these variables; did not consider engagement in other recreational activities or school-related screen media activity
Twait et al. (2019) (Israel)	Experiment	Screen-Based Media Use	attention Network Task	functional	Positive Impact: Children in the dialogic reading also demonstrated lower differences between P300 and N200 amplitudes for congruent compared to incongruent stimuli in the attention network task paradigm, indicating improved orienting attention and executive-control abilities in this group. These abilities were measured in the electroencephalogram testing. The dialogic reading group also showed improved executive control via N200, although this did not reach significance, compared to the screen-based group. Higher accuracy rates for both incongruent and congruent conditions were observed for the dialogic reading compared to the screen group.	did not collect neurobiological measures before the intervention; lack of a control group
Kostyrka-Allchorne et al. (2019) (England)	Experiment	Video Viewing and Editing	Go-no-Go	functional	Negative Impact: A comparison of N2 waveforms shown in Figure 2 suggests that the amplitude difference is confined to the fast-paced group. The significant interaction shows that the correct-wrong difference in P3 latency was reliably more positive in the fast-paced group than in the slow-paced group. Watching fast-paced videos has short-term consequences for the neural responses, resulted in weaker inhibition.	lack of a control group; experimental videos may have reduced the ecological validity
Li et al. (2019) (China)	Experiment	Virtual Visual Scenes	a reality judgment task	functional	Mixed Impact: The fNIRS data also revealed higher prefrontal activations in children than in adults while watching and judging the reality of events, especially in the mPFC and rPFC.When judging the real events or the events performed by real characters, the children showed higher prefrontal activations. Positive correlations were revealed between children’s mPFC activations and their judgment accuracy on RR events.	did not investigate additional social-brain areas; did not explore the different neural bases of children and adults; a small age-range sample
Hutton et al. (2020) (USA)	Survey	Screen-Based Media Use	a parent-reported ScreenQ35 survey; EVT-2; CTOPP-2; GRTR	structural	Negative Impact: Increased use of screen-based media in the context of the AAP guidelines was associated with lower microstructural integrity of brain white matter tracts that support language, executive functions, and emergent literacy skills, controlling for child age and household income. Screen use was also associated with lower scores on corresponding behavioral measures, controlling for age.	did not use objective measures
Zivan et al. (2019) (Israel)	Both	Screen-Based Media Use	a parent-reported Conners and BRIEF; the Naming subtest and the Matrix task from the WPPSI; Sky Search and Score subtests, from the TEA-Ch battery	functional	Negative Impact: The EEG functional connectivity of the theta band compared to the beta band, a pattern that was previously found to be related to attention deficits. In contrast to passive attendance to the screen, children listening to stories told by the interactive experimenter engaged the children using eye contact and gestures prompting them to stay alert.	lack of EEG data before the intervention
Rodriguez-Ayllon et al. (2019) (Netherlands)	Survey	Screen-Based Media Use	a parent-reported Screen time questionnaire	structural	No Impact: No association was found between individual or aggregate screen time variables and global DTI metrics.	did not allow drawing causal inferences; did not consider other environmental confounders; did not use objective measures
Li et al. (2020) (China)	Experiment	Virtual Visual Scenes	animated shows with frequent fantastical events	functional	Negative Impact: It indicated that viewing frequent fantastical events led to overloaded processing, which in turn impaired Chinese preschoolers’ EF task performance.	lack of a control group; differences between the two stimulus videos
Wetzel et al. (2021) (German)	Experiment	Internet and Pad Use	play the game Memory either with a human opponent in reality or with a virtual opponent on a tablet PC and to ignore a task-irrelevant sound sequence	functional	Positive Impact: Results demonstrate increased activity on the level of the auditory cortex in response to task-irrelevant sounds when children play with a tablet PC compared to playing the same game with a human partner. It indicates enhanced processing of all sound types independent from their novelty.	did not disentangle potential influencing factors; cannot rule out that the absence vs. presence of the experimenter between conditions influenced the results
Li et al. (2021) (China)	Both	Internet and Pad Use	DCCS Tasks; HLEP developed from HLEI	functional	Negative Impact: Researchers tend to suggest that this “heavy-user” activation pattern might be abnormal and very likely unhealthy, demonstrating the negative impact of tablet use on young children’s executive function.	fNIRS data acquisition limitation; a small sample size; did not consider gender as a variable; lack of longitudinal data
Chaarani et al. (2022) (USA)	Both	Games	a child self-report Screen time questionnaire; stop signal task (SST); n-back task	functional	Positive Impact: The VGs performed better on both fMRI tasks compared with the NVGs. Although the CBCL scores were elevated in children who play video games for 3 or more hours a day, the results raise the intriguing possibility that video gaming may provide a cognitive training experience with measurable neurocognitive effects.	did not consider game genre as a moderating variable; did not allow drawing causal inferences
Hutton et al. (2022) (USA)	Survey	Screen-Based Media Use	EVT-2; CTOPP-2, rapid object naming subscale; PIPA; GRTR	structural	Negative Impact: The overall CT is expected to have largely peaked, though not yet in higher-order areas. Despite limited statistical power, particularly when controlling for maternal education, significant (ROI-based) and/or marginally significant (whole-brain) associations were identified between higher screen-based media use and lesser CT involving both primary and higher-order areas.	a sample with largely of higher income and maternal education; did not discern mechanisms; lack of a range of cognitive-behavioral measures at the formative age
Zhao et al. (2022) (USA)	Survey	Screen-Based Media Use	a youth self-report SMA questionnaire	structural	No Impact: The JIVE analyses identified one joint component that was characterized by a covariation pattern common across surface areas and GM volumes in key brain regions comprising a previously described thalamus-PFC-brainstem circuit. The identified joint component was associated with total screen time and externalizing behaviors but not with total sleep problems, internalizing behaviors, and crystalized intelligence.	lack of longitudinal data; lack of direct measurement of types, patterns of SMA, sleep activity and performance on cognitive tasks
Lewin et al. (2023) (USA)	Both	Screen-Based Media Use	all Daily Screentime questionnaires; Go/No-Go task	functional	Negative Impact: Children with greater screen time exhibit less robust neural processes for inhibitory control.	did not allow drawing causal inferences; lack of longitudinal data; did not use objective measures of children’s screen use
Law et al. (2023) (Singapore)	Both	Screen-Based Media Use	a parent-reported Screen time questionnaire; NEPSY-II: naming inhibition, shifting, and working memory; a teacher-reported CBCL & BRIEF-2	functional	Negative Impact: Increased screen time in infancy is associated with impairments in cognitive processes critical for health, academic achievement, and future work success.	did not use objective measures of children’s screen use; did not include contextual influences
Chen et al. (2023) (USA)	Survey	Screen-Based Media Use	a child self-report Screen time questionnaire; a parent-reported Sports and Activities; Involvement Questionnaire; SAP scores; BAS scores provided by the ABCD dataset	functional	Negative Impact: The longer screen activity is associated with a delayed development of the inhibitory control system in the brain and longer screen exposure may amplify the negative effect of reward sensitivity on ICN development. It showed a negative change in the frontostriatal connectivity in individuals with relatively increased daily screen time. The delaying effect of screen exposure on the development of ICN is a general phenomenon that exists in children during preadolescence.	lack of access to the processed inhibitory task-based cortical network to subcortical regions data
Zhao et al. (2023) (USA)	Survey	Screen-Based Media Use	a parent-reported CBCL; a self-reported SMA questionnaire	structural	Negative Impact: Compared to individuals with low-frequency and moderate-frequency SMA, those with high-frequency SMA demonstrated a slower expansion in subcortical regions. This global structural co-development pattern provided a proxy measure of imbalanced brain structural development (mainly in GMV and cortical thickness) among cortical and subcortical regions.	did not consider different screen modalities beyond total screen use and additional potential influences; lack of Age-dependent effects; lack of longitudinal data; did not understand the brain correlates of gender difference

Collating, Summarizing, and Reporting the Results

An analytic framework or thematic construction is essential for mapping the strategies and themes found in studies identified for a scoping review (Arksey & O’Malley, 2005). Based on the thematical analysis described above, four themes of concern were identified in the 33 articles in this scoping study: (1) demographic information of the studies (countries, participants, research designs, neuroimaging technologies); (2) digital experience impacts the function of young children’s brains; (3) digital experience impacts the structure of young children’s brains. Most articles reported only one of the two types of impacts (functional or structural), and (4) whether the impact is longitudinal. Any conflicts over theme classification were resolved through a group discussion between the authors of this review until an agreement was reached.

Results

Demographic Information of the Studies

Altogether, 33 articles met the inclusion criteria of this study (see Tables 2 and 3). All the studies were published in the past 18 years (and the first was published in 2005), with four-fifths (84%) published in the past decade (2010-present). This subsection presents a descriptive overview of the demographic information of these studies.

Countries

Most studies (14 out of 33; 42.42%) were conducted in the United States. Eight were in Europe (24.24%): two from Germany, two from Spain, two from Israel, two from the Netherlands, and one from England. Altogether, 11 studies were from Asia-Pacific areas (33.33%): five from China, five from Japan, and one from Singapore (Figure 2).

Participants

A total of 30,100 participants were included in these studies, ranging from infants aged six months (Study #3) to adults aged 23.87 years old (Study #20 includes both children and adults). Most of them were cross-sectional studies (29/33; 87.88%) and focused on children aged from 6 to 286.47 months. Four studies (Studies #7, 9, 16, 20) collected longitudinal imaging data from children (aged 64.8 to 220.8 months) and adults. About 87.88% of the studies had a sample size between 1–3000, 9.09% had around 3001–6000 participants, and 3.03% had a large sample between 6000–10000 (see Figure 2).

Research Designs

Figure 3 shows that 16 studies adopted the experimental design, 10 were survey studies, and the other seven included both experiment and survey studies.

Figure 3. Design and technologies.

Neuroimaging Technologies

Figure 3 lists the neuroimaging technologies used by the 33 studies. In particular, nine studies adopted Magnetic Resonance Imaging (MRI), seven employed Functional Magnetic Resonance Imaging (fMRI), nine used Electroencephalogram (EEG), five took Functional Near-Infrared Spectroscopy (fNIRS), one reported Near-Infrared Spectroscopy (NIRS), and two deployed Diffusion Tensor Imaging (DTI). Technically, NIRS and fNIRS share the same neuroimaging technology, the term “NIRS” was used in the early times, and then it was replaced by fNIRS to emphasize functional scans. In this study, we followed the authors’ choice of terms to maintain the original meaning of their studies’ nature.

Digital Media

The primary digital media were screen-based media use (51.51%), followed by Games (21.21%), Virtual Visual Scenes (12.12%), Video Viewing and Editing (9.09%), Internet and Pad Use (6.06%).

Digital Experience Impacts the Function of Young Children’s Brains

About 23 studies reported the impact of digital experiences on the function of children’s brains. Among them, six studies reported positive effects, 15 studies demonstrated negative ones, and two reported mixed results. Figure 4 presents the neuroimaging findings of the impact of digital experience on brain functions.

Figure 4. Neuroimaging findings of the impact of digital experience on brain functions.

Note. Red represents the positive impact, blue reflects the negative impact, and purple refers to the mixed impact. The authors created with MedPeer (www.medpeer.cn).

Positive Impact

The EEG/ERP Evidence

Two EEG/ERP studies also reported the training and priming effects of digital experience on children’s brain function. In particular, Study #8 (Mondéjar et al., 2016) found that the frontal lobe (EF) was very active in the game phase, indicating that the brain relies more on the skills located in the frontal lobe. They also noticed that theta waves, related to new learning, increased in the frontal lobe and most users when they had to use a different game mechanic. Beta waves, which are related to memory recall, also increased in the frontal lobe and some parts of the back of the brain when they had to remember an answer to continue in the game. Study # 11 (Bergen et al., 2017) explored the impact of video game play on brain development by comparing how younger and older children responded to two educational game conditions: mixed and single. They found that younger children had more brain activity in the frontal lobe, which is related to attention, when they played the mixed game. This means that they had to pay more attention to the mixed game. This shows a developmental difference in how children react to different educational game conditions.

The fNIRS Evidence

Study #12 (Li et al., 2017) found that viewing fantastical events elicited higher activation in the dorsolateral prefrontal cortex, which indicated increased cognitive demand and impaired inhibitory control. However, this effect was attenuated when children touched fantastical events on a touch-screen device, suggesting that tactile feedback reduced the cognitive load. This study showed that the content and modality of mobile media could have differential impacts on children’s executive functions.

The fMRI Evidence

Three fMRI studies reported digital use’s positive training or priming effects on children’s brain function. In particular, Study #4 (Han et al., 2007) reported that cartoon clips elicited activation in the bilateral middle temporal cortex (MT) and posterior superior temporal sulcus (STS), bilateral superior parietal lobule, and medial prefrontal cortex (MPFC). Study #5 (Murray et al., 2006) revealed that TV violence viewing activated a network of regions implicated in attention, arousal, and salience; engaged a phylogenetically old brain system involved in the detection of fear or threat in the environment; was associated with activation of limbic and neocortical systems likely involved in the episodic encoding and retrieval of the environmental context associated with such threat; and was associated with activation of premotor regions possibly involved in the programming of motor plans (fight or flight). Recently, Chaarani et al. (2022) (#27) evaluated the effects of video gaming on children’s fMRI performance and behavior. They found that the video game group performed better on two fMRI tasks than the non-video game group, suggesting the cognitive benefits of video gaming. However, this study has not explored the connection between these behavioral problems and brain development, warranting further neuroimaging studies.

Negative Impact

The EEG/ERP Evidence

Six EEG/ERP studies reported the negative effects of digital experiences on children’s brain function. In particular, Twait et al. (2019) (Study #18) compared the effects of screen media and dialogic reading on children’s orienting attention and executive-control abilities. They found that the dialogic reading group showed enhanced EEG measures of these abilities, while the screen group did not. Meanwhile, Kostyrka-Allchorne et al. (2019) (Study #19) examined the impact of the editing pace of videos on children’s neural responses during the Sustained Attention to Response Task (SART). They found that watching fast-paced videos reduced the EEG amplitude of the no-go trials, indicating weaker inhibition. Zivan et al. (2019) (Study #22) found that the screen group had higher theta-band connectivity than beta-band connectivity, which was associated with attention deficits. In contrast, the story group without screen experience did not show this difference, indicating that interactive storytelling enhanced attentional engagement. Wetzel et al. (2021) (Study #25) analyzed the event-related potentials (ERPs) of children who played a card game with a virtual or a human opponent on a tablet PC. They found that playing with a virtual opponent increased the amplitude of the P3a component, which reflected increased processing of task-irrelevant auditory stimuli. Playing with a human opponent did not show this effect, suggesting that social interaction reduced distraction. Lewin et al. (2023) (#30) examined the relationship between child screen time and neural processes for inhibitory control. They found that after controlling for age, increased screen time was associated with reduced P2 and P3 amplitudes elicited by No-Go trials, indicating less robust inhibition. Law et al. (2023) (#31) investigated the association between infant screen use and cortical EEG activity before Age 2. They found that increased screen time in infancy was associated with impairments in cognitive processes critical for health, academic achievement, and future work success.

The fNIRS evidence

About five NIRS/fNIRS studies revealed the negative impact of digital experience on children’s brain function. In particular, Study #1 (Matsuda & Hiraki, 2006) reported a significant reduction of oxygenated hemoglobin (HbO) concentration changes in the dorsal prefrontal cortex (DPFC) during the onset of video game playing and a rapid restoration upon termination. Video game playing tended to suppress DPFC activity relative to the resting state, and adults and children older than infants shared this effect. Study #3 (Shimada & Hiraki, 2006) found that the infant’s motor areas, activated during their actions, were also engaged when observing the actions of others. These findings suggested different processing of body movements in these two contexts (live or TV), demonstrating that sensorimotor activity was more pronounced in response to the actions of others in the live setting but not in the TV setting. Li et al. (2019) (Study #20) measured the fNIRS activation of children and adults who watched and judged the reality of events on a screen. Children showed decreased prefrontal activation over time, especially in the dorsolateral, lateral, and right prefrontal cortex, while adults showed a stable activation pattern. Li et al. (2020) (Study #24) assessed the fNIRS activation of preschoolers who viewed frequent or infrequent fantastical events on a screen and performed an executive function (EF) task. They found that the frequent fantasy group had an increase in prefrontal activation over time, which reflected cognitive overload and impaired EF performance. Li et al. (2021) (Study #26) compared the fNIRS activation and EF performance of heavy and non-users of tablet devices. They found that the non-users performed better on the Dimensional Change Card Sort (DCCS) task and had higher activation in the prefrontal cortex (BA 9), which indicated normal and healthy brain functioning. The heavy-users performed worse on the DCCS task and had lower activation in BA 9, which indicated abnormal and unhealthy brain functioning.

The fMRI Evidence

Four MRI/fMRI studies revealed the negative impact of digital experience on children’s brain function, including functional connectivity. In particular, Study #13 (Horowitz-Kraus & Hutton, 2017) found that screen time was related to lower connectivity between the seed area and regions related to language and cognitive control. Higher reported screen time was correlated with decreased functional connectivity between the visual word form area and the regions related to language and cognitive control. Study # 14 (Baker et al., 2018) examined the cortical activity patterns of preschool-age children who engaged with digital math apps on a touch-screen device using fNIRS. The results showed that animation reduced the connectivity among different brain networks, especially compared to illustration. In contrast, illustration may be more effective in facilitating language development and learning at this age. Study #15 (Hutton et al., 2018) found that between-network connectivity was decreased for all networks for animation rather than illustration. This finding suggests substantial differences in functional brain network connectivity for animated and more traditional story formats in preschool-age children, reinforcing the appeal of illustrated storybooks at this age to provide efficient scaffolding for language. Chen et al. (2023) (#32) found that longer screen exposure time was associated with the underdevelopment of the inhibitory control system. They suggested that long-term prolonged daily screen exposure may negatively affect children’s cognitive development.

Mixed Impact

Using EEG, Study #2 (Baumgartner et al., 2006) found that high spatial presence experiences in an arousing and noninteractive VR world were linked to both enhanced activity in parietal/occipital regions of the brain and reduced activity in frontal regions involved in the executive system of the brain. Moreover, they observed increased autonomic somatic responses and activation in brain regions implicated in the somatic and visceral representation of the body state and emotion processing. In addition, Study # 6 (Cantlon et al., 2013) investigated the neural processes evoked by naturalistic educational television viewing using fMRI. They found that the right IPS was more mature than the left IPS in children, based on both natural viewing and traditional paradigm data. They concluded that the IPS (especially in the right hemisphere) responded in a content-specific manner to numerical information presented naturalistically and that the neural response’s amplitude and temporal pattern were related to children’s school-based math performance.

Digital Experience Impacts the Structure of Young Children’s Brains

About nine studies reported the impact of digital experiences on the structure of young children’s brains. Two studies reported no impact, five demonstrated negative ones, and two reported mixed results. No studies reported any positive impact. Figure 5 presents the neuroimaging findings of the impact of digital experience on brain structures.

Figure 5. Neuroimaging findings of the impact of digital experience on brain structures.

Note. Blue reflects the negative impact, purple refers to the mixed impact, and green marks the null results. The authors created with MedPeer (www.medpeer.cn).

No Impact

Contrary to the majority of the studies, two studies did not find any evidence of digital experience affecting brain structure. Specifically, Study #23 (Rodriguez-Ayllon et al., 2019) reported no significant correlations between any of the screen time variables and global Diffusion Tensor Imaging (DTI) metrics, indicating that screen time did not have a noticeable impact on the brain’s structural integrity as measured by DTI. Similarly, Study #29 (Zhao et al., 2022) demonstrated a consistent pattern of structural covariation across different developmental stages related to high-quantity screen media activity and externalizing behaviors. Moreover, this joint component was linked to total screen time and externalizing behaviors but not to total sleep problems, internalizing behaviors, or crystallized intelligence.

Negative Impact

The MRI Evidence

Four studies provided MRI evidence to demonstrate the negative impact of digital experience on the structure of children’s brains. In particular, Study #9 (Takeuchi et al., 2016) found that more extended Video Game Play (VGP) was associated with greater diffusion tensor imaging mean diffusivity (MD) in extensive regions and lower verbal intelligence, both cross-sectionally and longitudinally. A higher Performance IQ was associated with lower MD in extensive regions in the brain, and higher full-scale IQ (FSIQ) and Verbal IQ (VIQ) were both associated with lower MD in the left thalamus, left hippocampus, left putamen, left insula, left Heschl gyrus and associated white matter bundles. Study #16 (Takeuchi et al., 2018) reported a longitudinal finding that a higher frequency of internet use in the pre-experiment showed a significant negative correlation with changes in rGMV and rWMV in widespread anatomical clusters. These clusters included extensive bilateral perisylvian areas, the bilateral temporal pole, the bilateral cerebellum, bilateral medial temporal lobe structures (hippocampus and amygdala), bilateral basal ganglia structures, the bilateral inferior temporal lobe, the thalamus, the bilateral orbitofrontal gyrus and lateral prefrontal cortex, the insula, and the left lingual gyrus. The negative correlation extended to the white matter area adjacent to the gray matter cluster and cingulate areas. Study #28 (Hutton et al., 2022) suggested that higher media use is associated with differences in Cortical Thickness (CT) in both primary visual and higher-order association areas. The findings indicate that media use may have a measurable impact on brain structure, particularly in regions involved in visual processing and higher-order cognitive functions. Study #33 (Zhao et al., 2023) revealed an imbalanced impact in that individuals with high-frequency screen-media activity (SMA) demonstrated a slower expansion in subcortical regions, such as the brainstem and left putamen, compared to individuals with low-frequency and moderate-frequency SMA. This global structural co-development pattern suggested an imbalanced brain structural development, particularly in GMV and cortical thickness, among cortical and subcortical regions.

The DTI Evidence

Study #21 (Hutton et al., 2020) found that increased screen-based media use was associated with lower microstructural integrity of brain white matter tracts that support language, executive functions, and emergent literacy skills after controlling for child age and household income. Screen use was also associated with lower scores on corresponding behavioral measures after controlling for age.

Mixed Impact

The impact of digital experience on children’s brain structure was mixed in two studies. Specifically, Study #7 (Takeuchi et al., 2013) found beneficial and detrimental effects. On the positive side, TV viewing was associated with (a) increased regional gray matter volume (rGMV) of the frontopolar and medial prefrontal areas in both cross-sectional and longitudinal analyses; (b) enhanced rGMV/rWMV (regional white matter volume) of areas in the visual cortex in cross-sectional analyses; (c) enlarged rGMV of the hypothalamus/septum and sensorimotor areas in longitudinal analyses. On the negative side, TV viewing was correlated with reduced verbal intelligence quotient (IQ) in both cross-sectional and longitudinal analyses. Furthermore, Study #17 (Paulus et al., 2019) indicated that screen media activity (SMA) accounted for 37% of the variance in structural brain indices such as cortical thickness, sulcal depth, and gray matter volume. The finding supported the idea of SMA-related maturational coupling or structural correlation networks in the brain and provided evidence that individual differences in these networks had mixed implications for psychopathology and cognitive performance.

Prompt Effects vs. Long-Term Influences

This scoping review found that most of the reviewed studies reported digital experience’ prompt effects or short-term influences on brain functions (see Figure 6); only four studies (Studies #7, 9, 16, 20) provided longitudinal evidence to demonstrate its long-term impact on brain structures (see Figure 7). In addition, study #10 (Pujol et al., 2016) conducted a one-year longitudinal study to evaluate the impact of video gaming on brain structure and function. At the behavioral level, they found that video gamers, overall, did not demonstrate more problematic behavior than non-gamers. In particular, the survey study indicated that weekly gaming time was steadily associated with conduct problems, peer conflicts, and reduced prosocial abilities. At the cognitive level, video game use was associated with faster motor response to visual stimulation. At the neural level, they found structural and functional brain changes associated with gaming use evident concerning basal ganglia circuitry. Positively, video gaming was associated with higher functional connectivity in the putamen and caudate nucleus maps. These findings jointly indicate that the digital experience’s impact could be short-term and long-term.

Figure 6. Neuroimaging findings of the short-term impact of digital experience.

Note. Red represents the positive impact, blue reflects the negative impact, and purple refers to the mixed impact. The authors created with MedPeer (www.medpeer.cn).

Figure 7. Neuroimaging findings of the longitudinal impact of digital experience.

Note. Red represents the positive impact, blue reflects the negative impact, purple refers to the mixed impact, and green marks the null results. The authors created with MedPeer (www.medpeer.cn).

Discussion

This scoping review has provided a preliminary summary of the impact of digital experience on children’s developing brains, drawn from 33 journal articles identified as meeting the inclusion criteria outlined in the method section. This section will discuss these findings and elaborate on their implications for future studies, practical improvement, and policymaking.

Whether Digital Experience Can Shape Young Minds

First, this scoping review found that 23 studies reported the impact on children’s brain functions: six positive effects, 15 negative impacts, and two mixed results. In particular, six EEG/ERP (Kostyrka-Allchorne et al., 2019; Law et al., 2023; Lewin et al., 2023; Twait et al., 2019; Wetzel et al., 2021; Zivan et al., 2019), five fNIRS (Li et al., 2019, 2020, 2021; Matsuda & Hiraki, 2006; Shimada & Hiraki, 2006), and four fMRI (Baker et al., 2018; Chen et al., 2023; Horowitz-Kraus & Hutton, 2017; Hutton et al., 2020) reported the negative effects of digital experiences on children’s brain function, which includes functional connectivity, cognitive and language processing, and executive function. In contrast, two EEG/ERP studies (Bergen et al., 2017; Mondéjar et al., 2016), one fNIRS study (Li et al., 2017), and three fMRI studies (Chaarani et al., 2022; Han et al., 2007; Murray et al., 2006) reported the positive and training effects of digital experience on children’s brain function. In addition, two studies (Baumgartner et al., 2006; Cantlon et al., 2013) reported both positive and negative effects. These findings jointly indicate that early digital experience can positively and negatively shape children’s brain function, with more negative than positive effects. This result demonstrates two directions of functional plasticity of young brains: positive and negative adaptation to respond to the digital experience. However, the underlying neuromechanisms are still unclear and deserve further studies.

Second, this scoping review found that nine studies reported the impact of digital experiences on the structure of young children’s brains: two reported no impact, five suggested negative, and two presented mixed results. In particular, two studies (Rodriguez-Ayllon et al., 2019; Zhao et al., 2022) reported no noticeable impact on the brain’s structural integrity. This null result might be related to a Type II error, which means not rejecting the null hypothesis when it is false. This error could be caused by the problematic research design or methods. For example, the areas of interest in these studies might not include the real affected brain areas, or the experimental tasks might have failed to activate the target brain areas. Nevertheless, further studies are needed to verify this “null” finding. In contrast, four MRI studies (Hutton et al., 2022; Takeuchi et al., 2016, 2018; Zhao et al., 2023) and one DTI study (Hutton et al., 2020) found a negative impact. In addition, two studies (Paulus et al., 2019; Takeuchi et al., 2013) found both positive and negative effects, but no studies purely reported positive impacts. Therefore, we can conclude that digital experience has basically negative positive impacts on the structure of children’s brains. This result implies that structural plasticity is primarily a negative response to the early digital experience, which also deserves further investigation.

Last but not least, one study (Pujol et al., 2016) reported both structural and functional influences of digital experience. This one-year longitudinal study aims to evaluate the long-term impact of video gaming play (VGP) on brain structure and function with behavioral and neuroimaging evidence. The VGP experience can enhance children’s response speed to visual stimulation, resulting in structural and functional changes in basal ganglia circuitry. This training effect of the VGP experience is evidenced by the higher functional connectivity in the putamen and caudate nucleus maps. In conclusion, this evidence supports the positive and negative effects of digital experience on the structural and functional development of children’s brains, demonstrating both structural and functional plasticity.

Which Brain Area(s) and Functions are Affected

First, this scoping review found that altogether 15 studies (Baker et al., 2018; Baumgartner et al., 2006; Bergen et al., 2017; Chen et al., 2023; Kostyrka-Allchorne et al., 2019; Law et al., 2023; Lewin et al., 2023; Li et al., 2017, 2019, 2020, 2021; Matsuda & Hiraki, 2006; Mondéjar et al., 2016; Twait et al., 2019; Zivan et al., 2019) reported the impact of digital experience on the function and structure of the frontal lobe, especially the prefrontal cortex (PFC), the neural base of executive function. In particular, some studies (Li et al., 2017, 2021; Matsuda & Hiraki, 2006) focused on the dorsolateral prefrontal cortex (DPFC), which is the front part of the frontal lobe of the cerebral cortex and includes lateral part of Brodmann’s area 9 and 46. DPFC is involved in executive functions, such as working memory, cognitive flexibility, planning, inhibition, and abstract reasoning (Li et al., 2021). Therefore, we can conclude that early digital experience could primarily and significantly shape children’s PFC and executive function, both positively and negatively.

Second, this scoping review found that some studies reported the impact of digital experience on the parietal/temporal/occipital regions of the brain. In particular, Cantlon et al. (2013) found that the math-related digital experience would impact the intraparietal sulcus (IPS) on the lateral surface of the parietal lobe. IPS is extensively involved in perceptual-motor coordination, visual attention, processing symbolic numerical information, visuospatial working memory, and interpreting the intent of others. In addition, Han et al. (2007) found that cartoon viewing experience would enhance the activation of the bilateral middle temporal cortex (MT) and posterior superior temporal sulcus (STS), bilateral superior parietal lobule, and medial prefrontal cortex (MPFC): And Baumgartner et al. (2006) found that high spatial presence experiences in VR world would enhance the activation of parietal/occipital regions of the brain. Recently, Hutton et al. (2022) reported that higher media use could cause Cortical Thickness (CT) differences in occipital regions. Therefore, we can conclude that early digital experience could shape children’s parietal/temporal/occipital lobes structurally and functionally.

Third, this scoping review also found that digital experience could shape brain connectivity and networks structurally and functionally. In particular, Hutton and his colleagues (Horowitz-Kraus & Hutton, 2017; Hutton et al., 2020, Study #15, #21) found that screen time was related to lower connectivity between the seed area and other regions. They could cause structural and functional changes in language, visual, and cerebellar networks. Paulus et al. (2019) reported similar findings. Meanwhile, Takeuchi and his colleagues (Takeuchi et al., 2013, 2016, 2018) found that media experience would cause long-term structural changes in the brain networks, including bilateral perisylvian areas, the bilateral temporal pole, the bilateral cerebellum, bilateral medial temporal lobe structures (hippocampus and amygdala), bilateral basal ganglia structures, the bilateral inferior temporal lobe, the thalamus, the bilateral orbitofrontal gyrus and lateral prefrontal cortex, the insula, and the left lingual gyrus. Pujol et al. (2016) reported similar findings. Recently, Zhao et al. (2023) found that screen-media experience would cause a slower expansion in subcortical regions, such as the brainstem and left putamen.

In conclusion, the synthesized evidence supports that digital experience could cause structural and functional changes in children’s frontal, parietal, temporal, and occipital lobes, brain connectivity, and brain networks.

Whether Short-Term or Longitudinal

This scoping review found that most of the reviewed studies reported digital experience’ prompt effects or short-term influences on brain functions; only four studies (Studies #7, 9, 16, 20) provided longitudinal evidence to demonstrate its long-term impact on brain structures. The leading team is Takeuchi and his colleagues, who conducted at least three longitudinal studies to explore the long-term impact of TV viewing, Video Game playing, and Internet use, respectively. First, Takeuchi et al. (2013) found that TV viewing could cause increased regional gray matter volume of the frontopolar and medial prefrontal areas and the hypothalamus/septum and sensorimotor areas. However, TV viewing would also produce reduced verbal intelligence quotient (IQ) longitudinally. Later, Takeuchi et al. (2016) found that Video Game playing experience would cause greater diffusion tensor imaging mean diffusivity (MD) in extensive regions and lower verbal intelligence, longitudinally. Last, Takeuchi et al. (2018) found that frequent Internet use could cause a significant negative effect on gray and white matter volumes in extensive bilateral perisylvian areas, the bilateral temporal pole, the bilateral cerebellum, bilateral medial temporal lobe structures (hippocampus and amygdala), bilateral basal ganglia structures, the bilateral inferior temporal lobe, the thalamus, the bilateral orbitofrontal gyrus and lateral prefrontal cortex, the insula, and the left lingual gyrus. Meanwhile, Pujol et al. (2016) evaluated the 1-year long-term impact of video gaming on brain structure and function. They found structural and functional brain changes associated with gaming use concerning basal ganglia circuitry. And the positive impact is that video gaming was associated with higher functional connectivity in the putamen and caudate nucleus maps. Synthesizing these longitudinal findings, we can conclude that digital experience has both positive and negative impacts on children’s brains longitudinally. But, more longitudinal neuroimaging studies are needed to consolidate this conclusion.

Conclusions, Limitations, and Implications

In summary, this scoping review has achieved three major conclusions. First, digital experience does have positive and negative impacts on children’s brains, structurally and functionally. Second, digital experience could cause structural and functional changes in children’s frontal, parietal, temporal, and occipital lobes, brain connectivity, and brain networks. And the most vulnerable area is the prefrontal cortex and its associated executive function. Third, digital experience has positive and negative impacts on children’s brain structure longitudinally.

However, this scoping review has three major limitations. First, only 33 studies were analyzed in this scoping review. This small sample size might be because this topic is novel and emerging, and research technologies are also evolving. We anticipate that there will be an exploration of research in this area in the following years. Second, this scoping review only covered the academic journals published in English, leaving those published in other languages, such as Chinese, French, and German, unexamined. This is because we lack multilingual resources and time. Nevertheless, given that English is the lingua franca of academic publications, the impact of this minor limitation is limited. Third, without consolidated neuroimaging evidence through comparative studies, this scoping review has not addressed the critical questions, such as whether it is the early digital use (for example, screen time) or the cognitive processes (i.e., learning experience) that have driven the change of brain function and structure, and whether there are different effects of digital equipment types and the mode of use. As most studies reviewed use cross-sectional designs, causal inferences about the effects of early digital experience on brain changes should be made with caution.

Nevertheless, the findings of this scoping review do have some implications for future directions. First, more studies should examine the “dose effect” of digital experience. The existing studies adopted different measures of TV viewing, Video Game playing, and Internet use experience to understand their impact on brain development, currently and longitudinally. However, they neglected the relationship between the dose of this experience and the magnitude of the biological change it caused. A dose-effect study could demonstrate how the response of a population varies with different levels of exposure or doses to the digital experience. Future studies could employ logistic regression to model the dose-effect relationship and estimate parameters such as potency, efficacy, and threshold. Second, there should be more studies on the longitudinal impact on brain functions. Four studies (Studies #7, 9, 16, 20) just reported the long-term impact on children’s brain structure, and one study (Li et al., 2021) indicated that heavy Pad use would damage young children’s brain function, such as EF. These findings imply that digital experience might have a long-term impact on brain functions, which deserves future consolidating evidence. Third, there should be more brain network studies on the impact of digital experience on brain structures and functions, as the existing findings indicated that the impact could be holistic and comprehensive. Given that the EEG-NIRS-MRI coupling technologies are evolving rapidly and more brain network indices are emerging, we anticipate that there will be more brain network studies to explore how the digital experience shapes the whole brain currently and longitudinally.

Last but not least, The conclusions drawn from this investigation contain significant implications for practical improvement and policymaking. Foremost, it should be recognized by both educators and caregivers that children’s cognitive development may be influenced by their digital experiences. As such, they should supply suitable guidance, involvement, and backing for children’s digital use. Limiting their screen time is an effective but confronting way, more innovative, friendly, and practical strategies could be developed and implemented. Secondly, it is imperative for those in positions of policy-making to develop and execute policies grounded in empirical evidence to safeguard and enhance brain development in children as they navigate the digital era. This could also involve offering resources and incentives for the creation and examination of digital interventions aimed at bolstering brain growth in children, a topic widely deserving of further investigation.

Author Contributions

Conceptualization: D.W. and H.L.; Methodology: D.W. and H.L.; Data analysis: D.W., X.D., and D.L.; Tables and Figures: X.D. and D.L.; Writing: D.W., H.L., X.D., and D.L. All authors have read and agreed to the published version of the manuscript.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data will be provided by the authors upon a written request.