Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 May 1.
Published in final edited form as: Child Neuropsychol. 2022 Aug 11;29(4):666–685. doi: 10.1080/09297049.2022.2110577

Higher Access to Screens is Related to Decreased Functional Connectivity Between Neural Networks Associated with Basic Attention Skills and Cognitive Control in Children

Raya Meri 1, John Hutton 3, Rola Farah 1, Mark DiFrancesco 3, Leonid Gozman 6, Tzipi Horowitz-Kraus 1,2,4,5
PMCID: PMC10619703  NIHMSID: NIHMS1833332  PMID: 35957604

Abstract

Screen-based media has become a prevailing part of children’s lives. Different technologies provide limitless access to a wide range of content. This accessibility has immensely increased screen exposure among children, showing that this exposure is associated with decreased cognitive abilities. This study was designed to evaluate how the neurobiological correlates for different sub-components of screen exposure, such as level of access, content, and frequency, are related to different cognitive abilities.

Resting-state functional MRI data were collected in 29 native English-speaking children (8–12 years old), in addition to cognitive-behavioral measures. Functional connectivity measures within and between several networks related to cognitive control and attention were calculated [fronto-parietal (FP), cingulo-opercular (CO), dorsal attention (DAN), ventral attention (VAN), salience, default mode (DMN), cerebellar networks]. Sub-components of screen exposure were measured using the Screen-Q questionnaire.

Higher access to screens was related to lower functional connectivity between neural networks associated with basic attention skills and cognitive control (i.e., DAN and salience). In addition, higher levels of parent-child interaction during screen exposure were related to increased functional connectivity between networks related to cognitive control and learning (i.e., CO and cerebellar).

These findings suggest that screen exposure may reduce the engagement of basic attention and modulation of cognitive control networks and that higher levels of parent-child interaction engage cognitive control networks. An enhanced understanding of these processes can provide an important scientific basis for future educational and medical approaches regarding screen exposure.

Keywords: Brain networks, Child development, Cognitive control, Parent-child interaction, Resting-state fMRI, Screen exposure

1. Introduction

Child cognitive development is a summation of endogenous (both biological and psychological factors such as genetics, age, gender, and health) and exogenous factors (such as the child’s immediate environment, which includes school, peers, family, and mass media; see Bronfenbrenner’s ecological model (Bronfenbrenner, 1992)). In recent years, screen exposure has been another exogenous component that affects child cognitive development (Ajrezo, Wiener-Vacher, Bucci, & Bui-Quoc, 2016). Screen-based media has become a prevailing part of people’s lives, particularly children (Drouin, 2020; McDaniel, 2016; Newsham, 2020; Radesky & Christakis, 2016; J. C. Zimmerle, 2019). The availability of different technologies, including portable devices, is burgeoning and provides unprecedented access to a wide range of content (Hutton, Huang, Sahay, DeWitt, & Ittenbach, 2020). This accessibility has immensely increased screen exposure among young children, leading researchers to investigate its association with cognitive abilities using behavioral and neuroimaging tools.

1.1. Screen exposure in children- behavioral evidence

Screen-based media has broken out in the last 20 years and has been fundamentally affecting humans, especially children, in several aspects of life, including social interactions, emotional support, and cognitive abilities (Johnson, Cohen, Kasen, & Brook, 2007; Johnson, Cohen, Smailes, Kasen, & Brook, 2002; Pediatrics, 2013). According to the American Association of Pediatrics (AAP) recommendations, children above the age of 2 years can consume 1–2 hours of screen exposure daily (Pediatrics, 2013). Regardless of these recommendations, 90% of infants under two years are frequently exposed to screens while exceeding the recommended 2-hour limit in kindergarten (Rideout, 2017). This excessive screen exposure has been associated with numerous negative outcomes, including obesity (Mendoza, Zimmerman, & Christakis, 2007), language delays (Chonchaiya & Pruksananonda, 2008), attention problems (Zimmerman & Christakis, 2007), and deterioration of cognitive abilities (Lillard, Li, & Boguszewski, 2015). Hence, several studies have inspected the impact of screen exposure on cognitive abilities in children using behavioral and cognitive batteries. It was suggested that television viewing has adverse effects on children under two years, particularly on language and executive functions due to reduced parent-child interaction. Moreover, and despite many positive associations found for screen exposure among children, excessive screen exposure was suggested to replace fundamental cognitive activities, such as reading (Anderson & Subrahmanyam, 2017).

1.2. Screen exposure in children- neurobiological evidence

The rapid use of screens among young children and its effect on executive functions (EF) (Huber, Yeates, Meyer, Fleckhammer, & Kaufman, 2018), have raised interest in also examining its neurobiological association with children using several neuroimaging tools, such as structural and functional Magnetic Resonance Imaging (sMRI, fMRI, respectively). Several studies examined the relationship between screen exposure time and the involvement of visual processing in children, whereas others examined the cognitive aspects related to it. On the visual processing side, Murray et al. (2006) suggested that children ages 8-12-years who viewed violent television content recruited several brain regions involved in emotional regulation, attention, and episodic memory, such as the amygdala and the precuneus. Moreover, ‏Paulus et al. (2019) showed a decreased cortical thickness in visual processing regions in 9-year-old children. In another study, a decreased functional connection between the fusiform gyrus (part of the ventral stream also related to word recognition) and regions associated with EF (left anterior cingulate cortex, right insula, left inferior prefrontal cortex), and language (left and right angular gyrus, left and right inferior temporal gyrus) with increased screen time in 8-12-year-old children was shown (Horowitz‐Kraus & Hutton, 2018).

Regarding the EF aspect, the literature indicates that excessive screen exposure in 3-5-year-old children is related to decreased diffusivity in white matter tracts connecting cognitive control, language processing, and visual processing regions (arcuate fasciculus, inferior frontal occipital fasciculus) (Hutton, Dudley, Horowitz-Kraus, DeWitt, & Holland, 2020). In addition, several brain networks can be affected by excessive screen exposure in children.‏ A decreased functional connectivity between the dorsal attention network (DAN, related to alerting attention) and language networks and between the ventral attention network (VAN, related to orienting attention) and the visual perception network in 4-6-year-old children exposed to animated content versus Illustrated content was revealed (Hutton, Dudley, Horowitz-Kraus, DeWitt, & Holland, 2019). These findings point at the reliance on more basic attention networks while viewing animated content, rather than higher order cognitive networks.

Furthermore, it was found that children with and without reading difficulties showed altered functional connectivity between the salience network (associated with cognitive control and modulating different brain networks) and language and EF-related brain regions when correlating with the screen versus reading ratio (Horowitz-Kraus et al., 2020). The association between screen exposure and additional critical networks related to cognitive control has not been studied yet. These networks include the fronto-parietal network (FP, related to adaptive control and flexibility), the cingulo-opercular network (CO, related to error detection), the default mode network (DMN, the task-negative network), and the cerebellum.

Screen exposure was also related to reduced attention abilities in children. A recent electroencephalogram (EEG) study by Zivan et al. (2019) revealed that a higher ratio between theta/beta bands was found in children exposed to storytelling via screens compared to children exposed to storytelling by a storyteller (person), which was also related to decreased visual attention abilities. A similar pattern of excessive theta/beta ratio was previously associated with challenges in attention (Shi et al., 2012). Hence, the study by Zivan and colleagues strengthens these results and also links them to screen exposure. As joint interaction with the parent while using screens was found to improve learning rates in an early study by Salomon (1977), a question then arises regarding the neurobiological association between these different types of screen exposure in children.

The overall goal of the current study was to explore the relations between the type of screen exposure (level of access, content, frequency, and parent-child interaction) using the Screen-Q questionnaire (Hutton, Huang, et al., 2020) and children’s cognitive abilities, focusing on attention and cognitive control. Therefore, we aimed to investigate how different sub-components of screens are related to various cognitive abilities using behavioral and neuroimaging tools. We hypothesized that greater access to screens would be associated with decreased engagement of several networks related to cognitive control. We, therefore, anticipated that children would show reduced functional connectivity within and between DAN, salience, and cerebellar networks. We also hypothesized that greater parent-child interaction would be related to increased engagement of these networks, suggesting greater functional connectivity within and between them. Moreover, we anticipated that screen exposure would not be related to decreased engagement in other networks related to higher-order cognitive ability, including the CO and FP networks. We expect the main involvement would be of more basic attention and modulation networks than EF networks.

Methods

2.1. Participants

Twenty-nine native English-speaking children ages 8–12 years (mean age: 10 years, SD: 1.7 years, nine females), all right-handed, with no history of psychiatric or neurological disorders, participated in this study. There were no differences between the participants regarding average household income ($50,000-$100,000)or maternal level of education (mean=17.92 years, SD=2.22) as measured using A demographic questionnaire (following (Romeo, 2018)). Additionally, academic level as measured using the reading and math tests [letter-word (mean=113.03, SD=11.378) and the math fluency (mean=103.55, SD=13.78) subtests from Woodcock-Johnson battery (Woodcock, 2007), respectively, was within the normal range (85–115). The sample size was based on our previous experience (Horowitz‐Kraus & Hutton, 2018), and in order to reach an effect size of 0.5, it was estimated that 90% power would be achieved using a minimum of n=28 children (https://www.softpedia.com/get/Science-CAD/G-Power.shtml.).

2.2. Study procedure

We only recruited participants that met our inclusion criteria: (1) ages 8–12 years, (2) no history of developmental or psychiatric delays, (3) no history of learning disabilities, (4) intact vision and hearing, (5) within the normal range of verbal and non-verbal abilities. This information was provided by the parent or the guardian and was verified using neurological questionnaires. Attention deficits were exclusionary for this study and were assessed using the Conners parent questionnaire (Conners, 2008). In addition, all participants were screened for MRI suitability and were eligible to undergo an MRI scan.

After meeting the inclusion/exclusion criteria, participants underwent a series of cognitive, language, and reading tests, followed by an fMRI scan as part of a larger study design. Participants and their parents signed written informed assents and consent, respectively, and were compensated with $75 for their time and effort. The study was approved by the Cincinnati Children’s Hospital Medical Center Institutional Review Board committee.

2.3. Behavioral measures

Children were tested for intact verbal and non-verbal IQ using the Peabody Picture Vocabulary Test, 4th edition (PPVT-4, (Dunn & Dunn, 2007)) and the Test of Nonverbal Intelligence, 3rd edition (TONI-3, (Brown, Sherbenou, & Johnsen, 1997), respectively. In addition, all children were administered a comprehensive behavioral assessment to appraise cognitive abilities, as follows:

Cognitive control measures:

Participants were tested for several cognitive control measures, including; (1) speed of processing (subtest of symbol search from the Wechsler Intelligence Scale for Children (WISC, (Wechsler, 1999)), (2) working memory (the digit span task (WISC, (Wechsler, 1999)), (3) attention (sky search attention subtest from the Test of Everyday Attention for children (TEACH) (Manly, Robertson, Anderson, & Nimmo-Smith, 1999)). Moreover, the Behavioral Rating Inventory of Executive Functions questionnaire (BRIEF) was given to the parents to evaluate the child’s general cognitive control abilities (Gioia, Isquith, Guy, & Kenworthy, 2000), with lower scores reflecting higher cognitive control abilities).

Screen exposure measure:

Screen exposure was measured using the Screen-Q questionnaire (Hutton, Huang, et al., 2020). This questionnaire includes several subcomponents that were derived from the American Academy of Pediatrics (AAP) recommendations: (1) access to screens: portable monitor devices, excluded from bedrooms, (2) frequency of use: age of exposure to screens, limitation to 1 hour per day, (3) content: non-violent, slow pace, (4) dialogue: promote parent-child interaction and co-viewing. The Screen-Q questionnaire contains 17 items, with a total score range of 0–30, with the lower scores reflecting obedience to the AAP recommendations and higher scores reflecting more significant use of screens in all subcomponents except for the dialogue subcomponent, which reflects the opposite.

2.4. Neuroimaging measures

2.4.1. Neuroimaging resting-state condition

Ten-minute resting-state data were used to determine the functional connectivity within and between neural networks supporting cognitive control. Participants were instructed to look at a cross presented on the screen scan for two sessions of 5 minutes each. Participants were requested to keep their eyes open and avoid sleeping during the task. Wakefulness was determined using an eye tracker device and monitoring the participant using a video camera in the control room.

2.4.2. Neuroimaging data acquisition

Data were acquired using a 3T Philips Ingenia MRI scanner. A T1-weighted anatomical scan was performed using 8.05 ms repetition time (TR), 3.7 ms echo time (TE), contiguous slices with a 1 mm thickness, and 1 × 1× 1 mm3 voxel size. We used a multiband 4 Echo Planar sequence for the T2*-weighted BOLD fMRI scan with TR/TE = 700/30 ms, FOV= 20 × 20 cm, matrix= 68 × 67, voxel size of 2.4 × 2.4× 2.4 mm3 and slice thickness = 3 mm. For the resting-state scan, 868 whole-brain volumes were acquired for an overall time of 10 minutes.

1.5. Behavioral data analysis

One-sample t-test analyses were conducted for all cognitive-behavioral measures and screen sub-categories from the Screen-Q questionnaire. In addition, to determine the relations between behavioral-cognitive measures and sub-components of screen exposure, Pearson correlation analyses were conducted.

Neuroimaging data analysis

2.6.1. Functional MRI data pre-processing

The resting-state data were pre-processed using SPM-12 (Wellcome Department of Cognitive Neurology, London, UK). Pre-processing steps included the traditional slice time correction, realignment of all images to the first image to control motion correction (6 motion parameters), and segmentation of different tissue types, including grey matter, white matter, and cerebrospinal fluid. Normalization of all images to the Montreal Neurological Institute (MNI)-152 template version ICBM152 non-linear 2009c , and coregistration of the anatomical image to the mean aligned functional image. Smoothing was also applied using an 8-mm full width at half minimum (FWHM) spatial smoothing kernel. The functional connectivity resting-state analysis was achieved using CONN toolbox (Whitfield-Gabrieli and Nieto-Castanon, 2012). We controlled for motion-related artifacts by applying bandpass filtering between 0.08 and 0.9 Hz and linear regression by removing the six motion parameters, scrubbing, and removing the effect of rest. None of the participants were excluded as a result of excessive head motion after applying framewise displacement (FD) above 0.5 mm.

2.6.2. Defining the networks

As the focus of the current work was on the relationship between cognitive control and screen exposure, several related networks were defined in the analysis: VAN, DAN, CO, FP, DMN, salience, and cerebellar networks. All the networks were defined based on the coordinates of 264 regions of interest (ROIs) for the functional networks reported by Power’s atlas (Power et al., 2011). See Figure 1 and Table 1 for the networks’ coordinates.

Figure 1:

Figure 1:

Open in a new tab

Spatial maps for the neural networks. Images presenting the maps for neural networks: cingulo-opercular (CO) in light green, salience network in purple, default mode network (DMN) in dark green, dorsal attention (DAN) in yellow, ventral attention (VAN) in dark blue, frontoparietal (FP) in red, and the cerebellum in light blue. Images are presented in transverse orientation (L=left, R=right).

Table 1:

Regions of interest and their corresponding coordinates:

ROI X Y Z
Fronto-parietal Middle Frontal Gyrus −44 2 46
Middle Frontal Gyrus 48 25 27
Inferior Frontal Gyrus −47 11 23
Inferior Parietal Lobule −53 −49 43
Superior Frontal Gyrus −23 11 64
Middle Temporal Gyrus 58 −53 −14
Superior Frontal Gyrus 24 45 −15
Middle Frontal Gyrus 34 54 −13
Middle Frontal Gyrus 47 10 33
Inferior Frontal Gyrus −41 6 33
Middle Frontal Gyrus −42 38 21
Middle Frontal Gyrus 38 43 15
Inferior Parietal Lobule 49 −42 45
Superior Parietal Lobule −28 −58 48
Inferior Parietal Lobule 44 −53 47
Superior Frontal Gyrus 32 14 56
Inferior Parietal Lobule 37 −65 40
Inferior Parietal Lobule −42 −55 45
Middle Frontal Gyrus 40 18 40
Middle Frontal Gyrus −34 55 4
Middle Frontal Gyrus −42 45 −2
Inferior Parietal Lobule 33 −53 44
Middle Frontal Gyrus 43 49 −2
Middle Frontal Gyrus −42 25 30
Middle Frontal Gyrus −3 26 44
Cingulo-Opercular Medial Frontal Gyrus −3 2 53
Inferior Parietal Lobule 54 −28 −34
Middle Frontal Gyrus 19 −8 64
Superior Frontal Gyrus −16 −5 71
Cingulate Gyrus −10 −2 42
Insula 37 1 −4
Superior Frontal Gyrus 13 −1 70
Medial Frontal Gyrus 7 8 51
Precentral Gyrus −45 0 9
Superior Temporal Gyrus 49 8 −1
Claustrum −34 3 4
Superior Temporal Gyrus −51 8 −2
Cingulate Gyrus −5 18 34
Insula 36 10 1
Ventral attention network Superior Frontal Gyrus −10 11 67
Inferior Parietal Lobule 54 −43 22
Superior Temporal Gyrus −56 −50 10
Superior Temporal Gyrus −55 −40 14
Superior Temporal Gyrus 52 −33 8
Middle Temporal Gyrus 51 −29 −4
Superior Temporal Gyrus 56 −46 11
Inferior Frontal Gyrus 53 33 1
Inferior Frontal Gyrus −49 25 −1
Dorsal attention network Precuneus 10 −62 61
Middle Temporal Gyrus −52 −63 5
Precuneus 22 −65 48
Middle Temporal Gyrus 46 −59 4
Superior Parietal Lobule 25 −58 60
Sub-Gyral −33 −46 47
Precuneus −27 −71 37
Middle Frontal Gyrus −32 −1 54
Sub-Gyral −42 −60 -8
Superior Parietal Lobule −17 −59 64
Middle Frontal Gyrus 29 −5 54
Cerebellar network Declive −16 −65 −20
Culmen −32 −55 −25
Declive 22 −58 −23
Declive 11 −62 −18
Salience Paracentral Lobule 55 −45 37
Supramarginal Gyrus 42 0 47
Middle Frontal Gyrus 31 33 26
Sub-gyral 48 22 10
Inferior Frontal Gyrus −35 20 0
Extra-Nuclear 36 22 3
Insula 37 32 −2
Inferior Frontal Gyrus 34 16 −8
Extra-Nuclear −11 26 25
Anterior Cingulate −1 15 44
Cingulate Gyrus −28 52 21
Middle Frontal Gyrus 0 30 27
Cingulate Gyrus 5 23 37
Anterior Cingulate 10 22 27
Middle Frontal Gyrus 31 56 14
Superior Frontal Gyrus 26 50 27
Superior Frontal Gyrus −39 51 17
Open in a new tab

ROI region of interest. All ROIs and their coordinates are based on the study by Power et al. (2011).

To determine the functional connectivity patterns of these networks during the resting-state condition, functional connectivity measures between and within each network were calculated: (1) within networks: an average of all ROIs connections in each of the networks, (2) between networks: an average of all ROI-to-ROI functional connectivity measures between all pairs of ROI in all of the networks. All analyses were corrected for multiple comparisons using a False-Discovery Rate (FDR) correction P<.05.

1.6. Correlations between behavioral and neuroimaging measures

Several Pearson correlations were used to associate the screen exposure subcomponents (as measured using the Screen-Q) with functional connectivity within and between the cognitive networks.

3. Results

3.1. Behavioral measures

Children’s performance in the cognitive-behavioral battery was within the normal range (see Table 2). Pearson correlations between cognitive control and screen exposure measures revealed a significant positive correlation between subcomponents of Screen-Q and several cognitive control abilities. First, a significant positive correlation between screen content and working memory (digit span, WISC) (r=0.400, P= 0.039)] was found. In addition, a significant negative correlation between the dialogic subcomponent and speed of processing (coding, WISC) (r=-0.407, P=0.035) was observed. Overall, better EF abilities (working memory, speed of processing) were related to greater screen content measure and increased screen dialogue measure while exposed to screens, respectively. See Table 3.

Table 2.

Means and standard deviations for the non-verbal, verbal and cognitive measures.

Cognitive ability Measure Mean (SD) Range of normal values
General non-verbal IQ TONI (Percentile) 106.66 (19.01) 85–115
General verbal IQ PPVT (Standard Score) 119.46 (17.01) 85–115
Memory Digit Span, WISC (Standard Score) 10.93 (2.55) 7–13
Speed of processing Symbol Search, WISC (Standard Score) 11.31 (2.72) 7–13
Attention Sky Search Attention, Teach (Scaled Score) 8.46 (2.983) 7–13
Math fluency Math fluency, Woodcock-Johnson (Scaled Score) 103.55 (13.78) 90–110
Non-word reading fluency Letter-word, Woodcock-Johnson (Scaled Score) 113.03 (11.378) 90–110
Open in a new tab
‡

IQ intelligence quotient, PPVT peabody picture vocabulary test, SD standard deviation, TEACH test of everyday attention for children, TONI test of nonverbal intelligence, WISC wechsler intelligence scale for children

Table 3.

Pearson correlations between Screen Q subcomponents and IQ and cognitive control measures

Condition Verbal IQ [PPVT, SD R(P)] Non-verbal IQ [Toni Correct Responses R(P)] Working memory [WISC Digit Span SS R(P)] Speed of processing [WISC Coding SS R(P)] General EF [Brief PR GEC score R(P)]
Screen Q Access total 0.060 (0.771) −0.170 (0.396) −0.102 (0.613) 0.013 (0.950) 0.360 (0.065)
Screen Q Frequency total −0.376 (0.054) 0.0140 (0.476) −0.048 (0.807) −0.031 (0.875) 0.220 (0.261)
Screen Q Content total −0.175 (0.392) 0.176 (0.379) 0.400* (0.039) −0.284 (0.151) 0.198 (0.323)
Screen Q dialogic total −0.199 (0.330) 0.047 (0.816) 0.324 (0.099) −0.407* (0.035) 0.146 (0.469)
Screen Q total score 0.244 (0.241) 0.112 (0.396) 0.168 (0.412) −0.216 (0.289) 0.358 (0.073)
Open in a new tab
§

BRIEF behavioral rating inventory of executive functions, EF executive functions, IQ intelligence quotient, PPVT peabody picture vocabulary test, SD standard deviation, SS standard score, TONI test of nonverbal intelligence, WISC wechsler intelligence scale for children

*

P<0.05

3.2. Functional connectivity measures

Within and between networks’ functional connectivity for the whole group was calculated and revealed a positive functional connectivity between DAN and salience networks and between CO and cerebellar networks. See Figure 2 and Table 4 for more details.

Figure 2:

Figure 2:

Open in a new tab

Within and between networks’ functional connectivity matrix. A representation of the within and between networks’ functional connectivity correlation matrix. Blue and red colors represent negative and positive within and between networks’ functional connectivity values, respectively. Hotter colors represent more positive functional connectivity values; cooler colors represent more negative functional connectivity values.

Table 4.

Within and between networks functional connectivity

Within network connectivity Average SD
CO 0.433 0.068
FP 0.279 0.069
DAN 0.336 0.085
VAN 0.345 0.090
DMN 0.244 0.054
Salience 0.343 0.081
Cerebellar 0.506 0.123
Between networks connectivity
DAN-VAN 0.054 0.080
DAN-FP 0.091 0.057
DMN-Salience 0.006 0.063
DAN-salience 0.041 0.067
DAN-DMN −0.084 0.092
DAN-cerebellar 0.059 0.087
DAN-CO 0.062 0.086
FP-VAN 0.082 0.074
FP-DMN 0.055 0.082
FP-salience 0.134 0.066
FP-cerebellar −0.030 0.062
DMN-cerebellar −0.036 0.054
VAN-cerebellar 0.001 0.081
VAN-DMN 0.112 0.111
VAN-salience 0.159 0.082
Salience-cerebellar 0.019 0.074
CO-cerebellar 0.082 0.097
CO-DMN −0.121 0.104
CO-FP −0.015 0.108
CO-salience 0.213 0.895
CO-VAN 0.165 0.117
Open in a new tab
**

CO cingulo-opercular, DAN dorsal attention network, DMN default mode network, FP fronto parietal, VAN ventral attention network.

3.3. Correlations between behavioral and neuroimaging measures

Pearson correlations between Screen-Q subcomponents and between and within networks’ functional connectivity were conducted and revealed a significant negative correlation between access subcomponent and functional connectivity between DAN and salience networks (r=-0.444, P=0.020). Moreover, a significant negative correlation between screen access measures and within the cerebellar network functional connectivity was found (r=-0.442, P=0.021). Lastly, a significant negative correlation was found between dialogic measure and between CO and cerebellar networks’ functional connectivity (r=-0.486, P=0.010). Overall, higher screen access measure was related to decreased functional connectivity between networks related to attention and modulation of cognitive control, and increased functional connectivity between networks related to cognitive control and learning was associated with lower screen dialogue measure.

The correlation results are shown in Table 5 and Figures 3 and 4.

Table 5.

Pearson correlations between Screen Q subcomponents and between networks functional connectivity

Condition Between DAN Cerebellar R(P) Between DAN FP R(P) Between DAN salience R(P) Between CO Cerebellar R(P) within cerebellar R(P) within DAN R(P) within FP R(P)
Screen Q Access total −0.271 (0.171) 0.043 (0.830) −0.444* (0.020) −0.018 (0.928) −0.442* (0.021) −0.059 (0.771) −0.243 (0.223)
Screen Q Frequency total −0.046 (0.816) 0.055 (0.780) −0.008 (0.969) 0.215 (0.227) 0.083 (0.673) −0.018 (0.927) −0.098 (0.621)
Screen Q Content total −0.060 (0.716) −0.139 (0.489) −0.072 (0.720) −0.164 (0.414) −0.103 (0.609) −0.302 (0.126) 0.242 (0.224)
Screen Q dialogic total 0.009 (0.966) 0.022 (0.912) 0.005 (0.978) −0.486* (0.010) −0.079 (0.694) −0.018 (0.930) 0.296 (0.134)
Screen Q total score −0.132 (0.519) 0.049 (0.811) −0.157 (0.443) −0.132 (0.521) −0.286 (0.157) −0.151 (0.461) 0.178 (0.385)
Open in a new tab
††

CO cingulo-opercular, DAN dorsal attention network, FP fronto parietal.

*

P<.05

**

P<.01

Figure 3:

Figure 3:

Open in a new tab

Correlation matrices of the within networks’ functional connectivity and the screen sub-components. A representation of the correlation matrices of the within networks’ functional connectivity and the screen sub-components. (A) represents the access sub-component, (B) represents the dialogic sub-component, (C) represents the content sub-component, and (D) represents the frequency sub-component. The y-axis represents the participants and the x-axis represents the within networks’ functional connectivity. Participants closer to the origin have higher screen scores. A stronger correlation is noted in a hotter color.

Figure 4:

Figure 4:

Open in a new tab

Correlation matrices between networks’ functional connectivity and the screen sub-components. A representation of the correlation matrices between networks’ functional connectivity and the screen sub-components. (A) represents the access sub-component, (B) represents the dialogic sub-component, (C) represents the content sub-component, and (D) represents the frequency sub-component. The y-axis represents the participants (as participants closer to the origin have higher screen scores), and the x-axis represents the between networks’ functional connectivity. The blue and red colors represent the negative and positive correlations, respectively.

4. Discussion

The goal of the current study was to examine the relationship between cognitive abilities (or EF), sub-components of screen exposure (e.g., level of access, content, frequency, and dialogic), and the neurobiological correlates for these different sub-components of screens in 8-12-year-old English speaking children. In line with our hypotheses, significant correlations were found between screen exposure subcomponents and cognitive control abilities. Although the access-to-screen subcomponent was not correlated with behavioral or cognitive abilities, other screen subcomponents, such as the dialogic subcomponent, showed a negative correlation with the speed of processing scores and the content subcomponent, which was positively correlated with working memory abilities. Moreover, the correlations between the dialogic sub-component of screen exposure and the neuroimaging data revealed a decreased functional connectivity between DAN and salience networks and increased functional connectivity between CO and cerebellar networks.

4.1. Higher screen access in children is linked to decreased functional connectivity in networks related to attention and cognitive control abilities

The findings of our study revealed a negative correlation between the Screen-Q access subcomponent and between networks’ functional connectivity for DAN and salience. This result implies higher access to screens was related to decreased functional connectivity of neural networks supporting basic attention and modulation of cognitive control abilities. As previously suggested, DAN is related to alerting attention and is part of the attentional network model (Petersen & Posner, 2012). This model states that DAN and VAN networks interact with higher-order cognitive abilities associated with the CO and FP networks (Dosenbach, Fair, Cohen, Schlaggar, & Petersen, 2008). Moreover, part of the DAN includes the putative word form area (or the “fusiform gyrus) (Vogel, 2014), which was previously related to exceeding screen time in 8–12 years old children (Horowitz- Kraus, 2017), reinforcing our results. The current results suggest tight interaction between networks related to cognitive control (DAN) and to modulation of cognitive control (salience network). The role of the salience network is diverse and includes error monitoring and switching (Twait, Farah, & Horowitz-Kraus, 2018). Still, it is also responsible for monitoring performance and connectivity between different neural networks (Ham, Leff, de Boissezon, Joffe, & Sharp, 2013). Interestingly, the current study results suggest that higher access to screens among children may diminish the engagement of the DAN and salience networks. Moreover, the results indicate that the more access children have to screens, the lower the functional connectivity is within the cerebellar network, which is associated with learning (Doya, 2000). As we suggested, no correlations were found between FP and CO networks and Screen-Q access. This finding indicates that networks related to higher-order cognitive abilities were not engaged during higher screen exposure. This may be linked to the fact that screen exposure was assessed only during activities that required basic attention.

The questions related to accessibility to screens in the Screen-Q access measure include the existence of TV and screens in the child’s bedroom. Previous studies suggested that the presence of a TV in the child’s bedroom may cause sleep disturbances (Brockmann et al., 2016). In the current study, we found that higher access to screens is related to decreased between networks functional connectivity of DAN and salience networks and within the cerebellar network. The reduced engagement of the cognitive control networks (DAN, salience) and learning networks (cerebellar network) that was found in our study may be related to sleep deprivation which was reported to be associated with screen exposure. Sleep deprivation can subsequently lead to dysfunction in cognitive and learning abilities. These results significantly augment the AAP recommendations to limit screen exposure and accessibility of screens in the child’s room.

4.2. Parent-child interaction during screen exposure is associated with better cognitive abilities

In support of previous evidence concerning the importance of parent-child interaction to the cognitive, social, and emotional development of the child (Ginsburg, 2007; Padilla‐Walker et al., 2020), this study’s findings demonstrate a significant negative correlation between the Screen-Q dialogic subcomponent and CO and cerebellar between networks functional connectivity. This finding suggests that higher parent-child interaction is associated with increased functional connectivity between these two networks, as lower scores in the dialogic measure reflect higher parent-child interaction. The CO network plays a crucial part in the attentional network model and is responsible for error monitoring (Dosenbach et al., 2007). The result of this study may suggest that when interacting with an adult, the child might be able to monitor errors better, leading to better learning patterns. In addition, recent studies accounted for cerebellar involvement in cognitive, social-emotional, and learning abilities (Buckner, 2013; Keren‐Happuch, Chen, Ho, & Desmond, 2014), adding to its traditional role in motor function (Llinás & Welsh, 1993).

Similarly, Hutton et al. (2017) have also proposed a central role for the cerebellum in cognitive function and learning by demonstrating an increased activation in the cerebellum during storytelling in relation to maternal reading engagement. Our study is distinctive from Hutton’s by examining parent-child interaction during screen exposure rather than shared reading and enrolling older children (8-12-years-old vs. 3-5-years-old in Hutton’s study). However, both studies support the significance of healthy communication between parents and children and emphasize its influence on the child’s cognitive abilities and the role of the cerebellum in these processes. These results reinforce the involvement of cognitive control regions/networks, as was also indicated previously, showing that a lack of parent-child interaction leads to an opposite phenomenon to that of our current findings.‏ Farah et al. (2020) have implied increased functional connectivity between cognitive control and visualization regions in children born to mothers who suffered from maternal depression and had reduced interaction with them.

Moreover, our study revealed that a higher speed of processing abilities is related to increased dialogue measures. This correlation highlights the importance of joint interaction with the parent during screen exposure. To avoid gender bias, as we had more male participants (20) than females (9), we re-analyzed the behavioral and neuroimaging correlations while controlling for gender. When controlling for gender, the correlations between screen dialogue and processing speed were absent (data not shown). This might be due to the enhanced verbal development in females vs males contributing to the original significant correlation when analyzing the full cohort(Adani & Cepanec, 2019). This may point to the need for additional effort from the parents to engage males during screen co-viewing.

Taken together, the results of this study emphasize the critical role of exogenous factors (such as screen exposure) on child cognitive development, specifically related to cognitive control. The results also support the “technoference” phenomenon, suggesting that parental screen exposure may be associated with less interaction with their children. Here, the results show that although exposure to screens may be related to decreased cognitive abilities in the child, interaction with the child around the screen may be a protective factor and limit the technoference phenomenon (Joanna C Zimmerle, 2019).

4.3. Study’s limitations

Despite being of great importance to the recent literature regarding screen exposure in children, our study has several limitations. First, we collected neuroimaging and behavioral data at only one time point. More time-points (i.e., longitudinal studies) could reveal developmental changes in functional brain connectivity linked to screen exposure and point to causal relations between screen exposure and cognitive development. Second, screen exposure was measured using a questionnaire and not by direct measurement (e.g., observations etc.). Additionally, this questionnaire ignores multiple aspects of screen use that may impact child development, most prominently social media and active vs. passive viewing. In-depth parcellation of the type of activity done on the screen is essential in further research. Third, gender effects were not taken into account in our analyses. Future research should address the influence of gender differences, such as motivation and on-screen exposure. Furthermore, in the current study, we did not document several factors that may impact the participants’ screen exposure and co-viewing, such as type of schooling (private, public, or homeschooling). Future studies should take into consideration the various environmental, educational, and social factors of the participants to expand the understanding of the effect of environmental/academic factors on screen exposure and co-viewing at this age

4.4. Conclusions

The uniqueness of our study is reflected through the focus on various neural networks that are associated with attention and cognitive control abilities, as well as the focus on cognitive outcomes of screen exposure beyond the pre-school years. The results reinforce the importance of parent-child interaction even at a relatively older age (8–12 years), including screen exposure, highlighting its association with neural network alterations related to cognitive control. Additionally, our results support the AAP recommendations regarding screen exposure in children. Our findings can provide a valuable scientific basis for future educational and medical approaches regarding screen exposure in children. Nevertheless, an enhanced understanding of these processes is still desirable.

Highlights.

  • Higher screen access is associated with decreased between-networks functional connectivity in networks related to basic attention and modulation of cognitive control.

  • Joint-viewing the screens with a parent is associated with increased between-networks functional connectivity in cognitive control, learning and social-emotional-related networks.

  • There is a positive relationship between speed of processing abilities and screen co-viewing with the parent.

Acknowledgment:

This study was supported by the National Institute of Child Health and Human Development (R01 HD086011; PI: Horowitz-Kraus).

Footnotes

Conflict of interest statement: The authors have no conflicts of interest to report.

Ethics approval statement: The research protocols used in this study were approved by the Cincinnati Children’s Hospital Medical Center IRB committee.

Data availability statement:

The data will be available upon a reasonable request.

References

  1. Adani S, & Cepanec M (2019). Sex differences in early communication development: behavioral and neurobiological indicators of more vulnerable communication system development in boys. Croatian medical journal, 60(2), 141–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ajrezo L, Wiener-Vacher S, Bucci MP, & Bui-Quoc E (2016). Influence of screen exposure on vergence components from childhood to adolescence. International Journal of Ophthalmology & Eye Science, 4(1), 170–177. [Google Scholar]
  3. Anderson DR, & Subrahmanyam K (2017). Digital screen media and cognitive development. Pediatrics, 140(Supplement 2), S57–S61. [DOI] [PubMed] [Google Scholar]
  4. Brockmann PE, Diaz B, Damiani F, Villarroel L, Núñez F, & Bruni O (2016). Impact of television on the quality of sleep in preschool children. Sleep medicine, 20, 140–144. [DOI] [PubMed] [Google Scholar]
  5. Bronfenbrenner U (1992). Ecological systems theory: Jessica Kingsley Publishers. [Google Scholar]
  6. Brown L, Sherbenou RJ, & Johnsen SK (1997). TONI-3, test of nonverbal intelligence: A language-free measure of cognitive ability: Pro-ed
  7. Buckner RL (2013). The cerebellum and cognitive function: 25 years of insight from anatomy and neuroimaging. Neuron, 80(3), 807–815. [DOI] [PubMed] [Google Scholar]
  8. Chonchaiya W, & Pruksananonda C (2008). Television viewing associates with delayed language development. Acta Paediatrica, 97(7), 977–982. [DOI] [PubMed] [Google Scholar]
  9. Conners CK (2008). Conners ADHD Index (AI) Scale, Third Edition (Conners-3). North Towonada NY Multihealth System
  10. Dosenbach NU, Fair DA, Cohen AL, Schlaggar BL, & Petersen SE (2008). A dual-networks architecture of top-down control. Trends in cognitive sciences, 12(3), 99–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dosenbach NU, Fair DA, Miezin FM, Cohen AL, Wenger KK, Dosenbach RA, … Raichle ME (2007). Distinct brain networks for adaptive and stable task control in humans. Proceedings of the National Academy of Sciences, 104(26), 11073–11078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Doya K (2000). Complementary roles of basal ganglia and cerebellum in learning and motor control. Current opinion in neurobiology, 10(6), 732–739. [DOI] [PubMed] [Google Scholar]
  13. Drouin M, McDaniel BT,Pater J, Toscos T (2020). How Parents and Their Children Used Social Media and Technology at the Beginning of the COVID-19 Pandemic and Associations with Anxiety. CYBERPSYCHOLOGY,BEHAVIOR, AND SOCIAL NETWORKING, 00(00). doi: 10.1089/cyber.2020.0284 [DOI] [PubMed]
  14. Dunn L, & Dunn D (2007). PPVT-4 Manual Bloomington, MN: NCS Pearson. Inc.[Google Scholar]. [Google Scholar]
  15. Farah R, Greenwood P, Dudley J, Hutton J, Ammerman RT, Phelan K, … Horowitz-Kraus T (2020). Maternal depression is associated with altered functional connectivity between neural circuits related to visual, auditory, and cognitive processing during stories listening in preschoolers. Behavioral and Brain Functions, 16, 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ginsburg KR (2007). The importance of play in promoting healthy child development and maintaining strong parent-child bonds. Pediatrics, 119(1), 182–191. [DOI] [PubMed] [Google Scholar]
  17. Gioia GA, Isquith PK, Guy SC, & Kenworthy L (2000). Behavior rating inventory of executive function: BRIEF: Psychological Assessment Resources Odessa, FL. [Google Scholar]
  18. Ham T, Leff A, de Boissezon X, Joffe A, & Sharp DJ (2013). Cognitive control and the salience network: an investigation of error processing and effective connectivity. Journal of Neuroscience, 33(16), 7091–7098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Horowitz- Kraus T, & Hutton J . (2017). More screen time correlates with decreased functional connectivity in reading-related brain regions. Paper presented at the Human Brain Mapping Mapping Organization, Vancuver, Canada.
  20. Horowitz-Kraus T, DiFrancesco M, Greenwood P, Scott E, Vannest J, Hutton J, … Farah R (2020). Longer Screen Vs. Reading Time is Related to Greater Functional Connections Between the Salience Network and Executive Functions Regions in Children with Reading Difficulties Vs. Typical Readers. Child Psychiatry & Human Development, 1–12. [DOI] [PMC free article] [PubMed]
  21. Horowitz‐Kraus T, & Hutton JS (2018). Brain connectivity in children is increased by the time they spend reading books and decreased by the length of exposure to screen‐based media. Acta Paediatrica, 107(4), 685–693.0020https://www.softpedia.com/get/Science-CAD/G-Power.shtml. Retrieved from https://www.softpedia.com/get/Science-CAD/G-Power.shtml. [DOI] [PubMed] [Google Scholar]
  22. Huber B, Yeates M, Meyer D, Fleckhammer L, & Kaufman J (2018). The effects of screen media content on young children’s executive functioning. Journal of experimental child psychology, 170, 72–85. [DOI] [PubMed] [Google Scholar]
  23. Hutton JS, Dudley J, Horowitz-Kraus T, DeWitt T, & Holland SK (2019). Functional connectivity of attention, visual, and language networks during audio, illustrated, and animated stories in preschool-age children. Brain connectivity, 9(7), 580–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hutton JS, Dudley J, Horowitz-Kraus T, DeWitt T, & Holland SK (2020). Associations between screen-based media use and brain white matter integrity in preschool-aged children. JAMA pediatrics, 174(1), e193869–e193869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hutton JS, Huang G, Sahay RD, DeWitt T, & Ittenbach RF (2020). A novel, composite measure of screen-based media use in young children (ScreenQ) and associations with parenting practices and cognitive abilities. Pediatric Research, 1–8. [DOI] [PubMed]
  26. Hutton JS, Phelan K, Horowitz-Kraus T, Dudley J, Altaye M, DeWitt T, & Holland SK (2017). Story time turbocharger? Child engagement during shared reading and cerebellar activation and connectivity in preschool-age children listening to stories. PloS one, 12(5), e0177398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Johnson JG, Cohen P, Kasen S, & Brook JS (2007). Extensive television viewing and the development of attention and learning difficulties during adolescence. Archives of pediatrics & adolescent medicine, 161(5), 480–486. [DOI] [PubMed] [Google Scholar]
  28. Johnson JG, Cohen P, Smailes EM, Kasen S, & Brook JS (2002). Television viewing and aggressive behavior during adolescence and adulthood. Science, 295(5564), 2468–2471. [DOI] [PubMed] [Google Scholar]
  29. Keren‐Happuch E, Chen SHA, Ho MHR, & Desmond JE (2014). A meta‐analysis of cerebellar contributions to higher cognition from PET and fMRI studies. Human brain mapping, 35(2), 593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lillard AS, Li H, & Boguszewski K (2015). Television and children’s executive function. In Advances in child development and behavior (Vol. 48, pp. 219–248): Elsevier. [DOI] [PubMed] [Google Scholar]
  31. Llinás R, & Welsh JP (1993). On the cerebellum and motor learning. Current opinion in neurobiology, 3(6), 958–965. [DOI] [PubMed] [Google Scholar]
  32. Manly T, Robertson I, Anderson V, & Nimmo-Smith I (1999). The test of everyday attention for children (TEA-Ch). Bury St Edmunds, UK: Thames Valley Test Company. [DOI] [PubMed] [Google Scholar]
  33. McDaniel BT, & Coyne SM (2016). Technoference”: The interference of technology in couple relationships and implications for women’s personal and relational well-being. Psychology of Popular Media Culture, , 5(1), 85. [Google Scholar]
  34. Mendoza JA, Zimmerman FJ, & Christakis DA (2007). Television viewing, computer use, obesity, and adiposity in US preschool children. International Journal of Behavioral Nutrition and Physical Activity, 4(1), 44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Murray JP, Liotti M, Ingmundson PT, Mayberg HS, Pu Y, Zamarripa F, … Fox, P. T. (2006). Children’s brain activations while viewing televised violence revealed by fMRI. Media psychology, 8(1), 25–37. [Google Scholar]
  36. Newsham G, Drouin M, & McDaniel BT . (2020). Problematic Phone Use, Depression, and Technology Interference Among Mothers. Psychology of Popular Media, , 9(2), 117–124. 10.1037/ppm0000220. [DOI] [Google Scholar]
  37. Padilla‐Walker LM, Coyne SM, Booth MA, Domoff SE, Summers K, Schvaneveldt E, & Stockdale L (2020). Parent–child joint media engagement in infancy. Infancy, 25(5), 552–570. [DOI] [PubMed] [Google Scholar]
  38. Paulus MP, Squeglia LM, Bagot K, Jacobus J, Kuplicki R, Breslin FJ, … Bartsch, H. (2019). Screen media activity and brain structure in youth: evidence for diverse structural correlation networks from the ABCD study. Neuroimage, 185, 140–153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pediatrics A. A. o. (2013). Children, adolescents, and the media. Pediatrics, 132(5), 958–961. [DOI] [PubMed] [Google Scholar]
  40. Petersen SE, & Posner MI (2012). The attention system of the human brain: 20 years after. Annual review of neuroscience, 35, 73–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Power JD, Cohen AL, Nelson SM, Wig GS, Barnes KA, Church JA, … Schlaggar BL (2011). Functional network organization of the human brain. Neuron, 72(4), 665–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Radesky JS, & Christakis DA (2016). Increased screen time: implications for early childhood development and behavior. Pediatric Clinics, 63(5), 827–839. [DOI] [PubMed] [Google Scholar]
  43. Rideout V (2017). The Common Sense census: Media use by kids age zero to eight San Francisco, CA: Common Sense Media, 263–283. [Google Scholar]
  44. Romeo RR, Christodoulou JA, Halverson KK, Murtagh J, Cyr AB, Schimmel C, Chang P, Hook PE, Gabrieli JDE (2018). Socioeconomic Status and Reading Disability: Neuroanatomy and Plasticity in Response to Intervention. Cerebral Cortex 28(7), 2297–2312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Salomon G (1977). Effects of encouraging Israeli mothers to co-observe” Sesame Street” with their five-year-olds. Child development, 1146–1151.
  46. Shi T, Li X, Song J, Zhao N, Sun C, Xia W, … Tomoda, A. (2012). EEG characteristics and visual cognitive function of children with attention deficit hyperactivity disorder (ADHD). Brain and Development, 34(10), 806–811. [DOI] [PubMed] [Google Scholar]
  47. Twait E, Farah R, & Horowitz-Kraus T (2018). Decreased functional connectivity of the salience network during narrative comprehension in children with reading difficulties: An fMRI study. NeuroImage: Clinical, 20, 987–992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Vogel AC, Petersen SE, & Schlaggar BL (2014). The VWFA: it’s not just for words anymore. Front Hum Neurosci, 8, 88. doi: 10.3389/fnhum.2014.00088 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wechsler D (1999). Manual for the Wechsler abbreviated intelligence scale (WASI). San Antonio, TX: The Psychological Corporation. [Google Scholar]
  50. Woodcock RW, McGrew KS, & Mather N . (2007). Woodcock-Johnson Tests of Achievement, Three . In. Rolling Meadows, IL: Riverside Publishing. . [Google Scholar]
  51. Zimmerle JC (2019). Limiting Technoference: Healthy Screen Time Habits for New Parents. International Journal of Childbirth Education, 34(2), 54–59. [Google Scholar]
  52. Zimmerle JC (2019). Limiting Technoference: Healthy Screen Time Habits for New Parents. International Journal of Childbirth Education, 34(2). [Google Scholar]
  53. Zimmerman FJ, & Christakis DA (2007). Associations between content types of early media exposure and subsequent attentional problems. Pediatrics, 120(5), 986–992. [DOI] [PubMed] [Google Scholar]
  54. Zivan M, Bar S, Jing X, Hutton J, Farah R, & Horowitz-Kraus T (2019). Screen-exposure and altered brain activation related to attention in preschool children: An EEG study. Trends in Neuroscience and Education, 17, 100117. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data will be available upon a reasonable request.

RESOURCES