Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 14.
Published before final editing as: Biol Psychiatry Cogn Neurosci Neuroimaging. 2026 Mar 27:S2451-9022(26)00085-6. doi: 10.1016/j.bpsc.2026.03.012

The roles of delayed cortical maturation and lower anticipatory reward activation in predicting addictive screen use in youth

Katie J Paige 1, Mike Angstadt 1, Meghan E Martz 1, Luke W Hyde 2, Mary M Heitzeg 1, Chandra Sripada 1, Omid Kardan 1,*
PMCID: PMC13075549  NIHMSID: NIHMS2163765  PMID: 41905476

Abstract

Background:

Addictive screen use (ASU), above and beyond screen time, has been linked to significant mental health risks. Yet, little is known about the neural risk factors that may associate with ASU. We examined two neurodevelopmental factors—cognitive control and reward—highlighted in substance use research and their links to ASU.

Method:

We utilized resting-state and monetary incentive delay (MID) task fMRI data from the Adolescent Brain Cognitive Development (ABCD) Study® at baseline (Y0; ages 9–10) to predict addictive videogaming, addictive social media use, addictive phone use, and a composite measure of ASU at year two follow-up (Y2; ages 11–12). Cortical connectomic maturation was operationalized as distance from early-life and proximity to adult functional networks in an individual connectome to potentially index cognitive control development. This was supplemented by also assessing cognitive task performance. Nucleus Accumbens (NAc) activation in anticipation of reward in the MID task was used to assess reward processing.

Result:

Above and beyond total screen time and attention problems, lower connectomic maturation at Y0 associated with Y2 higher ASU composite and addictive videogaming. Analyses including task performance indicated cortical maturation was associated with both ASU and task performance, but we did not find cognitive task performance to be directly related to ASU. Additionally, lower NAc anticipatory reward activation at Y0 was very weakly associated with higher Y2 ASU.

Conclusion:

Delayed cortical network maturation and, to a lesser extent, lower anticipatory reward activation in 9-to-10-year-olds may be associated with addictive screen use in early adolescence, above-and-beyond parent-reported attention problems.

Keywords: Addictive videogaming, addictive phone use, addictive social media use, neurocognitive maturation, reward processing, adolescence

Introduction

Widespread engagement in screen-based activities among adolescents—including social media use, videogaming, and mobile phone use—has raised concerns about its potential psychological harms [1]. Emerging evidence suggests that addictive screen use behavior may be an important risk factor for poor youth mental health [2–5]. Addictive screen use refers to problematic patterns of screen use that include elements of addiction, such as mood modification, tolerance, withdrawal, conflict, relapse, and functional impairment [6–8]. Addictive screen use has been linked to significant mental health risks, including internalizing and externalizing symptoms, increased risk for alcohol use initiation, eating disorder symptoms, and suicidal behaviors [2,9,10]; yet, little is known about the neural factors that may be linked to these problematic patterns of screen use. The current study aims to elucidate associations between two potential neural risk factors – delayed neurocognitive maturation and blunted reward processing – and addictive screen use behaviors in a large population-based sample of adolescents. Further, research on neurodevelopmental risk factors for youth addictive screen use is in its early stages [11], and we extend past research in this area by addressing its three main limitations.

First, the majority of prior studies have examined screen time, rather than addictive screen use behaviors [11], when examining links with brain phenotypes. This is a critical limitation because emerging evidence suggests that addictive screen use is more strongly associated with risk for poor psychological adjustment [2–5] than screen time. Second, prior neuroimaging studies have been limited by small sample sizes. Indeed, a recent literature review [11] found 16 studies, 14 of which used sample sizes smaller than N=62. The only large study (N=2,532) investigated the relationship between screen time and youth white matter microstructure and reported no significant associations [12]. More recently, a study using Adolescent Brain Cognitive Development (ABCD) Study data found that lower activation in the bilateral caudate during reward anticipation was associated with greater addictive videogaming behaviors across time [13]. However, this study did not include addictive screen use other than videogaming (e.g., social media and phone use), did not assess neural measures outside of reward processing pathways, and did not account for total screen time. Third, previous studies in this area have primarily focused on the impact of screen use on neurocognitive development (e.g., does screen time in preschoolers predict subsequent effortful control?) [14]. While the impact of screens on brain development is of great public interest [15], there is compelling empirical evidence from studies employing causal inference methodologies, like biometric modeling of polygenic risk scores, that prolonged screen time is likely an early sign of neurodevelopmental deficits rather than a causal factor [16,17]. Therefore, it is important to identify neural factors that may predispose addictive screen use. Yet, this has rarely been done using longitudinal data.

Preliminary evidence suggests that addictive screen use may be associated with deficits in cognitive control and reward processing systems [18–22]. Additional support for the plausibility of these two neural risk factors comes from substance use neuroimaging research, wherein inhibitory control and reward neural pathways have been implicated in the development of adolescent substance use (for a review, see [23]).

In the cognitive control domain, research has shown that adolescents who engage in high levels of screen time have reduced resting-state connectivity in subcortical, frontal, and parietal areas [18,19,21] (see [11] for a review). While these regions are involved in attentional and control networks, no relationship between these reduced connectivity patterns and behavioral task performance has been established. Therefore, the validity of these reduced connectivity patterns as measures of deficits in control circuitry is unknown. Indeed, multiple large-scale brain networks are involved in cognitive control and its development [24,25]; thus, to advance the field, multi-network measures of neurocognitive development with established behavioral significance should be utilized [26].

In the reward processing domain, altered activation in the ventral striatum and specifically Nucleus Accumbens (NAc), has been widely implicated in addictive behaviors, including Substance Use Disorders [23], Gambling Disorder [27], and Internet Gaming Disorder [28]. For example, neuroimaging studies have found that social media “likes”, which symbolize social approval and connection, act on reward system circuitry [29,30]. Heightened sensitivity to reward, particularly in the context of peer evaluation [31,32], is a hallmark of adolescence [33]. In addition to receiving positive feedback from peers on social media, various screen time activities (e.g., scoring points in a videogame, scrolling online, etc.) provide immediate rewards. Therefore, adolescents who demonstrate deficits in reward processing (e.g. deviation from the typical developmental reactivity of NAc) may be especially vulnerable to engaging in addictive screen use behaviors compared to their peers. However, both substance use and excessive screen time research have produced mixed findings regarding the direction of the alteration in NAc activation when anticipating reward. Specifically, while some studies report lower NAc activation in adolescents and young adults with substance use [34,35] or frequent screen use [36], other studies have found higher NAc activation or no association when anticipating reward (see [37,38] for substance use; see [22] for videogaming). Despite mixed findings, evidence appears to more consistently support an association between adolescent addictive behaviors and NAc hypoactivation, rather than hyperactivation, during reward anticipation.

The current study uses a longitudinal neuroimaging design, a large population-based sample of adolescents (Ns > 3000), and controls for overall screen time to examine associations between neural factors at 9–10 and addictive screen use at 11–12 years of age. We hypothesize that delayed neurocognitive maturation, indicated by lower maturity of cortical brain networks, will be associated with higher levels of addictive screen use. Additionally, we hypothesize that blunted reward processing (i.e., lower activation in the NAc when anticipating reward) will be associated with higher levels of addictive screen use.

Methods

Participants

The Adolescent Brain Cognitive Development (ABCD) Study® is an ongoing longitudinal study that recruited 11,867 children across 21 sites in the US [39]. We utilized data from ABCD Study® Release 5.1 including baseline (Y0; ages 9–10) neuroimaging and two-year follow-up (Y2; ages 11–12) addictive screen use survey data. A total of 7396 and 9166 participants passed image quality control criteria for resting-state fMRI and Monetary Incentive Delay (MID) task fMRI, respectively (see Supplementary section 1 for exclusion criteria for rest and task fMRI). Addictive videogaming, addictive social media use, and addictive phone use questionnaires were administered only to participants who endorsed playing any video games (N = 8101; 68%), using any social media (N = 6023; 51%), or having a mobile phone (N = 7810; 66%), respectively. Therefore, analyses predicting each measure included different sample sizes based on availability of both ASU questionnaire and fMRI. Additional exclusions were applied based on missing covariates. Demographic characteristics are reported in Table 1 for resting-state and task fMRI in each screen use domain as well as the cross-domain ASU (maximum z-scored across the three domains). Distributions of these ASU values are shown in the right margins of Figure 1). Means and SDs of these distributions and the number of participants endorsing at least one item on each questionnaire are reported in Table 1.

Table 1.

Size, demographic characteristics, and basic addictive screen use statistics of the samples in the analyses.

Rest: Max z-scored ASU (N = 5683) Rest: Addictive Videogaming (N = 4522) Rest: Addictive Social Media Use (N = 3330) Rest: Addictive Phone Use (N = 4466)
Male 2905 (51.1%) 2629 (58.1%) 1489 (44.7%) 2102 (47.0%)
Female 2778 (49.1%) 1893 (41.9%) 1841 (55.3%) 2364 (53.0%)
White 3365 (59.2%) 2647 (58.5%) 1758 (52.8%) 2556 (57.2%)
Black 621 (10.9%) 522 (11.5%) 468 (14.1%) 539 (12.1%)
Hispanic 1038 (18.3%) 827 (18.3%) 699 (20.1%) 857 (19.2%)
Asian 96 (1.7%) 71 (1.6%) 48 (1.4%) 69 (1.5%)
Other 563 (10.0%) 455 (10.0%) 357 (10.7%) 445 (10.0%)
Endorse>0 N/A 3570 (79%) 2557 (77%) 4332 (97%)
Mean (SD) N/A 6.03 (6.06) 4.95 (5.22) 16.24 (8.70)
MID: Max z-scored ASU (N = 7209) MID: Addictive Videogaming (N = 5729) MID: Addictive Social Media Use (N = 4155) MID: Addictive Phone Use (N = 5532)
Male 3778 (52.4%) 3428 (59.8%) 1932 (46.5%) 2694 (48.7%)
Female 3431 (47.6%) 2301 (40.2%) 2223 (53.5%) 2838 (51.3%)
White 4050 (56.2%) 3173 (55.4%) 2072 (49.9%) 3012 (54.4%)
Black 837 (11.6%) 698 (12.2%) 632 (15.2%) 707 (12.8%)
Hispanic 1442 (20.0%) 1152 (20.1%) 930 (22.4%) 1139 (20.6%)
Asian 130 (1.8%) 97 (1.7%) 63 (1.5%) 99 (1.9%)
Other 750 (10.4%) 609 (10.6%) 458 (11%) 575 (10.4%)
Endorse>0 N/A 4587 (80%) 3171 (76%) 5376 (97%)
Mean (SD) N/A 6.22 (6.13) 5.03 (5.31) 16.43 (8.73)
Open in a new tab

Figure 1.

Figure 1.

Open in a new tab

Associations of Anchored Functional Connectivity (AFC) maturation score at Y0 with Y2 addictive screen use (max ASU), as well as addictive videogaming, addictive social media use, and addictive phone use across youth. βM1 and βM2 show standardized coefficients for AFC from mixed-effects regressions Model1 and Model2 (see Methods: covariates). All p-values are based on permutation tests.

fMRI data

Resting-state fMRI data were acquired across four separate runs (~5 min per run) (details are described in Hagler et al., 2019 [40]). The entire data pipeline was run through automated scripts on described in detail in [41]. Monetary incentive delay task [42] fMRI data were acquired in two runs (~6 min per run) (details are described in [39,40]). More details for processing steps of the rest and task fMRI data are in Supplementary section 1.

Neuroimaging measures

Resting-state and MID task fMRI data from baseline year (Y0; ages 9–10) were used to measure cortical functional maturation and NAc anticipatory reward activation, respectively.

Anchored Functional Connectomic maturation score (AFC):

Following [26], resting-state fMRI data were spatially down-sampled to 333 parcels in the Gordon parcellation for cortical regions [43]. Pearson correlations between activities in all pairs of these parcels results in a connectome per individual. Anchored functional connectomic maturation score (AFC) was calculated as the difference in the fit of the connectome to early life canonical networks [44] versus adult canonical networks [43]:

AFC=[Connwithin-Connbetween]Adult-Connwithin-ConnbetweenBaby,

Where, [Connwithin – Connbetween]) was calculated as the mean of the transformed signed correlation values between pairs of parcels that were inside networks (Connwithin) and those outside networks (Connbetween); and the superscripts Adult and Baby respectively refer to the networks defined in adult participants [43] and networks defined in 8–26-months-old participants [44]. These network assignments are shown in Supplementary section 2 (Figures S1 and S2).

Briefly, AFC estimates the neoteny/maturation status of the individual youth’s cortical networks by assessing the distance of the youth’s intrinsic cortical connectivity patterns from those expected (i.e., on average) in early-life versus adulthood. Lower AFC indicates a delay-like pattern in cortical maturation and has been associated with lower cognitive task performance both cross-sectionally and longitudinally from pre-to early adolesce [26]. Examples and more details about AFC are provided in Supplementary section 3. More details including validation of this measure and its relationship with performance in cognitive control tasks across and within youth can be found in [26].

ROI-specific activation in anticipation of reward:

We used the contrast beta weights for anticipation of large reward versus neutral trials of MID task in the Nucleus Accumbens Freesurfer ASEG segmentation in both right and left hemispheres for our main analyses. Additionally, in posthoc analyses, we used the beta weights for the same contrast but in the Caudate and Amgdala using Freesurfer segmentation and Insula based on the Desikan-Killiany cortical atlas (see Supplementary section 1).

Behavioral measures

Addictive screen use:

Addictive screen use questionnaires1 were not administered at the baseline assessment in ABCD and were first administered at Y2. Thus, in the current study, addictive screen use questionnaires from Y2 (ages 11–12) were utilized. For addictive videogaming, addictive social media use, and addictive phone use, summed scores across items were very highly correlated with scores derived from confirmatory factor analyses (see [3]; all rs > .99). Therefore, we used the summed scores for each domain (see Supplementary section 4 for more details). Additionally, we made a cross-domain ASU (max-ASU), which was the highest of the three questionnaires sums (each z-scored) to represent the participant’s overall propensity to addictive screen use on whatever screen they may have (most) access to.

Covariates & other variables:

The basic covariates (Model 1) included data collection site and familyID as random intercepts and scanner type, head motion (mean framewise displacement), amount of fMRI data (number of runs), parental history of drug use (problematic use binary yes/no), parental history of alcohol use (problematic use binary yes/no), highest parental education, household income, participant sex (binarized), age, and race/ethnicity (as a proxy for experiences of minorities in the US) as covariates. We also tested a second regression model (Model 2) with additional covariates of total screen time (self-reported total daily screen use for weekdays and weekends), attention problems (based on Child Behavior Checklist attention problems syndrome scale; ABCD variable: cbcl_scr_syn_attention_r) and general factor of psychopathology (using CBCL identical to the p-factor from [45]; see Supplementary section 5) at baseline.

Cognitive task performance:

Performance in tasks that tax cognitive control beyond only attention, including N-back and NIH-Toolbox tasks, were utilized for our statistical mediation analysis. This analysis is to assess whether cortical maturation is a cognitive control neural pathway to ASU or is independently related to it. Overall cognitive performance was quantified as average performance in the N-back task, and seven NIH-Toolbox tasks: Picture vocabulary, Flanker inhibitory control and attention, Pattern Comparison processing speed, Picture sequence memory, Oral reading recognition task, List sorting, and Dimensional change card sorting tasks. More details about these cognitive tasks can be found in [46,47], and Supplementary section 6.

Statistical analyses

Main analyses:

In the cognitive pathway study, we regressed the three addictive screen use measures separately on the AFC scores derived from functional connectivity data during rest at baseline. The lmer function from the lme4 package in R was used to conduct the mixed-effects regressions with site and familyID as random intercepts and scanner type and all other covariates as fixed effects. Due to non-normal distributions of addictive videogaming and addictive social media use values (and for consistency, also for addictive phone use), p-values for the regression coefficients were calculated based on permutations using null distributions generated from 5000 shuffled values for the dependent variable. To understand the specificity of the AFC’s relationship with each of the addictive screen use measures as opposed to general risk for poor mental health and screen use, the regressions were run with and without total screen time, attention problems, and general factor of psychopathology as covariates. Post hoc, a negative binomial regression using glmer.nb function was fit to separately estimate the effect size for addictive videogaming as count data, adjusted for all covariates. Finally, indirect-effects between cognitive task performance, AFC, and addictive screen use were estimated using 1000 bootstrapped mixed-effects regressions.

Mixed-effects regressions were also conducted in the reward pathway study, with NAc anticipatory activation as the main predictor instead of AFC. The permutation p-values in both studies were Holm-Bonferroni corrected for running 3 models after the cross-domain ASU model (3 domain-specific screen use measures).

Post-hoc analyses:

Due to the strong association between sex and addictive screen use measures, we repeated the regressions above but instead of having sex as a covariate, the samples were split based on the binarized sex.

Results

Addictive screen use and cortical connectomic maturation

The distributions of cross-domain and specific addictive screen use are shown on the right margins of Figure 1 (means and SD are reported in Table 1). Across all analyses, >76% of the surveyed participants endorsed at least one addictive questionnaire item (see Table 1).

We tested whether patterns of delay in cortical connectomic maturation (lower AFC scores) at Y0 predicted more ASU at Y2. The distributions of AFC scores are shown on the top margins of Figure 1 (AFC has Mean = .24, SD = .10 with differences within rounding error based on the analysis sub-sample). The relationships between ASU and AFC scores are represented in the scatterplots with fitted lines. We found a negative association between AFC at Y0 and the cross-domains addictive screen use at Y2 (N = 5683; βM1 = −.056, CI = [−.083, −.027], p < .001), as well as the more specific addictive videogaming at Y2 (N = 4522; βM1 = −.060, CI = [−.087, −.031], p < .001). These associations were adjusted for data collection site, familyID, scanner, head motion, sex, age, race/ethnicity, family income, parental education, and family history of alcohol and drug use. Additionally, these associations persisted above and beyond total screen time, general factor of psychopathology, and attention problems (βM2 = −.051, CI = [−.077, −.026] for max-ASU; and βM2 = −.059, CI = [−.087, −.032] for videogaming, ps < .001), showing specific contribution to predicting ASU rather than general screen time or attention dysregulation. We also found a negative association between AFC at Y0 and addictive phone use at Y2 (N = 4466; βM1 = −.039, CI = [−.069, −.008], p = .035), but this association did not persist above and beyond the participant’s total screen time and attention problems (Figure 1; βM2 = −.035, p = .072). Posthoc analyses separated by sex showed the relationship between AFC and addictive phone use to be significant in the full covariates model for only female youth (See Supplementary Section 7 and Figure S4). Addictive social media use did not show a specific relationship with AFC. All reported p-values are based on permutations due to skewed distributions of ASU domains. Negative binomial mixed-effects regression additionally confirmed the significant results for addictive videogaming as a count variable (βM2-logCount = −.028; odds-ratio = .97, CI = [.95, .99]; z = −4.07; p < .001).

Is delayed cortical maturation a cognitive control pathway to ASU?

Next, we ran indirect-effects regression models to assess the relationship between AFC, cognitive task performance (N-back and NIH-toolbox tasks), and ASU (both max-ASU and the addictive videogaming specifically). There was a consistent relationship between cognitive task performance and AFC (‘a’ paths in Figure 2) as well between AFC and ASU/addictive videogaming (‘b’ paths). This resulted in a modest but statistically significant indirect effect-only mediation between cognitive task performance and ASU through AFC (‘a*b’ in Figure 2). Surprisingly, however, there was no direct relationship between cognitive task performance and ASU/addictive videogaming (‘c’ path in Figure 2). To further investigate, we checked if any of the seven NIH-toolbox tasks or N-back at baseline were separately related to any of the ASU domains adjusted for the covariates. The results showed no relationships between performance scores in any of the tasks and ASU after adjusting for covariates, except maybe for the Picture Vocabulary task (β = −.031, p = .032, not significant after correction for multiple comparisons). Together, these analyses suggested that our AFC findings cannot be interpreted as a cognitive control neurodevelopmental pathway to ASU, rather this is a delayed cortical maturation pathway that is related independently to both cognitive task performance and ASU/addictive videogaming beyond family resources and youth attention problems.

Figure 2.

Figure 2.

Open in a new tab

Indirect-effect models showing statistical non-causal mediation such that AFC score is associated with cognitive task performance and max addictive screen use (top) or addictive videogaming (bottom). Supplementary section 8 (Figure S5) shows the version of these models with cognitive task performance as mediator and AFC as predictor.

Addictive screen use and anticipatory reward activation

Next, we examined the association between reward response when anticipating large reward contrasted with the neutral trials in the MID task at Y0 and addictive screen use at Y2. There was a modest relationship between lower NAc activation and more cross-domain ASU (N = 7209; βM2 = −.023, CI = [−.045, −.000], p = .029). However, as shown in Figure 3, this association was not statistically significant in the more limited samples with specific domains of ASU (addictive video gaming, addictive social media use or addictive phone use). These results did not change based on participants’ sex (see Supplementary section 7). Therefore, we found weak evidence supporting our hypothesis that blunted reward processing (lower NAc activation) is associated with ASU.

Figure 3.

Figure 3.

Open in a new tab

Associations of anticipatory reward activation in Nucleus Accumbens (NAc) with addictive cross-domain and specific domains of ASU. βM1 and βM2 show standardized coefficients for NAc activation from mixed-effects regressions Model1 and Model2 (see Methods: covariates). All p-values are based on permutation tests.

After these initial analyses, we conducted three post-hoc analyses with the caudate, amygdala, and insula anticipatory reward activations instead of NAc. Similar to NAc, after adjusting for covariates, we did not find an association between addictive use in any of the screen use domains and activation in the caudate, insula, or amygdala when anticipating reward. These results are shown in Supplementary section 9 (Figure S6).

Discussion

The present study examined neural correlates for addictive screen use in adolescents. We found associations between lower cortical connectomic maturation and higher levels of addictive screen use. This association was strongest for addictive videogaming, specifically. Results suggested weak evidence for an association between delayed cortical maturation and higher levels of addictive phone use; however, this association was driven by female adolescents. We also found modest evidence for a reward processing risk pathway, indicated by lower activation in the NAc when anticipating reward, to addictive screen use. Results add substantially to existing knowledge about potential neurodevelopmental risk factors for addictive screen use. Strengths of the current study include using large (Ns ≥ 3330), diverse, longitudinal neuroimaging data, considering multiple potential mechanisms of neurodevelopmental risk, adjusting for total screen time, general psychopathology, and parent-reported attention problems, and utilizing both cross-platform and platform-specific assessments of addictive screen use.

Delayed cortical maturation, indicated by low AFC, was associated with more addictive videogaming and addictive phone use, above and beyond parent-reported attention problems. However, analyses including performance scores across tasks of executive function demonstrated that AFC’s relationship with cognitive performance was independent of its association with addictive screen use. Specifically, cognitive task performance did not significantly associate with addictive screen use. Therefore, our AFC findings should not be interpreted as a cognitive control neurodevelopmental pathway to addictive screen use; rather, this may be better understood as a delayed cortical maturation pathway that is related independently to both cognitive task performance and addictive videogaming. Moreover, the modest indirect-only association between cognitive task performance and ASU through AFC is very small and should be interpreted cautiously. Significant neural findings are consistent with a recent review which found support for reduced connectivity among frontal and parietal areas involved in attentional and control networks and high levels of screen time in adolescents [11]. We extend this work by providing evidence that low connectomic cortical maturation, specifically, predicts addictive screen use behaviors, which have been more closely linked to poor mental health including suicidal behaviors [2], than total screen time. Importantly, despite adjusting for total screen time and general psychopathology/attention problems, we were unable to account for baseline levels of addictive videogaming and phone use when testing associations with AFC, as ABCD Study did not administer addictive screen use measures at their baseline assessment. Therefore, the directionality of associations remains unknown. Additionally, despite its unique contribution to addictive screen use over multiple family, cognitive, and psychopathology factors, delay-like patterns in functional maturation of cortical networks should be further assessed for clinical relevance as AFC’s strength of association with addictive screen use was small (e.g., based on negative binomial regression, about 3% less likely to endorse an additional addictive videogaming survey item for one SD increase in AFC).

There was weak evidence for our hypothesis that blunted reward processing would be associated with higher levels of addictive screen use. While there was a significant association between blunted reward processing at baseline and the addictive screen use composite at Y2, the effect was very small. Further, the associations between blunted reward processing and specific domains of addictive videogaming, social media, and mobile phone use were not significant. These null findings are inconsistent with previous research which has found links between altered reward processing and elevated screen time [11,13]. One possible reason our results diverge is the components of our measure of blunted reward processing, including the task and the brain region investigated. MID might not reflect the same quality of reward that youth with problematic screen use patterns may be hypo/hyper sensitive to. Furthermore, while many previous studies have utilized low NAc activation to index blunted reward processing [23,29], some research highlights the roles of other regions, including the orbitofrontal cortex [28], dACC, insula, and amygdala [11], in reward processing and its association to problematic screen use behaviors. On the other hand, we examined bilateral caudate, amygdala, and insula activations of reward anticipation in post-hoc analyses and results were also not significant. Our analyses have multiple differences with [13], which reported lower anticipatory reward activation in caudate measured at Y2 being related to addictive videogaming in the same sample. First, we used baseline MID fMRI data rather than Y2 which temporally separates the brain and behavioral measures. Second, we did not introduce multiple brain regions in the same regression, as anticipatory activations across brain regions are highly correlated and, for example, caudate activation adjusted for that of amygdala, NAc, insula, etc. would not be easily interpretable in the context of our hypotheses. Third, we included baseline total screen time, general psychopathology and parent-reported attention problems, as covariates to assess the contribution of altered reward processing specifically to addictive screen use. Still, identifying neural factors that increase risk for addictive screen use behaviors is a relatively new area of research, and future studies may aim to utilize adolescent samples that span a wider age range, as our study may not have captured the peak of reward sensitivity development.

In addition to the small effect sizes, results from the current study should be interpreted within the context of certain limitations. First, we did not examine other aspects of reward processing, such as connectivity across striatal and salience regions, as well as the frontal cortex. While some of these connectivity measures tend to be correlated with the anticipatory activations that were included in the current study, future work may aim to include more comprehensive neural measures. Second, while we used longitudinal data to test study aims, addictive screen use behaviors were not assessed at the baseline assessment in the ABCD study. Therefore, despite controlling for total screen time, general factor of psychopathology, and attention problems in youth, we could not control for prior levels of our specific dependent variables, limiting conclusions about the directionality of associations, temporal ordering, and causality. Future research should use repeated longitudinal measures and causal inference methodologies to tease apart the directionality and improve understanding of the causal nature of associations between neural deficits and addictive screen use. Past work suggests that genetic factors account for some shared risk [16,17,48]; however, all previous studies in this area which used causal inference methods to examine neural risk examined screen time as the outcome of interest, rather than addictive screen use. Given emerging evidence that addictive screen use is more strongly linked to poor psychosocial adjustment [2,9,10], future studies that use repeated longitudinal measures and causal inference methods to elucidate directionality of associations with neural pathways should examine addictive screen use, in addition to total screen time.

In conclusion, the current study offers insights that are especially relevant amidst significant increases in adolescent screen time behaviors in recent years [49,50]. Delay-like patterns in normative functional brain maturation may provide unique correlational predictive power for addictive screen use, especially addictive videogaming behaviors in general and addictive phone use in girls. Parents may consider monitoring their children’s screen use activities for addictive patterns of use (e.g., mood changes after stopping use, failing to meet schoolwork responsibilities due to screen use), even for children who may not display difficulties in performing tasks that demand cognitive control. The American Academy of Pediatrics recommends that parents develop a family media use plan, which may involve regular communication with adolescents about online activities and distinguishing healthy from addictive patterns of screen use behaviors [51,52]. Future research using repeated longitudinal measures and causal inference methodologies is needed to improve understanding of the neural cognitive control and reward causes and consequences of addictive screen use.

Supplementary Material

Supplementary Material

Acknowledgements

OK was supported by K01DA059598. KJP was supported by T32AA007477.

ABCD acknowledgements

The ABCD data used in this report came from NIMH Data Archive Digital Object Identifier 10.15154/8873-zj65. The ABCD Study is supported by the National Institutes of Health and additional federal partners under award numbers U01DA041022, U01DA041028, U01DA041048, U01DA041089, U01DA041106, U01DA041117, U01DA041120, U01DA041134, U01DA041148, U01DA041156, U01DA041174, U24DA041123, and U24DA041147.

Footnotes

Financial Disclosures

All authors report no biomedical financial interests or potential conflicts of interest.

1

We followed the terminology used by the ABCD Study questionnaires, but “problematic screen use pattern” may be a preferred phrase over “addictive screen use”.

Data and Code availability

Scripts to generate AFC scores from fMRI data are shared in our previous work https://doi.org/10.1016/j.dcn.2025.101543. Scripts to run the regression analyses and generate the figures in this study are shared at https://github.com/okardan/ASU_neurodev_risk. Data tables including behavioral and task fMRI measures used in the scripts can be downloaded from the ABCD Study website with an approved ABCD data use agreement.

References

  • 1.Domingues-Montanari S. Clinical and psychological effects of excessive screen time on children. J Paediatr Child Health. 2017;53(4):333–338. doi: 10.1111/jpc.13462 [DOI] [PubMed] [Google Scholar]
  • 2.Xiao Y, Meng Y, Brown TT, Keyes KM, Mann JJ. Addictive Screen Use Trajectories and Suicidal Behaviors, Suicidal Ideation, and Mental Health in US Youths. JAMA. Published online June 18, 2025. doi: 10.1001/jama.2025.7829 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Burén J, Nutley SB, Thorell LB. Screen time and addictive use of gaming and social media in relation to health outcomes. Front Psychol. 2023;14. doi: 10.3389/fpsyg.2023.1258784 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Dorrestein M, Nutley SB, Thorell LB. Screen Time, Addictive Use of Social Media, Motives for Social Media Use and Social Media Content: Interrelations and Associations with Psychosocial Problems. Int J Ment Health Addict. Published online April 28, 2025. doi: 10.1007/s11469-025-01491-5 [DOI] [Google Scholar]
  • 5.Howard SJ, Hayes N, Mallawaarachchi S, et al. A meta-analysis of self-regulation and digital recreation from birth to adolescence. Comput Hum Behav. 2025;163:108472. doi: 10.1016/j.chb.2024.108472 [DOI] [Google Scholar]
  • 6.Bagot K, Tomko R, Marshall AT, et al. Youth screen use in the ABCD® study. Dev Cogn Neurosci. 2022;57:101150. doi: 10.1016/j.dcn.2022.101150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Nagata JM, Singh G, Sajjad OM, et al. Social epidemiology of early adolescent problematic screen use in the United States. Pediatr Res. 2022;92(5):1443–1449. doi: 10.1038/s41390-022-02176-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Andreassen CS, Torsheim T, Brunborg GS, Pallesen S. Development of a Facebook Addiction Scale. Psychol Rep. 2012;110(2):501–517. doi: 10.2466/02.09.18.pr0.110.2.501-517 [DOI] [PubMed] [Google Scholar]
  • 9.Nagata JM, Smith N, Zamora G, et al. Problematic social media use and alcohol expectancies in early adolescents. BMC Public Health. 2023;23(1). doi: 10.1186/s12889-023-15298-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chu J, Ganson KT, Testa A, et al. Screen time, problematic screen use, and eating disorder symptoms among early adolescents: findings from the Adolescent Brain Cognitive Development (ABCD) Study. Eat Weight Disord - Stud Anorex Bulim Obes. 2024;29(1). doi: 10.1007/s40519-024-01685-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Marciano L, Camerini AL, Morese R. The Developing Brain in the Digital Era: A Scoping Review of Structural and Functional Correlates of Screen Time in Adolescence. Front Psychol. 2021;12. doi: 10.3389/fpsyg.2021.671817 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Rodriguez-Ayllon M, Derks IPM, Van Den Dries MA, et al. Associations of physical activity and screen time with white matter microstructure in children from the general population. NeuroImage. 2020;205:116258. doi: 10.1016/j.neuroimage.2019.116258 [DOI] [PubMed] [Google Scholar]
  • 13.Lopez DA, Foxe JJ, Van Wijngaarden E, Thompson WK, Freedman EG. The longitudinal association between reward processing and symptoms of video game addiction in the Adolescent Brain Cognitive Development Study. J Behav Addict. 2024;13(4):1051–1063. doi: 10.1556/2006.2024.00068 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Guellai B, Somogyi E, Esseily R, Chopin A. Effects of screen exposure on young children’s cognitive development: A review. Front Psychol. 2022;13. doi: 10.3389/fpsyg.2022.923370 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.COUNCIL ON COMMUNICATIONS AND MEDIA, Hill D, Ameenuddin N, et al. Media and Young Minds. Pediatrics. 2016;138(5). doi: 10.1542/peds.2016-2591 [DOI] [PubMed] [Google Scholar]
  • 16.Takahashi N, Tsuchiya KJ, Okumura A, et al. The association between screen time and genetic risks for neurodevelopmental disorders in children. Psychiatry Res. 2023;327:115395. doi: 10.1016/j.psychres.2023.115395 [DOI] [PubMed] [Google Scholar]
  • 17.Zhang Y, Choi KW, Delaney SW, Ge T, Pingault JB, Tiemeier H. Shared Genetic Risk in the Association of Screen Time With Psychiatric Problems in Children. JAMA Netw Open. 2023;6(11):e2341502. doi: 10.1001/jamanetworkopen.2023.41502 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Wang L, Shen H, Lei Y, et al. Altered default mode, fronto-parietal and salience networks in adolescents with Internet addiction. Addict Behav. 2017;70:1–6. doi: 10.1016/j.addbeh.2017.01.021 [DOI] [PubMed] [Google Scholar]
  • 19.Chun JW, Choi J, Cho H, et al. Role of Frontostriatal Connectivity in Adolescents With Excessive Smartphone Use. Front Psychiatry. 2018;9. doi: 10.3389/fpsyt.2018.00437 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Sherman LE, Payton AA, Hernandez LM, Greenfield PM, Dapretto M. The Power of the Like in Adolescence: Effects of Peer Influence on Neural and Behavioral Responses to Social Media. Psychol Sci. 2016;27(7):1027–1035. doi: 10.1177/0956797616645673 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Hong SB, Zalesky A, Cocchi L, et al. Decreased Functional Brain Connectivity in Adolescents with Internet Addiction. Zuo XN, ed. PLoS ONE. 2013;8(2):e57831. doi: 10.1371/journal.pone.0057831 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Palaus M, Marron EM, Viejo-Sobera R, Redolar-Ripoll D. Neural Basis of Video Gaming: A Systematic Review. Front Hum Neurosci. 2017;11:248. doi: 10.3389/fnhum.2017.00248 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Heitzeg MM, Cope LM, Martz ME, Hardee JE. Neuroimaging Risk Markers for Substance Abuse: Recent Findings on Inhibitory Control and Reward System Functioning. Curr Addict Rep. 2015;2(2):91–103. doi: 10.1007/s40429-015-0048-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Dwyer DB, Harrison BJ, Yücel M, et al. Large-Scale Brain Network Dynamics Supporting Adolescent Cognitive Control. J Neurosci. 2014;34(42):14096–14107. doi: 10.1523/JNEUROSCI.1634-14.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Kardan O, Stier AJ, Cardenas-Iniguez C, et al. Differences in the functional brain architecture of sustained attention and working memory in youth and adults. Cohen Kadosh R, ed. PLOS Biol. 2022;20(12):e3001938. doi: 10.1371/journal.pbio.3001938 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kardan O, Jones N, Wheelock MD, et al. Assessing neurocognitive maturation in early adolescence based on baby and adult functional brain landscapes. Dev Cogn Neurosci. 2025;73:101543. doi: 10.1016/j.dcn.2025.101543 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Wolfschlag M, Håkansson A. Drug-Induced Gambling Disorder: Epidemiology, Neurobiology, and Management. Pharm Med. 2023;37(1):37–52. doi: 10.1007/s40290-022-00453-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Park B, Han DH, Roh S. Neurobiological findings related to Internet use disorders. Psychiatry Clin Neurosci. 2017;71(7):467–478. doi: 10.1111/pcn.12422 [DOI] [PubMed] [Google Scholar]
  • 29.Sherman LE, Greenfield PM, Hernandez LM, Dapretto M. Peer Influence Via Instagram: Effects on Brain and Behavior in Adolescence and Young Adulthood. Child Dev. 2018;89(1):37–47. doi: 10.1111/cdev.12838 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Sherman LE, Hernandez LM, Greenfield PM, Dapretto M. What the brain ‘Likes’: neural correlates of providing feedback on social media. Soc Cogn Affect Neurosci. 2018;13(7):699–707. doi: 10.1093/scan/nsy051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Spear LP. Effects of adolescent alcohol consumption on the brain and behaviour. Nat Rev Neurosci. 2018;19(4):197–214. doi: 10.1038/nrn.2018.10 [DOI] [PubMed] [Google Scholar]
  • 32.Albert D, Chein J, Steinberg L. The Teenage Brain: Peer Influences on Adolescent Decision Making. Curr Dir Psychol Sci. 2013;22(2):114–120. doi: 10.1177/0963721412471347 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Foulkes L, & Blakemore SJ (2016). Is there heightened sensitivity to social reward in adolescence?. Current opinion in neurobiology, 40, 81–85. [DOI] [PubMed] [Google Scholar]
  • 34.Peters J, Bromberg U, Schneider S, et al. Lower Ventral Striatal Activation During Reward Anticipation in Adolescent Smokers. Am J Psychiatry. 2011;168(5):540–549. doi: 10.1176/appi.ajp.2010.10071024 [DOI] [PubMed] [Google Scholar]
  • 35.Martz ME, Trucco EM, Cope LM, et al. Association of Marijuana Use With Blunted Nucleus Accumbens Response to Reward Anticipation. JAMA Psychiatry. 2016;73(8):838. doi: 10.1001/jamapsychiatry.2016.1161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hahn T, Notebaert KH, Dresler T, Kowarsch L, Reif A, Fallgatter AJ. Linking Online Gaming and Addictive Behavior: Converging Evidence for a General Reward Deficiency in Frequent Online Gamers. Front Behav Neurosci. 2014;8. doi: 10.3389/fnbeh.2014.00385 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Karoly HC, Bryan AD, Weiland BJ, Mayer A, Dodd A, Feldstein Ewing SW. Does incentive-elicited nucleus accumbens activation differ by substance of abuse? An examination with adolescents. Dev Cogn Neurosci. 2015;16:5–15. doi: 10.1016/j.dcn.2015.05.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Skumlien M, Mokrysz C, Freeman TP, et al. Neural responses to reward anticipation and feedback in adult and adolescent cannabis users and controls. Neuropsychopharmacology. 2022;47(11):1976–1983. doi: 10.1038/s41386-022-01316-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Casey BJ, Cannonier T, Conley MI, et al. The Adolescent Brain Cognitive Development (ABCD) study: Imaging acquisition across 21 sites. Dev Cogn Neurosci. 2018;32:43–54. doi: 10.1016/j.dcn.2018.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hagler DJ, Hatton SeanN, Cornejo MD, et al. Image processing and analysis methods for the Adolescent Brain Cognitive Development Study. NeuroImage. 2019;202:116091. doi: 10.1016/j.neuroimage.2019.116091 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Sripada C, Angstadt M, Taxali A, et al. Brain-wide functional connectivity patterns support general cognitive ability and mediate effects of socioeconomic status in youth. Transl Psychiatry. 2021;11(1):571. doi: 10.1038/s41398-021-01704-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Knutson B, Westdorp A, Kaiser E, Hommer D. FMRI Visualization of Brain Activity during a Monetary Incentive Delay Task. NeuroImage. 2000;12(1):20–27. doi: 10.1006/nimg.2000.0593 [DOI] [PubMed] [Google Scholar]
  • 43.Gordon EM, Laumann TO, Adeyemo B, Huckins JF, Kelley WM, Petersen SE. Generation and Evaluation of a Cortical Area Parcellation from Resting-State Correlations. Cereb Cortex. 2016;26(1):288–303. doi: 10.1093/cercor/bhu239 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Kardan O, Kaplan S, Wheelock MD, et al. Resting-state functional connectivity identifies individuals and predicts age in 8-to-26-month-olds. Dev Cogn Neurosci. 2022;56:101123. doi: 10.1016/j.dcn.2022.101123 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Brislin SJ, Martz ME, Joshi S, et al. Differentiated nomological networks of internalizing, externalizing, and the general factor of psychopathology (‘ p factor’) in emerging adolescence in the ABCD study. Psychol Med. 2022;52(14):3051–3061. doi: 10.1017/S0033291720005103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Casey BJ, Cannonier T, Conley MI, Cohen AO, Barch DM, Heitzeg MM, … & Dale AM (2018). The adolescent brain cognitive development (ABCD) study: imaging acquisition across 21 sites. Developmental cognitive neuroscience, 32, 43–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Gershon RC, Wagster MV, Hendrie HC, Fox NA, Cook KF, & Nowinski CJ (2013). NIH toolbox for assessment of neurological and behavioral function. Neurology, 80(11_supplement_3), S2–S6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Yang A, Rolls ET, Dong G, et al. Longer screen time utilization is associated with the polygenic risk for Attention-deficit/hyperactivity disorder with mediation by brain white matter microstructure. eBioMedicine. 2022;80:104039. doi: 10.1016/j.ebiom.2022.104039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Harvey DL, Milton K, Jones AP, Atkin AJ. International trends in screen-based behaviours from 2012 to 2019. Prev Med. 2022;154:106909. doi: 10.1016/j.ypmed.2021.106909 [DOI] [PubMed] [Google Scholar]
  • 50.Kim S, Jo H, Son Y, et al. Nationwide Trends in Screen Time and Associated Risk Factors by Family Structures Among Adolescents, 2008–2022: Nationwide Cross-Sectional Study. JMIR Public Health Surveill. 2025;11:e57962. doi: 10.2196/57962 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Radesky J, Chassiakos Y (Linda) R, Ameenuddin N, Navsaria D, COUNCIL ON COMMUNICATION AND MEDIA. Digital Advertising to Children. Pediatrics. 2020;146(1):e20201681. doi: 10.1542/peds.2020-1681 [DOI] [PubMed] [Google Scholar]
  • 52.Reid Chassiakos Y (Linda), Radesky J, Christakis D, et al. Children and Adolescents and Digital Media. Pediatrics. 2016;138(5). doi: 10.1542/peds.2016-2593 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Material
NIHMS2163765-supplement-Supplementary_Material.pdf (1.2MB, pdf)

Data Availability Statement

Scripts to generate AFC scores from fMRI data are shared in our previous work https://doi.org/10.1016/j.dcn.2025.101543. Scripts to run the regression analyses and generate the figures in this study are shared at https://github.com/okardan/ASU_neurodev_risk. Data tables including behavioral and task fMRI measures used in the scripts can be downloaded from the ABCD Study website with an approved ABCD data use agreement.

RESOURCES