Neuroimaging the effects of smartphone (over-)use on brain function and structure-a review on the current state of MRI-based findings and a roadmap for future research

The paper addresses whether and how excessive smartphone use (often termed problematic smartphone use, smartphone use disorder, or smartphone addiction) relates to alterations in human brain structure and function. Contextualized within the rapid global adoption of smartphones and concerns about detrimental effects (e.g., reduced productivity and academic performance, elevated negative affect, and safety risks), the review examines if excessive use exhibits addiction-like features and associated neurobiological correlates. The authors discuss the ongoing debate about medium versus function (i.e., whether impacts stem from the device itself or specific app/content use such as social media or games) and terminology (problematic smartphone use vs. addiction vs. Smartphone Use Disorder, SmUD). The purpose is to synthesize MRI-based evidence on structural and functional brain differences linked to excessive smartphone use and outline a roadmap for rigorous future research.

The review compiles more than 20 MRI studies (with a marked increase since 2020) investigating associations between excessive smartphone use/SmUD and brain variations. Structural MRI work includes voxel-based morphometry (VBM) on T1-weighted images and diffusion tensor imaging (DTI) with tract-based spatial statistics to assess gray matter volumes, cortical morphology (including cortical folding), and white matter integrity. Functional MRI studies cover resting-state connectivity (intrinsic network organization) and task-based fMRI paradigms (e.g., facial emotion processing, cue reactivity, cognitive control, oddball tasks). The literature uses heterogeneous self-report instruments to assess problematic use (e.g., SAS/SAS-SV, SAPS, SPAI, MPAI), varies in analytic approaches (whole-brain vs. ROI; preprocessing pipelines), and often comprises small, cross-sectional samples. Summaries in Tables 1 and 2 report regional gray matter reductions (e.g., ACC, OFC, fusiform, parahippocampal areas, striatum/caudate), white-matter integrity differences (e.g., corpus callosum, internal capsule, hippocampal cingulum bundle), altered resting-state connectivity involving salience, default mode, central executive, striatal–limbic–frontal systems, and task-evoked alterations in prefrontal and cingulate regions during emotional and cognitive tasks. Despite some overlap with neural patterns reported in substance and behavioral addictions, replication across SmUD studies is limited, and consistent brain markers have not yet emerged.

This is a narrative review of the MRI literature on smartphone (over-)use/SmUD. The authors organized the evidence into structural MRI (VBM, DTI/cortical morphology) and functional MRI (resting-state and task-based) sections and summarized results in tables and a schematic figure. No formal systematic search protocol, inclusion/exclusion criteria, or meta-analytic methods are reported. The review emphasizes methodological heterogeneity across primary studies, including variability in SmUD assessments, image preprocessing and analysis pipelines, regions of interest vs. whole-brain analyses, and statistical correction practices. The authors integrate findings qualitatively and derive a roadmap for future research based on observed gaps and best practices from addiction neuroimaging.

- Structural MRI: Many studies report inverse associations between SmUD severity and regional gray matter volume, including in the anterior cingulate cortex (ACC), orbitofrontal cortex (OFC), fusiform gyrus, parahippocampal regions, and striatum (caudate). Group comparisons often show lower volumes in high SmUD groups. However, regions do not consistently replicate across studies, and analytic variability complicates interpretation. White-matter studies show reduced integrity (e.g., corpus callosum body fractional anisotropy inversely correlated with PSU; lower integrity in superior longitudinal fasciculus, internal/external capsules, hippocampal cingulum, fornix/stria terminalis). - Resting-state fMRI: Problematic users often show altered connectivity within and between large-scale networks (salience, default mode, central executive) and circuits involving striatal, limbic, and frontal regions. Examples include enhanced salience–DMN coupling and reduced salience–executive coupling; lower OFC–NAcc connectivity; reduced fronto-limbic connectivity; and altered dorsal ACC interactions with ventral attention and default mode networks. Findings overlap partly with patterns reported in substance and behavioral addictions but remain heterogeneous. - Task-based fMRI: Fewer studies exist. Reported findings include decreased dorsolateral prefrontal cortex and dorsal ACC responses to angry faces; enhanced but less differentiated fronto-parietal activation with poorer task performance in cognitive conflict tasks; reduced frontopolar activation and impaired distractor filtering in an oddball task; and cue-reactivity differences in ACC/medial prefrontal and temporal regions to smartphone cues. A study linked higher smartphone addiction scale scores to increased connectivity among emotional/cognitive control networks during facial emotion recognition. - Overall: Evidence suggests that excessive smartphone use associates with variations in brain structure and function in systems implicated in salience/reward, executive control, emotion, and habit/compulsive behavior. However, small sample sizes, cross-sectional designs, heterogeneous measures, and analytic variability preclude firm conclusions about specific, reliable neural markers.

The assembled findings suggest that excessive smartphone use may engage and alter neural circuits commonly implicated in addictive and affective-cognitive dysregulations, potentially mediating observed impairments in attention, executive control, and emotional processing. Yet, due to methodological limitations, it remains unclear whether observed neural differences reflect predispositions for excessive use, consequences of escalated use, or confounding comorbidities (e.g., depression/anxiety) and psychological traits (e.g., fear of missing out). The review underscores the need to disentangle medium- versus function-specific effects (e.g., social media vs. gaming vs. work email), to employ standardized and validated assessments of SmUD and its symptom dimensions, and to incorporate objective smartphone usage tracking alongside self-reports. Aligning research with frameworks and methods from addiction neuroscience (e.g., cue reactivity, executive control tasks, longitudinal designs) may clarify mechanisms and the extent to which SmUD maps onto established addiction-related neurobiology.

MRI studies indicate potential associations between smartphone (over-)use and alterations in brain structure and function across cortical and subcortical systems involved in reward/motivation, salience, executive control, and emotion. However, current evidence is fragmentary and does not establish causality or robust, replicable markers specific to SmUD. The authors propose a roadmap: (1) examine symptom dimensions/facets of SmUD and related constructs (e.g., FOMO), distinguishing SmUD-specific from transdiagnostic effects; (2) expand task-based fMRI targeting addiction-relevant domains (cue reactivity, executive functions, emotion/stress, natural reward); (3) integrate objective, tracked smartphone use data with neuroimaging; (4) conduct prospective longitudinal studies with repeated MRI to separate predisposition from consequence; (5) integrate multimodal brain measures (structural, functional, EEG/fNIRS/PET), hormones, and genetics; (6) develop a taxonomy of app- and context-specific smartphone use; (7) combine structural and functional MRI sources to link anatomy with function; (8) examine sex differences and lifespan effects; and (9) enhance reproducibility via adequate power, rigorous multiple comparison control, preregistration, and transparent data/code sharing. Such advances are needed to build a coherent neuroscientific framework for smartphone (over-)use.

Primary studies are predominantly cross-sectional with small samples, heterogeneous and non-standardized SmUD assessments, and variable imaging preprocessing/analysis pipelines; multiple comparison corrections are inconsistently applied. Few studies include objective smartphone usage tracking; most rely on self-reports that poorly capture true usage quantities. There is a lack of prospective longitudinal MRI with repeated measures, minimal replication, and limited task-based paradigms. Potential confounds (e.g., depression, anxiety, FOMO) are not consistently controlled. The diversity of smartphone functions and app use is rarely disentangled, and animal models are not available for mechanistic inference.