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

This review addresses whether and how excessive smartphone use—often discussed in terms such as problematic smartphone use, smartphone addiction, or Smartphone Use Disorder (SmUD)—is associated with measurable changes in human brain structure and function. In the context of more than six billion global smartphone subscriptions and pervasive integration of smartphones into work, social communication, navigation, and entertainment, concerns have risen about adverse cognitive, affective, and behavioral consequences (e.g., reduced productivity, lower academic achievement, increased negative emotionality, and safety risks like distracted driving). A key conceptual debate distinguishes the “medium” (smartphone) from the “functions” and contents (e.g., social media, gaming, email), which may differentially drive excessive use. The paper situates SmUD within an addiction framework (loss of control, preoccupation, continued use despite harm, functional impairment) while noting ongoing terminological disputes and the risk of over-pathologizing everyday behavior. Against this backdrop, the review synthesizes MRI-based evidence on associations between excessive smartphone use and brain alterations and outlines a roadmap to advance research on neurobiological, cognitive, and affective impacts.

The review collates >20 MRI studies (with a marked increase since 2020) examining associations between SmUD tendencies and brain structure/function. Structural MRI studies reported gray matter variations (often via voxel-based morphometry of T1 images) and white matter differences (diffusion tensor imaging with TBSS), with findings including reduced volumes or densities in orbitofrontal cortex, anterior cingulate cortex, insula, parahippocampal regions, frontal areas, and subcortical structures, alongside reduced white matter integrity in multiple tracts. Functional MRI studies included resting-state investigations of large-scale networks (salience, default mode, central executive, attention) and task-based paradigms probing emotion processing, attention, cognitive control, and cue reactivity. Across studies, alterations in connectivity and regional engagement—particularly within salience, executive control, default mode, and fronto-striatal systems—mirror patterns reported in substance and behavioral addictions. However, heterogeneity in SmUD assessments (e.g., SAS, SAPS, SPAI, MPAI), analytic pipelines, regions-of-interest versus whole-brain approaches, sample sizes, and multiple-comparison handling complicates synthesis and limits strong conclusions.

This is a narrative review of MRI-based research on smartphone (over-)use up to early 2023. It summarizes structural MRI methods (T1-weighted morphometry, voxel-based morphometry, cortical thickness/folding analyses; white matter via diffusion tensor imaging and tract-based spatial statistics) and functional MRI methods (resting-state connectivity of large-scale networks; task-based fMRI assessing emotion processing, attention/cognitive control, and smartphone cue reactivity). Studies employed both group comparisons (high vs. low SmUD) and dimensional analyses correlating symptom scores with neuroimaging measures. Smartphone use severity was assessed via diverse self-report instruments (SAS, SAPS, SPAI, MPAI, etc.), with few studies incorporating objective smartphone tracking. The review highlights methodological variability across processing pipelines, analytic strategies, and correction procedures, and notes the dominance of cross-sectional designs without repeated MRI measures.

- Structural MRI: Higher SmUD tendencies frequently associate with reduced gray matter in regions including orbitofrontal cortex (especially right lateral OFC), anterior cingulate cortex, anterior insula, inferior temporal and parahippocampal cortex, thalamus, and frontal gyri. Some studies report altered caudate volume and inconsistent gray matter density changes in insula and parietal regions. White matter studies show lower integrity (e.g., reduced fractional anisotropy/axial diffusivity) in tracts such as superior longitudinal fasciculus, corona radiata, internal/external capsule, sagittal stratum, fornix/stria terminalis, midbrain structures, hippocampal cingulum, and corpus callosum. - Resting-state fMRI: Problematic users often exhibit increased connectivity within and between salience and default mode networks, reduced salience–central executive connectivity, altered ACC coupling with ventral attention and default mode networks, and changes in network efficiency and temporal variability (e.g., higher static efficiency within frontoparietal/cingulo-opercular networks and reduced dynamic variability in attention networks correlating with SmUD scores). - Task-based fMRI: High SmUD groups show reduced engagement of dorsolateral prefrontal cortex and dorsal ACC during angry-face processing, impaired attentional control with generalized hyperactivation of fronto-parietal regions, lower frontopolar activation and poorer distractor filtering, and cue reactivity differences involving anterior cingulate, medial prefrontal, and temporal regions. Emotional face recognition paradigms link higher smartphone addiction scores to increased connectivity within emotional/cognitive control networks. - Overall: Findings align with addiction-related alterations in salience, executive control, default mode, and fronto-striatal systems, suggesting potential neural mediators of cognitive/affective dysregulations in SmUD. Yet evidence is inconsistent, often underpowered, and predominantly cross-sectional, precluding causal inference.

The reviewed MRI literature indicates that excessive smartphone use is associated with structural and functional brain variations in systems implicated in reward/motivation, salience attribution, executive control, emotion processing, and habit formation—paralleling substance and behavioral addictions. These associations may underlie observed cognitive and affective dysregulations (e.g., impaired attention, altered emotion processing) and potential progression toward compulsive use. However, heterogeneity in assessments, analytic pipelines, small samples, and cross-sectional designs hinder robust conclusions about directionality (predisposition vs. consequence). The roadmap emphasizes: studying symptom facets rather than only total SmUD scores and disentangling SmUD-specific effects from comorbid domains (e.g., depression, anxiety, FOMO); expanding task-based fMRI targeting cue reactivity, executive functions, emotion/stress, and natural reward processing; integrating objective smartphone tracking (digital phenotyping/mobile sensing) to complement self-report; implementing longitudinal designs with repeated MRI to separate predisposing markers from consequences of escalating use; multimodal integration (structural, functional, EEG/fNIRS/PET, hormones/genetics) to link anatomy with function; developing a taxonomy of use patterns and app contexts; examining sex and lifespan effects; and improving replicability via transparent data/code sharing, pre-registration, and designs beyond traditional case–control.

Current MRI evidence suggests associations between smartphone (over-)use and alterations in brain regions and networks related to reward/motivation, salience, executive control, and emotion, but remains fragmentary and largely cross-sectional. The field lacks overarching frameworks uniquely tailored to smartphone use, and cannot yet establish whether observed brain characteristics reflect predispositions or consequences of specific use patterns. To advance, research should employ prospective longitudinal designs, larger and replicated samples, rigorous multiple-comparison controls, objective smartphone-use tracking, and multimodal neuroimaging integrating structural, functional, and physiological measures. Targeted task-based paradigms and symptom-specific analyses, alongside consideration of sex and developmental stages, are needed to build robust, generalizable neurobiological models of smartphone (over-)use.

- Predominance of retrospective cross-sectional designs without repeated MRI measures, preventing causal inference and separation of predisposing vs. consequential brain changes. - Small, often underpowered samples and inconsistent application/reporting of multiple-comparison corrections. - Heterogeneity in SmUD assessment instruments (SAS, SAPS, SPAI, MPAI, etc.) and lack of consensus on conceptual framework, limiting comparability. - Variability in MRI processing pipelines and analytic strategies (ROI vs. whole-brain), which can drive differences in reported findings. - Limited task-based fMRI studies and scarcity of objective smartphone-use tracking data; reliance on self-report prone to inaccuracies. - Potential confounding by comorbid psychological domains (e.g., depression, anxiety, FOMO) not consistently controlled across studies. - Focus primarily on MRI, with limited multimodal integration across other neuroimaging and physiological measures.