War-related stress is associated with resting-state functional connectivity of cognitive control and sensory networks in children Authors: Sarit Roizman, Rola Farah, Noah Wenger, Mika Shapira Rootman, Rivka Tuval-Mashiach, Tzipi Horowitz-Kraus Published: 2026 ABSTRACT AbstractThe ability to read fluently relies on a synchronization between cognitive and sensory abilities and their corresponding brain networks. Environmental stress is related to reduced cognitive control abilities (i.e., executive functions; EF) as well as academic skills. Exposure to war-related events is recognized as a significant environmental stressor; however, whether such exposure directly affects the functional connectivity of brain networks underlying executive functions and sensory processing, both of which are critical for academic skills such as reading, remains unclear. This study aimed to assess the relationship between children’s exposure to war events and neurobiological networks supporting these abilities using resting-state functional connectivity. Here, forty-two children (5–13 years old) underwent fMRI resting state scans during the October 7th war (December 2023–2024) and were compared to fMRI datasets of thirty-seven age-matched children who were scanned prior to the war (January 2021- September 2023). Within- and between-functional connectivity (FC) in neural networks supporting EF (frontoparietal, cingulo-opercular, ventral, and dorsal attention) and reading (auditory and visual networks) were exported. War-related events were collected using a questionnaire together with EF, language, and reading measures. Behavioral results revealed significant reductions in the war-exposed group across several domains, including inhibitory control, speed of processing, phonemic awareness, and both word and non-word reading fluency (all p < .05). Neuroimaging results suggested a significant reduction in the between-network FC between EF and sensory networks and increased between-network FC in attention-sensory networks in the war-exposed group. Significant positive correlations within cognitive control networks and negative correlations between sensory networks were observed with elevated war-related events. The results suggest that greater exposure to war-related events is linked to cognitive decline and neurobiological changes associated with increased alertness and lower audio-visual integration, both of which are critical for academic success. Studies should further examine the connection between these neurobiological alterations and academic success.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-56279-x.Keywords: War-related events, Stress, Executive function, Sensory processing, Functional connectivitySubject terms: Attention, Cognitive control, Reading Introduction Reading fluency- a synchronization between cognitive and sensory networks Reading fluency, the ability to read accurately, quickly, and automatically, is critical for intact reading comprehension1. It was suggested that fluent reading is related to a synchronization between the sensory system modalities (i.e., the auditory and visual modalities) (for EEG studies see; for functional MRI studies, see2. In the past years, accumulated data pointed to the role of cognitive control (or executive functions) in reading fluency skills (for review see3. Neuroimaging studies support this point, demonstrating the increased functional connections (FC) between executive functions networks (cingulo-opercular; CO and frontoparietal; FP) during reading fluency improvements4, and between attention, executive functions, and sensory networks (e.g., correlations between the dorsal attention and visual processing5, the FP and visual processing6, and the CO and visual processing7. Reduced FC between the sensory system or within and between executive functions or attention networks were found to characterize non-fluent reading2,8. Although dyslexia, a specific case of reading difficulties, was previously related to the above-mentioned lower synchronization within the sensory and executive, and attention systems, a challenge in achieving reading fluency skills might also be attributed to an external, environmental interference such as COVID-19 times9, stress10, violence and trauma10, which fall under the category of adverse childhood experiences. Adverse childhood experiences: exposure to war-related events War-related events affect a person directly or indirectly. Direct impacts include being injured or suffering death, getting displaced, whereas indirect effects could include economic hardship, trouble with community infrastructure like healthcare and schooling, psychological trauma from witnessing violence, or being related to someone who suffered in the war11. Unfortunately, children are disproportionately affected by conflict; in 2021, approximately 449 million children lived in conflict zones12, which profoundly impacted their emotional and cognitive trajectories (Adverse Childhood Experiences [ACEs], 2014). Research has consistently shown that exposure to ACEs, including the chronic stress of war, leads to significant impairments in attention, executive function, and memory13. This disruption often translates into diminished academic achievement, with children at higher risk for learning disabilities and school absenteeism3. In war zones, children are exposed to multiple traumas and chronic stress, which hampers their cognitive development and academic performance14–16. These psychological disturbances interrupt essential cognitive functions such as concentration, memory, and attention17. War conditions also disrupt the natural social support systems, both at home and in school, making it difficult for children to receive the emotional and practical support they need. Families often face economic hardship, unstable living conditions, and frequent parental conflicts18. All of this has a significant effect on the children’s academic outcomes. Crucially, the neural networks that support academic skills, such as reading, are the same systems that are frequently compromised by chronic stress. For instance, exposure to adverse events is known to alter the sensitivity of attention networks (e.g., the Ventral attention network -VAN)19 and the efficiency of executive control networks (e.g., the FP network)20,21. Because fluent reading depends on the high-level functional synchronization between these cognitive-control systems and bottom-up sensory processing1,3, these networks are primary targets for investigating the neurobiological impact of war-related exposure on academic performance. War-related events and brain development Research investigating the impact of war zone deployment has traditionally identified significant volumetric reductions in brain regions critical for emotion regulation and cognitive control, such as the anterior cingulate cortex (ACC) and ventromedial prefrontal cortex (vmPFC)13,22,23. While the experiences of soldiers and civilian children differ in nature and duration, these findings establish a neurobiological baseline for the vulnerability of these neural hubs to extreme environmental stress. Extending this to pediatric populations, studies of children in conflict zones have found comparable structural brain changes, including reduced hippocampal volumes, associated with impaired learning24. Chronic stress in children has been linked to alterations in other brain regions as well. The amygdala, which is critical for emotional processing, often becomes hyperactive and enlarged in children exposed to prolonged stress, contributing to heightened anxiety and emotional dysregulation25. Additionally, the prefrontal cortex, responsible for executive functions such as decision-making and impulse control, can become underdeveloped, leading to difficulties in attention and problem-solving26. Previous neuroimaging studies27 demonstrated that children who experienced high levels of adversity showed reduced cortical thickness, particularly in the prefrontal cortex and temporal lobes. This reduction in cortical thickness is associated with impaired cognitive functions and increased risk of psychiatric disorders. Another study by De Bellis and Zisk28 highlighted that stress-induced neurobiological changes in children could have long-term consequences, increasing susceptibility to mental health issues such as depression, anxiety disorders, and post-traumatic stress disorder (PTSD). These brain changes underscore the critical need for early intervention and support for children exposed to high-stress environments. Overall, the evidence thus far suggests that chronic exposure to stress during critical periods of brain development can result in long-lasting structural and functional changes, potentially affecting cognitive, emotional, and psychological outcomes. While these structural studies provide a critical foundation for understanding the neurobiological impact of war-related stress, they do not capture the real-time synchronization between brain regions. Furthermore, functional alterations often manifest even in the absence of overt structural lesions, providing a more sensitive marker of environmental impact. Functional connectivity (FC) analysis offers a complementary perspective by highlighting how environmental stressors disrupt the communication within networks essential for cognitive and academic development. Despite the accumulated findings regarding the relationship between higher stress levels due to exposure to war-related events and children’s academic abilities, the neurobiological correlates of this exposure and its relation to children’s academic abilities are yet to be known. This study aims to determine the relations between adverse war-related events and child brain FC related to cognitive control, academic skills, and reading abilities. We hypothesize that children exposed to war-related events will show lower cognitive, language, and reading abilities. Furthermore, we hypothesize that children exposed to war-related events will exhibit significantly different FC patterns compared to non-exposed controls. Specifically, within the war-exposed group, we predict that a higher frequency of war-related events will positively correlate with within-network FC in cognitive control networks (FP and VAN). Conversely, we predict that increased exposure will negatively correlate with between-network FC between sensory processing networks (Auditory and Visual), reflecting a disruption in multisensory integration essential for reading fluency. Methods Participants A total of 79 children participated, divided to two separate groups: 1) 42 children in the war-exposed group, scanned between Oct 7 2023-Dec 2024 (mean age: 7.55, SD = 2.26); 11 females, and; 2) 37 in the none-war-exposed group, scanned between January 2021-September 2023 (mean age: 8.30, SD = 2.33; 14 females). The groups were balanced for age and gender and all were Caucasian from a middle-income household. Children were recruited from various regions in Israel in a manner comparable to that of the two research groups. Inclusion criteria included intact hearing and vision, and no psychiatric or neurological deficits. All participants had average socioeconomic status, were within the normal range of verbal and nonverbal IQ, and had no contraindications to MRI scanning. Study procedure Inclusion criteria for this study included healthy Hebrew-speaking children aged 5–13 years without neurological or psychiatric disorders, including attention difficulties or learning difficulties, and without contraindications for MRI scans. Children in both study’s conditions (exposed/non-exposed to war-related events) were recruited via online ads. Parents signed an informed written consent confirming their child’s participation in the study, and children provided verbal assents. The study procedure included a comprehensive behavioral assessment that evaluated children’s cognitive abilities, including EF, language processing, reading, and reading-related tasks. The war-exposed group (N = 42) was specifically assessed using the Exposure to War-related Events (EWE) questionnaire to quantify their level of trauma and environmental disruption. The non-exposed control group consisted of age-matched children scanned prior to October 2023 and, therefore, did not complete this war-related scale. This distinction allowed us to perform within-group correlations between the level of exposure and neurobiological markers, specifically within the affected population. The behavioral assessment lasted for about 1.5 h. Once the behavioral session was completed, children completed a 5-minute 3D T1-weighted anatomical scan and a 5-minute fMRI resting-state scan. The children were first desensitized before the scan to ensure they felt comfortable undergoing it. The study was conducted in the MRI Centers at Tel Aviv University and the Technion, Israel (based on a published protocol29, without any use of sedation or anesthesia during the MRI scans (following a desensitization protocol developed by the PI29). All children were compensated for their time and travel ($70). The study was approved by Schneider’s Children Hospital Medical Center (protocol number: 0531-16-RMC) and the Ministry of Health in Israel, as well as by the Technion’s ethical committee (protocol number: 08-2017). All experiments were performed in accordance with relevant guidelines and regulations. Behavioural measures Children’s cognitive abilities were assessed using age-appropriate measures. Young children aged 5–7 years were tested for verbal and non-verbal IQ using the Picture Naming Test and Matrix Reasoning subtest [WPPSI-IV30]. In addition to IQ testing, they completed a comprehensive behavioural battery that assessed language abilities, reading abilities, attention, and executive functioning (EF) skills, as follows: Language and language-related abilities were assessed using the Phonemic Awareness test [Shatil31], and Naming Letters and Objects tests31; Executive functions measures included processing speed (Symbol Search [WPPSI-IV30]; Naming Numbers-RAN [Aleftaf32], working memory (Digits Memory [CTOPP33], sustained attention (Score! and Sky Search [TEA-Ch]34, inhibitory control (Walk, Don’t Walk [TEA-Ch34]), and cognitive flexibility (Switching Colores/Animals35). Older children aged 7–13 years were tested for verbal and non-verbal IQ using the Picture Naming Test (PPVT); and the Matrix Reasoning subtest (WISC-IV36). As with the younger group, they completed a comprehensive behavioral battery. Language abilities were assessed using phonological manipulation tasks, including phoneme segmentation, merging, and omission (Aleftaf32). Reading abilities were assessed using the Decoding Words and Pseudowords subtests (Aleftaf32). Executive functions measures included processing speed (Symbol Search and Coding [WISC-IV]; Naming Numbers – RAN [Aleftaf]), working memory (Digits Memory [CTOPP]), sustained attention (Sky Search [TEA-Ch]), and cognitive flexibility (Stroop task [D-KEFS CW3]; and Naming Mixed Objects – RAS [Aleftaf]). The Exposure to War-related Events (EWE) questionnaire was inspired by the questionnaire assessing the number of traumatic events in adults37. This questionnaire (which followed the exposure to traumatic events in adults consisted of 12 items covering the following components: (1) exposure to sirens; (2) absence of a close person due to military reserve service; (3) home relocation; (4) lack of access to a shelter or safe place; (5) disruptions to daily routine (e.g., school closures or missed school days); and (6) exposure to war-related events through news, television, smartphones, or other media. Moreover, to determine the differences in neurobiological correlates for different types of stressors, the questions were grouped into four categories: 1) Close person who joined the service (q1-3); 2) Exposures to external traumatic stimuli (q4-6,10–12); 3) Transient from a familiar environment (q7); 4) Living in an unsafe environment (q8); 5) Routine change (q9). See supplemental materials 1 for the full list of questions. Neuroimaging parameters Data were collected at Tel Aviv University and at the Technion, utilizing a similar Siemens 3T MRI scanner. An anatomical T1-weighted scan was performed with the following parameters: repetition time (TR) = 8.05 ms, echo time (TE) = 3.7 ms, contiguous slices with a thickness of 1 mm, and a voxel size of 1 × 1 × 1 mm3. Anatomical T1-weighted images were primarily utilized for coregistration and normalization purposes to ensure accurate ROI placement. Analysis of structural metrics (e.g., cortical thickness or volume) was beyond the scope of this study, which focuses specifically on resting-state FC within and between cognitive and sensory networks. For the T2*-weighted BOLD fMRI scans, a multiband 4 Echo Planar Imaging (EPI) sequence was used with a TR of 1000 ms, TE of 28 ms, field of view (FOV) of 20 × 20 cm, matrix size of 100 × 100 × 68, voxel size of 2 × 2 × 2 mm3, and slice thickness of 2 mm. During the resting-state scan, a total of 300 whole-brain volumes were acquired over a duration of 5 min. Resting state paradigm Resting-state fMRI data were collected using the Inscape paradigm, which was designed to address the challenge of participants lying still for extended periods, a common issue that leads to head motion and drowsiness38,39. The Inscapes movie (5 min long) features abstract shapes and is designed to keep participants engaged without overwhelming them. No audio stimulation was used, and participants were instructed to keep their eyes open throughout the whole session. Functional MRI data preprocessing and post-processing Preprocessing of anatomical and functional MRI data was performed using the CONN toolbox40 and included the following steps: slice timing correction, realignment, segmentation, and coregistration. Functional images were coregistered to the participant’s corresponding high-resolution T1-weighted anatomical scan. Subsequently, all images were normalized to the MNI152 template to facilitate group-level comparisons. Following normalization, spatial smoothing was applied using a 5 mm Full-Width at Half-Maximum (FWHM) Gaussian kernel to enhance the signal-to-noise ratio. Preprocessing was conducted using a strict threshold (95th percentile), which flags the top 5% of time points (frames) showing the highest motion or signal changes (~ 0.5–0.9 mm). These flagged time points, along with global BOLD signal fluctuations exceeding ~ 3 standard deviations, were included as nuisance regressors (scrubbing) in the first-level General Linear Model to prevent motion-related artifacts from contaminating the FC estimates. Additionally, two participants were excluded due to incomplete scanning sessions, and the functional data for one participant was unusable due to excessive head motion exceeding the a priori scrubbing threshold. Power’s brain parcellation was applied to the coregistered functional images for analytical purposes. Power’s brain parcellation defines 264 ROIs and 14 functional networks. Following preprocessing, a denoising step was performed using the CONN toolbox to minimize the influence of non-neural signals on FC estimates. Nuisance variables, including five principal components from the white matter and cerebrospinal fluid masks (aCompCor method), the six realignment parameters, and their first-order derivatives, were regressed from the BOLD time series. Furthermore, the data underwent temporal band-pass filtering (0.008–0.09 Hz) to isolate low-frequency BOLD fluctuations and remove high-frequency physiological noise. Scrubbing was also applied as previously described to address motion outliers. Neural networks supporting reading fluency were selected for the analysis and included Cognitive networks [EF-related networks (FP, CO)], attention networks (DAN, VAN), and Sensory networks (visual, auditory). The selection of these specific networks was motivated by their dual vulnerability: they are the primary architecture for the cognitive-sensory synchronization required for fluent reading and are also networks previously shown to be highly sensitive to environmental stress, manifesting as increased alertness and altered executive control. See Fig. 1 for the networks. Masks for the cognitive control (EF and attention) and sensory networks. Data analyses Behavioural data analyses Descriptives (means and standard deviations) for the behavioral measures were calculated for each group. Independent samples t-tests were conducted to examine differences between the war-exposed and non-war-exposed groups across the behavioral measures. To determine how different types of war-related events were associated with FC patterns, scores per war-related event category were correlated with FC measures. Neuroimaging data analyses Several independent t-tests were conducted for the war-exposed vs. non-war-exposed groups to determine the difference in FC (within and between networks). To quantify the synchronization within and between the selected neural networks, FC metrics were calculated based on the Fisher-transformed bivariate Pearson correlation coefficients between the BOLD time series of ROIs defined by the Power parcellation. Within-network connectivity was defined as the mean correlation value across all possible ROI-to-ROI pairs within a single network (e.g., within the FP network). Between-network connectivity was defined as the mean correlation value between all ROI pairs belonging to two distinct networks (e.g., between the Auditory and VAN networks). These averaged Fisher-Z values were then used for subsequent group comparisons and correlation analyses with war-exposure scores . To determine the relationship between behavioral performance and neural synchronization, Pearson correlations were conducted between the behavioral measures showing significant group differences and the FC of the selected networks. Pearson correlations were conducted to determine the relations between the FC of the selected networks and war-related events in the war-exposed group. To assess the statistical robustness of these findings, a post-hoc power analysis was performed. With a sample size of N = 42 in the war-exposed group, the study has approximately 80% power to detect a medium effect size (r ≈ .40) at an alpha level of 0.05. Data were corrected for multiple comparisons for all the analyses in this section. Correlations between the scores for the war-event questionnaire and neuroimaging data analysis Pearson correlations were conducted to determine the relations between the FC of the selected networks and war-related events in the war-exposed group. To account for multiple comparisons across network pairs, all results were corrected using False Discovery Rate (FDR) correction at the network level (p < .05). Data were corrected for multiple comparisons for all the analyses in this section. Data will be available upon request from the PI (Horowitz-Kraus). Results Behavioral results T-test analyses suggested reduced cognitive and language skills in the war-exposed group, with significant differences found in inhibition (walk/don’t walk), speed of processing (symbol search), switching (naming RAS), phonemic awareness, and word/nonword reading. See Fig. 2; Table 1 for the results. Independent samples t-tests comparing behavioral measures in the war-exposed vs. non-exposed groups. Independent t-tests for cognitive, language and reading-related abilities in children who were not-exposed (left, X axis) and exposed (right, x axis) to war-related events. *p<.05; ***p<.001. Table 1Group differences between war-exposed and non-exposed children across cognitive, linguistic, and academic performance measures (controlling for nonverbal intelligence).DomainMeasure (Test)War-exposed groupMean (SD) – ANon-exposed groupMean (SD) – BContrastF(df) or t, pGeneral cognitiveabilitiesNonverbal intelligence(Matrix, WISC/WPPSI, scaled score)10.5(2.2)10(2.3)A > B-1.09, p=.275Verbal ability (Naming PPVT/ WPPSI, scaled score)0.53(0.198)0.70(0.228)A < B0.376, p = .708Speed of processingSymbol Search (WISC/ WPPSI, scaled score)9.48(0.39)10.80(0.417)A < B1.45, p=.015Working memoryDigits Memory (CTOPP, scaled score)9.22(0.463)10.20(0.495)A < B1.69, p = .095Visual attentionSky Search (TEA-Ch attention – overall attentional control, scaled score)8.28 (0.571)8.16 (0.630)A > B− 0.045, p =.964Auditory attentionScore! (TEA-Ch, scaled score)11.51 (0.613)12.67 (0.803)A < B1.82, p = .074InhibitionWalk Don’t Walk (TEA-Ch, scaled score)14.43(4.98)17.29(2.34)A < B2.19, p = .021Cognitive flexibilitySwitching Color/Animals(accuracy% of total)77.53(15.05)93.00(5.99)A < B4.57, p<.001Naming Mix Objects - RAS (Aleftaf, items Per Min, scaled score)-1.17 (0.167)-0.42 (0.281)A < B2.302, p=.026PhonemicPhonemic Awareness (Shatil, scaled score)0.36(0.91)1.25(1.68)A < B2.268, p=.028ReadingSingle Words Reading (Aleftaf, all responses per min, scaled score)-1.00(0.61)0.91(1.13)A < B5.85, p<.001Single Non-Words Reading (Aleftaf, all responses per min, scaled score)-1.39(0.87)0.97(1.10)A < B4.58, p<.001Note. SD = Standard Deviation; WISC/WPPSI = Wechsler Intelligence Scale for Children / Preschool and Primary Scale of Intelligence; PPVT = Peabody Picture Vocabulary Test; CTOPP = Comprehensive Test of Phonological Processing;TEA-Ch = Test of Everyday Attention for Children; D-KEFS=Delis–Kaplan Executive Function System, CW = Color-Word task; RAN = Rapid Automatized Naming;RAS = Rapid Alternating Stimulus; SHATIL=Functional Learning Deficit Screening Battery; Aleftaf = Reading & Writing Assessment Battery; Scores are scaled or standardized unless otherwise noted. Note that once the Levene assumption was not confirmed, an F test was applied. Contrast refers to the comparison of the non-exposed group (B) minus the war-exposed group (A). Positive t-values indicate higher scores in the non-exposed group. Group differences between war-exposed and non-exposed children across cognitive, linguistic, and academic performance measures (controlling for nonverbal intelligence). General cognitive abilities Nonverbal intelligence (Matrix, WISC/WPPSI, scaled score) 10.5 (2.2) 10 (2.3) Verbal ability (Naming PPVT/ WPPSI, scaled score) 0.53 (0.198) 0.70 (0.228) Symbol Search (WISC / WPPSI, scaled score) 9.48 (0.39) 10.80 (0.417) Digits Memory (CTOPP , scaled score) 9.22 (0.463) 10.20 (0.495) Sky Search (TEA-Ch attention – overall attentional control, scaled score) Score! (TEA-Ch, scaled score) Walk Don’t Walk (TEA-Ch, scaled score) 14.43 (4.98) 17.29 (2.34) Switching Color/Animals (accuracy% of total) 77.53 (15.05) 93.00 (5.99) Naming Mix Objects - RAS (Aleftaf, items Per Min, scaled score) Phonemic Awareness (Shatil, scaled score) 0.36 (0.91) 1.25 (1.68) Single Words Reading (Aleftaf, all responses per min, scaled score) -1.00 (0.61) 0.91 (1.13) 5.85, p<.001 Single Non-Words Reading (Aleftaf, all responses per min, scaled score) -1.39 (0.87) 0.97 (1.10) Note. SD = Standard Deviation; WISC/WPPSI = Wechsler Intelligence Scale for Children / Preschool and Primary Scale of Intelligence; PPVT = Peabody Picture Vocabulary Test; CTOPP = Comprehensive Test of Phonological Processing; TEA-Ch = Test of Everyday Attention for Children; D-KEFS=Delis–Kaplan Executive Function System, CW = Color-Word task; RAN = Rapid Automatized Naming; RAS = Rapid Alternating Stimulus; SHATIL=Functional Learning Deficit Screening Battery; Aleftaf = Reading & Writing Assessment Battery; Scores are scaled or standardized unless otherwise noted. Note that once the Levene assumption was not confirmed, an F test was applied. Contrast refers to the comparison of the non-exposed group (B) minus the war-exposed group (A). Positive t-values indicate higher scores in the non-exposed group. War-related questionnaire War exposure scores, calculated specifically for the 42 participants in the war-exposed group, showed a normal distribution (mean score 9.47, SD = 2.914). This indicates that most participants in this group experienced a moderate level of war-related exposure, with variance ranging from low-level news exposure to high-level environmental instability (Fig. 3). A normal distribution for the war exposure event questionnaire. Correlations between reading-related measures and war-related events Pearson correlations showed that a transient from a familiar environment due to the war was related to reduced phonological processing abilities (phonemic segmentation) (r = -.833, p = .001). Higher scores in the safety feeling category (i.e., lower safety feeling due to war) were related to lower naming scores (RAS) (r = -.291, p = .043). Neuroimaging results Differences between the two groups Independent t-test analyses demonstrated increased FC between attention-sensory (auditory-VAN, t = -1.818, p<.05) and a trend within the EF network (frontoparietal, t = -1.45, p = .075). See Fig. 4 and supplementary Tables 1, 2. Differences in FC between the group exposed to war-related events and the non-exposed group. Independent t-tests for between and within networks’ FC differences in children who were not-exposed (left, X axis) and exposed (right, x axis) to war-related events. Functional connectivity measures are noted in the Y-axis. *p<.05. Correlations between the war-related events questionnaire and neuroimaging data Correlation analysis between overall war exposure scores and network synchronization in the war-exposed group showed significant positive correlations between war-related events and the FC within EF and attention networks. Specifically, higher exposure to war-related events was associated with increased FC in EF and attention networks and reduced between-network FC between the sensory networks. See Fig. 5. Pearson correlation analyses between war-related events and FC measures. Pearson correlations for networks’ FC (Y axis) and an exposure to war related-events scores (X-axis). Correlations between subcategories of war-related events and neuroimaging measures Pearson correlation analysis between specific categories of war-related events and neural synchronization revealed a significant positive correlation between external war-related events and FC within the EF network (FP). Conversely, a significant negative correlation was found between these events and the FC between sensory networks. Results suggest that higher levels of external war-related events are associated with greater FC within the FP network (r = .33, p = .032) and reduced FC between the auditory-visual networks (r = -.248, p = .05). Correlations between cognitive-behavioral and neuroimaging measures To examine the relationship between behavioral performance and neural synchronization, Pearson correlations were computed between behavioral measures showing significant group differences and the FC of the selected networks. In the war-exposed group, a trend was observed where reduced inhibition scores (Walk/Don’t Walk) were associated with increased Auditory-VAN FC (r = -.26, p = .09), suggesting that heightened alertness may be linked to lower inhibitory control. Further studies with larger samples are needed to confirm these brain-behavior relationships. Discussion This study aimed to determine the relations between war-related events (type and quantity) and child brain FC related to cognitive control, academic skills (EF, attention, and sensory processing), and, more specifically, reading fluency skills. In line with our hypothesis, children exposed to war-related events showed lower cognitive and language abilities and different FC in cognitive control and sensory networks compared to those not exposed to war-related events. Within the group exposed to war-related events, more war-related events were related to a greater engagement of cognitive networks and a lower engagement of sensory processing networks. The results confirm the effect of external stress in critical school ages on children’s cognitive, language, and reading-related abilities. Results also provide neurobiological evidence for this effect on neural circuits supporting reading abilities. Exposure to war-related events is related to lower cognitive, language processing, and reading measures in children The current study’s results align with previous research pointing to the effect of unsafe and stressful environments on cognitive and academic skills, previously described in 625 adult students living in the south of Israel41. It was suggested that these students’ stress and sleep quality affected their performance. Others suggested a resilience of age in developing post-traumatic stress disorder (PTSD), with higher PTSD cases in younger vs. older adults during the Oct 7 war, even after controlling for basic demographics as well as pre-war psychological parameters42. These studies, although in young adults, position children, due to their young age and as children of the young adult population described above, in a vulnerable place to be affected by war-related events. Indeed, children aged 3–12 years in Israel exposed to continuous traumatic events during the Oct 7 war reported elevated post-traumatic stress symptoms, especially when their mothers reported elevated post-traumatic stress symptoms43. This effect was found to be more dramatic during early stages of development (i.e., under the age of 2 years44. These findings are in line with the ecological model for child development suggested by Bronfenbrenner, pointing at the importance of the close and far environments in shaping the child’s cognitive and emotional development. In the case of war, both the close (close family, home) and far environment (neighborhood, school) are becoming fragile and unstable, which may result in lower cognitive, linguistic, and reading achievements, which were found in the war-exposed group. This was especially evident in the significant correlations between increased unsafe feeling (i.e., no shelter in the house) and reduced scores in precursors for reading development (i.e., phonological processing and naming abilities). These results strengthen the importance of the environment, and not solely genetics, in developing the basic foundations for academic abilities in the child. War-related events are related to reduced synchronization between sensory and cognitive networks The reduced cognitive, linguistic, and reading scores found in the war-exposed (vs. non-exposed group) behaviorally did not reveal the mechanistic explanation for the possible causes for this reduced performance. Our functional results extend the existing structural literature by demonstrating that even in children without overt clinical diagnoses, war-related exposure is associated with altered network synchronization. This suggests that functional changes may precede or co-occur with structural alterations previously reported in war-exposed populations. Our functional MRI results shed light on the relations between exposure to war-related events and neural networks associated with cognitive control (EF, attention) and sensory processing. The increased FC between attention-sensory (auditory-VAN) and a higher level of war-related events was associated with higher within-network FC values within the EF (FP) and attention (VAN) networks, and reduced between-network FC between the sensory networks. The significant correlations observed between war exposure, Auditory-VAN connectivity, and reading scores suggest a candidate neurobiological pathway for academic decline. Specifically, we hypothesize a model where increased environmental stress disrupts sensory-attention synchronization, which in turn hinders fluent reading development. While the current study identifies these markers as significant co-occurrences, future research with larger cohorts is necessary to formally test this mediation. Our functional MRI results shed light on the relations between exposure to war-related events and neural networks associated with cognitive control (EF, attention) and sensory processing. Higher levels of war-related exposure were significantly associated with increased within-network FC (within the FP and within VAN) and increased between-network FC (Auditory-VAN), alongside a significant reduction in between-network sensory FC. The VAN was previously related to alertness19, which is expected in the context of exposure to war-related events. Interestingly, in addition to the traditional role of the FP network in EF abilities, increased FP FC was associated with increased levels of anxiety and stress as well as obsessive thoughts and anxiety20. Anxiety was also found to be related to reduced multisensory processing20. We suggest that increased war-related events are related to increased stress levels, which in turn are linked to increased anxiety and obsessive thoughts (FP), alertness (VAN), and reduced multisensory processing (auditory-visual processing). The correlation between the FP and audiovisual FC was specific to external war-related events such as sirens and the number of sirens, which probably triggered this feeling of anxiety. Interestingly, these networks are also related to reading fluency abilities, and hence it doesn’t come as a surprise that children who were exposed to more war-related events showed reduced reading abilities. To validate the connection between war-related events and anxiety level, an additional study including either physiological (cortisol) or a questionnaire should be administered. Study’s limitations The results of the current study should be taken into account in the context of the following limitations. First, due to ethical considerations, the study did not include children suffering from PTSD due to first-hand exposure to extreme war-related events (i.e., kidnapping, homicide). We can assume that the effect on these children might be more dramatic than the reported results, but it might also include additional effects. Future studies should include children who were exposed to more extreme war-related trauma events. Moreover, the comparison to the non-exposed group scanned before 2023 included children exposed to various environmental stimuli (such as other stressors, etc.), though not war-related, as well as IQ, education and family relationships. However, the within-group analysis (within war-exposed group) provided the specific relations between the level of war-related events and the type of events and children’s cognitive and neurobiological correlates. Third, while we identified robust correlations within the war-exposed group (N = 42), this sample size did not provide sufficient statistical power to perform a formal mediation analysis or to examine specific age-related effects to confirm the indirect effect of brain connectivity on the relationship between war exposure and reading fluency per age group. Future longitudinal studies with larger populations are required to statistically validate this mechanistic link. An additional important point is that the EWE questionnaire was not piloted or validated, as its goal was to quantify the level and type of exposure (rather than the severity of exposure). To assess the effect of war-related event severity on children, a validated, normalized questionnaire will be required. Lastly, while this power for the current study is sufficient for the primary correlation and t-test analyses reported, we acknowledge that it may be below the threshold required for more complex mediation or multi-factor ANOVA models, supporting future larger-scale studies focusing on children exposed to war-related events. Conclusions The current study demonstrated the relations between war-related events (amount and type) and FC of networks associated with EF, attention, and sensory processing. The altered synchronization in the child might be only the tip of the iceberg in the short- and long-term consequences of the effect of an exposure to war-related events: as parents and children also demonstrate brain-to-brain matching (fingerprints), can it be that war-related events may also affect the synchronization between the parent and the child? What might be the long-term consequences on the brain and academic abilities in children exposed to more war-related events, and whether the effect is reversible? Additional studies should be conducted to address these points. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements The authors would like to thank the families who participated in this study. Author contributions SR- formal analysis, wrote the paper, collected the data, prepared figuresadd space RF-formal analysis, wrote the paper, prepared figures add space RTM- wrote the paperNW- formal analysis, wrote the paper add spacerTHK-formal analysis, wrote the paper, collected the data, funding, data curation, prepared figuresadd space MSR- reviewed the paper, advice on analysisAll authors reviewed the manuscript. Funding This study was supported by the DFG add: The Diane and Guilford Glazer Foundation Data availability Data will be available upon request from the PI (Horowitz-Kraus). Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement Data will be available upon request from the PI (Horowitz-Kraus).