Of the 874 peer-reviewed studies in this collection on electromagnetic fields, 701 — 80.2% — reported biological effects. On children's brains specifically the direct evidence is thin and the honest summary is layered: tissue measurements show absorption properties change with age, animal work shows radiofrequency exposure producing DNA strand breaks in brain cells, one large birth cohort links prenatal phone exposure to behavioral difficulties while cautioning the link may be non-causal, and the largest brain-tumor study found no overall increase in risk.
Based on analysis of 903 peer-reviewed studies
Children's brains are fundamentally different from adult brains—not just smaller, but actively developing, forming new neural connections, and undergoing critical periods of growth. This raises important questions about how electromagnetic field exposure might affect the developing brain.
Researchers have approached this question through multiple methods: measuring how much RF energy children's brains absorb compared to adults, studying cognitive outcomes in children with various EMF exposures, and examining brain tissue effects in laboratory settings.
This page presents the scientific evidence on EMF exposure and childhood brain development.
Of the 874 peer-reviewed studies indexed here on electromagnetic fields, 701 — 80.2% — reported measurable biological effects. Very few examine children's brains directly. Children cannot be enrolled in controlled radiofrequency exposure experiments, so the evidence comes from animal brain tissue, tissue-property measurement, adult cognitive testing, and observational cohorts.
Peyman and colleagues (2001) measured the dielectric properties of ten rat tissues at six ages across 130 MHz to 10 GHz. Properties decreased with age, and the trend was most apparent in brain, skull and skin, driven by declining water content. Because those properties govern how radiofrequency energy enters and deposits in tissue, younger tissue interacts with these fields differently. This is the measured basis for treating children's exposure as a distinct question — tissue data from animals, not a measured dose to a child's brain.
Lai and Singh (1996) exposed rats at a spatial-averaged power density of 2 mW/cm², producing a whole-body specific absorption rate of 1.2 W/kg, and measured DNA strand breaks in individual brain cells four hours after exposure. Both single- and double-strand breaks increased, with no significant difference between pulsed and continuous-wave radiation. The authors positioned it as support for earlier in vitro and in vivo work showing radiofrequency effects on DNA.
A developing brain is building the structure it will use for life, and DNA damage during that period carries different consequences than the same damage in a mature, slowly dividing tissue. That reasoning is why the animal result gets weighted heavily in discussions of children — it is an inference about vulnerability, not a measured pediatric outcome.
Preece and colleagues (1999) tested 36 adults under about 1 W mean power at 915 MHz in a randomized cross-over design. Only choice reaction time changed, and it improved. Word, number and picture recall and spatial memory were unaffected.
The behavioral signal in children comes from Divan and colleagues (2008), who followed 13,159 children in the Danish National Birth Cohort to age seven. Children with possible prenatal or postnatal phone exposure showed greater odds of behavioral problems — emotional and hyperactivity difficulties — with the prenatal association stronger and surviving adjustment for confounders. The authors state that the associations may be non-causal and may reflect unmeasured confounding.
Hutter and colleagues (2006) added population evidence in adults, finding effects on wellbeing and cognitive performance among 365 residents near base stations at exposures far below recommended limits, with no known mechanism at those levels.
The INTERPHONE study ran an interview-based case-control study across 13 countries with 2,708 glioma and 2,409 meningioma cases and matched controls. It found no overall increase in either tumor type among mobile phone users — in fact reduced odds ratios that its authors attributed to participation bias or other methodological limitations. Suggestions of increased glioma risk appeared at the highest exposure levels, but the authors concluded that bias and error prevent a causal interpretation. That study concerns adults and tumors, not children and development, and it is the most substantial null result in this area.
Nothing here demonstrates that phone or WiFi exposure impairs a child's brain development. The tissue and DNA work is animal-based at exposures above typical use. The cognitive findings are in adults and are not uniformly negative. The behavioral cohort measures recalled phone use and carries the confounding its authors name.
The steps that follow from this evidence are the ones that would be sensible even if every study above were overturned: keep phones away from children's heads by using speaker mode or wired headphones, favor texting over calling, keep devices out of bedrooms overnight, and use wired connections where devices do not move. Each reduces exposure during the years when the tissue differences are largest, and none costs anything worth counting.
Sidorenko A, Tsaryuk V · 2000
Researchers studied the effects of microwave radiation and strychnine on cerebral biopotentials in narcotized rats. Strychnine and microwave radiation produced changes in spectral parameters of electrocorticogram, correlation dimension, and Kolmogorov entropy opposite to those induced by urethane.
Schirmacher A et al. · 2000
Krause CM et al. · 2000
Researchers examined the effects of electromagnetic fields from cellular phones on event-related desynchronization and synchronization responses in four EEG frequency bands during cognitive processing. Twenty-four subjects performed a visual sequential letter task with three working memory load conditions both with and without exposure to a digital 902 MHz EMF in counterbalanced order. The presence of EMF altered the ERD/ERS responses in the 6-8 and 8-10 Hz frequency bands but only when examined as a function of memory load and depending on whether the stimulus was a target.
Krause CM et al. · 2000
Researchers studied the effects of cellular phone EMF on EEG frequency bands in 16 normal subjects performing an auditory memory task. All subjects completed the memory task with and without exposure to a digital 902 MHz EMF in counterbalanced order. EMF exposure significantly increased EEG power in the 8-10 Hz frequency band only. EMF presence altered ERD/ERS responses across all studied frequency bands depending on time and memory task phase.
Koivisto M, Krause CM, Revonsuo A, Laine M, Hamalainen H · 2000
The influence of pulsed radiofrequency (RF) electromagnetic fields of digital GSM mobile phones on working memory in healthy subjects were studied. Memory load was varied from 0 to 3 items in an n-back task, and each subject was tested twice with and without RF exposure at 902 MHz and 217 Hz. The RF field speeded up response times when the memory load was three items, but no effects were observed with lower loads. The results suggest that RF fields have a measurable effect on human cognitive performance and encourage further studies.
Koivisto et al. · 2000
The study examined possible influences of a 902 MHz electromagnetic field emitted by cellular telephones on cognitive functioning in 48 healthy humans. A battery of 12 reaction time tasks was performed twice by each participant in a counterbalanced order, once with and once without exposure to the field. Exposure speeded up response times in simple reaction time and vigilance tasks and decreased the cognitive time needed in a mental arithmetic task. The results suggest that exposure may have a facilitatory effect on brain functioning, especially in tasks requiring attention and manipulation of information in working memory.
Hardell L, Nasman A, Pahlson A, Hallquist A. · 2000
The study included 233 men and women aged 20 to 80 years with histopathologically verified brain tumors, with two matched controls selected per case from the Swedish Population Register. Of these, 209 cases (90%) and 425 controls (91%) completed the exposure questionnaire. Medical diagnostic x-ray examination of the head and neck region yielded an odds ratio of 2.10 (95% CI 1.25-3.53) for a tumor induction period of 5 years or more. Ipsilateral cellular telephone use was associated with an odds ratio of 2.42 (95% CI 0.97-6.05) for tumors in the temporal, temporoparietal, and occipital lobes.
Freude, G, Ullsperger, P, Eggert, S, Ruppe, I, · 2000
Researchers studied the influence of electromagnetic fields emitted by cellular telephones on preparatory slow brain potentials in two experiments about six months apart. A significant decrease of slow brain potentials was found during EMF exposure in a complex visual monitoring task in the first experiment and this effect was replicated in the second experiment. The results indicated a selective EMF effect on particular aspects of human information processing but no influence on human performance, well-being and health.
Cox RA, Luxton LM · 2000
The study aimed to investigate cerebral symptoms from mobile telephones. Mobile phones affect the inner ear in 5-8% of users leading to dizziness, disorientation, nausea, headache and transient confusion.
Lebedeva NN et al. · 2000
EEG was recorded from 16 standard leads during eyes-open and eyes-closed conditions and analyzed with custom non-linear dynamics software that computed a multichannel global correlation dimension. A significant increase in global correlation dimension occurred during and after exposure, more pronounced with eyes closed. The authors interpreted this increase as a manifestation of cortex activation under phone EMF exposure.
Ivanova VIu, Martynova OV, Aleinik SV, Limarenko AV. · 2000
Researchers studied the effect of modulated electromagnetic fields on brainwave activity in awake cats by recording electroencephalograms from the skin surface in the vertex area using carbon electrodes. In normal cats, brainwave components above 20 Hz predominated, but during electromagnetic field exposure at 980 MHz and a power density of 30–50 microW/cm2, components in the 12–18 Hz range began to prevail instead. A similar redistribution of brainwave power was also seen with acoustic stimulation, and the researchers suggested a common neurophysiological mechanism underlying both effects.
Huber R et al. · 2000
Researchers exposed healthy young male subjects for 30 minutes to EMF at 900 MHz and a spatial peak specific absorption rate of 1 W/kg during the waking period before sleep. Compared with a sham exposure control condition, EEG spectral power during non-rapid eye movement sleep was increased, with the maximum rise in the 9.75-11.25 Hz and 12.5-13.25 Hz bands during the initial part of sleep. The researchers concluded that the changes of brain function induced by pulsed high-frequency EMF outlast the exposure period.
Adey WR et al. · 2000
Researchers exposed Fischer 344 rats to a frequency-modulated signal at 836.55 MHz in a two-year bioassay, testing for effects on spontaneous and chemically induced central nervous system tumors. The study used six groups of offspring totaling 540 rats, with some groups receiving the chemical ENU and others sham or field exposure. No FM field-mediated changes were observed in number, incidence, or histological type of either spontaneous or ENU-induced brain tumors, nor were gender differences detected in tumor numbers.
Noda Y, Mori A, Liburdy RP, Packer L · 2000
Researchers investigated the effect of pulsed magnetic fields on nitric oxide synthase (NOS) activity in the rat brain. A significant increase in NOS activity was observed in the cerebellum at 111.2±2.0% (P<0.05) for a 1 Gauss (0.1 mT) pulsed DC field, across five separate experiments. Neither AC nor static DC field treatment showed any significant change in NOS activity. In the hippocampus, cortex, medulla oblongata, hypothalamus, striatum, and midbrain, only a slight change or no effect was observed.
Unknown authors · 1999
Researchers conducted a double-blind laboratory study of 24 healthy young men to examine effects of intermittent versus continuous exposure to a 60 Hz, 28.3 microT magnetic field on EEG measures of night sleep. Intermittent exposure was associated with less total sleep time, reduced sleep efficiency, increased Stage II sleep, and decreased REM sleep, while continuous and sham exposure showed no such effects. Subjects exposed intermittently reported sleeping less well and feeling less rested in the morning compared to the other groups. All observed effects were significant at P < or = .04 or less.
Higashikubo R et al. · 1999
Researchers exposed Fischer 344 rats implanted with 9L gliosarcoma cells to 835.62 MHz FMCW or 847.74 MHz CDMA fields with a nominal brain specific absorption rate of 0.75 +/- 0.25 W/kg. The animals were exposed for 4 hours a day, 5 days a week, starting 4 weeks before tumor implantation and continuing up to 150 days after. Sham-exposed rats injected with 2 to 10 cells had a median survival of 70 days with 27% surviving to 150 days; those with 11 to 36 cells had a median survival of 52 days and 14% final survival; those with 37 to 100 cells had a median survival of 45 days and 0% survival. Statistical comparison showed no significant differences in survival curves between RF-exposed and sham-exposed animals in any of the three cell-dose groups.
Wu Y, Jia Y, Guo Y, Zheng Z · 1999
Wistar rats were divided into four groups with three exposure groups and one control group, and learning was measured with a Y-maze followed by neurotransmitter detection in brain areas. Learning ability decreased in exposure groups over three days of EMP exposure compared with controls. After one day, hippocampal 5-HT and DOPAC increased and hypothalamic dopamine increased while hypothalamic Adr decreased; after the second day, hippocampal Adr and 5-HT reduced; after the third day, hippocampal Adr remained lower than control.
Sidorenko AV · 1999
Researchers used the nonlinear dynamics method together with the spectral correlation method to process animal electrocorticograms. They found that the nonlinear dynamics method may be used to analyze electrocorticograms in different functional states, confirmed by increased correlation dimensionality in animals exposed to microwaves or given strychnine. Correlation dimensionality serves as a numerical criterion for comparing dynamic states of the system.
Preece et al. · 1999
Thirty-six participants completed a randomized three-way crossover study with two training sessions and three test sessions. The only cognitive measure that changed was choice reaction time, which became faster, while word, number, picture recall and spatial memory were unchanged.
Morrissey JJ et al. · 1999
Researchers investigated the effects of a one-hour exposure of mice to a 1.6-GHz radiofrequency signal, either continuous wave or pulse-modulated at 11 Hz (IRIDIUM), on c-fos gene expression in the brain. C-fos expression was not significantly elevated until exposure levels exceeded six times the peak dose and 30 times the whole-body average dose relative to human cellular telephone exposure limits. The pattern of c-fos elevation in limbic cortex and subcortex areas was consistent with a stress response due to thermal perception and restraint, not a direct interaction of IRIDIUM energy with brain tissue.
Lamble D, Kauranen T, Laakso M, Summala H · 1999
Nineteen participants aged 20 to 29 years drove at 80 km/h, 50 m behind a lead car on a 30 km motorway section. The lead car decelerated at 0.47 m/s² while participants either looked ahead, dialed numbers, or performed a memory and addition task. Detection ability was impaired by about 0.5 s in brake reaction time and almost 1 s in time-to-collision during the non-visual task. This impairment was similar to when drivers divided visual attention by dialing numbers.
Johnson EH, Chima SC, Muirhead DE · 1999
Researchers examined an adult squirrel monkey with a history of long-term exposure to microwave radiation that had a malignant tumor in the right cerebral cortex. Microscopy showed sheets of moderate-sized cells with clear cytoplasm and central nuclei, resembling an oligodendroglioma, with malignant features including nuclear pleomorphism, mitotic figures, and necrosis. Immunohistochemical analysis supported a neuronal origin with positivity for neurofilament proteins and neuron-specific enolase, while vimentin and glial fibrillary acidic protein were negative except in reactive astrocytes. The tumor showed strong positivity for Ki-67 antigen and p53 oncoprotein, indicating aggressive malignant nature, and was diagnosed as a cerebral primitive neuroectodermal tumor.
Hardell et al. · 1999
Researchers conducted a case-control study with 209 brain tumour cases and 425 controls in Sweden. Overall cellular telephone use yielded an odds ratio of 0.98 with a 95% confidence interval. Separate analyses for the digital GSM system and the analogue NMT system gave odds ratios of 0.97 and 0.94 respectively. A non-significantly increased risk was found for tumours in the temporal or occipital lobe on the same side as phone use. The increased risk was only seen for use of the NMT system.
Dreyer NA, Loughlin JE, Rothman KJ · 1999
The initial protocol called for continuing surveillance for mortality. Access to data beyond 1 year was blocked by a class-action lawsuit challenging the legality of monitoring product safety through record linkage.
de Seze R, Ayoub J, Peray P, Miro L, Touitou Y · 1999
Researchers evaluated maximum serum melatonin concentration, time of maximum, and area under the curve of the hormone profile. Thirty-eight men aged 20 to 32 years were exposed to GSM 900 or DCS 1800 phones for 2 hr/day, 5 days/wk, for 4 wk at maximum power. Blood samples were collected hourly during the night and every 3 hr in the daytime across four sessions at 15-day intervals. The melatonin circadian profile was not disrupted by the exposure.
Further Reading:
For a comprehensive exploration of EMF health effects and practical protection strategies, explore these books by R Blank and Dr. Martin Blank.
