The results cannot be generalized to cell phone radiation exposures which are more complex. Cell phones generate a mixture of ELF and RF EMF. A smart phone may emit Bluetooth, Wi-Fi, GSM, UMTS, and LTE signals, at multiple frequencies. For example, GSM is pulsed at 2 Hz, 8 Hz, and 217 Hz and the carrier wave may be at 1800-1900 MHz as well as 450-900 MHz. The cell phone's power system may also produce ELF magnetic fields.
Inter-individual and intra-individual variation of the effects of pulsed RF EMF exposure on the human sleep EEG
Lustenberger, C., Murbach, M., Tüshaus, L., Wehrle, F., Kuster, N., Achermann, P. and Huber, R. (2015), Inter-individual and intra-individual variation of the effects of pulsed RF EMF exposure on the human sleep EEG. Bioelectromagnetics. doi: 10.1002/bem.21893. First published online Feb 17, 2015.
Pulse-modulated radiofrequency electromagnetic fields (RF EMF) can alter brain activity during sleep; increases of electroencephalographic (EEG) power in the sleep spindle (13.75–15.25 Hz) and delta-theta (1.25–9 Hz) frequency range have been reported. These field effects show striking inter-individual differences. However, it is still unknown whether individual subjects react in a similar way when repeatedly exposed. Thus, our study aimed to investigate inter-individual variation and intra-individual stability of field effects.
To do so, we exposed 20 young male subjects twice for 30 min prior to sleep to the same amplitude modulated 900 MHz (2 Hz pulse, 20 Hz Gaussian low-pass filter and a ratio of peak-to-average of 4) RF EMF (spatial peak absorption of 2 W/kg averaged over 10 g) 2 weeks apart.
The topographical analysis of EEG power during all-night non-rapid eye movement sleep revealed: (1) exposure-related increases in delta-theta frequency range in several fronto-central electrodes; and (2) no differences in spindle frequency range. We did not observe reproducible within-subject RF EMF effects on sleep spindle and delta-theta activity in the sleep EEG and it remains unclear whether a biological trait of how the subjects' brains react to RF EMF exists.
A number of studies have consistently shown that near head exposure to pulse-modulated radiofrequency electromagnetic fields (RF EMF), such as those emitted by mobile phones, can alter brain physiology during sleep and waking [Kwon and Hämäläinen, 2011]. Even though most published studies vary in terms of exposure conditions (pulse-modulation frequencies, field strength, exposure duration, exposure side, etc.) relatively consistent effects have been observed in sleep electroencephalogram (EEG). Specifically, one of the most reliable effects observed in these studies is the increase of EEG power in the sleep spindle frequency (11–16 Hz) range [Borbély et al., 1999; Huber et al., 2000; Huber et al., 2002; Loughran et al., 2005; Regel et al., 2007; Loughran et al., 2012; Schmid et al., 2012a,]. However, the frequency range affected and the observed time-course of the effects remain variable. Recent studies also found increased EEG power in the delta (1–4.5 Hz) and theta (5–8 Hz) frequency range during sleep [Schmid et al., 2012b; Lustenberger et al., 2013], an effect that might be attributable to the pronounced delta and theta components in the pulse-modulation scheme or the attenuation of higher harmonics [Schmid et al., 2012b]. These studies revealed that (1) the effect of RF EMF exposure can outlast the exposure; and (2) the pulse modulation of the RF EMF is an important factor in generating the observed effects [Huber et al., 2002; Huber et al., 2005; Regel et al., 2007].
An important characteristic of observed field effects are striking inter-individual differences within [Loughran et al., 2005; Loughran et al., 2012; Schmid et al., 2012a;] and between studies in how the subject's brain reacted to field exposure. Mobile phone-induced effects on sleep EEG are therefore difficult to investigate and are presumably statistically undetectable with a small sample size. Why certain subjects show a decrease in power of a specific frequency range after field exposure while a majority showed an increase is unknown. Understanding the origin of these individual differences may be important to investigate the mechanisms underlying effects of EMF on brain activity. Loughran et al.  addressed these inter-individual differences and showed that a group of subjects that demonstrated increased spindle activity in the first study tended to react in a similar way to RF EMF exposure when retested several years later. This result provided a first indication that exposure-related increases in sleep spindle activity might be reproducible in a specific group of subjects. However, whether individual subjects tend to react in a similar way when repeatedly exposed is still unknown.
Our study aimed to investigate inter-individual variation and intra-individual stability of the effects of pulsed RF EMF exposure on human sleep EEG. We, therefore, exposed the same subjects 2 weeks apart to the same pulse-modulated RF EMF and tested the stability of the effect within the subjects. We used the identical exposure system and RF EMF properties used in Schmid et al. [2012b].
Each participant spent four study nights (one per week on the same weekday) in the sleep laboratory of the Institute of Pharmacology and Toxicology, University of Zurich. In two of those nights, subjects were exposed for 30 min to the same RF EMF (field; FD1 and FD2) prior to their scheduled bedtime. The two other nights served as sham condition (sham; SH1 and SH2). In a randomized, double-blind crossover design, subjects had one of the two condition orders: SH1-FD1-SH2-FD2 or FD1-SH1-FD2-SH2. After real or sham exposure, a high density EEG cap was applied to the subjects resulting in a 30 min time window between end of exposure and start of sleep recording (lights out). We recorded two subjects simultaneously. Therefore, exposure times were shifted because only one exposure unit was available. Thus, bedtimes were either 10:50 p.m. to 06:50 a.m. or Prior to exposure, subjects performed a 30 min declarative learning task and conducted an implicit motor sequence learning task during exposure. Subjects were retested on these tasks in the morning (results not shown). After exposure, subjects were asked whether they perceived a field or not and completed questionnaires concerning sleep quality and mood in the morning.
The 2 Hz pulse-modulated RF signal of Schmid et al. [2012b] was applied in this study. The signal had a carrier frequency of 900 MHz with a basic modulation frequency of 2 Hz and a peak-to-average ratio of 4. Higher harmonics were reduced by applying a 20 Hz Gaussian low-pass filter to a rectangular pulse sequence (see Fig. 1 of Schmid et al. [2012b] for details of the signal characteristics). In contrast to other studies, the EEG electrodes were mounted after the RF EMF exposure, as the utilized high density EEG net would significantly increase the uncertainty of induced fields due to coupling with incident fields ....
... obtain left hemispheric 900 MHz exposure of 2 W kg−1 peak spatial SAR averaged over 10 g (psSAR10g) in the whole head. An identical box-casing was installed on the right side to prevent knowledge about the RF EMF origin. For the targeted 2 W kg−1 psSAR10g, ... the lateral cortex is exposed to a psSAR1g of up to 2.12 W kg−1 ± 12% (SD, k = 1). The corresponding 0.62 W kg-1 ± 35% (SD, k = 1) of the thalamus indicates limited RF penetration depth at 900 MHz. For a robust double blind protocol, the physical exposure system was fully computer controlled.
The present study aimed to explore topographical distribution of 2 Hz pulse-modulated RF EMF induced effects on the sleep EEG to extend the findings of an earlier study that used the same RF EMF properties [Schmid et al., 2012b]. Moreover, we aimed to investigate the inter-individual variation and intra-individual stability of these effects.
Topographical analysis revealed several significant or trend-level changes in electrodes of fronto-central regions that showed increased power in delta-theta frequency range after exposure. In the spindle frequency range, only at two electrodes, we observed a trend of an EMF exposure-related increase. However, none of these electrodes remained significant when a supra-threshold cluster analysis was applied. Our study might have been underpowered to statistically detect such subtle topographical differences. We, thus, defined a specific fronto-central cluster that contained all significant, trend-level electrodes of the delta-theta, and spindle frequency range. Interestingly, earlier studies mainly found effects in left central electrodes that were also included in our cluster. Delta-theta activity but not spindle activity was significantly higher in field condition compared to sham for this specific fronto-central cluster. Thus, the previously observed increase in EEG power between 1.25 Hz and 9 Hz in the study by Schmid et al. [2012b] was similar/reproducible in our study, though slightly lower in magnitude (5.3% vs. ∼9%). It remains elusive whether such small changes in delta-theta activity are biologically relevant and therefore behavioral consequences of small delta-theta changes should be investigated in future studies. Interestingly, the extremely low frequency magnetic field (ELF MF) exposure (2 Hz pulsed with the same characteristics as the RF envelope) used in Schmid et al. [2012b] also led to an increase in this low-frequency range whereas spindle activity was not affected. Thus, a pulse-modulation (e.g. 2 Hz) seems to be essential for an increase in delta-theta activity. Another possible explanation might be that the applied attenuation of higher harmonics restricting modulation components to the low frequency range led to an effect on delta-theta activity during sleep as discussed in Schmid et al. [2012b]. Of course, these are still speculative explanations and alternative interpretations might apply.
Spindle activity was not significantly higher in the fronto-central cluster after field exposure as would have been expected from earlier studies [Borbély et al., 1999; Huber et al., 2000; Huber et al., 2002; Loughran et al., 2005; Regel et al., 2007; Loughran et al., 2012; Schmid et al., 2012a;]. Using an established spindle detection algorithm [Ferrarelli et al., 2007], additional analysis revealed that spindle density, amplitude, duration, and power were not affected by exposure (data not shown). Even though the exposure-related increase in sleep spindle activity is one of the most robust effects reported, a number of studies failed to find exposure-related increases in sleep spindle activity likely explainable by methodological issues [Mann and Röschke, 1996; Wagner et al., 1998; Wagner et al., 2000; Hinrichs et al., 2005; Fritzer et al., 2007; Lowden et al., 2011] ...
... Loughran et al.  found the effect to be largest in females. Thus, our null finding in sleep spindle activity might be explainable by the exclusion of female subjects in our study. Hence, further studies should investigate gender specific aspects of RF EMF effects on sleep spindle activity.
Previous research indicated that exposure-related effects on sleep EEG are reproducible in a specific group of participants [Loughran et al., 2012]. We took advantage of this finding for our specific study design and further addressed whether individual subjects (in addition to a group of subjects) react in a similar way by performing correlational analysis. Thus, even though there was no overall field effect in our study in the spindle frequency range, subjects might react similarly to the field when exposed again after 2 weeks. Our data clearly showed that subjects did not react in a similar way in the first and second exposure condition since no correlation between the effects in the spindle frequency range of these two exposures was observed as revealed by the correlation analysis and the group analysis. In addition, even though delta-theta activity was significantly higher after RF EMF exposure in a fronto-central cluster, we also found no correlation or group difference between the two exposure conditions in this frequency range. Our study did not provide any evidence that some subjects were more susceptible/sensitive to our specific RF EMF. Thus, it remains unclear whether a biological trait exists that predicts how subjects' brains would react to EMF exposure.
Joel M. Moskowitz, Ph.D., Director
Center for Family and Community Health
School of Public Health
University of California, Berkeley
Electromagnetic Radiation Safety
News Releases: http://pressroom.prlog.org/