Table 1.

Behavioural studies with radiofrequency (RF) fields reporting significant impairments of place learning and spatial memory.

Model	Exposure	Response	Comment	Reference
12-arm radial maze SD rat (n = 8) 250–300 g	2450 MHz, pulsed; 2 μs pulses at 500 pps 45 min/day, 10 days WBA SAR 0.6 W/kg Brain local SAR 0.5–2.0 W/kg Restrained	RF alone: more errors than sham. Pre-treatment with physostigmine or naltrexone: no difference exposed/sham. Pre-treatment with naloxone: no effect	Behaviour assessed after each daily exposure. SAR measured according to Chou et al. [7]	Lai et al. [8]
MWM SD rat (n = 11, 12) 2–3 months 250–300 g	2450 MHz pulsed; 2 μs pulses at 500 pps 60 min × 2/day, 3 days WBA SAR 1.2 W/kg Restrained	Increased escape times, no effect on speed; less time in correct quadrant during probe trial	Differences in probe trial not significant using ANOVA, but significant using Newman–Keuls post-hoc analysis. SAR measured according to Chou et al. [7]	Wang and Lai [9]
MWM SD rat (n = 8) 2–3 months 250–300 g	2450 MHz CW 60 min × 2/day, 3 days WBA SAR 1.2 W/kg Temporally incoherent magnetic noise at 6 µT Restrained	Increased escape times, less time in correct quadrant during probe trial; smaller changes after co-exposure with magnetic noise	Magnetic noise alone had no effect. SAR measured according to Chou et al. [7]	Lai [10]
MWM Wistar rat (n = 5) 3 months	Pulsed 2450 MHz ± glucocorticoid receptor antagonist RU468 3 h/day, 30 days Brain SAR 0.7 W/kg; WBA SAR 0.2 W/kg Free	RF: increased escape latency on day 4–6; RF + RU468: on day 6. RF: impaired memory RF + U468: no effects on memory	SAR calculated by calorimetry. Brain SAR seems doubtful	Li et al. [11]
MWM, activity wheel Parkes mouse (n = 5) 40 days	2450 MHz CW WBA SAR 0.03 W/kg 120 min/day, 30 days Restrained	No effect on escape latency during acquisition; less time in correct quadrant during probe trial. Phase shift in activity after 30 days, less active at night	Mice given 20 s to locate the escape platform Size of pool not reported. Modest group sizes, non-standard testing protocol. SAR calculated according to Gandhi [12]	Chaturvedi et al. [13]
MWM, 8-arm radial maze Wistar rat (n = 6) 3 months	2450 MHz, pulsed 10 µs, 800 pps WBA SAR 0.2 W/kg Brain SAR 0.7 W/kg 3 h/day for 30 days Free	Deficits in both tasks reduced by i.p. injection of glucose before each trial	SAR calculated using calorimetry	Lu et al. [14]
MWM Fischer rat (n = 6) 150–200 g	900 MHz WBA SAR 0.085 mW/kg 2 h/day, 5 days/week for 30 days Restrained	Increased time to locate, and decreased time in target quadrant in probe trial	SAR used is very low	Deshmukh et al. [15]
MWM Wistar rat (n = 10) 8 weeks	2856 MHz, pulsed 200 or 500 pps, pulse width 500 ns 6 min Average brain SAR 3.5, 7 or 35 W/kg Restrained	Increased escape latency at 6 h after exposure at 7 or 35 W/kg; and for 24 h after exposure at 35 W/kg. Reduced number of crossings of platform location in probe trial at 7 W/kg and 35 W/kg	Highest SAR caused a rise in brain temperature of 1.2 °C and in rectal temperature of 0.6 °C of anesthetized rats. SAR calculated using FDTD methods	Wang et al. [16]
MWM Wistar rat (n = 15) 200 g	2856 MHz, pulsed 500 pps pulse width 500 ns 5 min WBA SAR 14 W/kg Restrained	Increased escape latency 1 day, 2, 3 and 7 days after exposure. Non-significant increase after 6 h, 4 and 14 days.	Rise in body temperature of 0.3 °C. No probe trial. SAR calculated using FDTD methods	Qiao et al. [17]
MWM Swiss albino mouse (n = 6) 6–8 weeks	10 GHz WBA SAR 0.18 W/kg 2 h/day, 30 days Restrained	Increased escape latency	Two mice exposed together in same cage. No probe trial. SAR calculated according to Durney et al. [18]	Sharma et al. [19]
MWM Wistar rat (n = 15) 8 weeks	2856 MHz, pulsed 500 pps, pulse width 500 ns 6 min Brain SAR 35 W/kg WBA SAR 15 W/kg Restrained	Increased escape latency up to 18 months after exposure	Rise in brain temperature of 1.2 °C and in rectal temperature of 0.6 °C of anesthetized rats. No probe trial. SAR calculated using FDTD methods	Wang et al. [20]
MWM Wistar rat (n = 15) 4 weeks	2.856 GHz 6 min × 3 per week, 6 weeks PD 5, 10, 20 or 30 mW/cm2 (50, 100, 200, 300 W/m2) Restrained	Escape latency increased at 5 mW/cm2 at 14 days, at 10 mW/cm2 at 4, 14, 28 days and at 20 and 30 mW/cm2 at 3, 4, 14, 28 days after exposure. All exposed groups spent less time in the target quadrant in probe trial 5 days after exposure, and escape latency increased at 14 days after exposure	WBA SAR 1.5, 3, 6 or 9 W/kg estimated from Wang et al. [16,20]	Li et al. [21]
MWM, OFA, EPM, tail suspension, forced swim CD-1 mouse 5 weeks old	9.417 GHz 200 V/m (SAR 2 W/kg) 12 h/day from gestational day 3.5 to 18 Free	Impaired learning and memory only in male mice. Increased anxiety and decreased depression in males and females	Basis of SAR calculation not given	Zhang et al. [22]
MWM Fischer 344 rat (n = 6) 180 days	900, 1800 or 2450 MHz 2 h/day, 5 days/week for 90 days WBA SAR 0.59, 0.58, 0.67 mW/kg Restrained	All exposures impaired performance in probe trial, with increased time to target quadrant and decreased time in quadrant	Also increased HSP70 levels and increased DNA strand breaks Identical results published by Deshmukh et al. [23] following exposure for 90 days	Deshmukh et al. [24]
MWM SD rat (n = 9) 220–250 g	900 MHz CW 3 h/day for 14 or 28 days Average SAR in the head 2 W/kg, WBA SAR 0.016 W/kg. Restrained	No effects on learning, exposure for 28 days significantly impaired memory	Ultrastructural changes and increased serum albumin leakage. Effects attributed to changes in mkp-1/ERK pathway	Tang et al. [25]
MWM Swiss mouse (n = 20) 12 weeks	2450 MHz CW 2 h/day for 15, 30 for 60 days WBA SAR 0.0146 W/kg Restrained	Significantly increased escape latency during acquisition, impaired memory in probe trial. Deficits increased with increasing exposure time	No effect on rectal temperature. Exposure time-dependent changes in neuronal morphology, apoptosis, oxidative state	Shahin et al. [26]
MWM Wistar rat (n = 25) 200 g	1500 MHz, 2856 MHz or both sequentially 6 min/frequency WBA SAR 1.8/1.7 W/kg or 3.7/3.3 W/kg Free	Escape latency increased only at higher SAR (both frequencies). No increased effect with sequential exposure	Skin temperature increased by <1 °C (n = 4). Changes in EEG and hippocampal morphology at higher SAR	Tan et al. [27]
MWM Wistar rat (n = 15) 8 weeks	2586 MHz, pulsed 6 min/day, 5 days/week for 6 weeks Brain SAR 1.7, 3.5 or 7 W/kg Restrained	Effects only at 7 W/kg: escape latency increased; impaired memory in probe trial.	No change in measured body temperature (n = 10). Changes in EEG, NMDAR, hippocampal structure	Wang et al. [28]
MWM Swiss mouse (n = 6) 14 days	1000 MHz 2 h/day for 15 days WBA SAR 0.179 W/kg Lightly restrained	Deficits in learning and memory when tested at 6 weeks of age	Biochemical and histological changes	Sharma et al. [29]
Radial arm maze, 4/8 task Swiss mouse (n = 20) 12 weeks	2450 MHz CW 2 h/day for 15, 30 or 60 days WBA SAR 14.6 mW/kg	Exposure-time dependent increase in errors in working and reference memory	No increase in rectal temperature	Shahin et al. [30]

Abbreviations: ANOVA: analysis of variance; CW: continuous wave; EPM: elevated plus maze; FTDT: finite-difference time-domain; GSM: Global System for Mobile communication; i.p.: intraperitoneal; MWM: Morris water maze; NMDAR: N-methyl-D-aspartate receptor; OFA: open field arena; PD: power density; SAR: specific energy absorption rate; SD: Sprague Dawley; WBA: whole-body average. “No effects” means no statistically significant effects. “Restrained” means that the animals were held immobile during exposure, and “free” means that the animals were free to move during exposure. The age and/or weight of the animals is given at the start of exposure.