Resumo

Introduction: The ability to anticipate the timing of an event allows the organism to optimize behavior. Experimentally, it has been known that when the foreperiod (FP) between a warning cue and a target increases, so does the speed of responding to that target. This experimental phenomenon is called the “foreperiod effect” and refers to the increasing conditional probability that an event will occur over time. Several EEG studies have shown that a slow preparatory brain potential called the contingent negative variation (CNV) correlates with the FP effect. However, although the role of brain oscillations for sensorimotor processing has gained increased interest, the relationship between the CNV and these oscillations is still unknown. In the present experiment, we aimed to investigate the relationship between the foreperiod effect, the CNV and brain oscillations. Methods and Results: EEG was recorded continuously from 12 healthy, right-handed participants using 34 Ag/AgCl electrodes at 1000Hz. At the beginning of each trial, a grey square (warning signal) appeared and remained on a grey screen. After a FP, the grey ring became filled with a blue target. Participants were instructed to respond as quickly as possible to the target using a button-press response. Foreperiods of 1250 ms, 2250 ms and 3250 ms were randomized from trial to trial, according to a probability distribution, which remained constant throughout the block. Two probability distributions were used: U-Shape and Negative Skewed. In the U-Shape distribution, the probability of the target being presented after 1250, 2250 and 3250 was of 0.45, 0.10 and 0.45 respectively. In the Negative Skewed distribution, the target had a probability of 0.10, 0.45 and 0.45 of being presented after each FP. The mean reaction time (RT) for each FP and condition were submitted to a 2X3 ANOVA with factors Probability Distribution and FP. We found a significant main effect of FP (F(2, 22)=80.703, p<0.001), Prob. Distribution (F(1,11)=8.512, p<0.05) and a significant interaction (F(2,22)=11.661, p< 0.01). Further post-hoc tests revealed a significant difference between Probability Distributions only on the shortest FP (p<0.05). The CNV results revealed a similar pattern of effects at the Cz electrode. To investigate the brain oscillations underlying this effect, a continuous Morlet wavelet time–frequency decomposition was applied to all electrodes for each trial. We compared the Theta power (4 to 8 Hz) at electrode Cz for the two first FPs and the two Probability Distributions. We found a effect of the FP (F(1,11)=14.96, p=0<0.01) and a interaction between the factors (F(1,11)=5.20, p<0.05). Further tests revealed that the theta power was significantly higher in the U-Shape than in the Neg. Skew for the first FP (p<0.05) but not for the second. A similar pattern was found for the phase locking values between trials. We further found a phase-power coupling between Theta (5 and 6 Hz) and Beta power (15-30 Hz) that was modulated by both, the FP and the Probability Distribution. Conclusions: Our findings support the hypothesis that increasing temporal expectations can also modulate brain oscillations, especially at low frequencies. Moreover, it suggests that slow preparatory brain potentials as the CNV result from low frequency brain oscillations that can drive activity in higher frequencies.