Pharmacological dissociation of moderate and high contextual fear as assessed by freezing behavior and fear-potentiated startle.

The amplitude of the whole-body acoustic startle response is reliably enhanced when elicited in the presence of foreground signals, such as light, previously paired with footshocks. It has been shown that this enhancement is evident by moderate fear levels, but is less affected by high fear levels. Potentiation of the acoustic startle reflex has also been reported in the presence of background cues previously associated with footshocks. However, the effects of anxiolytic drugs on different levels of fear elicited by moderate and intense contextual fear conditioning associated with startle reflex have not been examined yet. To approach this issue, we examined the effects of the anxiolytic, midazolam, on two intensities of contextual fear; freezing behavior and the startle response to loud noise. First, we compared the magnitude of the freezing behavior and the startle amplitude during the testing sessions in groups of rats submitted to fear conditioning using 0.3 and 0.6 mA as unconditioned stimuli (10 stimuli of 1 s each, intertrial interval from 60 to 180 s). Afterwards, the effects of midazolam (0.5 and 1.0 mg/kg) were assessed in these two conditions. Rats showed a potentiated startle reflex and a significant freezing behavior to moderate fear conditioning, which were both attenuated by midazolam. Higher levels of fear conditioning caused more intense freezing behavior without enhancing the startle reflex. Whereas midazolam reduced this freezing response, the startle response was unaffected. These results are indicative that anxiolytic-sensitive freezing and fear-potentiated startle are triggered by moderate contextual fear conditioning, while contextual conditioning with the use of high footshocks causes a distinct pattern of behavioral responses, which is only partially affected by midazolam. Due to the differential sensitivity to midazolam of these two patterns of startle responses generated as a function of the intensity of contextual fear conditioning, it is proposed that they represent moderate and intense aversive states that may be related to anxiety or panic/phobic conditions, respectively.