Comments on the use of the startle reflex in psychopharmacological challenges: impact of baseline startle on measurement of fear-potentiated startle

Introduction
A large body of experimental literature indicates that the startle reflex is a
sensitive measure of fear and anxiety. In both humans and animals, the startle
reflex is increased when the startle-eliciting stimulus is presented in the presence of a cue that signals an aversive event, such as a shock (Davis and
Astrachan 1978; Grillon et al. 1991). In animals, drugs that reduce anxiety in
humans, such as diazepam and buspirone, reduce this so-called fearpotentiated
startle effect (Davis et al. 1993). This finding suggests that the fear-potentiated startle methodology is a useful procedure to investigate the anxiolytic properties
of drugs in humans. However, the emerging literature on the effects of
benzodiazepines on fear-potentiated startle to threat in humans has been contradictory. Benzodiazepines have been shown to reduce fear-potentiated startle in some studies (Bitsios et al. 1999; Riba et al. 2001), but not in others
(Riba et al. 1999; Baas et al. 2002). In all these studies, fear-potentiated startle was expressed as a difference score between startle amplitude in the threat
condition minus startle amplitude in a no threat control condition. The objective
of this paper is to call attention to the issue that difference scores may not always be appropriate to express fear-potentiated startle. A problem with
difference scores is that they may be dependent on baseline startle levels,
which are themselves affected (reduced) by the sedative side-effect of benzodiazepines (Rodriguez-Fornells et al. 1999). Possibly, a reduction of
absolute fear-potentiated startle by benzodiazepines could be due to a non-specific (i.e. non-anxiolytic) effect on baseline startle reactivity.