Updated meta-analysis of classical fear conditioning in the anxiety disorders.

The aim of the current study was twofold: (1) to systematically examine differences in fear conditioning between anxiety patients and healthy controls using meta-analytic methods, and (2) to examine the extent to which study characteristics may account for the variability in findings across studies. Forty-four studies (published between 1920 and 2013) with data on 963 anxiety disordered patients and 1,222 control subjects were obtained through PubMed and PsycINFO, as well as from a previous meta-analysis on fear conditioning (Lissek et al.). Results demonstrated robustly increased fear responses to conditioned safety cues (CS-) in anxiety patients compared to controls during acquisition. This effect may represent an impaired ability to inhibit fear in the presence of safety cues (CS-) and/or may signify an increased tendency in anxiety disordered patients to generalize fear responses to safe stimuli resembling the conditioned danger cue (CS+). In contrast, during extinction, patients show stronger fear responses to the CS+ and a trend toward increased discrimination learning (differentiation between the CS+ and CS-) compared to controls, indicating delayed and/or reduced extinction of fear in anxiety patients. Finally, none of the included study characteristics, such as the type of fear measure (subjective vs. psychophysiological index of fear), could account significantly for the variance in effect sizes across studies. Further research is needed to investigate the predictive value of fear extinction on treatment outcome, as extinction processes are thought to underlie the beneficial effects of exposure treatment in anxiety disorders.