(Bruder et al., submitted for publication). Ten studies found significantly smaller P3 amplitude in depressed patients as compared to healthy controls (Blackwood et al., 1987; Muir et al., 1991; Gangadhar et al., 1993; Ancy et al., 1996; Wagner et al., 1997; Yanai et al., 1997; Anderer et al., 2002; Röschke and Wagner, 2003; Urretavizcaya et al., 2003; Kawasaki et al., 2004), whereas five studies did not (Sara et al.,1994; Bruder et al.,1998; Vandoolaeghe et al.,1998; Kaustio et al., 2002; Kaiser et al., 2003). The difference in P3 amplitude between patients and controls had small to large effect sizes (Cohen's d), which ranged widely across studies from 0.11 to 2.25. The difference in findings across studies could be related to differences in patients' clinical features, with P3 reductions being more evident in depressed patients having a melancholic depression (Gangadhar et al., 1993; Ancy et al., 1996; Urretavizcaya et al., 2003), psychotic depression (Kaustio et al., 2002; Karaaslan et al., 2003), or suicidal features (Hansenne et al., 1996).
