Until relatively recently, there has been little interest in the empirical exploration of insight in clinical psychiatry. However, over the last 10–15 years, increasing numbers of studies have focused on this area. Such studies have, predominantly, set out to examine the relationship between patients' insight and clinical variables, such as prognosis (McEvoy et al., 1989a; Amador et al., 1993), treatment compliance (e.g. Bartkó et al., 1988; Buchanan, 1992; Cuffel et al., 1996; Mutsatsa et al., 2003) and severity of psychopathology (e.g. McEvoy et al., 1989b; Amador et al., 1993; 1994; Carroll et al., 1999; Goldberg et al., 2001; Drake et al., 2004), and have concentrated mainly on examining this in patients with psychoses (Amador & David, 1998). More recently, a number of studies have explored the relationship between patients' insight and neuropsychological impairment (e.g. Young et al., 1993; 1998; Cuesta & Peralta, 1994; Lysaker et al., 1994; 1998a; 2002; Marks et al., 2000; McCabe et al., 2002; Mintz et al., 2004) and, indeed, structural brain lesions as assessed, e.g. by magnetic resonance imaging (MRI) (Takai et al., 1992; Flashman et al., 2001; Rossell et al., 2003). Much of this work has yielded, as will be shown in this chapter, somewhat mixed and inconsistent results. Consequently, the relationship between patients' insight and various clinical variables remains unclear (Marková & Berrios, 1995a, b).