We wholeheartedly agree with Dike et al that the definition of ‘pathological liar’ is vague and confusing. Although pathological lying has been defined in several different ways, no specific psychological test is available. Hence we applied a symptom-based approach and defined individuals as ‘liars’ if they fulfilled: (a) criteria for pathological lying on the Psychopathy Checklist - Revised (PCL-R), (b) criteria for conning/manipulative behaviour on the PCL-R, (c) the deceit-fulness criterion for DSM-IV, or (d) criteria for malingering as reported in a self-report crime interview.

We maintain that our study did investigate at least one form of pathological lying. In a new analysis, we found that 42% of our liars had psychopathy, antisocial personality disorders or borderline personality disorder. These liars likely correspond to those Healy & Healy (Reference Healy and Healy1926) refer to as ‘secondary pathological liars’ - people whose lying is a complication of disorders such as those above. The other 58% of our group, who did not meet this comorbid requirement, probably correspond to the ‘primary pathological liars’ described by Healy & Healy - people who habitually lie but do not demonstrate symptoms of a clearly defined psychiatric disorder. This new analysis also revealed that liars with or without psychiatric disorders showed significantly increased prefrontal white matter volume compared with antisocial controls (P=0.003, P=0.01, two-tailed respectively) and normal controls (P=0.005, P=0.014 respectively). Although our study is a preliminary attempt to reveal brain abnormalities in people who lie, cheat and deceive we hope that it will stimulate interest in this important but understudied phenomenon.