Abstract Today, the gendarmerie in France is equipped with a specialist service to help with inquiries, the Department of Behavioural Science. This department assists with inquiries involving homicide and serial rape by looking purely at the psychological, and focusing in particular upon a suspect's behaviour during police interviews. It is in this constrained environment where a suspect is most likely to lie or clam up that the interviewer must obtain the truth from a suspect, a truth which can only be subjective. In preparation for a hearing, a criminological evaluation of the facts and a study of the personality of the suspect are conducted. The frequent absence of an element of proof in the case handled means that adopting a more rigorous approach including the psychological profile of the suspect is necessary. It is therefore crucial that between the interviewer and the suspect a relationship in which confidences are exchanged is cultivated. In this context, the search for the truth depends upon the detection of a lie. But lies can only be detected by those who help an inquiry. Currently, the search for the truth is still very often in convincing a suspect to confess, the aim being for the interviewer to ‘make the suspect talk’. Although the truth, it can be argued, is independent of a confession. In fact, to base evidence upon the lies, contradictions and inconsistencies in a case can sometimes provide more proof than the confession itself. To receive a truthful interview and confession from a suspect in a limited amount of time to provide evidence for a case is a difficult exercise. Knowing the dynamic of a case, it seems illusory to look for an absolute truth, especially during an interview with a suspect. It would be better to open up the possibilities without making errors and so that the truth may be eventually found.