法医学杂志 ›› 2019, Vol. 35 ›› Issue (3): 295-299.DOI: 10.12116/j.issn.1004-5619.2019.03.006

Previous Articles     Next Articles

Polygraph Accuracy of Control Question Test in Criminal Cases

SUN Zhen-yu1, ZHANG Fan2, ZHANG Fei-xia2, LUO Rong2, MAO Yuan-yi2, HU Ze-qing1, GU Yan1   

  1. 1. West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University, Chengdu 610041, China; 2. Criminal Detachment of Chengdu Public Security Bureau, Chengdu 610000, China
  • Online:2019-06-25 Published:2019-06-28

Abstract: Objective To explore polygraph accuracy of Control Question Test (CQT)and whether it could be influenced by examinee’s education level and type of violation of law. Methods Real cases of CQT (n=104) and the data from MAO (n=296) were collected. The polygraph accuracy of CQT was calculated. Variance analysis on three groups of different education levels was used to compare their age, and then the chi-square test was employed to compare polygraph accuracy among the groups. Independent sample t test was used to compare the age of subjects in the two groups of different types of violation of law, and then chi-square test was used to compare the true positive rate and true negative rate of lie detection after integration. Results In CQT lie detection of criminal cases, the true positive rate was 87.00%, the false negative rate was 13.00%, the true negative rate was 82.20%, and the false positive rate was 17.80%. There was no statistical significance in the differences between the true positive rate and the true negative rate (P>0.05). In CQT lie detection of the groups of different education levels, there was no statistical significance in the differences between the true positive rates (P>0.05) while the differences between the true negative rates had statistical significance (P<0.05). There was no statistical significance in the differences of both the true positive rates and the true negative rates between the violent violation of law and non-violent violation of law (P>0.05). Conclusion There is no significant difference between the efficiency of CQT lie detection of identifying criminals and excluding innocents. However, a comparatively high false positive rate and false negative rate still exist. The efficiency of CQT lie detection identifying criminals may not influenced by the examinee’s education level and type of violation of law, but its efficiency of excluding innocents may be influenced by the examinee’s education level.

Key words: forensic psychiatry, lie detection, Control Question Test (CQT), criminal investigation