Chun-Wei Hsu
University of Plymouth
Title: The effect of mental countermeasures on fMRI-based concealed information tests
Biography
Biography: Chun-Wei Hsu
Abstract
Concealed information paradigms (CITs) have been developed to determine if an individual is familiar with a certain piece of information such as a crime-related item. The main logic of CITs is that recognition of an item of interest (probe) will generate a differential response, compared to suitable control items (irrelevants) that can be detected by monitoring behavioral, psychophysiological, or neural variables.
An important issue is an extent to which countermeasures used by suspects can reduce the accuracy of the CIT. Recent work has focused on neural variables measured with functional magnetic resonance imaging (fMRI) because at first sight, such variables may seem more resistant to countermeasures than more peripheral variables. Previous work has shown that hybrid physical and mental countermeasures can decrease the accuracy of fMRI-based CITs, but questions remain as to whether purely mental countermeasures can do so as well. Existing evidence shows that attentional and memory strategies can decrease the accuracy with which one can use fMRI to detect successful recognition in standard face recognition tasks.
The aim of this fMRI study was to determine if such mental countermeasures are effective also with standard CITs. Participants (N=20) were tested under three conditions: no knowledge, concealed knowledge, and countermeasures. Results based on regions of interest defined in previous CIT studies showed that the area under the curve (AUC) for discriminating no knowledge and concealed knowledge cases with multi-voxel pattern analyses was 0.86 without countermeasures. Critically, memory and attentional countermeasures significantly reduced the AUC to 0.74.
These results indicate that purely mental countermeasures can reduce the accuracy of fMRI-based CITs, even without extensive training of participants.