Intentional retrieval suppression can conceal guilty knowledge in ERP memory detection tests

Zara M Bergström,Michael C Anderson,Marie Buda,Jon S Simons,Alan Richardson-Klavehn

doi:10.1016/j.biopsycho.2013.04.012

Abstract

Brain-activity markers of guilty knowledge have been promoted as accurate and reliable measures for establishing criminal culpability. Tests based on these markers interpret the presence or absence of memory-related neural activity as diagnostic of whether or not incriminating information is stored in a suspect's brain. This conclusion critically relies on the untested assumption that reminders of a crime uncontrollably elicit memory-related brain activity. However, recent research indicates that, in some circumstances, humans can control whether they remember a previous experience by intentionally suppressing retrieval. We examined whether people could use retrieval suppression to conceal neural evidence of incriminating memories as indexed by Event-Related Potentials (ERPs). When people were motivated to suppress crime retrieval, their memory-related ERP effects were significantly decreased, allowing guilty individuals to evade detection. Our findings indicate that brain measures of guilty knowledge may be under criminals’ intentional control and place limits on their use in legal settings.

Highlights

Recent suggestions that technological advances allow us to decode criminal guilt from brain activity data have generated intensive interdisciplinary debate within the scientific community (Garland & Glimcher, 2006; Greely & Illes, 2007; Sip, Roepstorff, McGregor, & Frith, 2008; Wolpe, Foster, & Langleben, 2005)
The validation test confirmed that crime objects were strongly encoded, and that presenting these crime probes in our memory detection test would be expected to strongly and reflexively elicit retrieval of the associated objects stored in memory
We reduced the probability of crime reminders to make them subjectively unexpected, in order to investigate whether probability-sensitive parietal Event-Related Potentials (ERPs) effects would be amenable to suppression

Summary

Introduction

Recent suggestions that technological advances allow us to decode criminal guilt from brain activity data have generated intensive interdisciplinary debate within the scientific community (Garland & Glimcher, 2006; Greely & Illes, 2007; Sip, Roepstorff, McGregor, & Frith, 2008; Wolpe, Foster, & Langleben, 2005). Despite widespread interest and discussion, empirical data concerning the validity of these brain activity-based methods is sparse. Motivation to hide their guilt, whereas real criminals may use countermeasure strategies to avoid detection. In view of the important societal, legal and ethical implications of brain activity crime detection, it is vital to validate these methods before they are widely adopted, and, in particular, to evaluate how well they work for uncooperative suspects motivated to conceal incriminating knowledge. ERPs have gained popularity as an alternative to traditional autonomic measures in memory detection studies (Lykken, 1959; see Ben-Shakhar & Elaad, 2003, for review), partly because the rapid and process-specific brain responses reflected in ERPs are believed to be more resistant to countermeasures than other physiological and behavioural measures Researchers have challenged this assumption, showing that ERP memory detection tests may be more vulnerable than previously thought

Objectives

Methods

Results

Discussion

Conclusion