Study Shows That Criminals Can Beat Some Lie Detectors with ‘Selective Memory’

This article came up in my news feed this morning, and I thought it was interesting in light of the recent sentencing of a man to 8 months in prison or teaching people how to beat polygraphs.

line-o-rama-the-simpsons-moe-szyslak-20091119034544332-000

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

from the journal Biological Psychology, September 2013

1-s2.0-S0301051112X00103-cov150h f

Zara M. Bergström, Michael C. Anderson, Marie Buda, Jon S. Simons, Alan Richardson-Klavehn, Intentional retrieval suppression can conceal guilty knowledge in ERP memory detection tests, Biological Psychology, Volume 94, Issue 1, September 2013, Pages 1-11, ISSN 0301-0511, http://dx.doi.org/10.1016/j.biopsycho.2013.04.012.

The paper’s section on the study’s practical implications provides a good plain-language summary of the researcher’s aim, “to experimentally demonstrate people’s capacity to control memory-related brain activity, we manipulated whether our volunteers should retrieve or suppress crime memories.” The study was conducted using Event Related Potentials, ERP, a form of “lie detector” in use for thirty years (The Truth Will Out: Interrogative Polygraphy (“Lie Detection”) With Event-Related Brain Potentials)

What the research demonstrated is that the results of this memory-based lie-detector test used to determine “guilty knowledge” by reading electrical activity in the brain associated with a person accessing actual memories can be manipulated by test subjects who have a reason to suppress memories related to their guilt. “[T]he presence or absence of memory-related brain activity appears to primarily track whether that person is having a subjective experience of remembering (cf. Allen and Mertens, 2009 and Rissman et al., 2010).

The authors’ full conclusion:

“An absence of memory-related brain activity to reminders thus indicates that suspects are not remembering an associated crime at that specific time, but does not determine that they have no such crime memories stored in their brain. An innocent verdict in a guilty knowledge test could arise because a suspect is truly innocent, because they have forgotten the particular details of the crime that are being tested, or because they are highly motivated to disguise their knowledge and are intentionally suppressing crime memories.

Of course, in a real crime setting, a suspect is more likely to attempt to suppress than intentionally retrieve crime memories. In contrast, volunteers participating in lab-based studies on guilty knowledge testing are likely more cooperative than real criminal suspects. Typical research volunteers without countermeasure instructions will have little motivation to suppress retrieval, and may even intentionally retrieve crime memories in response to demand characteristics (e.g. Orne, 1962). Such cooperation from volunteers would lead to an over-estimation of the test’s ability to detect memories in a real crime setting.

Nevertheless, despite methodological differences, the current research has important practical implications because it challenges the assumption that memory-related brain activity is outside of voluntary control, and hence more resistant to countermeasures than other physiological and behavioural measures (e.g. Lykken, 1998; see Ben-Shakhar, 2011). Our findings showed an unprecedented degree of success at intentional control over memory-related brain activity, with little special training, and that this control can be marshalled as an effective countermeasure during memory detection tests. These results thus demonstrate that in principle, retrieval suppression poses a challenge to guilt detection tests that rely on brain-activity markers of memory.

The ability of uncooperative suspects motivated to disguise their guilt to suppress memory-related neural activity established here raises concerns regarding the validity of ERP-based memory detection tests as a means to establish criminal guilt or innocence.”

 

Fig. 1. Group average mid-parietal ERPs and scalp maps contrasting different item types within blocks in Experiment one (left column) and Experiment two (right column). Topographic maps show the mean difference between probes and irrelevants between 450 and 800 ms.
Fig. 1. Group average mid-parietal ERPs and scalp maps contrasting different item types within blocks in Experiment one (left column) and Experiment two (right column). Topographic maps show the mean difference between probes and irrelevants between 450 and 800 ms.
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s