The False Truth

How to detect the events that are wrongly encoded as true in our minds.

Posted Jun 17, 2019

This post was authored in collaboration with Sara Boni, a medical student at Ferrara University in Italy.

How often do you lie? If your answer is “rarely” or “never,” this indicates that you are lying even now! We are not questioning your honesty. We believe that you, in most cases, answer questions in ways that you consider to be true, but deception is not always a conscious and deliberative process.

Research showed that lying, like many other behaviors and decisions, can be also generated automatically and with little consciousness and awareness. In particular, studies showed that a person can also “lie” because he or she recalls something that did not happen or it happened differently from the way it actually occurred.

Not convinced? Read the list of words reported below:

Grass, flowers, rain, green, water, earth, mountains, air, clouds, trees, leaves, walking, stars, sky.

Did you read the word “nature”? Your answer almost certainly is yes!

Studies indeed revealed that the majority of the people tend to recall the word “nature” even if it was not presented in the list, only because it is semantically connected to all the other words reported. This phenomenon is known as a “false memory” and described as “a falsification of remembrance in good faith”.

But, how can we detect the truthfulness of an event if this may be wrongly encoded as true in our mind?

Recently, researchers developed a new tool called the autobiographical Implicit Association Test (aIAT). The aIAT is a modified version of the well-known Implicit Association Test (IAT) that allows researchers to unconsciously assess whether an event is true for a given individual. For example, in an aIAT evaluating whether an individual stole a CD or not, respondents are asked to classify sentences representing four categories.

Two categories are logical categories and are represented by statements that are always true (e.g., I live in Boston) or always false (e.g., I live in New York) for them. Two other categories are represented by alternative versions of an autobiographical event (e.g., I stole a CD vs. I did not steal a CD), only one of the two being true. The aIAT includes two response conditions. In one condition, respondents classify sentences related to the event “I stole a CD” and to the dimension true with one response key, while classifying sentences related to the event “I did not steal a CD” and to the dimension false by using another response key.

In another condition, respondents use one response key to classify sentences related to the event “I stole a CD” and to the dimension false, and another response key to classify sentences related to the event “I did not steal a CD” and to the dimension true. Faster latencies in the former condition compared to the latter one, indicates a recognition of the event “I stole a CD” over “I did not steal a CD” as true.

The aIAT showed good accuracy in detecting true autobiographical events when those are really happened (about 91%) and also discriminated them from those memories that are erroneously recognized as truthful.

In one study, participants were presented with a Deese-Roediger-McDermott (DRM) task, i.e. a paradigm developed to produce false memories. First, they hear a list of words semantically associated (e.g., butter, food, eat, sandwich and rye) and then they administered a recognition test. The recognition test included words presented in the previous list (targets) and words that were non-presented but semantically associated (critical lures, e.g. bread).

After five minutes, participants were asked to perform two aIATs. The first aIAT (true-memory aIAT) evaluated the association of the targets (true memories) with the dimension “true,” whereas the second aIAT (false-memories aIAT) assessed the association of the critical lures (false memories) with the dimension “true.” Results showed that the aIAT detected a greater association of targets (i.e., presented words) than critical lures with the dimension “true,” indicating that although true and false memories are indistinguishable at the explicit level, the aIAT was able to detect at the implicit level a different association of these memories with the dimension “true”.

Human memory is prone to various kinds of distortions and illusions. Among these memory errors, people can incorrectly claim that they have recently experienced an event (e.g., seen or heard something) that never happened. This phenomenon, in contrast to deception, is often accompanied by a feeling of recollection which, at the explicit level, makes them indistinguishable from true memories. Recent research, however, seems to suggest that promising experimental tools may be used to address this issue and allows scientists to distinguish true and false memories at the unconscious level.


Marini, M., Agosta, S., Sartori, G. (2016). Electrophysiological Correlates of the Autobiographical Implicit Association Test (aIAT): Response Conflict and Conflict Resolution. Frontiers Human Neuroscience, 10:391.

Roediger, H.L.,andMcDermott, K. B.(1995). Creating false memories: remembering words not presented in list. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 803–814.

Sartori, G., Agosta, S., Zogmaister, C., Ferrara, S.D., Castiello, U. (2008). How to accurately detect autobiographical events. Psychological Science, 19 (8), 772–780.

More Posts