Discriminating deception from truth and misinformation: an intent-level approach

ABSTRACT

Deception detection has been studied for hundreds of years. A particularly challenging problem is to not only identify truth from deception, but also discriminate misinformation, i.e. errors, from deception. Misinformation has generally been ignored in the study of deception detection, but through analysing the foundations of deception, it may be possible to pinpoint a fundamental difference between deception and all other benign communications – namely, the intent of the speaker. We present a detection model that captures a speaker’s intent by measuring his patterns of reasoning. The reasoning patterns of deceivers may serve as indicators of intentional deception. Through empirical studies, these intent-driven reasoning patterns can identify as well as explain deceptive communications.

KEYWORDS:

• Deception detection
• misinformation
• reasoning
• Bayesian network
• user modelling

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Asynchronous communication refers to written communication that can be planned such as reports and emails. Synchronous communication refers to more instant communication such as instant messaging and audio conferencing.

2. Recall that we refer to arguments in this paper as just the propositions or statements found in a story and not the more complex organization of multiple propositions/statements found in argumentation.

3. Resnick et al. found that predicting with highly correlated agents (a high $\mathrm{\Delta }$) can improve the performance. However, having a higher $\mathrm{\Delta }$ will result in a smaller size of correlation network. In our work, for the simplicity of the framework and to facilitate further analysis on its behaviour, we choose to use all agents in the correlation networks (i.e. $\mathrm{\Delta }=0$).

Additional information

Funding

This work was supported in part by AFOSR Grant Nos. FA9550-12-1-0457, FA9550-10-1-0499, FA9550-09-1-0716, FA9550-07-1-0050, FA9550-15-1-0383 and ONR/Naval Postgraduate School Research Initiative, Grant No. N00244-15-1-0046.