Truth Bias and Regression toward the Mean Phenomenon in Detecting Deception

Abstract

A 2009 study by Masip, et al. contended that the truth bias appears in brief communications. They demonstrated a strong truth bias when truth-lie judgments were made at the beginning of the judged statement. Over time, a decrease in the truth bias and an increase in accuracy were observed. The improvement was explained by systematic information processing. The present paper suggests an alternative explanation, which rests on the phenomenon of regression toward the mean.

Regression toward the Mean or Heuristic Processing in Detecting Deception?: Reply to Elaad (2010)

“…This may also have limited the number of lie judgments that the observers made. For these reasons, the Moment 1 truth bias reported by Masip, et al (2009) is not as surprisingly extreme as Elaad (2010) suggests.…”

“…The reason is that some participants in the treatment group may have had extreme values on the first measurement due to random fluctuations, i.e., "a non-systematic variation in the observed values around a true mean (e.g., random measurement error, or random fluctuations in a subject)" (Barnett, van der Pols, & Dobson, 2004, p. 215, emphasis added). Although a cutoff point was not established to select Masip, et al's participants (2009), Elaad (2010 seems to suggest that many of the participants made truthfulness judgments at Moment 1 only because of random factors.…”

“…Recently, Elaad (2010) contended that as Masip, et al (2009) did not measure information processing specifically, their interpretation was questionable. Elaad offered an alternative explanation for the findings.…”