Telling lies: the irrepressible truth?

Abstract

Telling a lie takes longer than telling the truth but precisely why remains uncertain. We investigated two processes suggested to increase response times, namely the decision to lie and the construction of a lie response. In Experiments 1 and 2, participants were directed or chose whether to lie or tell the truth. A colored square was presented and participants had to name either the true color of the square or lie about it by claiming it was a different color. In both experiments we found that there was a greater difference between lying and telling the truth when participants were directed to lie compared to when they chose to lie. In Experiments 3 and 4, we compared response times when participants had only one possible lie option to a choice of two or three possible options. There was a greater lying latency effect when questions involved more than one possible lie response. Experiment 5 examined response choice mechanisms through the manipulation of lie plausibility. Overall, results demonstrate several distinct mechanisms that contribute to additional processing requirements when individuals tell a lie.

Figure 1. Example of trials in Experiment 1: a) Directed, b) Choice.

Figure 2. Response times of Experiment 1 as a function of type of instruction and honesty.

Note: Error bars are standard error.

Figure 3. Example of trials in Experiment 2: a) Directed, b) Choice.

Figure 4. Response times of Experiment 2 as a function of type of instruction and honesty.

Note: Error bars are standard error.

Figure 5. Example of a three-color lie trial from Experiment 3.

Figure 6. Response times of Experiment 3 as a function of number of response possibilities and honesty.

Note: Error bars are standard error.

Figure 7. Response times of Experiment 4 as a function of number of response possibilities and honesty.

Note: Error bars are standard error.

Figure 8. Response times of Experiment 5 as a function of truthful color and honesty.

Note: Error bars are standard error.