From junior to senior Pinocchio: A cross-sectional lifespan investigation of deception

Full text

Running head: DECEPTION ACROSS LIFE
From junior to senior Pinocchio: A cross-sectional lifespan investigation of deception
Evelyne Debey 1, Maarten De Schryver 1, Gordon D. Logan 2, Kristina Suchotzki 1, & Bruno
Verschuere 1,3,4
1. Faculty of Psychology and Educational Sciences, Department of Experimental-
Clinical and Health Psychology, Ghent University, Ghent, Belgium
2. Department of Psychology, Vanderbilt University, Nashville, USA
3. Faculty of Social and Behavioural Sciences, Department of Clinical Psychology,
University of Amsterdam, Amsterdam, The Netherlands
4. Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The
Netherlands
Reference: Debey, E., De Schryver, M., Logan, G. D., Suchotzki, K., & Verschuere, B.
(2015). From junior to senior Pinocchio: A cross-sectional lifespan investigation of deception.
Acta Psychologica, 160, 56-68.
Correspondence concerning this article should be addressed to Bruno Verschuere, Faculty of
Social and Behavioural Sciences, Department of Clinical Psychology, University of
Amsterdam, Amsterdam, The Netherlands. E-mail: B.J.Verschuere@uva.nl. Tel.: +31 (0)2
05256799.

1
Abstract
We present the first study to map deception across the entire lifespan. Specifically, we
investigated age-related difference in lying proficiency and lying frequency. A large
community sample (n = 1005) aged between 6 and 77 were surveyed on their lying frequency,
and performed a reaction-time (RT) based deception task to assess their lying proficiency.
Consistent with the inverted U-shaped pattern of age-related changes in inhibitory control that
we observed in a stop signal task, we found that lying proficiency improved during childhood
(in accuracy, not RTs), excelled in young adulthood (in accuracy and RTs), and worsened
throughout adulthood (in accuracy and RTs). Likewise, lying frequency increased in
childhood, peaked in adolescence, and decreased during adulthood. In sum, we observed
important age-related difference in deception, that generally fit with the U-shaped pattern of
age-related changes observed in inhibitory control. Theoretical and practical implications are
discussed from a cognitive view of deception.

2
Research Highlights
(3-5, each max 85 characters)
1. This is the first study to examine deception across the entire lifespan [76].
2. We tested 1005 Science museum visitors from 6 to 77 years [61].
3. Adolescents tell most lies, children and the elderly the least [65].
4. Inhibitory capacity matures from childhood to young adulthood, declines thereafter [85]
5. We observed important age-related differences in lying ability [65].

3
From junior to senior Pinocchio: A cross-sectional lifespan investigation of deception
Introduction
Executive control and lying
The ubiquity of lies in everyday life does not imply that lying is child’s play. Research
has shown that lying considerably challenges our cognitive capacities (Vrij & Granhag, 2012;
Walczyk, Igou, Dixon, & Tcholakian, 2013). The cognitive load that accompanies lying is, for
instance, reflected in slower response times and a higher number of errors, compared to truth
telling (i.e., lie effects; e.g., Van Bockstaele et al., 2012; Williams, Bott, Patrick, & Lewis,
2013; but see Suchotzki, Verschuere, Crombez, & De Houwer, 2013). Furthermore, lying
evokes more activity in the prefrontal cortex, a brain region that is crucially linked to
cognitive or executive control (Abe, 2011; Christ, Van Essen, Watson, Brubaker, &
McDermott, 2009; Farah, Hutchinson, Phelps, & Wagner, 2014). Miyake and colleagues
distinguished response inhibition, working memory updating, and shifting as the three main
executive functions (Miyake et al., 2000), and several lines of research support the
involvement of these functions in lying. Evidence has been found for the notion that the truth
response is activated first, thereby inducing response conflict and an increased need for
response inhibition to prevent the truth from slipping out (Debey, Ridderinkhof, De Houwer,
& Verschuere, 2014; Duran, Dale, & McNamara, 2010; Hadar, Makris, & Yarrow, 2012;
Vartanian et al., 2013). However, the truth may initially also be kept active in working
memory to help the formulation of an alternative, deceptive response (Ambach, Stark, &
Vaitl, 2011; Debey, De Houwer, & Verschuere, 2014; Visu-Petra, Miclea, & Visu-Petra,
2012). Finally, shifting may help to flexibly shift between the mental sets associated with
truthful and deceptive responses (Visu-Petra, Miclea, & Visu-Petra, 2012; Visu-Petra, Varga,
Miclea, & Visu-Petra, 2013).

4
Apart from a few developmental studies in children (< 17 years), deception research
has largely neglected that executive control is subject to changes throughout life that may give
rise to age-related changes in deception skills. The present study addressed this issue by
pursuing a first attempt to depict lying proficiency and lying frequency across the near entire
lifespan (targeted age range: 6–82).
Executive control across life
The relationship between age and executive control ability is best described as an
inverted U-curve: Executive control increases across childhood, peaks in late adolescence or
young adulthood, and declines thereafter (for reviews, see Craik & Bialystok, 2006; Jurado &
Rosselli, 2007; Zelazo, Craik, & Booth, 2004). Several theories have been proposed to
explain the cognitive differences at both ends of life. The frontal lobe hypothesis (West,
1996), attributes lifespan differences in executive control to specific age-related changes in
the frontal lobe, a brain region that is the last to mature and among the first regions to
deteriorate during ageing (Raz et al., 2005). The global speed hypothesis (Salthouse, 1996),
however, states that a general factor - global processing speed - determines the performance
levels that can be reached on most cognitive tasks (Bashore & Smulders, 1995). It is therefore
imperative to control for global processing speed when examining age-related changes in
executive control (e.g., Span, Ridderinkhof, & van der Molen, 2004; Williams, Ponesse,
Schachar, Logan, & Tannock, 1999).
Response inhibition across life
Response inhibition has been proposed to be the executive function that may be at the
heart of deception (Spence et al., 2001; Vartanian et al., 2013). We therefore looked more
closely into studies that examined age-related differences in inhibitory control. This literature
is characterized by three main limitations. First, whereas some studies controlled for

5
processing speed or – more general – baseline performance (e.g., Rush, Barch, & Braver,
2006; Troyer, Leach, & Strauss, 2006; Van der Elst, Van Boxtel, Van Breukelen, & Jolles,
2006; Williams et al., 1999), others did not (Collette, Germain, Hogge, & Van der Linden,
2009; Davidson, Amso, Anderson, & Diamond, 2006; May & Hascher, 1998; Schroeter,
Zysset, Wahl, & von Cramon, 2004). Second, most studies only examined inhibitory control
in a restricted age range (e.g., Bub, Masson, & Lalonde, 2006; Bunge, Dudukovic, Thomason,
Vaidya, & Gabrieli, 2002), or investigated lifespan changes in a discontinuous manner by
reducing the lifespan to a group of children, young adults, and older adults (e.g., Christ,
White, Mandernach, & Keys, 2001; van de Laar, van den Wildenberg, van Boxtel, & van der
Molen, 2011). Third, as highlighted by Band, van der Molen, and Logan (2003), many studies
are underpowered.
A limited number of studies addressed these main limitations. Van der Elst et al.
(2006; n = 1856; age range: 24–81) and Troyer et al. (2006; n = 272; age range: 18–94),
administered the Stroop task throughout adulthood. In the Stroop task, participants name the
ink color in which color words are presented. Because naming the word is the prepotent
response, response inhibition is needed on trials where the ink color is incongruent with the
word name (e.g., GREEN in red; MacLeod, 1991). Both studies showed that the Stroop
effects (i.e., incongruent minus congruent) in errors and reaction times (RTs) systematically
increased with age. Also in an adult sample (n = 304; age range: 20–86), Borella, Carretti, and
De Beni (2008) administered the Hayling sentence completion test (Burgess & Shallice,
1997), in which participants complete high-closed sentences, either with an expected word
(initiation condition) or with a word that is unrelated to the sentence content (inhibition
condition). A comparison of both conditions showed that with age, participants had more
difficulty to inhibit expected words. Williams and co-workers (1999) administered a global
stop-signal task to 275 participants that were between 6 and 81 years old. The stop-signal task

6
requires participants to perform a visual choice reaction-time task, but to withhold their
response on an infrequent number of trials (e.g., 25 %) in which a stop signal occurs shortly
after presentation of the stimulus (Verbruggen & Logan, 2008). Stop-signal reaction time
(SSRT; an estimated time to stop responses) decreased between early childhood and
adolescence, and increased again thereafter. However, the loss during adulthood was less
pronounced than the gain during childhood. Bedard and colleagues (2002), examining 317
participants aged 6 to 82, found a more symmetric U-shaped pattern with a modified stop-
signal task that required participants to suppress their response in the presence of one signal,
but not in the presence of another.
In sum, there is some evidence for an age-related inverted U-course of inhibitory
control, yet studies remain restricted in sample size, age range, and efforts to correct for
baseline performance.
Lying ability across life
From the idea that lying depends on executive control, one could infer the hypothesis
that deception ability will change with age in a similar vein as the executive functions it relies
on. So far, the relationship between age and lying has been primarily examined in studies that
focused on the development of lying in the age range between 2 and 16 years. A common
used paradigm to assess children’s lying ability is the temptation resistance paradigm (Lewis,
Stanger, & Sullivan, 1989; Talwar & Lee, 2002). In this paradigm, an experimenter asks
children not to peek at or play with a toy when left alone. Because the situation is very
tempting, many children transgress the instruction. Upon returning, the experimenter asks
whether the child has peeked at or played with the toy. If the child denies transgression,
follow-up questions (e.g., “What do you think the toy is?”) are asked to assess strategic lying,
that is, the ability to make statements that are consistent with the initial lie.
Of interest with regard to the impact of ageing on lying skills is a study of Abe and

7
colleagues (2009) in which participants with Parkinson’s disease and healthy controls were
instructed to lie or tell the truth about recognizing certain items. It was found that patients had
more difficulty to lie relative to healthy controls. The finding that this poorer performance in
patients was correlated with their prefrontal hypometabolism, points to the possibility that the
typical deterioration of the frontal lobe in old age may also hamper lying skills.
The observation that executive control ability predicts lying from the very first
appearances of lies substantiates the idea that executive control is a core component of lying.
It is therefore striking that no research has yet examined lying proficiency across the entire
lifespan.
Lying frequency across life
We also explored how lying frequency evolves over life. Several studies found that
people lie on average one to two times a day (DePaulo et al. 1996; Vrij, 2008). Some of these
studies reported upon how age impacts on lying frequency. Serota, Levine, and Boster (2010;
Study 1) surveyed 1000 U.S. adults (aged 18–65+) how many times they had lied in the past
24 hours. Participants lied on average 1.65 a day, and lying frequency decreased with aging.
The diary study of DePaulo and colleagues (1996) also found that young adults lie more often
than older adults. Levine and colleagues further found an increased lying frequency in
adolescents (aged 14–17; Levine, Serota, Carey, & Messer, 2013) as compared to the lying
frequency reported by adults in other studies. It is, however, important to note that Serota et
al. (2010; see also Halevy, Shalvi, & Verschuere, 2014) showed that the mean may provide a
biased statistic in lying frequency research: Many people reported not to lie, and a very small
proportion of the surveyed samples (i.e., ‘prolific liars’) appeared responsible for most of the
lies.
Although lying frequency has never been systematically investigated under the age of
14, developmental studies have shown that with increasing age, children become more

8
inclined to lie (Chandler, Fritz, & Hala, 1989; Talwar & Lee, 2002; Talwar, Murphy, & Lee,
2007; Wilson, Smith, & Ross, 2003).
Current study
Our primary goal was to investigate age-related changes in lying proficiency and lying
frequency. We predicted an age-related inverted U-course for both lying proficiency and lying
frequency. This prediction was based upon the previously observed age-related inverted U-
course for response inhibition capacity. As those studies were restricted in sample size, age
range, or efforts to correct for baseline performance, we also sought to replicate the age-
related inverted U-course for response inhibition capacity. Finally, we explored whether the
effect of age on lying proficiency and lying frequency would be related to changes in response
inhibition capacity. This study was run in a science museum, as a convenient means to assess
a large sample that has substantial variation in age.
We assessed inhibitory control by means of a global stop-signal task that was similar
to the one used in the lifespan study of Williams and colleagues (1999). While participants
make speeded classification judgements of two letters (X or O?), this tasks requires one to
withhold responding when an auditory stop signal is presented, allowing to estimate the time
to stop (SSRT; Verbruggen & Logan, 2008). Lying proficiency was measured by means of the
Sheffield lie test (Spence et al., 2001) that requires speeded yes/no responses to questions
(e.g., “Are you in Africa?”). A color cue on the screen instructs when to lie or tell the truth.
This paradigm has been successfully used with children (Otgaar, Verschuere, Meijer, & van
Oorsouw, 2012), adolescents (Verschuere, Spruyt, Meijer, & Otgaar, 2011), and adults
(Fullam, McKie, & Dolan, 2009; Kaylor-Hughes et al., 2011). In this task, one typically
observes significant lie effects in error rates and RTs, indicating that lying is typically slower
and more error-prone than truth telling (Spence et al., 2001; Van Bockstaele et al., 2012).
Lying frequency was assessed with the Serota self-report measure that asks for number of lies

9
told in the past 24 hours. Due to the expected non-normal distribution of lying frequency, we
will report mean frequencies, but focus on unbiased statistics (i.e., the median), and examine
whether the number of prolific liars changes with age.
Method
Participants
In August 2012, 1005 visitors of Science Center NEMO (Amsterdam, the
Netherlands), aged between 6 and 77 years were recruited in the study.1 From this original
sample, outlying participants were removed separately for the stop-signal task, the Sheffield
lie test, and the Serota questionnaire (see below for details).
Apparatus
The stop-signal task and Sheffield lie test were run on Dell latitude E5510 laptops (2.4
GHz Core i3 M730 processor; 15.6-inch color monitor). Closed headphones were used to
present the stop signals in the stop-signal task and the questions in the Sheffield lie test. The
stop-signal task was programmed and presented using the Tscope library for C/C++
programming (Stevens, Lammertyn, Verbruggen, & Vandierendonck, 2006). Stop-signal
functions were adapted from STOP-IT software (Verbruggen, Logan, & Stevens, 2008). The
Sheffield lie test was programmed and presented using Inquisit version 1.33 software.
Participants sat approximately 40 cm from the screen.
Questionnaires
A demographic questionnaire assessed gender, date of birth, mother language, and
education/profession.

10
Lying frequency was assessed with the Serota (2010) self-report lying frequency
questionnaire. Participants were asked how many times they had lied in the past 24 hours.
They had to specify the lie rate for five different receivers (family members, friends,
work/school- related contacts, acquaintances, and strangers) along two deceptions modes
(face-to-face, and other forms of deception [e.g., internet, telephone]). A sum score of the 10
mode-receiver combinations was calculated as an estimate of lying frequency.2
Tasks
Stop-signal task. The stimuli used in the stop-signal task were the uppercase letters
“X” and “O” (Courier, 48 pt., bold). They were presented in white in the middle of a black
screen. Depending on the response mapping that was counterbalanced across participants, the
letter X required a left response (“z” key) on a QWERTY keyboard and the letter O a right
response (“/” key), or vice versa. The stimuli remained on the screen for the maximum
response time of 2000 ms, followed by a blank screen of 300 ms. On a random selection of
25 % of the trials, a stop signal (750 Hz, 75 ms) was presented shortly after the stimulus
onset. The stop-signal delay (SSD; the time interval between the presentation of the stimulus
and the stop signal) was initially set at 250 ms and continuously adjusted according to a
staircase tracking algorithm (Levitt, 1970): The SSD decreased with 50 ms when inhibition
was unsuccessful, and increased with 50 ms after successful inhibition. This procedure allows
obtaining an approximate .50 probability of stopping.
The task started with two practice phases. The first practice phase (8 trials; 2 stop-
signal trials) focused on the go task. Participants were asked to respond to the letters as
quickly and accurately as possible. They were told that occasionally they would hear an
auditory signal, but were instructed to ignore it in this phase. In the second phase (16 trials; 4
stop-signal trials), the actual stop-signal task was explained. Participants were told that they
would have to try to stop their response when they heard the signal. They were also told that

11
sometimes it would be fairly easy to stop their response, whereas other times stopping would
be hard. To prevent participants from applying a waiting strategy, it was stressed that the
computer would delay the presentation of the stop signal when they started waiting for it.
After each practice phase, performance feedback was provided. When necessary, the practice
phases could be run again until the participants understood the task. After practicing,
participants performed the test phase with 160 trials (i.e., four blocks of 40 trials; 40 stop-
signal trials in total). In the middle of the task, a self-paced break was provided. In each block,
the letters X and O were presented equally often and followed equally often by a stop signal.
The task took approximately 10 minutes.
Sheffield lie test. Thirty general knowledge questions served as stimuli (see Appendix
A). Half of the questions could be truthfully answered with a “yes” response (e.g., “Is grass
green?”), whereas the other half required a truthful “no” response (e.g., “Can pigs fly?”; note
that we provide the English translation here, whereas the stimuli were presented in Dutch).
The questions were matched for length, and tested for understandability in a pilot study. On
each trial, a single prerecorded question (always produced by the same male Dutch speaker)
was presented through the headphones. At the same time, the “YES” and “NO” response
labels were presented on the screen according to the response mapping (YES = left, NO =
right, or vice versa). Both labels were presented in a blue or yellow color, and a color rule
(counterbalanced across participants) designated when to lie or tell the truth (blue = lying,
yellow = truth telling, or vice versa). From the start of the question, participants had 6 s to
give a response, and the RT reflects the time between the onset of the question and the
response. When no answer was given within the response deadline, the next question was
automatically presented. After a response, or after 6 s without a response, a blank screen of
200 ms was inserted before the next question was presented.
Before the start of the task, the experimenter provided the instructions that were also

12
presented on the screen. Participants first completed two practice phases. In the first practice
phase, 6 questions (3 yes- and 3 no-questions) that were not used in the test phase were
presented twice in a fixed order of increasing difficulty (truth/yes-questions, lie/yes-questions,
truth/no-questions, lie/no-questions). In the second practice phase, the same questions were
randomly presented. In both practice phases, a picture of an upward thumb, downward thumb,
or a clock were presented as feedback to a correct, incorrect, or too late response,
respectively. After each practice phase, a performance feedback screen enabled the
experimenters to see if the participant understood the task. If necessary, the task was
explained again and participants completed the practice phases a second time. The test phase
consisted of three blocks of 40 trials, so that each question required twice a truthful response,
and twice a deceptive response. Feedback was no longer provided after each trial. The
intertrial interval was set at 300 ms. Blocks were separated by a self-paced break. At the end
of the task, participants received feedback based on their RT lie effect (i.e., RTTRUTH – RTLIE),
classifying them as a good, mediocre, or bad liar. The task lasted approximately 10 minutes.
Procedure
Individuals who did not speak Dutch, who were younger than 6, or older than 82 were
not eligible to participate. Volunteers were shortly briefed about the experiment and signed an
informed consent. If participants were younger than 18, the informed consent was signed by
an accompanying parent. Participants were first asked to fill out the demographic and lying
questionnaires. Next, they entered a research room, where they performed the stop-signal
task, followed by the Sheffield lie test. Up to four participants could be tested simultaneously.
After completing the tasks, an oral and written debriefing was provided. Overall, the
experiment took approximately 30 minutes.
Results

13
Of 11 participants, day of birth were missing and hence their data could not be used in
the analyses. To allow for comparisons with previous lifespan studies on the stop-signal task
(Bedard et al., 2002; Williams et al., 1999), we divided participants into seven age groups (see
Table 1 for a description of the age groups).
For ANOVAs and hierarchical regressions, we calculated Cohen’s f as an effect size,
using the following formula: f = √[η2p/(1 – η2p)]. According to Cohen (1992), fs from 0.10,
0.25, and 0.40 represent small, medium, and large effects, respectively. For
t-tests, we used Cohen’s d as an effect size, with d values from 0.20, 0.50, and 0.80
representing small, medium, and large effects, respectively (Cohen, 1988). When Levene’s
test for equal variances was significant, degrees of freedom were adjusted.
In preliminary analyses, we also investigated the impact of gender. However, as
gender differences were not our primary focus, these analyses are only briefly described in
Footnote 3.
Stop-signal task
Data analysis. Because it takes a number of trials for the tracking algorithm to adjust
the SSD to the point where participants can successfully inhibit their responses on
approximately 50 % of the stop-signal trials, we discarded the first presentation block from
the analyses. The SSRT was estimated using the integration method (Logan, 1981;
Verbruggen, Chambers, & Logan, 2013). We removed RTs below 200 ms (1.40 %) and trials
with RT recording errors (0.02 %). Too late responses (1.34 %) were set at the maximum
response time of 2000 ms.
Spearman-Brown split-half reliability (i.e., odd vs. even) was high for RT on go trials
(i.e., GoRT; ρ = .99), and low for SSRT (ρ = .45). The low reliability of SSRT may be
attributed to a slowing strategy that some participants developed during the task, despite clear
instructions not to do so. Indeed, a repeated measures ANOVA on GoRT revealed a main

14
effect of Block, F(2, 1662) = 92.06, p < .001, f = 0.33, showing a gradual increase of GoRT
across blocks (ps < .001 in follow-up t-tests). This slowing strategy may also explain why the
proportion of responding given a stop signal (p[r|s]) was overall rather low (see Table 1). The
finding that p(r|s) reached .50 in another study in which participants completed the same task
(Suchotzki, Crombez, Debey, van Oorsouw, &Verschuere, 2014), supports the notion that not
task parameters, but strategy use accounts for our results. By means of simulations,
Verbruggen, Chambers, and Logan (2013) showed that in case of strategic slowing in the
stop-signal task, reliable and unbiased SSRTs can be obtained by estimating the SSRT for
smaller blocks using the integration method and then taking the average across blocks.
Accordingly, our estimation of the SSRT equals the average of the SSRTs calculated for each
of the three presentation blocks using the integration method.
We removed the data of three participants who did not (entirely) complete the stop-
signal task. Hundred forty participants were defined as outlying and excluded from the
analyses because they met at least one of the following lenient outlier criteria proposed by
Congdon et al. (2012): (1) a p(r|s) less than .25 or greater than .75, (2) no response on more
than 40 % of the go trials, (3) more than 10 % errors on go trials, and (4) an SSRT that was
negative or smaller than 50 ms. From the remaining sample, we additionally discarded 16
participants whose SSRT was 2.5 standard deviations (SDs) removed from the mean SSRT of
the age group they belonged to. Table 1 displays the remaining sample across the age groups.
Following Williams et al. (1999), we not only analyzed SSRT, but also GoRT as a measure of
global speed processing
Response execution (GoRT). A one-way ANOVA on GoRT with Age (7 age groups)
as predictor revealed a main effect of Age, F(6, 825) = 7.33, p < .001, f = 0.23. To further
examine the effect of age on GoRT, we conducted a hierarchical polynomial regression on
GoRT, where we subsequently entered the linear, quadric, and cubic functions of Age as

15
predictors (see Table 2, Analysis A). The quadratic function of Age proved to be a significant
predictor of GoRT. In line with this quadratic term, visual inspection of Figure 1 (and Table
1) shows that GoRT decreased during childhood and gradually increased again during the
course of adulthood. Planned comparisons of subsequent age groups revealed significant
decreases between early childhood and adolescence, ts ≥ 1.99, ps < .05, 0.27 ≥ ds ≤ 0.38.
Further, older adults were significantly slower than middle-aged adults, t(219.93) = 3.54, p <
.001, d = 0.43. All other comparisons were non-significant, ts ≤ 1.
Response inhibition (SSRT). A one-way ANOVA on SSRT with Age as predictor
yielded a main effect of Age F(6, 825) = 22.95, p < .001, f = 0.41. The relationship between
SSRT and age was further examined in a hierarchical polynomial regression on SSRT (see
Table 2, Analysis B). To control for baseline speed, we entered GoRT as a predictor in a first
step (Williams et al., 1999; see also McAuley et al., 2006; Rush, Barch, & Braver, 2006, for
similar control procedures in other inhibition tasks). In subsequent steps, the linear, quadratic,
and cubic functions of Age were entered as predictors. GoRT was no significant predictor of
SSRT. In contrast, the linear and quadratic functions of Age were found to significantly
contribute to SSRT. Figure 1 and Table 1 show that the SSRT decreased from early childhood
to young adulthood and thereafter increased again, albeit much more shallowly. Planned
comparisons between subsequent age groups revealed significant decreases in SSRT between
early childhood and young adulthood, ts ≥ 2.58, ps ≤ .01, 0.35 ≥ ds ≤ 0.54. In contrast, the
SSRT of subsequent age groups in adulthood did not differ from each other, ts < 1. A post-
hoc comparison between young adults and seniors showed only a trend toward an increase in
SSRT, t(127) = 1.80, p = .07, d = 0.32. Moreover, the SSRT in early childhood was
significantly larger than the SSRT of seniors, t(75.68) = 4.72, p < .001, d = 0.97.
Sheffield lie test

16
Data analysis. Trials with latencies below 300 ms (0.94 % of the original data set),
and trials without a response within the 6 s deadline (0.80 %) were discarded. For RT
analyses, we excluded trials with errors (10.58 %), and truth and lie trials with RTs that were
2.5 SDs removed from each individual’s mean RT on truth and lie trials (1.59 %).
Fourteen participants did not perform the Sheffield lie test, because they quit the
experiment during or after the stop-signal task. In an outlier analysis at participant level, we
sequentially removed (1) two participants whose mother language was not Dutch and who
clearly stated that they did not understand certain words in the task, (2) 28 participants who
had an error rate that was 2.5 SDs higher than the mean error rate of the age group they
belonged to, (3) 71 participants who had less than 60 % truth and lie trials left for the RT
analyses after the RT outlier analysis, and (4) six participants who had an overall RT that was
2.5 SDs away from the mean RT of the age group they belonged to. Table 3 shows the
distribution of the remaining sample across the age groups. Based on this sample, Spearman-
Brown split-half reliability (i.e., odd vs. even) was found to be acceptable (ρ = .68) for the RT
lie effect (see Table 3 for reliability coefficients per age group).
We first performed an overall analysis to examine potential effects of age on lying
proficiency. Mean error rates and RTs were subjected to repeated measures ANOVAs with
Deception (Lie vs. Truth) as within-subjects variable, and Age (7 age groups) as between-
subjects variable. Irrespective of the presence or absence of Age by Deception interactions
revealing age-related variations in lie effects, we aimed to draw a more accurate picture of the
relationship between age and lying by examining lying performance after accounting for the
age-related variance explained by truth telling (i.e., baseline) performance. We applied a
similar control procedure as applied in the SSRT analysis: For both errors and RTs, we fitted
a hierarchical polynomial regression on the lie value (e.g., RT on lie trials), in which we first
controlled for baseline performance by entering the truth value (e.g., RT on truth trials) as a

17
predictor, followed by the linear, quadratic, and cubic functions of Age in the subsequent
steps. We compared lying performance between age groups using the estimated lie values
after having partialled out the variance attributable to truth telling performance. Each
estimated lie value was calculated by (1) computing the predicted lie value based on a model
that included the truth value as the only predictor for the observed lie value (e.g., RT lying =
β0 + β1 . RT truth telling), and (2) subtracting this predicted lie value from the observed lie
value.
Errors. The ANOVA revealed a main effect of Deception, F(1, 866) = 315.54, p <
.001, f = 0.60, indicating the presence of an error lie effect, with more errors made on lie trials
than on truth trials (see Table 3). The main effect of Age, F(6, 866) = 49.41, p < .001, f =
0.59, pointed to age-related changes in the overall error rate. However, these main effects
were subsumed under a significant Deception by Age interaction, F(6, 866) = 2.69, p = .01, f
= 0.14, suggesting that the error lie effect differed among age groups. Planned comparisons
revealed that the error lie effect was significantly different from zero in all age groups, ts ≥
4.28, ps < .001, 0.48 ≥ ds ≤ 0.85. Visual inspection of the age-related error lie effects (see
Figure 2 and Table 3) shows a decrease between midchildhood and young adulthood, and an
increase thereafter. Comparing subsequent age groups disclosed a trend toward a smaller error
lie effect in adolescence than midchildhood, t(345) = 1.78, p = .08, d = 0.21. Further, the error
lie effect in young adulthood was significantly smaller compared to all other age groups, ts ≥
2.05, ps ≤ .04, 0.27 ≥ ds ≤ 0.56, except for adolescence, t(172) = 1.39, p = .17. The error lie
effects did not differ between the youngest children and the seniors, t < 1.
The hierarchical regression analysis revealed that the error rate on truth trials
explained a significant proportion of variance in the lie error rate (see Table 4, Analysis A).
After accounting for this baseline effect, both the linear and quadratic function of Age were
also found to be significant predictors of the lie error rate. The estimated lie error rates after

18
controlling for truth telling performance (Figure 2 and Table 3) showed a similar age-related
U-pattern as observed for the error lie effects. Planned comparisons between subsequent age
groups exposed significant decreases between midchildhood and young adulthood, ts ≥ 2.04,
ps ≤ .04, 0.25 ≥ ds ≤ 0.38. Further, there was a trend toward an increase between young- and
midadulthood, t(287) = 1.80, p = .07, d = 0.24 (other ts ≤ 1.52). However, seniors had a
significantly larger lie error rate than young adults, t(87.83) = 3.24, p < .01, d = 0.61. The lie
error rate of seniors did not differ from that of the youngest children, t < 1.
Reaction times. The ANOVA produced a main effect of Deception, F(1, 866) =
767.05, p < .001, f = 0.94, indicating the presence of an RT lie effect with slower responses on
lie trials than on truth trials (see Table 3). The main effect of Age, F(6, 866) = 32.83, p <
.001, f = 0.48, signified that overall speed varied among age groups. These main effects were
subsumed under a significant Deception by Age interaction, F(6, 866) = 24.98, p < .001, f =
0.42, suggesting that the RT lie effect differed between age groups. Planned comparisons
showed that the RT lie effect was significant in all age groups, ts ≥ 5.37, ps < .001, 0.71 ≥ ds
≤ 1.64. The RT lie effect followed an age-related S-shaped curve with an evident increase
between young- and midadulthood (see Figure 3 and Table 3). Comparisons of subsequent
age groups revealed that only this increase reached significance, t(282) = 4.99, p < .001, d =
0.66 (other ts ≤ 1.19). Additional comparisons disclosed no significant changes in the RT lie
effect between early childhood and young adulthood, ts < 1. There was a small trend toward a
larger RT lie effect in seniors than midadulthood, t(254) = 1.66, p = .10, d = 0.26.
The hierarchical regression analysis showed that after controlling for the significant
contribution of RT on truth trials, all functions of Age explained a significant proportion of
variance in RT on lie trials (see Table 4, Analysis B). Figure 3 and Table 3 display the
estimated lie RTs after accounting for truth telling performance, and confirm the age-related
S-shaped curve found for RT lie effects. Planned comparisons of subsequent age groups

19
demonstrated again that only the manifest lie RT increase between young and middle-aged
adults reached significance, t(282) = 5.10, p < .001, d = 0.68 (other ts ≤ 1.24). Post-hoc
comparisons further confirmed that there were no significant RT increases between early
childhood and young adulthood, ts < 1. The RT increase between midadulthood and seniors
was marginally significant, t(254) = 1.89, p = .06, d = 0.30.
Lying frequency
Data from one participant (a 12-year old boy) were removed because he reported an
implausible lying frequency of 315. One additional participant was excluded as data were
missing from his Serota questionnaire. Table 5 displays the remaining sample. As effect sizes
for the Mood’s median tests, we calculated Cramér’s V, for which values from .10, .30, and
.50 respectively reflect small, medium, and large effects (Cramér, 1999).
Table 5 shows that on average participants told about two lies a day, whereas the
median frequency was one lie a day. A Mood’s median test with Age as predictor showed that
the median lying frequency differed among age groups, χ2(6, N = 992) = 24.41, p < .001, V =
.06. Based on the absolute values, one could detect an increase of lying frequency during
childhood with a peak in adolescence, followed by a decrease during adulthood. Planned
comparisons showed that median lying frequency was higher in midchildhood than in early
childhood, χ2(1 , N = 397) = 6.31, p = .01, V = .13. Adolescents lied significantly more than
any other age group, 7.84 ≥ χ2s ≤ 18.91, p ≤ .01, .14 ≥ V ≤ .31, except for a trendwise
difference with young adulthood, χ2(1 , N = 184) = 3.21, p = .07, V = .13. The difference in
median frequency between young adulthood and seniors failed to reach significance, χ2(1, N =
144) = 1.31, p = .25. Further, median lying frequency was not different for early childhood
and seniors, χ2(1, N = 163) = 1.02, p = .31.
When examining the distribution of lies based on the entire sample, we found that
nearly half of the participants reported not to have lied in the past 24 hours (see Table 5). The

20
proportion of individuals who reported to have lied decreased as a function of the number of
lies, so that 50.67 % of the lies was told by 8.87 % prolific liars (see Figure 4). This non-
normal distribution was present in all age groups (see Table 5). Though a Chi-square test
showed that the proportion of prolific liars did not differ across age groups, χ2 (6, N = 992) =
8.14, p = .23, the age-related pattern was by and large similar to the reversed U-course found
in the median analysis, with a peak number of prolific liars in adolescence.
Correlational analysis
We produced a correlation matrix to examine the relationship between measures of the
stop-signal task, Sheffield lie test, and lying frequency (see Table 6). SSRT was incorporated
from the stop-signal task. From the Sheffield lie test, we included the estimated lie error rate
and estimated lie RT after accounting for the variance explained by truth telling performance.
Because lying frequency was not normally distributed, we used Spearman’s rho (ρ) as
correlation coefficient.
Evidence for a relation between executive (inhibitory) control and lying proficiency
was weak. Participants who had longer SSRTs tended to have higher estimated lie error rates,
but faster estimated lie RTs in the Sheffield lie test. However, given that values of .10, .30,
and .50 respectively reflect small, medium, and large correlations, the observed correlations
were very small. Although the analysis showed that individuals who reported a higher number
of lies also lied faster in the Sheffield lie test, that correlation can also be considered small.
Moreover, stop-signal task performance was not related to lying frequency.
Discussion
How does lying evolve over life? This was the primary question we aimed to address
in the current study. In a large sample aged 6–77, we measured accuracy and speed of lying in
the Sheffield lie test as measure of lying proficiency. The number of lies told in the past 24

21
hours provided a measure of lying frequency. Based on previous studies that highlighted the
role of inhibitory control in lying (Christ et al., 2009), and the finding that inhibitory control
typically follows an inverted U-curve with age (Jurado & Rosselli, 2007), we expected that
lying proficiency and lying frequency would follow an age-related U-curve. Because studies
examining age-related difference in inhibitory control were restricted in sample size, age
range, or efforts to correct for baseline performance, we also sought to replicate the age-
related U-curve shape in response inhibition capacity in a Stop-signal task.
Response inhibition capacity across life
Concentrating first on the stop-signal task, we found that the SSRT decreased from
early childhood to young adulthood. The SSRT increased again thereafter, though this decline
in inhibitory control only followed a trend. This pattern aligns with the results found in the
lifespan stop-signal task study of Williams and colleagues (1999), with two exceptions: (1) in
contrast with the study of Williams et al., GoRT did not significantly explain variance in
SSRT, and (2) SSRT peaked in young adulthood, whereas it already peaked in adolescence in
Williams’ study. Our study largely replicates the findings of Williams et al. in a larger
sample, and adds to the few lifespan studies that investigated the maturation and decline of
inhibitory control after controlling for general processing speed. The presence of age-related
differences in response inhibition (SSRT) after accounting for the variance attributable to
response execution (GoRT), together with the larger age-related trends for response execution
relative to response inhibition, coincide with the findings of Williams and others (Band &
Van der Molen, 2000; Bedard et al., 2002; Ridderinkof, Band, & Logan, 1999), and contrast
with the hypothesis that only global differences in speed would underlie age-related
differences in executive control (Salthouse, 1996). The relatively weaker age-related trends
for response inhibition compared to response execution may suggest that withholding planned
actions is relatively stable across life due to an early development and long preservation, or

22
may reflect the possibility that individual differences play a larger role than age differences
(Bedard et al., 2002). The observation that age affected inhibition more heavily in childhood
than in adulthood supports the notion that ageing is not merely development in reverse (Craik
& Bialystok, 2006; Sander, Lindenberger, & Werkle-Bergner, 2012).
Lying proficiency across life
The most basic finding that emerged from the analysis of the Sheffield lie task
that assessed lying proficiency is the observation that the error and RT lie effects as
previously observed in student populations were significantly different from zero in all age
groups. As such, our study forms strong support for the cognitive view of deception, as it
shows that the notion of lying being more cognitively demanding than truth telling applies to
the entire lifespan. However, analyzing lying performance after controlling for truth telling
performance showed that age groups differed in the way they can cope with these cognitive
demands. The instances of erroneously telling the truth on lie trials followed the expected age-
related U-shaped pattern: The lie error rate decreased with age during childhood, was lowest
in young adulthood, and thereafter increased at a relatively slower pace, so that error-
proneness did not differ between the youngest children and the seniors. In contrast, the RT
data revealed an unforeseen age-related S-shaped pattern, with RT lie effects that did not
significantly change from young childhood until young adulthood. The most plausible
explanation seems to relate to the reliability of the RT lie effect. As can be seen in Table 3,
the split-half reliability of the RT lie effect is the lowest in young childhood and
systematically increases with age. The relatively low reliability in childhood reflects a low
consistency in speed of responding, which may have masked their true (larger) RT lie effect.
That children had a low response consistency may suggest that the Sheffield lie test was
generally too difficult for them. Another, not mutually exclusive explanation may be that
children were too distracted during the Sheffield lie test. We often observed that children were

23
able to focus well during the stop-signal task, but that attention continuously dropped when
they had to perform the subsequent Sheffield lie test. Having to perform this second
cognitively demanding task, while knowing that more entertaining attractions were yet to be
explored in the science center, may have lowered their motivation to perform well. Based on
this motivational explanation, one could also argue that a speed-accuracy trade-off may have
further contributed to the small lie effect in children. Although children were still overall
slower than young adults, one cannot entire rule out the possibility that their motivation to
quickly finish the task may have encouraged children to weigh speed over accuracy. Such a
speed-accuracy trade-off may even have occurred apart from motivation, as a few studies
have shown that children often tend to apply a decision rule that is optimized for speed (Carp,
Fitzgerald, Taylor, & Weissman, 2012; Nardini, Bedford, & Mareschal, 2010). Future studies
are thus needed to test our research question again in a context that allows more consistent
responding within children, for example, by making the task less difficult (e.g., blocking truth
and lie trials would reduce the need for task switching), and/or by letting them perform the
task in more controlled, less distractible lab setting, and/or by rewarding them to equally
weigh speed and accuracy.
The age-related U-curve of lying accuracy and the decline of lying speed during
adulthood, are in line with the notion that lying requires executive control. The lack of strong
correlations between stop-signal task and Sheffield lie test performance seems to tone down
such an interpretation, as it casts doubt on the hypothesis that response inhibition is one of the
executive functions at the heart of lying. Yet, a closer look at the literature on response
inhibition suggests that our findings do not necessarily contradict the response inhibition
hypothesis of lying. Previous studies that found a correlation between deception ability and
inhibitory control in adult samples used the Stroop task to assess inhibitory control (e.g.,
Visu-Petra, Miclea, & Visu-Petra, 2012). Developmental studies that examined the relation

24
between executive control and lying administered several inhibition tasks, but only found a
significant correlation with the Stroop task (Evans, Xu & Lee, 2011; Talwar & Lee, 2008).
One explanation may be the cognitively and neurally multifaceted character of response
inhibition (Wager et al., 2005). Different response inhibition paradigms may tap into different
subcomponents of response inhibition that recruit partially distinct regions of the neural
inhibition network (Aron, 2011; Eagle, Bari, & Robbins, 2008; Jahfari et al., 2011; Nigg,
2000; Sebastian et al., 2013; Swick, Ashley, & Turken, 2011). The Stroop task and Simon
task, for example, would tap into ‘interference inhibition’, as they involve the inhibition of
unintentionally activated response tendencies. The Go/No-go task, on the other hand, where a
prepotent response tendency has to be inhibited on a minority of trials, would engage the
‘action withholding’ component. Finally, in the stop-signal task, ‘action cancelation’ would
be needed to inhibit responses that have already been initiated (Sebastian et al., 2013). When
coupling these insights to our study, the lack of a strong association between the deception
task and the stop-signal task may be explained by their reliance on different inhibitory
subcomponents. Whereas the stop-signal measures action withholding, lying in our Sheffield
lie test may have tapped more strongly into the component of interference inhibition as
conflict would arise between the truthful and deceptive response dimension. However, it is
possible that with questions that are less trivial and elicit more dominant truth responses,
interference inhibition may be preceded by action withholding or cancellation. Future
(lifespan) studies that allow a more elaborate investigation may test the above-mentioned
explanation by a more extensive assessment of the inhibition network.
Lying frequency across life
We also aimed to draw a lifespan picture of lying frequency. Three main findings
resulted from the Serota questionnaire that assessed participants’ lying frequency in the past
24 hours. First, the mean lying frequencies of two lies on average per day reported in

25
adulthood were in line with the typical finding that adults tell one to two lies a day (DePaulo
et al., 1996; Serota, Levine, & Boster, 2010). Second, the unbiased measure (i.e., median)
showed that lying frequency increased during childhood, peaked in adolescent years, and then
decreased into old age to the point that seniors lied equally often as the youngest children.
Third, in each age group, lies were non-normally distributed around the mean: (1) Many
participants did not lie, and (2) the distributions of lies were positively skewed, so that most
lies were told by a small group of prolific liars. With these results, we replicated and extended
prior findings (Halevy, Shalvi, & Verschuere, 2014; Levine et al., 2013; Serota, Levine, &
Boster, 2010).
The finding that the age-related changes in lying frequency were by and large similar
to the age-related changes in lying proficiency and SSRT seems to support the idea that better
lying proficiency due to better executive skills may boost the frequency to lie. However, the
correlational analysis did not favor such an account: Whereas lying proficiency only modestly
predicted lying frequency, stop-signal task performance proved no significant predictor. It
seems plausible that the small variety in lying frequency due to the non-normal distribution
made it difficult to find strong relationships. Irrespective of this explanation, the lack of a
relationship with stop-signal task performance may be explained by the fact that lying would
rely on a different inhibitory subcomponent than measured in the stop-signal task (cf. supra).
It is crucial to realize that some developmental studies have shown that next to
executive control (and closely related: theory of mind understanding), lying behavior can also
be influenced by one’s moral evaluation of lies (Jensen, Arnett, Feldman, & Cauffman, 2004;
Talwar & Lee, 2008; Xu, Bao, Fu, Talwar, & Lee, 2010). Future research that assesses lying
frequency would therefore do well by questioning participants about the type of lies told, and
how they morally evaluate and justify these lies. The increase of lies told in childhood may,
for instance, reflect a surge of prosocial, other-oriented lies as a result of an increasing

26
positive evaluation of prosocial lies (Heyman, Sweet, & Lee, 2009; Popliger, Talwar, &
Crossman, 2011; Xu et al., 2010). The peak lying frequency of adolescents may reflect a large
number of lies told to gain autonomy from their parents (Jensen et al., 2004; Perkins & Turiel,
2007). Finally, the decrease of lying frequency throughout adulthood would be consistent
with the finding that older adults become more focused on positive emotional experiences in
social situations and therefore become more sensitive to moral information (Carstensen, Fung,
& Charles, 2003; Narvaez, Radvansky, Lynchard, & Copeland, 2011).
Limitations
Several limitations of the current study can be recognized. First, shortening the tasks
may have sacrificed reliability. Second, the Sheffield lie test examines lying in a restricted
and artificial design, because (1) it only focuses on lying to yes/no questions, (2) it precludes
the emotional involvement of lying that may characterize lying in real life, and (3)
participants are instructed to lie. Such a design rules out the role of theory of mind
understanding (i.e., lying in real life requires the understanding that a false belief can be
instilled in others) and moral reasoning that have been linked to lying ability (e.g., Talwar &
Lee, 2008). However, we do think this restricted design was adequate to address our research
question, because it allowed measuring more purely the executive control aspects involved in
a basic form of lying. The use of such a basic form of lying may, however, explain why the
explained variance of age-related terms was relatively smaller than the variance explained by
baseline performance. Third, we could not examine the impact of education on lying
proficiency, because education levels were not assessed for all participants. One could expect
that the average education level in our study was similar to those in the studies that were also
run in science musea (Williams et al. (1999; Bedard et al. (2002). In these studies, the
majority of participants had high education levels. Such a sample restriction to higher
educated individuals may have prevented more pronounced U-curves in our study, as it has

27
been shown that the age-related cognitive decline is more pronounced in people with a low
level of education (Coffey, Saxton, Ratcliff, Bryan, & Lucke, 1999; Van der Elst et al., 2006).
Fourth, our research was limited to only one executive function, response inhibition capacity,
and should be extended with assessment of other executive functions (i.e., working memory
updating and shifting). Fifth, one could argue that the cross-sectional design may have
distorted our results because people with different ages were born and raised in a different
time period (Schaie, 2013). However, recent findings suggest that longitudinal designs may be
more misleading with regard to cognitive changes than cross-sectional designs (Salthouse,
2014). Whereas cross-sectional studies mostly find age-related U-curves in cognitive abilities,
longitudinal studies often observe that performance levels are maintained or even increase
with ageing (Salthouse, 2009; Schaie, 2013). This suggests that experience with a test on a
previous assessment may influence performance on a following assessment. In line with this
idea, some quasi-longitudinal studies tested the same birth cohort for the first time in different
years, and obtained patterns that best resemble those observed in cross-sectional studies
(Kaufman 2013; Salthouse, 2014). Consequently, we think our findings cannot be ascribed to
the specific research method we used.
Implications for research and practice
Our study points to an important implication for deception research. Deception studies
often use convenient samples of undergraduate participants. However, because our research
shows that young adults are overall the best liars, the true impact of the mentally taxing aspect
of lying (e.g., the role of executive control) and the effectiveness of cognition-based deception
techniques, may be underestimated. Our findings should therefore stimulate deception
researchers to attach more attention to the role of age by testing their hypotheses in a larger
age range.

28
Future research will tell whether the observed age-related pattern of lying proficiency
holds in other lie tests and in more realistic settings. If our findings are generalizable, then
also practitioners would gain from taking into account the age of suspects. Our findings
suggest that whereas the lies of older suspects would be relatively easy to catch, young adults
would be more successful in getting away with their lies. Given that crimes are mostly
committed by adolescents and young adults (Loeber & Farrington, 2014), our findings concur
with the plea to use strategies to increase cognitive load to be able to detect deception
(Walczyk et al., 2013), perhaps particularly so in young adults for whom lie detection may be
a more difficult task.

29
Acknowledgements
We thank the staff of Science Center NEMO, especially Diana Issidorides, for giving
us the opportunity to run this study. We also cordially thank Minki Gyles, Liza Ruijs, Silva
van Schagen, Job Wagemaker, Nikki Wortel, and Peter Paul van Zenderen for assisting in the
collection of the data.
Evelyne Debey was supported by Ghent University Grant BOF01D01010. Maarten
De Schryver is funded by Ghent University Methusalem Grant BOF09/01M00209. Kristina
Suchotzki was supported by an ECRP Grant (09-ECRP-025; FWO Grant ESF 3G099310).

30
References
Abe, N. (2011). How the brain shapes deception: An integrated review of the literature.
Neuroscientist, 17, 560–574. doi:10.1177/1073858410393359
Abe, N., Fujii, T., Hirayama, K., Takeda, A., Hosokai, Y., Ishioka, T., … Mori, E. (2009). Do
parkinsonian patients have trouble telling lies? The neurobiological basis of
deceptive behaviour. Brain, 132, 1386–1395. doi:10.1093/brain/awp052
Ambach, W., Stark, R., & Vaitl, D. (2011). An interfering n-back task facilitates the
detection of concealed information with EDA but impedes it with
cardiopulmonary physiology. International Journal of Psychophysiology, 80,
217–226. doi:10.1016/j.ijpsycho.2011.03.010
Aron, A. R. (2011). From reactive to proactive and selective control: developing a richer
model for stopping inappropriate responses. Biological Psychiatry, 69, e55–e68.
doi:10.1016/j.biopsych.2010.07.024
Band, G. P., Ridderinkhof, K. R., & Segalowitz, S. (2002). Explaining neurocognitive
aging: Is one factor enough? Brain and Cognition, 49, 259–267. doi:10.1006/
brcg.2001.1499.
Band, G. P. H., van der Molen, M. W., & Logan, G. D. (2003). Horse-race model
simulations studies of the stop signal procedure. Acta Psychologica, 112, 105–
142. doi:10.1016/S0001-6918(02)00079-3
Bashore, T. R., & Smulders, F. (1995). Do general slowing functions mask local slowing
effects? A chronopsychophysiological perspective. In P. A. Allen, & T. R. Bashore
(Eds.), Age differences in word and language processing (pp. 390–425).
Amsterdam, Netherlands: North-Holland/Elsevier Science Publishers.
Bedard, A. C., Nichols, S., Barbosa, J. A., Schachar, R., Logan, G. D., & Tannock, R. (2002).
The development of selective inhibitory control across the life span.

31
Developmental Neuropsychology, 21, 93–111. doi: 10.1207/S15326942DN2101
_5
Bradley, M. M., & Lang, P. J. (1994). Measuring emotion: The Self-Assessment Manikin
and the Semantic Differential. Journal of Behavior Therapy and Experimental
Psychiatry, 25, 49–59.doi:10.1016/0005-7916(94)90063-9
Bub, D. N., Masson, M. E. J., & Lalonde, C. E. (2006). Cognitive control in children:
Stroop interference and suppression of word reading. Psychological Science, 17,
351–357. doi:10.1111/j.1467-9280.2006.01710.x
Bunge, S. A., Dudukovic, N. M., Thomason, M. E., Vaidya, C. J., & Gabrieli, J. D. E. (2002).
Immature frontal lobe contributions to cognitive control in children: evidence
from fMRI. Neuron, 33, 301–311. doi:10.1016/S0896-6273(01)00583-9
Burgess, P. W., & Shallice, T. (1997). The Hayling and Brixton test. Hartcourt
Assessment. The Psychological Corporation.
Carlson, S. M., & Moses, L. J. (2001). Individual differences in inhibitory control and
children’s theory of mind. Child Development, 72, 1032–1053. doi:10.1111/
1467-8624.00333
Carp, J., Fitzgerald, K. D., Taylor, S. F., & Weissman, D. H. (2012). Removing the effect of
response time on brain activity reveals developmental differences in conflict
processing in the posterior medial prefrontal cortex. NeuroImage, 59, 853-860. doi:
10.1016/j.neuroimage.2011.07.064
Carstensen, L. L., Fung, H., & Charles, S. (2003). Socioemotional selectivity theory and
the regulation of emotion in the second half of life. Motivation and Emotion, 27,
103–123. doi:10.1023/A:1024569803230
Chandler, M. J., Fritz, A. S., & Hala, S. M. (1989). Small-scale deceit: Deception as a
marker of two-, three-, and four-year-olds’ early theories of mind. Child

32
Development, 60, 1263–1277. doi:10.2307/1130919
Christ, S. E., Van Essen, D. C., Watson, J. M., Brubaker, L. E., & McDermott, K. B. (2009).
The contributions of prefrontal cortex and executive control to deception:
Evidence from activation likelihood estimate meta-analyses. Cerebral Cortex, 19,
1557–1566. doi:10.1093/cercor/bhn189
Christ, S. E., White, D. A., Mandernach, T., & Keys, B. A. (2001). Inhibitory control
across the lifespan. Developmental Neuropsychology, 20, 653–669. doi:10.1207/
875656401753549852
Coffey, C. E., Saxton, J. A., Ratcliff, G., Bryan, R. N., & Lucke, J. F. (1999). Relation of
education to brain size in normal aging: Implications for the reserve hypothesis.
Neurology, 53, 189–196. doi:10.1212/WNL.53.1.189
Cohen, J. (1988). Statistical power analysis for the behavioural sciences. Hillsdale, NJ:
Lawrence Erlbaum.
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. doi:
10.1037//0033-2909.112.1.155
Collette, F., Germain, S., Hogge, M., & Van der Linden, M. (2009). Inhibitory control of
memory in normal ageing: Dissociation between impaired intentional and
preserved unintentional processes. Memory, 17, 104–122. doi:10.1080/0965821
0802574146
Congdon, E., Mumford, J. A., Cohen, J. R., Galvan, A., Canli, T., & Poldrack, R. A. (2012).
Measurement and reliability of response inhibition. Frontiers in Psychology, 3,
37. doi:10.3389/fpsyg.2012.00037
Craik, F. I. M., & Bialystok, E. (2006). Cognition through the lifespan: Mechanisms of
change. Trends in Cognitive Sciences, 10, 131–138. doi:10.1016/j.tics.2006.01.
007

33
Cramér, H. (1999). Mathematical Methods of Statistics. Princeton, NJ: Princeton
University Press.
Davidson, M. C., Amso, D., Anderson, L. C., & Diamond, A. (2006) Development of
cognitive control and executive functions from 4 to 13 years: Evidence from
manipulations of memory, inhibition, and task switching. Neuropsychologia, 44,
2037–2078. doi:10.1016/j.neuropsychologia.2006.02.006
Debey, E., De Houwer, J., & Verschuere, B. (2014). Lying relies on the truth.
Cognition, 32, 324–334. doi: 10.1016/j.cognition.2014.04.009
Debey, E., Ridderinkhof, K. R., De Houwer, J., & Verschuere, B. (2014). Suppressing the
truth as a mechanism of deception: Delta plots reveal the role of response
inhibition in lying. Manuscript submitted for publication.
DePaulo, B. M., Kashy, D. A., Kirkenol, S. E., Wyer, M. W., & Epstein, J. A. (1996). Lying
in everyday life. Journal of Personality and Social Psychology, 70, 979–995. doi:
10.1037//0022-3514.70.5.979
Duran, N. D., Dale, R., & McNamara, D. S. (2010). The action dynamics of overcoming
the truth. Psychonomic Bulletin & Review, 17, 486–491. doi:10.3758/PBR.
17.4.486
Eagle, D. M., Bari, A., & Robbins, T. W. (2008). The neuropsychopharmacology of action
inhibition: cross-species translation of the stop-signal and go/no-go tasks.
Psychopharmacology, 199, 439–456. doi:10.1007/s00213-008-1127-6
Evans, A. D., Xu, F., & Lee, K. (2011). When all signs point to you: Lies told in the face of
evidence. Developmental Psychology, 47, 39–49. doi:10.1037/a0020787
Farah, M. J., Hutchinson, J. B., Phelps, E. A., & Wagner, A. D. (2014). Functional MRI-
based lie detection: scientific and societal challenges. Nature Reviews

34
Neuroscience, 15, 123–131. doi:10.1038/nrn3665
Fullam, R. S., McKie, S., & Dolan, M. C. (2009). Psychopathic traits and deception:
functional magnetic resonance imaging study. British Journal of Psychiatry, 194,
229–235. doi:10.1192/bjp.bp.108.053199
Hadar, A. A., Makris, S., & Yarrow, K. (2012). The truth-telling motor cortex: Response
competition in M1 discloses deceptive behaviour. Biological Psychology, 89,
495–502. doi:10.1016/j.biopsycho.2011.12.019
Hale, S. (1990). A global developmental trend in cognitive processing speed. Child
Development, 61, 653–663. doi:10.2307/1130951
Halevy, R., Shalvi, S., & Verschuere, B. (2014). Being honest about dishonesty:
Correlating self-reports and actual lying. Human Communication Research,
40, 54–72. doi:10.1111/hcre.12019
Heyman, G. D., Sweet, M. A., & Lee, K. (2009). Children’s reasoning about lie-telling and
truth-telling in politeness contexts. Social Development, 18, 728–746. doi:
10.1111/j.1467-9507.2008.00495.x
Jahfari, S., Waldorp, L., van den Wildenberg, W. P. M., Scholte, H. S., Ridderinkhof, K.
R., & Forstmann, B. U. (2011). Effective connectivity reveals important roles for
both the hyperdirect (fronto-subthalamic) and the indirect (fronto-striatal–
pallidal) frontobasal ganglia pathways during response inhibition. Journal of
Neuroscience, 31, 6891–6899. doi:10.1523/JNEUROSCI.5253-10.2011
Jensen, L. A., Arnett, J. J., Feldman, S. S., & Cauffman, E. (2004). The right to do wrong:
Lying to parents among adolescents and emerging adults. Journal of Youth &
Adolescence, 33, 101–112. doi:10.1023/B:JOYO.0000013422.48100.5a
Jurado, M. B., & Rosselli, M. (2007). The elusive nature of executive functions: A review
of our current understanding. Neuropsychology Review, 17, 213–233. doi:

35
10.1007/s11065-007-9040-z
Kaufman, A. S. (2013). Clinical applications II: Age and intelligence across the adult life
span. In E. O. Lichtenberg & A. S. Kaufman (Eds.), Essentials of WAIS IV
Assessment. Hoboken, NJ: Wiley.
Kaylor-Hughes, C. J., Lankappa, S. T., Fung, R., Hope-Urwin, A. E., Wilkinson, I. D., &
Spence, S. A. (2011). The functional anatomical distinction between truth telling
and deception is preserved among people with schizophrenia. Criminal
Behaviour and Mental Health, 21, 8–20. doi:10.1002/cbm.785
Levine, T. R., Serota, K. B., Carey, F., & Messer, D. (2013). Teenagers lie a lot: A further
investigation into the prevalence of lying. Communication Research Reports, 30,
211–220. doi:10.1080/08824096.2013.806254
Levitt, H. (1970). Transformed up-down methods in psychoacoustics. The Journal of
the Acoustical Society of America, 49, 467–477. doi:10.1121/1.1912375
Lewis, M., Stanger, C., & Sullivan, M. W. (1989). Deception in 3-year-olds.
Developmental Psychology, 25, 439–443. doi:10.1037//0012-1649.25.3.439
Loeber, R., & Farrington, D. P. (2014). Age-Crime Curve. In Encyclopedia of Criminology
and Criminal Justice (pp. 12–18). New York, NY: Springer New York.
Logan, G. D. (1981). Attention, automaticity, and the ability to stop a speeded choice
response. In J. Long, & A. D. Baddeley (Eds.), Attention and performance IX (pp.
205–222). Hillsdale, NJ: Erlbaum.
MacLeod, C. M. (1991). Half a century of research on the Stroop Effect: an integrative
review. Psychological Bulletin, 109, 163–203. doi:10.1037//0033-2909.109.2.
163
Massidda, D. (2013). Package ‘retimes’ - Reaction Time Analysis. R package version 0.1-
2. Retrieved from http://cran.r-project.org/web/packages/retimes/retimes.pdf

36
McAuley, T., Yap, M., Christ, S. E., & White, D. A. (2006). Revisiting inhibitory control
across the life span: Insights from the Ex-Gaussian distribution. Developmental
Neuropsychology, 29, 447–458. doi: 10.1207/s15326942dn2903_4
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D.
(2000). The unity and diversity of executive functions and their contributions to
complex frontal lobe tasks: A latent variable analysis. Cognitive Psychology, 41,
49–100. doi:10.1006/cogp.1999.0734
Nardini, M., Bedford, R., & Mareschal, D. (2010). Fusion of visual cues is not mandatory in
children. Proceedings of the National Academy of Sciences of the United States of
America, 107, 17041–17046. doi: 10.1073/pnas.1001699107
Narvaez, D., Radvansky, G. A., Lynchard, N. A., & Copeland, D. E. (2011). Are older
adults more attuned to morally charged information? Experimental Aging
Research, 37, 398–434. doi:10.1080/0361073X.2011.590756
Nigg, J. T. (2000). On inhibition/disinhibition in developmental psychopathology: views
from cognitive and personality psychology and a working inhibition taxonomy.
Psychological Bulletin, 126, 220–246. doi:10.1037/0033-2909.126.2.220
Otgaar, H., Verschuere, B., Meijer, E. H., & Van Oorsouw, K. (2012). The origin of
children’s implanted false memories: Memory traces or compliance? Acta
Psychologica, 139, 397–403. doi:10.1016/j.actpsy.2012.01.002
Perkins, S. A., & Turiel, E. (2007). To lie or not to lie: To whom and under what
circumstances? Child Development, 78, 609–621. doi:10.1111/j.1467-8624.2007.
01017.x
Popliger, M., Talwar, V., & Crossman, A. (2011). Predictors of children’s prosocial lie-
telling: Motivation, socialization variables, and moral understanding. Journal of
Experimental Child Psychology, 110, 373–392. doi:10.1016/j.jecp.2011.05.003

37
Ratcliff, R. (1979). Group reaction time distributions and an analysis of distribution
statistics. Psychological Bulletin, 86, 446–461. doi:10.1037//0033-2909.86.3.446
Raz, N., Lindenberger, U., Rodrigue, K. M., Kennedy, K. M., Head, D., Williamson, A.,
Dahle, C., … Acker, J. D. (2005). Regional brain changes in aging healthy
adults: General trends, individual differences and modifiers. Cerebral Cortex, 15,
1676–1689. doi:10.1093/cercor/bhi044
Ridderinkhof, K. R., Band, G. P. H., & Logan, G. D. (1999). A study of adaptive behavior:
Effects of age and irrelevant information on the ability to inhibit one’s actions.
Acta Psychologica, 101, 315–337. doi:10.1016/s0001-6918(99)00010-4
Rosenthal, R., & Rosnow, R. L. (1991). Essentials of behavioral research (2nd ed.). New
York, NY: McGraw-Hill.
Rush, B. K., Barch, D. M., & Braver, T. S. (2006). Accounting for cognitive aging: Context
processing, inhibition or processing speed? Neuropsychology, Development and
Cognition. Section B: Aging, Neuropsychology and Cognition, 13, 588–610. doi:
10.1080/13825580600680703
Schaie, K. W. (2013). Developmental influences on adult intelligence: The Seattle
Longitudinal Study (2nd ed.). New York, NY: Oxford University Press.
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in
cognition. Psychological Review, 103, 403–437. doi:10.1037//0033-295X.103.3.
403
Salthouse, T. A. (2009). When does age-related cognitive decline begin? Neurobiology
of Aging, 30, 507–514. doi:10.1016/j.neurobiolaging.2008.09.023
Salthouse (2014). Why are there different age relations in cross-sectional and longitudinal
comparisons of cognitive functioning? Current Directions in Psychological Science,
23, 252. doi: 10.1177/0963721414535212

38
Sander, M. C., Lindenberger, U., & Werkle-Bergner, M. (2012). Lifespan age differences
in working memory: A two-component framework. Neuroscience &
Biobehavioral Reviews, 36, 2007–2033. doi:10.1016/j.neubiorev.2012.06.004
Schroeter, M. L., Zysset, S., Wahl, M. M., & von Cramon, D. Y. (2004). Prefrontal
activation due to Stroop interference increases during development—An event-
related fNIRS study. NeuroImage, 23, 1317–1325. doi:10.1016/j.neuroimage.20
04.08.001
Sebastian, A., Pohl, M. F., Klöppel, S., Feige, B., Lange, T., Stahl, C., … Tüscher, O.
(2013). Disentangling common and specific neural subprocesses of response
inhibition. NeuroImage, 64, 601–615. doi:10.1016/j.neuroimage.2012.09.020
Serota, K. B., Levine, T. R., & Boster, F. J. (2010). The prevalence of lying in America:
Three studies of self-reported lies. Human Communication Research, 36, 2–25.
doi:10.1111/j.1468-2958.2009.01366.x
Span, M. M., Ridderinkhof, K. R., & van der Molen, M. W. (2004). Age-related changes
in the efficiency of cognitive processing across the life span. Acta Psychologica,
117, 155–183. doi:10.1016/j.actpsy.2004.05.005
Spence, S. A., Farrow, T. F. D., Herford, A. E., Wilkinson, I. D., Zheng, Y., & Woodruff,
P. W. R. (2001). Behavioural and functional anatomical correlates of deception in
humans. NeuroReport, 12, 2849‒2853. doi:10.1097/00001756-200109170-
00019
Stevens, M., Lammertyn, J., Verbruggen, F., & Vandierendonck, A. (2006). Tscope: A C
library for programming cognitive experiments on the MS Windows platform.
Behavior Research Methods, 38, 280–286. doi:10.3758/BF03192779
Suchotzki, K., Verschuere, B., Crombez, G., & De Houwer, J. (2013). Reaction time
measures in deception research: comparing the effects of irrelevant and

39
relevant stimulus-response compatibility. Acta Psychologica, 144, 224–231.
doi:10.1016/j.actpsy.2013.06.014
Suchotzki, K., Crombez, G.., Debey, E., van Oorsouw, K., & Verschuere, B.. (2014). In vino
veritas? – Alcohol, Response Inhibition and Lying. Alcohol & Alcoholism, 50, 74-81.
doi: 10.1093/alcalc/agu079.
Swick, D., Ashley, V., & Turken, A. U. (2011). Are the neural correlates of stopping and
not going identical? Quantitative meta-analysis of two response inhibition tasks.
NeuroImage, 56, 1655–1656. doi:j.neuroimage.2011.02.070
Talwar, V., & Lee, K. (2002). Development of lying to conceal a transgression: Children’s
control of expressive behavior during verbal deception. International Journal of
Behavioral Development, 26, 436–444. doi:10.1080/01650250143000373
Talwar, V., & Lee, K. (2008). Social and cognitive correlates of children’s lying behavior.
Child Development, 79, 866–881. doi:10.1111/j.1467-8624.2008.01164.x
Talwar, V., Murphy, S., & Lee, K. (2007). White lie-telling in children for politeness
purposes. International Journal of Behavioral Development, 31, 1–11. doi:10.11
77/0165025406073530
Troyer, A. K., Leach, L., & Strauss, E. (2006). Aging and response inhibition: Normative
data for the Victoria Stroop test. Aging, Neuropsychology, and Cognition, 13,
20–35. doi:10.1080/138255890968187
Van Bockstaele, B., Verschuere, B., Moens, T., Suchotzki, K., Debey, E., & Spruyt, A.
(2012). Learning to lie: effects of practice on the cognitive cost of lying. Frontiers
in Psychology, 3, 526. doi:10.3389/fpsyg.2012.00526
van de Laar, M. C., van den Wildenberg, W. P. M., van Boxtel, G. J. M, & van der Molen,
M. W. (2011). Lifespan changes in global and selective stopping and
performance adjustments. Frontiers in Psychology, 2, 357. doi:10.3389/fpsyg.

40
2011.00357
Van der Elst, W., Van Boxtel, M. P. J., Van Breukelen, G. J. P., & Jolles, J. (2006). The
Stroop color–word test: Influence of age, sex, and education; and normative
data for a large sample across the adult age range. Assessment, 13, 62–76.
doi:10.1080/138255890968187
Vartanian, O., Kwantes, P. J., Mandel, D. R., Bouak, F., Nakashima, A., Smith, I., & Lam,
Q. (2013). Right inferior frontal gyrus activation as a neural marker of successful
lying. Frontiers in Human Neuroscience, 7, 616. doi:10.3389/fnhum.2013.00616
Verbruggen, F., Chambers, C. D., & Logan, G. (2013). Fictitious inhibitory differences:
How skewness and slowing distort the estimation of stopping latencies.
Psychological Science, 24, 352–362. doi:10.1177/0956797612457390
Verbruggen, F., & Logan, G. D. (2008). Response inhibition in the stop-signal paradigm.
Trends in Cognitive Sciences, 12, 418–424. doi:10.1016/j.tics.2008.07.005
Verbruggen, F., Logan, G. D., & Stevens, M. A. (2008). STOP-IT: Windows executable
software for the stop-signal paradigm. Behavior Research Methods, 40, 479–
483. doi:10.3758/BRM.40.2.479
Verschuere, B., Spruyt, A., Meijer, E. H., & Otgaar, H. (2011). The ease of lying.
Consciousness and Cognition, 20, 908–911. doi:10.1016/j.concog.2010.10.023
Visu-Petra, G., Miclea, M., & Visu-Petra, L. (2012). RT-based detection of concealed
information in relation to individual differences in executive functioning. Applied
Cognitive Psychology, 26, 342–351. doi:10.1002/acp.1827
Visu-Petra, G., Varga, M., Miclea, M., & Visu-Petra, L. (2013). When interference helps:
increasing executive load to facilitate deception detection in the concealed
information test. Frontiers in Psychology, 4, 146. doi:10.3389/fpsyg.2013.00146
Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities. Chichester, UK:

41
Wiley.
Vrij, A., & Granhag, P. A. (2012). Eliciting cues to deception and truth: What matters are
the questions asked. Journal of Applied Research in Memory and Cognition, 1,
119–117. doi:10.1016/j.jarmac.2012.02.004
Wager, T. D., Sylvester, C. Y., Lacey, S. C., Nee, D. E., Franklin, M., & Jonides, J. (2005).
Common and unique components of response inhibition revealed by fMRI.
NeuroImage, 27, 323–340. doi:10.1016/j.neuroimage.2005.01.054
Walczyk, J. J., Igou, F. P., Dixon, A. P, & Tcholakian, T. (2013). Advancing lie detection by
inducing cognitive load on liars: a review of relevant theories and techniques
guided by lessons from polygraph-based approaches. Frontiers in Psychology, 4,
14. doi:10.3389/fpsyg.2013.00014
West, R. L. (1996). An application of prefrontal cortex function theory to cognitive
aging. Psychological Bulletin, 120, 272–292. doi:10.1037//0033-2909.120.2.272
Williams, E. J., Bott, L. A., Patrick, J., & Lewis, M. B. (2013). Telling Lies: The irrepressible
truth? PLoS ONE, 8, e60713. doi:10.1371/journal.pone.0060713
Williams, B. R., Ponesse, J. S., Schachar, R. J., Logan, G. D., & Tannock, R. (1999).
Development of inhibitory control across the life-span. Developmental
Psychology, 35, 205–213. doi:10.1037//0012-1649.35.1.205
Wilson, A. E., Smith, M. D., & Ross, H. S. (2003). The nature and effects of young
children’s lies. Social Development, 12, 21–45. doi:10.1111/1467-9507.00220
Xu, F., Bao, X., Fu, G., Talwar, V., & Lee, K. (2010). Lying and truth-telling in children:
From concept to action. Child Development, 81, 581–596. doi:10.1111/j.1467-
8624.2009.01417.x
Zelazo, P. D., Craik, F. I. M., & Booth, L. (2004). Executive function across the life span.
Acta Psychologica, 115, 167–183. doi:10.1016/j.actpsy.2003.12.005

42
Footnotes
1 This research was part of Science Live, the innovative research program of Science
Center NEMO that enables scientists to carry out real, publishable, peer-reviewed research
using NEMO visitors as volunteers.
2 For exploratory reasons, we also assessed subjective lying proficiency and difficulty
using 10-point Likert scales: “How good do you think you are at lying?” (1 = definitely not
good, 10 = very good), and “How difficult is lying to you?” (1 = not at all difficult, 10 = very
difficult). Also emotionality of lying was assessed by asking participants how they generally
feel when they lie by means of the 9-point Self-Assessment Manniken (SAM; Bradley
& Lang, 1994) rating scales for valence (1 = very sad, 9 = very happy) and arousal (1 = very
calm, 9 = very aroused). With age, lying was rated as being more difficult, more arousing, and
more negative. Participants who lied more often rated themselves as relatively good liars,
found lying more easy, and experienced less negative and aroused feelings during lying
compared to participants who lied less. Interestingly, participants who lied faster in the
Sheffield lie test also rated themselves as better liars, found lying less difficult, and/or rated
lying less negatively. However, these correlations were small (full results can be obtained on
request).
3 To examine potential gender differences in the stop-signal task, we performed
univariate ANOVAs on GoRT and SSRT with Age (7 age groups) and Sex (Male vs. Female)
as predictors. In the GoRT analysis, the main effect of Sex proved significant, F(1, 818) =
4.29, p = .04, f = 0.07, indicating that male participants were slower to respond (M = 829 ms,
SD = 268) than female participants (M = 772 ms, SD = 235). No effects of Sex were found for
SSRT, Fs < 1.
For the Sheffield lie test, we performed repeated measures ANOVAs on error rates

43
and mean RTs with Deception (Truth vs. Lie), Age, and Sex as predictors. In the error
analysis, no effects of Sex were found, Fs < 1. The RT analysis yielded a main effect of Sex,
F(1, 859) = 10.92, p < .01, f = 0.11, indicating that female participants (M = 3421 ms, SD =
359) were overall slower to respond than male participants (M = 3366 ms, SD = 336).

48
Table 1
Variables of interest from the stop-signal task
Age
(years) Description n % female SSRT
M (SD)
SSD
(ms) p(r|s) GoRT
(ms)
GoAcc
(%)
6–8 Early childhood 47 43 299 (105) 536 (170) .46 (.06) 910 (163) 95.15 (2.89)
9–12 Midchildhood 247 53 257 (86) 500 (215) .46 (.07) 827 (227) 97.12 (2.58)
13–17 Adolescence 89 54 229 (66) 461 (260) .45 (.08) 761 (284) 97.23 (2.70)
18–29 Young adulthood 77 69 196 (54) 473 (238) .46 (.07) 718 (265) 98.86 (1.74)
30–44 Midadulthood 201 68 203 (56) 475 (217) .45 (.07) 733 (238) 98.78 (1.75)
45–59 Older adulthood 119 50 211 (60) 562 (238) .43 (.08) 841 (276) 98.68 (1.80)
60–77 Seniors 52 67 215 (66) 578 (229) .43 (.09) 858 (248) 98.82 (1.33)
6–77 Total 832 58 228 (77) 503 (227) .45 (.08) 796 (251) 97.91 (2.42)
Note. SSRT = stop-signal reaction time (ms); SSD = stop-signal delay (ms); p(r|s) = proportion of responses given a stop signal; GoRT = reaction time on Go trials (ms); GoAcc =
accuracy on Go trials as a percentage of correct Go trials.

49
Table 2
Hierarchical regression analyses predicting global processing speed (GoRT) and response
inhibition capacity (SSRT) by (the linear, quadric, and cubic functions of) age.
Analysis and step
Cumulative
R
F for R
∆R2
∆F
p for ∆F
ß
t for ß
A. GoRT
A. Age
.056
2.61
.003
2.61
.11
–0.06
1.61
A. Age2
.210
19.21
.041
35.71
< .001
1.10
5.98
A. Age3
.211
12.81
.000
0.05
.82
–0.15
0.23
B. SSRT
B. GoRT
.037
1.12
.001
1.12
.29
0.04
1.06
B. Age
.315
45.77
.098
90.30
< .001
–0.31
9.50
B. Age2
.374
44.83
.040
38.78
< .001
1.11
6.23
B. Age3
.374
33.72
.000
0.48
.49
–0.45
0.69
Note. Age = linear function of Age; Age2 = quadratic function of Age; Age3 = cubic function of Age.

50
Table 3
Error rates and mean reaction times in the Sheffield lie test
Error rate (%)
Reaction times (ms)
Age group n % female
Truth
M (SD)
Lie
M (SD)
Lie effect
M (SD)
Estimated lie
M (SD)
Truth
M (SD)
Lie
M (SD)
Lie effect
M (SD)
Spearman-
Brown’s ρ
of RT lie effect
Early childhood 62 45
15.00 (8.50) 20.20 (9.41) 5.20 (7.78) 1.91 (7.40)
3703 (354) 3838 (381) 135 (186) .38
Midchildhood 252 54
11.60 (7.49) 16.86 (8.61) 5.26 (7.28) 1.32 (6.95)
3331 (333) 3463 (365) 133 (186) .51
Adolescence 95 53
10.82 (6.81) 14.59 (8.12) 3.78 (5.87) –0.31 (5.77)
3163 (293) 3301 (321) 138 (143) .52
Young adulthood 79 66
5.97 (5.00) 8.55 (6.23) 2.58 (5.36) –2.43 (5.18)
3100 (313) 3246 (398) 146 (181) .68
Midadulthood 205 68
4.96 (3.97) 9.03 (6.52) 4.07 (5.54) –1.14 (5.53)
3229 (260) 3507 (329) 278 (207) .72
Older adulthood 129 49
5.77 (5.06) 10.40 (6.90) 4.63 (6.20) –0.42 (6.02)
3277 (281) 3584 (350) 307 (219) .74
Seniors 51 69
6.63 (5.74) 12.65 (6.82) 6.03 (7.11) 1.14 (6.68)
3527 (365) 3859 (394) 332 (203) .73
Total 873 58
8.53 (7.00) 13.06 (8.52) 4.53 (6.50) 0.00 (6.36)
3298 (340) 3504 (391) 206 (208) .68

51
Table 4
Hierarchical regression analyses predicting lying proficiency (lying values for errors and
RTs) by (the linear, quadric, and cubic functions of) age.
Analysis and step
Cumulative
R
F for R
∆R2
∆F
p for ∆F
ß
t for ß
A. Errors lying (%)
A. Errors truth telling
.665
689.12
.442
689.12
< .001
0.67
26.25
A. Age
.672
357.36
.009
14.74
< .001
–0.11
3.84
A. Age2
.682
252.39
.015
23.75
< .001
0.66
4.87
A. Age3
.683
190.03
.001
2.04
.154
0.69
1.43
B. Mean RT lying (ms)
B. Mean RT truth telling
.847
2212.70
.718
2212.70
< .001
0.85
47.04
B. Age
.868
1328.82
.036
126.39
< .001
0.19
11.24
B. Age2
.869
890.85
.001
4.43
.036
0.21
2.10
B. Age3
.870
674.82
.002
7.31
.007
–0.88
2.70
Note. Age = linear function of Age; Age2 = quadratic function of Age; Age3 = cubic function of Age.

48
Table 5
Descriptives of self-reported lying frequency
Age group n %
female
Lying
frequency
M (SD)
Lying
frequency
Mdn % no lies % 1–5 lies % > 5 lies
Early childhood 102 53 1.75 (4.82) 0 63.73 29.41 6.86
Midchildhood 295 54 2.59 (4.92) 1 42.71 43.39 13.90
Adolescence 101 53 2.80 (3.08) 2 25.74 59.41 14.85
Young adulthood 83 67 1.94 (2.62) 1 37.35 51.81 10.84
Midadulthood 216 67 2.06 (5.22) 1 49.07 43.06 7.87
Older adulthood 134 49 1.82 (3.16) 0 50.75 38.81 10.45
Seniors 61 69 1.57 (2.47) 0 55.74 34.43 9.84
Total 992 58 2.19 (4.34) 1 45.97 43.04 10.99

49
Table 6
Correlations between variables of interest
1 2 3
1. SSRT – – –
2. Error % lying .08* – –
3. RT lying –.09* .02 –
4. Lying frequency –.01 –.03 –.10**
Note. Error % lying and RT lying represent the estimated values after baseline control.
* p < .05, ** p < .01, *** p < .001

50
Figures
Figure 1. GoRT and SSRT. EC = Early childhood; MC = Midchildhood; ADO =
Adolescence; YA = Young adulthood; MA = Midadulthood; OA = Older adulthood; S =
Seniors (see also Figures 2–3).

51
Figure 2. Observed error rates for truth telling and lying, error lie effects, and estimated
error rates for lying after baseline control.

52
Figure 3. Observed mean RTs for truth telling and lying, RT lie effects, and estimated mean
RTs for lying after baseline control.

53
Figure 4. Lying frequency.

54
Appendix A.
Questions used in the Sheffield lie test.
Questions requiring a truthful
“yes” response
Questions requiring a truthful
“no” response
Is water wet?
Is water dry?
Is ice cold?
Is ice warm?
Can birds fly?
Can pigs fly?
Is a crocodile an animal?
Is a computer an animal?
Is Amsterdam in the Netherlands?
Is Amsterdam in Switzerland?
Are giants big?
Are giants small?
Do cars have four wheels?
Do cars have six wheels?
Is an igloo made of ice?
Is an igloo made of stone?
Is sausage meat?
Is salad meat?
Is stone hard?
Is stone soft?
Is fire warm?
Is fire wet?
Is milk white?
Is milk green?
Are bananas yellow?
Are bananas red?
Is grass green?
Is grass blue?
Does a butcher sell meat?
Does a butcher sell bread?