Dorsolateral prefrontal cortex specifically processes general – but not personal – knowledge deception: Multiple brain networks for lying

Dorsolateral prefrontal cortex specifically processes general – but not personal – knowledge deception: Multiple brain networks for lying

Behavioural Brain Research 211 (2010) 164–168 Contents lists available at ScienceDirect Behavioural Brain Research journal homepage: www.elsevier.co...

323KB Sizes 0 Downloads 2 Views

Behavioural Brain Research 211 (2010) 164–168

Contents lists available at ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Research report

Dorsolateral prefrontal cortex specifically processes general – but not personal – knowledge deception: Multiple brain networks for lying Francesca Mameli a , Simona Mrakic-Sposta a , Maurizio Vergari b , Manuela Fumagalli a,c , Margherita Macis a , Roberta Ferrucci a , Francesco Nordio d,f , Dario Consonni d , Giuseppe Sartori e , Alberto Priori a,b,c,∗ a

Centro Clinico per le Neuronanotecnologie e la Neurostimolazione, Fondazione IRCCS Cà Granda–Ospedale Maggiore Policlinico, Via F. Sforza, 35, 20122 Milan, Italy Unità Operativa di Neurofisiologia Clinica, Fondazione IRCCS Ospedale Maggiore Policlinico, Mangiagalli e Regina Elena, Via F. Sforza, 35, 20122 Milan, Italy c Università di Milano, Dipartimento di Scienze Neurologiche, Via F. Sforza, 35, 20122 Milan, Italy d Unità di Epidemiologia, Fondazione IRCCS Ospedale Maggiore Policlinico, Mangiagalli e Regina Elena, Via F. Sforza, 35, 20122 Milan, Italy e Dipartimento di Psicologia Generale, Università di Padova, Via Venezia, 12, 35100 Padua, Italy f Dipartimento di Clinica Medica, Nefrologia e Scienze della Prevenzione Università degli Studi di Parma, Via Gramsci 14, 43100, Parma, Italy b

a r t i c l e

i n f o

Article history: Received 10 December 2009 Received in revised form 6 March 2010 Accepted 12 March 2010 Available online 20 March 2010 Keywords: Deception Lies Dorsolateral prefrontal cortex Transcranial direct current stimulation tDCS Behaviour Frontal lobe

a b s t r a c t Despite intensive research into ways of detecting deception in legal, moral and clinical contexts, few experimental data are available on the neural substrate for the different types of lies. We used transcranial direct current stimulation (tDCS) to modulate dorsolateral prefrontal cortex (DLPFC) function and to assess its influence on various types of lies. Twenty healthy volunteers were tested before and after tDCS (anodal and sham). In each session the Guilty Knowledge Task and Visual Attention Task were administered at baseline and immediately after tDCS ended. A computer-controlled task was used to evaluate truthful responses and lie responses to questions referring to personal information and general knowledge. Dependent variables collected were reaction times (RTs) and accuracy. At baseline the RTs were significantly longer for lies than for truthful responses. After sham stimulation, lie responses remained unchanged (p = 0.24) but after anodal tDCS, RTs decreased significantly only for lies involving general knowledge (p = 0.02). tDCS left the Visual Attention Task unaffected. These findings show that manipulating DLPFC function with tDCS specifically modulates deceptive responses for general information leaving those on personal information unaffected. Multiple cortical networks intervene in deception involving general and personal knowledge. Deception referring to general and personal knowledge probably involves multiple cortical networks. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Reliable ways of detecting deception have attracted growing interest for legal, moral and clinical purposes [3,4,22,24,34]. Research on deception conducted over the past 10 years has focused primarily on lie detection techniques intended to study deceptive processes in the brain [1,7,13,14,31,32]. Functional magnetic resonance imaging (fMRI) studies show that lying activates the anterior cingulate, dorsolateral and ventromedial prefrontal cortex, and parietal cortex [1,7,13,14,25]. The activated regions nevertheless differ widely across studies probably owing to the

∗ Corresponding author at: Clinical Center for Neuronanotechnology and Neurostimulation, Fondazione IRCCS Ospedale Maggiore Policlinico, Department of Neurological Sciences, University of Milan, Via F. Sforza 35, 20122 Milan, Italy. Tel.: +39 02 55033621; fax: +39 02 55033800. E-mail address: [email protected] (A. Priori). 0166-4328/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2010.03.024

different tasks and experimental protocols used [16]. For example, in some studies the investigators used auditory or visual stimuli, the subjects responded verbally or pressing a button, and the task involved working with various sorts of content, such as playing cards, autobiographical knowledge, or rehearsed scenarios. Other studies used brain stimulation to investigate changes in cortical excitability during deception [11,12,15,27]. In one study, transcranial magnetic stimulation (TMS) was delivered to the left and right motor cortices, and motor evoked potentials (MEPs) were recorded before the question and immediately after questions that elicited a false response [15]. The task required subjects to give ‘Yes’ or ‘No’ answers to questions involving simple lies or truths (“Are you a man?” or “Is the moon round?”) or complex lies or truths (“How old are you?” or “How many legs has a dog?”). The results suggested that generating a false response was associated with increased excitability in the corticospinal tract, perhaps reflecting increased motor readiness or a general effect of arousal.

F. Mameli et al. / Behavioural Brain Research 211 (2010) 164–168

165

Three further studies used transcranial direct current stimulation (tDCS) to alter deceptive responses directly [11,12,27]. Preliminary findings, suggested that cathodal tDCS applied to the anterior prefrontal cortex (aPFC) (FP2 according to the international 10–20 EEG system) speeds up lie telling but not truth telling [11]. The first systematic study describing tDCS-modulated lying [27] showed that anodal tDCS on the dorsolateral prefrontal cortex (DLPFC) (F3/F4 according to the 10–20 EEG system) slows up the speed of lie production but not of the truthful responses. Finally, in a study, extending their previous findings Karim et al. [12], confirmed that cathodal tDCS on the aPFC speeds up lies. Collectively, these experiments show the causal involvement of the PFC in lie responses and provide evidence that tDCS can interfere in a polarity-dependent way with the ability to lie. Available data provide no information on how tDCS affects lies concerning personal information and general knowledge. In this study we investigated whether DLPFC–tDCS specifically influences cognitive processing of general knowledge deception or also influences processing of personal information deception. To do so we delivered anodal and sham tDCS to the DLPFC in healthy subjects. Before and after tDCS participants did the Guilty Knowledge Task (GKT) [17,18], a simple, fast computerized task that requires subjects to answer truthfully or lie to questions referring to personal information and general knowledge. Our experiments investigated instructed lies, a basic lie during which we manipulated the response reversal without manipulating belief [13,14,27]. To investigate the specificity of tDCS-induced changes for lying, we also tested subjects with a Visual Attention Task (VAT). 2. Materials and methods 2.1. Subjects The study was approved by the ethics committee for the Fondazione IRCCS Cà Granda–Ospedale Maggiore Policlinico, Via F. Sforza, 35, 20122, Milan, Italy and was conducted according to the Declaration of Helsinki. After written informed consent, 20 healthy volunteers were tested (men 9, women 11; [mean ± SD] age 29.7 ± 8.3 years; education 16.6 ± 2.7 years). 2.2. Experimental design Because our previous study showed that cathodal tDCS to the DLPFC failed to modulate lie responses [27], in these experiments we used only anodal tDCS. All participants received anodal and sham (placebo) tDCS, tested during two separate experimental sessions held at least 1 week apart. The two types of stimulation were presented in counterbalanced order across subjects, half of the participants received first anodal tDCS and then sham; for the other half the order was reversed. The subject and the examiner who did the ratings were blind to the type of tDCS delivered in each session. The experimenter who applied real or sham tDCS differed from the experimenter who assessed outcome measures. In each session two cognitive tasks, the GKT [13,14,27] and the VAT [26], were tested at baseline (pre-stimulation) and immediately after tDCS ended (poststimulation) (Fig. 1). We used a modified version of the GKT [27]. In the GKT subjects had to answer truthfully or lie to questions referring to personal information (“Are you in Milan?”)

Fig. 2. Guilty Knowledge Task (GKT). The task sequence for personal information (top) and general knowledge (bottom) responses.

or to general knowledge (“Is an apple a fruit?”). Hence, after presenting each question, the computer asked the participant to lie or to respond truthfully. In a total of 80 trials, the stimuli were 40 personal information (PI) questions and 40 general knowledge (GK) questions, presented in random order. Of the 40 stimuli used, 20 required a truthful response to PI questions (Tpi: responding truthfully to a personal question) and 20 to GK questions (Tgk: responding truthfully to a general knowledge question); 20 stimuli required subjects to lie to personal questions (Lpi: lying to a personal question) and 20 to general knowledge questions (Lgk: lying to a general knowledge question). The dependent variables collected were the reaction times (RTs) and accuracy (expressed as a percentage of correct responses) (Fig. 2). To investigate whether the tDCS-induced effects were specific for lying or reflected changes in arousal or attention, we administered the VAT before and after tDCS. We used an exogenous cue version of the Posner task [26] for studying attention using a computer-controlled procedure (CogLab Software, Inc.). In this task, the subjects responded to targets that appeared at one of two locations on either side of the fixation mark. Before the target appeared, one of these locations was cued so that subjects focused their attention on this location. This experimental procedure used three types of cues: valid cues (Vc), invalid cues (Ic) and neutral cues (Nc). Vc appeared on the same side as the target, Ic appeared on the side opposite to the target, and Nc always appeared without the target. The subjects were asked to answer by pressing the n key on the keyboard. As the dependent variable we evaluated the reaction times (RTs). 2.3. Transcranial direct current stimulation (tDCS)

Fig. 1. Experimental design. Each participant did the cognitive tasks before transcranial direct current stimulation (tDCS) and immediately thereafter.

The tDCS was bilaterally delivered with constant direct current stimulator (HDCstim, Newronika, Milan, Italy) connected to three sponge electrodes, two placed on the scalp over bilateral DLPFC side and the other one over the right deltoid muscle [2,5,21]. Scalp electrodes were positioned over F3 and F4 according to the 10–20 EEG international system [27]. To avoid confounding biases arising from two electrodes with opposite polarities over the scalp, we used a non-cephalic reference electrode for tDCS [2,5,21] (Fig. 1). The electrodes used for tDCS were thick (0.3 cm), rectangular saline-soaked synthetic sponges (scalp electrodes 32 cm2 ; deltoid electrode 64 cm2 ). The stimulus was an anodal DC at 2 mA intensity delivered for 15 min over the DLPFC bilater-

166

F. Mameli et al. / Behavioural Brain Research 211 (2010) 164–168

Fig. 3. Reaction times (RT) for truthful (left) and lying (right) responses in the Guilty Knowledge Task (GKT). The histograms are mean RTs for the tasks after tDCS expressed as % of baseline (y axis) for anodal and sham tDCS (x axis). Error bars are SEM. *p < 0.05.

ally (0.03 C/cm2 ). We ramped the current up over the first 5 s of stimulation and down over the last 5 s and kept tDCS below perceptual threshold throughout the experimental session. For sham stimulation, electrode placement was identical to that used for real stimulation but the stimulator was turned off after 10 s. To ensure that subjects could not distinguish real from sham stimulation, they felt the initial itching sensation under both conditions and were therefore blind to the stimulation condition [6]. 2.4. Statistical analysis The dependent variables were RTs and accuracy. A repeated-measures analysis of variance (ANOVA) was run using Greenhouse–Geisser corrections for both tasks. In the VAT, a three-way within-subjects analysis was run with tDCS (anodal, sham), response type (Vc, Ic, Nc) and time (pre- and post-stimulation) as factors. The same three-way analysis was run for the GKT, with the factors tDCS (anodal, sham), response type (Tpi, Tgk, Lpi, Lgk) and time (pre- and post-tDCS). Post hoc analysis with Tukey’s Honest Significant Difference (HSD) test was used to assess differences between the dependent variables measured at baseline and after tDCS for each task. Pearson’s correlation coefficient was calculated to check the correlation between continuous variables and differences in means were assessed using a two-tailed ttest (p ≤ 0.05). All statistical data were analyzed with STATA 10 (StataCorp. Stata Statistical Software: Release 10, College Station, TX, StataCorp LP, 2007). All results are expressed as mean ± standard error of mean (SEM).

3. Results All the subjects tolerated the experiments well and none of them reported adverse effects. Neither sham nor anodal tDCS induced significant changes in VAT RTs (F1,57 = 0.8, p = 0.37). In the GKT neither anodal nor sham tDCS influenced response accuracy (F3.48 = 2.3, p = 0.12). At baseline RTs were significantly longer for lie responses than for truthful responses (955.8 ± 53.3 ms vs 1131.5 ± 79.2 ms; p = 0.001), and RTs were significantly shorter for personal lies than for general knowledge lies (1074.1 ± 70.8 ms vs 1189.0 ± 90.1 ms; p = 0.009). In the truth condition RTs for personal information at baseline were significantly shorter than those for general knowledge (916.1 ± 50.3 ms vs 995.5 ± 59.0 ms; p = 0.007). After anodal DLPFC–tDCS repeated-measures ANOVA identified significant two- and three-way interactions for RTs in the GKT (time × tDCS: F1,64 = 4.3, p = 0.04; time × tDCS × response type: F3,64 = 2.6, p = 0.05). Post hoc analysis confirmed that tDCS speeded up lies to general knowledge (1244.5 ± 110.3 ms vs 1083.5 ± 97.2 ms; F2,64 = 2.8; p = 0.02). Conversely, tDCS failed to change the RTs for personal lies (1033.6 ± 64.3 vs 1114.5 ± 91.7; p = 0.92). In the truth condition neither sham nor anodal tDCS significantly interfered with RTs (Tpi: 939.3 ± 55.6 vs 892.9 ± 53.7; p = 0.89; Tgk: 938.6 ± 72.6 vs 952.5 ± 63.0; p = 0.89) (Fig. 3). No significant differences were found between RTs for general knowledge lies in males and females (F3,45 = 0.2, p = 0.83). Finally, Pearson coefficient disclosed a weak correlation between age and RTs for general knowledge lies (R = 0.4, p = 0.05) and no correlation between age and personal information lies (R = 0.2, p = 0.40).

4. Discussion Our findings in healthy subjects undergoing tDCS during a cognitive task eliciting truthful or lying responses to questions referring to personal information and general knowledge show that whereas tDCS over the human DLPFC has no effect on the cognitive processing of personal information deception, it specifically speeds up general knowledge deception. Our findings are in line with neuroimaging studies indicating the role of the prefrontal cortex in deceptive behaviour [1,7,13] and with previous observations that tDCS can influence deception [27]. Because anodal tDCS interfered with lies for general knowledge responses and left lies for personal information responses unchanged, we assume that modulating lies referring to general knowledge is a specific tDCS-induced effect. Also, because tDCS induced no significant changes in VAT data, we exclude a possible tDCS-induced effect on attentive processes. As expected, RTs in the GKT at baseline were longer for false responses than for true responses. The RTs for lie responses confirm the interference effect when subjects are falsifying information [23]. Under these circumstances, the brain seems to be designed to tell the truth so that, when we lie, control processes such as those in charge of monitoring and inhibiting responses are activated to ‘block’ truthful information and produce false information. The longer RTs when our subjects were lying than when they were telling the truth might reflect the cognitive overload. In line with previous experimental data [7,27], we found that the DLPFC differentially modulated the two types of lies tested. Lies referring to personal information and general information therefore involve different neural mechanisms. This hypothesis receives support from a functional magnetic resonance imaging (fMRI) study [23], showing different neural patterns for autobiographical and non-autobiographical lies. When subjects had to falsify autobiographical responses the superior and mesial prefrontal, the posterior cingulate, precuneus, and temporal and parietal cortices were recruited. Autobiographical lies also elicited more robust effects than did non-autobiographical lies [23]. Several fMRI studies have tried to identify the brain structures involved in the dichotomy between autobiographical vs non-autobiographical or personal vs non-personal stimuli [9,19,23,33]. When subjects received stimuli requiring autobiographical or personal responses fMRI scans showed activity in the medial prefrontal lobe. The mesial prefrontal area has been implicated in self-identity, first-person perspective and emotional processing [9,19,23,33]. Considerable evidence shows that this brain region has functions related to episodic memory [8,10,20]. The task we used to assess deception comprises questions about general knowledge involving semantic memory and questions about personal information, involving episodic memory. Because neuroimaging studies showed activation in DLPFC during semantic retrieval [35], given the brain area we stimulated, we expected DLPFC–tDCS to modulate semantic processes and therefore alter

F. Mameli et al. / Behavioural Brain Research 211 (2010) 164–168

all general knowledge responses. But unexpectedly, it modulated only false responses to general knowledge questions. Hence, presumably tDCS interfered not with brain retrieval of semantic and episodic information but specifically with the correlate of deception for general knowledge deception. The different activation in the brain for autobiographical vs non-autobiographical stimuli suggests a greater conflict and increased cognitive control when subjects falsify information about themselves than when they falsify information of no personal significance. Hence, we hypothesize that tDCS applied to the DLPFC modulates general knowledge deception but not personal information deception because personal information has more complex mechanisms and involves mesial structures in the frontal lobe. From a cognitive neurobiological perspective, liars’ behaviour may be seen as an exercise in behavioural control [30]. Deception involves multiple cognitive processes necessary to generate deception including inhibition (the truthful response must be blocked), set shifting, memory and conflict monitoring (conflict between the automatic truthful response and the lie response required by the instructions). In this study, instead of manipulating overall deception we manipulated only the end stages, two sub-processes, namely inhibition of truthful responses (a process still present also when participants are instructed to lie) and producing a lie while pretending not to know. These cognitive functions are modulated by complex structures including the DLPFC, anterior cingulated cortex, and inferior frontal cortex [7,13,14]. We chose to use the GKT not only because it is suitable for use in an experimental setting (it is short, repeatable and reliable also when delivered to subjects at short intervals) but particularly because it focusses only on the basic cognitive processes of lying. The experimental paradigm we used [13,14,27] excludes those cognitive components that characterize more complex lies, for example planning a complex narrative, holding in working memory interim production of the story, and intentionality. But the greatest advantage of the GKT is that it focusses on a limited set of processes, primarily those involving response selection and inhibition [29]. Our study confirms neuroimaging studies underlining the crucial role of the DLPFC in selecting and inhibiting responses during deception. Our findings directly confirm that the DLPFC is involved in non-personal lies but also indirectly confirm that mesial structures are strongly involved in personal information. Our findings show that tDCS modulates DLPFC functions, altering RTs without changing response accuracy. Hence tDCS probably interfered with information processing without changing the outcome of the response. Finally, in our previous study [27], anodal tDCS had an opposite effect, i.e. anodal DC slowed lie responses. The discrepancy probably arises from methodological differences. First, whereas in our previous study [27] the GKT was based on material learned before the execution of the task, in the present experiment GKT concerned information stored in long-term memory. Furthermore, whereas in our previous study we delivered tDCS at 1 mA for 10 min, in the experiments described here we applied 2 mA for 15 min. Delivering tDCS for longer and at higher intensity might have elicited different changes in cortical excitability and therefore different behavioural changes. The larger the amount of current applied on the scalp, the more distant is the effect. tDCS applied to the prefrontal cortex at 2 mA might therefore influence other processes involved in other features of lying. Although Karim et al. [12] found that cathodal tDCS facilitated lying responses reducing the RTs, because the experimental procedures differed, the two studies are not comparable. For example, Karim et al. [12] delivered tDCS during the task, used a cephalic reference electrode, stimulated the anterior prefrontal cortex (FP2 according to the international 10–20 EEG system), used a role-

167

play task, and delivered stimulation at lower intensity than we did. Despite these methodological differences, the finding common to our study and previous studies on tDCS and deception [12,27] is that tDCS without affecting true answers elicits significant changes in lie generation. In conclusion, our tDCS experiments show that the neural processes underlying deception for personal information differ from those concerning general knowledge information and argue that multiple brain networks exist for lying. Given the effect induced by tDCS on general knowledge responses, it would be interesting in future experiments to assess a double dissociation by stimulating the medial frontal regions so as to evaluate its effect on personal information responses. Although the studies in this field need to be expanded and intensified to deepen the knowledge on the neural correlates of deception, tDCS opens the way to research investigating deception in healthy subjects for potential applications in the forensic field and in pathological liars. Given its simplicity and safety tDCS [28] could be used also in more ‘ecological’ and ‘realistic’ experiments on deception. References [1] Abe N, Suzuki M, Tsukiura T, Mori E, Yamaguchi K, Itoh M, et al. Dissociable roles of prefrontal and anterior cingulate cortices in deception. Cereb Cortex 2006;16:192–9. [2] Cogiamanian F, Marceglia S, Ardolino G, Barbieri S, Priori A. Improved isometric force endurance after transcranial direct current stimulation over the human motor cortical areas. Eur J Neurosci 2007;26:242–9. [3] Edelstein RS, Luten TL, Ekman P, Goodman GS. Detecting lies in children and adults. Law Hum Behav 2006;30:1–10. [4] Farrow TF, Reilly R, Rahman TA, Herford AE, Woodruff PW, Spence SA. Sex and personality traits influence the difference between time taken to tell the truth or lie. Percept Mot Skills 2003;97:451–60. [5] Ferrucci R, Marceglia S, Vergari M, Cogiamanian F, Mrakic-Sposta S, Mameli F, et al. Cerebellar transcranial direct current stimulation impairs the practice-dependent proficiency increase in working memory. J Cogn Neurosci 2008;20:1687–97. [6] Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol 2006;117:845–50. [7] Ganis G, Kosslyn SM, Stose S, Thompson WL, Yurgelun-Todd DA. Neural correlates of different types of deception: an fMRI investigation. Cereb Cortex 2003;13:830–6. [8] Grasby PM, Frith CD, Friston K, Frackowiak RS, Dolan RJ. Activation of the human hippocampal formation during auditory-verbal long-term memory function. Neurosci Lett 1993;163:185–8. [9] Greene JD, Sommerville RB, Nystrom LE, Darley JM, Cohen JD. An fMRI investigation of emotional engagement in moral judgment. Science 2001;293: 2105–8. [10] Henson RN, Shallice T, Dolan RJ. Right prefrontal cortex and episodic memory retrieval: a functional MRI test of the monitoring hypothesis. Brain 1999;122(Pt 7):1367–81. [11] Karim AA, Schneider M, Dockery C, Weber C, Braun C, Birbaumer N. The truth about lying: inhibition of the anterior prefrontal cortex improves deceptive behavior. Psychophysiology 2006;43:S50. [12] Karim AA, Schneider M, Lotze M, Veit R, Sauseng P, Braun C, et al. The truth about lying: inhibition of the anterior prefrontal cortex improves deceptive behavior. Cereb Cortex 2009;20(1):205–13. [13] Langleben DD, Loughead JW, Bilker WB, Ruparel K, Childress AR, Busch SI, et al. Telling truth from lie in individual subjects with fast event-related fMRI. Hum Brain Mapp 2005;26:262–72. [14] Langleben DD, Schroeder L, Maldjian JA, Gur RC, McDonald S, Ragland JD, et al. Brain activity during simulated deception: an event-related functional magnetic resonance study. Neuroimage 2002;15:727–32. [15] Lo YL, Fook-Chong S, Tan EK. Increased cortical excitability in human deception. Neuroreport 2003;14:1021–4. [16] Luber B, Fisher C, Appelbaum PS, Ploesser M, Lisanby SH. Non-invasive brain stimulation in the detection of deception: scientific challenges and ethical consequences. Behav Sci Law 2009;27:191–208. [17] Lykken D. The GSR in the detection of guilt. J Appl Psychol 1959;43:385–8. [18] Lykken D. The validity of the guilty knowledge technique: the effects of faking. J Appl Psychol 1960;44:258–62. [19] Maddock RJ, Garrett AS, Buonocore MH. Posterior cingulate cortex activation by emotional words: fMRI evidence from a valence decision task. Hum Brain Mapp 2003;18:30–41. [20] Maddock RJ, Garrett AS, Buonocore MH. Remembering familiar people: the posterior cingulate cortex and autobiographical memory retrieval. Neuroscience 2001;104:667–76.

168

F. Mameli et al. / Behavioural Brain Research 211 (2010) 164–168

[21] Monti A, Cogiamanian F, Marceglia S, Ferrucci R, Mameli F, Mrakic-Sposta S, et al. Improved naming after transcranial direct current stimulation in aphasia. J Neurol Neurosurg Psychiatry 2008;79:451–3. [22] Newman ML, Pennebaker JW, Berry DS, Richards JM. Lying words: predicting deception from linguistic styles. Pers Soc Psychol Bull 2003;29: 665–75. [23] Nunez JM, Casey BJ, Egner T, Hare T, Hirsch J. Intentional false responding shares neural substrates with response conflict and cognitive control. Neuroimage 2005;25:267–77. [24] Pavlidis I, Eberhardt NL, Levine JA. Seeing through the face of deception. Nature 2002;415:35. [25] Phan KL, Magalhaes A, Ziemlewicz TJ, Fitzgerald DA, Green C, Smith W. Neural correlates of telling lies: a functional magnetic resonance imaging study at 4 Tesla. Acad Radiol 2005;12:164–72. [26] Posner MI. Orienting of attention. Q J Exp Psychol 1980;32:3–25. [27] Priori A, Mameli F, Cogiamanian F, Marceglia S, Tiriticco M, Mrakic-Sposta S, et al. Lie-specific involvement of dorsolateral prefrontal cortex in deception. Cereb Cortex 2008;18:451–5.

[28] Priori A, Hallett M, Rothwell JC. Repetitive transcranial magnetic stimulation or transcranial direct current stimulation? Brain Stimul Basic Transl Clin Res Neuromodul 2009;2:241–5. [29] Sip KE, Roepstorff A, McGregor W, Frith CD. Detecting deception: the scope and limits. Trends Cogn Sci 2008;12:48–53. [30] Spence SA. The deceptive brain. J R Soc Med 2004;97:6–9. [31] Spence SA, Farrow TF, Herford AE, Wilkinson ID, Zheng Y, Woodruff PW. Behavioural and functional anatomical correlates of deception in humans. Neuroreport 2001;12:2849–53. [32] Spence SA, Hunter MD, Farrow TF, Green RD, Leung DH, Hughes CJ, et al. A cognitive neurobiological account of deception: evidence from functional neuroimaging. Philos Trans R Soc Lond B: Biol Sci 2004;359:1755–62. [33] Vogeley K, Fink GR. Neural correlates of the first-person-perspective. Trends Cogn Sci 2003;7:38–42. [34] Vrij A, Mann S, Fisher RP. An empirical test of the behaviour analysis interview. Law Hum Behav 2006;30:329–45. [35] Wiggs CL, Weisberg J, Martin A. Neural correlates of semantic and episodic memory retrieval. Neuropsychologia 1999;37:103–18.