Deception rate in a “lying game”: Different effects of excitatory repetitive transcranial magnetic stimulation of right and left dorsolateral prefrontal cortex not found with inhibitory stimulation

Deception rate in a “lying game”: Different effects of excitatory repetitive transcranial magnetic stimulation of right and left dorsolateral prefrontal cortex not found with inhibitory stimulation

Neuroscience Letters 583 (2014) 21–25 Contents lists available at ScienceDirect Neuroscience Letters journal homepage: www.elsevier.com/locate/neule...

547KB Sizes 5 Downloads 26 Views

Neuroscience Letters 583 (2014) 21–25

Contents lists available at ScienceDirect

Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet

Deception rate in a “lying game”: Different effects of excitatory repetitive transcranial magnetic stimulation of right and left dorsolateral prefrontal cortex not found with inhibitory stimulation Inga Karton a,b,∗ , Annegrete Palu a , Kerli Jõks a , Talis Bachmann b a b

University of Tartu, Institute of Psychology, Näituse 2, Tartu 50409, Estonia University of Tartu (Tallinn branch), Institute of Public Law, Kaarli puiestee 3, Tallinn 10119, Estonia

h i g h l i g h t s • Left-DLPFC excitation by TMS compared to right-DLPFC excitation decreases lying. • Excitation protocol of TMS is more systematic compared with the inhibition protocol. • Right hemisphere is more susceptible to opposite effects of stimulation types compared with the left-hemisphere effects.

a r t i c l e

i n f o

Article history: Received 12 July 2014 Received in revised form 22 August 2014 Accepted 7 September 2014 Available online 16 September 2014 Keywords: rTMS Deception DLPFC Lying game

a b s t r a c t Knowing the brain processes involved in lying is the key point in today’s deception detection studies. We have previously found that stimulating the dorsolateral prefrontal cortex (DLPFC) with repetitive transcranial magnetic stimulation (rTMS) affects the rate of spontaneous lying in simple behavioural tasks. The main idea of this study was to examine the role of rTMS applied to the DLPFC in the behavioural conditions where subjects were better motivated to lie compared to our earlier studies and where all possible conditions (inhibition of left and right DLPFC with 1-Hz and sham; excitation of left and right DLPFC with 10-Hz and sham) were administered to the same subjects. It was expected that excitation of the left DLPFC with rTMS decreases and excitation of the right DLPFC increases the rate of lying and that inhibitory stimulation reverses the effects. As was expected, excitation of the left DLPFC decreased lying compared to excitation of the right DLPFC, but contrary to the expectation, inhibition had no different effects. These findings suggest that propensity to lie can be manipulated by non-invasive excitatory brain stimulation by TMS targeted at DLPFC and the direction of the effect depends on the cortical target locus. © 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction The importance of studying deceptive behaviour lies primarily in the necessity to be able to detect it. The possibilities of lie detection can be advanced once the brain processes involved in lying are known. Lying is a complex cognitive activity as it requires more effort than speaking the truth [1–3]. This kind of information management entails the implementation and control of the executive functions [3,4], which are essential for deceptive behaviour. Lying does not consist solely of expressing a lie [5]: it is also important to avoid speaking the truth, which may, in some cases, require

∗ Corresponding author at: University of Tartu, Institute of Psychology, Näituse 2, Tartu 50409, Estonia. Tel.: +372 737 6611. E-mail address: [email protected] (I. Karton). http://dx.doi.org/10.1016/j.neulet.2014.09.020 0304-3940/© 2014 Elsevier Ireland Ltd. All rights reserved.

even more control [1,6]. In recent times, this topic has been studied extensively within the framework of neurosciences targeting the mechanisms of cognitive processes. However, a lot remains unclear in the study of lying and many contradictory research results have been produced, in which correlational relations have been investigated the most while causal relations have been examined significantly less [7]. The aforementioned correlational relations have been studied with the aid of various neuroimaging methods [5,6,8], during which different brain processes are comparatively assessed while subjects respond to stimuli truthfully or by lying. It cannot be claimed without a doubt that certain brain activity is caused by deceptive behaviour or vice versa because lying activates areas of the brain which are involved in many different cognitive processes [2,3,9,10]. The studies capable of observing some causal effects are mostly carried out by implementing either of the two non-invasive

22

I. Karton et al. / Neuroscience Letters 583 (2014) 21–25

neurostimulation techniques: transcranial direct current stimulation (tDCS) and transcranial magnetic stimulation (TMS). The principal advantage of TMS is that it helps to investigate causal relations between the activation or inhibition of a certain brain region and monitored behaviour, and thus verify the validity of neuroimaging methods [1,10]. TMS uses the electromagnetic induction principle, which changes certain short-term brain activity in a specific area of the cerebral cortex [11]. When the behaviour of a subject changes after stimulating certain brain regions, it is possible to assume a causal connection between them. The biggest problem in determining whether lying has its corresponding brain region or specific brain mechanism is the fact that these regions may be and often are active without deception as well [3]; this is because deception is based on several different cognitive processes that are linked to the same region [5]. An increasing number of authors have studied and claimed that processes of lying are associated with the dorsolateral prefrontal cortex (DLPFC) e.g., [2,6,8,9,12]. The functionality of right and left DLPFC is different: right DLPFC is involved in cognitive control, avoidance and behavioural inhibition [13–15] while left DLPFC participates in reality monitoring, approach motivation, strategic behaviour, naming and execution [8,9,16,17]. As the prefrontal cortex is asymmetrical in its function, it is possible to assume that the rTMS applied to the left and right DLPFC also brings about varying changes in behaviour [18]. We have previously [19,20] found that stimulating the DLPFC with rTMS affects the rate of spontaneous lying in simple behavioural tasks. Spontaneous choice to lie more or less can be influenced by brain stimulation. Where subjects had freedom to name presented stimulus-objects (red and blue coloured circles) either veridically or nonveridically the amount of truthful answers can be manipulated by inhibitory off-line 1-Hz repetitive transcranial magnetic stimulation targeted at DLPFC: inhibition of the left DLPFC with rTMS increased the relative rate of lying, but inhibition of the right DLPFC decreased it [19]. In the second study [20] the subjects were allowed to report the name of the shape (circle or square) of the object they actually saw or report the name of the object they did not actually see, therefore producing a non-truthful response. When trains of 10-Hz pulses were delivered to the right DLPFC, propensity to lie increased while similar left-hemisphere DLPFC stimulation did not change the rate of untruthful responses. In order to retest those earlier findings and develop this subject further, we conducted a study reported here in which we examined all four conditions (the excitation and inhibition of the left and right DLPFC with rTMS) together in the same subjects and using a larger sample. But in order to develop the method and test for the generality of the effects, we replaced the task requiring simple spontaneous object-naming responses with lying in a more highly motivated and engaging task context. However, we preserved the subjects’ free will in regard to whether, how much and when to lie. In accord with most of our above mentioned earlier results we hypothesise that excitation of the left DLPFC with rTMS decreases the rate of lying and excitation of the right DLPFC increases the rate of lying; also, to the contrary, we expect that inhibition of the left DLPFC with rTMS increases and inhibition of the right DLPFC decreases the rate of lying. (Even though the specific hypothesis about the effect of excitation of left DLPFC was not confirmed in the earlier study, there is no reason for the contrary assumption.)

2. Method 2.1. Subjects Seventeen healthy right-handed volunteers participated in the study, sixteen of them (12 females) were included in the data analysis (age of subjects, 20–47 years, M = 25.6, SD = 7.85). One subject

was excluded due to problems with following the instructions. Subjects visited the lab altogether on five days: four experiment days and one a pre-experiment day when his/her motor threshold (MT) was estimated, and a written informed consent document introduced and signed. The study was approved by the Research Ethics Committee of the University of Tartu and was conducted according to the principles set in the Declaration of Helsinki. The subjects each time received 5-8 EUR in compensation for participation; the precise amount of money depended on the collected points, which was another element in the attempt to increase the motivation to lie. 2.2. Experiment The experimental task consisted in playing a so-called ‘circle game’, which was created in our lab and was a further development of our previous experimental protocol [19]. During the game, red and blue circles appeared in quasi-random order on the computer monitor for 100 ms per each appearance. The time between appearance of the circles and possibility to respond was set at 400 ms. Each trial started with a fixation interval with a variable length of 1000/1500 ms (Fig. 1A) The subjects’ behavioural task was to name the colour of the circle. In each trial, a circle (diameter 13 cm) was presented on a SUN CM751U monitor (1024 × 768 pixels; 100 Hz refresh rate) from the viewing distance of approximately one metre. The slight variability in viewing distance appeared due to the precondition that participants were allowed to choose a distance that was comfortable for them to look at the screen and perceive the stimuli as they liked. During the entire game, a total of 240 circles were presented, either of the colours was used 120 times. Participants were instructed that the purpose of the game is to collect as many points as possible and that seeing and naming the red circles will give one point per each red circle. However, they were additionally told that there is also possibility to lie and name the blue circles as red. Thus, one point was earned when the subject answered “red”, no points were earned for answering “blue”. The subjects were free to choose whether to lie or not. Importantly, during the game there were also occasional checks. For every untrue answer detected by the experiment program, the subject lost 5 points. 40 control trials were divided equally between trials with red and trials with blue circles. To answer, the subjects’ used two fingers on their right hand: they pressed the right arrow key to answer red and the left arrow key to answer blue. The score and number of trials executed were visible during the entire experiment. Also, feedback from control trials was shown after each control trial. 2.3. Transcranial magnetic stimulation The experimental task was accompanied by “off-line” 1-Hz and 10-Hz rTMS to inhibit or excite the left or right DLPFC (BA 9, see Fig. 1B). Each subject participated in four experimental sessions, carried out on different days. All subjects received all stimulation conditions (left and right 1-Hz and sham; left and right 10-Hz and sham) and the conditions were counterbalanced between subjects. In the part of the experiment with inhibitory stimulation (in two separate days) the subjects received – prior to each experimental block – a train of 1-Hz rTMS (360 pulses over the course of 6 min) or sham stimulation to the left or to the right DLPFC followed by the experimental task which contained the naming of 120 circles. Each block was carried out twice. In the part of the experiment consisting of excitatory stimulation (spanning over two days) the subjects received 24 one-second trains of 10-Hz or sham stimulation to the left or to the right DLPFC, followed by the 10 s long time ‘window’ without rTMS. During this ‘window’ subjects performed the experimental tasks naming five circles per ‘window’.

I. Karton et al. / Neuroscience Letters 583 (2014) 21–25

23

Fig. 1. Illustration of the display screens showing typical stimuli, and trial count and points scoring areas (A); and the target locations for TMS in frontal cortex: left and right DLPFC (BA 9) (B).

Depending on the terms of the experimental protocol the subjects’ DLPFC was stimulated either by rTMS or by imitating it in the sham condition. To mask the coil-generated clicks and to reduce differences between real rTMS and sham stimulation, white noise was played over earphones in both cases. For sham stimulation the white noise was mixed with an audio recording of the TMS clicks while the coil was pressed perpendicularly to the subject’s head and no real magnetic pulses were delivered. MRI-assisted NBS (Navigated Brain Stimulation; Nexstim Ltd., Helsinki, Finland) system with a figureof-eight coil was used for stimulation. The stimulation intensity was set at 100% of the individual MT. There was a 10 min break between the sham- and TMS blocks. Although the experimental task and stimulation target (left or right DLPFC) remained the same during all blocks, the TMS status related condition (TMS, sham) was varied. The break was essential to avoid a sustained inhibitory effect of TMS during sham condition: according to Robertson and colleagues [10] and Thut and Pascual-Leone [21] the effect of stimulation in DLFPC is diminishing after 5–10 min from the end of stimulation. 3. Results At first we looked at to what extent subjects responded according to the instruction by producing appropriate responses. The appropriate responses were: red responded as red, blue responded as blue and blue responded as red (a lying response allowed by the protocol). Red circle responded to as blue qualified as a behavioural mistake, and the subjects lost the point as this was not consistent with the experimental protocol. We corrected the deception level for performance errors by subtracting the amount of mistakes (per individual and condition) from appropriate responses. The amount of mistakes accounted for 0.72% of all responses through all conditions. The errors per condition were the same in the excitation sham and TMS and inhibition sham conditions, in the inhibition TMS condition subjects made 1.6 times more mistakes compared with the sham condition. In the following analysis we focused on the lying range. During the game altogether the subjects responded to 1920 trials, 240 trials per condition (excitation – left and right, sham and TMS; and inhibition – left and right, sham and TMS) where half of the trials consisted in presenting blue circles. Thus, the maximal possible rate for lying was 960. The actual overall rate of lying varied from 86 to 774 (M = 472.69; SD = 191.03), in the excitation conditions this rate varied from 20 to 408 (M = 230.94; SD = 102.90) and in the inhibition conditions from 66 to 366 (M = 241.75; SD = 92.29). For the analysis of the effects of TMS on deceptive responding we first averaged the results of sham conditions (rate of lying varied from 47 to 383; M = 244.25; SD = 94.89; the sham conditions did not differ [F(1,15) = 1.44, p = 0.248]) and then subtracted the overall average from the number of deceptive responses in the TMS conditions. Repeated-measures two-way analysis of variance (ANOVA)

Fig. 2. Mean differences between averaged deceptive responses of the sham conditions subtracted from the deceptive responses in the TMS conditions.

was used to assess the effects of experimental conditions. Planned comparisons were carried out via dependent samples t-test. When we looked at all data together fitted into the same model, we found that although the main effect was not significant neither for rTMS protocol (1-Hz vs 10-Hz) nor for stimulation hemisphere (left vs right DLPFC) (p > 0.05), the interaction between rTMS protocol and stimulation hemisphere was significant [F(1,15) = 5.494, p = 0.033; G2 = 0.0138]. A t-test for dependent samples showed significant differences of excitatory stimulation between hemispheres [t(15) = −2.521, p = 0.023, Cohen’s d = 0.630] (Fig. 2). Also we interviewed the subjects after the experiment to get to know if they understood the difference between stimulation and sham conditions; did they use any strategy to play the game; whether they were motivated and what motivated them the most; how they felt during the game. In response to these questions the subjects replied that they did notice the difference between TMS and sham stimulation in all cases, but in most of the cases (11 from 16) they did not realise that there was no stimulation at all during the sham condition – they thought that it was some other type of stimulation or just did not wonder about that. During the game all subjects tried out

24

I. Karton et al. / Neuroscience Letters 583 (2014) 21–25

different strategies and changed them – they were engaged into planning most of the time. The main motivator was the hazardous excitement, not as much the amount of money they might win. All subjects also reported some negative emotional arousal (e.g. frustration) when they got caught and lost their points.

4. Discussion The main idea of this study was to examine the role of rTMS applied to the DLPFC in the behavioural conditions where subjects were better motivated to lie compared to our earlier studies. As we expected, excitation had different effects when stimulation was applied to the right and left hemisphere. Excitation of the left DLPFC with rTMS tended to decrease lying compared to right DLPFC stimulation which showed the lying rate slightly increased. This result showing the differences between hemispheres is consistent with our previous finding [20] and adds generality to this observed regularity by changing the behavioural context and motivational demand. However, our present finding that excitation of the left DLPFC decreased the amount of lying was expected but not confirmed in our earlier study [20]. Surprisingly, we did not find an inter-hemispheric opposite effect in case of inhibition as we did in our first study [19]. Our hypotheses were supported only partially. The reason that no difference of the effect was found between left- and righthemisphere stimulation in the rTMS inhibitory condition may be predominantly methodological: that is to say 1-Hz rTMS has a lingering effect which reduces the comparative laterality effects between DLPFCs especially when cognitive control is involved [22,23]. The duration of the rTMS effect is mainly dependent on the duration and the strength of the magnetic stimulation. Thus, the one reason why in our first experiment [19] the behavioural differences occurred may be related to the length of the behavioural task: in our first experiment the task performance lasted several times less than in this experiment whereas the duration of stimulation was even half a minute shorter. Therefore, in the present case the rTMS application was not enough to induce a significant enough change in behaviour. In any case, the present results must be interpreted cautiously. While our previous studies were first and foremost an attempt to manipulate people’s free choices in a simple naming task, in this experiment, however, conscious lying has been included within the context of the risk-taking game, which is much more complex, and which relates more likely to the involvement of the entire network for which DLPFC is an important part. Thus in this more complex case we still cannot be sure which function may be superior or provide crucial input to behaviour control and this is probably the main reason why the results differ. Deceptive behaviour itself is a complex cognitive process which besides expressing the lie consists in also avoiding telling the truth [1,5,6]. Therefore, excitation of the right DLPFC, which is known to be involved in cognitive control and behavioural inhibition [14,15], will improve the control capacity or readiness. This means that producing a response that goes counter to the perceived reality is relatively easier. Also as right DLPFC is responsible for the actual mental generation and evaluation of sequences as it requires the integration of individual moves while taking into account their interdependencies [24], the excitation of this cortical area may make subjects prefer the more risky alternatives and therefore lie more. The results about excitation of right DLFPC continue to be inconsistent: it is found that after anodal tDCS over the right DLPFC the subjects either made safer and conservative decisions [25,26] or they did not [27]. Our results showed that excitation of the left DLPFC decreased the rate of non-truthful responses. This may appear surprising,

provided that some earlier findings about risky behaviour [27] indicate to the opposite expectation. Perhaps when lying could be considered as a real risk this effect may overweigh other effects of DLPFC modulation. Moreover, because TMS exerts strong contralateral inhibitory effects on the homologous cortical areas this is equivalent to right DLPFC inhibition and this result is to be expected according to our hypotheses. Left DLPFC is also involved in reality monitoring and strategic planning [9,17] and activated more for higher demands on goal hierarchy [24]. Consequently, its excitation may bring forward the more natural (inner) goal – not to lie, as was also mentioned by subjects, and the lying rate decreased. The fact that in our previous study [20] left DLPFC excitation did not decrease lying can be interpreted on the grounds of motivational factors: in that study the task was simply to name the visible objects with lying allowed; in this study lying was motivated, being a goal if one wants to win over the computer and/or get a high score in the game. Furthermore, in the present task spontaneity of responses was more controlled by strategic considerations on the subjects’ side based on evaluation of the game-situation. 5. Conclusions Consistent with our earlier findings, the propensity to lie can be manipulated by non-invasive brain stimulation by TMS targeted at DLPFC. The effects depend on (i) stimulated hemisphere and (ii) type of stimulation. Left DLPFC excitation tends to decrease lying contrary to right DLPFC excitation. It seems that the excitation protocol is more systematic compared with the inhibition protocol. It also appears that whether right-hemisphere effects or left-hemisphere effects are more or less robust depends on the type of task as evidenced by the differences between the results of different studies. Acknowledgements This work was supported by the Estonian Science Agency, project SF0180027s12 (TSHPH0027) and IUT20-40 (TSHPH14140I). We also thank Endel Põder for contributing to the creation of the game and Anu Einberg, Carolina Murd and Renate Rutiku for help they provided through various stages of this research. References [1] B. Verschuere, T. Schuhmann, A.T. Sack, Does the inferior frontal sulcus play a functional role in deception? A neuronavigated theta-burst transcranial magnetic stimulation study, Front. Hum. Neurosci. 6 (284) (2012) 1–7. [2] A.A. Karim, M. Schneider, A. Lotze, R. Veit, P. Sauseng, C. Braun, N. Birbaumer, The truth about lying: inhibition of the anterior prefrontal cortex improves deceptive behavior, Cereb. Cortex 20 (1) (2010) 205–213. [3] K.E. Sip, A. Roepstorff, W. McGregor, C.D. Frith, Detecting deception: the scope and limits, Trends Cogn. Sci. 12 (2) (2008) 48–53. [4] E. Debey, B. Verschuere, G. Crombez, Lying and executive control: an experimental investigation using ego depletion and goal neglect, Acta Psychol. 140 (2) (2012) 133–141. [5] K.E. Sip, M. Lynge, M. Wallentin, W.B. McGregor, C.D. Frith, A. Roepstorff, The production and detection of deception in an interactive game, Neuropsychology 48 (12) (2010) 3619–3626. [6] D.D. Langleben, L. Schroeder, J.A. Maldjian, R.C. Gur, S. McDonald, J.D. Ragland, C.P. O’Brien, A.R. Childress, Brain activity during simulated deception: an event-related functional magnetic resonance study, NeuroImage 15 (3) (2002) 727–732. [7] G. Ganis, J.P. Keenan, The cognitive neuroscience of deception, Social Neurosci. 4 (6) (2009) 465–472. [8] A. Ito, N. Abe, T. Fujii, A. Hayashi, A. Ueno, S. Mugikura, S. Takahashi, E. Mori, The contribution of the dorsolateral prefrontal cortex to the preparation for deception and truth-telling, Brain Res. 1464 (2012) 43–52. [9] N. Abe, How the brain shapes deception: an integrated review of the literature, The Neuroscientist 17 (5) (2011) 560–574. [10] E.M. Robertson, H. Théoret, A. Pascual-Leone, Studies in cognition: the problems solved and created by transcranial magnetic stimulation, J. Cogn. Neurosci. 5 (7) (2003) 948–960. [11] M. Hallett, Transcranial magnetic stimulation: a primer, Neuron 55 (2007) 187–199.

I. Karton et al. / Neuroscience Letters 583 (2014) 21–25 [12] A. Priori, F. Mameli, F. Cogiamanian, S. Marceglia, M. Tiriticco, S. Mrakic-Sposta, R. Ferrucci, S. Zago, D. Polezzi, G. Sartori, Lie-specific involvement of dorsolateral prefrontal cortex in deception, Cereb. Cortex 18 (2) (2008) 451–455. [13] D. Knoch, E. Fehr, Resisting the power of temptations: the right prefrontal cortex and self-control, Ann. N.Y. Acad. Sci. 1104 (2007) 123–134. [14] O. Vartaniana, P. Kwantesa, D.R. Mandela, Lying in the scanner: localized inhibition predicts lying skill, Neurosci. Lett. 529 (2012) 18–22. [15] A.J. Shackman, B.W. McMenamin, J.S. Maxwell, L.L. Greischar, R.J. Davidson, Right dorsolateral prefrontal cortical activity and behavioral inhibition, Psychol. Sci. 20 (2009) 1500–1506. [16] E.T. Berkman, M.D. Lieberman, Approaching the bad and avoiding the good: lateral prefrontal cortical asymmetry distinguishes between action and valence, J. Cogn. Neurosci. 22 (2010) 1970–1979. [17] N. Steinbeis, B.C. Bernhardt, T. Singer, Impulse control and underlying functions of the left DLPFC mediate age-related and age-independent individual differences in strategic social behavior, Neuron 73 (2012) 1040–1051. [18] D. Knoch, V. Treyer, M. Regard, R.M. Müri, A. Buck, B. Weber, Lateralized and frequency-dependent effects of prefrontal rTMS on regional cerebral blood flow, NeuroImage 31 (2006) 641–648. [19] I. Karton, T. Bachmann, Effect of prefrontal transcranial magnetic stimulation on spontaneous truth-telling, Behav. Brain Res. 225 (2011) 209–214. [20] I. Karton, J.-M. Rinne, T. Bachmann, Facilitating the right but not left DLPFC by TMS decreases truthfulness of object-naming responses, Behav. Brain Res. 271 (2014) 89–93.

25

[21] G. Thut, A. Pascual-Leone, A review of combined TMS-EEG studies to characterize lasting effects of repetitive TMS and assess their usefulness in cognitive and clinical neuroscience, Brain. Topogr. 22 (2010) 219–232. [22] T. Torii, A. Sato, M. Iwahashi, K. Iramina, Transition of after effect on P300 by short-term rTMS to prefrontal cortex, IEEE Trans. Magn. 48 (11) (2012) 2873–2876. [23] M. Hansenne, O. Laloyaux, S. Mardaga, M. Ansseau, Impact of low frequency transcranial magnetic stimulation on event-related brain potentials, Biol. Psychol. 67 (2004) 331–341. [24] C.P. Kaller, B. Rahm, J. Spreer, C. Weiller, J.M. Unterrainer, Dissociable contributions of left and right dorsolateral prefrontal cortex in planning, Cereb. Cortex 21 (2011) 307–317. [25] S. Fecteau, D. Knoch, F. Fregni, N. Sultani, P. Boggio, A. Pascual-Leone, Diminishing risk-taking behavior by modulating activity in the prefrontal cortex: a direct current stimulation study, J. Neurosci. 27 (2007) 12500– 12505. [26] P.S. Boggio, S. Zaghi, A.B. Villani, S. Fecteau, A. Pascual-Leone, F. Fregni, Modulation of risk-taking in marijuana users by transcranial direct current stimulation (tDCS) of the dorsolateral prefrontal cortex (DLPFC), Drug Alcohol Depend. 112 (2010) 220–225. [27] T. Sela, A. Kilim, M. Lavidor, Transcranial alternating current stimulation increases risk-taking behavior in the Balloon Analog Risk Task, Front. Neurosci. 14 (2012).