Introduction
Repetitive transcranial magnetic stimulation (rTMS) is a non-invasive technique causing transient disruptions in brain activity by delivering strong magnetic pulses to particular areas in the human brain cortex (Kluger & Triggs, Reference Kluger and Triggs2007). Previous research (Post & Keck, Reference Post and Keck2001) has demonstrated that low-frequency (LF) rTMS (⩽1 Hz) is capable of decreasing cortical excitability, whereas high-frequency (HF) rTMS (>1 Hz) increases it. Yet, recent research (Siebner et al. Reference Siebner, Lang, Rizzo, Nitsche, Paulus, Lemon and Rothwell2004; Silvanto et al. Reference Silvanto, Cattaneo, Battelli and Pascual-Leone2008) has emphasized that the baseline cortical activation state of the targeted region may also modulate the activating or disrupting effects of rTMS.
Over the last decade, this technique has emerged as a promising treatment procedure for major depressive disorder (McNamara et al. Reference McNamara, Ray, Arthurs and Boniface2001; Holtzheimer et al. Reference Holtzheimer, Russo and Avery2002; Avery et al. Reference Avery, Holtzheimer, Fawaz, Russo, Neumaier, Dunner, Haynor, Claypoole, Wajdik and Roy-Byrne2006; Rachid & Bertschy, Reference Rachid and Bertschy2006; Bortolomasi et al. Reference Bortolomasi, Minelli, Fuggetta, Perini, Comencini, Fiaschi and Manganotti2007). Research findings have suggested the effectiveness of multi-session rTMS in reducing depressive symptoms when HF or LF magnetic fields are generated respectively in the left (Kozel & George, Reference Kozel and George2002; Gershon et al. Reference Gershon, Dannon and Grunhaus2003) or right (Klein et al. Reference Klein, Kreinin, Chistyakov, Koren, Mecz, Marmur, Ben-Shachar and Feinsod1999; Fitzgerald et al. Reference Fitzgerald, Benitez, de Castella, Daskalakis, Brown and Kulkarni2006) dorsolateral prefrontal cortex (DLPFC). On the other hand, as stated by Loo & Mitchell (Reference Loo and Mitchell2005) and in light of recent reports of negative rTMS trials (e.g. Herwig et al. Reference Herwig, Fallgatter, Höppner, Eschweiler, Kron, Hajak, Padberg, Naderi-Heiden, Abler, Eichhammer, Grossheinrich, Hay, Kammer, Langguth, Laske, Plewnia, Richter, Schulz, Unterecker, Zinke, Spitzer and Schönfeldt-Lecuona2007; Eranti et al. Reference Eranti, Mogg, Pluck, Landau, Purvis, Brown, Howard, Knapp, Philpot, Rabe-Hesketh, Romeo, Rothwell, Edwards and McLoughlin2007), there is much scope for the further refinement and development of rTMS as an antidepressant treatment strategy. We believe that a possible avenue to further this refinement is to investigate possible underlying neurocognitive working mechanisms.
Neuroimaging studies in depression (Mayberg, Reference Mayberg1997, Reference Mayberg2007; Drevets, Reference Drevets2000; Leppänen, Reference Leppänen2006) have demonstrated dysfunctional activation patterns within the DLPFC and other cortical and subcortical brain regions (e.g. the anterior cingulate cortex and amygdala). These anatomical structures are all components of the neural circuit involved in the interplay between emotional and attentional information processing (Liotti & Mayberg, Reference Liotti and Mayberg2001; Taylor & Fragopanagos, Reference Taylor and Fragopanagos2005). More specifically, it has been found that the DLPFC is important in the implementation of top-down attentional control (MacDonald et al. Reference MacDonald, Cohen, Stenger and Carter2000, Vanderhasselt et al. Reference Vanderhasselt, De Raedt, Baeken, Leyman and D'Haenen2006). Dysfunctional activation patterns within this region might lead to a diminished ability to inhibit or disengage attention away from intrusive emotional information (Davidson et al. Reference Davidson, Pizzagalli, Nitschke and Putman2002), that is, information that corresponds with the slumbering negative schemata of depression-prone individuals (Ingram et al. Reference Ingram, Miranda and Zegal1998). Indeed, these biases in the attentional processing of emotional information have reliably been observed within recent behavioural studies examining depressive subjects (Koster et al. Reference Koster, De Raedt, Goeleven, Franck and Crombez2005; Goeleven et al. Reference Goeleven, De Raedt, Baert and Koster2006; Leyman et al. Reference Leyman, De Raedt, Schacht and Koster2007), underscoring that the inability to inhibit negative information might function as an important vulnerability in the development and maintainance of depressive disorder.
Based on the above reasoning, multiple sessions of rTMS applied over the DLPFC may also cause – apart from the reported antidepressant effects – improvements in attentional control, modulated by changes in cortical brain excitability within stimulated prefrontal regions (Luborzewski et al. Reference Luborzewski, Schubert, Seifert, Danker-Hopfe, Brakemeier, Schlattmann, Anghelescu, Colla and Bajbouj2007). Improvements in general cognitive functioning after rTMS treatment that correlate with positive mood effects have repeatedly been reported within recent clinical studies (Padberg et al. Reference Padberg, Zwanzger, Thoma, Kathmann, Haag, Greenberg, Hampel and Möller1999; Little et al. Reference Little, Kimbrell, Wassermann, Grafman, Figueras, Dunn, Danielson, Repella, Huggins, George and Post2000; Speer et al. Reference Speer, Figueras, Demian, Kimbrell, Wasserman and Post2001; Martis et al. Reference Martis, Alam, Dowd, Hill, Sharma, Rosen, Pliskin, Martin, Carson and Janicak2003; Fabre et al. Reference Fabre, Galinowski, Oppenheim, Gallarda, Meder, De Montigny, Olié and Poirier2004; Hausmann et al. Reference Hausmann, Pascual-Leone, Kemmler, Rupp, Lechner-Schoner, Kramer-Reinstadler, Walpoth, Mechtcheriakov, Conca and Weiss2004; O'Conner et al. Reference O'Conner, Jerskey, Robertson, Brenninkmeyer, Ozdemir and Leone2005; Schulze-Rauschenbach et al. Reference Schulze-Rauschenbach, Harms, Schlaepfer, Maier, Falkai and Wagner2005). However, so far, almost no research has attempted to gain more insight into the immediate effects of prefrontal rTMS on the attentional processing of emotional information, thereby enhancing our understanding of the potential underlying working mechanisms of this promising treatment tool.
The most straightforward way to start investigating this research question is by assessing the temporary effects of one session of prefrontal HF-rTMS in a sample of healthy volunteers. More specifically, this study will examine possible alternations in the inhibitory control for emotional information in two separate experiments, stimulating the left and right DLPFC, and additionally, will also control for potential associated changes in subjective mood reports. Although the present study is, to our knowledge, the first exploring this research question, some predictions can be made based on previous findings in the literature.
First, in line with recent research examining immediate mood effects of one session of HF prefrontal rTMS in healthy volunteers (Mosimann et al. Reference Mosimann, Rihs, Engeler, Fisch and Schlaepfer2000; Baeken et al. Reference Baeken, Leyman, De Raedt, Vanderhasselt and D'Haenen2006), no significant mood changes were expected in both stimulation conditions.
Second, with regard to the expected changes in attentional performance, differential predictions can be made depending on the site of prefrontal stimulation.
Previous studies applying HF-rTMS over the left DLPFC were able to demonstrate immediate positive effects on several aspects of global executive functioning in depressive patient samples (e.g. Martin et al. Reference Martin, Barbanoj, Schlaepfer, Thompson, Perez and Kulisevsky2003). Indeed, this site of stimulation is predominantly chosen in depression treatment and has been identified within recent neuropsychological and neuroimaging studies as being dominantly involved in top-down attentional control (MacDonald et al. Reference MacDonald, Cohen, Stenger and Carter2000). In line with these findings, a recent study by Boggio et al. (Reference Boggio, Bermpohl, Vergara, Muniz, Nahas, Leme, Rigonatti and Fregni2007) was able to demonstrate specific improvements in performance on an affective go/no-go task after anodal transcranial direct current stimulation of the left DLPFC. More specifically, these authors observed an increased attention to stimuli of positive emotional content in a depressive patient sample. However, in a recent fMRI study by Rounis et al. (Reference Rounis, Stephan, Lee, Siebner, Pesenti, Friston, Rothwell and Frackowiak2006), one session of HF-rTMS applied over the left DLPFC in a sample of healthy volunteers did not lead to improvements or changes in attentional reorienting.
Contrary to the growing number of studies examining the immediate effects of left prefrontal HF-rTMS on global attentional functioning, to date almost no research has focused on examining the immediate effects of right prefrontal HF-rTMS. The study by Rounis et al. (Reference Rounis, Stephan, Lee, Siebner, Pesenti, Friston, Rothwell and Frackowiak2006) is, to our knowledge, the first to indicate a differential involvement of left versus right DLPFC in attentional processing. These authors demonstrated a deficiency in the reorienting of attention during performance of a cued visual choice reaction-time task (Posner, Reference Posner1980) after HF-rTMS over the right DLPFC.
Because Rounis et al. (Reference Rounis, Stephan, Lee, Siebner, Pesenti, Friston, Rothwell and Frackowiak2006) did not focus on examining the immediate effects of prefrontal rTMS on the attentional processing of emotional information, the present study aimed at extending this research. More specifically, we examined whether prefrontal HF-rTMS in healthy volunteers might cause immediate changes in an important cognitive vulnerability factor for depression, i.e. a deficient inhibition of negative information. Potential changes in inhibitory control were examined, using an experimental paradigm that enables direct quantification of the strength of inhibitory processes toward emotional information before and after rTMS, i.e. the Negative Affective Priming (NAP) task (Joormann, Reference Joormann2004; Gotlib et al. Reference Gotlib, Yue and Joormann2005; Goeleven et al. Reference Goeleven, De Raedt, Baert and Koster2006).
To summarize, because to date no previous study has examined the immediate effects of left and right prefrontal HF-rTMS on the attentional processing of emotional information and self-reported mood in healthy subjects, the present study can offer new and useful insights into (1) the involvement of prefrontal circuits in the attentional processing of emotional information and (2) the underlying working mechanisms of rTMS treatment. Based on results by Rounis et al. (Reference Rounis, Stephan, Lee, Siebner, Pesenti, Friston, Rothwell and Frackowiak2006) it was hypothesized that HF-rTMS of the DLPFC in healthy subjects would (1) induce no changes in inhibitory processing of emotional information after left HF-rTMS but (2) might cause deficits in the inhibitory processing of negative information after right HF-rTMS. Moreover, we hypothesized that HF-rTMS over the left or right DLPFC will (3) not lead to instant mood changes.
Experiment 1: Method
Subjects
Eighteen healthy, drug-free, right-handed female volunteers aged between 19 and 24 years (mean age=21.1 years, s.d.=1.45 years) were recruited to participate in this experiment. Subjects received a complete description of the procedure of the study which was approved by the local institutional ethics committee. All subjects subsequently gave written informed consent. Prior to inclusion in the study, subjects were carefully screened in order to meet safety criteria for HF-rTMS (Wassermann, Reference Wassermann1998). They underwent a thorough physical and psychiatric examination. Psychiatric disorders were excluded using the Mini International Neuropsychiatric Interview (MINI); a structured clinical interview performed by a trained psychiatrist (Pinninti et al. Reference Pinninti, Madison, Musser and Rissmiller2003). All participants also completed the Beck Depression Inventory (BDI; Beck et al. Reference Beck, Ward, Mendelson, Mock and Erlbaugh1961), a 21-item self-report measure of the severity of depressive symptoms. All participants had a score below 10, which is indicative of the absence of depression (mean=1.67, s.d.=2.68). Apart from the exclusion of psychiatric disorders, additional exclusion criteria were a history of epileptic seizures and neurosurgical interventions, having a pacemaker or other metal or magnetic implants and being pregnant. Finally, all included participants had to be medication free. Only birth-control pills were allowed. Handedness was assessed with the hand preference scale of Van Strien (Reference Van Strien2001). Subjects were financially compensated.
Study design
Participants received one session of HF-rTMS over the left DLPFC. Each subject also received one session of placebo (sham) rTMS stimulation, separated by an interval of 1 week from the active stimulation session. The sequence of sessions was randomized across subjects and a single-blind crossover design was used in order to exclude non-specific effects.
In order to evaluate temporary changes in mood before (T pre), immediately after (T post) and ±40 min after (T post40) each rTMS (active/sham) session, mood ratings were administered using five visual analogue scales (VAS) providing measures of sadness, fatigue, tension, anger and vigour (McCormack et al. Reference McCormack, David, Horne and Sheater1988). Participants were asked to describe how they felt ‘at that moment’ by indicating on horizontal 100 cm lines whether they experienced the five above-mentioned mood states, from ‘totally not’ to ‘very much’.
Before (T pre) and ±60 min after (T post60) stimulation or placebo, inhibitory processing of emotional information was measured using the NAP task.
Because the present study is part of a larger project investigating the influence of rTMS on different neurocognitive markers, before and immediately after each rTMS (active/sham) session, additional cognitive tasks and functional magnetic resonance imaging measures were also administered. Yet, these measures were not used for the purposes of the present study.
rTMS
For the application of rTMS we used a Magstim high-speed magnetic stimulator (Magstim Company Limited, Whitland, Carmarthenshire, UK) connected to a figure-of-eight-shaped coil. Before stimulation the identification of the precise stimulation location of the left or right DLPFC (Brodmann area 9/46) and the optimal position of the coil were determined for each subject using MRI non-stereotactic guidance. Perpendicular to this point the precise stimulation site on the skull was marked and stimulated. Next, a stimulation intensity of 110% of the subject's motor threshold of the right abductor pollicis brevis muscle was determined using electromyography.
In a HF stimulation session (10 Hz), subjects received 40 trains of 3.9 s duration, separated by an intertrain interval of 26.1 s (1560 pulses per session). During placebo stimulation, the coil was placed at an angle of 90°, resting on the scalp with only one edge. During stimulation, all subjects wore earplugs. Before and during stimulation subjects were also blindfolded in order to ensure that the altering of the orientation of the coil with respect to the scalp in the placebo condition was effectively blinded. Safety guidelines were followed based on recent available safety studies on rTMS (Wassermann, Reference Wassermann1998; Anand & Hotson, Reference Anand and Hotson2002).
NAP task
Inhibitory processing of emotional information was measured using the NAP task. During the administration of the NAP task, subjects were seated at a 60 cm viewing distance from an IBM-compatible computer with a 72 Hz, 17 inch (43 cm) colour monitor. The task was programmed using inquisit software (2001 version 1.33; Millisecond Software, Seattle, WA, USA). At the start of each trial, subjects were instructed to look at a fixation cross that was displayed for 1000 ms in the middle of the computer screen. Thereafter, two emotional faces were presented in the upper and the lower half of the screen, one picture surrounded by a grey frame and one by a black frame. At each trial, subjects had to evaluate the valence (positive or negative) of the emotional expression of the target picture in the grey or black frame (randomized across subjects) by pressing one of two corresponding keys and had to ignore the distractor picture. Both facial expressions remained on the screen until responding. All individual trials were separated by a 1000 ms blank screen.
In this multi-stimulus task, a complete NAP sequence includes two separate trials: a prime and a probe trial. Importantly, participants were not aware of this difference between prime and probe trials. However, within experimental conditions, distractors in the prime trial correspond with the emotional valence of targets in the probe trial. Due to this manipulation, the NAP effect can be measured, involving a slowdown in responding to an item that has previously been inhibited, which is an index of inhibitory functioning toward affective material (e.g. Wentura, Reference Wentura1999). This delay in responding is not expected within control conditions, in which there is no similarity between prime distractors and probe targets. Table 1 provides an overview of the different conditions used in the NAP task. Subjects first completed 32 practice trials, followed by a sequence of 256 test trials, divided into eight blocks of 16 prime and probe trials. The order of NAP sequences within the blocks was randomized. Furthermore, the spatial position of the target and the distracter in both the prime trials and the probe trials were randomly assigned from trial to trial, with an equal number of presentations for each condition.
Table 1. Control and experimental conditions for negative and positive trials in the NAP task
NAP, Negative Affective Priming; +, happy facial expression; −, sad facial expression; N, neutral facial expression.
The entire task lasted approximately 20 min. The 88 coloured pictorial stimuli used in this paradigm were carefully selected on valence and arousal ratings based on a prior validation study of the Karolinska Directed Emotional Faces database (Goeleven et al. Reference Goeleven, De Raedt, Leyman and Verschuere2008). In the present task, 33 happy, 33 sad and 22 neutral faces were presented in random order. The neutral faces were used as distractors in the probe trials. Facial expressions were 5 cm wide by 5.5 cm high and were surrounded by a 3 mm coloured frame. Responses to prime and probe trials were recorded, but only responses to the probe trials were analysed.
Data analysis
In order to analyse changes in subjective mood ratings after HF-rTMS over the left or right DLPFC, a multivariate repeated-measures analysis of variance with stimulation (active v. placebo rTMS) and time (T pre, T post and T post40) as within-subjects factors was conducted on the different VAS, used as multiple dependent variables (i.e. fatigue, tension, anger, vigour and depression).
Second, a repeated-measures analysis of variance with valence (negative v. positive), stimulation (active v. placebo rTMS) and time (T prev. T post60) as within-subject factors was used to examine the NAP scores (i.e. individual mean reaction times in the experimental condition minus individual mean reaction times in the control condition).
For all analyses the significance level was set at an α level of 0.05. Estimates of effect size are also reported (partial eta squared: ηp2 or Cohen's d). All procedures were two-tailed and analyses were conducted with SPSS version 12.0 (SPSS Inc., Chicago, IL, USA).
Experiment 1: Results
Effects of HF-rTMS over the left DLPFC on mood
The relevant mean mood ratings for each VAS are presented in Table 2. Due to some missing values, five subjects were removed from analysis.
Table 2. VAS measures (cm) before (Tpre), immediately (Tpost) and 40 min after (Tpost40) the different rTMS conditions
VAS, Visual analogue scale; rTMS, repetitive transcranial magnetic stimulation.
Values are given as mean (standard deviation).
The crucial two-way interaction effect between stimulation and time was not significant [F(10, 3)=3.81, p=0.15, ηp2=0.93]. Analyses also revealed no significant overall effects of time or stimulation (F's<1.1, all ηp2<0.75).
Effects of HF-rTMS over the left DLPFC on inhibitory processing
An overview of the mean response times for the NAP conditions before and after the different rTMS conditions, as well as the mean NAP scores for negative and positive facial expressions, is presented in Table 3. Due to technical problems, NAP scores from one subject were missing. Only NAP sequences that were correct in both prime and probe trials were further processed. Therefore, when an error was made on one trial in a NAP sequence (consisting of a prime trial and a probe trial), the other trial was also discarded from further processing. In this way, 6.40% of all trials over the different conditions were removed. Thereafter, responses shorter than 300 ms and longer than 2000 ms were considered as outliers (reflecting anticipatory and delayed responding, respectively; see Goeleven et al. Reference Goeleven, De Raedt, Baert and Koster2006) and were also removed from the analyses (5.8%).
Table 3. Response times (ms) for the NAP conditions before and after the different rTMS conditions
NAP, Negative Affective Priming; rTMS, repetitive transcranial magnetic stimulation.
Values are given as mean (standard deviation).
a NAP score=experimental condition – control condition. A positive NAP score indicates effective inhibition of emotional information, whereas the smaller the score, the more that inhibitory control becomes impaired.
Repeated-measures analysis of variance revealed no significant three-way interaction effect between stimulation, time and valence (F<1, ηp2=0.001). We also could not demonstrate a significant main effect of stimulation, time or valence (F's<1, ηp2<0.06), nor did we find any significant two-way interaction (all F's<1, all ηp2<0.02).
Experiment 1: Discussion
The findings of experiment 1 showed that one session of HF-rTMS over the left DLPFC had no immediate effects on the attentional processing of emotional information in healthy volunteers, or on their subjective mood reports.
The absence of a significant change in inhibitory control after one session of HF-rTMS over the left DLPFC is in line with previous findings by Rounis et al. (Reference Rounis, Stephan, Lee, Siebner, Pesenti, Friston, Rothwell and Frackowiak2006). These authors were also unable to demonstrate significant changes in attentional reorienting during administration of a non-emotional cued choice reaction-time task after HF-rTMS of the left DLPFC in a sample of healthy volunteers. The absence of improved attentional control in both studies can possibly be explained by a ceiling effect in healthy subjects. In these subjects, inhibitory control and attentional reorienting is already functioning well, with additional improvements not being expected contrary to the improvements found in studies examining attentional functioning in depressive patient samples after left magnetic stimulation (e.g. Boggio et al. Reference Boggio, Bermpohl, Vergara, Muniz, Nahas, Leme, Rigonatti and Fregni2007).
The present experiment was also not able to demonstrate immediate changes in self-reported mood. Although some previous studies have found immediate mood changes after one single session of left prefrontal HF-rTMS in healthy volunteers (e.g. Pascual-Leone et al. Reference Pascual-Leone, Catala and Pascual-Leone1996), these results are in line with recent evidence reported in well-designed methodologically sound studies (Mosimann et al. Reference Mosimann, Rihs, Engeler, Fisch and Schlaepfer2000; Baeken et al. Reference Baeken, Leyman, De Raedt, Vanderhasselt and D'Haenen2006) that failed to induce immediate mood changes after one session of HF-rTMS over the left DLPFC in healthy volunteers.
In light of the absence of left prefrontal rTMS effects on mood and attentional processing of emotional information in healthy volunteers, a second experiment was conducted to check whether one session of HF-rTMS applied over the contralateral side of the DLPFC would result into similar or differential effects.
Experiment 2: Method
Subjects
Twenty-two healthy, drug-free, right-handed female subjects aged between 20 and 30 years (mean age=24 years, s.d.=2.33 years) were recruited to participate in the second experiment. None of them had participated in the previous experiment. As in the first experiment, participants were carefully screened and gave written informed consent. None of the participants reported a history of psychiatric problems, with scores on the BDI below 6 indicative of the absence of depression (mean=2.18, s.d.=1.81). Subjects were again financially compensated.
Study design and data analysis
The second experiment followed the exact same procedure as the first one, except that participants now received one session of HF-rTMS over the right DLPFCFootnote †. The statistical approaches were also the same as in the first experiment.
Experiment 2: Results
Effects of HF-rTMS over the right DLPFC on mood
The relevant mean mood ratings for each VAS are again presented in Table 2. Due to some missing values, two subjects were removed from the analysis.
The crucial two-way interaction effect between stimulation and time was not significant [F(10, 10)=1.13, p=0.42, ηp2=0.53]. Analyses also revealed no significant overall effects of time or stimulation (F's<1, all ηp2<0.40).
Effects of HF-rTMS over the right DLPFC on inhibitory processing
An overview of the mean response times for the NAP conditions before and after the different rTMS conditions, as well as the mean NAP scores for negative and positive facial expressions, is presented in Table 3. Also in this experiment, errors (4.19%) and outliers (1.89%) were discarded from further analyses.
In line with our hypothesis, a significant three-way interaction effect was found [F(1, 21)=4.20, p=0.05, ηp2=0.17]. More specifically, subjects showed a significant decrease in NAP scores for negative faces after HF-rTMS over the right DLPFC [t(21)=2.13, p<0.05, d=0.55] compared with baseline measures, revealing less effective inhibition of negative information. NAP scores for negative faces after active rTMS also significantly differed from scores after placebo rTMS [t(21)=3.04, p<0.01, d=0.84], whereas no differences between both stimulation conditions were found before rTMS (t<1, d=0.005). A significant change in inhibitory processing of positive information was not found, nor did we find any significant effects in the placebo condition (all t's<1.5, all d's<0.47).
Experiment 2: Discussion
Although no effects were found after HF-rTMS of the left DLPFC, the present experiment was able to offer evidence for a significant deficit in the inhibitory processing of negative information after stimulating the right DLPFC.
These results are in accordance with the hypothesis that lateralized disturbances in the homeostasis of frontal brain activity – with a relative higher activation pattern in the right compared with the left DLPFC consistently found in major depression (Drevets, Reference Drevets2000; Henriques & Davidson, Reference Henriques and Davidson1991; Tomarken & Keener, Reference Tomarken and Keener1998) – can lead to impairments and biases in the inhibition of mood-congruent stimuli (Heller & Nitschke, Reference Heller and Nitschke1997). Indeed, increased activation patterns within the right prefrontal brain regions have frequently been associated with increased processing of negative emotions and stimuli with a negative emotional content (Davidson et al. Reference Davidson, Pizzagalli, Nitschke and Putman2002). Moreover, these results are in line with previous research demonstrating disturbances in reorienting attention towards non-emotional information after right DLPFC conditioning in healthy volunteers (Rounis et al. Reference Rounis, Stephan, Lee, Siebner, Pesenti, Friston, Rothwell and Frackowiak2006).
This cognitive impairment could not be explained as an effect of negative mood. The absence of changes in self-reported mood again replicate those reported in a recent study by Baeken et al. (Reference Baeken, Leyman, De Raedt, Vanderhasselt and D'Haenen2008), indicating no immediate mood effects after one session of HF prefrontal rTMS in healthy volunteers.
General discussion
The present study aimed at contributing to a deeper understanding of the underlying cognitive processes that might mediate positive treatment effects of rTMS in depressive pathology. More specifically, we investigated the effects of one session of HF-rTMS applied over the left and right DLPFC on the inhibitory processing of emotional information and mood in two samples of healthy subjects. Using this straightforward experimental design, we were able to demonstrate that one session of right unilateral prefrontal stimulation causes inhibitory deficits only in the processing of negative information, whereas no significant changes in inhibitory control were found when stimulating the left DLPFC. Furthermore, this study failed to demonstrate any significant acute mood changes that could account for the above findings. The implications of these results will be discussed in more detail below.
Consistent with our hypothesis, it was found that HF stimulation of the right DLPFC resulted into an acute impaired inhibitory processing of negative information in healthy volunteers. This cognitive impairment in the processing of emotional information is commonly reported in depressed individuals (Joormann, Reference Joormann2004; Goeleven et al. Reference Goeleven, De Raedt, Baert and Koster2006) and has, as already been stated, been associated with disturbances in activation patterns within prefrontal brain regions (Levin et al. Reference Levin, Heller, Mohanty, Herrington and Miller2007). Although, based on our behavioural results, the present study cannot make inferences about the associated changes in brain activation patterns; it can be assumed that HF right rTMS might have resulted into a similar ‘depression-specific’ prefrontal brain asymmetry with higher relative right versus left activity.
To our knowledge, this study is the first suggesting a relationship between heightened activity within right prefrontal areas and specific deficiencies in the attentional processing of negative emotional information. Apart from the present study, to date only two studies have examined the effect of prefrontal rTMS on the attentional modulation of emotional information in healthy subjects (d'Alfonso et al. Reference d'Alfonso, van Honk, Hermans, Postma and de Haan2000; van Honk et al. Reference van Honk, Schutter, d'Alfonso, Kessels and de Haan2002). Using LF-rTMS, both studies provided an important contribution to the experimental rTMS research by evidencing prefrontal involvement in emotional goal-directed behaviour, demonstrating the involvement of the left and right PFC in the withdrawal and approach-related emotions anger and fear respectively. However, these studies explored a somewhat different hypothesis from the present study. Furthermore, the above studies differed from the present study with respect to the experimental task employed to measure attentional control processes. Using a prototypical emotional Stroop task, ‘controlled suppression’ of a dominant response (i.e. naming the coloured word) is measured, whereas within negative priming tasks, as the one used in the present study, more automatic ‘reactive inhibition’ is assessed (Miyake et al. Reference Miyake, Friedman, Emerson, Witzki and Howerter2000), the latter possibly tapping more closely onto the cognitive vulnerability suggested in depressive pathology.
The present study can be regarded as a first step in exploring the intriguing research question of whether underlying changes in cognitive processing of emotional information could be responsible for the well-established therapeutic effects of rTMS in depressive pathology (Rachid & Bertschy, Reference Rachid and Bertschy2006). Because in the present study no immediate changes in self-reported mood were found, the speculative assumption can be made that in depressive patients, one session of left HF-rTMS may restore the imbalance between right and left prefrontal activity, leading to primary changes in attentional control, with secondary changes in self-reported mood only appearing after multiple treatment sessions. In order to make more firm conclusions, it is evident that in future research this study has to be replicated within a group of clinically depressed patients, applying multiple sessions of HF-rTMS over the left DLPFC. Moreover, fMRI data might additionally reveal associated changes in activation patterns of specific neural circuits (dorsal frontocingulate–amygdala connectivity) mediating emotion/cognition interactions. Next, because this study only investigated healthy female subjects, in order to ensure homogeneity of the study sample, generalization of the present findings is limited. Therefore, this study is also in need of replication in a mixed sample of female and male subjects. Finally, although in the present study placebo stimulation was performed at an angle of 90°, causing minimal stimulation of the right and left DLPFC, a somewhat different sensation compared with active stimulation might have been experienced. However, differences in perceived amplitude between both stimulation conditions were minimized using earplugs and blindfolding participants. Moreover, in the present study it was impossible to voluntary control responses made on the NAP task because subjects were not aware which of the presented trials were prime or probe trials, with their succession order being the crucial manipulation of the task.
To summarize, the findings of the present study have suggested a specific involvement of the right DLPFC in the inhibitory processing of emotional information in a sample of healthy female volunteers. More specifically, HF stimulation of the right DLPFC caused impairments in the ability to inhibit negative information, in line with a characteristic cognitive vulnerability found in depressive pathology, whereas HF-rTMS of the left DLPFC did not lead to significant changes in inhibitory control. Because alternations in the inhibitory processing of negative information after one stimulation session were not accompanied by immediate changes in self-reported mood, this study might also have offered a first glance at possible causal effects of the frequently reported therapeutic effects of rTMS treatment in depressive pathology.
Acknowledgements
The authors are grateful to Professor Hugo D'haenen for his assistance with the study design and regret his sudden death. This research was supported by a grant from the Scientific Fund W. Gepts AZ-VUB.
Declaration of Interest
None.
