Neural dissociation of automatic and controlled temporal preparation by transcranial magnetic stimulation
Ángel Correaa,b*, Giorgia Conac, Sandra Arbulad, Antonino Vallesid,e, Patrizia Bisiacchic,e.
a Centro de Investigación Mente, Cerebro y Comportamiento, Universidad de Granada, Granada
18071, Spain.
b Departamento de Psicología Experimental, Universidad de Granada, Granada 18071, Spain.
c Dipartimento di Psicologia Generale. Università degli Studi di Padova, Padova 35100, Italy.
d Dipartimento di Neuroscienze: SNPSRR, Università degli Studi di Padova, Padova 35100, Italy.
e Centro di Neuroscienze Cognitive, University of Padua, Padova 35100, Italy.
* Corresponding author. Tel.: +34958247881; fax: +34958246239.
E-‐mail address: [email protected]; http://www.ugr.es/~act
NOTICE: This is the authors’ version of a work that was accepted for publication in the Neuropsychologia journal, and can be used for scholarly non-‐commercial purposes. Changes resulting from the publishing process, such as editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Definitive version will be subsequently published by the journal.
Correa, A., Cona, G., Arbula, S., Vallesi, A. & Bisiacchi, P. (in press). Neural dissociation of automatic and controlled temporal preparation by transcranial magnetic stimulation. Neuropsychologia.
Abstract
Recent neuropsychological evidence suggested a role for the right prefrontal cortex in temporal orienting of attention guided by symbolic cues, and the left prefrontal cortex in preparation guided by rhythms. We tested this hypothesis by comparing the effects of 1-‐Hz repetitive transcranial magnetic stimulation (TMS) over prefrontal regions on the performances of two temporal preparation tasks, one using symbolic cues (short vs. long lines) and the other using regular rhythms (fast vs. slow pace) to indicate when (early vs. late) a target would be most likely to appear. Stimulation site was either the left dorsolateral prefrontal cortex (DLPFC), right DLPFC, or sham condition. The results showed that frontal TMS produced differential effects as a function of type of cuing. In symbolic cuing, TMS on either left or right frontal sites (vs. sham) increased temporal orienting effects by reducing reaction times in valid trials. In rhythmic cuing, however, frontal TMS did not influence performance. This dissociation between two forms of temporal preparation suggests a specific role for the DLPFC in the ability of temporal orienting, but not in preparation guided by rhythms.
Keywords: TMS, temporal orienting, rhythm, DLPFC, foreperiod, enhancement
INTRODUCTION
Recent evidence suggests that different sources of temporal information are exploited by specialized neural mechanisms to generate predictions and optimize behaviour, that is, “temporal preparation” (see Nobre, Correa, & Coull, 2007, for a review). The field of temporal preparation thus includes a variety of phenomena where neurobehavioural function is influenced by predictions based on specific temporal relationships between events.
The temporal orienting of attention is an example of such temporal preparation phenomena. Research on temporal orienting has revealed enhanced performance and brain activity for a stimulus appearing after an expected rather than an unexpected interval (Coull, Frith, Buchel, & Nobre, 2000; Coull & Nobre, 1998; Coull, Davranche, Nazarian, & Vidal, 2013; Davranche, Nazarian, Vidal, &
Coull, 2011; for reviews see Correa, 2010; Coull & Nobre, 2008). In a temporal orienting procedure, expectations are based on the presentation of a symbolic cue, which informs on the duration of the “foreperiod” (preparatory interval between cue and target stimulus). The temporal orienting effect (or ‘validity effect) can be measured by comparing behavioural responses to target stimuli appearing after validly vs. invalidly cued foreperiods. That is, the cue facilitates preparation by predicting when the target is most likely to appear.
Other studies using temporally regular, isochronous sequences of stimuli (hereafter referred to as “rhythms”) have also reported changes in performance and brain activity as a function of the matching between the inter-‐onset interval (interval between the onset of two stimuli forming the rhythmic sequence) and the foreperiod (Barnes & Jones, 2000; Bolger, Coull, & Schön, 2014; Breska &
Deouell, 2014; Correa & Nobre, 2008; Cravo, Rohenkohl, Wyart, & Nobre, 2013;
Doherty, Rao, Mesulam, & Nobre, 2005; Jones, Moynihan, MacKenzie, & Puente, 2002; Lange, 2009, 2010; Sanabria, Capizzi, & Correa, 2011; Sanabria & Correa, 2013; Schmidt-‐Kassow, Schubotz, & Kotz, 2009; Schwartze, Rothermich, Schmidt-‐Kassow, & Kotz, 2011). The dynamic attending theory (Large & Jones, 1999) can explain these findings by assuming that regularity provided by
rhythmic stimulation entrains attention to focus on points in time matching the structure of the rhythm.
These two forms of temporal preparation, temporal orienting and rhythms, have been dissociated in behavioural (Rohenkohl, Coull, & Nobre, 2011) and event-‐
related potentials (ERP) studies (Breska & Deouell, 2014). For example, preparation is impaired when a working memory task is performed concurrently with a temporal orienting task (Capizzi, Correa, & Sanabria, 2013; Capizzi, Sanabria, & Correa, 2012). In contrast, preparation based on rhythms is not interfered by a concurrent working memory task in a dual task design (de la Rosa, Sanabria, Capizzi, & Correa, 2012). These findings suggest the involvement of controlled processing in temporal orienting, and automatic processing in preparation guided by rhythms (Logan, 1979). Highly similar dissociations have been reported in the time perception literature, for example, between absolute (single interval) and beat-‐based (sequential) timing mechanisms (Teki, Grube, &
Griffiths, 2011; Teki, Grube, Kumar, & Griffiths, 2011), or between controlled vs.
automatic timing (reviewed by Koch, Oliveri, & Caltagirone, 2009). Note, however, that the use of time for prediction (temporal preparation) and for the explicit reproduction of time intervals (time perception) do not necessarily involve the same functions and neural structures (Coull et al., 2013). Therefore, the current work will focus on functions related to temporal preparation.
On the basis of current neuroimaging data, however, it is difficult to establish a clear dissociation between brain networks involved in temporal orienting vs.
rhythms. A network including subcortical structures (putamen, cerebellum), frontal (e.g. dorsolateral prefrontal cortex – DLPFC) and parietal (left intraparietal sulcus) cortices commonly shows activation associated with both temporal orienting and rhythms in separate studies (Coull & Nobre, 2008). In these studies, the role of DLPFC in temporal orienting and rhythms has received less attention with respect to parietal areas, probably because the available evidence is contradictory so far. In particular, both left and right DLPFC were shown to be active in some temporal orienting studies (Coull et al., 2013; Coull &
Nobre, 1998-‐ when comparing temporal orienting vs. rest conditions), while only
the left DLPFC was active in a rhythm experiment (Marchant & Driver, 2013);
Bolger et al. (2014) and Davranche et al. (2011) did not report any involvement of the DLPFC in rhythm and temporal orienting tasks.
The implication of the right DLPFC in temporal preparation converges with research on the neural basis of the variable foreperiod effect. In a variable foreperiod design (i.e., when the foreperiod varies randomly and equiprobably across trials), responses are typically faster for current longer foreperiods (“foreperiod effect”), and they are slower for longer preceding foreperiods, especially for current short foreperiods (asymmetric sequential effects) (Niemi &
Näätänen, 1981; Woodrow, 1914). The role of the right DLPFC in the foreperiod effect has been clearly established in a study showing that virtual lesions of this area induced by transcranial magnetic stimulation (TMS) can impair the foreperiod effect (Vallesi, Shallice, & Walsh, 2007). Neuropsychological (Vallesi, Mussoni, et al., 2007; Stuss et al., 2005) and neuroimaging (Vallesi, McIntosh, Shallice, & Stuss, 2009; see also Coull et al., 2000) studies further support this finding.
According to the studies reviewed above, two main questions emerge. Can the neural networks underlying temporal orienting vs. rhythms be dissociated by means of a single experimental design? What is the specific role of the DLPFC in these two mechanisms of temporal preparation?
The first question was addressed by a neuropsychological study (Triviño, Arnedo, Lupiáñez, Chirivella, & Correa, 2011), which found that right prefrontal patients (N = 10) could use rhythms but not symbolic cues to develop temporal preparation. In contrast, left prefrontal patients (N = 5) could use symbolic cues for temporal orienting, but they did not show significant effects of preparation guided by rhythms. However, this latter result was based on a small sample of five patients only. Therefore, the aim of the current study was to test for the double neural dissociation between type of cues driving temporal preparation and laterality of prefrontal cortex, as suggested by Triviño and colleagues
(2011). Regarding the second question, we used the TMS methodology to address the role of DLPFC in these effects in healthy participants.
The current research applied TMS in a within-‐subjects design, where participants received stimulation on either left DLPFC, right DLPFC or sham condition, in different days to allow a wash out period and therefore avoid TMS carry over effects. We used the two temporal preparation tasks as in Triviño and colleagues (2011), temporal orienting and rhythms. Hence, following the classic
“virtual lesion” model (Pascual-‐Leone, Bartres-‐Faz, & Keenan, 1999) our predictions were based on their neuropsychological results. That is, by considering that TMS produces local virtual lesions, we expected that the reduced cortical excitability typically observed after low-‐frequency (1 Hz) stimulation would simulate the behavioural deficit as reported after a lesion in the same brain structure. Specifically, our main hypothesis considered that inhibition of the right DLPFC by TMS should impair temporal orienting but not rhythm effects. Additionally, this study allowed us to test further the role of left DLPFC in preparation guided by rhythms. On the basis of our previous results from five patients, we expected that left prefrontal TMS should impair rhythm but not temporal orienting effects. To summarize, we expected an interaction between Cue, TMS site and Validity, such that validity effects should be reduced after right vs. sham TMS for symbolic rather than rhythm cues, while validity effects should be reduced after left vs. sham TMS for rhythm rather than symbolic cues.
METHODS
Participants
Twelve volunteer participants took part in the experiment (4 females; mean age:
26.5; range: 22-‐42). All participants were right-‐handed and were checked for TMS exclusion criteria (Rossi, Hallett, Rossini, Pascual-‐Leone, & Safety of TMS Consensus Group, 2009). All had normal or corrected-‐to-‐normal vision and no auditory or neurological impairment. They gave informed written consent, and were paid 30 euro for their participation. The study was conducted in
accordance with the declaration of Helsinki, and was approved by the ethical committee of the Department of General Psychology, University of Padua.
Stimuli and Procedure
Stimulus presentation and data recording were controlled by E-‐prime software (Schneider, Eschman, & Zuccolotto, 2002). Participants were seated in front of a colour monitor screen, with a distance of approximately 57 cm. Two temporal preparation tasks, one with rhythmic cues and the other with symbolic cues, were administered in counterbalanced order across participants. Each task lasted about 10–15 min. Since the tasks were similar to those utilized in Triviño and colleagues’ study, in the present work only essential information on task stimuli and procedure is provided (please see Triviño et al., 2011, for more details).
In both Symbolic and Rhythm cue tasks, each trial included a fixation point, a temporal cue, and a target, all presented at the centre of a black screen. In the Symbolic cue task, the temporal cue consisted of either a short or long red line.
Participants were instructed that the short and the long lines indicated an early (after 400 ms) and a late (after 1400 ms) occurrence of the target, respectively.
In the Rhythm cue task, the temporal cue comprised two horizontal red lines of the same length, which appeared simultaneously, disappearing 5 times at either a short (i.e. every 400 ms) or long (i.e. every 1400) pace. Participants were instructed to expect the target appearing early when the rhythm pace was fast, and late when the rhythm was slow.
In both tasks, the target was either the letter ‘O’ or the letter ‘X’, which were presented with equal probability. Participants had to press the ‘B’ key as soon as they detected any of the two letters. Therefore, unlike the tasks used in Triviño et al.’s study (2011) that comprised both go and no-‐go target, this experiment included only go targets to simplify the design. However, in order to be able to compare the results with our previous studies, we decided to keep using both letters instead of just one. In both tasks, each trial began with the fixation point presented for a random interval (range: 500 -‐ 1500 ms). In the Symbolic cued
task, a 50 ms long temporal cue (short or long red line) was followed by a blank screen lasting either for 350 or 1350 ms based on the foreperiod on that trial. In the Rhythm cue task, the temporal cue (two lines) appeared for 50 ms and disappeared five times for 350 or 1350 ms (depending on the rhythm pace condition). On the last appearance of the two lines they became thicker to warn participants about the upcoming target. This additional warning of the rhythmic condition was already included in the previous neuropsychological study to facilitate temporal preparation for patients (Triviño et al. 2011), and was held in the current study just for comparison purposes across studies. Note, however, that we have found similar patterns of preparation regardless of the presence/absence of this final warning (Sanabria et al., 2011; Sanabria & Correa, 2013). After this, the screen remained blank for 350 or 1350 ms as in the Symbolic cue task. Then, the target appeared for 100 ms and was followed by a blank screen until the participant’s response or for a maximum duration of 2000 ms. Each trial ended with a 500 ms black screen.
Both Symbolic and Rhythm cue tasks included one practice block and 4 experimental blocks. The practice block comprised 16 ‘early’ cues followed by 16
‘late’ cues. All cues were valid to encourage participants to rely on the predictive value of the cues. In each of the 4 experimental blocks the valid cues were 75% of the total and the temporal expectancy was manipulated across blocks to optimise temporal orienting effects (Correa, Lupiáñez, & Tudela, 2006). In half of the blocks, the cue indicated that the target would likely appear after 400 ms (‘early’ cue condition), while in the other half the cue indicated that the target would likely appear after 1400 ms (‘late’ cue condition) with respect to cue onset. ‘Early’ and ‘late’ cue blocks were alternated and presented in a counterbalanced order across participants. Each experimental block included 32 randomly presented trials.
TMS protocol
Stimulation was applied using a Magstim 200 magnetic stimulator model (Magstim, UK) equipped with a figure-‐of-‐eight coil (70mm outer diameter).
The coil was placed on the scalp of participants based on their MRI images by using the Brainsight stereotaxic neuronavigation system (Rogue Research, Inc., Montreal, Canada) coupled with a Polaris Vicra infrared camera system (NDI, Waterloo, Canada). The coil position was monitored on-‐line during the stimulation. Participants wore also a swimming cap on which the location found in the coregistration procedure was marked with a coloured spot. A chin support was given during TMS stimulation to prevent or reduce head movements.
Stimulation intensity used during the experiment was set at 100% of each participant’s motor threshold. Motor threshold was determined at the optimal scalp position over the hand area of the left and right motor cortices, and was defined as the minimum intensity that can induce a visible twitch of a muscle of the hand contralateral to the stimulated motor cortex at least five out of ten times (Rossini et al., 1994). The resulting mean intensity was 57% of the maximum stimulator output (range 52–65%).
For each participant, target stimulation sites were the right and the left DLPFC.
The identification of the target position was driven by previous TMS studies that have stimulated DLPFC regions (e.g., Hasan, Galea, Casula, Falkai, Bestmann, &
Rothwell, 2013; Rizzo, Sandrini, & Papagno, 2007; Sandrini, Rossini, & Miniussi, 2008), and then slightly adjusted according to individual MRI scans. The resulting mean Talairach coordinates of the stimulation points across participants were: X=±37, Y=30, Z=38, corresponding to right and left DLPFC regions (Figure 1). Sham stimulation site corresponded to Fz according to the 10–20 International System.
Figure 1. Anatomical location of the three sites of magnetic stimulation (right DLPFC, left DLPFC and central, sham condition).
In the sham stimulation the coil was angled slightly off the head such that the superficial scalp muscles were activated to simulate the sensation and acoustic artifacts of TMS. This was accomplished by tilting the coil with the two wings of the figure-‐eight coil touching the scalp (two-‐wing sham) at an angulation of 45 degrees from the plane tangential to the scalp (Lisanby, Gutman, Luber, Schroeder, & Sackeim, 2001). The stimulation site order (left DLPFC, right DLPFC, sham) was counterbalanced across participants. The three stimulation sessions were run on different days, with at least one day between sessions. At the end of the last session, participants reported not to have perceived relevant differences between experimental and sham stimulations.
An off-‐line TMS paradigm was chosen rather than an online one to avoid any influence of the proprioceptive sensation and the sound given by the TMS on the RTs (Terao et al., 1997; see also Vallesi, Shallice, et al., 2007) and thus on the foreperiod-‐related phenomena. In each session, a single 10-‐min train of 1 Hz repetitive TMS (rTMS) was delivered, resulting in a total of 600 pulses.
Immediately after the TMS, the cognitive tasks were presented. For each participant, the order between Symbolic and Rhythm cue tasks was kept
constant across the TMS sessions. This order was however counterbalanced across participants.
Data analysis
The influence of frontal TMS on temporal orienting and rhythm effects was measured by a repeated-‐measures ANOVA with Cue (symbolic, rhythm), TMS Site (sham, left DLPFC, right DLPFC) and Validity of cue (valid, invalid) as within-‐
subject factors, and mean RT as dependent variable. As it is usual in temporal orienting research (e.g., Nobre, 2001; Triviño et al., 2011), the analysis focused on short foreperiod data, given that validity effects are larger or unique to the short foreperiod. This analysis did not include practice trials, anticipations (2.84%), misses (0.42%), and trials with RTs below 100 ms and above 1200 ms (1.97%).
Given that the current design involved variable foreperiod durations, it was also possible to test the influence of frontal TMS on foreperiod and sequential effects by a TMS (sham, left DLPFC, right DLPFC) x Current foreperiod (short – 400 ms, long – 1400 ms) x Previous foreperiod (short, long) ANOVA, by collapsing the two cue conditions. Note, however, that this foreperiod manipulation was inserted in a temporal cuing paradigm and was not specifically designed to measure foreperiod and sequential effects, so that results may differ from classical research using neutral warning signals (see Capizzi et al., 2013, for a discussion on this issue). In order to study sequential effects, the first trial of each block was not included in this analysis.
RESULTS
The Cue x TMS x Validity ANOVA showed a main effect of cue, F(1,11) = 27.32, p
< .001, η2 = .71, with symbolic cues leading to faster RTs than rhythms (329 vs.
358 ms). The typical validity effect was replicated, as revealed by faster RTs at valid (316 ms) vs. invalid (370 ms) trials, F(1,11) = 95.86, p < .001, ηp2 = .90. The main effect of the TMS and the Cue x TMS interaction were not significant (Fs <
1). Most relevant to our predictions was the finding of a significant interaction between cue, TMS and validity, F(2,22) = 3.55, p = .046, ηp2 = .24 (Figure 2).
Figure 2. Mean reaction time (± standard error of the mean) as a function of TMS site (sham, left DLPFC, right DLPFC), type of temporal cue (symbolic, rhythmic) and cue validity (valid – white, invalid – black).
Further analyses of this interaction revealed that TMS influenced the validity effect produced by symbolic cues only, TMS x Validity: F(2,22) = 5.45, p = .01, ηp2
= .33, rather than by rhythmic cues, TMS x Validity: F < 1. Planned comparisons in the symbolic cue condition revealed larger validity effects for TMS applied on both left DLPFC, F(1,11) = 9.95, p = .01, and right DLPFC, F(1,11) = 7.97, p = .02, in relation to the sham TMS condition. There was no difference between left and right TMS, F < 1. As can be observed in Figure 2, this interaction was mainly driven by the valid condition, in which RTs were faster after TMS on either left (p
= .06) or right (p = .04) DLPFC as compared to the sham condition.
The TMS x Current foreperiod x Previous foreperiod ANOVA replicated both the typical variable foreperiod effect, as indicated by faster RTs at long (313 ms) vs.
short (337 ms) foreperiods, F(1,11) = 22.48, p < .001, ηp2 = .67, and sequential
effects (previous foreperiod: F(1,11) = 63.03, p < .001, ηp2 = .85; current x previous foreperiod: F(1,11) = 79.68, p < .001, ηp2 = .88). However, these effects were not influenced significantly by the TMS manipulation (TMS x foreperiod:
F(2,22) = 1.66, p = .21; TMS x previous foreperiod: F < 1; TMS x previous foreperiod x current foreperiod: F(2,22) = 1.56, p = .23).
DISCUSSION
A neuropsychological study by Triviño and colleagues (2011) reported that right prefrontal patients showed impaired temporal orienting, but temporal preparation was preserved when these patients were shown a rhythm as temporal cue. On the other hand, in a small group of five patients with left prefrontal lesions, temporal preparation based on rhythms was impaired, while temporal orienting was preserved. The current research aimed to test this preliminary evidence of a double dissociation by using a larger sample, exploiting the high spatial resolution of TMS to directly manipulate the activity in DLPFC regions in order to establish causal relationships between these regions and temporal preparation mechanisms, and by following a within-‐subjects design where every participant was evaluated in each of the three TMS conditions.
Specifically, we aimed to dissociate, for the first time, the neural bases of temporal orienting vs. rhythms by applying repetitive TMS on either left or right sites of the DLPFC while participants performed a task following Triviño and colleagues’ design (2011). We expected to find impaired temporal orienting after TMS on the right DLPFC, and impaired effects of rhythm after TMS on the left DLPFC. The main finding revealed that frontal TMS altered behavioural performance selectively as a function of type of temporal cuing. In symbolic cuing, frontal TMS (both on the left and right DLPFC) increased validity effects in relation to the sham condition. On the other hand, in rhythmic cuing, frontal TMS did not influence task performance.
This dissociation between temporal preparation guided by symbolic cues vs.
rhythms partially supported our hypothesis. In particular, temporal orienting is
more sensitive to frontal TMS than preparation based on rhythms. Considering the classic involvement of frontal areas in executive control, this result converges with research showing that rhythms (but not symbolic cues) can enhance preparation automatically, regardless of task instructions (Rohenkohl et al., 2011) and intentional temporal expectations (Breska & Deouell, 2014), despite the interference of a simultaneous working memory task in a dual task procedure (Capizzi et al., 2013, 2012; de la Rosa et al., 2012), and without the need of a functional right prefrontal cortex (Triviño et al., 2011). Moreover, this process of rhythmic preparation seems to develop independently of explicit timing processes. Koch and colleagues (Koch, Olivieri, Torriero, & Caltagirone, 2003) found that 1-‐Hz rTMS on the right DLPFC, but not left DLPFC, produced underestimation of intervals between 5 and 15 s. The right DLPFC has therefore been associated with controlled timing, involving attention and working memory processes, where time estimation is demanded, for example, in the context of long durations (in the seconds range) and high cognitive load. In the current study, rhythmic preparation involving timing of shorter intervals (350 and 1350 ms) was not influenced by TMS on the right DLPFC. Although comparisons across these different experimental settings may be taken with caution, a plausible explanation would assume the involvement of different timing processes underlying the usage of time for prediction (temporal preparation) and for explicit estimation, according to a recent neural dissociation between temporal reproduction and temporal orienting tasks, both using intervals in the milliseconds range (from 600 to 1400 ms; Coull et al., 2013).
The current study does not support a role of the left DLPFC in rhythmic preparation (in contrast to Triviño et al., 2011). The divergence between the previous neuropsychological and the current results may be due to differences in the spatial resolution and homogeneity of the damaged/stimulated area. Note that the left frontal group in Triviño et al.’s study included a sample of five patients with more heterogeneous lesions over the frontal cortex (and grouping was based on radiological reports rather than structural imaging) as compared to the more precise stimulation of TMS on the left DLPFC area. Moreover, other studies ascribe the divergence between neuropsychological and TMS results to
differences in neural reorganisation, thus precluding a full, unequivocal correspondence between damaged and stimulated areas (Pascual-‐Leone, Walsh,
& Rothwell, 2000; Walsh, Ellison, Battelli, & Cowey, 1998). That is, while patients may compensate for a focal lesion with plastic reorganisation of the brain, the transient nature of the TMS does not allow enough time for such a reorganisation in experimental subjects. Therefore, this differential reorganisation may increase the spatial divergence between long-‐term real lesions (where other areas may be involved to compensate for a function) and short-‐term virtual lesions.
Nevertheless, the current null result should be taken with caution as it is based on a small sample (N = 12). In any case, the finding that rhythmic preparation was not affected by frontal TMS in our specific experimental setting it is not surprising according to neuroimaging studies pointing to subcortical rather than cortical structures, such as the putamen (Geiser, Notter, & Gabrieli, 2012;
Marchant & Driver, 2013), and to more posterior rather than frontal areas, such as the premotor (Schubotz, von Cramon, & Lohmann, 2003; Schubotz & von Cramon, 2001) or parietal cortex (Bolger et al., 2014; Marchant & Driver, 2013) in the ability to anticipate events unfolding in a rhythmic structure.
On the other hand, the effect of TMS on temporal orienting followed an unexpected direction according to the classic virtual lesion approach (Pascual-‐
Leone et al., 1999), as we found larger rather than smaller validity effects after frontal stimulation. Although the current design did not include a neutral cue condition to elucidate whether larger validity effects were due to smaller costs or larger benefits, planned comparisons revealed that frontal TMS reduced RTs on the valid as compared to the invalid condition. An explanation for this counterintuitive result relies on recent conceptions on the neural effects of TMS that go beyond the ‘virtual lesion’ approach (Luber & Lisanby, 2014; Pascual-‐
Leone et al., 2000; Silvanto & Muggleton, 2008). Traditionally, the effects of TMS have been associated with disrupted performance by assuming a direct relationship between cortical excitability and behaviour. Given that low-‐
frequency repetitive TMS (e.g., 1 Hz) generally reduces cortical excitability of a potentially task-‐relevant area, it makes sense to predict that task performance would be disrupted by such stimulation parameters on this area.
However, a recent review has identified more than sixty publications reporting enhancing effects of TMS (Luber & Lisanby, 2014), thus illustrating the complexity of the relationship between the (not yet well-‐known) neurophysiological mechanisms underlying TMS effects and the translation into behavioural function. Although the TMS parameters used in the current study generally produce behavioural interference, the current findings are not an isolated instance, and several studies have found that 1-‐Hz rTMS can enhance cognitive performance in, for example, visual discrimination (Waterston & Pack, 2010) or picture naming tasks (Mottaghy, Sparing, & Töpper, 2006).
Therefore, the finding of larger validity effects in the symbolic cue condition after stimulation of the DLPFC can be interpreted as an enhancing effect of the TMS.
This effect cannot be accounted for by the ‘intersensory facilitation’ phenomenon (Terao et al., 1997), since we followed an offline protocol and the behavioural enhancements were relative to the sham condition, which was equated to the experimental conditions in sensory and motor aspects. Interestingly, the TMS enhancement was functionally specific, as it involved temporal orienting only, but not preparation guided by rhythms. The interpretation in terms of TMS enhancement therefore emphasizes the relevance of the DLPFC (and probably other areas functionally connected to it) for the temporal orienting function, and it allows reconciling the current work with previous research on the neuropsychological basis of temporal orienting (Triviño et al., 2011; Triviño, Correa, Arnedo, & Lupiáñez, 2010).
The results also revealed a lack of specificity between left and right frontal TMS on the temporal orienting effect. This may be a consequence of TMS effects not only on the area directly stimulated, but also on interconnected brain areas.
Indeed, research combining TMS and neuroimaging measures has revealed that the effects of TMS can change the activation state of a brain network including the stimulated area and other areas functionally connected (Ruff, Driver, &
Bestmann, 2009). Thus, it may be possible that left and right frontal cortices of a healthy brain naturally work in coordination during temporal orienting. This
hypothesis makes sense on the basis of the finding of bilateral activation of the DLPFC during temporal orienting in a recent study (Coull et al., 2013). Another interesting issue to be addressed in future TMS studies would be the determination of the role of other areas claimed to be fundamental for temporal orienting, such as the left IPS (Davranche et al., 2011), which probably are functionally connected to the DLPFC.
Finally, the analysis of foreperiod and sequential effects showed no effects of TMS. This result replicated previous research only partially. In Vallesi and colleagues’ study (2007), TMS on the right DLPFC impaired the foreperiod effect while sequential effects remained intact in a variable foreperiod design. A significant modulation of the foreperiod effect was not observed in the current study. As it has been argued elsewhere (Capizzi et al., 2013), the current temporal orienting design may not be an optimal procedure to study foreperiod effects. However, the automaticity of sequential effects was not disconfirmed here, which similarly to the effect of the rhythm, was not affected by TMS interference on the DLPFC. This similarity might suggest the involvement of a mechanism for automatic temporal preparation, which can be commonly triggered either by regular sequences within a trial (rhythms) or by temporal sequences of foreperiods across trials (i.e., sequential effects as a result of a rhythm across trials).
Conclusions
To our knowledge, this is the first TMS study on temporal orienting and temporal preparation guided by rhythms. The current findings show a role of both left and right DLPFC in the ability for temporal orienting. The current study did not find evidence suggesting that temporal preparation based on rhythms depends on the DLPFC, in contrast to the role of the left PFC suggested by a previous neuropsychological study (Triviño et al. 2011). These results provide novel evidence following a TMS approach to dissociate between temporal preparation guided by cues vs. rhythms.
ACKNOWLEDGEMENTS
This work was supported by a Spanish grant from the Plan Nacional I+D+i, (PSI2010-‐15399; Ministerio de Ciencia e Innovación) to A.C and by a grant from the Bial Foundation 84/12 to P.B.; A.V. and S.A. are funded by an ERC starting
grant (LEX-‐MEA, GA# 313692) to A.V.
REFERENCES
Barnes, R., & Jones, M. (2000). Expectancy, attention, and time. Cognitive Psychology, 41, 254–311.
Bolger, D., Coull, J. T., & Schön, D. (2014). Metrical rhythm implicitly orients attention in time as indexed by improved target detection and left inferior parietal activation. Journal of Cognitive Neuroscience, 26(3), 593-‐605.
doi:10.1162/jocn_a_00511
Breska, A., & Deouell, L. Y. (2014). Automatic Bias of Temporal Expectations following Temporally Regular Input Independently of High-‐level Temporal Expectation. Journal of Cognitive Neuroscience.
doi:10.1162/jocn_a_00564
Capizzi, M., Correa, A., & Sanabria, D. (2013). Temporal orienting of attention is interfered by concurrent working memory updating. Neuropsychologia.
doi:10.1016/j.neuropsychologia.2012.10.005
Capizzi, M., Sanabria, D., & Correa, A. (2012). Dissociating controlled from automatic processing in temporal preparation. Cognition, 123, 293-‐302.
doi:10.1016/j.cognition.2012.02.005
Correa, A. (2010). Enhancing behavioural performance by visual temporal orienting. En A. C. Nobre & J. T. Coull (Eds.), Attention and Time (pp. 357-‐
370). Oxford: Oxford University Press.
Correa, A., Lupiáñez, J., & Tudela, P. (2006). The attentional mechanism of temporal orienting: Determinants and attributes. Experimental Brain Research, 169(1), 58-‐68.
Correa, A., & Nobre, A. C. (2008). Neural modulation by regularity and passage of time. Journal of Neurophysiology, 100(3), 1649-‐1655.
Coull, J., Davranche, K., Nazarian, B., & Vidal, F. (2013). Functional anatomy of timing differs for production versus prediction of time intervals.
Neuropsychologia, 51(2), 309-‐319.
doi:10.1016/j.neuropsychologia.2012.08.017
Coull, J., Frith, C. D., Buchel, C., & Nobre, A. C. (2000). Orienting attention in time:
Behavioural and neuroanatomical distinction between exogenous and endogenous shifts. Neuropsychologia, 38(6), 808-‐819.
Coull, J., & Nobre, A. (2008). Dissociating explicit timing from temporal
expectation with fMRI. Current opinion in neurobiology, 18(2), 137-‐144.
doi:10.1016/j.conb.2008.07.011
Coull, J., & Nobre, A. C. (1998). Where and when to pay attention: The neural systems for directing attention to spatial locations and to time intervals as revealed by both PET and fMRI. The Journal of Neuroscience, 18(18), 7426-‐7435.
Cravo, A. M., Rohenkohl, G., Wyart, V., & Nobre, A. C. (2013). Temporal
expectation enhances contrast sensitivity by phase entrainment of low-‐
frequency oscillations in visual cortex. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 33(9), 4002-‐4010.
doi:10.1523/JNEUROSCI.4675-‐12.2013
Davranche, K., Nazarian, B., Vidal, F., & Coull, J. (2011). Orienting attention in time activates left intraparietal sulcus for both perceptual and motor task goals. Journal of Cognitive Neuroscience, 23(11), 3318-‐3330.
doi:10.1162/jocn_a_00030
De la Rosa, M. D., Sanabria, D., Capizzi, M., & Correa, A. (2012). Temporal preparation driven by rhythms is resistant to working memory
interference. Frontiers in Cognition, 3, 308.
doi:10.3389/fpsyg.2012.00308
Doherty, J. R., Rao, A., Mesulam, M. M., & Nobre, A. C. (2005). Synergistic effect of combined temporal and spatial expectations on visual attention. The Journal of Neuroscience, 25, 8259-‐8266.
Geiser, E., Notter, M., & Gabrieli, J. D. E. (2012). A corticostriatal neural system enhances auditory perception through temporal context processing. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 32(18), 6177-‐6182. doi:10.1523/JNEUROSCI.5153-‐11.2012
Jones, M., Moynihan, H., MacKenzie, N., & Puente, J. (2002). Temporal aspects of stimulus-‐driven attending in dynamic arrays. Psychological Science, 13(4), 313-‐319.
Koch, G., Oliveri, M., & Caltagirone, C. (2009). Neural networks engaged in milliseconds and seconds time processing: evidence from transcranial magnetic stimulation and patients with cortical or subcortical
dysfunction. Philosophical Transactions of the Royal Society of London.
Series B, Biological Sciences, 364(1525), 1907-‐1918.
doi:10.1098/rstb.2009.0018
Koch, G., Olivieri, M., Torriero, S., & Caltagirone, C. (2003). Underestimation of time perception after repetitive transcranial magnetic stimulation.
Neurology, 60, 1844-‐1846.
Lange, K. (2009). Brain correlates of early auditory processing are attenuated by expectations for time and pitch. Brain and Cognition, 69(1), 127-‐137.
doi:10.1016/j.bandc.2008.06.004
Lange, K. (2010). Can a regular context induce temporal orienting to a target sound? International Journal of Psychophysiology: Official Journal of the International Organization of Psychophysiology, 78(3), 231-‐238.
doi:10.1016/j.ijpsycho.2010.08.003
Large, E. W., & Jones, M. R. (1999). The dynamics of attending: How we track time varying events. Psychological Review, 106, 119-‐159.
Lisanby, S. H., Gutman, D., Luber, B., Schroeder, C., & Sackeim, H. A. (2001). Sham TMS: intracerebral measurement of the induced electrical field and the induction of motor-‐evoked potentials. Biological Psychiatry, 49(5), 460-‐
463.
Logan, G. D. (1979). On the use of a concurrent memory load to measure attention and automaticity. Journal of Experimental Psychology: Human Perception and Performance, 5(2), 189-‐207. doi:10.1037/0096-‐
1523.5.2.189
Luber, B., & Lisanby, S. H. (2014). Enhancement of human cognitive performance using transcranial magnetic stimulation (TMS). NeuroImage, 85 Pt 3, 961-‐
970. doi:10.1016/j.neuroimage.2013.06.007
Marchant, J. L., & Driver, J. (2013). Visual and Audiovisual Effects of Isochronous Timing on Visual Perception and Brain Activity. Cerebral Cortex, 23(6), 1290-‐1298. doi:10.1093/cercor/bhs095
Mottaghy, F. M., Sparing, R., & Töpper, R. (2006). Enhancing picture naming with transcranial magnetic stimulation. Behavioural Neurology, 17(3-‐4), 177-‐
186.
Niemi, P., & Näätänen, R. (1981). Foreperiod and simple reaction time.
Psychological Bulletin, 89, 133-‐162.
Nobre, A. C. (2001). Orienting attention to instants in time. Neuropsychologia, 39, 1317-‐1328.
Nobre, A. C., Correa, A., & Coull, J. T. (2007). The hazards of time. Current Opinion in Neurobiology, 17, 1-‐6.
Pascual-‐Leone, A., Bartres-‐Faz, D., & Keenan, J. P. (1999). Transcranial magnetic stimulation: studying the brain-‐behaviour relationship by induction of
«virtual lesions». Philosophical Transactions of the Royal Society of London.
Series B, Biological Sciences, 354(1387), 1229-‐1238.
doi:10.1098/rstb.1999.0476
Pascual-‐Leone, A., Walsh, V., & Rothwell, J. (2000). Transcranial magnetic stimulation in cognitive neuroscience-‐-‐virtual lesion, chronometry, and functional connectivity. Current Opinion in Neurobiology, 10(2), 232-‐237.
Rohenkohl, G., Coull, J. T., & Nobre, A. C. (2011). Behavioural dissociation
between exogenous and endogenous temporal orienting of attention. PloS One, 6(1), e14620. doi:10.1371/journal.pone.0014620
Rossi, S., Hallett, M., Rossini, P. M., Pascual-‐Leone, A., & Safety of TMS Consensus Group. (2009). Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 120(12), 2008-‐2039.
doi:10.1016/j.clinph.2009.08.016
Ruff, C. C., Driver, J., & Bestmann, S. (2009). Combining TMS and fMRI: from
«virtual lesions» to functional-‐network accounts of cognition. Cortex; a Journal Devoted to the Study of the Nervous System and Behavior, 45(9), 1043-‐1049. doi:10.1016/j.cortex.2008.10.012
Sanabria, D., Capizzi, M., & Correa, A. (2011). Rhythms that speed you up. J Exp Psychol Hum Percept Perform, 37, 236-‐244.
Sanabria, D., & Correa, Á. (2013). Electrophysiological evidence of temporal preparation driven by rhythms in audition. Biological Psychology, 92(2), 98-‐105. doi:10.1016/j.biopsycho.2012.11.012
Schmidt-‐Kassow, M., Schubotz, R. I., & Kotz, S. A. (2009). Attention and entrainment: P3b varies as a function of temporal predictability.
Neuroreport, 20(1), 31-‐36. doi:10.1097/WNR.0b013e32831b4287 Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-‐Prime user’s guide.
Pittsburgh: Psychology Software Tools Inc.
Schubotz, R. I., & von Cramon, D. Y. (2001). Functional organization of the lateral premotor cortex: fMRI reveals different regions activated by anticipation of object properties, location and speed. Cognitive Brain Research, 11, 97-‐
112.
Schubotz, R. I., von Cramon, D. Y., & Lohmann, G. (2003). Auditory what, where, and when: a sensory somatotopy in lateral premotor cortex. Neuroimage, 20(1), 173-‐185.
Schwartze, M., Rothermich, K., Schmidt-‐Kassow, M., & Kotz, S. A. (2011).
Temporal regularity effects on pre-‐attentive and attentive processing of deviance. Biological Psychology, 87(1), 146-‐151.
doi:10.1016/j.biopsycho.2011.02.021
Silvanto, J., & Muggleton, N. G. (2008). New light through old windows: moving beyond the «virtual lesion» approach to transcranial magnetic
stimulation. NeuroImage, 39(2), 549-‐552.
doi:10.1016/j.neuroimage.2007.09.008
