Modulating lexical and semantic processing by transcranial direct current stimulation

Abstract

Here we aim to evaluate the ability of transcranial direct current stimulation (tDCS), which is applied over Wernickeâ[euro](TM)s area and its right homologue, to influence lexical decisions and semantic priming and establish an involvement for temporo-parietal areas in lexical and semantic processing. Thirty-two subjects (17 women) completed a lexical decision task and a semantic priming task while receiving 20Â min of bilateral tDCS stimulation (right anodal/left cathodal or left anodal/right cathodal stimulation) or sham stimulation. We hypothesized that right anodal/left cathodal stimulation over temporo-parietal areas would selectively interrupt the typical lexical processing dominance of the left hemisphere and facilitate mediated priming, while left anodal/right cathodal stimulation would selectively facilitate lexical processing and direct priming. Results showed impaired lexical processing under right anodal/left cathodal stimulation in comparison with sham and left anodal/right cathodal stimulation. Results are discussed in light of previous findings and hemispheric lateralization models.[PUBLICATION ABSTRACT]

Exp Brain Res (2013) 226:121135 DOI 10.1007/s00221-013-3416-5

RESEARCH ARTICLE

Modulating lexical and semantic processing by transcranial direct current stimulation

Keren Weltman Michal Lavidor

Received: 2 May 2012 / Accepted: 9 January 2013 / Published online: 1 February 2013 Springer-Verlag Berlin Heidelberg 2013

Introduction

The left hemispheres (LH) tendency toward language was one of the earliest observations of brain asymmetry and was well established during the twentieth century by various methodologies (Beeman and Chiarello 1998). However, in the last few decades, accumulating evidence suggests that the right hemisphere (RH) has a signicant role in lexical and semantic processing (Lindell 2006; Mitchell and Crow 2005).

The LH specialization for word recognition may be due to any number of cognitive factors. Orthographic, phonological, and semantic characteristics of a word may all affect the uency of its retrieval, as evidenced by both behavioral and physiological research (Binder et al. 2003). Several models have been proposed in order to explain the psycholinguistic effects on word retrieval and how they may affect the two cerebral hemispheres differently. Chiarello and Hoyer (1988) describe how lexical decision is characterized by prelexical, lexical, and post-lexical processing stages. The pre- and post-lexical stages involve stimulus encoding and response making, respectively, and are not believed to differ between the cerebral hemispheres. However, the lexical stage involves matching the lexical stimulus with stored entries, which is termed lexical access, an event that may well be LH specialized (Zouridakis et al. 1998). It is during lexical access that semantic and phono-logical characteristics have their most clear impact, perhaps in different ways for the two hemispheres. Phonological processing has also been demonstrated to differ between the hemispheres, with the LH demonstrating a greater capability for phonological output (Chiarello and Hoyer 1988; Iacoboni and Zaidel 1996).

There is evidence that hemispheric differences also affect semantic (and not only phonological) processes in

Abstract Here we aim to evaluate the ability of transcranial direct current stimulation (tDCS), which is applied over Wernickes area and its right homologue, to inuence lexical decisions and semantic priming and establish an involvement for temporo-parietal areas in lexical and semantic processing. Thirty-two subjects (17 women) completed a lexical decision task and a semantic priming task while receiving 20 min of bilateral tDCS stimulation (right anodal/left cathodal or left anodal/right cathodal stimulation) or sham stimulation. We hypothesized that right anodal/left cathodal stimulation over temporo-parietal areas would selectively interrupt the typical lexical processing dominance of the left hemisphere and facilitate mediated priming, while left anodal/right cathodal stimulation would selectively facilitate lexical processing and direct priming. Results showed impaired lexical processing under right anodal/left cathodal stimulation in comparison with sham and left anodal/right cathodal stimulation. Results are discussed in light of previous ndings and hemispheric lateralization models.

Keywords tDCS Wernicke's area Lexical Semantic

Language

K. Weltman M. Lavidor (*)

Department of Psychology, Bar-Ilan University, 52900 Ramat Gan, Israele-mail: [email protected]

M. Lavidor

The Gonda Multidisciplinary Brain Research Center, Bar-Ilan University, Ramat Gan, Israel

M. Lavidor

Department of Psychology, University of Hull, Hull HU6 7RX, UK

1 3

122 Exp Brain Res (2013) 226:121135

word recognition. One of the most common methods of examining semantic processing is the semantic priming paradigm. Target words which are presented after semantically related words are processed more quickly than when they follow unrelated words. This effect is called the semantic priming effect. It is a rich source of information about the mental lexicon and is used in order to investigate how words are accessed, comprehended, and integrated with other words in sentence comprehension (Neely 1991).

Mediated priming (also known as two-step or indirect priming) is a facilitated response to a target (e.g., lion) following a prime (e.g., stripes) that is indirectly related to the target via a connecting mediator (e.g., tiger) (Balota and Lorch 1986). In the brain, temporal and inferior prefrontal brain regions have been found to be consistently related to semantic priming, but whereas left regions are related to classic semantic priming, right or bilateral activations are related to mediated priming. Tivarus et al. (2006) investigated the effect of varying semantic distances by using controlled semantic priming with a long stimulus onset asynchrony (SOA = 700 ms). They found

effects of direct and indirect priming on behavioral as well as on neural levels: The direct and indirect semantic relationship modulated brain activation within left inferior frontal and bilateral fronto-temporal regions, respectively. In another study, Kuperberg et al. (2007) investigated controlled lexico-semantic processing of directly and indirectly related words with an SOA of 800 ms. They reported a widespread bilateral temporo-occipital response suppression for indirectly related words, whereas directly related words mainly led to inferior left prefrontal and temporal response suppression. Sass et al. (2009) used a short SOA of 350 ms, and while no behavioral mediated priming was found, they found neural correlates between directly related pairs and left-lateralized activations in fronto-temporo-parietal areas and between indirect priming and right-hemispheric fronto-parietal signal changes.

One of the most inuential conceptualizations of lateralized semantic processing was made by Jung-Beeman (2005) who suggested the Bilateral semantic Activation, Integration and Selection (BAIS) model as a theoretical framework for understanding the latest ndings on brain asymmetry. According to the model, bilateral semantic processes of activation, integration, and selection interact in order to process language. These processes occur in qualitatively different ways in each hemisphere: While semantic activation in the LH is ne in its nature (it rapidly focuses on dominant features which are tightly linked to the input), semantic activation in the RH is coarser (it entails weak activation of multiple concepts which are remotely

associated with the input). In accordance with this view, a RH advantage for remote semantic relations has been consistently reported.

Previous divided visual eld semantic priming studies in which the target was presented to either the left or the right visual eld (for a review, see Beeman and Chiarello 1998) found facilitatory effects in comparison with a neutral control condition for related word pairs in both hemispheres, but inhibitory effects for unrelated pairs only in the LH (Chiarello and Hoyer 1988). Hence, RH semantic processing can be characterized by a lack of inhibition for irrelevant semantic information. In a later study, Chiarello and Richards (1992) assessed semantic priming for words that are members of the same category, but are not strongly associated. They observed a semantic priming effect in the RH, not the LH. Nakagawa (1991) further examined hemispheric specialization in semantic processing by varying semantic distance between prime and target word. Nakagawa found that close associates elicited signicant facilitatory effects in either hemisphere. In contrast, strong inhibitory effects were found for remote or unrelated targets in the LH, whereas there was no sign of inhibition in the RH (see also Beeman et al. 1994). In addition, priming effects were obtained for directly related targets in either hemisphere; in contrast, indirectly related words were facilitated only in the RH, whereas there was a tendency for inhibition in the LH (Weisbrod et al. 1998).

In contrast to these ndings, other studies have found no difference between the RH and the LH in mediated priming (Chiarello and Richards 1992). Further evidence for hemispheric specialization comes from studies in people with RH brain damage, which consistently report specic abnormalities in semantic activation tasks as well as in natural language tasks which require semantic integration (Mitchell and Crow 2005).

Another line of evidence includes brain imaging studies. These studies often reveal a strong neural activity in the LH when it comes to language processing tasks, but also a weaker signal in right homologous regions (Bookheimer 2002). Furthermore, imaging studies and especially fMRI increasingly report greater RH versus LH activity while subjects perform higher-level language tasks (Jung-Beeman 2005). More specically, a semantic priming study which examined brain activity during semantic processing by using ERP found a semantic priming effect in the inferior fronto-temporal area bilaterally for directly semantic related words, but only in the RH for indirectly semantic related words (Kiefer et al. 1998). Also, as reviewed above, fMRI studies have found that directly linked words led to left-lateralized activations in fronto-temporo-parietal areas (Kuperberg et al. 2007; Sass et al. 2009; Tivarus et al. 2006) and indirect priming

1 3

Exp Brain Res (2013) 226:121135

123

led to right-hemispheric fronto-parietal signal changes (Sass et al. 2009).

An additional, newer source of evidence is noninvasive brain stimulation (NIBS) such as transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS), a recently revived noninvasive brain stimulation technique. In the healthy brain, NIBS can be used in order to delineate the functional signicance of a particular brain region for the respective task under study (Floel et al. 2011). NIBS methods have been proven to effectively inuence cognitive processes, including several language functions. Transcranial magnetic stimulation over left temporo-parietal language areas (i.e., Wernickes area, which is considered as the long-term storage of the conceptual and semantic knowledge; Andoh et al. 2008; Mottaghy et al. 2006) has been reported to modulate performance in different types of language tasks, including semantic ones. In addition, magnetic stimulation of the right homologue to Wernickes area successfully modulated semantic functions (Pobric et al. 2008). Also, tDCS over the left peri-sylvian area (i.e., Wernickes area) has been recently reported to improve language learning in healthy participants (Floel et al. 2008).

In addition, TMS over the left inferior prefrontal cortex (LIPC) impairs high-order semantic processing (Whitney et al. 2011) and repetition priming (Thiel et al. 2005). However, semantic priming effects under noninvasive brain stimulation such as TMS or tDCS have not been previously reported.tDCS utilizes persistent direct current stimulation as opposed to the phasic electrical responses which are initiated by the TMS coil and is applied by using scalp electrodes which pass the electrical current between a positively charged anode and a negatively charged cathode. It is a noninvasive, painless cortical stimulation technique (Nitsche and Paulus 2000). Depending on the polarity of the current ow, brain excitability can either be increased or be decreased. Anodal stimulation is considered to produce an enhancement of cortical excitability, while cathodal stimulation, at least when applied over motor cortex, appears to have the opposite effect (Nitsche and Paulus 2000).

In light of these previous studies, the main aim of the present study was to evaluate the ability of tDCS as applied over Wernickes area and its right homologue to inuence lexical access and semantic priming and to establish an involvement for temporo-parietal areas in processing semantic relations. Bilateral stimulation was useddirect current stimulation over the left and right temporo-parietal areas simultaneously. This stimulation method enabled us to control the hemispheric dominance by presumably increasing cortical excitability in right language areas while decreasing it on the left, and vice versa.

We hypothesized that right anodal/left cathodal stimulation would interrupt the lexical processing dominance of the LH (Lieber 1976). Specically, we predicted poorer lexical performance under active right anodal/left cathodal stimulation when compared to sham and left anodal stimulation. We also expected to nd an opposite effect for the left anodal/right cathodal stimulation due to the LHs assumed superiority in lexical processing. We expected the stimulation to affect not only lexical processing, but also the semantic and association processing, in particular increased mediated priming under right anodal/left cathodal stimulation based on the reviewed previous ndings of RH involvement in mediated priming. Note that the unrelated baseline condition of the semantic priming paradigm may differ from neutral condition and from pure lexical decision tasks, as it is composed of additional underlying processes such as semantic matching and other strategic processes (McNamara 2005) and even inhibition (Lecardeur et al. 2007). This embedded inhibition, which differs the unrelated condition from a neutral condition (i.e., a single lexical decision trial), might be selectively affected by tDCS due to its polarity-sensitive effects (Jacobson et al. 2012).

Methods

Subjects

Thirty-two subjects (17 women) were randomly divided into two groups of 16. In the rst group (9 women and 7 men), the mean age (SD) was 25.4 years (2.4 years) and the mean education was 13.4 years (1.5 years). In the second group (8 women and 8 men), the mean age was 25.8 years (6.6) and the mean education was 13.2 years (0.7 years). All participants were right-handed, native speakers of Hebrew, with normal or corrected-to-normal vision and no known learning disabilities or ADHD. All were naive regarding the nature of the experiment and gave a written informed consent before taking part in the study, which was approved by the Bar-Ilan IRB committee.

Stimuli and design

Each subject participated in two sessionsa sham (active placebo) session and an active stimulation session. Each session began with practice trials, followed by 20 min of stimulation in one of the tDCS conditions and divided into 11 min of stimulation and 6 min of stimulation during task performance: a lexical decision task and a semantic priming task in a counterbalanced order.

A mixed design was employed, with stimulation condition (right anodal/left cathodal and left anodal/right

1 3

124 Exp Brain Res (2013) 226:121135

cathodal) as the between-subject factor. For the lexical decision task, stimuli lexicality and stimulation type (active and sham) were the within-subject variables. For the semantic priming task, the relation type between the words (direct relation, indirect mediated relation, and no relation) and stimulation type (active and sham) served as the within-subject variables.

Lexical decision task

The stimuli consisted of 96 visually presented ve-letter stringshalf of the targets were real Hebrew words (as above), and the other half were pseudo-words which were spelled accordingly. The stimulus pool was divided into two equally long lists that were presented in two separate blocks and equated in log word frequency [list 10.75 (0.7), list 21 (0.75)] and concreteness levels [list 14.8 (1.6), list 24.7 (1.6)].

Semantic priming task

The stimuli consisted of 234 visually presented pairs of three- to six-letter strings (primes and targets). Primes were always Hebrew words. Half of the targets were real Hebrew words, and the other half were pronounceable pseudo-words that were spelled accordingly. The pseudo-words were derived from real words by substituting one or two letters. Each pair belonged to one of four relation types: direct (wool-sheep, rain-cloud), mediated (snow-black, barrel-grapes), unrelated, and word/ pseudo-word pairs. There was a 1:1 ratio of word targets to pseudo-word targets over the duration of the task. The pairs were divided into two separate lists or blocks which were equated in average word length, log word frequency, and concreteness score. The matched lists were used in the active and sham sessions in a counterbalanced order. Each list consisted of all four relation types, and target words were matched across relation types for concreteness, log word frequency, and word length. The matched mean lexical properties of the stimulus lists are presented in Table 1. Overall, 39 stimuli were presented per condition. For the benet of Hebrew researchers, the matched word pairs are presented in Appendix, with English translation of all the word pairs.

All the stimulus material was drawn from a Hebrew association database (Rubinsten et al. 2005). Direct relations were drawn from the database, while only those pairs in which none of the subjects produced the target word were accepted as mediated and related and were included in the stimulus list. Mediated (or indirectly) related word pairs were dened as strongly related to a strongly related concept. Hence, in all the mediated

pairs, both the association between the prime and the mediator and the association between the mediator and the target were strong (e.g., snow-white-black). The criterion for the unrelated pairs was the absence of association between prime and target, and in addition the absence of a mediated word. All subjects received the same stimuli, which were separated into two lists which were randomly ordered for each subject. Since stimulus material was drawn from a Hebrew association database, the number of words which were available for pairing was limited and we could not use the common prime and target recombination method.

Table 1 Mean frequency, concreteness, and length of the different stimulus lists which were used in the semantic priming experiments per list (upper table) and the semantic priming condition (lower table)

List 1 List 2

Log word frequency

Prime 0.93 0.90 SD 0.57 0.63 Target 1.08 1.04 SD 0.54 0.58 Word length

Prime 4.18 4.18 SD 0.86 0.80 Target 4.20 4.20 SD 0.86 1.00 Concreteness

Prime 5.95 5.94 SD 0.67 0.62 Target 5.94 5.70 SD 0.70 0.83

Direct Mediated Unrelated Pseudo

Log word frequency

Prime 0.81 0.96 0.93 0.93 SD 0.60 0.60 0.64 0.60 Target 1.10 1.10 0.92SD 0.50 0.65 0.51Word length

Prime 4.10 4.10 4.20 4.20 SD 0.82 0.70 0.90 0.85 Target 4.10 4.20 4.20SD 0.88 0.76 0.96 Concreteness

Prime 5.90 5.80 6.00 5.98 SD 0.75 0.78 0.55 0.62 Target 5.85 5.67 5.95SD 0.78 0.86 0.67

1 3

Exp Brain Res (2013) 226:121135

125

A half-visual eld pilot study was conducted (N = 21)

in order to validate the semantic priming stimulus list and to assess the hemispheric contribution to semantic and associative processing in healthy subjects. Priming magnitude was dened as the difference between the unrelated targets and the associated targets, separately for direct and for mediated targets. Facilitation would be of higher accuracy and faster RT in the associated targets compared to the unrelated targets, and a reversed pattern would be inhibition. The results showed the RHs specialization in remote, mediated association (the mean accuracy rate of 86.9 % in the RVF for mediated association in comparison with 92.8 % in the LVF; t(20) = 3.42, p < .003);

it thereby replicated, for the rst time in Hebrew, previous half-visual eld studies (Beeman et al. 1994; Chiarello and Richards 1992; Nakagawa 1991; Weisbrod et al. 1998) that found advantage for remote relations in the RH over the LH. A signicant interaction was found for the visual eld and priming type in the accuracy variable [F(1, 20) = 9.233, p < 0.007], averages of 8.3 % in

the direct relation and 7.5 % in the mediated relationin the LVF(RH)in comparison with 12.4 % in the direct and 3 % in the mediatedin the RVF(LH). A subsequent Bonferroni post hoc comparison revealed a signicant difference between direct and mediated priming in the RVF/LH [t(20) = 3.39, p < 0.003], but not in the LVF/RH

[t(20) = 1.00, p < 0.922]. Also, a signicant difference

was found between mediated and unrelated conditions only in the LVF (RH) [t(20) = 3.05, p < 0.006]. The direct

condition was signicantly different from the unrelated condition in both visual elds [t(20) = 2.3, p < 0.032, for

LVF(RH); and t(20) = 4.773, p < 0.001 for RVF(LH)].

In addition, the results demonstrated a signicant main effect for priming type [F(2, 20) = 11.57, p < .0001 for

accuracy rate; F(2, 20) = 7.89, p < .01 for RTs]. Here

the priming effect resulted from improved performance to mediated primes in the LVF and not poorer baseline, as performance for the unrelated condition was similar in RVF (84 %) and LVF (85 %). Also, signicant RVF advantage for direct relations was not found in this pilot experiment although accuracy of directly related targets in RVF (mean accuracy rate of 96.3 %) was higher than in the LVF (93.3 %).

tDCS

Bilateral stimulations were applied by passing a current between the left and right posterior temporo-parietal (Wernickes area, BA 22) scalp regions in both anodal and cathodal directions. The direction of current ow was opposite in the two hemispheres, leading to anodal ow in the left Wernickes area, whereas the right Wernickes

area received cathodal ow in one bilateral condition, and cathodal ow in the right Wernickes area, whereas the left Wernickes area received anodal ow in the complementary condition.

For the active stimulation condition, a direct current of 1.5 mA intensity was induced by two 5 7 cm saline-

soaked, synthetic sponge electrodes and delivered by a battery-driven, constant-current stimulator (Magstim LtD, Wales). The current had a ramp-up time of 30 s, was held at 1.5 mA for 20 min, and then ramped down over 30 s. For sham stimulation, the current was ramped up over 30 s and then ramped down over 30 s. CP5 and CP6 scalp positions were marked on the Lycra hat according to the EEG 1020 system in order to indicate the stimulation areas which were to be used during the semantic and lexical tasks. These positions were correlated with the left and right Wernickes areas, respectively (Floel et al. 2008). For the right (homologue of) Wernicke enhancement, the anode electrode was placed over the CP6 position and the cathode electrode over the CP5 position (Wernicke). For the second stimulation condition (left Wernicke enhancement), polarity was reversed. The tDCS application was single-blinded as subjects did not know whether they have active or sham stimulation.

Procedure

Stimuli were displayed in a darkened room on a 340 270 mm LCD monitor with a 60 Hz refresh rate at

a viewing distance of 57 cm. Both primes and targets were presented in the center of the screen. Stimulus presentation and data collection were automatically controlled by the computer program (E-prime, version 1).

For the lexical decision task, each trial initiated with a 1,000-ms presentation of a xation cross, followed by a 120-ms presentation of the target, followed by a 2,000-ms blank screen. The subjects responses consisted of pressing one of two buttons on the computers keyboard when presented with the blank screen (number 1 or 2 buttons, counterbalanced across subjects). For the semantic priming task, subjects were told that they would be presented with a word rst and subsequently with a string of characters. This second string of characters could be a word or a pseudo-word. The task was simply to read the rst word silently and then to decide as quickly and accurately as possible whether or not the second string of letters was a real Hebrew word. Each trial started with a xation cross which was presented for 500 ms, followed by a prime word which was presented for 750 ms and then a target word which was presented for 180 ms, followed by a 2,000-ms blank screen. The subjects responses consisted of pressing one of two buttons on the computers

1 3

126 Exp Brain Res (2013) 226:121135

keyboard. All subjects responded with their right hand. Half of them used the index nger for yes-responses (number 1 button) and the middle nger for no-responses (number 2 button), and the other half pressed 2 for yesand 1 for no-responses.

Exposure durations of the targets in the lexical decision (120 ms) were shorter than target presentation times in the semantic priming (180 ms) in order to equate the difculty level of the two tasks, based on pilot studies. Both durations were brief in order to replicate typical times in lateralization studies; however, here lateralization was dened by the stimulation rather than by visual elds. Prime type and target lexicality were randomly presented within stimulation conditions.

Statistical analysis

A mixed design ANOVA repeated measures statistical analysis was conducted. The dependent variables were reaction time for correct responses and accuracy (the percentage of correct responses over all trials). The within-subject variables were target lexicality for the lexical decision task and the relation type between the words and stimulation type for the semantic priming task, and the between-subject factor was a stimulation condition (group). In addition, for the semantic priming task, semantic measures were calculated in order to evaluate priming, whereas the no relation condition was used as reference to the other two related conditions. Only correct responses were analyzed in the RT analysis. Correct RTs were normally distributed, and therefore, no ltering was required.

Results

Lexical decision task

A signicant three-way interaction between lexicality (word/pseudo-word), stimulation type (sham/active), and stimulation group (right anodal/left cathodal, left anodal/right cathodal) was found for accuracy rates (F(1, 30) = 7.44, p < .011, p2 = 0.2) but not for RTs [F(1,

30) = 0.7, p < .407]. There were no additional effects or

interactions, apart from lexicality main effect for accuracy [F(1, 31) = 6.52, p < .042, p2 = 0.133], with higher

accuracy for pseudo-words (mean = 95 %) than for words

(mean = 92.5 %). See accuracy and mean RTs in Table 2.

There was no trade-off between RT and accuracy for the lexical decision task (Pearson correlation coefcients in the different conditions ranged between .11 and .07,

nonsignicant).

Consideration of performance according to stimulation group revealed a signicant difference for words between sham and active stimulation for right anodal/left cathodal stimulation [t(15) = 2.92, p < .010, Bonferroni correction;

an average decrease from 93.5 % accuracy rate in the sham condition to 89.8 % in the active condition], but not for the left anodal/right cathodal stimulation group, where the difference between active and sham was not signi-cant. However, it was in the opposite direction (an average improvement for words, from a 91.6 % accuracy rate in the sham condition to 93 % in the active condition; see Fig. 1). In contrast to words, pseudo-word processing was not signicantly affected by the stimulation group and condition (for right anodal/left cathodal stimulation: t(15) = 1.85,

Table 2 Mean accuracy rates and RT in msec (and standard deviation) for the lexical decision task, as a functionof stimulation and stimuli lexicality

Group Word Pseudo-word

Active Sham Active Sham

Right anodal/left cathodal

Mean accuracy 0.89 0.93 0.95 0.93 SD 0.04 0.03 0.02 0.02 Left anodal/right

cathodal

Mean accuracy 0.93 0.91 0.96 0.96 SD 0.06 0.06 0.04 0.03 Right anodal/left

cathodal

Mean RT (ms) 472.8 451.9 509.1 521.8 SD 88.5 70.2 77.7 77.7 Left anodal/right

cathodal

Mean RT (ms) 440.8 444.5 475.7 476.4 SD 84.9 77.9 74.4 90.9

1 3

Exp Brain Res (2013) 226:121135

127

Fig. 1 Stimulation effects on accuracy rates (and SEM) for words in the lexical decision task. The X-axis indicates the different stimulation groups: left anodal/right cathodal and right anodal/left cathodal. The Y-axis indicates the participants percentage of word identication accuracy. Right anodal/left cathodal stimulation signicantly decreased accuracy when compared to sham (**p < .01)

Table 3 Mean accuracy rates (and standard deviation) and RT in msec for the semantic priming task, as a function of stimulation and prime type

p < .084, an average increase from 93.5 % accuracy rate in the sham condition to 95.8 % in the active condition; for the left anodal/right cathodal stimulation: t(15) = .194,

p < .85). See accuracy and mean RTs in Table 2.

Semantic priming task

A signicant three-way interaction between relation type (direct/mediated/unrelated), stimulation (sham/active), and

stimulation group (right anodal/left cathodal, left anodal/ right cathodal) was found for accuracy rates [F(2, 30) = 5.47, p < .007, p2 = 0.154]. For reaction times, the

interaction was not signicant [F(2, 30) = 3.62, p < .067]

(see Table 3). There were no additional effects or interactions, apart from a prime type main effect for accuracy [F(1, 31) = 7.14, p < .035, p2 = 0.182]. There was no trade

off between RT and accuracy in the semantic priming task (Pearson correlation coefcients in the different conditions ranged between .16 and .11, nonsignicant).

Post hoc comparisons with Bonferroni corrections (p < .050) were carried out for the accuracy measure in order to reveal the source of the three-way interaction. Post hoc comparisons revealed a signicant difference between sham and active stimulation for unrelated pairs only [t(15) = 3.01, p < 0.009, see Fig. 2]. For mediated

priming and right anodal/left cathodal stimulation, a signicant priming effect was found under active stimulation [t(15) = 4.681, p < .001, 98 % for mediated vs. 88 % for

unrelated] but not under sham stimulation [t(15) = 0.728,

p < .478, 96.7 % for mediated vs. 95.5 % for unrelated]. Mediated priming effects under left anodal/right cathodal stimulation did not signicantly differ from sham.

Small yet signicant direct priming was evident for the two stimulation groups. In the right anodal group, a direct priming effect was found under both active stimulation [t(15) = 2.9, p < .004, 93 % for direct vs. 88.2 % for unre

lated] and sham stimulation [t(15) = 3.12, p < .008, 99 %

for direct vs. 95.5 % for unrelated]. In the left anodal/right cathodal group, a direct priming effect was found under both active [t(15) = 2.7, p < .006, 98 % for direct vs. 94.5 %

for unrelated] and sham stimulation [t(15) = 2.24, p < .020,

Group Direct Mediated Pseudo-word Unrelated

Active Sham Active Sham Active Sham Active Sham

Right anodal/ left cathodal

Acc 0.93 0.99 0.98 0.96 0.96 0.97 0.87 0.95 SD 0.02 0.01 0.03 0.05 0.02 0.02 0.07 0.04 Left anodal/

right cathodalAcc 0.98 0.95 0.97 0.96 0.97 0.97 0.94 0.93 SD 0.03 0.02 0.03 0.04 0.02 0.02 0.06 0.06 Right anodal/

left cathodalRT 568.2 560.9 548.6 551.4 577.1 574.9 574.6 557.1 SD 97.1 87.2 104.5 81.3 100.1 108.7 113.3 94.7 Left anodal/

right cathodalRT 552.2 563.5 554.1 550.8 585.4 574.4 561.8 557.9 SD 160.0 90.9 141.4 94.7 117.7 126.4 126.3 103.9

1 3

128 Exp Brain Res (2013) 226:121135

Fig. 2 Stimulation effectson mean accuracy rates (and SEM) in the semantic priming task. The X-axis indicates the different stimulation groups: left anodal/right cathodal and right anodal/left cathodal and the three prime types. The Y-axis indicates the participants percentage of correct decision. For unrelated pairs, a signicant decrease in accuracy was found under active right stimulation but not under sham (***p < .001)

95 % for direct vs. 93 % for unrelated]. Increases in direct priming during active stimulation did not differ according to stimulation side.

Discussion

The present study tested the effect of direct current stimulation over Wernickes area and its right homologue on lexical and semantic processing. Our hypothesis was that right anodal/left cathodal stimulation would interrupt the lexical processing dominance of the LH. We also expected to nd an opposite effect for the left anodal/right cathodal stimulation. As for the semantic priming, we hypothesized that the anodal stimulation of the RH will have a selective effect on the processing of mediated (but not direct) priming.

Lexical decision

In accordance with our hypothesis, the right anodal stimulation impaired lexical processing, whereas some nonsigni-cant improvement compared to sham was found in the left anodal/right cathodal stimulation. Hence, our results support previous reports on LH advantage in lexical decisions (Lieber 1976) and provide evidence for an involvement of posterior temporal areas (including Wernickes area and its right homologue and other adjacent posterior temporal areas) in lexical decision. It is also in accordance with neuroimaging ndings (Rossell et al. 2001) and previous NIBS

ndings which connect left temporo-parietal areas to language processing, including word detection (Andoh et al. 2008), picture naming (Mottaghy et al. 2006), word generation (Knecht et al. 2002), and associative learning (Floel et al. 2008).

Our ndings reect more interference to lexical processes following right anodal/left cathodal rather than improvement following left anodal/right cathodal stimulation, though there are some nonsignicant advantages in processing words under left anodal/right cathodal when compared to sham stimulation (at least the pattern is reversed for the right anodal/left cathodal stimulation). This may be due to a ceiling effect, as the accuracy rates were already very high in the sham condition, so improvement magnitude is limited. Another plausible explanation for this may be the simultaneous cathodal stimulation over the right Wernickes area. Such stimulation can produce inhibition and therefore impair the inter-hemispheric interaction. In line with this possibility, our ndings are consistent with a previous tDCS study which reported that cathodal stimulation in the right Wernickes area that was administered during language treatment improved auditory verbal comprehension in subacute stroke patients (You et al. 2011).

Semantic priming

We hypothesized a larger mediated priming effect under right anodal/left cathodal stimulation in comparison with sham stimulation and with left anodal/right cathodal stimulation.

1 3

Exp Brain Res (2013) 226:121135

129

In line with the hypothesis, a signicantly larger mediated priming effect was found under right anodal/left cathodal stimulation.

Closer examination of the priming task nding revealed that the signicant right anodal/left cathodal stimulation effects were the strongest in the unrelated condition, hence the lexical baseline that the direct and mediated conditions refer to. Similarly to the stimulations effect on lexical decisions, the right anodal/left cathodal stimulation signicantly diminished the ability to respond correctly when the prime and target words were unrelated. Since priming is calculated by subtracting the related from the unrelated condition, the inevitable result of lower unrelated scores is larger priming effects. The opposite is of course true for higher unrelated scores. However, the magnitude of the direct priming was not affected by this change in unrelated accuracy as the right anodal/left cathodal stimulation also impaired the accuracy of directly related targets. A plausible explanation for this may be the simultaneous cathodal stimulation over Wernickes area. Such stimulation can produce inhibition and therefore impair the proper lexical processing of the target words, hence the poorer performance when it came to directly related and unrelated targets which both rely on left-lateralized mechanisms (Koivisto and Laine 2000).

The increased mediated priming effect under right anodal/left cathodal stimulation may indicate the role of the right language areas in processing indirect semantic associations. Furthermore, such an increase was not found for direct priming. Hence, the signicant increase may represent the summation of two opposite inuences: an increased mediated processing effect due to RH activation/LH inhibition and decreased word recognition due to the same activation/inhibition pattern. There seems to be support for this interpretation in the standard lexical decision data, in that right anodal combined with left cathodal montage decreased accuracy. Applying unilateral stimulation may resolve this possibility.

Another future direction is increasing the difculty level of the priming task. A degradation of the target words, for instance, can be used. These methods of presentation are thought to increase priming effects (Massaro et al. 1978) and therefore may overcome the stimulations lexical level inuence and uncover the hypothesized inuence on the semantic level.

Additionally, strategic processes were likely to inuence the semantic priming effects, considering the long SOA (750 ms) that was used (Neely 1991). Controlled, strategic processes of semantic priming have been related to prefrontal regions, such as the anterior cingulate cortex

(ACC) and left inferior prefrontal cortex (LIPC). Lesions to the LIPC resulted in intact automatic semantic priming but impaired controlled priming, which suggests LIPC is involved in strategic processes of semantic priming (Hagoort 1997). Also, the LIPC was reported to be more active when subjects perform semantic judgments for words in comparison with nonsemantic judgments for the same words (Gabrieli et al. 1996), and rTMS over LIPC impaired high-order semantic processing (Whitney et al. 2011). Rossell et al. (2001) suggested that the rostral ACC monitors automatic lexical access to semantic relations, whereas dorso-caudal ACC is presumably related to the controlled processes (such as expectancy generation and post-lexical semantic matching). Thus, using short SOA and allowing only automatic processes may result in a more signicant inuence. Alternatively, stimulation of prefrontal areas can also be applied. However, this will not allow for the exploration of the automatic spread of activation processes. Future experimental designs can test long and short SOA with stimulation applied over Wernickes area (and its homologue) and left and right IPC. This will allow testing for both the controlled and spreading activation semantic processing.

Conclusions

We applied tDCS over Wernickes area and its right homo-logue, bilaterally. Right anodal/left cathodal stimulation signicantly and selectively impaired the lexical processing of words and semantic judgment of unrelated word pairs as well. Although the stimulation method does not allow us to draw a specic conclusion regarding the underlying brain mechanism, our results support previous reports on a LH advantage in lexical decisions and provide evidence that overactive temporo-parietal RH areas interfere with the swift lexical processing that is typically conducted by the LH. In addition, the bilateral stimulation neural effect can be conceptualized as a shift in the normal hemispheric asymmetry. Thus, it may serve as a model for brain language abnormalities such as schizophrenia (Mitchell and Crow 2005).

Acknowledgments This study was supported by the Israeli Academy of Sciences grant no. 100/10, the Israeli Center of Research Excellence (I-CORE) in Cognition (I-CORE Program 51/11), and an ERC starting grant which was awarded to ML (Inspire 200512).

Appendix

See Table 4.

1 3

130 Exp Brain Res (2013) 226:121135

Table 4 Word stimuli for the semantic priming task

Prime Target Translated prime

Translated target

Relation type

Target frequency

Prime frequency

Prime length

Target length

List Log frequency target

Log frequency prime

Yellow Sun Direct 46 3 30 4 1 1.66 1.48 Couch Living room Direct 7 4 3 3 1 0.85 0.48 Trees Wood Direct 17 3 19 4 1 1.23 1.28 Yolk Egg Direct 15 4 2 5 1 1.18 0.3 Apron Kitchen Direct 18 4 1 5 1 1.26 0 Clouds Rain Direct 29 3 4 5 1 1.46 0.6 Cradle Baby Direct 37 5 1 5 1 1.57 0 Paper Newspaper Direct 68 5 56 4 1 1.83 1.75 Inhaler Asthma Direct 2 5 1 4 1 0.3 0 Autumn Fall Direct 2 4 8 4 1 0.3 0.9 Wool Ship Direct 9 4 8 3 1 0.95 0.9 Hunger Food Direct 184 4 17 3 1 2.26 1.23 Lampshade Lamp Direct 6 5 1 4 1 0.78 0 Humus Pita Direct 4 4 14 5 1 0.6 1.15 Needle Pin Direct 3 4 3 3 1 0.48 0.48 Cow Milk Direct 63 3 3 3 1 1.8 0.48 Smoke Cigarette Direct 7 6 19 3 1 0.85 1.28 Laundry Clean Direct 36 3 13 5 1 1.56 1.11 Africa Asia Direct 30 4 68 6 1 1.48 1.83 Helmet Motorbike Direct 6 5 4 4 1 0.78 0.6 Climate Pineapple Mediated 5 4 8 5 1 0.7 0.9 Hut Grove Mediated 1 5 1 4 1 0 0 Heavy Feather Mediated 3 4 10 5 1 0.48 1 Angle Hearing Mediated 6 5 13 5 1 0.78 1.11 Barrel Grapes Mediated 9 5 33 4 1 0.95 1.52 Ball Square Mediated 4 5 108 4 1 0.6 2.03 Priest Bell Mediated 3 5 4 4 1 0.48 0.6 Pillow Peacock Mediated 1 4 4 4 1 0 0.6 Copper Strong Mediated 82 3 5 5 1 1.91 0.7 Airport Jet Mediated 52 4 3 3 1 1.72 0.48 Rock Corner Mediated 35 4 32 3 1 1.54 1.51 Snow Black Mediated 76 4 25 3 1 1.88 1.4 Clown Tent Mediated 9 4 4 4 1 0.95 0.6 West Middle Mediated 37 5 23 4 1 1.57 1.36 Bee Apple Mediated 17 4 14 5 1 1.23 1.15 Apartment Dishes Mediated 76 4 57 4 1 1.88 1.76 Spinage Grass Mediated 8 3 4 3 1 0.9 0.6 Surfboard Fish Mediated 41 4 4 4 1 1.61 0.6 Pearls Neck Mediated 4 5 11 5 1 0.6 1.04 Coffee Night Mediated 145 4 140 3 1 2.16 2.15 Knight Shell Unrelated 2 3 4 4 1 0.3 0.6 Dough Stem Unrelated 2 5 18 3 1 0.3 1.26 Store Tribe Unrelated 10 3 7 4 1 1 0.85 Medal Vinger Unrelated 9 4 7 4 1 0.95 0.85 Skirt Menu Unrelated 32 5 4 5 1 1.51 0.6 Bun Barricade Unrelated 15 5 1 6 1 1.18 0 Scarf Cream Unrelated 4 4 3 4 1 0.6 0.48

1 3

Exp Brain Res (2013) 226:121135

Table 4 continued

Prime Target Translated prime

131

Log frequency prime

Telephone Brain Unrelated 33 3 102 5 1 1.52 2.01 Wich Train Unrelated 30 5 3 5 1 1.48 0.48 Step Bird Unrelated 9 5 13 5 1 0.95 1.11 Sink Noise Unrelated 24 3 2 4 1 1.38 0.3 Stone Coat Unrelated 9 4 70 3 1 0.95 1.85 Street Blue Unrelated 8 6 54 4 1 0.9 1.73 Hat Shark Unrelated 3 4 13 4 1 0.48 1.11 Sheet Narcissus Unrelated 6 5 2 4 1 0.78 0.3 Note Toy Unrelated 11 5 39 3 1 1.04 1.59 Boot Cigar Unrelated 5 4 1 3 1 0.7 0 Palm tree Tanks Unrelated 8 5 15 3 1 0.9 1.18 Spoon Journey Unrelated 106 3 39 4 1 2.03 1.59 Sled Cotton Unrelated 6 5 1 5 1 0.78 0 Thorn Pseudo-word 4 3 3 1 0.48 Hatch Pseudo-word 3 2 4 1 0.3 Thumb Pseudo-word 5 1 4 1 0 Building Pseudo-word 4 48 5 1 1.68 Young goat Pseudo-word 4 26 3 1 1.41 Case Pseudo-word 4 4 4 1 0.6 Shoulder Pseudo-word 3 8 3 1 0.9 Closet Pseudo-word 3 15 4 1 1.18 Palace Pseudo-word 4 10 5 1 1 Hump Pseudo-word 5 1 4 1 0 Police Pseudo-word 3 40 5 1 1.6 Door Pseudo-word 3 26 3 1 1.41 Sugar Pseudo-word 5 54 4 1 1.73 Helicopter Pseudo-word 5 9 4 1 0.95 Bed Pseudo-word 4 13 4 1 1.11 Gum Pseudo-word 4 5 5 1 0.7 Orbit Pseudo-word 5 38 5 1 1.58 Juce Pseudo-word 5 32 3 1 1.51 Metal Pseudo-word 4 15 4 1 1.18 Water Pseudo-word 4 165 3 1 2.22 Steak Pseudo-word 6 5 5 1 0.7 Waterfall Pseudo-word 3 2 3 1 0.3 Sprinkler Pseudo-word 4 1 5 1 0 Spice Pseudo-word 5 3 5 1 0.48 Basement Pseudo-word 4 5 4 1 0.7 Truck Pseudo-word 5 11 5 1 1.04 Beverage Pseudo-word 4 14 4 1 1.15 Doorpost Pseudo-word 6 1 5 1 0 Kiss Pseudo-word 5 8 5 1 0.9 Crow Pseudo-word 5 3 4 1 0.48 Dollars Pseudo-word 3 81 6 1 1.91 Bus Pseudo-word 3 27 6 1 1.43 Point Pseudo-word 4 35 5 1 1.54 Smell Pseudo-word 4 26 3 1 1.41

Translated target

Relation type

Target frequency

Prime frequency

Prime length

Target length

List Log frequency target

1 3

132 Exp Brain Res (2013) 226:121135

Table 4 continued

Prime Target Translated prime

Translated target

Relation type

Target frequency

Prime frequency

Prime length

Target length

List Log frequency target

Log frequency prime

Dance Pseudo-word 5 20 5 1 1.3 Property Pseudo-word 6 17 4 1 1.23 Ride Pseudo-word 5 9 5 1 0.95 Harvest Pseudo-word 5 7 4 1 0.85 Elephant Pseudo-word 4 14 3 1 1.15 Skiing Pseudo-word 5 23 3 1 1.36 Cube Pseudo-word 4 2 5 1 0.3 Scream Pseudo-word 4 1 5 1 0 Cherry Pseudo-word 4 7 6 1 0.85 Salad Pseudo-word 4 24 3 1 1.38 Match Pseudo-word 5 1 5 1 0 Cloak Pseudo-word 5 2 5 1 0.3 Ice cream Pseudo-word 3 13 5 1 1.11 Walk Pseudo-word 4 23 5 1 1.36 Guitar Pseudo-word 3 12 5 1 1.08 Hill Pseudo-word 5 7 4 1 0.85 Orchestra Pseudo-word 3 13 6 1 1.4 Bucket Pseudo-word 3 17 3 1 1.23 Pool Pseudo-word 4 12 5 1 1.08 Mud Pseudo-word 3 9 3 1 0.95 Sliding Pseudo-word 4 29 5 1 1.46 Rhinoceros Pseudo-word 3 1 4 1 0 Bubble Pseudo-word 6 4 4 1 0.6 Jeep Pseudo-word 3 10 3 1 1 Iron Pseudo-word 6 32 4 1 1.51 Etrog Pseudo-word 3 1 5 1 0 Tiles Red Direct 66 4 3 5 2 1.82 0.48 Flour Wheat Direct 4 4 24 3 2 0.6 1.38 Carpet Magic Direct 8 5 1 4 2 0.9 0 Sour Sweet Direct 30 4 7 4 2 1.48 0.85 Pencil case Pencil Direct 3 6 1 4 2 0.48 0 Soldier Uniform Direct 9 4 51 4 2 0.95 1.71 Ring Finger Direct 16 4 7 4 2 1.2 0.85 Planter Plant Direct 24 3 2 4 2 1.38 0.3 Table Chair Direct 5 3 53 5 2 0.7 1.72 Suite Tie Direct 3 5 5 4 2 0.48 0.7 Pot Cocking Direct 24 5 9 3 2 1.38 0.95 Bald spot Hair Direct 43 4 3 4 2 1.63 0.48 Surprise Party Direct 31 5 24 5 2 1.49 1.38 City Village Direct 75 3 66 3 2 1.88 1.82 Cheetah Tiger Direct 9 3 1 5 2 0.95 0 Clown Purim Direct 8 5 4 4 2 0.9 0.6 Piano Grand Direct 4 3 14 5 2 0.6 1.15 Horse Ride Direct 9 5 33 3 2 0.95 1.52 Butcher shop Meat Direct 64 3 1 5 2 1.81 0 Orchard Orange Mediated 10 4 8 4 2 1 0.9 Springboard Swimming Mediated 275 5 1 5 2 2.44 0

1 3

Exp Brain Res (2013) 226:121135

Table 4 continued

Prime Target Translated prime

133

Log frequency prime

Frog King Mediated 41 3 7 5 2 1.61 0.85 Ball Diet Mediated 22 5 4 4 2 1.34 0.6 Perfume Taste Mediated 11 4 122 4 2 1.04 2.09 Carpenter Green Mediated 46 3 7 4 2 1.66 0.85 Fortune Wallet Mediated 2 4 99 3 2 0.3 2 Shoemaker Socks Mediated 5 6 4 5 2 0.7 0.6 Rake Digging Mediated 2 5 1 5 2 0.3 0 Coal White Mediated 140 3 9 3 2 2.15 0.95 Morning Dark Mediated 11 4 122 4 2 1.04 2.09 Button Rose Mediated 8 5 15 5 2 0.9 1.18 Rug Ceiling Mediated 4 4 6 4 2 0.6 0.78 Cat Barking Mediated 1 5 15 4 2 0 1.18 Scholarship Bank Mediated 144 3 3 4 2 2.16 0.48 Shutter Glass Mediated 26 5 1 4 2 1.41 0 Bag Belly Mediated 24 3 2 5 2 1.38 0.3 Bolt Handle Mediated 3 4 5 4 2 0.48 0.7 Oven Candles Mediated 17 4 8 4 2 1.23 0.9 Jaw Violin Unrelated 7 5 2 3 2 0.85 0.3 Hose Desert Unrelated 5 5 12 5 2 0.7 1.08 Oil Mug Unrelated 4 3 19 3 2 0.6 1.28 Compass Food Unrelated 94 4 1 5 2 1.97 0 Tractor Dog Unrelated 27 3 4 6 2 1.43 0.6 Labyrinth Pier Unrelated 1 3 5 4 2 0 0.7 Sukkah Earring Unrelated 1 4 3 4 2 0 0.48 Test Soap Unrelated 13 4 38 4 2 1.11 1.58 Pan Church Unrelated 3 6 4 4 2 0.48 0.6 Bench Plate Unrelated 12 4 25 4 2 1.08 1.4 Car Whale Unrelated 1 6 69 6 2 0 1.84 Diary Jellysh Unrelated 2 5 136 4 2 0.3 2.13 Medal Road Unrelated 51 4 7 5 2 1.71 0.85 Pliers Corn Unrelated 11 4 1 3 2 1.04 0 Turn Cinnamon Unrelated 9 6 25 5 2 0.95 1.4 Crying Tavor Unrelated 4 4 11 3 2 0.6 1.04 Eight Anvil Unrelated 6 3 118 5 2 0.78 2.07 Pumpkin Mail Unrelated 62 4 9 4 2 1.79 0.95 Kite Raspberry Unrelated 7 3 1 6 2 0.85 0 Molting Pseudo-word 5 1 5 2 0 Camera Pseudo-word 3 19 5 2 1.28 Shopping bag Pseudo-word 6 12 4 2 1.08 Tobacco Pseudo-word 3 9 3 2 0.95 Fig Pseudo-word 3 3 4 2 0.48 Body Pseudo-word 4 81 3 2 1.91 Coin Pseudo-word 3 28 4 2 1.45 Dove Pseudo-word 5 22 4 2 1.34 Wolf Pseudo-word 4 67 3 2 1.83 Lever Pseudo-word 5 6 4 2 0.78 Napkin Pseudo-word 4 1 4 2 0

Translated target

Relation type

Target frequency

Prime frequency

Prime length

Target length

List Log frequency target

1 3

134 Exp Brain Res (2013) 226:121135

Table 4 continued

Prime Target Translated prime

Translated target

Relation type

Target frequency

Prime frequency

Prime length

Target length

List Log frequency target

Log frequency prime

Monastery Pseudo-word 5 4 4 2 0.6 Flute Pseudo-word 4 10 4 2 1 Cinema Pseudo-word 3 152 6 2 2.18 Box Pseudo-word 4 1 5 2 0 Envelope Pseudo-word 3 5 5 2 0.7 Boat Pseudo-word 3 3 4 2 0.48 Paintbrush Pseudo-word 6 2 5 2 0.3 Pasture Pseudo-word 5 1 3 2 0 Nock Pseudo-word 3 1 5 2 0 Breathing Pseudo-word 5 26 5 2 1.41 Ant Pseudo-word 3 2 4 2 0.3 Song Pseudo-word 3 90 3 2 1.95 Lish Pseudo-word 3 3 5 2 0.48 Cloud Pseudo-word 3 6 3 2 0.78 Snail Pseudo-word 5 3 5 2 0.48 Root Pseudo-word 4 12 4 2 1.08 Clock Pseudo-word 4 28 4 2 1.45 Spike Pseudo-word 4 4 5 2 0.6 Fox Pseudo-word 6 5 4 2 0.7 Plum Pseudo-word 4 2 4 2 0.3 Robot Pseudo-word 5 8 5 2 0.9 Ground Pseudo-word 5 50 4 2 1.7 Jump Pseudo-word 4 15 5 2 1.18 Radio Pseudo-word 5 132 4 2 2.12 Scooter Pseudo-word 5 2 5 2 0.3 Foot Pseudo-word 5 49 3 2 1.69 Olive harvest Pseudo-word 6 2 4 2 0.3 Faucet Pseudo-word 3 8 3 2 0.9 Pedal Pseudo-word 4 1 5 2 0 Hail Pseudo-word 5 2 3 2 0.3 Skiff Pseudo-word 6 2 5 2 0.3 Bottle Pseudo-word 6 19 5 2 1.28 Cereal Pseudo-word 4 15 3 2 1.18 Drought Pseudo-word 5 38 5 2 1.58 Dragging Pseudo-word 3 1 5 2 0 Through Pseudo-word 6 5 4 2 0.7 Onion Pseudo-word 4 33 3 2 1.52 Camel Pseudo-word 6 10 3 2 1 Screw Pseudo-word 3 53 4 2 1.72 Land Pseudo-word 4 81 3 2 1.91 Duck Pseudo-word 4 5 5 2 0.7 Hare Pseudo-word 6 1 5 2 0 Tail Pseudo-word 6 5 3 2 0.7 Form Pseudo-word 4 35 4 2 1.54 Hike Pseudo-word 4 50 4 2 1.7 Olive Pseudo-word 5 40 3 2 1.6

1 3

Exp Brain Res (2013) 226:121135

135

References

Andoh J, Artiges E, Pallier C, Rivire D, Mangin JF, PaillreMartinot ML et al (2008) Priming frequencies of transcranial magnetic stimulation over Wernickes area modulate word detection. Cereb Cortex 18:210216

Balota DA, Lorch RF Jr (1986) Depth of automatic spreading activation. J Exp Psychol Learn Mem Cogn 12:336345

Beeman M, Chiarello C (1998) Right hemisphere language comprehension: perspectives from cognitive neuroscience. Erlbaum, Mahwah

Beeman M, Friedman RB, Grafman J, Perez E, Diamond S, Lindsay

MB (1994) Summation priming and coarse semantic coding in the right hemisphere. J Cogn Neurosci 6(1):2645Binder JR, McKiernan KA, Parsons ME, Westbury CF, Possing

ET, Kaufman JN, Buchanan L (2003) Neural correlates of lexical access during visual word recognition. J Cogn Neurosci 15:372393Bookheimer S (2002) Functional MRI of language: new approaches to understanding the cortical organization of semantic processing. Annu Rev Neurosci 25:151188Chiarello C, Hoyer WJ (1988) Adult age differences in implicit and explicit memory: time course and coding effects. Psychol Aging 3:358366Chiarello C, Richards L (1992) Another look at categorical priming in the cerebral hemispheres. Neuropsychologia 30:381392Floel A, Roesser N, Michka O, Knecht S, Breitenstein C (2008) Noninvasive brain stimulation improves language learning. J Cogn Neurosci 20:14151422Floel A, Meinzer M, Kirstein R, Nijhof S, Deppe M, Knecht S, Breitenstein C (2011) Short-term anomia training and electrical brain stimulation. Stroke 42(7):20652067Gabrieli JDE, Desmond JE, Demb JB, Wagner AD, Stone MV, Vaidya

CJ, Glover GH (1996) Functional magnetic resonance imaging of semantic memory processes in the frontal lobes. Psychol Sci 7(5):278283Hagoort P (1997) Semantic priming in Brocas aphasics at a short

SOA: no support for an automatic access decit. Brain Lang 56(2):287300Iacoboni M, Zaidel E (1996) Hemispheric independence in word recognition: evidence from unilateral and bilateral presentations. Brain Lang 53:121140Jacobson L, Koslowsky M, Lavidor M (2012) tDCS polarity effects in motor and cognitive domains: a meta-analytical review. Exp Brain Res 216:110Jung-Beeman M (2005) Bilateral brain processes for comprehending natural language. Trends Cogn Sci 9:512518Kiefer M, Weisbrod M, Kern I, Maier S, Spitzer M (1998) Right hemisphere activation during indirect semantic priming. Brain Lang 64:377408Knecht S, Floel A, Drager B, Breitenstein C, Sommer J, Henning-sen H, Ringelstein EB (2002) Degree of language lateralization determines susceptibility to unilateral brain lesions. Nat Neurosci 5(7):695699Koivisto M, Laine M (2000) Hemispheric asymmetries in activation and integration of categorical information. Laterality 5:121 Kuperberg GR, Deckersbach T, Holt DJ, Goff D, West WC (2007)

Increased temporal and prefrontal activity in response to semantic associations in schizophrenia. Arch Gen Psychiatry 64:138151 Lecardeur L, Giffard B, Laisney M, Brazo P, Delamillieure P, Eustache F, Dollfus S (2007) Semantic hyperpriming in schizophrenic patients: increased facilitation or impaired inhibition in semantic association processing? Schiz Res 89(13):243250

Lieber L (1976) Lexical decisions in the right and left cerebral hemispheres. Brain Lang 3(3):443450

Lindell AK (2006) In your right mind: right hemisphere contributions to language processing and production. Neuropsychol Rev 16(3):131148

Massaro DW, Jones RD, Lipscomb C, Scholz R (1978) Role of prior knowledge on naming and lexical decisions with good and poor stimulus information. J Exp Psychol Learn 4(5):498512

McNamara TP (2005) Semantic priming. Psychology Press, New York Mitchell RL, Crow TJ (2005) Right hemisphere language functions and schizophrenia: the forgotten hemisphere? Brain 128:963978 Mottaghy FM, Sparing R, Topper R (2006) Enhancing picture naming with transcranial magnetic stimulation. Behav Neurol 17:177186 Nakagawa A (1991) Role of anterior and posterior attention networks in hemispheric asymmetries during lexical decisions. J Cogn Neurosci 3:313321Neely JH (1991) Semantic priming effects in visual word recognition:

A selective review of current ndings and theories. In: Besner D, Humphreys GW (eds) Basic processes in reading: visual word recognition. Erlbaum, Hillsdale, pp 264336Nitsche MA, Paulus W (2000) Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol 527(3):633639Pobric G, Mashal N, Faust M, Lavidor M (2008) The role of the right cerebral hemisphere in processing novel metaphoric expressions: a TMS study. J Cogn Neurosci 20(1):170181Rossell SL, Bullmore ET, Williams SC, David AS (2001) Brain activation during automatic and controlled processing of semantic relations: a priming experiment using lexical-decision. Neuropsychologia 39(11):11671176Rubinsten O, Anaki D, Henik A, Drori S, Faran Y (2005) Free association norms in the Hebrew language. In: Henik A, Rubinsten O, Anaki D (eds) Word norms in Hebrew. Ben-Gurion University of the Negev, Beer-Sheva, pp 1734Sass K, Krach S, Sachs O, Kircher T (2009) Liontigerstripes: neural correlates of indirect semantic priming across processing modalities. NeuroImage 45(1):224236Thiel A, Haupt WF, Habedank B, Winhuisen L, Herholz K, Kessler J,

Markowitsch HJ (2005) Neuroimaging-guided rTMS of the left inferior frontal gyrus interferes with repetition priming. NeuroImage 25(3):815823Tivarus ME, Ibinson JW, Hillier A, Schmalbrock P, Beversdorf DQ

(2006) An fMRI study of semantic priming: modulation of brain activity by varying semantic distances. Cogn Behav Neurol 19(4):194200Weisbrod M, Maier S, Harig S, Himmelsbach U, Spitzer M (1998)

Lateralised semantic and indirect priming effects in people with schizophrenia. Br J Psychiatry 172:142146Whitney C, Kirk M, OSullivan J, Lambon Ralph MA, Jefferies E

(2011) The neural organization of semantic control: TMS evidence for a distributed network in left inferior frontal and posterior middle temporal gyrus. Cereb Cortex 21(5):10661075You DS, Kim DY, Chun MH, Jung SE, Park SJ (2011) Cathodal transcranial direct current stimulation of the right Wernickes area improves comprehension in subacute stroke patients. Brain Cogn 119:15Zouridakis G, Simos P, Breir J, Papanicolaou A (1998) Functional hemispheric asymmetry assessment in a visual language task using MEG. Brain Topogr 11(1):5765

1 3