Something to talk about: Enhancement of linguistic cohesion throughtdCS in chronic non fluent aphasia

Paola Marangolo a,b,n, Valentina Fiori b, Serena Campana b, Maria Antonietta Calpagnano b,Carmelina Razzano b, Carlo Caltagirone b,c, Andrea Marini b,d,n

a Facoltà di Medicina, Università Politecnica Marche, Ancona, Italyb IRCCS Fondazione Santa Lucia, Roma, Italyc Università di Tor Vergata, Roma, Italyd Dipartimento di Scienze Umane, Università di Udine, Udine, Italy

a r t i c l e i n f o

Article history:Received 22 December 2012Received in revised form20 November 2013Accepted 2 December 2013Available online 11 December 2013

Keywords:Left frontal gyrustDCSBroca's areaBrain stimulationAphasia rehabilitationCohesionPragmatics

a b s t r a c t

Several studies have shown that the modulation of cortical activity through transcranial direct currentstimulation (tDCS) enhances naming performance in persons with aphasia. In this study, we investigatedthe potential effects of tDCS in improving spontaneous speech and the ability to use connective words toestablish cohesion among adjacent utterances in a group of eight participants with chronic non fluentaphasia. They were administered five short videoclips representing everyday life contexts and twopicture description tasks. Three videoclips were used to elicit spontaneous conversation during thetreatment, while the remaining tasks were presented to the patients only before and after the therapy.Patients were required to talk about each videoclip, with the help of a therapist, while they were treatedwith tDCS (20 min, 1 mA) over the left hemisphere in three different conditions: anodic tDCS over theBroca's area, anodic tDCS over the Wernicke's area and a sham condition. Each experimental conditionwas performed for ten consecutive daily sessions with 14 days of intersession interval. Only after Broca'sstimulation, patients showed a greater improvement in producing words that enhanced the cohesion oftheir speech samples (i.e., pronouns, ellipses, word repetitions, conjunctions). Beneficial effects of thestimulation were generalized also to contexts presented to the patients at the beginning and at the end ofthe therapy sessions. Our data further confirm the key role of the left inferior frontal gyrus in bindingwords into a coherent speech. We believe that positive tDCS effects may be further extended to differentlinguistic domains, useful to promote language recovery.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Over the last few years, aphasiology has witnessed a great dealof change. Indeed, the advances in linguistic theory and theevidence coming from neuropsychology and cognitive neuroscienceare dramatically boosting our knowledge about the linguistic,cognitive and neural underpinnings of the human ability to gen-erate a discourse or take part to a conversation. As a consequence,clinicians need to devise novel ways to assess the linguistic outputof their patients (Andreetta, Cantagallo, & Marini, 2012; Armstrong,2000; Boles, 1998; Chapman & Ulatowska, 1989, 1992; Marini,Andreetta, Del Tin, & Carlomagno, 2011) and innovative rehabilita-tion protocols aimed at recovering not only their linguistic produc-tion but also their communicative skills (e.g. Marangolo, 2010;

Marini, Caltagirone, Pasqualetti, & Carlomagno, 2007). Formerly,the research on discourse production in persons with aphasiafocused on the quantity of information they could convey withtheir speech samples (Ulatowska, North, & Macaluso-Hayes, 1981;Ulatowska, Freedman-Stern, Doyel, Macaluso-Haynes, & North,1983). It soon became clear, however, that, when compared tohealthy speakers, non fluent aphasic individuals tend to producefewer (and shorter) complex sentences with a general reduction ofinformation (Ulatowska et al., 1981, 1983). But production is notonly a matter of informativeness. Indeed, the production of con-nected discourse or a contribution to a conversation rest also on theability to link the utterances by means of cohesive connectives.These are linguistic devices that link distinct utterances so toprovide the continuity of the meanings conveyed by a discourseor conversation (Halliday & Hasan, 1976). One of the most fre-quently used cohesive devices is “reference”. As to this point, animportant distinction has been introduced between the concepts ofexophoric and endophoric reference (Halliday & Hasan, 1976).Exophoric reference consists in the ability to refer to someone or

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/neuropsychologia

Neuropsychologia

0028-3932/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.neuropsychologia.2013.12.003

n Corresponding authors at: Facoltà di Medicina, Università Politecnica Marche,Ancona, Italy. Tel.: þ39 712 206093; fax: þ39 2206 214.

E-mail addresses: p.marangolo@univpm.it (P. Marangolo),andrea.marini@uniud.it (A. Marini).

Neuropsychologia 53 (2014) 246–256

something (e.g., objects, ideas, persons) that is not directly detect-able from the words used in the utterances (e.g., “Put it here” or “Idon't believe it”). As such, exophoric reference is not a cohesivedevice, as it does not bind elements together into a text. Rather, itsmain function is to refer to a state or item that has not beenintroduced verbally in the preceding utterances but must beinferred by the extralinguistic context. On the contrary, endophoricreference points to concepts that have been previously mentionedin the flow of discourse. As such, this type of reference is needed toestablish cohesion among the utterances that form a connecteddiscourse. Endophoric reference can be established in differentways: through the use of lexical repetitions (e.g., “I saw a boy inthe garden. The boy was climbing a tree”), anaphoras (pronounsreferring to someone that has been previously mentioned, e.g., “he[the boy] was about to fall”), cataphoras (pronouns linking forwardto a referent in the following utterances, e.g., “I had told him, butMarco did not listen to me”), lexical substitutions (e.g., “I wasworried about the child [the boy]), ellipses (i.e., omissions as in “Iran 5 miles on the first day and 8 [miles] on the second),conjunctions (e.g., “and”, “or”) and words sharing some semanticrelation with previously uttered lexical items (e.g., “The little man[the boy] was climbing a tree”) (Halliday & Hasan, 1976). Obviously,a good communicator must be competent in both exophoric andendophoric reference. However, the latter is particularly interestingfrom a clinical point of view. Indeed, a verbal exchange in which theinterlocutors discuss about the same topic and build on each others’contributions would be considered more cohesive than one inwhich each partner describes his/her physical environment withno reference to its own or the partner's prior utterances (Rochester& Martin, 1977). A number of studies have suggested that cohesiveties might be used to distinguish among different syndromes.Indeed, individuals with schizophrenia with thought disorder useties differently than non-thought-disordered ones (Rochester &Martin, 1979). Similarly, persons with Alzheimer's disease usecohesive ties differently than healthy individuals (Ripich, Terell &Spinelli, 1983; Ripich, Vertes, Whitehouse, & Fulton, 1988). Theresearch on cohesion in individuals with aphasia is at best scanty,controversial and limited to individuals with fluent forms ofaphasia. For example, if Bloom, Borod, Santschi-Haywood, Pickand Obler (1996) reported normal cohesive levels in the languagesamples produced by a group of persons with fluent aphasia, in aprevious study by Glosser and Deser (1990) individuals with similardeficits used often pronouns with antecedents (see also Ultaowskaet al., 1981, 1983). Interestingly, they interpreted the reduced abilityto provide adequate cohesive ties among utterances as a conse-quence of a deficit in lexical retrieval rather than a real problem inintersentential organisation (see also Bates, Hamby & Zurif, 1983).Overall, it might be hypothesized that aphasic patients engaged inconversational settings tend to use egocentric speech and do nottailor their utterances to avoid ambiguity for their addressee(Marangolo, 2010). For instance, they might opt for words withexophoric reference, such as first and second person pronouns (e.g.“I have done this too”) and general questions (“but what are youdoing?”), which more likely refer to their own ideas or theirphysical environment than to topics shared with their interlocutors.

Therefore, both clinicians and researchers in the field of aphasiarehabilitation are gradually becoming aware of the need to includein their protocols also tests for the assessment of discourse levelabilities such as those involved in the establishment of cohesionamong adjacent utterances. This goes along with the need toprovide new ways to treat also this aspect of verbal communication.Indeed, several investigators have stressed the importance of theinclusion of pragmatic treatments in severe chronic aphasic patientsto elicit verbal communication (Basso, 2010; Lai, 1993; Marangolo,2010; Marini & Carlomagno, 2004; Wilkinson & Wielaert, 2012).For example, the Conversational Therapy approach prompts a

natural conversation between the therapist and the aphasic patient,a condition of communicative exchange, in which both speakersparticipate using their available communicative resources (Basso,2010; Grice, 1975; Marangolo, 2010). However, to date, no studieshave explicitly investigated if a pragmatic treatment of this kindmight enhance the patient's ability to use cohesive devices.

Parallel to this growing interest in the way spontaneouslanguage is processed in daily communicative interactions, inmore recent years, the development of new technologies hasprovided both professional therapists and educators with innova-tive instruments. In the field of aphasia, a small but growing bodyof evidence has already shown that noninvasive brain stimulation,such as transcranial magnetic stimulation (TMS) or transcranialdirect current stimulation (tDCS), coupled with language trainingcan exert beneficial effects in the treatment of naming deficits(Naeser et al., 2005; Baker, Rorden, & Fridriksson, 2010; Fiori et al.,2011, Fridriksson, Richardson, Baker, & Rorden, 2011; Marangoloet al., 2013). In the context of language processing, recent inves-tigations have suggested a potential role of the left inferior frontalgyrus and of the adjacent cortex in the selection and unification ofoperations by which individual pieces of lexical information arebound together into a meaningful discourse (Hagoort, 2005;Marini & Urgesi, 2012). Therefore, we might assume that couplinga pragmatic treatment with repeated stimulation over the leftfrontal gyrus may exert a positive influence also in the recovery ofcohesive units.

This study was designed to investigate whether cohesionanalysis can be used to evaluate changes in the language of eightindividuals with chronic nonfluent aphasia after an intensiverehabilitation treatment based on a Conversational Therapyapproach. Namely, the research addressed two major issues: 1.assess the efficacy of an intensive pragmatic treatment in enhan-cing discourse cohesion; 2. investigate whether the combined useof tDCS applied over the left inferior frontal gyrus (i.e., Broca'sarea) and the language treatment would enhance the recovery oflinguistic cohesive connectives in a group of persons with chronicnon fluent aphasia.

2. Materials and methods

2.1. Participants

2.1.1. Control groupTwenty healthy individuals (10 males and 10 females) matched for age (40 to

75 years) and education level (13 to 17 years) with the participants with aphasiawere enrolled in the experiment. They were all native Italian speakers with nohistory of neurological or psychiatric illness.

2.1.2. Aphasic groupEight participants (5 males and 3 females) who had suffered a single left

hemisphere stroke were included in the study. Inclusion criteria for the study werenative Italian proficiency, pre-morbid right handedness, a single left hemisphericstroke at least 6 months prior to the investigation, and no acute or chronicneurological symptoms requiring medication. The data analyzed in the currentstudy were collected in accordance with the Helsinky Declaration and theInstitutional Review Board of the IRCCS Fondazione Santa Lucia, Rome, Italy. Priorto participation, all patients signed informed consent forms.

2.2. Neuropsychological assessment

Their linguistic skills were assessed using standardized language tests (theBattery for the analysis of aphasic disorders, BADA test (Miceli, Laudanna, Burani, &Capasso, 1994) and the Token test (De Renzi & Vignolo, 1962). They were alsoadministered a Neuropsychological Battery (Orsini et al., 1987; Zimmermann &Fimm, 1994), which excluded the presence of attention and memory deficits thatmight have confounded the data (see Table 1).

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256 247

2.3. Clinical data

In all patients, the MRI scans revealed an ischemic lesion involving the lefthemisphere. The lesion mapping analysis indicated that the areas of maximal lesionoverlap were localised in the capsula estrema, the claustrum, the capsula esternaand the putamen (see Fig. 1). The eight patients were diagnosed with non fluentaphasia because of their reduced spontaneous speech with short sentences andfrequent word-finding difficulties. They had no articulatory deficits that might haveconfounded the data (see Table 1).

2.4. Materials

We prepared five short videoclips (lasting 15 min each) that reproducedcommon everyday life situations. Three videoclips were used to elicit spontaneousconversation in the patients during the treatment (at the seaside, at school, aroundthe city) (T-videoclips). The remaining two (shopping at the supermarket and the

housekeepers) together with two cartoons story with six pictures each (the “Flowerpot”, Huber & Gleber, 1982; and the “Quarrel”, Nicholas & Brookshire, 1993) werepresented to the patients only before and after the therapy to control for general-ization effects (G-videoclips).

2.5. Procedure

Prior to the experiment, the five videoclips and the cartoons were shown to thecontrol group. Each subject was asked to freely describe each context accurately,with no interference from the examiner. Each language sample was tape-recordedand transcribed verbatim.

2.6. Transcranial direct current stimulation (tDCS)

tDCS was applied using a battery driven Eldith (neuroConn GmbH) Program-mable Direct Current Stimulator with a pair of surface-soaked sponge electrodes

Table 1Sociodemographic and Clinical data of the eight non fluent aphasic subjects. Results in the neuropsychological battery for attention (Zimmermann & Fimm, 1994), memorydeficits (Orsini et al., 1987) and Token Test (De Renzi & Vignolo, 1962) are also reported.

Subjects Sex Age Ed. Level Timepost-onset

Type of aphasia Attentional abilities(scores in percentile45 unimpaired)

Memory Token Test(correctresponses)

WM (cut/off 572)STM (cut/off 772)LTM (cut/off 5.5)

B.C. F 63 8 3 Year and 5 months Non fluent Alertness (tot): 80 WM: 5 16/36Sustained Att (tot): 81 STM: 6Selective Att (tot): 57 LTM: 11

F.S. F 71 5 1 Year and 8 months Non fluent Alertness (tot): 30 WM: 5 22/36Sustained Att (tot): 25 STM: 6Selective Att (tot): 35 LTM: 13

P.C. M 65 9 1 Year Non fluent Alertness (tot): 59 WM: 4 9/367 Months Sustained Att (tot): 75 STM: 6

Selective Att (tot): 55 LTM: 12P.F. M 44 13 7 Years Non fluent Alertness (tot): 99 WM: 5 17/36

Sustained Att (tot): 66 STM: 7Selective Att (tot): 84 LTM: 13

A.C M 64 13 4 Years Non fluent Alertness (tot): 20 WM: 4 19/365 Months Sustained Att (tot): 24 STM: 6

Selective Att (tot): 18 LTM: 6N.M. F 65 13 3 Years Non fluent Alertness (tot): 97 WM: 6 18/36

7 Months Sustained Att (tot): 68 STM: 6Selective Att (tot): 86 LTM: 11

P.M. M 52 13 1 Year Non fluent Alertness (tot): 87 WM: 5 12/362 Months Sustained Att (tot): 52 STM: 5

Selective Att (tot): 63 LTM: 11R.L. M 61 11 4 Years Non fluent Alertness (tot): 96 WM: 5 10/36

7 Months Sustained Att (tot): 57 STM: 6Selective Att (tot): 52 LTM: 12

Legend: Ed. Level¼educational level, WM¼working memory, STM¼short term memory, LTM¼ long term memory.

Fig. 1. Lesion descriptions for each aphasic patient. The figure shows the MRI acquisitions of all patients. Top line: Sagittal and axial views of colour coded probability mapof lesion overlap (range 1% purple to 91% white). Individual volume lesions were drawn manually on the re-oriented brain volume transformed into MNI standardizedstereotaxic coordinate system using a computational semi-automatic procedure of REGISTER software provided by Brain Imaging Center, Montreal Neurological Institute,McGill University. Averaging the labelled voxels of the individual lesion volumes re-aligned in MNI space generated the probability map revealing the localisation of areas ofpercentage of lesion overlap. Maximal overlap includes the capsula estrema, the claustrum, the capsula esterna and the putamen. The inferior frontal gyrus (including Broca'sarea) and the superior temporal gyrus (including Wernicke's area) were similary damaged both having about 37% of lesion overlap; percentage increases in thesupramarginal gyrus where there was about 62% of lesion overlap. B.C.’s lesion is localized in the left fronto-temporo cortices and part of the parietal cortex, including themiddle frontal gyrus, the inferior frontal gyrus (Broca's area), part of the lateral frontal pole and of the pre-central gyrus, the temporal pole, the full extension of the superiortemporal gyrus (Wernicke's area), most part of the post-central gyrus, the angular gyrus and part of the supramarginal gyrus. The lesion also includes the insula. F.S.’s lesionis localized mainly in the left temporal-insular region. A further damaged area is at the level of the homolateral frontal lobe (mesial portion) involving the white matterrunning under the middle frontal gyrus. P.C.’s lesion is localized in the left fronto-temporal parietal cortices. The cortical lesion includes mainly the frontal pole, andposteriorly part of the superior temporal gyrus and part of the inferior and superior parietal lobe. The lesion also involves the insula. P.F.’s lesion is localized mainly in the leftfronto-temporal cortices, with a minor involvement of the homolateral parietal cortex. The cortical lesion includes mainly the temporal pole, and posteriorly part of thesuperior temporal gyrus (Wernicke's area) and part of the middle temporal gyrus. The lesion also includes the insula. A.C.’s lesion is localized in the left fronto-temporalparietal cortices. The cortical lesion includes the middle frontal gyrus, the inferior frontal gyrus (Broca's area), the anterior portion of the superior temporal gyrus, and theangular gyrus. The lesion also involves the insula. N.M.’s lesion is localized in the left fronto-temporal parietal cortices. The cortical lesion includes mainly the frontal poleand the inferior frontal gyrus (Broca's area), the temporal pole, the full extension of the superior temporal gyrus (Wernicke's area), and part of the middle temporal gyrus, theangular gyrus and part of the supramarginal gyrus, and the superior parietal gyrus. The lesion also involves the insula. P.M.'s lesion is localized in the left fronto-temporocortices, involving the inferior frontal gyrus (Broca's area), and the temporal pole. The lesion also includes the insula. R.L’s lesion is localized in the left temporo-parieto-occipital cortices, including mainly the temporal pole, the full extension of the superior temporal gyrus (Wernicke’s area), part of the middle temporal lobe, the angular andthe supramarginal gyri, the inferior parietal lobule and the superior occipital gyrus. (For interpretation of the references to color in this figure legend, the reader is referred tothe web version of this article.)

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256248

(5 cm�7 cm). A constant current of 1 mA intensity was applied on the skin for20 min. If applied according to safety guidelines, tDCS is considered to be a safebrain stimulation technique with minor adverse effects (Poreisz, Boros, Antal, &Paulus, 2007). Two different electrode stimulation positions were used: the F5 of

the extended International 10–20 system for EEG electrode placement, whichcorresponds best to Broca's area (Naeser et al., 2010; Nishitani, Schürmann,Amunts, & Hari, 2005), and the CP5 of the extended International 10–20 systemfor EEG electrode placement, which has been found to correspond best to

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256 249

Wernicke's area (Fiori et al., 2011; Oliveri et al., 1999). In both conditions thereference electrode was placed over the contralateral frontopolar cortex (Nitsche &Paulus, 2000; Sparing, Dafotakis, Meister, Thirugnanasambandam, & Fink, 2008).

Each T-videoclip was assigned to a different stimulation condition. Overall,three different stimulation conditions were carried out: (1) anodic (F5-A) stimula-tion over the Broca's area; (2) anodic (CP5-A) stimulation over the Wernicke's area;and (3) a sham condition (F5/CP5 S). Sham stimulation was performed exactly likeanodic stimulation over the Broca's or Wernicke's area, but the stimulator wasturned off after 30 s. It has been shown that this procedure makes it possible toblind subjects as to the respective stimulation condition (Gandiga, Hummel, &Cohen, 2006). Each condition was performed in ten consecutive daily sessions(Monday–Friday, weekend off, Monday–Friday) over 3 months, with 14 days ofintersession interval, while the subjects underwent the “conversational therapy”treatment (see Fig. 2). The order of presentation of the T-videoclips and ofconditions was randomized across subjects.

2.7. Language treatment

To measure baseline performance, for each experimental condition, at thebeginning and at the end of the treatment sessions, all patients were asked todescribe the correspondent videoclip without the therapist's help.

The main objective of the conversational therapy was to set up a naturalconversation with the patient in which both interlocutors participated using theiravailable communicative resources. Both patient and therapist were left free to use anycommunicative means (e.g., gestures, drawings, orthographic or phonological cues) toexchange salient information about the videoclip. The therapist was instructed toaccept all the information provided by the patient and tried to relate it to the topic ofconversation in order to improve its content and informativeness. The goal of thetherapy was to make the patient more informative day-by-day with the context and totrain him/her to talk about the video without the therapist's support.

In order to control for the potential generalization of treatment effects, at thebeginning and at the end of each experimental condition, the participants were re-administered the language tests and asked to describe the three G-videoclips andthe cartoon-picture stories without the therapist's support. Each language samplewas tape-recorded and transcribed verbatim. The transcriptions did not includefillers such as “um, er, ah” and the examiner's prompt.

3. Data analysis

Data were analyzed using SPSS 13.0 software. First, we examinedwhether an increase in speech productivity (as measured in terms ofproduction of nouns, verbs, adjectives, adverbs and sentences) could

be found for the T-videoclips after each treatment condition. The t-tests comparison showed that only the stimulation of Broca's areainduced a significant increase in nouns, verbs and sentences (seeTable 2).

Second, we focused on the effect of stimulation on the ability toselect and use adequate cohesive endophoric devices. For thisreason, two different analyses were run. The former compared theresults achieved by the aphasic group with the data collected inthe healthy group for the different contexts (T and G-videoclipsand the two cartoons). For this analysis, before and after eachtreatment session, the number of correct endophoric referencesproduced by each aphasic for the T and G-videoclips and for thetwo cartoons (see Table 3) was divided by the mean numbercollected in the healthy control group for the same contexts (seeTable 4). The final result was converted into a mean percentage ofcorrect responses and then analyzed through a 2�3 repeated-measures ANOVA (ANOVArm) run separately for the T-videoclipsand for each G-videoclip and cartoon. Since the healthy group didnot produce cataphoras and very few lexical substitutions werepresent, these endophoric references were not analyzed.

To control if an overall increase in the rate at which endophoricreferences were used was independent from a more generalincrease in the amount of speech production, a final analysis wasrun. First, for each aphasic, we calculated the number of utterancesproduced in each context and for each condition (Broca vs.Wernicke vs. Sham). The number of endophoric references pro-duced by each aphasic participant was divided by the number ofutterances collected (see Table 5) and then analyzed through a2�3 repeated-measures ANOVA (ANOVArm) run separately for theT-videoclips and for each G-videoclip and cartoon.

3.1. Treatment

The comparison between the results achieved by the aphasicgroup with the data collected in the healthy group for the differentcontexts showed a significant effect of Time (baseline (T1) vs. endof treatment (T10), F(1,7)¼15.91; p¼ .007) and Condition (anodic

Fig. 2. Localization of the tDCS area (A) and overview of study design (B): one videoclip was used for the anodic Broca’s stimulation, one for the anodic Wernicke’sstimulation and a third one for the sham condition. Each condition was performed in ten consecutive daily sessions over 3 months, with 14 days of intersession interval,while the subjects underwent the “conversational therapy” treatment.

Table 2Mean number of nouns, verbs, adjectives, adverbs and sentences (7SEM, standard error of the mean) produced before and after each treatment condition by the aphasicgroup in the T-videoclips (AdjþAdv¼adjectives and adverbs; paired t-test, n¼o .05; nn¼o .01, nnn¼o .001; the order of n refers to the following comparisons: pre-postBroca/post Broca-post Wernicke/post Broca-post Sham).

Pre-Broca Post-Broca Pre-Wernicke Post-Wernicke Pre-Sham Post-Sham

Nouns 28 (76) 51 (78)nnn/n/nn 28 (76) 38 (77) 29 (76) 34 (76)Verbs 20 (75) 38 (76)nn/nn/nn 21 (74) 26 (75) 20 (74) 25 (74)AdjþAdv 12 (74) 26 (77)nn/nn/n 13 (74) 16 (75) 13 (74) 17 (74)Sentences 3 (71) 14 (74)nn/n/nn 2 (71) 6 (73) 2 (71) 5 (71)

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256250

Broca's area vs. anodic Wernicke's area vs. Sham, F(2,14)¼6.60;p¼ .012). Subjects’ performance significantly improved at the endof treatment with respect to baseline [mean¼58%, SEM¼3 (T10)vs mean¼18%, SEM¼10 (T1) p¼ .007]. Moreover, the mean num-ber of endophoric references in the anodic Broca's condition wassignificantly greater than in the other two conditions (mean¼51%,SEM¼13 (anodic Broca's) vs. mean¼35%, SEM¼9 (anodic Wer-nicke's) vs. mean¼27%, SEM¼8 (Sham) p¼ .012). The interactionof time� condition (F(2,14)¼4.59; p¼ .033) was also significant. TheScheffè post-hoc test revealed that the mean number of endopho-ric references significantly improved at the end of treatment ineach condition with respect to baseline (differences between T10vs. T1 for Broca's condition: 64%, p¼ .000, Wernicke's condition:30%, p¼ .011, and Sham: 25%, p¼ .027). However, while no sig-nificant differences emerged in the mean number of endophoric

references between the three conditions at baseline (differencesbetween Broca vs. Wernicke¼�1%, p¼ .646; differences betweenBroca vs. Sham¼5%, p¼ .878; differences between Wernicke vs.Sham¼6%, p¼ .542), at the end of treatment, the mean number ofendophoric references was significantly greater in the anodicBroca's condition with respect to the other two conditions, whichdid not differ from each other (differences between Broca vs.Wernicke¼33%; p¼ .006; differences between Broca vs. Sham¼44%;p¼ .001; differences between Sham vs. Wernicke¼�11%; p¼ .294)(see Fig. 3).

Moreover, the amount of improvement found for the Broca'scondition persisted also when we compared for each condition thenumber of endophoric references produced by each aphasic toutterances. Indeed, while no significant differences emergedbetween the three conditions at baseline (differences betweenBroca vs. Wernicke¼0, p¼1; differences between Broca vs.Sham¼0.2, p¼ .104; differences between Wernicke vs. Sham¼0.2,p¼ .074), the mean number of endophoric references to utterancessignificantly improved at the end of treatment with respect tobaseline only in the Broca's condition (differences between T10 vs.T1 Broca's condition: 0.3, p¼ .001, Wernicke's condition: 0.1,p¼ .448, and Sham: 0.1, p¼ .448) (see Fig. 4).

3.2. Generalization of the treatment

The comparison between the results achieved by the aphasicgroup with the data collected in the healthy group for the differentcontexts revealed a generalization of the effects of treatment butonly in the left Broca's condition for the two videos and the twocartoons presented to the patients only before and after thetreatment. In this condition, for all context, at the end of treat-ment, the mean percentage of endophoric references was

Table 3Number of endophoric references produced by each aphasic patient in the T and G-videoclips and in the two cartoons before and after each treatment session.

Participants Pre-Broca Post-Broca Pre-Wernicke Post-Wernicke Pre-Sham Post-Sham

T-videoclipsB.C. 5 32 4 15 3 10F.S. 5 21 5 13 3 9P.C. 6 25 4 16 4 6P.F. 8 30 10 20 6 10A.C. 3 13 6 13 0 7N.M. 7 25 9 19 6 17P.M. 6 14 7 12 6 5R.L. 10 34 7 18 6 15Mean 6 24 6 16 4 10

G-videoclipsShopping at the supermarket/the housekeepersB.C. 6/4 10/11 7/5 8/6 6/6 5/5F.S. 2/6 7/12 3/5 1/7 5/7 6/7P.C. 0/1 6/6 0/2 2/3 0/3 2/2P.F. 2/1 7/5 4/1 5/2 4/1 3/0A.C. 0/1 5/4 0/1 1/0 2/1 1/2N.M. 4/2 9/7 6/3 6/4 7/4 5/5P.M. 0/0 5/5 3/1 1/1 0/1 1/1R.L. 3/3 10/9 5/4 3/5 4/5 4/5Mean 2/2 7/7 4/3 3/4 4/4 3/3

CartoonsFlower pot/quarrelB.C. 6/6 11/13 5/6 6/6 6/5 6/7F.S. 5/8 10/15 4/8 6/9 4/7 5/7P.C. 2/0 6/3 2/0 3/1 1/0 1/1P.F. 6/6 11/10 5/5 6/4 6/7 8/8A.C. 1/0 7/4 1/0 3/1 1/0 0/1N.M. 5/4 10/8 3/5 5/6 5/6 8/8P.M. 2/6 7/10 3/6 3/6 2/5 2/6R.L. 4/7 8/11 5/6 7/8 3/7 4/9Mean 4/5 9/9 4/5 5/5 4/5 4/6

Table 4Mean number (7SD) of endophoric references produced by the control group foreach context (SD¼standard deviation).

Anaphoras Conjunctions Ellipses Wordrepetition

CartoonsFlower pot 5 (73) 4 (72) 5 (71) 3 (72)Quarrel 5 (72) 6 (74) 3 (73) 3 (71)G-VideoclipsShopping at thesupermarket

2 (71) 4 (73) 4 (73) 1 (71)

The housekeepers 3 (72) 3 (72) 10 (74) 0T-VideoclipsAround the city 6 (71) 11 (77) 20 (76) 0At the seaside 3 (73) 11 (75) 5 (72) 1 (71)At school 7 (73) 10 (75) 6 (71) 0

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256 251

significantly greater with respect to the other two conditions(differences between Broca vs. Wernicke¼30%; p¼ .04; differencesbetween Broca vs. Sham¼29%; p¼ .04; differences between Shamvs. Wernicke¼1%; p¼ .92 for the “Supermarket” video; differencesbetween Broca vs. Wernicke¼20%; p¼ .002; differences betweenBroca vs. Sham¼21%; p¼ .001; differences between Sham vs.Wernicke¼1%; p¼ .74 for the “Housekeepers” video; differencesbetween Broca vs. Wernicke¼23%; p¼ .003; differences betweenBroca vs. Sham¼25%; p¼ .002; differences between Sham vs.Wernicke ¼�2%; p¼ .79 for the “Flower Pot” cartoon and differ-ences between Broca vs. Wernicke¼23%; p¼ .001; differencesbetween Broca vs. Sham¼15%; p¼ .015; differences between Shamvs. Wernicke¼8%; p¼ . 14 for “the Quarrel” cartoon) (see Fig. 5).

Moreover, as for the T-videos, the mean number of endophoricreferences to utterances significantly improved for each context atthe end of treatment with respect to baseline only in the Broca'scondition (differences between T10 vs. T1 Broca's condition: 0.4,p¼ .007, Wernicke's condition: 0.1, p¼ .927, and Sham: 0.1, p¼ .819for the “Supermarket” video; differences between T10 vs. T1Broca's condition: 0.6, po .001, Wernicke's condition: 0.1,p¼ .803, and Sham: 0, p¼1 for the “Housekeepers” video;

Table 5Number of endophoric references to utterance produced by each aphasic patient in the T and G-videoclips and in the two cartoons before and after each treatment session.

Participants Pre-Broca Post-Broca Pre-Wernicke Post-Wernicke Pre-Sham Post-Sham

T-videoclipsB.C. 0.3 0.7 0.3 0.4 0.1 0.3F.S. 0.3 07 0,2 0,3 0,2 0.2P.C. 1 1.7 1 1.1 1 1.2P.F. 0.2 0.7 0.3 0.4 0.2 0.2A.C. 1 1.4 1.2 1.6 0 0.5N.M. 0.4 0.5 0.6 0.7 0.3 0.5P.M. 0.7 1 0.5 0.5 0.7 0.4R.L. 0.5 0.7 0.4 0.5 0.4 0.6Mean 0.5 0.9 0.6 0.7 0.4 0.5

G-videoclipsShopping at the supermarket/the housekeepersB.C. 0.8/0.5 0.9/1.2 0.8/0.7 1/0.8 0.8/0.8 0.6/0.6F.S. 0.4/0.4 0.7/0.9 0.4/0.4 0.1/0.7 0.6/0.5 0.7/0.5P.C. 0/0.5 1/0.9 0/0.7 1/1 0/0.6 1/0.7P.F. 1/0.1 1/0.5 0.8/0.1 0.8/0.2 0.6/0.1 0.6/0A.C. 0/0.3 0.8/0.8 0/0.2 0.3/0 0.5/0.3 0.5/0.5N.M. 0.8/0.2 1.1/0.8 1.2/0.4 1.2/0.4 0.8/0.5 0.6/0.6P.M. 0/0 0.5/0.6 0.4/0.2 0.1/0.2 0/0.1 0.2/0.1R.L. 1/0.4 1.1/1.5 0.6/0.4 0.4/0.6 0.5/0.6 0.5/0.6Mean 0.5/0.3 0.9/0.9 0.5/0.4 0.6/0.5 0.5/0.4 0.6/0.4

CartoonsFlower pot/quarrelB.C. 0.6/0.8 0.8/0.9 0.7/0.6 0.9/0.6 0.7/0.5 0.7/0.6F.S. 0.8/0.6 0.8/0.9 0.7/0.6 0.8/0.8 0.6/0.6 0.6/0.5P.C. 0.5/0.0 0.8/0.6 0.4/0 0.5/0.2 0.3/0 0.3/0.1P.F. 0.6/0.9 0.9/0.8 0.8/0.6 0.8/0.6 0.8/0.7 0.9/0.7A.C. 0.3/0 1/0.5 0.3/0 0.6/0.2 0.3/0 0/0.3N.M. 0.6/1 0.9/1.6 0.3/0.7 0.6/0.7 0.8/0.9 0.8/0.8P.M. 0.5/0.7 1/1 0.5/0.8 0.4/0.8 0.5/0.7 0.5/0.9R.L. 0.4/0.7 0.5/0.8 0.4/0.7 0.5/1 0.3/1 0.6/0.8Mean 0.5/0.5 0.8/0.9 0.5/0.5 0.6/0.6 0.5/0.6 0.5/0.6

Fig. 3. Mean percentage of correct endophoric references produced by the aphasicgroup at baseline (T1) and at the end of treatment (T10) for the left Wernicke’s,Broca’s and sham conditions (n¼o .01), respectively. Error bars represent standarderror of the mean.

Fig. 4. Mean number of endophoric references to utterance produced for theT-videos by the aphasic group before and after the treatment for the Broca’s,Wernicke’s and sham conditions, respectively.

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256252

differences between T10 vs. T1 Broca's condition: 0.3, p¼ .032,Wernicke's condition: 0.1, p¼ .706 and Sham: 0, p¼1 for the“Flower Pot” cartoon and differences between T10 vs. T1 Broca'scondition: 0.4, p¼ .012, Wernicke's condition: 0.1, p¼ .684, andSham: 0, p¼1 for “the Quarrel” cartoon) (see Fig. 6).

4. Discussion

The aim of this study was to determine whether in personswith chronic non fluent aphasia an intensive treatment based on aconversational therapy approach coupled with tDCS applied overthe left inferior frontal gyrus improves the levels of discoursecohesion by enhancing the use of linguistic connectives. The mainfinding is that after the treatment the patients produced descrip-tions with an increased number of words and, most importantly, amore accurate use of cohesive devices.

On the first assessment before the therapy protocol they couldproduce only a limited output characterized by few words andmany ungrammatical sentences teaming with self-referentialexpressions. At the end of the rehabilitative program, theymanaged to produce a higher percentage of anaphoras, ellipses,conjunctions and word repetitions (see Fig. 7).

Interestingly, in accordance with the results from our controlgroup, the aphasic group did not produce cataphoras and very fewlexical substitutions were present.

From a developmental perspective, the ability to adapt thelinguistic output to a given context and topic and to share acommon perspective with an interlocutor is considered as a highproficiency level in language acquisition. Children initially useegocentric expressions referred to their own environment andthen utterances that refer to the content chosen with theirinterlocutor (Haslett, 1983; Hickmann & Hendriks, 1999). Indeed,for communication to succeed, speakers must be cooperative(Grice, 1975), not egocentric, and able to produce utterancesrelated to each other through cohesive devices. The view thatlanguage production is a cooperative process has been furtherelaborated by Clark et al. (Clark, 1996; Clark & Marshall, 1981;Clark & Murphy, 1982) who argued that speakers formulateutterances by consulting information that is mutually shared withtheir partners on a common ground. According to Clark andMarshall (1981), speakers keep track of the shared informationwith a particular addressee by constructing a detailed model of theaddressee's knowledge and beliefs and constantly update it. Inorder to achieve this, they need to correctly select cohesiveelements that refer to previously mentioned topics and to bindthe different elements in a coherent way (Appelt, 1985; Appelt &Kronfeld, 1987). In our study, after the treatment, the aphasicpatients were able to use these referring expressions and toexchange this information on the basis of the mutual knowledgeof the argument with their interlocutor.

A second major result of our study was that after the stimula-tion over the left inferior frontal gyrus, the patients’ ability toproduce cohesive speech showed the greatest improvement. Sincethis improvement was accompanied by an increase in speechproductivity we wondered if the subjects’ actual ability to usecohesive devices was independent from an overall increase inspeech production. Indeed, a significant increase in the rate atwhich the endophoric references were used to utterance wasfound only for the Broca's condition confirming that the stimula-tion increased the “reference density” of subjects’ speech.

As mentioned in the Introduction, the improvement in theproduction of cohesive ties is in line with recent neuroimagingstudies suggesting that Broca's area and the adjacent left inferiorfrontal cortex plays a major role in binding single-word informa-tion into a unified interpretation of multiword utterances(Hagoort, 2003, 2005). In order to produce coherent speechsamples, the different elements need to be unified through theuse of linguistic devices, which refer back to things or ideasmentioned earlier. Indeed, it has been recently demonstrated thatthe left inferior frontal gyrus is a key region in pronoun processingand in maintaining co-reference between a pronoun and itsantecedent within utterances (Hammer, Jansma, Tempelmann, &Munte, 2011).

A further confirmation of the crucial role of Broca's area inenhancing discourse cohesion comes from our data on the gen-eralization effects. Only after the Broca's condition, beneficialbenefits, as measured in terms of endophoric references, general-ized also to contexts that had not been used during therapy (seeFig. 7).

It has been suggested that long-lasting functional changes inthe cortex as the result of electrical stimulation are the conse-quence of modulation of the strength of synaptic connections (i.e.synaptic plasticity, Nitsche & Paulus, 2000). In our study, thedecision to stimulate the left-damaged hemisphere regions wasrelated to previous results showing that the stimulation of perile-sional spared language areas close to the stimulation site inchronic aphasic patients may enhance functional improvement(Baker et al., 2010; Fiori et al., 2011; Fridriksson et al., 2011;

Fig. 5. Mean percentage of correct endophoric references produced by the aphasicgroup for the G-videos and cartoons before and after the treatment for the Broca’s,Wernicke’s and sham conditions, respectively.

Fig. 6. Mean number of endophoric references to utterance produced for theG-videos and cartoons by the aphasic group before and after the treatment for theBroca’s, Wernicke’s and sham conditions, respectively.

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256 253

Marangolo et al. 2011; Marangolo et al., 2013). These results are inline with the hypothesis that, in chronic patients, languagerecovery may be associated with the reactivation of left-hemispheric perilesional structures (Saur et al., 2006, 2008;Warburton, Price, Swinburn, & Wise, 1999; Winhuisen et al.2007). Although current modeling studies have suggested thatthe flow of current in conventional tDCS can be quite diffuse anddifficult to predict (Datta, Baker, Bikson, & Fridriksson, 2011),others studies have indicated that during tDCS the current isdistributed around the targeted region (e.g., Holland et al., 2011).

Similar results have been found in different fMRI studies tomeasure tDCS effects during the stimulation of the motor cortex(Antal, Polania, Schmidt-Samoa, Dechent, & Paulus, 2011; Antal,Kovacs, Chaieb, Paulus, & Greenlee, 2012). Therefore, we mightspeculate that, in our patients, tDCS has enhanced the capacity ofthe left hemispheric spared areas close to the stimulated region tomake compensatory plastic changes resulting in improved perfor-mance. Indeed, the lesion mapping analysis indicated that theareas of maximal lesion overlap were not localised into thetargeted regions suggesting that some perilesional spared tissue

Fig. 7. Mean percentage of correct anaphoras, conjunctions, ellipses and word repetition produced for the T–G videos and cartoons by the aphasic group before and after thetreatment for the Broca’s, Wernicke’s and sham conditions, respectively.

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256254

might have contributed to the recovery process (see Fig. 1).Moreover, as the same percentage of lesion overlap was presentin the two stimulated areas (37%), we can exclude the possibilitythat the more efficient recovery observed after the left Broca'sstimulation was due to greater sparing of this region to cerebraldamage compared to left Wernicke's area (see Fig. 1).

We are aware that the present approach, due to the smallsample and to the lack of functional magnetic resonance imagingdata, does not allow to draw firm conclusions about the under-lying neural mechanisms by which tDCS affected subjects’ perfor-mance. However, this study allows to draw some importantconclusions about language rehabilitation in persons with chronicnon fluent aphasia. Indeed, it confirms several previous reportsthat highlight the importance to use a more ecological approach tolanguage treatment. Moreover, this study clearly confirms that thelinguistic outcome of non fluent aphasic patients can be success-fully improved by coupling specific treatment approaches withtDCS.

References

Andreetta, S., Cantagallo, A., & Marini, A. (2012). Narrative discourse in anomicaphasia. Neuropsychologia, 50, 1787–1793.

Antal, A., Kovacs, G., Chaieb, L., Paulus, W., & Greenlee, M. W. (2012). Cathodalstimulation of human MTþ leads to elevated fMRI signal: A tDCS-fMRI study.Restoration Neurology of Neuroscience, 30, 255–263.

Antal, A., Polania, R., Schmidt-Samoa, C., Dechent, P., & Paulus, W. (2011).Transcranial direct current stimulation over the primary motor cortex duringfMRI. NeuroImage, 15, 590–596.

Appelt, D. E. (1985). Planning English referring expressions. Artificial Intelligence, 26,1–33.

Appelt, D. E., & Kronfeld, A. (1987). A computational model of referring. In:Proceedings of the 10th international joint conference on artificial intelligence,(pp. 640–647).

Armstrong, E. (2000). Aphasic discourse analysis: The story so far. Aphasiology, 14,875–892.

Baker, J. M., Rorden, C., & Fridriksson, J. (2010). Using transcranial direct-currentstimulation to treat stroke patients with aphasia. Stroke, 41, 1229–1236.

Basso, A. (2010). “Natural” conversation: A treatment for severe aphasia. Aphasiol-ogy, 24, 466–479.

Bates, E., Hamby, F., & Zurif, E. (1983). The effects of focal brain damage onpragmatic expression. Canadian Journal of Psychology, 37, 59–84.

Bloom, R. L., Borod, J. C., Santschi-Haywood, C., Pick, L. H., & Obler, L. K. (1996). Leftand right hemispheric contributions to discourse coherence and cohesion.International Journal of Neuroscience, 88, 125–140.

Boles, L. (1998). Conversational discourse analysis as a method for evaluatingprogress in aphasia: A case report. Journal of Communication Disorders, 31,261–273.

Chapman, S. B., & Ulatowska, H. K. (1989). Discourse in aphasia: Integration deficitsin processing reference. Brain and Language, 36, 651–668.

Chapman, S. B., & Ulatowska, H. K. (1992). Methodology for discourse managementin the treatment of aphasia. Clinics in Communication Disorders, 2, 64–81.

Clark, H. H., & Marshall, C. R. (1981). Definite reference and mutual knowledge. In:A. K. Joshi, B. Webber, & I. Sag (Eds.), Elements of discourse understanding (pp.10–63). Cambridge, UK: Cambridge University Press.

Clark, H. H., & Murphy, G. L. (1982). Audience design in meaning and reference. In: J.F.L. Ny, & W. Kintsch (Eds.), Language and comprehension (pp. 287–299).

Clark, H. H. (1996). Using language. Cambridge, UK: Cambridge University Press.Datta, A., Baker, J. M., Bikson, M., & Fridriksson, J. (2011). Individualized model

predicts brain current flow during transcranial direct-current stimulationtreatment in responsive stroke patient. Brain Stimulation, 4, 169–174.

De Renzi, E., & Vignolo, L. A. (1962). The Token Test: A sensitive test to detectreceptive disturbances in aphasia. Brain, 85, 665–678.

Fiori, V., Coccia, M., Marinelli, C. V., Vecchi, V., Bonifazi, S., Ceravolo, M. G., et al.(2011). Transcranial direct current stimulation improves word retrieval inhealthy and nonfluent aphasic subjects. Journal of Cognitive Neuroscience, 23,2309–2323.

Fridriksson, J., Richardson, J. D., Baker, J. M., & Rorden, C. (2011). Transcranial directcurrent stimulation improves naming reaction time in fluent aphasia: Adouble-blind, sham-controlled study. Stroke, 42, 819–821.

Gandiga, P. C., Hummel, F. C., & Cohen, L. G. (2006). Transcranial DC stimulation(tDCS): A tool for double-blind sham-controlled clinical studies in brainstimulation. Clinical Neurophysiology, 117, 845–850.

Glosser, G., & Deser, T. (1990). Patterns of discourse production among neurologicalpatients with fluent language disorders. Brain and Language, 40, 67–88.

Grice, H. P. (1975). Logic and conversation. In: P. Cole, & J. P. Morgan (Eds.), Syntaxand semantics: Speech acts (pp. 41–58). New York: Academic Press.

Hagoort, P. (2003). How the brain solves the binding problem for language: Aneurocomputational model of syntactic processing. NeuroImage, 20, 18–29.

Hagoort, P. (2005). On Broca, brain, and binding: A new framework. Trends inCognitive Sciences, 9, 416–423.

Halliday, M. A. K., & Hasan, R. (1976). Cohesion in English. New York: Longman.Hammer, A., Jansma, B. M., Tempelmann, C., & Munte, T. F. (2011). Neural

mechanisms of anaphoric reference revealed by fMRI. Frontiers in Psychology,32, 1–9.

Haslett, B. (1983). Children's strategies for maintaining cohesion in their writtenand oral stories. Communication Education, 32, 91–106.

Hickmann, M., & Hendriks, H. (1999). Cohesion and anaphora in children'snarratives: A comparison of English, French, German, and Mandarin Chinese.Journal of Child Language, 26, 419–452.

Holland, R., Leff, A. P., Josephs, O., Galea, J. M., Desikan, M., & Price, C. J. (2011).Speech facilitation by left inferior frontal cortex stimulation. Current Biology, 21,1403–1407.

Huber, W., & Gleber, J. (1982). Linguistic and non-linguistic processing of narrativesin aphasia. Brain and Language, 16, 1–18.

Lai, G. (1993). Conversazionalismo. Torino: Bollati Boringhieri.Marangolo, P. (2010). Riabilitazione nella fase degli esiti. Acta Phoniatrica Latina, 32,

289–300.Marangolo, P., Marinelli, C. V., Bonifazi, S., Fiori, V., Ceravolo, M. G., Provinciali, L.,

et al. (2011). Electrical stimulation over the left inferior frontal gyrus (IFG)determines long-term effects in the recovery of speech apraxia in three chronicaphasics. Behavioral Brain Research, 225, 498–504.

Marangolo, P., Fiori, V., DiPaola, M., Cipollari, S., Razzano, C., & Oliveri, M. (2013).Differential processing of the left frontal and temporal regions in verb naming:A tDCS treatment study. Restorative Neurology and Neuroscience, 31, 63–72.

Marini, A., & Carlomagno, S. (2004). Analisi del discorso e patologia del linguaggio.Milan, Italy: Springer Verlag Italia.

Marini, A., Caltagirone, C., Pasqualetti, P., & Carlomagno, S. (2007). Patterns oflanguage improvement in adults with non-chronic non-fluent aphasia afterspecific therapies. Aphasiology, 21, 164–186.

Marini, A., Andreetta, S., Del Tin, S., & Carlomagno, S. (2011). A multi-level approachto the analysis of narrative language in aphasia. Aphasiology, 25, 1372–1392.

Marini, A., & Urgesi, C. (2012). Please get to the point! A cortical correlate oflinguistic informativeness. Journal of Cognitive Neuroscience, 24, 2211–2222.

Miceli, G., Laudanna, A., Burani, C., & Capasso, R. (1994). Batteria per l’analisi deideficit afasici BADA. Roma, Italy: CEPSAG, Policlinico Gemelli, Università Catto-lica del Sacro Cuore.

Naeser, M. A., Martin, P. I., Nicholas, M., Baker, E. H., Seekins, H., Kobayashi, M., et al.(2005). Improved picture naming in chronic aphasia after TMS to part of rightBroca's area: An open-protocol study. Brain and Language, 93, 95–105.

Naeser, M. A., Martin, P. I., Treglia, E., Ho, M., Kaplan, E., Bashir, S., et al. (2010).Research with rTMS in the treatment of aphasia. Restorative Neurology andNeuroscience, 28, 511–529.

Nicholas, L. E., & Brookshire, R. H. (1993). A system for quantifying the informa-tiveness and efficiency of the connected speech of adults with aphasia. Journalof Speech and Hearing Research, 36, 338–350.

Nishitani, N., Schürmann, M., Amunts, K., & Hari, R. (2005). Broca's region: Fromaction to language. Physiology, 20, 60–69.

Nitsche, M. A., & Paulus, W. (2000). Excitability changes induced in the humanmotor cortex by weak transcranial direct current stimulation. The Journal ofPhysiology, 527, 633–639.

Oliveri, M., Rossini, P. M., Traversa, R., Cicinelli, P., Filippi, M. M., Pasqualetti, P., et al.(1999). Left frontal transcranial magnetic stimulation reduces contralesionalextinction in patients with unilateral right brain damage. Brain, 122, 1731–1739.

Orsini, A., Grossi, D., Capitani, E., Laiacona, M., Papagno, C., & Vallar, G. (1987).Verbal and spatial immediate memory span: Normative data from 1355 adultsand 1112 children. Italian Journal of Neurological Sciences, 8, 537–548.

Poreisz, C., Boros, K., Antal, A., & Paulus, W. (2007). Safety aspects of transcranialdirect current stimulation concerning healthy subjects and patients. BrainResearch Bulletin, 72, 208–214.

Ripich, N. D., Terell, B. Y., & Spinelli, F. (1983). Discourse cohesion in senile dementiaof the Alzheimer type. In: RH Brookshire clinical aphasiology conferenceproceedings (pp. 316–321). Minneapolis BRK Publishers.

Ripich, N. D., Vertes, D., Whitehouse, P., & Fulton, S. (1988). Conversational discoursepatterns in senile dementia of the Alzheimer's type patients. Paper presented at theAmerican Speech Language. Hearing Association Annual Convention Boston MA.

Rochester, S., & Martin, J. (1977). The art of referring: The speaker's use of nounphrases to instruct the listener. In: R. Freedle (Ed.), New directions in discourseprocessing (pp. 245–270). Norwood, N. J.: Ablex Publishing.

Rochester, S., & Martin, J. (1979). Crazy talk: A study of discourse of schizophrenicspeakers. New York: Plenum Press.

Saur, D., Lange, R., Baumgaertner, A., Schraknepper, V., Willmes, K., Rijntjes, M.,et al. (2006). Dynamics of language reorganization after stroke. Brain, 129,1371–1384.

Saur, D., Kreher, B. W., Schnell, S., Kümmerer, D., Kellmeyer, P., Vry, M. S., et al.(2008). Ventral and dorsal pathways for language. Proceedings of the NationalAcademy of Sciences of the United States of America, 105, 18035–18040.

Sparing, R., Dafotakis, M., Meister, I. G., Thirugnanasambandam, N., & Fink, G. R.(2008). Enhancing language performance with non-invasive brain stimulation-a transcranial direct current stimulation study in healthy humans. Neuropsy-chologia, 46, 261–268.

Ulatowska, H. K., North, A. J., & Macaluso-Hayes, S. (1981). Production of narrativediscourse and procedural discourse in aphasia. Brain and Language, 13,345–371.

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256 255

Ulatowska, H. K., Freedman-Stern, R., Doyel, A. W., Macaluso-Haynes, S., & North, A.J. (1983). Production of narrative discourse in aphasia. Brain and Language, 19,317–334.

Warburton, E., Price, C. J., Swinburn, K., & Wise, R. J. (1999). Mechanisms of recoveryfrom aphasia: Evidence from positron emission tomography studies. Journal ofNeurology Neurosurgery and Psychiatry, 66, 155–161.

Wilkinson, R., & Wielaert, S. (2012). Rehabilitation targeted at everyday commu-nication: Can we change the talk of people with aphasia and their significant

others within conversation? Archives of Physical Medicine and Rehabilitation, 93,S70–S76.

Winhuisen, L., Thiel, A., Schumacher, B., Kessler, J., Rudolf, J., Haupt, W. F., et al.(2007). The right inferior frontal gyrus and post-stroke aphasia: A follow upinvestigation. Stroke, 38, 1286–1292.

Zimmermann, P., & Fimm, B. (1994). Tests d’evaluation de l’attention (TEA). Psytest.

P. Marangolo et al. / Neuropsychologia 53 (2014) 246–256256