Brain & Language 140 (2015) 35–48

Contents lists available at ScienceDirect

Brain & Language

journal homepage: www.elsevier.com/locate/b&l

The Dutch Linguistic Intraoperative Protocol: A valid linguistic approach

to awake brain surgery
E. De Witte a, D. Satoer b,c, E. Robert d, H. Colle d, S. Verheyen e, E. Visch-Brink b, P. Mariën a,f,⇑
a
Clinical and Experimental Neurolinguistics, Centre for Linguistics, Vrije Universiteit Brussel, Brussels, Belgium
b
Department of Neurosurgery, Erasmus University Medical Centre, Rotterdam, The Netherlands
c
Department of Neurolinguistics, Groningen Center of Expertise for Language and Communication Disorders (GELC), University of Groningen, The Netherlands
d
Department of Neurosurgery, AZ Sint-Lucas, Ghent, Belgium
e
Department of Psychology, Catholic University Leuven, Belgium
f
Department of Neurology and Memory Clinic, ZNA Middelheim, Antwerp, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: Intraoperative direct electrical stimulation (DES) is increasingly used in patients operated on for tumours
Accepted 25 October 2014 in eloquent areas. Although a positive impact of DES on postoperative linguistic outcome is generally
Available online 17 December 2014 advocated, information about the neurolinguistic methods applied in awake surgery is scarce. We
developed for the first time a standardised Dutch linguistic test battery (measuring phonology,
Keywords: semantics, syntax) to reliably identify the critical language zones in detail. A normative study was carried
Awake surgery out in a control group of 250 native Dutch-speaking healthy adults.
Intraoperative language testing
In addition, the clinical application of the Dutch Linguistic Intraoperative Protocol (DuLIP) was
Language mapping
Brain tumour
demonstrated by means of anatomo-functional models and five case studies. A set of DuLIP tests was
Direct electrical stimulation selected for each patient depending on the tumour location and degree of linguistic impairment. DuLIP
Corticosubcortical mapping is a valid test battery for pre-, intraoperative and postoperative language testing and facilitates
Standardised test battery intraoperative mapping of eloquent language regions that are variably located.
Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction necessary for an adequate communication. Verb–noun association

tasks, auditory and visual comprehension tasks, repetition, reading,
Intraoperative language mapping during awake surgery is and writing tasks are occasionally used in experimental settings
considered to be the ‘gold standard’ since a positive impact of direct and some studies described a more tailored selection taking into
electrical stimulation (DES) on postoperative linguistic outcome is account tumour location (Bello et al., 2007, 2008; Bertani et al.,
generally advocated (De Witt Hamer, Robles, Zwinderman, 2009; Duffau et al., 2003; Duffau, Lopes, et al., 2005) (for a review
Duffau, & Berger, 2012; Duffau, Moritz-Gasser, & Gatignol, 2009; see De Witte & Mariën, 2013). As a result, test batteries including
Ilmberger et al., 2008; Pereira et al., 2009; Sanai & Berger, 2010; specific and sensitive tasks are needed for intraoperative language
Sanai, Mirzadeh, & Berger, 2008; Spena et al., 2010). DES enables testing in order to preserve language function. Polczynska (2009)
to identify critical cortical and subcortical language areas and path- suggested three sets of ‘home made’ tasks consisting of: (1)
ways, which cannot be resected without post-operative language ‘grammar-focused tests’, (2) ‘non-dominant right-hemisphere
deficits (De Witt Hamer et al., 2012; Duffau, 2007; Duffau, Lopes, tests’, and (3) ‘tests for subcortical stimulation’. Coello et al.
et al., 2005; Spena et al., 2010). During DES, picture naming (2013) proposed a minimal common protocol consisting of lan-
(Metz-Lutz et al., 1991) and counting are commonly used as guage tasks such as picture naming, semantic association and read-
language paradigms (Bertani et al., 2009; Duffau, 2007). However, ing. In the study of Miceli, Capasso, Monti, Santini, and Talacchi
this approach, in which word-finding abilities and automatic (2012), the following language tasks were included to assess lexi-
speech are investigated in a restricted way, does not cover the cal-semantic processing: naming (both objects and actions), lexical
integral variety of expressive and receptive language functions, decision (both auditory and visual) and word–picture matching
(nouns and verbs, auditory and visual). Papagno et al. (2012)
⇑ Corresponding author at: Department of Clinical & Experimental Neurolinguis- suggested a neuropsychological battery (Milano-Bicocca Battery
tics, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium. MIBIB) to assess cognition (language, memory, executive func-
E-mail address: peter.marien5@telenet.be (P. Mariën). tions, apraxia, spatial cognition) in low-grade glioma’s pre- and

http://dx.doi.org/10.1016/j.bandl.2014.10.011
0093-934X/Ó 2014 Elsevier Inc. All rights reserved.
36 E. De Witte et al. / Brain & Language 140 (2015) 35–48

postoperatively. Intraoperatively, they use naming of objects, Table 1

actions and people. However, in general little is known about the Demographic characteristics of the 250 healthy participants.

feasibility, the construction, the validity of the tasks, and no Demographics Mean (median) SD Range
normative data are available. The selection of the type of language (A) Age, education, MMSE (mean, SD, range)
paradigms during DES procedures is of crucial importance because Age in years 51.29 (54.00) 19.81 18–92
of its direct effect on clinical outcome (Hamberger, Seidel, Education in years 13.78 (14.00) 3.27 6–23
Mckhann, Perrine, & Goodman, 2005). Therefore the sensitivity of MMSE 29.16 (30.00) 1.18 24–30

each language task used during DES needs to be investigated Demographics Groups Number of subjects Percentage
(Satoer et al., 2012; Talacchi, Santini, Savazzi, & Gerosa, 2011; (B) Gender, age, education, province, handedness groups
Talacchi et al., 2012; Teixidor et al., 2007). We attempted to achieve Gender M 101 40.40
this goal by developing for the first time a Dutch standardised F 149 59.60
Age 18–54 y 129 51.60
neurolinguistic test battery for awake surgery in critical language
55+ y 121 48.40
areas. The language test battery, called DuLIP (Dutch Linguistic Education level 612 y 104 41.60
Intraoperative Protocol) includes a variety of phonological, >12 y 146 58.40
semantic, syntactic and articulatory tests, standardised in a healthy Province West Flanders 44 17.60
East Flanders 114 45.60
control population of 250 neurologically healthy adults (see part 2
Antwerp 20 8.00
and 3). The clinical use of DuLIP in awake surgery is illustrated with Flemish Brabant 40 16.00
a ‘location-function-task’ model and this model is applied to select Limburg 32 12.80
an appropriate intraoperative language test battery to assess Handedness L 13 5.20
linguistic functions in patients with tumoural lesions in eloquent A 8 3.20
R 229 91.60
areas (see part 4).
(A) Legend: SD = standard deviation; MMSE = Mini Mental State Examination.
(B) Legend: M = male, F = female, y = year, L = left-handed, A = ambidexter
2. Methods (Laterality Quotient P 40 en 6+40), R = right-handed.

2.1. Procedure

The first two authors of this manuscript and a group of master Belgium (provinces: West-Flanders, East-Flanders, Antwerp,
level students in clinical linguistics, speech and language pathol- Flemish Brabant, Limburg). The demographic data of the healthy
ogy and clinical psychology were trained to administer the set of participants are summarised in Table 1.
linguistic tests of the DuLIP, consisting of one object naming test,
four phonological tests (repetition words, phonological odd word 2.3. Test administration, registration and correction
out, phonological fluency, phonological sentence judgment), eight
semantic tests (semantic odd word out, semantic picture out, The tests were administered in an identical order. Practice items
semantic association, sentence completions (closed and broad con- were included to ensure that participants understood the task.
text), semantic fluency animals + jobs, semantic sentence judg- Visual stimuli were presented in a powerpoint presentation on a
ment), five syntactic1 tests (action naming, verb generation, laptop screen (Table 2). A division was made between tasks for
syntactic fluency, syntactic sentence judgment I & II) and one artic- use during DES stimulation within 4 s (A) and tasks for use during
ulatory test (verbal diadochokinesis). 250 healthy volunteers partic- resection (no time restriction) (B). Since corticosubcortical stimu-
ipated in the normative study. All test data were quantitatively (see lation may maximally last 4 s, all language tasks should ideally
statistics) and qualitatively analysed. be performed within this time frame. However, certain tasks
measuring important language aspects such as the fluency tasks
2.2. Subjects (language initiation) and judgment tasks (language perception)
are not feasible within 4 s. Since these tasks provide valuable
250 healthy adult volunteers entered the normative study for information they were included in the protocol and used during
the DuLIP. Inclusion criteria consisted of (1) Dutch as a mother ton- resection in between stimulations. Of course, the tasks for stimula-
gue, (2) no (history of) cardiovascular, neurological, psychiatric, or tion (A) are also very relevant to apply during resection. Test items
developmental language and/or speech disorders, (3) no toxic sub- for use during DES were presented every 4 s accompanied by a
stance abuse (no drug or alcohol abuse), (4) normal vision, (5) nor- beep called ‘laser sound’. To avoid fatigue the DuLIP protocol was
mal hearing, (6) no use of sleep medication, psychofarmaca or assessed in two parts (A & B) at two different test moments. The
neuroleptic drugs, and (7) a score above 24/30 on the Mini Mental total length of both assessments (heteroanamnesis, EHI, MMSE,
State Examination (MMSE) (generally accepted cut-off for DuLIP) was 1.5 h. The answers were recorded verbatim and tran-
impairment is 23; Crum, Anthony, Bassett, & Folstein, 1993; scribed. One point was given if the target answer was correctly pro-
Folstein, Folstein, & McHugh, 1975) (http://faculty.pepperdine.edu/ duced and within the given time frame. Self-corrections were
shimels/Courses/Files/MMSE.pdf). The participants were informed allowed if given within the 4 s time frame. One repetition of an audi-
about the goals of the project and gave a written informed consent. tory task on request of the participant was allowed. The accuracy of
The study was approved by the ethical committee of the Vrije the scoring of the tests was checked by one of the first two authors
Universiteit Brussel. The number of participants per age group and was based on qualitative terms agreed on a priori. Uncertainties
was defined on the basis of current statistics about the prevalence regarding the correctness of test items were discussed in group with
of supratentorial tumours in the Flemish adult population (30% the other authors until a consensus was reached.
between 18 and 49 years, 55% between 50 and 74 years, 15%
75 years or older (Van Eycken & De Wever, 2006). Both male 2.4. Description of tests
(18–92 y) and female (18–86 y) participants were included. We
recruited subjects from a variety of culturally and philosophically The phonological, semantic, syntactic, naming tasks and articu-
different organizations over different rural and urban regions in latory task of the DuLIP are presented in Table 2. Taking into
account the constraints of the awake procedure (fixed position,
1
The term syntax is used to define tasks that measure grammatical aspects. limited time) we carefully selected existing language tests
 E. De Witte et al. / Brain & Language 140 (2015) 35–48 37

Table 2
Intraoperative linguistic tasks from DuLIP.

Timing of Linguistic Task Stimuli (examples)

assessment(s) level
DURING DES Phonology – repetition of words: – discussie (discussion)
(in 4 s)
 repetition of 3-syllabic words with alternating word – wortel (carrot)
accents
 repetition of 2-syllabic words – individu (individual)
 repetition of words with phonemic similarities – programma (programme)
 repetition of words with consonant clusters – wijn, pijn, lijn, kat (wine, pain, line, cat) ? answer kat (cat)
– reading with phonological odd word outa
Semantics – reading with semantic odd word outa – been, arm, raam, voet (leg, arm, window, foot) ? answer raam
(window)
a
– naming with semantic odd word out – pictures of ‘borstel, hond, kat’ (brush, dog, cat) ? answer borstal
(brush)
– semantic association taska – auto, fiets . . . (car, bike . . .) ? answer e.g. bus
– sentence completion (semantically induced sentences)a – Hij snijdt met een . . . (He cuts with a . . .) ? answer e.g. mes (knife)
Syntax – verb generationa – bal (ball) ? gooien/werpen (to throw)
– action naming (3rd person singular, transitive verbs), 60 – de man . . . (the man . . .) ? answer e.g. loopt (e.g. runs) ? picture of
actionsa (RUG-Vrije Universiteit Brussel) a man who is running
Articulation – verbal diadochokinesis test – repeat/papapa/, /pataka/, /papopu/, /pafpafpaf/, /dafnaflaf/, /
– praxis pafpofpuf/, /pafpaspaf/ (5x)
Naming – object naming, 100 objectsa – black and white drawings of objects
Dit is . . . (this is) ? answer hond (dog) ? picture of a dog
NOT DURING Phonology – phonological sentence judgment – De hokkel eet een gersie. (The hokkel eats a gersie.) ? ‘hokkel’,
DES ‘gersie’ non-existing words
– phonological fluency – Letters D, A, T
Semantics – semantic sentence judgment – De gieter smeert een boterham. (The watering-can prepares a
sandwich.) ? wrong sentence
– semantic fluency – animals/jobs
– sentence completion (less semantically induced sentences)a – Om 5 uur . . .(At 5 O’clock. . .) ? possible answer: ga ik naar huis (I go
home)
Syntax – syntactic sentence judgment – Er was over niets meer van het scheerapparaat. (There was over
nothing left of the razer) ? wrong word order
– verbal fluency (verbs) – verbs
a
Presented with powerpoint slides.

(i.e. phonological fluency & semantic fluency, see below for a who are not able to repeat 3-syllable words due to e.g. pre-existing
description), adapted language tasks (e.g. naming task) and com- language problems. The word list consists of different complexity
posed new tasks (e.g. semantic picture out) to include in the DuLIP levels. To increase the degree of articulatory and phonological
protocol. The Dutch databases CELEX (Baayen, Piepenbrock, & van complexity, we included words without consonant clusters
Rijn, 1993), SUBTLEX-NL (Keuleers, Brysbaert, & New, 2010) and (75 words for 3 syllable list, 100 words for 2 syllable list) (e.g.
Positiewoordenboek (Weijters, 1983) were used to control for the mandarijn – mandarine) and words with consonant clusters (e.g.
linguistic variables frequency, imageability (Van Loon-Vervoorn, projectiel – projectile) (50 words for 3 syllable list, 25 words for 2
1989), word length, word form (see linguistic variables per test). syllable list) as well as phonologically similar words without
As a rule, non-compound singular nouns were selected, and words consonant clusters (40 words) (e.g. karamel – caramel) and with
that could evoke emotional reactions such as ‘sadness, brain, death’ consonant clusters (10 words) (e.g. crimineel – criminal). Phonolog-
were avoided. Per test, detailed information is given regarding the ically or semantically related words were not consecutively pre-
design of the items and the linguistic variables within the linguistic sented to avoid perseveration. Studies have shown that words
levels to be assessed. The qualitative terms to consider a test item with similar phonemes (especially vowels) and words with conso-
as correct or incorrect were also described. In addition, we nant clusters are more prone to errors (Nespoulous & Moureau,
attempted to motivate the selection of tasks in relation to lesion 1998; Shattuck-Hufnagel & Klatt, 1980).
location. Repetition processes can be distorted after stimulation of the
left arcuate fascicle (AF) but also after stimulation of the left infe-
A. Tasks performed during DES within a time frame of 4 s per rior supramarginal gyrus (Maldonado, Moritz-Gasser, & Duffau,
item 2011; Moritz-Gasser & Duffau, 2013).

A.1. Phonological tests A.1.2. Reading with phonological odd word out
A.1.1. Repetition of words
The aim of this phonological selection task is to assess phono-
The aim of the word repetition task is to assess word production logical awareness via the lexical input route. Participants covertly
according to the phonological input- and output route. Participants read a list of four one-syllable words (25 slides) presented on a lap-
were instructed to repeat 2 or 3-syllable words. In the awake set- top screen and read aloud the word that does not phonologically fit
ting 3-syllable words are easier to catch by the patient and the (Fig. 1A). In order to present words with equal consonant clusters,
neurosurgeon than 2-syllabic words because of the varying intona- the syllable structure of each of the items was controlled. Evidence
tional pattern in Dutch. A list of 125 3-syllable words with varying in the literature on low grade gliomas shows that the insula, the
syllable stress was composed, containing stimuli such as, oefening inferior supramarginal gyrus, the AF and the inferior longitudinal
(exercise), discussie (discussion), avontuur (adventure). 125 2-sylla- fascicle (ILF) are crucially involved in phonological decision (Bello
ble words were used in the task to allow examination of patients et al., 2007; Maldonado et al., 2011; Moritz-Gasser & Duffau,
38 E. De Witte et al. / Brain & Language 140 (2015) 35–48

bal (ball) been (leg)

hal (hall) arm (arm)

val (fall) raam (window)

weg (road) voet (foot)

A B C
De man … (The man …)

bus auto … De bakker verkoopt …

(bus car ...) (The baker sells …)

D E F
Fig. 1. (A) Reading with phonological odd word out. The target item is weg (road) because it is the only word that does not rhyme on –al (bal, hal, val); (B) reading with
semantic odd word out. The target item is raam (window) because it is not semantically related to the three others representing body parts i.e. been (leg), arm (arm), voet
(foot); (C) naming with semantic picture out. The target item is borstel (brush) because it does not belong to the semantic category (animals) to which the other two pictures
(dog, cat) belong; (D) semantic association task. The target item is a semantically related word such as fiets (bike) or trein (train) that fits within the same semantic category
transport. Weg (road) is an incorrect target answer although it is semantically related it is not a means of transport; (E) sentence completion (closed context). Possible target
answer is brood (bread): De bakker verkoopt brood (The baker sells bread). De bakker verkoopt schoenen (The baker sells shoes) is an incorrect answer; (F) action naming. The
target item is loopt (synonyms such as rent, holt are also correct): De man loopt (The man runs).

2013). In this task the occipital lobe is also involved since the pre- for DES in posterior as well as frontal brain regions and to map
sented words need to be read (Gras-Combe, Moritz-Gasser, Herbet, the IFOF (Coello et al., 2013). Participants add a semantically
& Duffau, 2012; Richardson, Seghier, Leff, Thomas, & Price, 2011). related word to a list of two semantically related nouns consisting
of 1 or 2 syllables (25 items) (Fig. 1D).
A.2. Semantic tests
A.2.1. Reading with semantic odd word out A.2.4. Sentence completion (closed context)

The aim of this test is to assess semantic judgment via the The aim of the test is to assess the production of parts of speech
lexical visual input route. Verbal semantic knowledge, semantic elicited by a specific context and to measure reading, language
processing and reading are assessed. These functions may be dynamics and production of semantically and syntactically appro-
distorted after stimulation of the temporal (posterior and superior) priate speech. Participants read sentences out loud and complete it
regions, the prefrontal (orbitofrontal and dorsolateral) regions in a meaningful way (25 items) (Fig. 1E). Reduction of spontaneous
(Duffau, Gatignol, et al., 2005), the parietal regions (Roux et al., speech has been found after stimulation of the supplementary
2004) and the occipital regions (Gras-Combe et al., 2012; motor area (Fontaine, Capelle, & Duffau, 2002), Broca’s area, the
Richardson et al., 2011). At the subcortical level stimulation of insula and the subcallosal fascicle and the ILF (Bello et al., 2007).
the uncinate fascicle (UF), the ILF (Bello et al., 2007), and the
inferior fronto-occipital fascicle (IFOF) may result in disruption of A.3. Syntactic tests
semantic judgment (Bello et al., 2007; Duffau, Gatignol, et al., A.3.1. Action naming
2005; Moritz-Gasser, Herbet, & Duffau, 2013). Participants are
instructed to read aloud the word that does not semantically fit The test aims to assess syntactic (and semantic) processing. On
in a list consisting of 4 one-syllabic or 4 two-syllabic words each slide a short introductive phrase is presented that has to be
(25 items) (Fig. 1B). completed with a finite verb in the third person singular depicted
by black and white drawings of an action (Fig. 1F). The reading
A.2.2. Naming with semantic odd picture out aloud of the introductive phase is used to ensure that distorted
verb naming is not due to seizures induced by the stimulation.
The aim of this test is to assess non-verbal semantic judgment Verb assessment in the third person present singular allows
and naming. As in the semantic odd word out task, semantic on-line monitoring of the grammatical processes of person and
knowledge and processing is assessed but pictures (Snodgrass & number agreement between the subject and the action verb.
Vanderwart, 1980) are used instead of written words. Therefore, Specific action naming sites are mainly identified in Broca’s area
this task can be used to stimulate temporal, prefrontal regions as and in the posterior midfrontal region (Lubrano, Filleron,
well as for the IFOF and the ILF (Moritz-Gasser et al., 2013). Partic- Demonet, & Roux, 2014). Test items are presented in 3 blocks of
ipants name the picture that does not semantically fit in a series 20 items (60 items in total) followed by a resting period. The
consisting of 3 black and white drawings of objects and animals selection of items was based on a cooperative study between
(25 items) (Fig. 1C). Two of the three pictures belong to the same Roelien Bastiaanse (University of Groningen) and Peter Mariën
semantic category. (Vrije Universiteit Brussel) (Rofes, 2012).

A.2.3. Semantic association task A.3.2. Verb generation

The aim of this test is to assess lexical-semantic processing. Verb generation is commonly used during function Magnetic
Reading, semantic knowledge and generating words according a Resonance Imaging (fMRI) to locate Broca’s area (Engström,
given concept are assessed. As a result this task may be relevant Ragnehed, Lundberg, & Söderfeldt, 2004). The test aims to assess
 E. De Witte et al. / Brain & Language 140 (2015) 35–48 39

semantic association as well as syntactic competence and can be B. Tasks performed with no time frame
used during stimulation of the perisylvian regions (Bertani et al., B.1. Phonological tests
2009; Bizzi et al., 2008). Participants produce verbs that are B.1.1. Phonological fluency
semantically related with the singular nouns presented on a laptop
screen (50 items). The stimuli were selected from ‘Spreektaal, The aim of this test requiring generating words with a given
Woordfrequenties in Gesproken Nederlands’ (De Jong, 1979), phoneme is to assess flexibility of phonological thought and inter-
which contains lists of words most frequently used in spoken nal driven language which are typically located in the frontal areas
Dutch. Next to frequency and morphological category, two other (Baldo, Schwartz, Wilkins, & Dronkers, 2006) involving the supple-
selection criteria were used to sample the nouns: imageability mentary motor areas (Satoer, Kloet, Vincent, Dirven, & Visch-Brink,
and word length. Sixty highly imageable nouns of maximally 2 2013). Participants produce within one minute as many words as
syllables were selected from the corpus. Ten items were deleted possible beginning with a particular phoneme. The phonemes D,
from the final list because in Dutch some (typically associated) A, T were chosen (Schmand, Groenink, & van den Dungen, 2008).
target verbs strongly relate to the morphological form of the item
due to an identical wordstem (e.g. fiets-fietsen (bike-to bike), B.1.2. Phonological sentence
spel-spelen (play-to play)). If they were included, the patient only
had to add ‘ –en’, which is too easy. The aim of this test is to assess phonological awareness by dis-
criminating between auditorily presented phonologically correct
A.4. Verbal diadochokinesis test and incorrect sentences (n = 30) including non-words (e.g. De hok-
kel eet een gersie – The hokkel eats a gersie). The stimuli were
The verbal diadochokinesis test aims to investigate the planning homemade. Phonological decoding and verbal short-term memory
and coordination abilities of the motor speech system. It is a are measured. No comparable task has been reported in the litera-
sensitive task to assess articulation which is generally represented ture so far. This task could be informative to assess function of the
in the inferior frontal gyrus, frontal ventral premotor areas and language areas in the middle posterior temporal area (Hickok,
anterior insular regions. Because of its limited plasticity these 2012) and at the subcortical level including the AF, SLF and the
regions should be preserved during surgery (Van Geemen, ILF (Bello et al., 2007; Maldonado et al., 2011; Moritz-Gasser &
Herbet, Moritz-Gasser, & Duffau, 2013). In addition, the test can Duffau, 2013).
be used during subcortical stimulation of the part of the superior
longitudinal fascicle (SLF) that connects the supramarginal gyrus B.2. Semantic tests
with the ventral premotor cortex (Maldonado et al., 2011) as well B.2.1. Semantic fluency
as the corticospinal tract (CST) (Bello et al., 2007). Participants
repeat 9 sequences of monosyllables as fast as possible 5 times The aim of this task is to assess word generation within a spe-
in a row. Both alternating (alternating different phonemes) cific conceptual category measuring the semantic knowledge rep-
(DAF PAF LAF) and sequential diadochokinesis (repeating identical resented in temporal areas (Baldo et al., 2006). The participants
phonemes) (PAF PAF PAF) are used. have to produce in one minute as many words as possible within
a given semantic category. The semantic categories animals and
A.5. Object naming jobs were chosen according to the data collection of Luteijn and
Barelds (2004) in a Dutch population (GIT I, II).
Object naming is typically applied during intraoperative map-
ping with DES (for a review see De Witte & Mariën, 2013). Naming B.2.2. Judgment of semantically anomalous and correct
errors have been described after stimulations in the left perisylvian sentences
regions and frontal premotor cortex and subcortically in the AF, ILF,
IFOF, subcallosal fascicle, UF (Bello et al., 2007; Coello et al., 2013; The aim of this test is to assess via the auditory input route,
Moritz-Gasser & Duffau, 2013). Specific object naming sites are semantic awareness, knowledge and comprehension, by discrimi-
mainly localised in Broca’s area and the temporal cortex (Lubrano nating between semantically correct and incorrect sentences
et al., 2014). (n = 50) with violations of semantic selection restrictions (e.g. De
The aim of this visual object naming task is to assess word gieter smeert een boterham – The watering-can prepares a sand-
retrieval. Participants name 100 black and white drawings of wich). This task can be used for temporal cortical and subcortical
objects and animals depicted on the screen after producing an areas (Bello et al., 2007; Bertani et al., 2009; Pereira et al., 2009;
introductive phrase ‘dit is een . . . (this is a . . .)’. The introductive Santini et al., 2012). The sentences were based on BOX part 6 of
phase is used for intraoperative condition to ensure that distorted Visch-Brink and Bajema (2001).
naming is not due to seizures induced by the stimulation. Black
and white drawings were selected from Boston Naming Test B.2.3. Semantic sentence completion (broad context)
(BNT) (Kaplan, Goodglass, & Weintraub, 1983; Mariën, Mampaey,
Vervaet, Saerens, & De Deyn, 1998), the Dutch versions of the Psy- The aim of the task is to assess speech production within a sen-
cholinguistic Assessments of Language Processing in Aphasia tence frame with obligatory and less obligatory parts of speech. It
(PALPA-NL 52) (Bastiaanse, Bosje, & Visch-Brink, 1995), ATP (Audi- measures reading, language dynamics, and production of semanti-
tief Taalbegripsprogramma) (Bastiaanse, 2010) and Graded Nam- cally (and syntactically) appropriate speech. The task can be used
ing Test (GNT) (Warrington, 1997). Both living and non-living in the supplementary motor area, the temporoparietal regions,
things are included in the test since they may activate different the insula, the subcallosal fascicle and the ILF (Bello et al., 2007;
regions (Martin & Chao, 2001). High, middle and low frequent Fontaine et al., 2002). Subjects read out loud 25 sentences, which
words were selected and different word lengths were used. As such have to be completed in a meaningful way. A phrase instead of a
the task allows assessment and manipulation of lexical retrieval single word (see A.2.4) has to be added and the sentence has to
and phonological encoding. Test items were randomly presented be grammatically correct (with conjugated verbs). For example
in blocks of 25 followed by a resting period. Om 5 uur rijdt de buurman naar zijn werk (At 5 O’clock the
40 E. De Witte et al. / Brain & Language 140 (2015) 35–48

neighbour drives to his work) is a correct answer. Samen met Lies responses. First, we accounted for the percentages of correct
lopen (Running together with Lies) is not a correct sentence in answers of each item and excluded items with less than 80% of pic-
Dutch. In contrast to the sentences of subtest A.2.4, the sentences ture-name agreement from further analysis. After the naming
included in this task do not strongly trigger a particular semantic agreement study 100 out of 150 test items were used for object
response but allow a diversity of answers. naming and 60 out of 100 items were included for action naming.

B.3. Syntactic tests

3. Results
B.3.1. Syntactic fluency
No significant difference in mean age (p = 0.369) and mean
The aim of this task is to assess the flexibility of syntactic
years of education (p = 0.061) was found between the male and
thought mainly organised in frontal areas (inferior, middle) but
female participants of this study. In general, the healthy subjects
also in the parietal lobe (Ardila, 2012). The subject has to produce
with a normal cognitive level obtained high scores and a ceiling
within one minute as many verbs as possible.
effect was found for all tasks but the fluency tasks (Appendix 1).
The tasks that elicited most errors in the study population were
B.3.2. Syntactic sentence judgment
the naming tasks, and some of the semantic tasks: the semantic
odd word out task, the semantic sentence completion task (broad
This task aims to assess syntactic awareness in frontal areas by
context) and the semantic association task. However, these tasks
discriminating between syntactically correct and incorrect
still have an accuracy rate (number of correct items divided by
sentences. Twenty-five stimulus sentences were selected from
total number of items multiplied by 100) of more than 90% except
the Werkwoorden-en Zinnentest (WEZT) (Bastiaanse, Maas, &
for the semantic association task (83.44%). The semantic associa-
Rispens, 2000) (e.g. De hond heeft de man geslaan – The dog has
tion task turned out to be the most difficult task with a lot of
beaten the man) and 25 sentences were home-made. The home-
‘delayed answers’ (not within 4 s = error) (see Appendices 2–4).
made sentences included word order errors (e.g. Er was over niets
For most of the DuLIP tasks there was a significant effect of age
meer van het scheerapparaat – There was over nothing left of the
(18–54, 55 y) (the higher the age, the lower the scores) and years of
raser), 8 sentences with verb inflection errors (e.g. Wij was van plan
education (612 y, >12 y) (the higher the education level, the higher
naar de film te gaan – We was planning to go to the movie) and 2
the scores), whereas gender groups (M, F) did not differ signifi-
sentences with incorrect pronouns (e.g. Dat is niet van jullie, dat
cantly (Appendix 1). There were some exceptions. No significant
is van wij – This is not from you, this is from we).
differences were observed between age/education/gender groups
on the subtest ‘verbal diadochokinesis accuracy’. No significant
2.5. Statistics
effect was found for education effect in the semantic sentence
completion task (closed context) (p = .001) and no significant effect
The data obtained in the cohort of 250 healthy participants
for age was found in the phonemic judgment task (p = .029), but
were analysed with SPSS Statistics (v. 22). The dependent variables
the p-value was smaller than the generally used alpha .05 in both
included the number of correct answers per test. For the analyses
tasks.
of the results, two gender groups (male, female), two age groups
and two education groups were compared. The age groups were
divided according to the median, i.e. 54 (18–54 y, >55 y). Education 4. Clinical application of DuLIP in tumour population
groups consisted of less or equal to 12 years of education and more
than 12 years of education. The choice to present the data in this 4.1. Procedure
way was based both on content and statistical criteria. There is a
lot of discussion about the age at which cognitive decline starts, To illustrate the clinical use of the DuLIP protocol the pre-, intra-
however, a lot of researchers suggest somewhere around the age and postoperative data of 5 patients with tumours in different elo-
of 55 (Rönnlund, Nyberg, Bäckman, & Nilsson, 2005). Since more quent brain areas are demonstrated. In the pre- and postoperative
than 12 years of education implies continuation of education after phase (6 weeks and 6 months postsurgery), the DuLIP Protocol is
the age of 18 years (bachelor or master degree), this cross over administered together with tasks for the other cognitive (memory,
point is frequently used in normative studies (Miatton, Wolters, attention, visuoperception, execution) functions (2 h of testing
Lannoo, & Vingerhoets, 2004). From the statistical point of view, divided into two test moments). The normative data and especially
this division of age and education allows for group comparisons the cut-off scores are used to define the presence and severity of a
with sufficient power (p = >.80). pre- and/or postoperative language impairment.
The non-parametric Mann–Whitney U test was used to measure In the intraoperative phase, the asleep–awake–asleep proce-
the effects of age/education/gender on the ordinal test scores. dure with corticosubcortical mapping is performed (Duffau,
A p-value of 6.00089948 (Sidak, 1967) was used as criterion of 2007). The mean time of the awake procedure is one hour and a
significance since multiple testing was involved (19  3 for age, half. For intraoperative testing, an individual-tailored battery of
education, gender) (see Appendix 1). The normative data were split language tasks from DuLIP is composed based on the patient’s
for, age, education and/or gender if these groups differed characteristics (preoperative language level, job) and the tumour
significantly. Mean, median, ranges and cut-off scores were then characteristics (location) (Coello et al., 2013). If relevant, other cog-
calculated per linguistic test both at the level of the entire group nitive tasks or non-language tasks are selected. The tailored test
and at the level of age/education combinations (see Appendices battery is practised in the preoperative phase. The items the
2–4). We defined the cut-off scores according to the generally patient is unable to perform correctly in the preoperative assess-
accepted method in clinical practice: i.e. the raw scores that go ments are left out of the set for intraoperative testing to ensure
together with percentile 7 (equal to standard deviation 1.5.) (clinically that the errors in the awake setting are due to cortical stimulation
impaired) and percentile 2 (equal to standard deviation 2.0) and not caused by a pre-existing deficit (Duffau, 2007). It is gener-
(pathologically impaired) (Palmer, Boone, Lesser, & Wohl, 1998). ally accepted that transient language deficits occurring during 3
With respect to object and action naming we additionally per- stimulations (with subsequent normalisation) is sufficient to
formed a picture naming agreement study (Kemmerer & Tranel, ensure that the stimulated site is essential for language processing.
2000). Therefore, the data were classified into correct and incorrect The selection of appropriate language tasks taking into account the
 E. De Witte et al. / Brain & Language 140 (2015) 35–48 41

tumour location is a complex process since speech and language dominant hemisphere. These models may form the starting-point
are represented in parallel, distributed and interactive corticosub- for the selection of language tasks in patients with brain tumours.
cortical networks (Catani & Thiebaut de Schotten, 2008; Coello In addition to the frequent use of naming tasks, other language task
et al., 2013; Hickok, 2012) (see Section 4.2). from the DuLIP may be selected based on the cortical and
subcortical areas involved (e.g. semantic sentence completion in
4.2. Task selection per cortical site and subcortical pathway the SMA area, repetition for AF) (see Section 4.3). During cortical
stimulation the tasks can be applied in serial order depending on
The construction of anatomoclinical correlations in patients the location of the stimulation (e.g. verb generation in frontal area,
with brain tumours is difficult given the fact that brain plasticity semantic picture out in temporal area). At the subcortical level, we
mechanisms play an important role (Duffau, 2008, 2009, 2013). suggest to apply tasks in parallel (e.g. semantic picture out and
Based on the available literature, studies and clinical practice we motor function by touching the item on the screen) and alternating
constructed models that broadly represent the functions and order (switching between different tasks) to not overlook any
preferred language tasks per cortical site (Fig. 2, Table 3) and function since subcortical pathways connect several cortical and
subcortical pathway (Fig. 3, Table 4) of the (left) language subcortical areas.

Fig. 2. Per coloured cortical site the function(s) and most suitable language tasks from the DuLIP are shown in Table 3. (B) Ant = anterior; Med = medial; Post = posterior;
SMG = supramarginal gyrus; ANG = angular gyrus; a = anterior; p = posterior; LAT = lateral; sup = superior; inf = inferior.

Table 3
Selection of DuLIP tasks per cortical site.

Frontal Regions Function(s) Intraoperative language tasks from DuLIP

inferior frontal gyrus Articulatory processing/ syntax/ (writing) Verbal diadochokinesis, repetition/
verb generation, action naming
posterior midfrontal gyrus Action naming/ (writing) Action naming

supplementary motor area Language initiation Sentence completion (closed and broad context),
(posterior superior frontal gyrus) fluency
precentral gyrus Motor network Repetition, verbal diadochokinesis

Temporal Regions Function Intraoperative language tasks

posterior superior temporal gyrus semantics, auditory comprehension, semantic picture out, semantic judgment, object
naming living objects naming
middle posterior superior temporal Phonological network Phonological odd word out, phonological judgment
sulcus
middle inferior temporal gyrus Lexical interface (link phonological – Semantic judgment, object naming
semantic information), naming non-living
objects
anterior middle temporal gyrus Famous face naming, (memory) Naming

Parietal Regions Function(s) Intraoperative language tasks

supramarginal gyrus Reading, naming, semantics, (memory) Reading + sentence completion, semantic
(SMG) association, naming
angular gyrus Reading, writing, (calculation) Reading + sentence completion
(ANG)
42 E. De Witte et al. / Brain & Language 140 (2015) 35–48

4.3. Case reports Deliganis, Dobbins, & Keles, 1994) in the frontal lobe (Fig. 4CD
and added in Table 5). At 6 weeks and 6 months postsurgery the
4.3.1. Case 1 DuLIP protocol was repeated and no aggravation of language
A 48-year-old right-handed (Edinburgh Handedness Inventory occurred. As in the preoperative phase phonological (RS = 23; <Pc
(EHI) = +100; Oldfield, 1978) man with an educational level of 2) and syntactic fluency (RS = 7; <Pc 2) were reduced 6 weeks
13 years developed a seizure and right sensory-motor syndrome postsurgery. Six months postsurgery improvement was noticed
after a period with paroxysms of severe headaches and dizziness. but defective results for phonological (RS = 24; <Pc 2) and syntactic
MRI and Diffusion Tensor Imaging (DTI) of the brain revealed a fluency (RS = 13; <Pc 2) were still present.
tumour in the left inferior frontal gyrus, near to Broca’s area
(Fig. 4AB), the UF, AF and frontal part of the IFOF. Preoperative
4.3.2. Case 2
functional MRI using a verb generation task and finger tapping task
A tumoural mass lesion in the left temporal lobe extending to
showed bilateral activation. Preoperative language testing revealed
the insula and opercular regions was identified in a 64-year old
impaired results for phonological (Raw Score (RS) = 22; <Percentile
right-handed (EHI + 80) woman with an educational level of
(Pc) 2) and syntactic fluency (RS = 9; <Pc 2) and verb generation
19 years (Fig. 5AB). She was admitted to the hospital because of
(RS = 40/50; <Pc 2). For intraoperative use, the verb generation
a grand mal seizure with right hemiparesis and language problems.
(selection of correct items) and repetition task were selected
because of the involvement of the inferior frontal gyrus and AF.
Semantic association and object naming task were included to
identify the IFOF and UF (Table 5). Cortical stimulation of inferior
and middle frontal gyrus did not induce language deficits. Facial/
hand sensations and motor reactions of face/hand were elicited
respectively after and before the central gyrus. During subcortical
stimulation and tumour resection no language deficits were found.
Postoperative MRI demonstrates a partial resection (Berger,

Fig. 4. Case 1 – preoperative axial T2-weighted MR of the brain disclosing a

tumoural mass lesion in the left prefrontal lobe enchroaching upon the precentral
gyrus of the left hemisphere, close to Broca’s area (A and B). Postoperative axial T2-
Fig. 3. Per subcortical pathway the function(s) and suitable language tasks from the weighted MR brain images demonstrating the resection cavity in the prefrontal lobe
DuLIP are shown in Table 4. (C and D).

Table 4
Selection of DuLIP tasks per subcortical pathway.

Subcortical pathways Function(s) Intraoperative language tasks from DuLIP

subcallosal fascicle Initiation of speech Fluency, sentence completion

inferior fronto-occipital fascicle Semantics, reading, judgment Semantic association, semantic odd word/picture
out, semantic judgment
inferior longitudinal fascicle Reading Reading + sentence completion, naming
Phonology/semantics
superior longitudinal fascicle (arcuate Articulatory processing/ Phonology Verbal diadochokinesis, repetition/ phonological
fascicle) odd word out
uncinate fascicle Famous face naming, semantics Naming, semantic odd picture out

corticospinal tract Motor speech Verbal diadochokinesis, repetition

 E. De Witte et al. / Brain & Language 140 (2015) 35–48 43

Fig. 5. Case 2 – preoperative axial T2-weighted MR of the brain demonstrating a Fig. 6. Case 3 – preoperative axial T2-weighted MR of the brain showing a tumoural
tumoural mass lesion involving the left medial and posterior insular region resection cavity in the medial and anterior prefrontal area of the left hemisphere
extending to the anterior pole of the left temporal lobe, with compression of the left close to the premotor and supplementary motor area (A). Significant regrowth of
lateral ventricle (A and B). Postoperative axial T2-weighted MR brain images the tumour is shown with invasion of the corpus callosum and extension to the
showing partial resection of the tumour in the left anterior pole of the temporal prefrontal right hemisphere (B). Postoperative axial T2-weighted MR brain images
lobe (C and D). demonstrating enlargement of the resection cavity in the left anterior frontal area
with a for/backward diameter of 7 cm and a transversal diameter of 4.7 cm (C and
D).

She regularly complained of word findings problems, memory arrest during object naming). Subcortical stimulation triggered
problems and visual, auditory and tactile hallucinations. DTI phonological paraphasias and word finding problems at the board-
indicated that the CST, the AF and IFOF were located adjacent to ers of the AF (repetition task) and IFOF (semantic picture out task).
the tumour. Preoperative fMRI (verb generation) showed left hemi- Resection was interrupted when the insula was reached. Postoper-
sphere language dominance and a mass lesion encroaching upon ative MRI showed partial resection of the tumour in the left anterior
Broca’s area and Wernicke’s area. pole of the temporal lobe (Fig. 5CD).
Preoperatively, in-depth examination of linguistic functions dis- At 6 weeks postsurgery, in addition to the disorder in semantic
closed defective results for semantic fluency of animals (RS = 12; fluency, which was still present, a number of novel language defi-
<Pc 2). Consequently, an intraoperative test battery was composed. cits were observed including object naming (RS = 61/100; <Pc 2),
The verb generation and action naming task were selected for action naming (RS = 32/60; <Pc 2) and comprehension problems
stimulation of the frontal areas (insular region), the repetition task indicated by the semantic sentence judgment score (RS = 42/50;
for stimulation of the AF. The semantic tasks (semantic picture out, <Pc 2). She also presented with fluency problems (phonological:
semantic sentence judgment, object naming) were selected for RS = 12; <Pc 2; semantic: RS = 9; <Pc 2; syntactic (RS = 12; <Pc 7).
stimulation of the posterior areas (Wernicke’s area) and the IFOF At 3 months postsurgery, the word finding difficulties, expressed
(Table 5). in object and action naming and perception problems disappeared.
During cortical stimulation, essential language areas were Language fluency improved but remained impaired for semantic
identified in the inferior frontal gyrus (speech arrests during action fluency of animals (RS = 11; <Pc 2) as was in the preoperative
naming) and in the anterior middle temporal gyrus (delay, speech phase.

Table 5
Selection of intraoperative DuLIP tasks per case study.

Case Location MRI Location fMRI Location DTI Intraoperative language tests Extent of resectiona
1 L inferior frontal gyrus Bilateral activation UF, IFOF, AF Repetition, verb generation, object naming, semantic association Partial
2 L temporoinsular L dominance CST, IFOF, AF Repetition, verb generation, naming (actions + objects), semantic Partial
picture out, semantic sentence judgment
3 L frontal, SMA L dominance IFOF, CST, SLF Fluency, semantic sentence completion, repetition, verbal Subtotal
diadochokinesis
4 L temporoparietal L dominance CST, AF, IFOF Repetition, semantic picture out, semantic sentence judgment, Subtotal
reading + semantic sentence completion, object naming
5 L parietooccipital R dominance AF, SLF Reading + semantic sentence completion, semantic picture out, Subtotal
naming, repetition, phonological odd word out (calculation,
vision)

Legend: Nr = Number of case, SMA = supplementary motor area, L = left, R = right, UF = Uncinate Fasciculus, AF = Arcuate Fasciculus, SLF = Superior Longitudinal Fasciculus,
CST = corticospinal tract.
a
According to the classification from Berger et al. (1994).
44 E. De Witte et al. / Brain & Language 140 (2015) 35–48

4.3.3. Case 3 Preoperative fMRI (verb generation, finger tapping) in this left-
A 38-year-old right-handed (EHI = +100) man with an educa- handed patient showed left hemisphere language dominance.
tional level of 12 years underwent repeat surgery after regrowth Wernicke’s area was located close to the posterior part of the
of a tumoural lesion in the left frontal lobe close to the supplemen- tumour. Preoperatively, in-depth examination of linguistic
tary motor area (SMA) and the premotor cortex (Fig. 6AB). DTI functions was entirely normal. Semantic tasks (semantic picture
showed involvement of the SLF and IFOF. The CST was adjacent out, semantic sentence judgment), reading task (reading +
to the tumour. fMRI (verb generation, finger tapping) revealed left semantic sentence completion), naming (object) and repetition
hemisphere language dominance with activations in the superior were included in the intraoperative test battery for these eloquent
and middle temporal gyrus and the inferior and middle frontal regions and pathways (Table 5). Cortical stimulation elicited
gyrus. In the preoperative phase, no linguistic problems were speech arrests in the inferior frontal gyrus (during repetition)
identified. The selected tasks from DuLIP for intraoperative map- and anomia and delayed word retrieval in the medial and inferior
ping were fluency tasks and semantic sentence completion (SMA, temporal gyrus (during naming). The preoperative fMRI disclosing
premotor cortex, IFOF) and repetition & verbal diadochokinesis activation in Wernicke’s area did not correlate with the positive
(SLF, CST) (see Table 5). During cortical stimulation of the left pre- intraoperative stimulation points (superior and middle temporal
central and inferior frontal gyrus no linguistic deficits were found. gyrus). During subcortical stimulation, word finding problems
Subcortical stimulation in the vicinity of the CST resulted during (during semantic picture out task) and delayed word retrieval
repetition tasks in speech arrests and motor reactions of the right (during semantic sentence completion) were found at the boarders
hand. After stimulation and resection near the SMA initiation of the IFOF and phonological parhaphasia at the boarders of the AF
difficulties appeared during the sentence completion task. On the (during repetition). Resection was stopped according to these
postoperative MRI a subtotal enlarged resection was found in the functional boundaries, consequently subtotal resection was found
left anterior frontal area (Fig. 6CD). In the acute phase after surgery on postoperative MRI in the left temporoparietal area (Fig. 7CD).
a dynamic aphasia (characterised by a selective disorder in the In the first week after surgery, the patient could speak fluently
spontaneous speech) typical for an SMA syndrome was found. but in a clinical examination some naming and comprehension
problems at a linguistic complex level were noticed. At 6 weeks
Extensive testing at 6 weeks revealed marked phonological
postsurgery, the following DuLIP tasks were severely impaired:
(RS = 15; <Pc 2), semantic (animals: RS = 13; <Pc 2), and syntactic
object naming (RS = 78/100; <Pc 2), action naming (RS = 48/60;
fluency (RS = 4; <Pc 2) problems. At six months postsurgery,
<Pc 2), semantic fluency (animals) (RS = 14; <Pc 2), syntactic
reassessment of language functions disclosed normal results.
fluency (RS = 18; <Pc 2), semantic odd word out (RS = 15; <Pc 2),
semantic picture out (RS = 23/25; <Pc 2) and semantic sentence
4.3.4. Case 4 judgment (RS = 46/50; <Pc 2). After six months the naming and
A 34-year-old left-handed (EHI 66, 67) man with an educa- semantic deficits resolved and only mild fluency problems
tional level of 16 years was transferred to the hospital following persisted in speech (syntactic fluency was still disturbed:
a grand mal seizure. MRI of the brain revealed a tumoural mass RS = 18; <Pc 2).
lesion in the left temporoparietal lobe (Fig. 7AB). DTI indicated that
the AF and CST were relocated upwards due to mass effects 4.3.5. Case 5
(AF to cranial, CST to anterior) and that the IFOF was involved. A 53-year-old right-handed (EHI = +80) man with an educa-
tional level of 10 years complained of visual problems (troubled
sight) and progressive memory problems. MRI of the brain
revealed a left posterior parietal glioma (Fig. 8AB). DTI showed that
the tumour was adjacent to the AF (that connects Broca’s and
Wernicke’s area) and involved the anterior and posterior indirect
segment of the SLF (Catani & Mesulam, 2008). Preoperative fMRI
(verb generation, finger tapping) showed reversed, right hemi-
sphere language dominance. Concise preoperative language testing
yielded normal results except for slight semantic fluency deficits
(animals: RS: 16; <Pc 7). Phonological tasks (phonological odd
word out, repetition) were used during surgery when the AF was
approached, whereas semantic and reading tasks (semantic picture
out, semantic sentence completion, naming) were selected for DES
in the parietal regions. Non-language tasks such as calculation
(self-made) and a line bisection task were added (Roux et al.,
2011). Speech arrest and dysarthria was triggered after stimulation
in the area located between the left angular gyrus and supramar-
ginal gyrus (semantic picture out). Anomia was found following
stimulation of the posterior middle temporal gyrus. Delayed word
retrieval was found after stimulation of the junction between the
postcentral gyrus and supramarginal gyrus. Paresthesias of the
right hand were elicited by stimulation near the postcentral gyrus.
During subcortical stimulation to inferior reading was slightly
delayed (semantic sentence completion) and calculation problems
for complex tasks occurred during stimulation of the deeper pari-
etal regions. Consequently, the resection was interrupted and a
Fig. 7. Case 4 – preoperative axial T2-weighted MR of the brain demonstrating a subtotal resection was confirmed in the posterior parietal lobe on
tumoural mass lesion in the left lateral ventricle extending to the temporoparie-
tooccipital region (A). A slight midline shift to the right and dilatation of the left
the postoperative MRI (Fig. 8CD).
lateral ventricle are observed (B). Postoperative axial T2-weighted MR brain images In the first days after surgery the patient presented with
showing the resection cavity in the left temporoparietal area (4–2 cm) (C and D). delayed reading and calculation problems. Language assessments
 E. De Witte et al. / Brain & Language 140 (2015) 35–48 45

judgment task (p = .029). A significant difference, however, was

obtained when using a p-value of 0.05. Therefore, cut-off scores
were calculated in all four categories (age/education groups) for
every DuLIP task (except for verbal diadochokinesis task which
did not show any age or education effect).
Qualitative analysis of the data revealed that the few errors
made by the control group could not be systematically related to
specific items. Therefore, no items were deleted after standardisa-
tion. Only for the semantic association task the percentage of cor-
rect answers was below 90% (83.44%) and the most frequent
mistakes were delayed responses. This task seems to be the most
difficult one since intact reading, semantic knowledge and lan-
guage initiation are necessary to produce the target answer within
4 s. Moreover, as is the case for all DuLIP tasks this test also relies
on memory, attention and executive function.

5.2. Clinical application of DuLIP

The normative data allow to analyse and compare the pre- and
postoperative results of the patient group with the norm group.
Intraoperative selection of DuLIP tasks was based on the preoper-
ative patient level and recent insights in the functional neuroanat-
omy subserving linguistic processes in brain tumour populations.
Although neural language representation is a complex phenome-
Fig. 8. Case 5 – preoperative axial T2-weighted MR brain of the brain disclosing a non and subject to brain plasticity mechanisms, some guidelines
tumoural mass lesion in the left posterior parietal lobe (A and B). Postoperative might be proposed regarding structure-to-function relations at
axial T2-weighted MR brain images reveals subtotal tumour resection in the left
parietooccipital area (C and D).
the cortical and subcortical level. For cortical tumours located in
dominant frontal brain regions, phonology, syntax and language
initiation should be assessed. For tumours located in dominant
at 6 weeks revealed defective scores for semantic fluency (RS = 16; temporal regions, semantic tasks that measure both comprehen-
<Pc 2) and reading & semantic sentence completion (RS = 18; <Pc sion and production seem indispensable (e.g. semantic picture
2). At six months postsurgery, remission of fluency, reading and out task). For tumours situated in the dominant parietal regions,
calculation problems was observed. semantic tasks and reading tests add valuable information (see
Fig. 2, Table 3). Since the plastic potential of the white matter is
limited (Duffau, 2013) and subcortical injury is an independent
5. Discussion predictor of postoperative neurological deficits (Trinh et al.,
2013), subcortical mapping is even more important. As the IFOF,
This study reports the rationale, development, standardisation ILF, SLF, CST and subcallosal fascicle are essential language path-
and application of DuLIP, the first Dutch linguistic test battery for ways involved in respectively semantics, reading, phonology,
pre-, intra- and postoperative use in awake brain surgery. DuLIP motor speech, language initiation, an appropriate set of DuLIP tests
includes tasks that relate to a broad variety of linguistic levels should be selected (see Fig. 3, Table 4). We have restricted to the
and modalities and as such allows assessment of various linguistic (left) dominant language hemisphere, however, the right hemi-
functions in detail. The normative data and clinical application of sphere is also involved in language functions such as prosody
DuLIP as well as some limitations and future needs are discussed and processing unfamiliar sentences that contain distant semantic
in the sections below. relationships (Polczynska, 2009). Polczynska (2009) therefore sug-
gested two types of tests for the non-dominant right hemisphere
5.1. Normative data of DuLIP (1) tests investigating alternative meanings of words and broader
semantic relationships, (2) affective prosody tests.
Quantitative analysis of data showed a ceiling effect for all but As demonstrated in the case descriptions, DuLIP may be a useful
one of the linguistic tests (fluency tasks). The target answers in the and reliable tool in the pre-, intra- and postoperative phase of
fluency tasks were numerous and there was no maximum score. In awake surgery. Preoperatively, no linguistic deficits were found
contrast to the other linguistic tasks more variance between sub- in cases 3 and 4 which may be due to typically slow growth of
jects was found. Since the primary objective of DuLIP is the selec- low-grade gliomas allowing neural reorganisation (preoperative
tion of a set of relevant tasks and test items in the preoperative plasticity). Cases 1, 2 and 5 presented verbal fluency problems
phase, DuLIP was constructed in such a way that patients with which is in agreement with a large outcome study of Satoer et al.
mild to moderate aphasia could perform to some degree all tasks (2014). Intraoperative corticosubcortical stimulation induced lan-
within 4 s. Consequently, normal participants obtained high scores guage deficits in all but one case (case 1). In three out of the four
(De Witte & Mariën, 2013). As expected, task performance differed cases (case 1, 2, 4) intraoperative findings did not correlate with
significantly according to age and education. The higher the age the preoperative fMRI results. As a consequence, the surgical strategy
lower the test score and the lower the education level the lower had to be adapted. The absence of a strong correlation between
the scores. In various studies similar findings have been found fMRI and DES findings in our cases highlights the importance of
(Brucki, 2004; Mariën et al., 1998; Mathuranath, 2003; Snitz, intraoperative stimulation and testing. Postoperatively, the extent
2009; Zec, Burkett, Markwell, & Larsen, 2007). According to the of the resection was calculated and subtotal and partial resections
Sidak correction, education did not significantly effect performance were performed. The cases in which essential language sites were
on the semantic sentence completion task (closed context) identified intraoperatively (cases 2–5) developed transient
(p = .001) and age did not have a significant effect on the phonemic language deficits at 6 weeks postsurgery. In agreement with the
46 E. De Witte et al. / Brain & Language 140 (2015) 35–48

observations of Duffau (2007) and Trinh et al. (2013) the postoper- To identify the exact benefits of the DuLIP protocol, the out-
ative linguistic impairments in these patients resolved after come of a group of patients in which a set of DuLIP tests is used
6 months. Naming and verbal fluency were most frequently versus a matched control group of patients in which only naming
impaired and seem to be sensitive parameters to identify (Satoer is assessed should be studied. Retrospective studies or compari-
et al., 2012, 2014). To conclude, the selected DuLIP tasks were sons of methods between neurosurgical departments should be
feasible in the pre-, intra- and postoperative phase and elicited conducted to further investigate this issue. Although the normative
language deficits during stimulation. data of the DuLIP set can only be used in the Dutch-speaking part
of Belgium and the Netherlands, the tasks, however, may be
adapted to different languages.
5.3. Limitations of the study Apart from language and sensorimotor functions, higher-order
cognitive functions (e.g. memory, attention/executive functioning,
Some tasks should be refined in order to better assess the spatial awareness, calculation) and visual functions should also be
function targeted. The phonological odd word out task might be (briefly) tested in awake procedures (see case 5) in order to ward
too easy since some answers can be visually identified without the postoperative outcome and Quality of Life of the patient
using phonological knowledge (e.g. bal, hal, val, weg). A solution (Coello et al., 2013; Duffau, 2010).
might be inclusion of items with no transparent phonology/
orthography correspondence, such as ‘jack’, in Dutch rhyming on
‘bek’ and ‘vrek’, but not on ‘dak’ (e.g. bek, jack, vrek, dak) or to pres- 6. Conclusion
ent the items auditorily. The syntactic fluency and verb generation
tasks are not pure syntactic tasks but also measure lexical seman- In this study a set of linguistic tasks (DuLIP) to assess a variety
tics. However, they rely more strongly on syntactic processing than of linguistic functions during awake surgery was carefully
semantic fluency as verb lemmas contain more information, may constructed and standardised in a cohort of 250 Dutch healthy
be harder to retrieve, and require more grammatical encoding than controls. With DuLIP, a valid linguistic test battery and methodo-
noun lemmas (Bastiaanse & van Zonneveld, 2004). The action nam- logically sound instrument has become available to reliably evalu-
ing task is a more efficient task to assess grammatical processing ate linguistic functions in the pre-, intra- and postoperative phase
since the verb needs to be inflected in a short sentence. The seman- of awake surgery. Based on the tumour localisation (preoperative
tic association task seems to be a difficult task to perform within neuroimaging data) and the patient’s personal characteristics
4 s even in the normal population and under normal conditions. (preoperative language level, job), DuLIP allows a tailored selection
Therefore, this task appears to be more suitable during tumour per patient. This valid linguistic approach enhances the reliability
resection (and not during stimulation) at the subcortical level to of the neurosurgical procedure and patient outcome.
control the semantic level.
Since only 5 case studies were included in this study no sound Acknowledgments
conclusions can be drawn regarding the most sensitive tasks for
pre-, intra- and postoperative testing. As the total DuLIP protocol Elke De Witte is a Ph.D. fellow of the Research Foundation –
takes 1h30 to perform, the total of pre- and postoperative cognitive Flanders (FWO). Steven Verheyen is a postdoctoral fellow of the
assessments require two test moments. In comparison to naming, Research Foundation – Flanders (FWO).
application of a selection of DuLIP tasks does not substantially pro- We want to thank Prof. Dr. Roelien Bastiaanse and Drs. Adria
long the awake procedure. The intraoperative cortical procedure is Rofes for their cooperation in the development and analysis of
only increased with a few minutes since in general only one addi- the action naming task.
tional task is required to identify positive language sites. During
subcortical stimulation the selected tasks are presented in alterna-
Appendix A. Supplementary material
tion which does not lead to a substantial prolongation of the surgi-
cal time.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bandl.2014.
5.4. Future directions 10.011.

In order to improve current knowledge on anatomoclinical

References
correlations and sensitivity of the DuLIP tasks (inter-location-
sensitivity and pre- and postoperative sensitivity), a longitudinal Ardila, A. (2012). Interaction between lexical and grammatical language systems in
neurolinguistic follow-up study in a large patient group is needed. the brain. Physics of Life Reviews, 9(2), 198–214. http://dx.doi.org/10.1016/
Results from these studies will shed light on the most sensitive j.plrev.2012.05.001.
Baayen, R. H., Piepenbrock, R., & van Rijn, H. (1993). The CELEX lexical database (CD-
DuLIP tasks and only these tasks will be further administered. ROM). Philadelphia, PA: Linguistic Data Consortium, University of Pennsylvania.
Future studies using the DuLIP protocol during pre- and postopera- Baldo, J. V., Schwartz, S., Wilkins, D., & Dronkers, N. F. (2006). Role of frontal versus
tive fMRI assessments may be relevant for both clinical (e.g. to plan temporal cortex in verbal fluency as revealed by voxel-based lesion symptom
mapping. Journal of the International Neuropsychological Society: JINS, 12(6),
or to exclude the possibility of awake surgery) and scientific 896–900. http://dx.doi.org/10.1017/S1355617706061078.
purposes (e.g. to study linguistic processes, anatomoclinical Bastiaanse, R., Bosje, M., & Visch-Brink, E. G. (1995). PALPA: De Nederlandse versie
correlations, brain plasticity mechanisms). Since verb generation (PALPA: The Dutch version). Hove: Lawrence Erlbaum.
Bastiaanse, R. (2010). ATP: Auditief Taalbegripsprogramma complete set. Houten, The
and naming tasks are often used as fMRI language paradigms Netherlands: Bohn Stafleu van Logum.
(Ojemann, Ojemann, & Lettich, 2002; Smits, Visch-Brink, Bastiaanse, R., Maas, E., & Rispens, J. (2000). Werkwoorden-en Zinnentest. WEZT.
Schraa-Tam, Koudstaal, & van der Lugt, 2006), the DuLIP protocol Lisse, The Netherlands: Swets Test Publishers.
Bastiaanse, R., & van Zonneveld, R. (2004). Broca’s aphasia, verbs and the mental
might be used in future research to define the most sensitive tasks
lexicon. Brain and Language, 90(1–3), 198–202. http://dx.doi.org/10.1016/
for preoperative fMRI. However, it should be noted that the correla- S0093-934X(03)00432-2.
tion between preoperative fMRI and intraoperative DES is poor for Bello, L., Gallucci, M., Fava, M., Carrabba, G., Giussani, C., Acerbi, F., et al. (2007).
language (66%). As a result preoperative fMRI cannot be reliably Intraoperative subcortical language tract mapping guides surgical removal of
gliomas involving speech areas. Neurosurgery, 60(1), 67–82.
used without the combination with DES to make critical surgical Bello, L., Gambini, A., Castellano, A., Carrabba, G., Acerbi, F., Fava, E., et al. (2008).
decisions (Roux et al., 2003). Motor and language DTI Fiber Tracking combined with intraoperative
 E. De Witte et al. / Brain & Language 140 (2015) 35–48 47

subcortical mapping for surgical removal of gliomas. NeuroImage, 39(1), Hamberger, M. J., Seidel, W. T., Mckhann, G. M., Perrine, K., & Goodman, R. R. (2005).
369–382. Brain stimulation reveals critical auditory naming cortex. Brain, 128(11),
Berger, M. S., Deliganis, A. V., Dobbins, J., & Keles, G. E. (1994). The effect of extent of 2742–2749.
resection on recurrence in patients with low grade cerebral hemisphere Hickok, G. (2012). The cortical organization of speech processing: Feedback control
gliomas. Cancer, 74(6), 1784–1791. and predictive coding the context of a dual-stream model. Journal of
Bertani, G., Fava, E., Casaceli, G., Carrabba, G., Casarotti, A., Papagno, C., et al. (2009). Communication Disorders, 45(6), 393–402. http://dx.doi.org/10.1016/j.jcomdis.
Intraoperative mapping and monitoring of brain functions for the resection of 2012.06.004.
low-grade gliomas: Technical considerations. Neurosurgical Focus, 27(4). Ilmberger, J., Ruge, M., Kreth, F.-W., Briegel, J., Reulen, H.-J., & Tonn, J.-C. (2008).
Bizzi, A., Blasi, V., Falini, A., Ferroli, P., Cadioli, M., Danesi, U., et al. (2008). Intraoperative mapping of language functions: A longitudinal neurolinguistic
Presurgical functional MR imaging of language and motor functions: Validation analysis. Journal of Neurosurgery, 109(4), 583–592.
with intraoperative electrocortical mapping. Radiology, 248(2), 579–589. http:// Kaplan, E. F., Goodglass, H., & Weintraub, S. (1983). The Boston naming test (2nd ed.).
dx.doi.org/10.1148/radiol.2482071214. Philadelphia: Lea and Fibiger.
Brucki, S. M. D. (2004). Category fluency test: Effects of age, gender and education Kemmerer, D., & Tranel, D. (2000). Verb retrieval in brain-damaged subjects: 1.
on total scores, clustering and switching in Brazilian Portuguese-speaking Analysis of stimulus, lexical, and conceptual factors. Brain and Language, 73(3),
subjects. Brazilian Journal of Medical and Biological Research, 37(12), 1771. 347–392. http://dx.doi.org/10.1006/brln.2000.2311.
Catani, M., & Mesulam, M. (2008). The arcuate fasciculus and the disconnection Keuleers, E., Brysbaert, M., & New, B. (2010). SUBTLEX-NL: A new measure for Dutch
theme in language and aphasia: History and current state. Cortex; A Journal word frequency based on film subtitles. Behavior Research Methods, 42(3),
Devoted to the Study of the Nervous System and Behavior, 44(8), 953–961. http:// 643–650. http://dx.doi.org/10.3758/BRM.42.3.643.
dx.doi.org/10.1016/j.cortex.2008.04.002. Lubrano, V., Filleron, T., Demonet, J.-F., & Roux, F.-E. (2014). Anatomical correlates
Catani, M., & Thiebaut de Schotten, M. (2008). A diffusion tensor imaging for category-specific naming of objects and actions: A brain stimulation
tractography atlas for virtual in vivo dissections. Cortex; A Journal Devoted to mapping study. Human Brain Mapping, 35(2), 429–443. http://dx.doi.org/
the Study of the Nervous System and Behavior, 44(8), 1105–1132. http:// 10.1002/hbm.22189.
dx.doi.org/10.1016/j.cortex.2008.05.004. Luteijn, F., & Barelds, D. (2004). Groninger Intelligentie Test II (GIT II). Amsterdam, The
Coello, A. F., Moritz-Gasser, S., Martino, J., Martinoni, M., Matsuda, R., & Duffau, H. Nederlands: Pearson.
(2013). Selection of intraoperative tasks for awake mapping based on Maldonado, I. L., Moritz-Gasser, S., & Duffau, H. (2011). Does the left superior
relationships between tumor location and functional networks. Journal of longitudinal fascicle subserve language semantics? A brain electrostimulation
Neurosurgery, 119(6), 1380–1394. http://dx.doi.org/10.3171/2013.6.JNS122470. study. Brain Structure and Function, 216(3), 263–274. http://dx.doi.org/10.1007/
Crum, R. M., Anthony, J. C., Bassett, S. S., & Folstein, M. F. (1993). Population-based s00429-011-0309-x.
norms for the Mini-Mental State Examination by age and educational level. Mariën, P., Mampaey, E., Vervaet, A., Saerens, J., & De Deyn, P. P. (1998). Normative
JAMA: The Journal of the American Medical Association, 269(18), 2386–2391. data for the Boston Naming Test in native Dutch-speaking Belgian elderly. Brain
De Jong, E. D. (Ed.). (1979). Spreektaal: Woordfrequenties in gesproken Nederlands. and Language, 65(3), 447–467. http://dx.doi.org/10.1006/brln.1998.2000.
Utrecht, The Netherlands: Boh, Scheltema & Holkema. Martin, A., & Chao, L. L. (2001). Semantic memory and the brain: Structure and
De Witt Hamer, P. C., Robles, S. G., Zwinderman, A. H., Duffau, H., & Berger, M. S. processes. Current Opinion in Neurobiology, 11(2), 194–201. http://dx.doi.org/
(2012). Impact of intraoperative stimulation brain mapping on glioma surgery 10.1016/S0959-4388(00)00196-3.
outcome: A meta-analysis. Journal of Clinical Oncology: Official Journal of the Mathuranath, P. S. (2003). Effects of age, education and gender on verbal fluency.
American Society of Clinical Oncology, 30(20), 2559–2565. http://dx.doi.org/ Journal of Clinical and Experimental Neuropsychology, 25(8), 1057.
10.1200/JCO.2011.38.4818. Metz-Lutz, M. N., Kremin, H., Deloche, G., Hannequin, D., Ferrand, L., & Perrier, D.
De Witte, E., & Mariën, P. (2013). The neurolinguistic approach to awake surgery (1991). Standardisation d’un test de dénomination orale: Contrôle des effets de
reviewed. Clinical Neurology and Neurosurgery, 115(2), 127–145. http:// l’âge, du sexe et du niveau de scolarité chez les sujets adultes normaux. Rev de
dx.doi.org/10.1016/j.clineuro.2012.09.015. Neuropsychol, 1, 73–95.
Duffau, H. (2007). Contribution of cortical and subcortical electrostimulation in Miatton, M., Wolters, M., Lannoo, E., & Vingerhoets, G. (2004). Updated and
brain glioma surgery: Methodological and functional considerations. extended Flemish normative data of commonly used neuropsychological tests.
Neurophysiologie Clinique/Clinical Neurophysiology, 37, 373–382. Psychologica Belgica, 44(3), 189–216.
Duffau, H. (2008). The anatomo-functional connectivity of language revisited new Miceli, G., Capasso, R., Monti, A., Santini, B., & Talacchi, A. (2012). Language testing
insights provided by electrostimulation and tractography. Neuropsychologia, in brain tumor patients. Journal of Neuro-Oncology, 108(2), 247–252. http://
46(4), 927–934. dx.doi.org/10.1007/s11060-012-0810-y.
Duffau, H. (2009). Surgery of low-grade gliomas: Towards a ‘‘functional Moritz-Gasser, S., & Duffau, H. (2013). The anatomo-functional connectivity of word
neurooncology’’. Current Opinion in Oncology, 21(6), 543–549. repetition: Insights provided by awake brain tumor surgery. Frontiers in Human
Duffau, H. (2010). Awake surgery for nonlanguage mapping. Neurosurgery, 66(3), Neuroscience, 7. http://dx.doi.org/10.3389/fnhum.2013.00405.
523–528. Moritz-Gasser, S., Herbet, G., & Duffau, H. (2013). Mapping the connectivity
Duffau, H. (2013). The huge plastic potential of adult brain and the role of underlying multimodal (verbal and non-verbal) semantic processing: A brain
connectomics: New insights provided by serial mappings in glioma surgery. electrostimulation study. Neuropsychologia, 51(10), 1814–1822. http://
Cortex; A Journal Devoted to the Study of the Nervous System and Behavior. http:// dx.doi.org/10.1016/j.neuropsychologia.2013.06.007.
dx.doi.org/10.1016/j.cortex.2013.08.005. Nespoulous, J.-L., & Moureau, N. (1998). ‘‘Repair strategies’’ and the production of
Duffau, H., Capelle, L., Denvil, D., Sichez, N., Gatignol, P., Taillandier, L., et al. (2003). segmental errors in aphasia: Epentheses vs. syncopes in consonantal clusters. In
Usefulness of intraoperative electrical subcortical mapping during surgery for E. Visch-Brink & R. Bastiaanse (Eds.), Linguistic levels in aphasiology
low-grade gliomas located within eloquent brain regions: Functional results in (pp. 133–145). San Diego-London: Singular Publishing Group.
a consecutive series of 103 patients. Journal of Neurosurgery, 98(4), 764–778. Ojemann, J. G., Ojemann, G. A., & Lettich, E. (2002). Cortical stimulation mapping of
Duffau, H., Gatignol, P., Mandonnet, E., Peruzzi, P., Tzourio-Mazoyer, N., & Capelle, L. language cortex by using a verb generation task: Effects of learning and
(2005). New insights into the anatomo-functional connectivity of the semantic comparison to mapping based on object naming. Journal of Neurosurgery, 97(1),
system: A study using cortico-subcortical electrostimulations. Brain, 128, 33–38. http://dx.doi.org/10.3171/jns.2002.97.1.0033.
797–810. Palmer, B. W., Boone, K. B., Lesser, I. M., & Wohl, M. A. (1998). Base rates of
Duffau, H., Lopes, M., Arthuis, F., Bitar, A., Sichez, J., Van Effenterre, R., et al. (2005). ‘‘impaired’’ neuropsychological test performance among healthy older adults.
Contribution of intraoperative electrical stimulations in surgery of low grade Archives of Clinical Neuropsychology, 13(6), 503–511. http://dx.doi.org/10.1016/
gliomas: A comparative study between two series without (1985–96) and with S0887-6177(97)00037-1.
(1996–2003) functional mapping in the same institution. Journal of Neurology Papagno, C., Casarotti, A., Comi, A., Gallucci, M., Riva, M., & Bello, L. (2012).
Neurosurgery and Psychiatry, 76(6), 845–851. Measuring clinical outcomes in neuro-oncology. A battery to evaluate low-
Duffau, H., Moritz-Gasser, S., & Gatignol, P. (2009). Functional outcome after grade gliomas (LGG). Journal of Neuro-Oncology, 108(2), 269–275. http://
language mapping for insular World Health Organization Grade II gliomas in dx.doi.org/10.1007/s11060-012-0824-5.
the dominant hemisphere: Experience with 24 patients. Neurosurgical Focus, Pereira, L. C. M., Oliveira, K. M. L., Abbate, G. L., Sugai, R., Ferreira, J. A., & Motta, L. A. (
27(2). (2009). Outcome of fully awake craniotomy for lesions near the eloquent
Engström, M., Ragnehed, M., Lundberg, P., & Söderfeldt, B. (2004). Paradigm design cortex: Analysis of a prospective surgical series of 79 supratentorial
of sensory-motor and language tests in clinical fMRI. Neurophysiologie Clinique = primary brain tumors with long follow-up. Acta Neurochirurgica, 151(10),
Clinical Neurophysiology, 34(6), 267–277. http://dx.doi.org/10.1016/ 1215–1230.
j.neucli.2004.09.006. Polczynska, M. (2009). New tests for language mapping with intraoperative
Folstein, M., Folstein, S., & McHugh, P. (1975). Mini-Mental State: A practical electrical stimulation of the brain to preserve language in individuals with
method for grading the cognitive state of patients for the clinician. Journal of tumors and epilepsy: A preliminary follow-up study. Poznan Studies in
Psychiatric Research, 12, 189–198. Contemporary Linguistics, 45(2), 261–279.
Fontaine, D., Capelle, L., & Duffau, H. (2002). Somatotopy of the supplementary Richardson, F. M., Seghier, M. L., Leff, A. P., Thomas, M. S. C., & Price, C. J. (2011).
motor area: Evidence from correlation of the extent of surgical resection with Multiple routes from occipital to temporal cortices during reading. The Journal
the clinical patterns of deficit. Neurosurgery, 50(2), 297–303. of Neuroscience: The Official Journal of the Society for Neuroscience, 31(22),
Gras-Combe, G., Moritz-Gasser, S., Herbet, G., & Duffau, H. (2012). Intraoperative 8239–8247. http://dx.doi.org/10.1523/JNEUROSCI.6519-10.2011.
subcortical electrical mapping of optic radiations in awake surgery for glioma Rofes, A. (2012). The verb in sentence context test: Standardization and application in
involving visual pathways. Journal of Neurosurgery, 117(3), 466–473. http:// awake neurosurgery. Unpublished master dissertation. University of Groningen,
dx.doi.org/10.3171/2012.6.JNS111981. Groningen.
48 E. De Witte et al. / Brain & Language 140 (2015) 35–48

Rönnlund, M., Nyberg, L., Bäckman, L., & Nilsson, L.-G. (2005). Stability, growth, and easy-to-implement paradigms for patient care and clinical research.
decline in adult life span development of declarative memory: Cross-sectional Radiographics: A Review Publication of the Radiological Society of North America
and longitudinal data from a population-based study. Psychology and Aging, Inc, 26(Suppl. 1), S145–S158. http://dx.doi.org/10.1148/rg.26si065507.
20(1), 3–18. http://dx.doi.org/10.1037/0882-7974.20.1.3. Snitz, B. E. (2009). Effects of age, gender, education and race on two tests of language
Roux, F.-E., Boulanouar, K., Lotterie, J.-A., Mejdoubi, M., LeSage, J. P., & Berry, I. ability in community-based older adults. International Psychogeriatrics, 21(6), 1051.
(2003). Language functional magnetic resonance imaging in preoperative Snodgrass, J. G., & Vanderwart, M. (1980). A standardized set of 260 pictures: Norms
assessment of language areas: Correlation with direct cortical stimulation. for name agreement, image agreement, familiarity, and visual complexity.
Neurosurgery, 52(6), 1335–1347. Journal of Experimental Psychology: Human Learning and Memory, 6(2), 174–215.
Roux, F.-E., Dufor, O., Lauwers-Cances, V., Boukhatem, L., Brauge, D., Draper, L., et al. Spena, G., Nava, A., Cassini, F., Pepoli, A., Bruno, M., D’Agata, F., et al. (2010).
(2011). Electrostimulation mapping of spatial neglect. Neurosurgery, 69(6), Preoperative and intraoperative brain mapping for the resection of eloquent-
1218–1231. area tumors. A prospective analysis of methodology, correlation, and usefulness
Roux, F., Lubrano, V., Lauwers-Cances, V., Tremoulet, M., Mascott, C., & Demonet, J. based on clinical outcomes. Acta Neurochirurgica, 152(11), 1835–1846.
(2004). Intra-operative mapping of cortical areas involved in reading in mono- Talacchi, A., d’Avella, D., Denaro, L., Santini, B., Meneghelli, P., et al. (2012). Cognitive
and bilingual patients RID B-3227-2008. Brain, 127, 1796–1810. outcome as part and parcel of clinical outcome in brain tumor surgery. Journal
Sanai, N., & Berger, M. S. (2010). Intraoperative stimulation techniques for of Neuro-Oncology, 108(2), 327–332. http://dx.doi.org/10.1007/s11060-012-
functional pathway preservation and glioma resection. Neurosurgical Focus, 0818-3.
28(2), E1. Talacchi, A., Santini, B., Savazzi, S., & Gerosa, M. (2011). Cognitive effects of tumour
Sanai, N., Mirzadeh, Z., & Berger, M. S. (2008). Functional outcome after language and surgical treatment in glioma patients. Journal of Neuro-Oncology, 103(3),
mapping for glioma resection. New England Journal of Medicine, 358(1), 18–27. 541–549.
Santini, B., Talacchi, A., Squintani, G., Casagrande, F., Capasso, R., & Miceli, G. (2012). Teixidor, P., Gatignol, P., Leroy, M., Masuet-Aumatell, C., Capelle, L., & Duffau, H.
Cognitive outcome after awake surgery for tumors in language areas. Journal of (2007). Assessment of verbal working memory before and after surgery for low-
Neuro-Oncology. <http://www.ncbi.nlm.nih.gov/pubmed/22350433>. grade glioma. Journal of Neuro-Oncology, 81(3), 305–313.
Satoer, D., Kloet, A., Vincent, A., Dirven, C., & Visch-Brink, E. (2013). Dynamic Trinh, V. T., Fahim, D. K., Shah, K., Tummala, S., McCutcheon, I. E., Sawaya, R., et al.
aphasia following low-grade glioma surgery near the supplementary motor (2013). Subcortical injury is an independent predictor of worsening
area: A selective spontaneous speech deficit. Neurocase. http://dx.doi.org/ neurological deficits following awake craniotomy procedures. Neurosurgery,
10.1080/13554794.2013.841954. 72(2), 160–169. http://dx.doi.org/10.1227/NEU.0b013e31827b9a11.
Satoer, D., Visch-Brink, E., Smits, M., Kloet, A., Looman, C., Dirven, C., et al. (2014). Van Eycken & De Wever (2006). Cancer incidence and survival in flanders, 2000–2001.
Long-term evaluation of cognition after glioma surgery in eloquent areas. Brussels: Flemish Cancer Registry Network (VLK).
Journal of Neuro-Oncology, 116(1), 153–160. http://dx.doi.org/10.1007/s11060- Van Geemen, K., Herbet, G., Moritz-Gasser, S., & Duffau, H. (2013). Limited plastic
013-1275-3. potential of the left ventral premotor cortex in speech articulation: Evidence
Satoer, D., Vork, J., Visch-Brink, E., Smits, M., Dirven, C., & Vincent, A. (2012). From intraoperative awake mapping in glioma patients. Human Brain Mapping.
Cognitive functioning early after surgery of gliomas in eloquent areas. Journal of http://dx.doi.org/10.1002/hbm.22275.
Neurosurgery, 117(5), 831–838. http://dx.doi.org/10.3171/2012.7.JNS12263. Van Loon-Vervoorn, W. A. (1989). Eigenschappen van basiswoorden. Lisse, The
Schmand, B., Groenink, S. C., & van den Dungen, M. (2008). Letterfluency: Netherlands: Swets and Zeitlinger.
Psychometrische eigenschappen en Nederlandse normen. Tijdschrift voor Visch-Brink, E., & Bajema, I. (2001). BOX. Lexicaal semantische afasietherapie.
Gerontologie en Geriatrie, 39(2), 64–74. http://dx.doi.org/10.1007/BF03078128. Amsterdam, The Netherlands: Pearson.
Shattuck-Hufnagel, S., & Klatt, D. H. (1980). How single phoneme error data rule out Warrington, E. K. (1997). The graded naming test: A restandardisation.
two models of error generation. In V. A. Fromkin (Ed.), Errors in linguistic Neuropsychological Rehabilitation, 7(2), 143–146.
performance: Slips of the tongue, ear, pen, and hand (pp. 35–46). New York- Weijters, T. (1983). Positiewoordenboek van de Nederlandse Taal. Lisse, The
London-Toronto-Sydney-San Francisco: Academic Press. Netherlands: Swets and Zeitlinger.
Sidak, Z. K. (1967). Rectangular confidence regions for the means of multivariate Zec, R. F., Burkett, N. R., Markwell, S. J., & Larsen, D. L. (2007). A cross-sectional study
normal distributions. Journal of the American Statistical Association, 62, of the effects of age, education, and gender on the Boston Naming Test. The
626–633. Clinical Neuropsychologist, 21(4), 587–616. http://dx.doi.org/10.1080/1385404
Smits, M., Visch-Brink, E., Schraa-Tam, C. K., Koudstaal, P. J., & van der Lugt, 0701220028.
A. (2006). Functional MR imaging of language processing: An overview of