Thiebautdeschotten 2015
Thiebautdeschotten 2015
Thiebautdeschotten 2015
doi: 10.1093/cercor/bhv173
Original Article
ORIGINAL ARTICLE
Abstract
On the 50th anniversary of Norman Geschwind’s seminal paper entitled ‘Disconnexion syndrome in animal and man’, we pay
tribute to his ideas by applying contemporary tractography methods to understand white matter disconnection in 3 classic cases
that made history in behavioral neurology. We first documented the locus and extent of the brain lesion from the computerized
tomography of Phineas Gage’s skull and the magnetic resonance images of Louis Victor Leborgne’s brain, Broca’s first patient, and
Henry Gustave Molaison. We then applied the reconstructed lesions to an atlas of white matter connections obtained from
diffusion tractography of 129 healthy adults. Our results showed that in all 3 patients, disruption extended to connections
projecting to areas distant from the lesion. We confirmed that the damaged tracts link areas that in contemporary neuroscience
are considered functionally engaged for tasks related to emotion and decision-making (Gage), language production (Leborgne),
and declarative memory (Molaison). Our findings suggest that even historic cases should be reappraised within a disconnection
framework whose principles were plainly established by the associationist schools in the last 2 centuries.
Key words: behavioral neurology, brain lesion, diaschisis, disconnection syndromes, white matter
1
2 | Cerebral Cortex
Localization by area is an oversimplification of the actual became available for clinical anatomical correlation studies
workings of the brain (Damasio and Damasio 1989; Mesulam (Gainotti et al. 1972; Naeser et al. 1982). Patients were studied
1999; Catani, Dell’acqua, Bizzi et al. 2012). The localizationist with extensive neuropsychological batteries of tests to document
bias stems from 2 main limitations. First, the overall idea of their symptoms in detail, coupled with lesion mapping and
equating localization of symptoms with localization of functions group analysis (Damasio and Damasio 1989). Both cortical and
may be incorrect. This point was already raised by several subcortical lesion localization were considered a crucial contri-
authors defending associationist theories, who argued that it is bution to the clinical presentation. Thanks to Geschwind’s vi-
entirely possible that some symptoms can be explained by a sec- sion, the anatomy of white matter connections derived from
ondary effect on other regions distant from the site of the damage 19th century postmortem dissections was revisited in the living
but functionally impaired (Wernicke 1874; Jackson 1881; human brain and disconnection lesions to specific tracts were
Lichtheim 1885; Jackson 1894; Dejerine 1895; Monakow 1914). considered a valid mechanism for newly described syndromes
Second, for a long time we have been unable to map lesions (e.g., tactile agnosia; Geschwind and Kaplan 1962). Geschwind’s
onto discrete circuits due to a lack of methods for visualizing sin- premature death in 1984 robbed him of the opportunity to appre-
gle tracts in the living human brain (Catani and ffytche 2005). ciate the tremendous impact that his ideas, coupled with meth-
Indeed, modern neuroimaging has shown that many complex odological advancements in the field of white matter imaging,
functions rely on the coordinated activity of distant regions con- had on contemporary behavioral neurology. This advance is par-
nected by long-range fibers coursing through the cerebral white ticularly striking when we consider the development of diffusion
matter. Damage either to cortical areas or to underlying connec- MRI, which provides unprecedented access to the anatomy of
tions has far-reaching consequences on distant regions (Baron white matter pathways in the living human brain. One advantage
Figure 1. CAT scan of Gage’s skull registered to MNI152 space. The trajectory of the bar through the skull is indicated in red.
Phineas Gage, Louis Victor Leborgne, and Henry Gustave speak. Leborgne was born in Morêt-sur-Loing and lost his mother
Molaison: Clinical History and Imaging Processing at the age of 3. He moved to Paris with his family when he was 11
Figure 2. T1-weighted MRI images of Leborgne’s brain, registered to MNI152 space. The damage produced by the stroke is indicated in red.
registered with the MNI152 using the affine and elastic deform- In 1993, Corkin and colleagues collected T1-weighted MRI
ation provided in the Statistical Parametric Mapping 8 software images (1 × 1 × 1 mm) of Molaison’s brain using a 1.5 T scanner
package (SPM8; http://www.fil.ion.ucl.ac.uk); a mask of the lesion (General Electric Signa, Milwaukee, WI, USA) (details of the acqui-
was used to exclude the contribution of the damaged voxels to sition in Corkin et al. 1997). We used these images to define the
the registration (Friston et al. 1995; Brett et al. 2001). lesions in Molaison’s brain using ALI (Seghier et al. 2008)
(Fig. 3). The lesion analysis and registration to the MNI152 were
the same as for Leborgne’s data set.
Henry Gustave Molaison (1926–2008)
Molaison had petit mal seizures that began at age 10 and grand
Mapping Disconnections in Phineas Gage, Louis Victor
mal seizures that began at age 15. During adolescence, the epilep-
Leborgne, and Henry Gustave Molaison
tic attacks became more severe and were uncontrolled with
pharmacological treatment (Mauguiere and Corkin 2015). His The next step in the analysis was to map lesions from each pa-
family doctor advised his parents to consult William Beecher tient onto tractography reconstructions of white matter path-
Scoville, a neurosurgeon at the Hartford Hospital in Connecticut, ways obtained from a group of healthy controls. We first
USA. At that time, Scoville was performing psychosurgical proce- obtained diffusion data sets and tract reconstructions, then
dures on patients with psychosis, consisting of the unilateral used complementary approaches to map the disconnections
removal of medial temporal lobe structures. Scoville made the (tract specific vs. data-driven), and later conducted meta-
fortuitous observation that the operation was effective in redu- analyses for each patient to validate the disconnection results
cing seizures in 2 psychotic women who had epilepsy (Scoville with complementary fMRI activation studies published in the
et al. 1953). Molaison’s EEG results did not indicate an epileptic literature.
focus, but showed diffuse bilateral activity, on the basis of
which Scoville decided to perform the experimental procedure Diffusion-Weighted Imaging Acquisition
in both left and right medial temporal lobes. The treatment pal- We recruited 129 healthy, right-handed volunteers (59 male, 70
liated his seizures, but unexpectedly left him with a severe and female) aged 18–79 years and diffusion MRI scans were obtained
lasting anterograde amnesia (Scoville and Milner 1957). This de- from each participant. We acquired 60 contiguous near-axial
clarative memory impairment affected his ability to record new slices on a 3T GE Signa HDx TwinSpeed system (General Electric,
events and facts postoperatively. Molaison was able to maintain Milwaukee, WI, USA) with the following parameters: rostro-
information online for about 30 s, but his ability to convert short- caudal phase encoding, voxel size 2.4 × 2.4 × 2.4 mm, matrix
term memories into long-term memories was lost (Corkin 1984). 128 × 128, slices 60, NEX 1, TE 93.4 ms, b-value 3000 s/mm2, 60
Revisiting Disconnection Syndromes Thiebaut de Schotten et al. | 5
tracts were reconstructed (see Fig. 4 for a visual summary of all Phineas Gage
these connections).
The atlas-based analysis showed that the trajectory of the bar
For each tract, binary visitation maps were created by assigning
impacted several frontal lobe tracts, including primarily the
each voxel a value of 1 or 0, depending on whether the voxel was
uncinate fasciculus (43.1%; z = 2.287; P = 0.022), frontal superior
intersected by the streamlines of the tract. For each participant,
(47.1%; z = 2.552; P = 0.011) and inferior longitudinal (50.5%;
convergence (CS) maps contrasting for white matter (Dell’acqua
z = 2.782; P = 0.005) tracts, frontal aslant tract (42.9%; z = 2.275; P =
et al. 2006) were registered to the MNI152 template provided
0.023), and frontal orbitopolar tract (34.4%; z = 1.702; P = 0.089)
with the FMRIB Software Library package (FSL, http://www.
(Table 1) (Fig. 5). Other partially affected tracts included the
fmrib.ox.ac.uk/fsl/). For the registration, we combined affine with
anterior thalamic projections (23%; z = 0.94; P = 0.347), fronto-
diffeomorphic deformations (Avants et al. 2007; Klein et al. 2009)
striatal projections (22.8%; z = 0.93; P = 0.352), and fronto-
using Advance Normalization Tools (ANTs, http://www.picsl.
pontine projections (22.5%; z = 0.91; P = 0.363). Table 1 contains a
upenn.edu/ANTS/). Binary visitation maps of each dissected
complete list of all analyzed tracts.
tract were normalized to MNI space using both affine and diffeo-
The lesion-based analysis indicated direct damage to the or-
morphic deformations. Normalized binary visitation maps were
bitofrontal cortex, dorsolateral prefrontal cortex, and temporopo-
then averaged to create percentage overlap maps, and we used
lar cortex. In addition, the lesion disconnected several areas not
50% overlap maps for the localization and quantification of the
directly affected by the tamping iron. These areas include the
lesions (Thiebaut de Schotten et al. 2014). We quantified the sever-
frontal pole, posterior inferolateral frontal cortex, anterior and
ity of the disconnection by measuring the proportion of the tract
posterior cingulate, pre-supplementary motor area, precuneus,
disconnected (Thiebaut de Schotten et al. 2008) using Tractotron
posterior temporal, and dorsolateral occipital cortices. Other sub-
Table 1 Percentage of damage to each tract represented and the corresponding z-score for each case
cortices, retrosplenial cortex, orbitofrontal cortex, and gyrus rec- of neurology to better understand the contribution of disconnec-
tus (Fig. 8a). Other subcortical disconnected regions included the tion to their cognitive and behavioral syndromes. Our analysis
mammillary bodies and septal nuclei. showed that in all 3 cases, the damage compromised both
Using the meta-analysis of fMRI studies carried out by Spaniol short- and long-range connections, suggesting that disconnec-
et al. (2009), we found that the areas activated during encoding tion mechanisms occurred beyond the lesion site. Below we dis-
and retrieval of declarative memories overlapped the cortical cuss these findings in greater detail for each patient.
projections of the tracts compromised in Molaison’s brain.
These areas included the posterior parahippocampal and retro- Phineas Gage
splenial cortices, anterior and posterior cingulate gyrus, the dor-
somedial prefrontal cortex, precuneus, orbitofrontal cortex, Previous computerized reconstructions of Gage’s skull indicated
mammillary bodies, and anterior thalamic nuclei (Fig. 8b). damage localized to frontal to cortical regions (orbitofrontal and
dorsolateral prefrontal cortex) (Damasio et al. 1994; Ratiu and
Talos 2004a,b) and underlying white matter (Van Horn et al.
2012). In our study, we identified disconnections in 3 major
Discussion white matter networks: the uncinate fasciculus linking the
Modern approaches to brain function rely on the ability to map orbitofrontal to the anterior temporal lobe; frontal intralobar
the complexity of brain networks underlying cognition and be- networks (connections between frontal regions); and the
havior. In our study, we revisited 3 seminal cases in the history fronto-striatal-thalamo-frontal network.
8 | Cerebral Cortex
The uncinate fasciculus connects the anterior temporal regions et al. 2013). In the original accounts, Harlow did not report
(entorhinal cortex, amygdala, temporopolar cortex) with medial whether Gage showed impairment in naming or semantic
and lateral orbitofrontal cortex (Crosby et al. 1962). Temporal re- knowledge. One may speculate that damage to the uncinate fas-
gions connected by the uncinate fasciculus are involved in epi- ciculus was limited to the medial “limbic” portion of the uncinate
sodic and semantic memory and emotion (Von Der Heide et al. fasciculus, leaving the most lateral projections to Broca’s area
2013; Murray et al. 2014). The orbitofrontal regions are associated intact.
with response inhibition, mood regulation, and reward (Berlin The frontal intralobar networks include 3 sets of connections
et al. 2004; Kramer et al. 2013; Kumfor et al. 2013). Lesions between different regions of the frontal lobe: the fronto-orbitopolar
to the anterior temporal and orbitofrontal regions or their con- tract, frontal aslant tract, and frontal superior and inferior longitu-
nections often cause mood and behavioral symptoms. In trau- dinal tracts (Catani, Dell’acqua, Vergani et al. 2012; Thiebaut de
matic brain injury, for example, patients with lesions to this Schotten et al. 2012).
anterior orbitofrontal-temporal network show socially inappro- The frontal orbitopolar tract represents a transmodal network
priate and disinhibited behavior, impulsivity, compulsive eating, for binding memories and emotions with olfactory, taste, visual,
reduced control of emotional response, reduced empathy, rigid- and auditory inputs. Multisensory association and limbic inte-
ity, and perseveration (Zappala et al. 2012; Dal Monte et al. gration are important to guide complex cognitive and behavioral
2014). Patients with anterior temporal lobe epilepsy may also functions, such as reward behavior associated with sensory and
manifest delusions and hallucinations. Damage to the uncinate abstract reinforcers (e.g., monetary gain and loss) (Kringelbach
fasciculus and its cortical projections has been reported in 2005) or response inhibition (e.g., go-no-go tasks) (Iversen and
children with conduct disorder (Sarkar et al. 2013) and adults Mishkin 1970).
with psychopathy (Craig et al. 2009). Hence, the disconnection The frontal aslant tract connects Broca’s territory with medial
of the left uncinate fasciculus in Gage may account for some of frontal areas (including the pre-supplementary motor area and
the behavioral manifestations reported by Harlow (1868). The un- cingulate cortex) (Lawes et al. 2008; Oishi et al. 2008; Ford et al.
cinate fasciculus has been also associated with semantic deficits 2010; Guevara et al. 2011). In patients with traumatic brain injury,
in patients with neurodegenerative disorders (Catani, Dell’acqua damage to the frontal aslant tract is correlated with impaired
Revisiting Disconnection Syndromes Thiebaut de Schotten et al. | 9
Figure 7. 3D reconstruction of Leborgne’s brain in MNI space. (a) The lesion (red) damaged both cortical structures ( posterior inferior frontal cortex) and subcortical white
matter ( perisylvian pathways). (b) 3D reconstruction of the MNI152 template, with a blue-to-orange gradient indicating the probability of disconnection of those areas not
directly affected by the lesion; red color indicate damage caused by the lesion. (c) Meta-analysis of functional MRI studies reporting activations related to the performance
of fluency tasks (75 studies).
response inhibition (Bonnelle et al. 2012). Interestingly, the initiation problems, from which patients often recover due to
strength of activation of the inferior frontal gyrus and the the bilateral distribution of this tract (see discussion for Leborgne).
pre-supplementary motor area during fMRI-based response The frontal superior and inferior longitudinal tracts connect re-
inhibition tasks has been associated with recurrent antisocial gions of the frontal lobe involved in decision-making at different
behavior (Aharoni et al. 2013). Other deficits include speech levels, from a low-processing level in the posterior frontal
10 | Cerebral Cortex
regions to a high-processing level in more anterior frontal regions (1906), Henry Head (1926), and many others. Their dissent was
(Badre and D’Esposito 2007; Badre 2008; Christoff et al. 2009). Over- based on empirical evidence of the existence of patients with
all, these longitudinal tracts permit the anatomical binding neces- non-fluent aphasia without damage to Broca’s area. Broca was
sary for complex cognitive control (Koechlin et al. 1999, 2003; also criticized for not performing dissections of the whole brain
Koechlin and Summerfield 2007). While more posterior frontal re- but limiting his investigation to the cortical surface. Indeed,
gions appeared intact in Gage’s brain, damage to the connections when computerized tomography (CT) and magnetic resonance
between posterior and anterior frontal regions could explain his imaging (MRI) scans of Leborgne’s brain were published, it was
deficits in high-level cognitive control. evident that the lesion extended well beyond the inferior frontal
The fronto-striatal and thalamo-frontal networks form gyrus to include large regions of the underlying white matter
parallel cortico-subcortical loops for motor, associative, and (Castaigne et al. 1980; Signoret et al. 1984; Cabanis et al. 1994;
limbic processing (Alexander et al. 1986; Lehericy et al. 2004; Dronkers et al. 2007). Our study confirmed that the extensive le-
Schmahmann and Pandya 2008). sion in Leborgne’s brain affected almost all dorsolateral tracts of
The associative circuit subsumes the dorsolateral prefrontal the left hemisphere, including the arcuate fasciculus and frontal
cortex, dorsal caudate nucleus, internal pallidum, and ventral an- aslant tract, both of which support language.
terior thalamic nuclei. Lesions to the associative circuit impair at- The long segment of the arcuate fasciculus connects Wer-
tention, working memory, strategy formation, and cognitive nicke’s with Broca’s region, whereas the anterior segment of
flexibility (Stuss and Benson 1984). The limbic circuit incorporates the arcuate fasciculus (or third branch of the superior longitudin-
the medial and orbitofrontal cortices, ventral striatum (i.e., nu- al fasciculus) connects Broca’s to Geschwind’s territory (in the
cleus accumbens), external and internal pallidum, and mediodor- inferior parietal lobule) (Catani et al. 2005). In addition, the frontal
sal thalamic nucleus. Functions of the limbic loop overlap with aslant tract connects Broca’s to the pre-supplementary area.
those of the fronto-orbitopolar tract described above. Gage had sig- These 3 tracts constitute a complex network dedicated to speech
nificant damage to this loop, although the exact extent of the le- production (Roelofs 2014). In Leborgne, whose only verbal output
sion is difficult to quantify using our indirect approach. was limited to a few words, the lesion to these 3 tracts explains
Overall, our analysis uncovered extensive frontal lobe damage his poor verbal fluency. Further, our analysis suggested that Le-
in Gage’s brain. This abnormality extended beyond the orbito- borgne’s pathology extended to tracts that are not part of the lan-
frontal and dorsolateral cortices, which were directly damaged guage system. The left cortico-spinal tract, for example, was
by the bar. The atlas-based approach identified several tracts damaged at different levels (corona radiata, internal capsule),
affected by the lesion, and the lesion-based approach showed which accounts for his right hemiplegia.
that the dysfunction impacted an extended network of areas It is difficult to say whether damage to other tracts, such as
that are commonly activated during the performance of deci- the uncinate fasciculus, frontal inferior longitudinal fasciculus,
sion-making, emotion processing, and reward tasks. fronto-orbitopolar tracts, and superior longitudinal fasciculus
had an impact on Leborgne’s behavior. Broca did not report any
other significant impairment and noted that Leborgne had nor-
Louis Victor Leborgne
mal intelligence. There is no mention, for example, of limb aprax-
Soon after Broca’s (1861a,b) publication, the concept of a center ia, which is usually associated with left hemisphere damage to
for spoken language was harshly criticized by Pierre Marie the superior longitudinal fasciculus. Similarly, damage to the
Revisiting Disconnection Syndromes Thiebaut de Schotten et al. | 11
frontal orbitopolar tract may have caused behavioral problems times, and damage to temporal neocortex. Semantic memory re-
that were not reported in the case notes. It is also true that Broca’s lies on a large, distributed cortical network that includes areas in
knowledge of the patient was minimal and limited to a surgical the inferior frontal gyrus and the temporal pole bilaterally (Mar-
consultation for the gangrenous leg. In the absence of more tin and Chao 2001; Hoffman et al. 2014). The latter 2 regions are
detailed clinical notes, speculation about possible symptoms interconnected through the uncinate fasciculus and the anterior
caused by tracts not directly involved in language is risky. commissure, which were damaged in Molaison’s brain. Bilateral
abnormalities of the uncinate fasciculus have been associated
with semantic deficits in neurodegenerative disorders (Mum-
Henry Molaison
mery et al. 1999; Compston 2011a,b; Catani, Mesulam et al.
Molaison’s groundbreaking case established that bilateral medial 2013), and the disconnection of these tracts may have contribu-
temporal lobe lesions cause severe amnesia (Scoville and Milner ted to his shortcomings on certain semantic tasks.
1957). MRI studies carried out in 1992 and 1993 showed that the Axonal tracing studies in animals show that the anterior fi-
resection included the medial temporal polar cortex, most of bers of the anterior commissure also project to olfactory regions
the amygdaloid complex, and all of the entorhinal cortex (Corkin (e.g., olfactory bulb, anterior perforated substance, etc.) (Crosby
et al. 1997). Also removed were the anterior ∼2 cm of the dentate et al. 1962). These anterior olfactory-linked fibers seem to be pre-
gyrus, hippocampus, and subicular complex, and the rostral por- sent in humans (Kiernan 1998; Di Virgilio et al. 1999), and the
tions of perirhinal and parahippocampal cortices. damage to these tracts that we have identified in H.M. may
Molaison’s memory impairment was more severe than have contributed to his deficit in odor quality discrimination
that of amnesic patients with selective hippocampal lesions and recognition (Eichenbaum et al. 1983).
generated after these steps provided an overall estimation of the pure white matter lesions the cortex is not affected. This suggests
tracts’ anatomy, it may not precisely match the exact individual that a network dysfunction is the common denominator for all
anatomy of the 3 patients. brain disorders, and a tract-based nomenclature should be pre-
Further, the diffusion-weighted imaging that we have used ferred to a cortical localizationism. However, syndromes certain-
for this analysis is based on a population of 18- to 79-year-old ly result from a dysfunction of an extended network of cortical
healthy participants and is not age or sex matched for each his- and subcortical areas connected by several tracts (Damasio and
toric patient. Age-related changes in volume and diffusion indi- Damasio 1989). Hence, mapping the disconnection in patients
ces of white matter pathways have been reported in previous should not lead to an underestimation of the role of the cortex.
studies (Stadlbauer et al. 2008; Rojkova et al. 2015). Similarly, Indeed, our lesion-based analysis (Figs 6a, 7b, and 8a) revealed
sex-related differences have been reported for the right arcuate cortical regions that were directly or indirectly affected by the
fasciculus (Catani et al. 2007). These differences can lead to disconnection.
under- or overestimation of the exact extension of the lesion.
For this reason, we have included in our interpretation those
tracts that showed a trend towards significant disconnection.
Conclusions
It should also be noted that tract estimation was based on Today as 50 years ago, the clinico-anatomical correlation method
tractography, which has many flaws. Even with more advanced remains pivotal in our understanding of the complex relations
diffusion methods, like spherical deconvolution, artifacts can between brain and behavior (Mah et al. 2014). The disconnection
occur due to partial volume effects and difficulty in reconstruct- paradigm, as envisaged by Geschwind in his landmark paper and
ing complex anatomical configurations (e.g., crossing, kissing, revitalized today by the availability of methods for mapping con-
and supported within the framework of a subsidy granted to the Brett M, Leff AP, Rorden C, Ashburner J. 2001. Spatial normaliza-
HSE by the Government of the Russian Federation for the imple- tion of brain images with focal lesions using cost function
mentation of the Global Competitiveness Program. Conflict of masking. Neuroimage. 14:486–500.
Interest: None declared. Broca P. 1861a. Perte de la parole, ramollissement chronique et
destruction partielle du lobe antérieur gauche du cerveau.
Bull Soc Anthropol. 2:235–238. 301–321.
Broca P. 1861b. Remarques sur le siège de la faculté du langage
References articulé; suivies d’une observation d’aphémie ( perte de la
Acosta-Cabronero J, Alley S, Williams GB, Pengas G, Nestor PJ. parole). Bull Soc Anthropol. 36:330–357.
2012. Diffusion tensor metrics as biomarkers in Alzheimer’s Brodmann K. 1909. Vergleichende Lokalisationslehre der
disease. PLoS ONE. 7:e49072. Großhirnrinde: in ihren Prinzipien dargest. auf Grund d.
Adey W, Meyer M. 1952. An experimental study of the hippocam- Zellenbaues. Leipzig, Germany: Barth.
pal afferent pathways from prefrontal and cingulae areas in Cabanis E, Iba-Zizen M, Abelanet R, Monod-Broca P, Signoret J.
the monkey. J Anat. 86:58–74. 1994. “Tan-Tan” the first Paul Broca’s patient with “Aphemia”
Aggleton JP. 2008. EPS Mid-Career Award 2006. Understanding an- (1861): CT (1979), & MRI (1994) of the brain. Nancy, France: Euro-
terograde amnesia: disconnections and hidden lesions. Q J pean Society of Neuroradiology.
Exp Psychol (Colchester). 61:1441–1471. Carrera E, Tononi G. 2014. Diaschisis: past, present, future. Brain.
Aharoni E, Vincent GM, Harenski CL, Calhoun VD, Sinnott- 137:2408–2422.
Armstrong W, Gazzaniga MS, Kiehl KA. 2013. Neuroprediction Castaigne P, Lhermitte F, Signoret JL, Abelanet R. Description et
Compston A. 2009. A short history of clinical neurology. In: Eichenbaum H, Morton TH, Potter H, Corkin S. 1983. Selective ol-
Donaghy M, editor. Brain’s diseases of the nervous system. factory deficits in case H.M. Brain. 106(Pt 2):459–472.
Oxford: Oxford University Press. Finger S. 1994. Origins of neuroscience. New York: Oxford Univer-
Corkin S. 1984. Consequences of bilateral medial temporal lobec- sity Press.
tomy: clinical course and experimental findings in H.M. Ford A, McGregor KM, Case K, Crosson B, White KD. 2010. Struc-
Semin Neurol. 4:1–11. tural connectivity of Broca’s area and medial frontal cortex.
Corkin S. 2013. Permanent present tense: the unforgettable life of Neuroimage. 52:1230–1237.
the amnesic patient H.M. Philadelphia: Basic Books. Forkel SJ, Thiebaut de Schotten M, Dell’Acqua F, Kalra L,
Corkin S, Amaral DG, González RG, Johnson KA, Hyman BT. 1997. Murphy DG, Williams SC, Catani M. 2014. Anatomical predic-
H. M.’s medial temporal lobe lesion: findings from magnetic tors of aphasia recovery: a tractography study of bilateral peri-
resonance imaging. J Neurosci. 17:3964–3979. sylvian language networks. Brain. 137:2027–2039.
Craig MC, Catani M, Deeley Q, Latham R, Daly E, Kanaan R, Friston KJ, Ashburner J, Frith CD, Poline JB, Heather JD,
Picchioni M, McGuire PK, Fahy T, Murphy DG. 2009. Altered Frackowiak RSJ. 1995. Spatial registration and normalization
connections on the road to psychopathy. Mol Psychiatry. of images. Hum Brain Mapp. 3:165–189.
14:946–953. 907. Gaffan D, Easton A, Parker A. 2002. Interaction of inferior tem-
Crinion J, Ashburner J, Leff A, Brett M, Price CJ, Friston KJ. 2007. poral cortex with frontal cortex and basal forebrain: double
Spatial normalization of lesioned brains: performance dissociation in strategy implementation and associative
evaluation and impact on fMRI analyses. NeuroImage. learning. J Neurosci. 22:7288–7296.
37:866–875. Gaffan D, Parker A, Easton A. 2001. Dense amnesia in the monkey
Klein A, Andersson J, Ardekani BA, Ashburner J, Avants B, Mesulam MM. 1990. Large-scale neurocognitive networks and
Chiang MC, Christensen GE, Collins DL, Gee J, Hellier P, et al. distributed processing for attention, language, and memory.
2009. Evaluation of 14 nonlinear deformation algorithms ap- Ann Neurol. 28:597–613.
plied to human brain MRI registration. NeuroImage. 46:786–802. Mesulam MM. 1999. Spatial attention and neglect: parietal, front-
Koechlin E, Basso G, Pietrini P, Panzer S, Grafman J. 1999. The role al and cingulate contributions to the mental representation
of the anterior prefrontal cortex in human cognition. Nature. and attentional targeting of salient extrapersonal events.
399:148–151. Philos Trans R Soc Lond B Biol Sci. 354:1325–1346.
Koechlin E, Ody C, Kouneiher F. 2003. The architecture of cogni- Metzler-Baddeley C, Hunt S, Jones DK, Leemans A, Aggleton JP,
tive control in the human prefrontal cortex. Science. O’Sullivan MJ. 2012. Temporal association tracts and the
302:1181–1185. breakdown of episodic memory in mild cognitive impair-
Koechlin E, Summerfield C. 2007. An information theoretical ap- ment. Neurology. 79:2233–2240.
proach to prefrontal executive function. Trends Cogn Sci. Monakow CV. 1914. Die Lokalisation im Grosshirn und der
11:229–235. Funktion durch kortikale Herde. Wiesbaden: Verlag von J.F.
Kolb B, Gibb R. 2014. Searching for the principles of brain plasti- Bergmann.
city and behavior. Cortex. 58:251–260. Monakow CV. 1897. Gehirnpathologie. Vienna: Hölder A.
Kramer UM, Solbakk AK, Funderud I, Lovstad M, Endestad T, Moulin CJ, Souchay C, Morris RG. 2013. The cognitive neuropsych-
Knight RT. 2013. The role of the lateral prefrontal cortex in in- ology of recollection. Cortex. 49:1445–1451.
hibitory motor control. Cortex. 49:837–849. Mummery CJ, Patterson K, Wise RJ, Vandenberghe R, Price CJ,
Kringelbach ML. 2005. The human orbitofrontal cortex: linking re- Hodges JR. 1999. Disrupted temporal lobe connections in se-
of the brain: parallel observations from diffusion spectrum Thiebaut de Schotten M, Kinkingnehun S, Delmaire C, Lehericy S,
imaging and autoradiography. Brain. 130:630–653. Duffau H, Thivard L, Volle E, Levy R, Dubois B, Bartolomeo P.
Schmolck H, Kensinger EA, Corkin S, Squire LR. 2002. Semantic 2008. Visualization of disconnection syndromes in humans.
knowledge in patient H.M. and other patients with bilateral Cortex. 44:1097–1103.
medial and lateral temporal lobe lesions. Hippocampus. Thiebaut de Schotten M, Tomaiuolo F, Aiello M, Merola S,
12:520–533. Silvetti M, Lecce F, Bartolomeo P, Doricchi F. 2014. Damage to
Schrier AM, Sperry RW. 1959. Visuomotor integration in split- white matter pathways in subacute and chronic spatial neg-
brain cats. Science. 14:1275–1276. lect: a group study and 2 single-case studies with complete vir-
Scoville W, Milner B. 1957. Loss of recent memory after bilateral tual “in vivo” tractography dissection. Cereb Cortex. 24:691–706.
hippocampal lesions. J Neurol Neurosurg Psychiatr. 20:11–21. Van Horn JD, Irimia A, Torgerson CM, Chambers MC, Kikinis R,
Scoville WB, Dunsmore RH, Liberson WT, Henry CE, Pepe A. 1953. Toga AW. 2012. Mapping connectivity damage in the case of
Observations on medial temporal lobotomy and uncotomy in Phineas Gage. PLoS ONE. 7:e37454.
the treatment of psychotic states; preliminary review of 19 Vann SD, Albasser MM. 2011. Hippocampus and neocortex:
operative cases compared with 60 frontal lobotomy and recognition and spatial memory. Curr Opin Neurobiol.
undercutting cases. Res Publ Assoc Res Nerv Ment Dis. 21:440–445.
31:347–373. Vogt C, Vogt O. 1926. Die vergleichend-architektonische und ver-
Seghier ML, Ramlackhansingh A, Crinion J, Leff AP, Price CJ. 2008. gleichend-reizphysiologische Felderung der Grosshirnrinde
Lesion identification using unified segmentation-normalisa- unter besonderer Berücksichtigung der menschlichen.
tion models and fuzzy clustering. NeuroImage. 41:1253–1266. Naturwissenschaften. 14:1190–1194.