ANRV314-NE30-19 ARI 20 May 2007 15:43

Development of Neural
Systems for Reading
Bradley L. Schlaggar1,∗
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

and Bruce D. McCandliss2,∗

Departments of Neurology, Radiology, Anatomy & Neurobiology, and Pediatrics,
1
Washington University School of Medicine, St. Louis, Missouri 63110;
email: [email protected]
2
Sackler Institute for Developmental Psychobiology, Weill Medical College of
Cornell University, New York, NY 10021; email: [email protected]

Annu. Rev. Neurosci. 2007. 30:475–503 Key Words

First published online as a Review in Advance on fMRI, diffusion tensor imaging, event-related potentials, dyslexia,
March 22, 2007
genetics, perceptual expertise
The Annual Review of Neuroscience is online at
neuro.annualreviews.org Abstract
This article’s doi: Functional and structural neuroimaging studies of adult readers have
10.1146/annurev.neuro.28.061604.135645
provided a deeper understanding of the neural basis of reading, yet
Copyright  c 2007 by Annual Reviews. such findings also elicit new questions about how developing neural
All rights reserved
systems come to support this learned ability. A developmental cogni-
0147-006X/07/0721-0475$20.00 tive neuroscience approach provides insights into how skilled reading
∗
Authors contributed equally to this chapter emerges in the developing brain, yet also raises new methodologi-
cal challenges. This review focuses on functional changes that occur
during reading acquisition in cortical regions associated with both
the perception of visual words and spoken language, and it examines
how such functional changes differ within developmental reading
disabilities. We integrate these findings within an interactive spe-
cialization framework of functional development and propose that
such a framework may provide insights into how individual differ-
ences at several levels of observation (genetics, white matter tract
structure, functional organization of language, cultural organization
of writing systems) impact the emergence of neural systems involved
in reading ability and disability.

475
 ANRV314-NE30-19 ARI 20 May 2007 15:43

Contents
INTRODUCTION . . . . . . . . . . . . . . . . . 476 FOR READING
OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . 478 DEVELOPMENT. . . . . . . . . . . . . . . 486
FUNCTIONAL NEUROSCIENCE APPROACHES
NEUROANATOMY OF TO UNDERSTANDING
WORD READING IN SKILLED INDIVIDUAL DIFFERENCES
ADULT READERS . . . . . . . . . . . . . . 478 IN READING
Functional Significance of the DEVELOPMENT. . . . . . . . . . . . . . . 487
VWFA . . . . . . . . . . . . . . . . . . . . . . . . 479 Functional Imaging of Individual
Studies of Letter-Sound Differences in Reading . . . . . . . . 487
Integration in Skilled Readers . . 480 Individual Differences in White
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

CHALLENGES TO STUDYING Matter Tract Properties . . . . . . . . 489

THE DEVELOPMENT OF Genetic and Cultural
READING-RELATED TASKS Contributions to Individual
USING FUNCTIONAL Differences in Reading
NEUROIMAGING. . . . . . . . . . . . . . 481 Development . . . . . . . . . . . . . . . . . 491
SPECIALIZATION OF THE RESEARCH ON
VWFA IS IMPLICATED IN INTERVENTIONS FOR
THE DEVELOPMENT OF READING DISABILITY IN
PERCEPTUAL EXPERTISE IN CHILDREN . . . . . . . . . . . . . . . . . . . . 493
READING . . . . . . . . . . . . . . . . . . . . . . 483 CONCLUDING REMARKS:
INTEGRATION OF INTERACTIVE
ORTHOGRAPHIC AND SPECIALIZATION AS A
PHONOLOGICAL CONCEPTUAL FRAMEWORK
PROCESSING IS IMPORTANT FOR READING
DEVELOPMENT. . . . . . . . . . . . . . . 494

“One of the strong claims of developmental nitive operations associated with lower-level
cognitive neuroscience is that a comprehen- perceptual processes and higher-level lan-
sive understanding of mature cognition can- guage function to specific brain systems (e.g.,
not be attained without understanding both Posner et al. 1988). More recently, develop-
normal and abnormal development of the mental cognitive neuroscience investigations
human brain. . .we cannot understand how have begun to examine the transformations
the mature system works until we under- in functional brain organization that support
stand how it is constructed in development, the emergence of reading skill. This work
and we cannot fully understand that process is beginning to address questions concerning
of normal construction without understand- how this evolutionarily recent human ability
ing how development can go awry.” emerges from changes within and between
– Johnson & Pennington (1999)
brain systems associated with visual percep-
tion and language abilities, how learning ex-
periences and maturation impact these neural
INTRODUCTION changes, and how individual differences at the
Reading ability has served as a model sys- genetic and neural systems level influence the
tem within cognitive science for linking cog- emergence of this skill.

476 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

A traditional approach to understanding from preexisting visual perception and lan-

the neurobiology of reading ability and dis- guage abilities. For example, interpretation of
ability has been to investigate the conse- orthography, the written form of language,
PET: positron
quences of focal brain lesions on the skilled places unique demands on the brain’s visual emission
adult reader (see Warrington & Shallice 1980 object-processing systems. Linking orthog- tomography
for review). More recently, functional neu- raphy with phonology, the sound structure fMRI: functional
roimaging, including positron emission to- of language, is the sine qua non of reading magnetic resonance
mography (PET) and functional magnetic acquisition (Share 1995) and implicates the imaging
resonance imaging (fMRI), as well as event- formation of functional connections between ERP: event-related
related potentials (ERP), have been used to visual object processing systems and systems potential
investigate the functional neuroanatomy of involved in processing spoken language. Orthography: the
reading skills in adults. Results from these A developmental cognitive neuroscience graphic structure of
approaches have provided a foundation for approach to investigating the emergence of language
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

understanding the neural basis of skilled read- reading ability and the associated changes in Phonology: the
ing and have been used to support theoreti- functional brain organization that accompany sound structure of
language
cal distinctions between component processes it provide unique perspective on these is-
of reading linked to particular cortical re- sues. Investigations of developmental changes
gions. In studying adults exclusively, how- within and between cortical regions associated
ever, such approaches to linking cognitive with orthographic and phonological aspects
functions to particular brain systems often of word reading help elucidate the neural
rely on simplifying assumptions, which be- basis for perceptual expertise that supports
comes problematic when considering issues of rapid word recognition. Such investigations
how functions emerge within brain systems. also help delineate the neural basis for the
For example, Johnson and coworkers (2002) well-established impact of individual differ-
note that many theories of structure/function ences in phonological skill on the develop-
relationships in adults either provide little ment of reading ability (Share 1995). Ba-
insight into how such relationships emerge sic research into maturational changes in the
or rely on the “static assumption,” which brain during the years in which reading exper-
presumes that functional properties of brain tise develops may provide important insights
regions observed in adulthood necessarily re- into how these changes influence learning and
flect those observed during human develop- functional specialization in cortical regions.
ment. Such assumptions may prove especially Finally, a comprehensive account of reading
problematic in explaining the development of development will eventually need to incor-
reading ability, in which the major compo- porate molecular (such as genetic susceptibil-
nent processes (i.e., visual word form recogni- ity for reading disability) and neurobiologi-
tion, letter-sound association) depend entirely cal (such as detailed information about white
on the rather recent development of writ- matter tract structure) factors that influence
ing systems and implicate learning processes individual differences in the capacity to ac-
that rely on explicit teaching of this cultural quire reading skill. Such studies may provide
invention. a more integrated understanding of develop-
To understand how the brain’s organiza- mental disorders of reading and set forth a
tion for skilled reading emerges, one must ex- theoretical basis for optimizing remediation
amine the changes that the developing brain strategies. Such issues, as described below, are
undergoes in the process of acquiring the skill. only now being addressed.
Some of the major challenges involve charac- Over the past decade, the field of devel-
terizing the nature of the progressive changes opmental cognitive neuroscience has enjoyed
that occur over the early reading years and remarkable growth and a concomitant surge
understanding how reading abilities emerge in methodological development for brain

www.annualreviews.org • Development of Reading 477

 ANRV314-NE30-19 ARI 20 May 2007 15:43

mapping. Consequently, although nascent, Next, the third and fourth sections address
the state of a developmental cognitive neuro- the developmental specialization of brain sys-
science of reading can be discussed. Although tems that occur as children develop from naı̈ve
reading development implicates many forms to expert readers.
of cognitive and neural processes that un- The fifth section, Neuroscience Ap-
dergo changes across development and skill proaches to Understanding Individual Dif-
acquisition, the scope of this review specifi- ferences in Reading Development, examines
cally focuses on factors that may contribute several sources of individual differences that
to the developmental changes that occur be- impact the development of reading ability and
tween brain systems that support the integra- extends these observations to help explain the
tion of the visual and phonological aspects neural bases of developmental reading dis-
of deciphering written words. The develop- abilities. A related sixth section describes re-
ment of this connection building is best ex- cent research linking intervention efforts for
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

plained perhaps within an interactive special- children with reading disabilities with obser-
ization scheme ( Johnson 2001, Johnson et al. vations of changes in patterns of functional
2002) that has implications for both visual per- activation.
ceptual processes as well as for phonological The concluding section discusses how the
processes, as both of these systems undergo concept of interactive specialization and re-
changes in the transition from the novice child lated computational models of reading may
to the skilled adult reader. provide a framework for relating individual
differences in neural systems to individual dif-
ferences in the functional organization of the
OVERVIEW expert adult state.
This review, organized into seven sections,
brings together recent insights from multiple
areas of neuroscientific investigation that are FUNCTIONAL
germane to understanding the development NEUROANATOMY OF WORD
of neural processes necessary to relate print READING IN SKILLED ADULT
to speech. We focus on this specific aspect of READERS
reading because it is both an essential subskill Notions of the functional localization of com-
for reading development and also a useful ve- ponent reading functions emerged in the era
hicle to illustrate connections across multiple of classic studies of language function in the
levels of observation in relating neural struc- late nineteenth century. Specifically, the idea
ture and function in reading development. of separate neural routes to support different
The first section discusses the functional aspects of reading was motivated by observa-
organization of reading in skilled adult read- tions of the consequences from focal brain le-
ers as a natural end-state of development to sions noted by Dejerine (Dejerine 1891). He
set the stage for a review of developmental described patients with different forms of ac-
investigations. quired reading difficulty (i.e., dyslexia) due
The second section, Challenges to Study- to focal lesions in the left posterior hemi-
ing the Development of Reading-Related sphere, contrasting effects of ventral lesions
Tasks Using Functional Neuroimaging, de- (i.e. fusiform gyrus), and dorsal lesions (i.e.,
lineates some confounds inherent in inter- left angular gyrus). Clinically distinct forms
preting developmental changes within fMRI of acquired dyslexia associated with different
and ERP studies and describes several ap- focal lesions ultimately led to the idea that
proaches to differentiating the contributions there is both a direct route from print to
of various factors that impact brain activity meaning as well as an indirect route medi-
across development. ated by associations between letters and the

478 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

sounds of language (Marshall & Newcome phonological elements within words (Fiez &
1973, Warrington & Shallice 1980, Harm & Petersen 1998, Poldrack et al. 1999).
Seidenberg 2004).
VWFA: visual word
PET and fMRI investigations of such pro- form area
cesses have contributed to numerous stud- Functional Significance of the VWFA
Prelexical: relating
ies attempting to characterize the functional Dejerine’s pioneering work associating some to units within words
circuitry supporting skilled adult word read- specific aspects of reading function with the
Pseudowords:
ing. Convergent patterns across these stud- left mid-fusiform gyrus anticipated one of the pronounceable
ies are evident in both qualitative and quan- most intriguing structure-function debates in strings of letters
titative meta-analyses (Fiez & Petersen 1998, cognitive neuroscience: the significance and without meaning
Turkeltaub et al. 2002, Jobard et al. 2003, Bol- specificity of the so-called visual word form
ger et al. 2005). In summarizing this work, area. This appellation comes from the obser-
we focus here on the ventral (orthographic) vation across multiple studies demonstrating
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

and the dorsal (phonological) systems because that contrasting visual words with other com-
they represent the computations most directly plex unfamiliar stimuli leads to increased ac-
relevant to the specialization of reading abil- tivity within a particular brain region, typi-
ity. cally involving a left occipitotemporal region
The visual orthographic regions are lo- centered on the mid-fusiform gyrus (for re-
cated ventrally in the extrastriate cortex. The view see McCandliss et al. 2003). There is
bilateral regions are thought to support initial little controversy surrounding this principle
visual processing and feed into the more an- observation and its replication across studies
terior left-lateralized region, which has been (McCandliss et al. 2003, Bolger et al. 2005),
termed by some as the “visual word form although current debate centers on the con-
area,” or VWFA (Cohen et al. 2000, 2002; tribution of this area to nonreading tasks and
McCandliss et al. 2003), and is argued to the precise functional significance this activ-
process a prelexical representation of let- ity contributes to reading function (for review
ter patterns within visual words and pseu- see Price & Devlin 2003, Cohen et al. 2004).
dowords (see Functional Significance of the To investigate the functional significance
VWFA). The putative VWFA is among the of the VWFA for word recognition, some
most consistently activated regions in quanti- studies have used higher spatial resolution
tative meta-analytic studies of adult reading. techniques (intracranial electrodes, high res-
The phonological system can be function- olution fMRI) to demonstrate increased ac-
ally divided into two components: a left dor- tivity for letter strings over other classes of
sal posterior component and a left anterior visual stimuli presented under equivalent con-
component. The left dorsal posterior compo- ditions (Allison et al. 1999, Baker et al. 2005).
nent, also referred to as the perisylvian region, When such effects are reported, they are typ-
includes the supramarginal gyrus, angular ically described in terms of relative strengths
gyrus, and superior temporal cortex (Rumsey of activation, rather than demonstrating ex-
et al. 1997). In adult imaging and ERP stud- clusivity of function for the domain of visual
ies, this system tends to have greater activity words. Evidence appears to support prefer-
while reading pseudowords than real words ential, but not exclusive, processing of word
and is thought to function as an integrative re- form–related stimuli in the putative VWFA,
gion linking orthography to phonology (Pugh which we discuss below as supporting a form
et al. 2001). The left anterior component, not of perceptual expertise for visual words.
emphasized in this review, includes the infe- Recent research on activation of the
rior frontal gyrus extending into the dorsal VWFA for visual word recognition implicates
premotor cortex and has been associated with a role for this region in the bottom-up per-
speech production as well as active analysis of ceptual encoding of orthographic properties

www.annualreviews.org • Development of Reading 479

 ANRV314-NE30-19 ARI 20 May 2007 15:43

of letter strings, demonstrating a paramet- see also Raij et al. 2000) demonstrated a role
ric relationship between activity in this re- for heteromodal (i.e., responsive to both vi-
gion and the statistical properties by which sual and auditory input) portions of the left
letters are organized into word forms within superior temporal cortex in the integration of
a given writing system (Dehaene et al. 2004, orthographic and phonological processing. In
Binder et al. 2006). However, several studies these experiments, visual letters and auditory
have demonstrated that although this region letter sounds were presented simultaneously,
is not typically responsive to passive presenta- with half the trials presenting a match between
tion of auditory words, this region’s activation the letters in these two modalities. Responses
level can be modulated by top-down atten- in these hetermodal regions were enhanced
tional and/or linguistic processes (Booth et al. when the identity of the visual and audi-
2003, Devlin et al. 2006). Integrating these tory letters matched. These findings suggest
two sets of observations is the notion that a form of experience-based plasticity in the
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

word-specific activation within the extrastri- response properties of these regions because
ate regions may reflect a “gateway” through associations between letters and sounds are
which critical invariant information extracted inherent only in the subjects’ reading experi-
from visual forms makes contact with linguis- ence. Furthermore, these results have recently
tic representations (Carr & Posner 1995). Ac- been extended to demonstrate that dyslexic
tivity within the VWFA may still reflect many adults show deficits in this experience-driven
properties that are general to visual recogni- effect, suggesting that disruptions in phenom-
tion and may also be influenced by top-down ena associated with letter-sound integration
attention processes associated with increased may be linked to disruptions in skilled read-
connectivity between phonological and se- ing (Blau et al. 2007).
mantic systems outside the visual system. We The link between experience and the for-
propose below that a primary function of the mation of such sensitivities was further probed
VWFA during reading is to support a form by Hashimoto & Sakai (2004), who investi-
of perceptual expertise for visual word recog- gated the cortical plasticity of skilled adult
nition that enables rapid perception of visual readers being trained to learn new letter-
words in one’s own language. However, a great sound associates in a short-term training
deal remains to be understood regarding how paradigm. In this repeated measures study,
this specialization develops. the primary sites of training-related change
were in a region near the left ventral occipi-
totemporal cortex in the mid-fusiform gyrus
Studies of Letter-Sound Integration and another in the left parietal/occipital cor-
in Skilled Readers tex near the angular gyrus. The study also
In addition to representing visual word forms showed an increase in the effective connec-
efficiently, reading also requires the forma- tivity (a measure of task-related correlated ac-
tion of highly specific mappings between vi- tivity) of these two regions.
sual print representations (orthography) and Taken together, these studies suggest a spe-
representations related to the sounds of lan- cific cortical circuitry involved in integrat-
guage (phonology). A number of studies have ing orthographic and phonological informa-
begun to investigate interactions between oc- tion and provide initial insight into how this
cipital and temporal regions that may be cru- circuitry changes with learning in the skilled
cially involved in the functional organization reader. The regions implicated are consis-
of reading ability, as such regions are impli- tently related to orthographic and phonologi-
cated in letter-sound integration. cal processing in the functional mapping stud-
In a study of skilled adult readers, Blom- ies reviewed above. However, demonstrations
ert and coworkers (Van Atteveldt et al. 2004; of experience-dependent changes in skilled

480 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

readers might differ dramatically from begin- One common concern is that age-group
ning readers forming such connections for the imaging differences could merely be reduced
first time. More direct links to development to age-related changes in performance on an
blood oxygenation
are needed, although developmental studies activation task such as time on task or number level dependent
raise significant challenges of their own. of error trials (Schlaggar et al. 2002, Murphy (BOLD): fMRI
& Garavan 2004). The existence of age- signal
related changes in performance on a reading-
CHALLENGES TO STUDYING related activation task at the time of imaging
THE DEVELOPMENT OF does not necessarily equate with age-related
READING-RELATED TASKS changes in overall reading skill. A distinc-
USING FUNCTIONAL tion must be made between stable measures
NEUROIMAGING of skill (such as psychometric tests) and in-
We now turn our attention to child stud- scanner performance measures (i.e., reaction
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

ies that examine the development of func- time and accuracy) within a particular session.
tional activation within brain systems asso- The burden on the field is to generate schemes
ciated with the emergence of skilled visual to differentiate these effects. This burden is
word recognition. In doing so, we focus on not unique to developmental studies; it is ger-
several interpretational challenges inherent mane to contrasts across the lifespan (e.g.,
within developmental cognitive neuroscience Grossman et al. 2002) and for comparing
research. patient to control populations (Weinberger
The ability to study typically develop- 1998, Price & Friston 1999, Johnson et al.
ing children with functional neuroanatomical 2002). Multiple strategies, each with different
techniques became more available with the strengths and weaknesses, exist to differenti-
advent of fMRI, which uses the blood oxy- ate activation patterns that may be associated
genation level dependent (BOLD) signal to with the separable influences of age, skill, and
indirectly measure neural activity related to performance effects (Palmer et al. 2004, Casey
execution of a cognitive task such as word et al. 2005).
recognition. fMRI studies of the developmen- In one approach to studying developmen-
tal functional neuroanatomy of reading have tal changes in fMRI activity to visual words,
provided evidence related to the developmen- Eden and coworkers (Turkeltaub et al. 2003)
tal dynamics that occur in brain circuits that attempted to minimize age-associated per-
support reading (Booth et al. 2001, 2004; formance confounds that typically accom-
Schlaggar et al. 2002; Shaywitz et al. 2002, pany explicit word reading tasks. They em-
2004; Gaillard et al. 2003a,b; Turkeltaub et al. ployed a strategy of studying reading as an
2003; Brown et al. 2005). A central focus implicit process, contrasting activation to vi-
of these studies is on how developmental sual words versus novel letter-like characters
changes in brain activity within particular cor- that served as control stimuli. To avoid de-
tical regions are linked to the emergence of velopmental confounds between reading skill
reading skill. However, concomitant changes and execution of the activation task, subjects
in brain maturation and changes in perfor- of all ages completed the simple task of mak-
mance dynamics within the activation task ing a physical judgment about parts of letters
also influence brain activity patterns over de- that appeared with equal frequency in both
velopment, posing significant challenges to stimulus types. Activity attributed to implicit
research in this area. Although no consisent word reading was measured by contrasting
solution to these challenges has emerged, a fMRI signals for words minus letter-like stim-
number of approaches to differentiating the uli. The presumption with using an implicit
contributions of these different effects have stimulus contrast approach is that general
been investigated. developmental differences for task difficulty

www.annualreviews.org • Development of Reading 481

 ANRV314-NE30-19 ARI 20 May 2007 15:43

(including differences in error number, re- Implementation of this strategy resulted

action time) will not influence the contrast in the characterization of brain regions with
between the stimulus classes. Unfortunately, activity profiles common across age, regard-
such approaches are not impervious to perfor- less of performance, as age/performance in-
mance effects because this very paradigm has dependent. Regions with group differences
demonstrated longer reaction times for more were classified as either age related (i.e., dif-
word-like stimuli in adults (Price et al. 1996, ference present in both performance-matched
Binder et al. 2006). An additional complica- and non-matched comparisons) or perfor-
tion to this approach is that differences in the mance related (i.e., difference present in
difficulty of primary tasks are known to modu- only the non-matched comparison). Age-
late the degree of implicit processing of irrel- related regions where activity increased with
evant stimuli (Lavie & Tsal 1994). Thus, de- age were found in left frontal and parietal
velopmental differences in the ability to deal cortex. Most age-related regions, however,
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

with the demands of the primary task can in- showed decreased activity with increased age.
directly influence implicit word processing. These regions were distributed bilaterally and
An alternative approach to segregating were concentrated in bilateral ventral occip-
these influences involves contrasting sub- itotemporal cortex. Performance-related re-
sets of the subject population according to gions were also found in bilateral extrastri-
age and performance levels and performing ate cortex including a region in the right
multiple analyses. Such approaches, called ventral extrastriate cortex—a region previ-
“performance-matching,” allow differentia- ously demonstrated to have a reduced in-
tion of activation changes associated with volvement in reading over the course of read-
maturation (i.e., age) from those associated ing development (Turkeltaub et al. 2003) and
with performance dynamics. Schlaggar and skill acquisition (Shaywitz et al. 2002). An-
coworkers demonstrated the potential of this other performance-related region in the left
approach within a large cross-sectional event- parietal-occipital-temporal junction has also
related fMRI study (Schlaggar et al. 2002, been consistently implicated in developmen-
Brown et al. 2005) designed to examine dif- tal dyslexia (e.g., Temple 2002).
ferences between adults and school-age chil- This analysis strategy provides a way of dis-
dren in lexical processing of visually presented entangling changes that might be linked to
words. In-scanner performance differences task performance and changes that might re-
were observed across accuracy and reaction- flect maturational differences in the way the
time profiles for adults and children, but with brain increases activity in regions related to
significant overlap between the age groups. computations in reading-related tasks. As ex-
This overlap enabled the formation of two pected, a set of regions exhibited responses
age-group analyses, one with a performance- that were equivalent across all age and perfor-
matched subset and another with unmatched mance groups. In addition, regions in ventral
performance subgroups. Performance match- occipitotemporal extrastriate cortex typically
ing across groups that differ (e.g., by age, associated with visual orthographic process-
diagnosis, or treatment) can introduce con- ing exhibited dynamic patterns of change as-
founds by applying opposing selection biases. sociated with age in some regions and per-
To protect against such confounds, these in- formance differences in other regions (see
vestigators required that the observed age- Figure 1). This pattern of strong recruit-
group differences in regional activation must ment of extrastriate regions in young readers,
be present in both performance-matched and which gradually diminishes with age, could
non-matched group comparisons for those reflect a form of protracted development in
differences to be considered driven by subject analysis of visual words, for which top-down
age rather than by in-scanner performance. influences are necessary for word reading

482 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Figure 1
Select left extrastriate regions from a large developmental fMRI study of word generation tasks
(Schlaggar et al. 2002, Brown et al. 2005) showing regions where activity differs by age but not
performance (red) and where activity is present independent of age or performance (green). Left
hemisphere viewed from an oblique-lateral perspective is shown in (a). The same hemisphere is viewed
from a ventral (b) and posterior (c) perspective. (d ) Event-related fMRI time courses for different age
groups in an age-related region with activity decreasing with age. Peak BOLD activity as a function of
age is shown in scatterplots with fitted curves for both the region showing decreasing activity (e) and a
region (circled on the posterior view) activated in the task that did not change with age ( f ).

(see Specialization of the VWFA Is Implicated described for the transition from novice to
in the Development of Perceptual Expertise expert in adults after skill learning (Raichle
in Reading). et al. 1994). Understanding age-related differ-
The presence of regions of the brain that ences is crucial for revealing the developmen-
show functional activation differences be- tal processes engaged during learning/skill
tween adults and children, even when per- acquisition and the interplay of maturation
formance dynamics are matched, can be con- with learning dynamics across reading devel-
sidered a behavioral phenocopy; the notion opment.
that identical performance is observed across
groups yet is supported by different under-
lying neural mechanisms. In principle, these SPECIALIZATION OF THE VWFA
age-related differences in underlying neural IS IMPLICATED IN THE
mechanisms would be invisible to behavioral DEVELOPMENT OF
assessment alone, demonstrating the added PERCEPTUAL EXPERTISE IN
value of functional imaging to revealing typi- READING
cal and atypical development. Age-related re- Functional specialization of the VWFA is
gions may constitute developmentally tran- thought to emerge during acquisition of read-
sient functional scaffolding analogous to that ing expertise (McCandliss et al. 2003, Pugh

www.annualreviews.org • Development of Reading 483

 ANRV314-NE30-19 ARI 20 May 2007 15:43

et al. 2001). The development of a specialized Future research will be required to isolate
ability to process letters, letter strings, and vi- whether the observed changes are systemati-
sual words could begin as a general object- cally linked to improvements in general read-
RD: reading-
disabled/reading recognition problem, using a large number ing skill, improvements in in-scanner task per-
disability of visual processing mechanisms. Through formance, or maturational changes that might
NI: nonimpaired experience and maturation of the underly- modulate how the task elicits activity in a given
ing neural substrate, and through exposure region.
to patterned orthographic information within Developmental functional imaging results
a specific writing system, the processing sys- from several groups are beginning to con-
tem could more efficiently utilize and tune verge on a central conclusion regarding de-
the most appropriate mechanisms that focus velopmental changes in extrastriate regions
on the most informative aspects of the in- involved in reading. During the ages when
put, while decreasing use of the less appropri- reading skill is acquired, a transition occurs
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

ate mechanisms. In this way, visual processing from bilateral extrastriate region involvement
mechanisms can build up a neural representa- for reading to a predominance of left (rela-
tion of information specific to the writing sys- tive to right) ventral occipitotemporal cortex
tem that is most relevant to the task of reading. involvement. This finding was evident in sev-
This specialization in visual processing can eral large cross-sectional developmental fMRI
be reflected across development as a change studies of reading-related tasks (Schlaggar
in relative activation in extrastriate regions, et al. 2002, Shaywitz et al. 2002, Brown
which could play out as a reduction of activ- et al. 2005) described above. Additionally,
ity in some regions, with retention of a sim- as described previously, Eden and cowork-
ilar level of activity, or even greater activity, ers (Turkeltaub et al. 2003) reported a sim-
in other regions. This idea is consistent with ilar pattern in their large cross-sectional de-
reports of changes in brain activity believed velopmental fMRI study that related age and
to occur in adults as they acquire percep- various measures of linguistic skill to activa-
tual expertise for novel items (e.g., Gauthier tion elicited by an implicit reading task. They
2001). showed an age-related decline in right extras-
Indeed, developmental fMRI studies (re- triate activity, whereas homologous left corti-
viewed in detail below) have examined cal regions maintained their level of activity
changes in neural responses in children stud- across age groups.
ied across several ages during the acquisi- Another critical aspect of understanding
tion of reading skill. Studying both reading- the development of function in the VWFA in-
disabled (RD) and nonimpaired (NI) read- volves distinguishing between activity reflect-
ers, Shaywitz and coworkers investigated the ing early perceptual processes and later post-
relationship of BOLD signals to reading perceptual processes, as well as characterizing
skill and age. They describe increased activ- changes that occur as the perception of visual
ity in a left ventral occipitotemporal region words emerges from a slow effortful process
(likely including the VWFA), which corre- to the rapid automatic process that occurs in
lated more robustly with reading skill than the mind of a skilled reader within the span
with age (Sandak et al. 2004). In a two-year of several hundred milliseconds. Because the
longitudinal component of RD children who fMRI signal is based on changes in deoxy-
had received intensive intervention leading to genated hemoglobin that slowly accumulate
improved skill, the putative VWFA showed in seconds after changes in neural activity, this
increased activity over the interval, which sug- measure has an inherent temporal sluggish-
gested the change in regional activity is related ness that limits one’s ability to differentiate ac-
to change in skill level as well as impacted by tivations associated with early perceptual pro-
educational activities (Shaywitz et al. 2004). cesses versus later post-perceptual processes

484 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

and limits one’s ability to trace changes in flu- sponse to visual word forms and the relation-
ency during development. ship between these signals and the rise of fast
Thus investigators must supplement fMRI perceptual specializations for reading (Posner
investigations with event-related potential & McCandliss 1999; Maurer et al. 2005b,
(ERP) measures that are more sensitive to 2006). Brandeis and coworkers (Maurer et al.
the temporal dynamics of brain activity at the 2005b) used this method to contrast responses
ms scale. ERP studies in adults have shown to words and visual control stimuli (e.g. novel
that within 200 ms of viewing a visual word, characters that resembled letters) in children
electrical activity recorded over left posterior just before the onset of reading instruction
inferior regions of skilled readers responds in school. Before the onset of formal school-
differently to visual words versus control stim- ing in reading, children produced a delayed
uli (i.e., strings of novel letter-like characters). N170-like component (peaking at 220 ms) in
This sensitivity, present in the negative deflec- response to both visual words and control vi-
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

tion over posterior-occipito-temporal chan- sual stimuli. Unlike adults and older children,
nels at 170 ms (i.e. the N170 component), however, this ERP response in preschool chil-
is thought to reflect fast perceptual special- dren showed no sensitivity to visual words or
ization for processing visual word forms in letters. Mere familiarity with letters by many
adults (Rossion et al. 2003, Simon et al. 2004, of the kindergarten children was insufficient
Maurer et al. 2005a). Researchers have found to produce this N170-like response. In fact,
similar N170 effects in several other forms the subset of children most familiar with let-
of visual expertise that result from extensive ters produced some modulation of the N170,
training experiences with a particular class of but with a right lateralized pattern. Presum-
stimuli (Tanaka & Curran 2001), although ably, experiences taking place between kinder-
such effects tend to be bilateral. Thus the left garten and adulthood progressively lead to
lateralization of the N170 response to visual functional tuning of the N170, which be-
words has been proposed as a phenomena that comes somewhat specific to the writing sys-
differentiates visual word expertise from other tem in which an individual is trained.
forms of visual expertise effects (Maurer & These same children were followed lon-
McCandliss 2007). gitudinally to examine the impact of ∼1.5
The left-lateralized N170 ERP response years of formal schooling on the sensitivity
has been linked to neural activity within of the N170 to visual word form process-
the left ventral occipitotemporal region, ing (Maurer et al. 2006). Across these years,
which includes the VWFA. A recent joint an N170-like response demonstrated strong
fMRI/ERP study (Brem et al. 2006) showed amplitude differences between the same vi-
systematic correlations between individual sual words and control stimuli they had seen
differences in the magnitude of the N170 for 18 months previously, directly demonstrating
words and the magnitude of the BOLD re- the impact of development and experience on
sponse to words in the VWFA. Furthermore, the response properties of the N170. As dis-
source estimation for word effects localized cussed above, this design does not directly dif-
maximal activity during the N170 to the left ferentiate maturation and learning-related ef-
ventral temporal area (Maurer et al. 2005b). fects, yet the contrast effect between visual
Thus, part of the observed fMRI BOLD re- words and control stimuli suggests that ex-
sponse to visual words may reflect early per- periences with visual words drive changes in
ceptual analysis procedures characteristic of N170 responses that are not observed in the
the expert reader. control stimuli. In addition, individual dif-
Developmental reading studies have used ferences in reading skill (i.e., fluency) were
ERP recordings to examine more directly the systematically correlated with the degree to
experience-dependent nature of the N170 re- which N170 responses came to resemble the

www.annualreviews.org • Development of Reading 485

 ANRV314-NE30-19 ARI 20 May 2007 15:43

adult-like pattern. Another critical issue in of language. The developmental changes de-
considering the development of visual word scribed above in ventral occipitotemporal re-
recognition skills and the underlying changes gions associated with visual word form pro-
in neural processing concerns developments cessing may be accompanied over the course
that occur between adolescence and adult- of reading development by changes in the
hood, as such systems continue to improve dorsal phonological (i.e. perisylvian) regions
in fluency. When the same N170 paradigm consistently implicated as critical to access-
described above was used to contrast ado- ing speech sounds associated with letters (Raij
lescents (i.e., ∼16 years of age) and adults, et al. 2000, Van Atteveldt et al. 2004). Pugh
the adolescent N170 response showed adult- and coworkers (2001) have proposed a model
like topographies in general. However, the of skilled reading acquisition that highlights
latency of the N170 was shorter for adults the importance of phonological systems in the
than for adolescents when visual words were functional development of orthographic re-
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

presented, but not when the control stimuli gions such as the left VWFA. In this scheme,
were presented. These findings provide fur- the novice/child reader implements the dor-
ther support for a role for specific experience sal phonological system to sound out novel or
with visual words driving changes even late in irregular words. With increasing reading ex-
development as fluency in word recognition is perience, however, the dorsal system, via an
progressively refined (Brem et al. 2006). Fur- interactive process, eventually serves to train
thermore, Brem and coworkers demonstrated the ventral visual word form system to rec-
correlations across individuals in this adoles- ognize orthographic patterns related to high
cent age group between word-related N170 frequency and irregular words so that they can
responses and word-related fMRI responses be processed rapidly.
in the VWFA, suggesting further links be- Investigators have also described a sim-
tween late developments in this cortical re- ilar developmental progression within con-
gion and functional changes in fast perceptual nectionist models of reading development.
processes applied to visual words. This approach instantiates explicit computer
These findings converge to support the models that “learn” to relate print to speech
notion that the development of fluent word and meaning. Early in the learning process
recognition is systematically related to func- the predominance of changes within connec-
tional refinements in early perceptual pro- tion weights (analogous to the strength and
cesses. These novel perceptual abilities, which reliability of neural connections) capitalize
are triggered during the first few hundred ms on statistical patterns that systematically re-
of processing a visual word, undergo consider- late orthography to phonology. Later in the
able experience-dependent refinement man- learning process, however, when such reg-
ifest as more focal, left-lateralized patterns ularities are well learned, the predominant
as reading experience develops. This process changes in connection weights occur between
continues through adolescence and into early orthography and semantic units (Seidenberg
adulthood and is linked to the development of 2005). Such models are explanatory such that
efficiency of visual word recognition. they explicitly relate the observed changes
in learning dynamics to differences in the
INTEGRATION OF statistical consistencies between mapping
ORTHOGRAPHIC AND print to speech versus mapping print to
PHONOLOGICAL PROCESSING meaning.
IS IMPORTANT FOR READING Another approach to studying brain activ-
DEVELOPMENT ity associated with the integration of ortho-
Reading ultimately requires expertise in link- graphic and phonological processing involves
ing phonological and orthographical aspects contrasting explicit task demands across

486 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

visual and auditory words. In a study of skilled NEUROSCIENCE APPROACHES

adult readers, Booth and coworkers (2002) TO UNDERSTANDING
presented auditory and visual words within a INDIVIDUAL DIFFERENCES IN
Phoneme: the
rhyming task to focus attention on phonology READING DEVELOPMENT smallest contrastive
versus within a spelling task to focus on or- Recent work investigating brain mechanisms sound unit of spoken
thography. Consistent with previous findings, of reading difficulties may provide an ex-
language
the auditory-rhyming task activated left supe- planatory framework for the wide range of
rior temporal gyrus, and the visual-spelling individual differences evident in the read-
task activated left fusiform. When condi- ing acquisition process and the etiology of
tions required integrating orthography and developmental reading disabilities estimated
phonology, as in the visual-rhyming or the to impact ∼5%–15% of the U.S. population
auditory-spelling condition, performance on (Rutter 1978, Stanovich 1986).
both of these tasks was associated with acti-
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

The prevailing theory of reading disabil-

vation levels in supramarginal/angular gyrus ity, the core phonological deficit hypothesis,
regions in a manner suggesting that these re- holds that subtle language deficits introduce
gions were associated with the integration of difficulties in accessing and manipulating
phonological and orthographic codes. When sounds within speech at the phoneme (i.e.,
this work was extended to children (aged sound elements roughly corresponding to let-
9–12 years), Booth and coworkers (2004) ters) level, which result in atypical processing
found that children demonstrated many of of visual words forms. This view has substan-
the same effects, although tasks that required tial empirical support from cognitive studies,
orthographic and phonological integration and the most commonly observed cognitive
produced stronger activation in the angu- disruption associated with RD in the English
lar gyrus in the adults than in the children. language is the presence of phonological pro-
These results suggest that this form of inte- cessing difficulties such as rhyming skills and
gration continues to develop well after late other tasks that require accessing and manipu-
childhood. Although this interpretation may lating sounds within words (Shankweiler et al.
appear to contradict the prediction of age/ 1999, NRP 2000). (For alternate theories, see
skill-related decreased activity in the dorsal Ramus et al. 2003.) These phonological pro-
system (i.e., Pugh et al. 2001), this work sets cessing deficits may be sufficiently subtle to
the stage for investigation of the relationship allow apparently typical development of au-
between changes in the phonological system, ral/oral aspects of language, but may be sub-
changes in the orthographic system, and in- stantial enough to prevent adequate mapping
tegration across these two systems as it re- of orthography onto phonology, leading to
lates to development and skill acquisition for reading disability.
reading.
Additional empirical work within this do-
main needs to examine such hypotheses di-
rectly in terms of developmental changes in Functional Imaging of Individual
cortical activation associated with phonology, Differences in Reading
orthography, and the integration across these An extensive literature of adult functional
two forms of information. A developmental neuroimaging studies examining the activa-
cognitive neuroscience account of how these tion correlates of RD has emerged (Habib
systems change and come to interact over 2000, McCandliss & Noble 2003). Converg-
the course of typical development may prove ing findings support two central insights into
crucial in constructing frameworks for un- individual differences related to phonologi-
derstanding atypical development of reading cal processing of linguistic information and
abilities. processing of visual word forms. First, studies

www.annualreviews.org • Development of Reading 487

 ANRV314-NE30-19 ARI 20 May 2007 15:43

that target phonological processing skills Several other developmental fMRI stud-
demonstrate atypically small or absent mod- ies on reading disability have been conducted.
ulation of functional activation in left perisyl- Although such studies generally rely on RD
MEG: magnetoen-
cephalography vian regions in RD adults (e.g., Temple 2002). versus NI group comparisons using substan-
Second, in studies that isolate activity associ- tially smaller samples of children, they pro-
ated with visual words versus lower-level con- vide some support for the previously reported
trol conditions, RD adults produce less robust findings of NI > RD activation for left peri-
activation in left ventral occipitotemporal cor- sylvian (Temple 2001, Backes et al. 2002) and
tex (e.g., Brunswick 1999, Paulesu et al. 2001). left ventral occipitotemporal regions (Temple
Progress has been made in extending func- et al. 2001, but see Backes et al. 2002). An ad-
tional neuroimaging studies of developmen- ditional study (n = 17 per group) failed to
tal dyslexia earlier into development, demon- demonstrate group differences in any region
strating some continuity of findings across except for left frontal regions, and such dif-
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

adults, adolescents, and children. The largest ferences were specific to reading-related tasks
cross-sectional developmental fMRI study of (Georgiewa et al. 1999).
reading development reported to date exam- Parallel findings regarding group compar-
ined 70 RD and 74 NI children ranging in isons between RD and NI groups have been
age from 7 to 17 years (Shaywitz et al. 2002). reported in MEG investigations, providing
RD children showed reduced reading-related insights into the time course of these atypical
activity in left ventral occipitotemporal and activations. MEG is closely related to ERP
left perisylvian regions including the mid- methods and shares the same high temporal
temporal and angular gyrus. This pattern of resolution, while providing some advantages
group effects was restricted primarily to these for localizing neural activity. In NI adults,
regions when the activation task involved ac- the contrast between visual words and low-
tively reading words to make semantic deci- level visual control stimuli produces activation
sions versus a control condition, yet similar localized to left ventral occipitotemporal re-
regions were implicated in a pseudoword- gions within ∼150–200 ms. RD adults, how-
rhyming task. Across all subjects and both ever, fail to show such differentiation within
reading tasks, correlation analyses that first re- this timeframe (Helenius et al. 1999), suggest-
gressed out age effects revealed a positive rela- ing that RD affects even the early percep-
tionship between reading skill and activation tual analysis of visual words forms. Later in
in left ventral occipitotemporal regions. Fur- the timeline, ∼200–400 ms after stimulus pre-
thermore, for the semantic task, negative cor- sentation, group differences emerge over left
relations were evident between reading skill posterior superior temporal regions (Salmelin
and activity in right ventral occipitotemporal et al. 1996), demonstrating reduced signal
regions, such that the poorest readers showed strength in RD adults.
an increased tendency to recruit right hemi- MEG studies in typically developing
sphere regions during this task. children demonstrate similar responses in
The finding that skill level and brain activ- left perisylvian regions when reading words
ity correlate, even when age is regressed out, (Simos et al. 2000a) or pseudowords (Simos
suggests that much of the observed individual et al. 2000b). However, children fail to
differences in brain activity may be related to demonstrate the left-lateralized ventral occip-
the emergence of reading skill across this pop- itotemporal response demonstrated by adults
ulation. Group differences between RD and during the first several hundred ms of view-
NI children in these regions may be more in- ing a visual word or pseudoword, suggest-
fluenced by the relationship between cogni- ing that the adult phenomenon emerges
tive skills and activation patterns than by age- during development of reading skill acquisi-
related effects. tion. Contrasts between NI and RD children

488 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

generally demonstrate patterns in left perisyl- reduction in left fusiform activity in RD ado-
vian regions similar to those seen in adults. lescents was not simply an effect of age or
However, RD children demonstrate strong skill; the effect appears to be a characteris-
DTI: diffusion
right perisylvian activity that is not observed tic of RD. A related study of younger RD tensor imaging
in NI children. In an additional experiment children (i.e., fifth graders) contrasted against
FA: fractional
that involved listening to words, RD and NI both typically developing age-matched con- anisotropy
children showed equivalent responses in left trols and even younger reading-matched con-
and right superior temporal gyrus regions, trol groups (i.e., third graders) failed to reveal
suggesting that the observed group differ- differences in left occipitotemporal regions
ences are somewhat specific to reading but (Hoeft et al. 2006). These results support
could be attributable to performance differ- the notion that some aspects of VWFA spe-
ences. cialization may arise quite late in develop-
As reviewed above (see Challenges to ment and further suggest that brain activity
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Studying the Development of Reading- patterns associated with RD cannot simply

Related Tasks Using Functional Neuroimag- be reduced to delays in reading skill. Such
ing), investigations of brain activation pat- designs may prove critical to establishing
terns that differentiate typically developing which aspects of atypical brain activity re-
readers from dyslexic children necessarily ported in RD are associated with delayed
face a multitude of confounding factors re- reading development versus those aspects
lated to the intrinsic relationship between that represent atypical pathways of functional
task performance, reading skill, and brain ac- development.
tivity. By definition, RD readers cannot be
equated on reading skill with same-age, typ-
ically developing children. Furthermore, any Individual Differences in White
reading-related task performed as an activa- Matter Tract Properties
tion paradigm is likely to show some perfor- Properties of long axonal connections be-
mance differences. One recent set of stud- tween regions may systematically account for
ies has attempted to address these issues in individual differences in cognitive skill and the
the investigation of RD by employing two development of skills such as reading. Diffu-
groups of controls: one group that is matched sion tensor imaging (DTI) employs 3D ten-
on chronological age and a second control sors to provide a measure of the degree to
group comprised of younger, typically de- which diffusion is constrained by white mat-
veloping readers matched on reading ability ter tracts, resulting in anisotropic patterns of
(Hoeft et al. 2006, 2007). Across two inno- diffusion. High fractional anisotropy (FA) val-
vative studies following this approach, these ues correspond to voxels dominated by dense
researchers demonstrated that many of the large, well-myelinated axon fibers oriented in
commonly reported differences in activation a coherent direction, and variations in these
patterns between RD and similarly aged NI properties lead to reductions in FA (Basser
groups are also present when contrasting RD et al. 1994).
groups with younger controls matched on Correlations between FA values in left
reading ability. For example, Gabrielli and temporal white matter tracts and reading abil-
coworkers (Hoeft et al. 2007) showed that ity quantified by standardized measures were
RD adolescents demonstrated significant re- first reported in adults ranging from skilled
ductions in left fusiform regions during a vi- readers to poor readers with a history of de-
sual word rhyming task. Furthermore, they velopmental dyslexia (Klingberg et al. 2000).
showed that this effect was present when con- Three developmental studies have since ex-
trasting the RD group to both age-matched amined relationships between FA values and
and reading-matched controls. Therefore, the standardized reading scores in children at ages

www.annualreviews.org • Development of Reading 489

 ANRV314-NE30-19 ARI 20 May 2007 15:43

when persistent reading disabilities are typi- drome but is on a continuum with typical
cally first diagnosed. Two of these studies in- reading.
cluded children predominantly in the aver- The specificity of this structure-function
age to above-average range (Beaulieu et al. relationship was also investigated, both in
2005, Deutsch et al. 2005), and one study re- terms of the degree to which the implicated
cruited children predominantly in the aver- white matter tracts were region-specific ver-
age to impaired range (Niogi & McCandliss sus systemic throughout the brain, as well as
2006). All three studies converged on the cen- the degree to which the cognitive differences
tral finding of a strong positive correlation were specific to the reading domain. Each of
between standardized reading scores and FA the published studies demonstrated a simi-
in a left hemisphere region centered on the lar region in the left hemisphere but showed
superior corona radiata (SCR) at the level of weaker or nonsignificant relationships in the
the corpus callosum. These effects were ro- right homologue. When factors such as IQ
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

bust, accounting for 29%–42% of variance in were controlled for, the results remained sig-
reading skills across studies. The same rela- nificant. Finally, Niogi & McCandliss (2006)
tionship that held between FA and reading demonstrated a “statistical double dissocia-
ability demonstrated in the normal range was tion” between the left SCR reading rela-
also present in the more extreme range of tionship and another structure-function re-
scores, both impaired and above average. This lationship between short-term memory and
structural anatomical contribution to read- a frontal white matter tract structure (see
ing skill is more consistent with the view Figure 2). These findings strongly suggest
that reading disability is not a discrete syn- that reading ability has a specific association

Figure 2
Variability in children’s white matter tract microstructure demonstrates a “statistical double dissociation”
in the patterns of correlation with two cognitive domains. Red sections indicate regions of interest used
for both data selection and seed volumes for fiber tracking for (a) left SCR (blue) and (b) bilateral ACR
(green). Bivariate scatterplots demonstrate standardized scores on X-axes for reading (left) and short-term
memory (right). Y-axes indicate fractional anisotropy (FA) values for SCR (blue) and bilateral ACR (green).
Trend lines indicating significant correlations are presented as solid lines.

490 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

with SCR structure that is independent of at with reading-related skill variance indepen-
least one other high-level linguistic cognitive dent from general intelligence (Schumacher
ability and is not found in other white matter et al. 2006). Multiple genes likely conspire to
regions. modulate susceptibility to dyslexia, and mul-
Although the same general region is im- tiple candidate genes are under investigation.
plicated across multiple studies, researchers Two candidate genes found at 6p22 are linked
have reported some discrepancies regarding to susceptibility to dyslexia. The pattern of ex-
the nature of the associated fiber tract. Two pression of both genes in the mature system
studies that performed fiber-tracking analy- is not restricted to reading-related brain re-
ses (Beaulieu et al. 2005, Niogi & McCandliss gions. Recent work suggests that disruption
2006) converged on the conclusion that the of normal function of either gene during cor-
dominant tract represents a superior-inferior tical development interferes with migration of
tract connecting the left internal capsule with cortical neurons (Meng et al. 2005, McGrath
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

left motor and premotor regions. This conclu- et al. 2006).

sion contrasts earlier speculations that this re- Although all the susceptibility loci identi-
gion implicates the superior longitudinal fas- fied to date likely contribute to the dyslexic
ciculus (Klingberg et al. 2000). Such findings phenotype, the identification of functional
present the new challenge of understanding candidate genes, whose products can poten-
the way in which this left-lateralized white tially be linked to specific influences on the
matter tract is systematically related to read- wiring of the brain, leads to a consideration
ing ability. One possible mechanism is that of how directly these genes may be linked to
subtle motor disruption early in life could al- reading development. Evidence supports the
ter the development of language and reading contention that the component processes of
skills emerging later in life, as suggested by reading (e.g., orthographic skill, phonolog-
recent findings in children genetically at risk ical skill, phonemic awareness, word recog-
for reading disability (Viholainen et al. 2006). nition) are heritable (Gayan & Olson 2001,
Grigorenko 2001). For example, Knopik et al.
(2002), using a twin study approach, employed
Genetic and Cultural Contributions linkage analysis at the dyslexia susceptibility
to Individual Differences in Reading locus on 6p to quantify the heritability of
Development the above component processes. They found
Genetic susceptibility provides another clear evidence for the role of a locus on 6p
source of individual variation in properties that influences orthographic and phonologi-
of the neural systems for reading and has cal skills, as well as phonemic awareness. In ad-
a profound impact on the emergence of dition, the influence on phonemic awareness
reading ability. Evidence for a genetic role appeared stronger for individuals with higher
in developmental dyslexia has been available IQs. These results suggest that a single sus-
for more than 50 years (Halgren 1950). The ceptibility locus is unlikely to independently
identification of multiple dyslexia suscep- influence a single component process of
tibility loci strongly supports the heritable reading.
basis of dyslexia with estimates of heritability Examining the development of differential
ranging from ∼45% to 70%. Recent progress phenotypic expression in genetically at-risk
has identified multiple susceptibility loci groups provides further support for the pre-
with additional ones quite likely to follow vailing view that the influence of genetics on
(McGrath et al. 2006). reading ability is mediated via subtle deficits in
Across studies, the dyslexia susceptibility language development. Scarborough (1990)
locus most reliably identified at present ap- examined the early development of chil-
pears to be DYX2 on 6p22, which is associated dren at familial risk for dyslexia who did

www.annualreviews.org • Development of Reading 491

 ANRV314-NE30-19 ARI 20 May 2007 15:43

or did not go on to demonstrate school-age Although researchers are directing much

reading difficulties and found that affected attention toward identifying genetic risks
children demonstrated differences in speech for reading disability, an often overlooked
accuracy as early as 2.5 years of age and aspect of reading development that must
showed deficits in phonemic awareness and be considered when constructing theories
letter sound knowledge by age 5. relating brain activity to reading function is
The Jyvaskyla Longitudinal Study the fact that writing systems and languages
(Lyytinen et al. 2004) has been following differ from one culture to the next. Ac-
genetically at-risk children from birth to cordingly, the degree to which one might
school age to track the development of the generalize findings across cultures remains a
phenotypic expression of heritable reading crucial empirical issue. Investigators recently
disability. Currently available data suggest identified a cultural effect in a functional
that although many early developmental lan- neuroimaging study of reading pseudowords,
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

guage assays correlate with early reading de- which demonstrated differential activation
velopment, deficiencies in receptive language across a network of reading-related regions
ability at age 3.5 years are strongly associated dependent on whether the subject’s native
with early reading difficulties at age 7 years language for reading was Italian (a shallow or-
for the at-risk group, yet are uncorrelated in thography with predictable spelling-to-sound
the control group. Such studies may provide patterns) or English (a deep orthography
early phenotypic markers for the expression containing less consistent mappings between
of genetic risk for reading disability. letters and sound) (Paulesu et al. 2001).
The Jyvaskyla study also investigated ERP Extending the concept of cultural differences
responses for at-risk and control infants. Pre- in writing systems to nonalphabetic writing
sentation of repeated simple speech sounds systems, the meta-analysis of reading-related
with an occasional deviant stimulus that short- studies performed by Bolger and colleagues
ened the vowel duration resulted in ERP re- (2005) showed that one of the most consistent
sponses to the mismatched stimuli that dif- regions of activation across writing systems is
ferentiated the two groups. Control infants in the left VWFA, and the greatest variability
demonstrated responses predominantly over among activations is in the left phonological
the left hemisphere, and at-risk infants over system. Considering that these results reflect
the right hemisphere, suggesting early hemi- language-dependent phonological demands,
spheric differences in auditory speech pro- these findings fit with a substantial literature
cessing. Furthermore, right hemisphere ERP implicating phonological ability in reading
responses to a more complex speech sound skill acquisition and as a remediation target
pattern (i.e., /ga/) correlated with deficits of (NRP 2000).
receptive language skills at 2.5 years of age for However, Tan et al. (2005) note that this
both groups of infants (Guttorm et al. 2005), emphasis on phonology is based largely on
yet in a larger sample, these responses were Western/alphabetic languages, which engage
more pronounced in the at-risk group (Gut- letter-sound processing. In contrast, logo-
torm et al. 2003). As this at-risk sample is graphic characters in Chinese are associ-
now old enough to start formal instruction ated with elementary aspects of word mean-
in reading, their success in reading develop- ing and place little emphasis on phonology.
ment can be retrospectively linked to their They investigated the relative contributions
ERP responses collected near birth, provid- of phonological and orthographic abilities to
ing a new opportunity to investigate whether Chinese reading development. In contrast to
infant ERP signals contain predictive infor- English, they found a stronger relationship
mation regarding the expression of reading between a child’s reading and character writ-
difficulties in at-risk populations. ing abilities and a weaker relationship between

492 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

their phonological skills and reading abilities. the school year (Shaywitz et al. 2004), and a
Such results emphasize the importance of MEG study spanning 2 months of intensive
orthographic awareness over phonological phonological skill intervention (Simos et al.
awareness as a critical individual difference 2002). Such studies combine the pragmatic
impacting the development of reading in Chi- challenges and attrition rates of two difficult
nese. A recent fMRI study of Chinese reading research enterprises: child neuroimaging
development contrasting activity between RD and educational intervention. Thus, these
and NI in children (Siok et al. 2004) demon- initial studies have been marked by small
strated that reading disabilities were associ- sample sizes, inconsistent outcome measures,
ated not with decreased activity in left poste- and lack of randomized assignment or skill-
rior dorsal or ventral regions as typically seen matched controls. Nonetheless, most of the
in alphabetic writing systems, but rather with reported studies demonstrate an important
decreased activity in a left middle frontal re- first-order principle: Current imaging proto-
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

gion, which the authors attributed to the de- cols for children are sensitive to changes that
mands of coordinating spatial and verbal in- occur on the order of weeks to months. Con-
formation with logographic characters. These sequently, this initial wave of studies provides
studies provide insights into the interaction a form of proof-of-concept, validating the
of cultural-specific features of writing systems potential use of neuroimaging to study the
on the neural systems involved. Furthermore, neural basis of educationally based changes
they highlight the notion that as different (McCandliss & Wolmetz 2004).
writing systems place different computational Critical assessment of these initial stud-
demands on reading development, different ies also raises several concerns regarding
brain systems will likely be engaged. interpretation of observed dynamic changes
in brain activity and its association with a
RESEARCH ON particular learning activity. First, many of the
INTERVENTIONS FOR activation tasks used to elicit neural responses
READING DISABILITY IN are fundamentally related to reading ability.
CHILDREN The associated dynamic changes in brain
Numerous reading intervention studies have activity may be highly sensitive to changes
been reported in the educational and cogni- in both task performance and reading skill,
tive science literature. These studies range making the result difficult to interpret. For
from contrasts between entire programs example, Simos et al. (2002) demonstrated
based on different theoretical approaches that for a particular task, training significantly
to attempts at isolating particular critical improved the performance of RD children
elements of effectiveness (Bus et al. 1999). from their initial pretest level of chance.
Recently, several studies have examined the This scenario is relevant to the perfomance-
impact of reading intervention programs on matching issue discussed above, highlighting
RD children and the changes in brain activity difficulties for making inferences about
before and after intervention, typically span- the functional significance of the observed
ning several months. These studies include changes in neural responses and the added
an fMRI investigation of a computer-based value neuroimaging might provide beyond
auditory-linguistic intervention program the observation of skill and performance
(Fast ForWord Language) spanning 2 gains following intervention.
months (Temple et al. 2003), an ERP study Second, direct comparison of changes in
of a 7-week nonlinguistic auditory training brain activity before and after intervention can
program (Kujala et al. 2001), an fMRI study result in widespread patterns of activity dif-
of a classroom-based intervention spanning ferences throughout multiple brain regions.
14 days (Aylward et al. 2003) or spanning For example, Temple et al. (2003) reported

www.annualreviews.org • Development of Reading 493

 ANRV314-NE30-19 ARI 20 May 2007 15:43

significant changes in 17 regions, some of for coding-specific brain pathways ( Johnson

which incorporated more than 8500 voxels. 2001, Johnson et al. 2002). IS posits that the
Interpretation of the relevance of observing emergence of functional development (i.e.,
change within a single region of interest must the relationship between the neuroanatomi-
therefore be made in the context of large-scale cal and perceptual/cognitive development) of
changes in activity. the brain is driven largely by “changes in
Future intervention studies will need to in- the interactions between several brain regions
troduce additional procedures for establishing that were already partially active.” ( Johnson
the connection between changes in activation 2001, p. 479). The adjustments of interactivity
patterns and changes in skill and performance within and between regions provide the basis
dynamics to allow inferences about processing for a form of systems-level plasticity allow-
to be made. They will also need to contrast al- ing novel computations that support emerg-
ternate intervention protocols to provide the ing behavioral abilities.
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

basis for examining the connection between Applied to reading, IS provides a frame-
an intervention procedure and changes in work for considering how a relatively new cul-
cognitive skills, and the neurological changes tural invention of reading can lead to shaping
that support these skills. Contrasts between a novel pattern of integration across extant
interventions not only provide a control for perceptual/cognitive mechanisms to result in
general effects of participating in an interven- a new behavioral ability. When a child comes
tion, but also provide a means for designing to the task of reading he/she already has
contrasting interventions to contribute to spe- well-established visual processes and linguis-
cific hypothesis testing and theory building. tic processes, yet these systems are not inte-
Such contrasts may take the form of mod- grated in a way that supports reading; that
ified versions of interventions in which the is, there are no specific mappings from par-
theory-driven active ingredient is withheld ticular visual inputs to particular elementary
(e.g., Friel-Patti et al. 2001) or may contrast speech sounds. The demands of learning to
the impact of two interventions using dif- read, played out by an interactive process,
ferent theoretical perspectives (e.g., Pokorni place stress on extant circuits for visual ob-
et al. 2004). ject processing and phonology, leading them
to change in interesting ways. Novel func-
tional connections also form between extant
CONCLUDING REMARKS: circuits to form a new functional circuit, as
INTERACTIVE seen in the case of the multisensory integra-
SPECIALIZATION AS A tion studies of letters and letter-sounds (Van
CONCEPTUAL FRAMEWORK Atteveldt et al. 2004). Furthermore, the ob-
FOR READING DEVELOPMENT servation of progressive lateralization and fo-
A conceptual framework for human brain calization of visual regions may also reflect a
development termed interactive specializa- process of interactive specialization. The de-
tion (IS) is useful for thinking about how velopment of a specialized ability to process
cortical circuits associated with visual and orthography could begin as a general object
linguistic processes change and form spe- recognition problem, utilizing a large array of
cific links in the child’s brain during the processing mechanisms, and integration with
course of learning to read. This emergentist left-lateralized phonological systems could in-
approach to the development of functional crease processing demands on specific forms
specialization may help explain the develop- of visual information, leading to both the left
ment of the rich functional specialization ob- lateralization of the VWFA as well as an in-
served in adults given the relatively limited creased sensitivity to commonly recurring let-
information present in genetic information ter patterns in this system noted above. Thus,

494 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

through reading experience and maturation of ing models of systems-level changes in func-
the underlying neural substrate, the process- tional processes associated with reading de-
ing system could more efficiently utilize and velopment. Such models have been used to
tune the most appropriate mechanisms (i.e., study how learning to read changes repre-
VWFA), eventually decreasing the demands sentations within neural systems dedicated to
on regions that were initially critical (i.e., the representing phonology, orthography, and the
dorsal phonological system). The wide range mappings between them. For example, Harm
of evidence for functional changes in cortical & Seidenberg (1999) presented a model of
regions that occur across the years of reading reading development that allowed learning
development reviewed above is generally con- interactions to take place between neuronal
sistent with this account of changes that oc- populations associated with visual print repre-
cur in functional circuitry over the course of sentations and regions associated with phono-
development. logical representations. Probing the model at
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Another aspect of IS provides a plausi- different points in the learning process al-
ble framework for considering how the child’s lowed these researchers to examine the rise
brain can take advantage of “primitive biases” of functional specializations within these sys-
or mechanisms for perception to attend to tems, as well the development of association
salient environmental stimuli and, in the con- patterns between the spelling and sound sys-
text of experience, to sculpt interregional re- tems. This model also provided an explicit
lationships. A relevant example of a primitive computational account of an interactive spe-
bias is the left hemispheric lateralization of cialization process in which learning to read
preverbal infant speech perception (Dehaene- improved the precision of phonological rep-
Lambertz et al. 2002). If reading builds on ex- resentations.
tant language structures, the tendency for lan- The IS framework, combined with con-
guage processing to be left lateralized could nectionist models of reading development,
represent an important primitive bias for left may prove to be relevant to understanding
hemisphere reading mechanisms to emerge the development of the neural basis of read-
and could explain why a left-lateralized sub- ing disability. Perhaps one of the fundamental
cortical white matter measure of structure challenges to understanding this disabilty in-
(such as DTI) is linked to individual dif- volves creating an integrated neuroscientific
ferences in reading ability. The presence of account of how specific individual differences
evidence of hemispheric biases in typically in properties of the nervous system observed
developing infants, however, does not neces- at the molecular or anatomical level (e.g., dif-
sarily imply the presence of specified circuitry ferences in specific genetic loadings or differ-
for carrying out language function in early ences in white matter microstructure) give rise
infancy. For example, infants with perinatal to the wide range of individual differences ob-
brain injury demonstrate that other brain re- served at the cognitive level, from skilled read-
gions have the capacity to implement the nec- ing to reading disability. Harm and coworkers
essary operations for language and eventu- (Harm et al. 2003, Harm & Seidenberg 1999)
ally implement novel pathways for reading provided a quantitative connectionist model
(Cohen et al. 2004, Fair et al. 2007). that allowed them to investigate how para-
One limitation of the IS framework is that metric manipulations of the neural properties
it fails to provide explicit, quantifiable predic- of the phonological system could lead to dif-
tions regarding behavior or changes in func- ferences in the emergence of reading perfor-
tion associated with particular neural systems. mance, which could be quantified in terms of
This framework, however, is highly consistent standardized reading tests used to assess read-
with connectionist modeling efforts, which do ing development in children. As they paramet-
provide explicit, quantifiable computer learn- rically increased the level of neural deficits in

www.annualreviews.org • Development of Reading 495

 ANRV314-NE30-19 ARI 20 May 2007 15:43

the phonological system prior to learning to employed several cutting-edge imaging tools
read, the model began to exhibit several symp- to provide initial insights into the nature
toms of developmental dyslexia, as assessed by of changes that occur in typically develop-
quantitiative measures. Their analysis of the ing readers as well as the nature of individ-
model over the course of reading development ual differences that help account for reading
demonstrated how differences in phonogi- disabilities. We argue that a deep under-
cal properties of the model led to systematic standing of how the developing child ac-
changes in the association pathways between quires reading skill requires a convergence of
orthography and phonology. Such models are psychological and neuroscientific approaches.
beginning to provide quantifiable predictions Attaining this level of understanding poses
regarding pathways by which individual dif- challenging issues regarding the development
ferences at the level of neural circuitry can of functional specialization, as well as chal-
impact the emergence of reading ability (or lenging interpretational issues regarding how
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

disability). Another promising aspect of this to relate changes in regional brain activity to
approach is that connectionist models pro- changes in reading ability. Separating contri-
vide a means of testing the impact of partic- butions of maturation, overall skill, and in-
ular reading intervention strategies for read- scanner performance on an activation task
ing disabilty, promising novel insights into the presents significant challenges to the field, al-
mechanisms of change associated with effec- though substantial progress has been made in
tive intervention (Harm et al. 2003). approaching these challenges. Furthermore,
An important future direction for this re- combining observations across the relative
search will involve building more explicit links strengths of imaging methods such as fMRI,
between quantifiable aspects of these models MEG, DTI, and ERP provides the basis for
and empirical observations from developmen- more complete theories of online processing
tal neuroimaging studies, such as develop- dynamics, developmental change, and indi-
mental changes and individual differences in vidual differences. Eventually, the field must
specific white matter tracts potentially related develop explicit theories to explain how ex-
to functional connectivity (i.e., inter-regional perience in reading leads to changes in func-
correlated activity measured at rest) between tional organization. Such theories must span
cortical regions, as well as patterns of cortical multiple levels of observation, including ge-
activity related to cognitive processing within netic, anatomical, physiological, and psycho-
particular regions. Such developments may logical, to explain how individual variation at
eventually help create an integrative theory any level can lead to differences in the emer-
of reading development and disability capa- gence of reading ability. Conceptual frame-
ble of relating observations at these different works such as IS and computational accounts
neural and cognitive levels, helping to inform such as connectionist models of reading pro-
remediation efforts. vide some progress toward these goals, but
These are early days in the study of the additional developments that constrain such
development of neural systems for reading. frameworks/models with neurobiological ob-
Nonetheless, investigators have successfully servations are needed.

FUTURE ISSUES
1. Future studies of visual word form processes will elucidate both the development of
perceptual expertise for letter identities and statistical patterns by which letters are
associated with pronunciations and grouped together into visual words.

496 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

2. Cognitive and neural function models need to account for multiple dimensions of
individual differences (e.g., genetic, sociocultural, cognitive) and their interactions in
the development of reading to drive more specific connections between theory and
intervention practice.
3. As training and intervention studies emerge beyond the proof-of-concept phase, such
studies will provide constraints on theories by directly testing intervention principles
in randomized controlled trials that isolate particular variables of interest.
4. Critical data are needed from pre-literate children to separate the influence of indi-
vidual differences on reading from the influence of reading development on individual
differences. This may require methodological advances in functional neuroimaging
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.

approaches to infant and preschool populations.

Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

5. Psychological models and hypotheses concerning the mental computations in read-

ing need to make more direct contact with investigations of functional significance
of activation in particular brain regions. Specifically, connectionist models of brain
activity may be well poised to provide insight into the dynamic relationship between
brain development and skill learning.
6. Functional connectivity MRI (based on very low frequency interregional correlated
activity at rest) may help reveal the development of reading circuits without confounds
of task-related performance differences.

ACKNOWLEDGMENTS
The authors thank Steve Petersen, Jason Zevin, Urs Maurer, Fumiko Hoeft, Vera Blau, Alecia
Vogel, Rebecca Coalson, Jessica Church, and Cathy Yun for helpful discussions and comments
on the manuscript. Both authors are recipients of the John Merck Scholars Fund Award. Support
for this work was also provided by NSF-REC-0337715 (B.D.M.), R01 DC007694 (B.D.M.),
NIH NSADA (B.L.S.), R21 NS41255 (B.L.S.), and the Burroughs Wellcome Fund (B.L.S.).

LITERATURE CITED
Allison T, Puce A, Spencer DD, McCarthy G. 1999. Electrophysiological studies of human
face perception. I: Potentials generated in occipitotemporal cortex by face and non-face
stimuli. Cereb. Cortex 9:415–30
Aylward EH, Richards TL, Berninger VW, Nagy WE, Field KM, et al. 2003. Instructional
treatment associated with changes in brain activation in children with dyslexia. Neurology
61:212–19
Backes W, Vuurman E, Wennekes R, Spronk P, Wuisman M, et al. 2002. Atypical brain
activation of reading processes in children with developmental dyslexia. J. Child Neurol.
17:867–71
Baker C, Liu J, Wald L, Benner T, Kanwisher N. 2005. Experience dependent selectivity for
visual letter strings in English and Hebrew readers. Soc. Neurosci. (Abstr.)
Basser PJ, Mattiello J, LeBihan D. 1994. MR diffusion tensor spectroscopy and imaging.
Biophys. J. 66:259–67
Beaulieu C, Plewes C, Paulson LA, Roy D, Snook L, et al. 2005. Imaging brain connectivity
in children with diverse reading ability. Neuroimage 25:1266–71

www.annualreviews.org • Development of Reading 497

 ANRV314-NE30-19 ARI 20 May 2007 15:43

Binder JR, Medler DA, Westbury CF, Leibenthal E, Buchanan L. 2006. Tuning of the human
left fusiform gyrus to sublexical orthographic structure. Neuroimage 33:739–48
Blau V, van Atteveldt N, Formisano E, Goebel R, Blomert L. 2007. Brain correlates of anoma-
lous letter/speech sound integration in dyslexia. Hum. Brain Mapp. (Abstr)
Bolger DJ, Perfetti CA, Schneider W. 2005. Cross-cultural effect on the brain revisited: uni-
versal structures plus writing system variation. Hum. Brain Mapp. 25:92–104
Booth JR, Burman DD, Meyer JR, Gitelman DR, Parrish TB, et al. 2002. Functional anatomy
of intra- and cross-modal lexical tasks. Neuroimage 16:7–22
Booth JR, Burman DD, Meyer JR, Gitelman DR, Parrish TB, et al. 2004. Development of
brain mechanisms for processing orthographic and phonologic representations. J. Cogn.
Neurosci. 16:1234–49
Booth JR, Burman DD, Meyer JR, Lei Z, Choy J, et al. 2003. Modality-specific and
-independent developmental differences in the neural substrate for lexical processing.
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

J. Neurolinguist. 16:383–405
Booth JR, Burman DD, Van Santen FW, Harasaki Y, Gitelman DR, et al. 2001. The de-
velopment of specialized brain systems in reading and oral-language. Child Neuropsychol.
7:119–41
Brem S, Bucher K, Halder P, Summers P, Dietrich T, et al. 2006. Evidence for developmental
changes in the visual word processing network beyond adolescence. Neuroimage 29:822–37
Brown TT, Lugar HM, Coalson RS, Miezin FM, Petersen SE, et al. 2005. Developmental
changes in human cerebral functional organization for word generation. Cereb. Cortex
15:275–90
Brunswick N, McCrory E, Price CJ, Frith CD, Frith U. 1999. Explicit and implicit processing
of words and pseudowords by adult developmental dyslexics: a search for Wernicke’s
Wortschatz? Brain 122(Pt. 10):1901–17
Bus AG, Ijzendoorn MH. 1999. Phonological awareness and early reading: a meta-analysis of
experimental training studies. J. Ed. Psychol. 91:403–14
Carr TH, Posner MI. 1995. The impact of learning to read on the functional anatomy of
language processing. In Language and Literacy: Comparative Approaches, ed. B de Gelder,
J Morais, pp. 267–94. Cambridge, MA: MIT Press
Casey BJ, Galvan A, Hare TA. 2005. Changes in cerebral functional organization during cog-
nitive development. Curr. Opin. Neurobiol. 15:239–44
Cohen L, Dehaene S, Naccache L, Lehericy S, Dehaene-Lambertz G, et al. 2000. The visual
word form area: spatial and temporal characterization of an initial stage of reading in
normal subjects and posterior split-brain patients. Brain 123(Pt. 2):291–307
Cohen L, Lehericy S, Chochon F, Lemer C, Rivaud S, et al. 2002. Language-specific tuning
of visual cortex? Functional properties of the visual word form area. Brain 125:1054–69
Cohen L, Lehericy S, Henry C, Bourgeois M, Larroque C, et al. 2004. Learning to read
without a left occipital lobe: right-hemispheric shift of visual word form area. Ann. Neurol.
56:890–94
Coltheart M, Curtis B, Atkins P, Haller M. 1993. Models of reading aloud: dual-route and
parallel-distributed-processing approaches. Psychol. Rev. 100:589–608
Dehaene S, Jobert A, Naccache L, Ciuciu P, Poline JB, et al. 2004. Letter binding and invariant
recognition of masked words: behavioral and neuroimaging evidence. Psychol. Sci. 15:307–
13
Dehaene-Lambertz G, Dehaene S, Hertz-Pannier L. 2002. Functional neuroimaging of speech
perception in infants. Science 298:2013–15
Dejerine J. 1891. Sur un cas de cécité verbale avec agraphie, suivi d’autopsie. C. R. Soc. Biol.
4:61–90

498 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

Deutsch GK, Dougherty RF, Bammer R, Siok WT, Gabrieli JD, et al. 2005. Children’s read-
ing performance is correlated with white matter structure measured by diffusion tensor
imaging. Cortex 41:354–63
Devlin JT, Jamison HL, Gonnerman LM, Matthews PM. 2006. The role of the posterior
fusiform gyrus in reading. J. Cogn. Neurosci. 18:911–22
Fair DA, Brown TT, Petersen SE, Schlaggar BL. 2007. fMRI reveals novel functional neu-
roanatomy in a child with perinatal stroke. Neurology 67:2246–49
Fiez JA, Petersen SE. 1998. Neuroimaging studies of word reading. Proc. Natl. Acad. Sci. USA
95:914–21
Friel-Patti S, Frome Loeb D, Gillam RB. 2001. Looking ahead: an introduction to five ex-
ploratory studies of Fast ForWord. Am. J. Speech-Lang. Path. 10:195–202
Gaillard WD, Balsamo LM, Ibrahim Z, Sachs BC, Xu B. 2003a. fMRI identifies regional
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.

specialization of neural networks for reading in young children. Neurology 60:94–100

Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Gaillard WD, Sachs BC, Whitnah JR, Ahmad Z, Balsamo LM, et al. 2003b. Developmental
aspects of language processing: fMRI of verbal fluency in children and adults. Hum. Brain
Mapp. 18:176–85
Gauthier I, Nelson CA. 2001. The development of face expertise. Curr. Opin. Neurobiol. 11:219–
24
Gayan J, Olson RK. 2001. Genetic and environmental influences on orthographic and phono-
logical skills in children with reading disabilities. Dev. Neuropsychol. 20:483–507
Georgiewa P, Rzanny R, Hopf JM, Knab R, Glauche V, et al. 1999. fMRI during word pro-
cessing in dyslexic and normal reading children. Neuroreport 10:3459–65
Grigorenko EL. 2001. Developmental dyslexia: an update on genes, brains, and environments.
J. Child Psychol. Psychiatr. 42:91–125
Grossman M, Cooke A, DeVita C, Alsop D, Detre J, et al. 2002. Age-related changes in working
memory during sentence comprehension: an fMRI study. Neuroimage 15:302–17
Guttorm TK, Leppanen PH, Poikkeus AM, Eklund KM, Lyytinen P, et al. 2005. Brain event-
related potentials (ERPs) measured at birth predict later language development in children
with and without familial risk for dyslexia. Cortex 41:291–303
Guttorm TK, Leppanen PH, Tolvanen A, Lyytinen H. 2003. Event-related potentials in new-
borns with and without familial risk for dyslexia: principal component analysis reveals
differences between the groups. J. Neural. Transm. 110:1059–74
Habib M. 2000. The neurological basis of developmental dyslexia: an overview and working
hypothesis. Brain 123(Pt. 12):2373–99
Halgren B. 1950. Specific dyslexia (congenital word-blindness); a clinical and genetic study.
Acta Psychiatr. Neurol. Suppl. 65:1–287
Harm WM, McCandliss BD, Seidenberg MS. 2003. Modeling the success and failures of
interventions for disabled readers. Sci. Stud. Read. 7:155–82
Harm MW, Seidenberg MS. 1999. Phonology, reading acquisition, and dyslexia: insights from
connectionist models. Psychol. Rev. 106:491–528
Harm MW, Seidenberg MS. 2004. Computing the meanings of words in reading: cooperative
division of labor between visual and phonological processes. Psychol. Rev. 111:662–720
Hashimoto R, Sakai KL. 2004. Learning letters in adulthood: direct visualization of cortical
plasticity for forming a new link between orthography and phonology. Neuron 42:311–22
Helenius P, Tarkiainen A, Cornelissen P, Hansen PC, Salmelin R. 1999. Dissociation of normal
feature analysis and deficient processing of letter-strings in dyslexic adults. Cereb. Cortex
9:476–83

www.annualreviews.org • Development of Reading 499

 ANRV314-NE30-19 ARI 20 May 2007 15:43

Hoeft F, Hernadez A, McMillon G, Taylor-Hill H, Marindale J, et al. 2006. Neural basis of

dyslexia: a comparison between dyslexic and non-dyslexic children equated for reading
ability. J. Neurosci. 26:1–9
Hoeft F, Melyer A, Hernadez A, Juel C, Taylor-Hill H, et al. 2007. Functional and morpho-
metric brain dissociation between dyslexia and reading ability. Proc. Nat. Acad. Sci. USA.
In press
Jobard G, Crivello F, Tzourio-Mazoyer N. 2003. Evaluation of the dual route theory of reading:
a metanalysis of 35 neuroimaging studies. Neuroimage 20:693–712
Johnson MH. 2001. Functional brain development in humans. Nat. Rev. Neurosci. 2:475–83
Johnson MH, Halit H, Grice SJ, Karmiloff-Smith A. 2002. Neuroimaging of typical and
atypical development: a perspective from multiple levels of analysis. Dev. Psychopathol.
14:521–36
Johnson MH, Pennington BF. 1999. Series foreword. In Neurodevelopmental Disorders, ed. H
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Tager-Flusberg, p. ix. Cambridge, MA: MIT Press

Klingberg T, Hedehus M, Temple E, Salz T, Gabrieli JD, et al. 2000. Microstructure of
temporo-parietal white matter as a basis for reading ability: evidence from diffusion tensor
magnetic resonance imaging. Neuron 25:493–500
Knopik VS, Smith SD, Cardon L, Pennington B, Gayan J, et al. 2002. Differential genetic
etiology of reading component processes as a function of IQ. Behav. Genet. 32:181–98
Kujala T, Karma K, Ceponiene R, Belitz S, Turkkila P, et al. 2001. Plastic neural changes and
reading improvement caused by audiovisual training in reading-impaired children. Proc.
Natl. Acad. Sci. USA 98:10509–14
Lavie N, Tsal Y. 1994. Perceptual load as a major determinant of the locus of selection in visual
attention. Percep. Psychophys. 56:183–97
Lyytinen H, Aro M, Eklund K, Erskine J, Guttorm T, et al. 2004. The development of children
at familial risk for dyslexia: birth to early school age. Ann. Dyslexia 54:184–220
Marshall JC, Newcombe F. 1973. Patterns of paralexia: a psycholinguistic approach. J. Psychol-
ing. Res. 2:175–99
Maurer U, Brandeis D, McCandliss BD. 2005a. Fast, visual specialization for reading in English
revealed by the topography of the N170 ERP response. Behav. Brain Funct. 1:13
Maurer U, Brem S, Bucher K, Brandeis D. 2005b. Emerging neurophysiological specialization
for letter strings. J. Cogn. Neurosci. 17:1532–52
Maurer U, Brem S, Kranz F, Bucher K, Benz R, et al. 2006. Coarse neural tuning for print
peaks when children learn to read. Neuroimage 33:749–58
Maurer U, McCandliss BD. 2007. The development of visual expertise for words: the con-
tribution of electrophysiology. In Single-Word Reading: Cognitive, Behavioral and Biological
Perspectives, ed. EL Grigorenko, A Naples. Mahwah, NJ: Erlbaum. In press
McCandliss BD, Cohen L, Dehaene S. 2003. The visual word form area: expertise for reading
in the fusiform gyrus. Trends Cogn. Sci. 7:293–99
McCandliss BD, Noble KG. 2003. The development of reading impairment: a cognitive neu-
roscience model. Ment. Retard. Dev. Disabil. Res. Rev. 9:196–204
McCandliss BD, Wolmetz M. 2004. Developmental psychobiology of reading disability. In
Developmental Psychobiology, ed. BJ Casey, pp. 69–110. Washington, DC: Am. Psychiatr.
McGrath LM, Smith SD, Pennington BF. 2006. Breakthroughs in the search for dyslexia
candidate genes. Trends Mol. Med. 12:333–41
Meng H, Smith SD, Hager K, Held M, Liu J, et al. 2005. DCDC2 is associated with reading
disability and modulates neuronal development in the brain. Proc. Natl. Acad. Sci. USA
102:17053–58

500 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

Murphy K, Garavan H. 2004. Artifactual fMRI group and condition differences driven by
performance confounds. Neuroimage 21:219–28
Natl. Read. Panel. 2000. Teaching Children to Read: An Evidence-Based Assessment of the Scientific
Research Literature on Reading and its Implications for Reading Instruction. Washington, DC:
NICHD
Niogi SN, McCandliss BD. 2006. Left lateralized white matter microstructure accounts for
individual differences in reading ability and disability. Neuropsychologia 44:2178–88
Palmer ED, Brown TT, Petersen SE, Schlaggar BL. 2004. Investigation of the functional
neuroanatomy of single word reading and its development. Sci. Stud. Read. 8:203–23
Paulesu E, Demonet JF, Fazio F, McCrory E, Chanoine V, et al. 2001. Dyslexia: cultural
diversity and biological unity. Science 291:2064–65
Pokorni JL, Worthington CK, Jamison PL. 2004. Phonological awareness intervention: com-
parison of Fast ForWord, Earobics, and LiPS. J. Ed. Res. 97:147–57
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Poldrack RA, Wagner AD, Prull MW, Desmond JE, Glover GH, et al. 1999. Functional
specialization for semantic and phonological processing in the left inferior prefrontal
cortex. Neuroimage 10:15–35
Posner MI, McCandliss BD. 1999. Brain circuitry during reading. In Converging Methods for
Understanding Reading and Dyslexia, ed. R Klein, P McMullen, pp. 305–38. Cambridge,
MA: MIT Press
Posner MI, Petersen SE, Fox PT, Raichle ME. 1988. Localization of cognitive operations in
the human brain. Science 240:1627–31
Price CJ, Devlin JT. 2003. The myth of the visual word form area. Neuroimage 19:473–81
Price CJ, Friston KJ. 1999. Scanning patients with tasks they can perform. Hum. Brain Mapp.
8:102–8
Price CJ, Wise RJS, Frackowiak RSJ. 1996. Demonstrating the implicit processing of visually
presented words and pseudowords. Cereb. Cortex 6:62–70
Pugh KR, Mencl WE, Jenner AR, Katz L, Frost SJ, et al. 2001. Neurobiological studies of
reading and reading disability. J. Commun. Disord. 34:479–92
Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, et al. 1994. Practice-related
changes in human brain functional anatomy during nonmotor learning. Cereb. Cortex 4:8–
26
Raij T, Uutela K, Hari R. 2000. Audiovisual integration of letters in the human brain. Neuron
28:617–25
Ramus F, Rosen S, Dakin SC, Day BL, Castellote JM, et al. 2003. Theories of developmental
dyslexia: insights from a multiple case study of dyslexic adults. Brain 126:841–65
Rossion B, Joyce CA, Cottrell GW, Tarr MJ. 2003. Early lateralization and orientation tuning
for face, word, and object processing in the visual cortex. Neuroimage 20:1609–24
Rumsey JM, Horwitz B, Donohue BC, Nace K, Maisog JM, et al. 1997. Phonological and
orthographic components of word recognition: a PET-rCBF study. Brain 120:739–59
Rutter LD. 1978. More health education needed. Br. Med. J. 1:370
Salmelin R, Service E, Kiesila P, Uutela K, Salonen O. 1996. Impaired visual word processing
in dyslexia revealed with magnetoencephalography. Ann. Neurol. 40:157–62
Sandak R, Mencl WE, Frost SJ, Pugh K. 2004. The neurobiological basis of skillled and
impaired reading: recent findings and new directions. Sci. Stud. Read. 8:273–92
Scarborough HS. 1990. Very early language deficits in dyslexic children. Child Dev. 61:1728–
43
Schlaggar BL, Brown TT, Lugar HM, Visscher KM, Miezin FM, et al. 2002. Functional
neuroanatomical differences between adults and school-age children in the processing of
single words. Science 296:1476–79

www.annualreviews.org • Development of Reading 501

 ANRV314-NE30-19 ARI 20 May 2007 15:43

Schumacher J, Anthoni H, Dahdouh F, Konig IR, Hillmer AM, et al. 2006. Strong genetic
evidence of DCDC2 as a susceptibility gene for dyslexia. Am. J. Hum. Genet. 78:52–
62
Seidenberg MS. 2005. Connectionist models of word reading. Curr. Dir. Psycholog. Sci. 14:238–
42
Shankweiler D, Lundquist E, Katz L, Stuebing KK, Fletcher JM, et al. 1999. Comprehension
and decoding: patterns of association in children with reading difficulties. Sci. Stud. Read.
3:69–94
Share DL. 1995. Phonological recoding and self-teaching: The sine qua non of reading acqui-
sition. Cognition 55:151–218
Shaywitz BA, Shaywitz SE, Blachman BA, Pugh KR, Fulbright RK, et al. 2004. Development of
left occipitotemporal systems for skilled reading in children after a phonologically-based
intervention. Biol. Psychiatr. 55:926–33
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Shaywitz BA, Shaywitz SE, Pugh KR, Mencl WE, Fulbright RK, et al. 2002. Disruption of
posterior brain systems for reading in children with developmental dyslexia. Biol. Psychiatr.
52:101–10
Simon G, Bernard C, Largy P, Lalonde R, Rebai M. 2004. Chronometry of visual word
recognition during passive and lexical decision tasks: an ERP investigation. Int. J. Neurosci.
114:1401–32
Simos PG, Breier JI, Fletcher JM, Foorman BR, Bergman E, et al. 2000a. Brain activation
profiles in dyslexic children during non-word reading: a magnetic source imaging study.
Neurosci. Lett. 290:61–65
Simos PG, Breier JI, Fletcher JM, Foorman BR, Castillo EM, et al. 2002. Brain mechanisms
for reading words and pseudowords: an integrated approach. Cereb. Cortex 12:297–305
Simos PG, Breier JI, Wheless JW, Maggio WW, Fletcher JM, et al. 2000b. Brain mechanisms
for reading: the role of the superior temporal gyrus in word and pseudoword naming.
Neuroreport 11:2443–47
Siok WT, Perfetti CA, Zhen J, Tan LH. 2004. Biological abnormality of impaired reading is
constrained by culture. Nature. 431:71–76
Stanovich KE. 1986. Matthew effects in reading: some consequences of individual differences
in the acquisition of literacy. Reading Res. Q. 21:360–406
Tan LH, Spinks JA, Eden GF, Perfetti CA, Siok WT. 2005. Reading depends on writing, in
Chinese. Proc. Natl. Acad. Sci. USA 102:8781–85
Tanaka JW, Curran T. 2001. A neural basis for expert object recognition. Psychol. Sci. 12:43–47
Temple E. 2002. Brain mechanisms in normal and dyslexic readers. Curr. Opin. Neurobiol.
12:178–83
Temple E, Deutsch GK, Poldrack RA, Miller SL, Tallal P, et al. 2003. Neural deficits in children
with dyslexia ameliorated by behavioral remediation: evidence from functional MRI. Proc.
Natl. Acad. Sci. USA 100:2860–65
Temple E, Poldrack RA, Salidis J, Deutsch GK, Tallal P, et al. 2001. Disrupted neural re-
sponses to phonological and orthographic processing in dyslexic children: an fMRI study.
Neuroreport 12:299–307
Turkeltaub PE, Eden GF, Jones KM, Zeffiro TA. 2002. Meta-analysis of the functional neu-
roanatomy of single-word reading: method and validation. Neuroimage 16:765–80
Turkeltaub PE, Gareau L, Flowers DL, Zeffiro TA, Eden GF. 2003. Development of neural
mechanisms for reading. Nat. Neurosci. 6:767–73
van Atteveldt N, Formisano E, Goebel R, Blomert L. 2004. Integration of letters and speech
sounds in the human brain. Neuron 43:271–82

502 Schlaggar · McCandliss

 ANRV314-NE30-19 ARI 20 May 2007 15:43

Viholainen H, Ahonen T, Lyytinen P, Cantell M, Tolvanen A, et al. 2006. Early motor devel-
opment and later language and reading skills in children at risk of familial dyslexia. Dev.
Med. Child Neurol. 48:367–73
Warrington ET, Shallice T. 1980. Word form dyslexia. Brain 103:99–112
Weinberger DR, Berman KF. 1998. Prefrontal function in schizophrenia: confounds and con-
troversies. In The Prefrontal Cortex: Executive and Cognitive Functions, ed. AC Roberts, L
Weiskrantz, pp. 165–80. New York: Oxford Univ. Press
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

www.annualreviews.org • Development of Reading 503

 AR314-FM ARI 20 May 2007 16:26

Annual Review of
Neuroscience

Contents Volume 30, 2007

by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Information Processing in the Primate Retina: Circuitry and Coding

G.D. Field and E.J. Chichilnisky p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p1
Orbitofrontal Cortex and Its Contribution to Decision-Making
Jonathan D. Wallis p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 31
Fundamental Components of Attention
Eric I. Knudsen p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 57
Anatomical and Physiological Plasticity of Dendritic Spines
Veronica A. Alvarez and Bernardo L. Sabatini p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 79
Visual Perception and Memory: A New View of Medial Temporal
Lobe Function in Primates and Rodents
Elisabeth A. Murray, Timothy J. Bussey, and Lisa M. Saksida p p p p p p p p p p p p p p p p p p p p p p p p 99
The Medial Temporal Lobe and Recognition Memory
H. Eichenbaum, A.P. Yonelinas, and C. Ranganath p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p123
Why Is Wallerian Degeneration in the CNS So Slow?
Mauricio E. Vargas and Ben A. Barres p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p153
The Head Direction Signal: Origins and Sensory-Motor Integration
Jeffrey S. Taube p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p181
Peripheral Regeneration
Zu-Lin Chen, Wei-Ming Yu, and Sidney Strickland p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p209
Neuron-Glial Interactions in Blood-Brain Barrier Formation
Swati Banerjee and Manzoor A. Bhat p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p235
Multiple Dopamine Functions at Different Time Courses
Wolfram Schultz p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p259
Ventral Tegmental Area Neurons in Learned Appetitive Behavior and
Positive Reinforcement
Howard L. Fields, Gregory O. Hjelmstad, Elyssa B. Margolis,
and Saleem M. Nicola p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p289

v
 AR314-FM ARI 20 May 2007 16:26

Copper and Iron Disorders of the Brain

Erik Madsen and Jonathan D. Gitlin p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p317
The Micromachinery of Mechanotransduction in Hair Cells
Melissa A. Vollrath, Kelvin Y. Kwan, and David P. Corey p p p p p p p p p p p p p p p p p p p p p p p p p p p p p339
Neurobiology of Feeding and Energy Expenditure
Qian Gao and Tamas L. Horvath p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p367
Mechanisms that Regulate Establishment, Maintenance, and
Remodeling of Dendritic Fields
Jay Z. Parrish, Kazuo Emoto, Michael D. Kim, and Yuh Nung Jan p p p p p p p p p p p p p p p p p399
by CORNELL UNIVERSITY MEDICAL COLLEGE on 07/11/07. For personal use only.
Annu. Rev. Neurosci. 2007.30:475-503. Downloaded from arjournals.annualreviews.org

Dynamic Aspects of CNS Synapse Formation

A. Kimberley McAllister p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p425
Adhesion Molecules in the Nervous System: Structural Insights into
Function and Diversity
Lawrence Shapiro, James Love, and David R. Colman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p451
Development of Neural Systems for Reading
Bradley L. Schlaggar and Bruce D. McCandliss p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p475
Molecular Architecture of Smell and Taste in Drosophila
Leslie B. Vosshall and Reinhard F. Stocker p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p505
The Neural Basis of Decision Making
Joshua I. Gold and Michael N. Shadlen p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p535
Trinucleotide Repeat Disorders
Harry T. Orr and Huda Y. Zoghbi p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p575

Indexes

Cumulative Index of Contributing Authors, Volumes 21–30 p p p p p p p p p p p p p p p p p p p p p p p p623

Cumulative Index of Chapter Titles, Volumes 21–30 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p627

Errata

An online log of corrections to Annual Review of Neuroscience chapters (if any, 1997
to the present) may be found at http://neuro.annualreviews.org/

vi Contents