Event-related potentials during auditory language processing in
congenitally blind and sighted people
Brigitte Ro
¨
der
a,
*, Frank Ro
¨
sler
a
, Helen J. Neville
b
a
Biological and Experimental Psychology, Philipps-University Marburg, Gutenbergstrasse 18, D-35032 Marburg, Germany
b
Psychology Department, University of Oregon, Eugene, USA
Received 23 June 1999; received in revised form 6 March 2000; accepted 9 March 2000
Abstract
While behavioral studies have documented delayed language acquisition in blind children, other studies have revealed better
speech discrimination abilities for blind than sighted adults. Several brain imaging studies have provided evidence for cortical
reorganization due to visual deprivation but the cerebral organization of language in blind humans is not known yet.
We hypothesized that the increasing specialization of language systems normally observed during development may not take
place to the same degree in blind individuals since posterior visual areas do not receive their adequate input. On the other hand,
we hypothesized that blind people, due to their greater reliance upon the auditory language signal, may process speech faster
than sighted people.
To test these assumptions, event-related potentials were recorded while 11 congenitally blind and 11 sighted adults matched in
age, gender, handedness and education were engaged in a language task. Participants listened to sentences in order to decide
after each sentence if it was meaningful or not.
Incongruous sentence-final words elicited an N400 e€ect in both groups. The N400 e€ect had a left-lateralized fronto-central
scalp distribution in the sighted but a symmetric and broad topography in the blind. Furthermore, the N400 e€ect started earlier
in the blind than in the sighted. Closed class compared to open class sentence middle words elicited a more pronounced late
negativity in the blind than in the sighted.
These results suggest that blind people process auditory language stimuli faster than sighted people and that some language
functions may be reorganized in the blind. 7 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Blindness; Neural plasticity; Language perception; Cross-modal compensation
1. Introduction
Blind people rely much more upon auditory
language functions than sighted people, both in acquir-
ing knowledge about the world and in social situ-
ations. Therefore, one could hypothesize that verbal
skills are better developed in the blind. In fact,
Chomsky (1980, in [1]) argued that blind children due
to their greater need of spoken language should
acquire it faster than their sighted counterparts. This
assumption is, however, in disagreement with the
results of a large number of developmental studies
which showed deviations in the language development
and/or delayed rather than accelerated language acqui-
sition for blind children (e.g., [1,14,36,44,87]). For
example, a delayed acquisition has relatively consist-
ently been reported for the referential vocabulary (e.g.,
pronouns or deictic and locational terms; [45,87]). The
lack of overgeneralizations, on the other side, is a sali-
ent deviation in the language acquisition of blind chil-
dren, which some authors attributed to di€erences in
the underlying categorization schemata caused by the
absence of visual experience [1,45]. Furthermore,
Neuropsychologia 38 (2000) 1482–1502
0028-3932/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved.
PII: S0028-3932(00 )00057 -9
www.elsevier.com/locate/neuropsychologia
* Corresponding author. Tel.: +49-6421-282-3723; fax: +49-6421-
282-8948.
E-mail address: roeder@mailer.uni-marburg.edu (B. Ro
¨
der).

although it is in general assumed that the underlying
concepts represented by words are similar in blind and
sighted individuals [87], some authors speculate that
semantic networks are less elaborated in the blind
[1,64]. It has been hypothesized that these represen-
tational di€erences as well as other deviations during
language acquisition occur as a secondary consequence
of visual deprivation. For example, there is evidence
for a changed communication style between a blind
child and its parents (e.g., attention cannot be caught
and directed by eye contact) [1], which was found to
be characterized by a ‘more restricted’ and a less rich
language input [1]. Warren (1994) [87] calls this situ-
ation an altered ‘sociolinguistic environment’ for blind
children. Furthermore, other authors have speculated
that a less richly elaborated semantic network (that
was made responsible for the lack of overgeneraliza-
tions and the higher proportion of specific concepts
found in blind children as well) could be the conse-
quence of the less ecient way blind children have to
use to acquire knowledge about the world that is hap-
tic-serial rather than visual-parallel [64]. This assump-
tion is supported by the finding that blind children
overextend words on the basis of haptic-kinesthetic
features, i.e., overgeneralizations occur for facts/con-
cepts for which they have rich experience. Neverthe-
less, blind people use visual terms as well. Since it was
assumed that they have no ‘visual’ concepts this
language usage was called ‘verbalism’, i.e., the use of
words for which one has no meaning (Cutsforth, 1932
in [87]). However, the work of Landau and Gleitman
[36] implies that this conclusion may be premature.
They showed that visual verbs are used by blind
people to express actions within their dominant mo-
dality, either in the haptic or the auditory sense [45].
For instance, ‘look’ stands for ‘explore with the domi-
nant sense’ or ‘see you’ is used for ‘meet you’ etc. [36].
In summary, several authors think, like Mills, that
there is ‘‘ . . . no evidence that blindness alone will
necessarily cause language retardation in the clinical
sense’’ [45], and other authors [1] attributed the fact
that blind adults seem to have normal language to the
high plasticity of the human cognitive system. Further-
more, Mills noted that blind people use auditory
language even in more situations than sighted people
do [45] (e.g., ‘talking books’, screen readers for compu-
ters, etc.) which could improve their language percep-
tion skills. In fact, enhanced speech discrimination in
the context of a noisy background has been observed
in blind compared to sighted individuals [50,59].
Because this superiority of the blind was not ac-
companied by a lower absolute threshold in standard
audiometric tests, it was speculated that it was a result
of a more ecient use or improved high-order proces-
sing of the language input [26,50,59]. In general, the
auditory speech signal is less redundant for blind than
for sighted people who in every-day conversation
usually have access to additional visual cues, e.g., lip
movements, facial expressions etc. These facial cues
can enhance speech intelligibility in sighted people,
particularly in situations with a degraded signal-to-
noise ratio (e.g., [8]). Thus, given the assumption that
blind people can use auditory information more e-
ciently (what was called perceptual rather than sensory
compensation [50]), it seems not too surprising that
blind people score higher in situations where a
degraded auditory signal but no facial information is
available. There is, in fact, some evidence from beha-
vioral and neurophysiological studies that simple
sounds like tones are processed faster by blind com-
pared to sighted individuals [59,69]. The question
arises whether these enhanced auditory-perceptual
skills transfer to and result in improved ‘higher’ cogni-
tive skills. For example, it has been proposed that
specific language impairments in children originate
from elementary auditory deficits rather than specific
language or cognitive processing problems [43,81,90].
Likewise, the intact language of children with the so-
called Williams-syndrome (despite their general retar-
dation) has been hypothesized to be linked to their
hypersensitivity to sounds [57]. Thus, it is conceivable
that improved auditory skills transfer to and improve
language perception in the blind. In fact, enhanced
voice recognition and memory for voices has been
reported for blind compared to sighted people [7,67].
Furthermore, enhanced auditory localization abilities
in the blind have been found by several authors
[50,66,71]. Therefore, given some changes at the audi-
tory-perceptual level, in the blind it is reasonable to
hypothesize that language processing and/or its cer-
ebral organization is altered in the blind as well. How-
ever, linguistic functions proper (e.g., phonology,
semantics and syntax), although often studied in blind
children, have hardly ever been investigated in blind
adults.
1.1. Event-related potentials to study language
In the past years there has been an increasing
number of studies using event-related potentials
(ERPs) to gain insight into both cerebral organiz-
ation and the timing of brain systems engaged in
language processing in general. ERPs are positive
and negative potential deflections, which can be
noninvasively picked up from the scalp. They reflect
a relative activation (or deactivation) of neural cor-
tical networks and are particularly useful to study
fast cognitive processes like language because they
can be recorded while a participant is engaged in
language comprehension without introducing ad-
B. Ro
¨
der et al. / Neuropsychologia 38 (2000) 1482–1502 1483