Elsevier

Brain and Language

Volume 86, Issue 2, August 2003, Pages 272-286
Brain and Language

Cortical localisation of the visual and auditory word form areas: A reconsideration of the evidence

https://doi.org/10.1016/S0093-934X(02)00544-8Get rights and content

Abstract

In this paper we examine the evidence for human brain areas dedicated to visual or auditory word form processing by comparing cortical activation for auditory word repetition, reading, picture naming, and environmental sound naming. Both reading and auditory word repetition activated left lateralised regions in the frontal operculum (Broca’s area), posterior superior temporal gyrus (Wernicke’s area), posterior inferior temporal cortex, and a region in the mid superior temporal sulcus relative to baseline conditions that controlled for sensory input and motor output processing. In addition, auditory word repetition increased activation in a lateral region of the left mid superior temporal gyrus but critically, this area is not specific to auditory word processing, it is also activated in response to environmental sounds. There were no reading specific activations, even in the areas previously claimed as visual word form areas: activations were either common to reading and auditory word repetition or common to reading and picture naming. We conclude that there is no current evidence for cortical sites dedicated to visual or auditory word form processing.

Introduction

In this paper, we review the evidence for human brain areas that are specialised for auditory and visual word form processing and then present a new functional imaging study that tests the various claims by comparing cortical activation for auditory word repetition and reading. The first anatomical models of word processing were based on data from post-mortem studies of patients with localised cortical lesions and specific language deficits (see Geshwind, 1965). The left posterior superior temporal cortex (Wernicke’s area) was designated the site of “auditory images of speech” and the left angular gyrus was designated as a “visual memory center for words” (Dejerine, 1891, Dejerine, 1892). According to the neurological model (see Geshwind, 1965), reading involved visual processing in occipital cortices, followed by word identification in the left angular gyrus, conversion to auditory word forms in Wernicke’s area, and speech output in the left posterior inferior frontal cortex (Broca’s area). Thus, the neurological model presupposed areas that were specific to word form processing and separate from visual or auditory processing in general. Auditory word form processing was localised to Wernicke’s area and visual word form processing was localised to the left angular gyrus. Some cognitive models of visual and auditory word form processing also propose word form modules (e.g., the auditory and visual input lexicons in the model by Patterson & Shewell, 1987). However, these models do not make any predictions as to how word form processing occurs at the neural level; they are consistent either with the existence of modular word form areas (as proposed by the neurological model) or the function of the word form lexicons arising from distributed neural processing.

The concept of modular word form areas has been pursued in several functional imaging studies (Cohen et al., 2000; Howard et al., 1992; Petersen, Fox, Posner, Mintum, & Raichle, 1989). Most functional neuroimaging studies of auditory word repetition (Giraud & Price, 2001; Howard et al., 1992; Petersen, Fox, Posner, Mintum, & Raichle, 1988a; Petersen, Fox, Posner, Mintum, & Raichle, 1988b; Price et al., 1996a) have reported reliable activation in the left posterior superior temporal cortex that is likely to correspond to Wernicke’s area. However, activation in the left angular gyrus is not consistently seen during reading aloud as would be expected if this region corresponded to a visual word form area. This has led to three distinct and contrasting proposals regarding the cortical localisation of the visual word form area (Cohen et al., 2000; Howard et al., 1992; Petersen, Fox, Synder, & Raichle, 1990). The evidence for each claim will be discussed in turn below.

The first imaging study of reading, reported by Petersen et al. (1990) proposed that the left medial extrastriate cortex housed the visual word form system because (i) this area was more active for viewing words and pseudowords relative to fixation but not for viewing consonant letter strings or falsefonts relative to fixation; and (ii) it is consistent with the left occipital lesions seen in alexia without agraphia. However, other studies have shown that the same left extrastriate region is activated by pictures of objects and other complex visual stimuli (e.g., Bookheimer, Zeffiro, Blaxton, Gaillard, & Theodore, 1995; Indefrey et al., 1997; Moore & Price, 1999). Therefore, left medial extrastriate activation is not specific to orthographic input and may reflect early visual processing that is modulated by the demands of the task (Mechelli, Humphreys, Mayall, Olson, & Price, 2000).

The second hypothetical visual word form area to be advocated on the basis of functional imaging data was proposed by Howard et al. (1992). They compared brain activation for (1) reading aloud relative to viewing falsefonts and saying “Crime”; and (2) auditory word repetition relative to hearing reversed words and saying “Crime.” Reading activated a left posterior middle temporal area that was not observed for auditory word repetition. This area was linked to the visual input lexicon (another term for the visual word form system). However, the difference between reading and auditory word repetition in the Howard et al. (1992) study was not significant. Therefore there was no evidence to suggest that this area was more involved in visual word form processing than auditory word form processing.

The third area to be linked to the visual word form system is the left posterior inferior temporal cortex (Cohen et al., 2000). This group contrasted activation for reading words presented in the left and right visual hemi-fields and found that irrespective of stimulus presentation, reading (relative to fixation) activated a left lateralised posterior inferior temporal area that is often damaged in patients with alexia without agraphia.1 However, other imaging studies (Moore & Price, 1999; Price & Friston, 1997b) have found that the same left inferior temporal area is activated by picture, letter and colour naming which do not involve visual word form input. Lesion and electrical stimulation studies also indicate that the left posterior inferior temporal cortex is not specific to word form analysis. For instance, the linguistic role of the inferior temporal cortex was first noted in electrical stimulation studies (see Burnstine et al., 1990; Lüders et al., 1986, Lüders et al., 1991) and the surrounding area was referred to as the “basal temporal language area.” Furthermore, although lesions that selectively damage the left posterior inferior temporal cortices are rare, Raymer, Foundas, and Maher (1997) describe a patient with acute damage to the left posterior inferior temporal lobe who suffered from anomia with impaired picture naming and naming to definition as well as reading. Thus, although the left posterior inferior temporal area is clearly involved in visual word processing, activation may reflect semantic and phonological processes that are not specific to reading.

In summary, three left lateralised regions (in the medial extrastriate, posterior middle temporal, and posterior inferior temporal cortices) have been linked to visual word form processing. However, other functional imaging studies show that the same areas also respond to non-lexical stimuli. Activation in the putative visual word form areas may therefore be evoked by (i) processes other than visual word form processing (e.g., early visual, semantic or phonological processes that are not specific to reading although they are required by reading); (ii) processes in addition to visual word form processing (i.e., the same area has different functions either because it has overlapping neural populations or because the function depends on activation in other distributed areas); or (iii) word form processing only but in close proximity to other regions that sustain other functions (i.e., segregation of word form areas will depend on techniques with higher spatial resolution).

Evaluating these hypotheses is not possible in a single imaging experiment. In this paper we attempt to replicate previous findings. In addition to comparing cortical activation for auditory word repetition and reading, the generalisability and specificity of the results was tested by including picture naming and environmental sound naming into the same design. Thirteen subjects were scanned during both auditory word repetition and reading; six subjects were scanned during auditory word repetition and environmental sound naming (full study reported in Giraud & Price, 2001); and eight subjects were scanned during reading and picture naming (full study reported in Moore & Price, 1999). The statistical analysis was then able to identify common areas for auditory word repetition and reading, and areas that were specific to auditory or visual word form processing.

We predicted that (1) both auditory word repetition and reading would activate the left posterior inferior frontal cortex (Broca’s area) and the left dorsal superior temporal gyrus (Wernicke’s area) and (2) auditory word repetition would specifically activate the superior temporal sulcus. With respect to reading, our regions of interest were the left medial extrastriate and posterior middle and inferior temporal cortices, as proposed by Petersen et al. (1990), Howard et al. (1992), and Cohen et al. (2000).

Section snippets

Methods

The data from three different PET experiments were analysed with a single multi-study, multi-factorial statistical model.2 All the subjects participated in 12 scans at the Wellcome Department of Cognitive Neurology on the same ECAT EXACT HR+ (model 962) PET scanner (Siemens/CTI, Knoxville, TN, USA) with collimating septa retracted; and all aspects of image processing were

Activation common to auditory word repetition and reading

As predicted, there was common activation for auditory word repetition and reading in the left posterior frontal cortex (Broca’s area) and the left posterior superior temporal cortex (Wernicke’s area). In addition, both auditory word repetition and reading activated the left posterior inferior temporal cortex (LPITC) and a central region of the left superior temporal sulcus. See part A of Table 1 for details of the anatomical localisation and Z scores. Fig. 1 illustrates the activations in

Discussion

The aim of this study was to evaluate the evidence for cortical sites associated with the auditory and visual word form systems. To this end, we compared activations elicited by reading and auditory word repetition and identified the neural regions (i) common to both reading and auditory word repetition, (ii) more active for auditory word repetition than reading, (iii) more active for reading than auditory word repetition. By including statistical comparisons between naming words, pictures, and

Acknowledgements

This work was supported by the Wellcome Trust.

References (43)

  • J.R. Binder et al.

    Human temporal lobe activation by speech and nonspeech sounds

    Cerebral Cortex

    (2000)
  • A. Bischoff-Grethe et al.

    Conscious and unconscious processing of nonverbal predictability in Wernicke’s area

    Journal of Neuroscience

    (2000)
  • S.Y. Bookheimer et al.

    Regional cerebral blood flow during object naming and word reading

    Human Brain Mapping

    (1995)
  • L. Cohen et al.

    The visual word form area: Spatial and temporal characterization of an initial stage of reading in normal subjects and posterior split-brain patients

    Brain

    (2000)
  • E. De Renzi et al.

    The pattern of neuropsychological impairment associated with left posterior cerebral artery infarcts

    Brain

    (1987)
  • J. Dejerine

    Sur un cas de cecite verbale avec agraphie, suivi d’autopsie

    Memoires Societe Biologique

    (1891)
  • J. Dejerine

    Contribution a l’etude anatomoclinique et clinique des differentes varietes de cecite verbal

    CR Hebdomadaire des Sceances et Memories de la Societe de Biologie

    (1892)
  • N.F. Dronkers

    A new brain region for co-ordinating speech articulation

    Nature

    (1996)
  • A. Foundas et al.

    Anomia: case studies with lesion localisation

    Neurocase

    (1998)
  • K.J. Friston et al.

    Statistical parametric maps in functional imaging: A general linear approach

    Human Brain Mapping

    (1995)
  • K.J. Friston et al.

    Spatial realignment and normalization of images

    Human Brain Mapping

    (1996)
  • Cited by (113)

    • Differential activation of the visual word form area during auditory phoneme perception in youth with dyslexia

      2020, Neuropsychologia
      Citation Excerpt :

      Support for the possibility of a developmental change in the recruitment of the left vOTC for auditory language tasks comes from findings of greater involvement of this region during both visual and auditory word processing tasks in children but only during the visual tasks in adults (Booth et al., 2001; Church et al., 2008). Of note, greater activation in this area for auditory words relative to nonspeech stimuli has also been reported in adults using tasks such as rhyming (Yoncheva et al., 2010), one-back matching (Ludersdorfer et al., 2013), and repetition (Price et al., 2003); however, in these studies, the differential activation across tasks actually reflected less de-activation relative to baseline in the word conditions than in the nonspeech conditions. Complex auditory tasks or selective attention to the auditory modality have been associated with suppression of activation in visual regions (Laurienti et al., 2002; Mozolic et al., 2008).

    View all citing articles on Scopus
    View full text