Measuring word recognition in reading: eye movements and event-related potentials

https://doi.org/10.1016/j.tics.2003.09.010Get rights and content

Abstract

The investigation of visual word recognition has been a major accomplishment of cognitive science. Two on-line methodologies, eye movements and event-related potentials, stand out in the search for the holy grail – an absolute time measure of when, how and why we recognize visual words while reading. Although each technique has its own experimental limitations, we suggest, by means of review and comparison, that these two methodologies can be used in complementary ways to produce a better picture of the mental action we call reading.

Section snippets

Eye movements (EMs)

An important contribution to understanding on-line visual language processing has come from EM research where reading behavior is measured via the position, duration, and sequence of eye fixations in text. The EM technique offers certain advantages over traditional behavioral techniques. EMs are a normal part of reading. First, subjects do not have to make decisions about words they read, or name them aloud – procedures that disrupt the flow of reading when used as a secondary task. Second, the

Event-related potentials (ERPs)

ERPs are stimulus-locked averages of the electroencephalogram (EEG) across many presentations of stimuli. They provide a continuous millisecond-by-millisecond record of electrical changes related to on-going perceptual and cognitive processing and can thus index changes related to word recognition in real time 9, 10. ERPs recorded at the scalp via high-density (e.g. 128-channel) sensor arrays provide a measure of processing with fine-grained (external) spatial as well as temporal resolution [11]

Comparisons of EMs and ERPs

Eye fixation time in reading sets constraints for processing but without sufficient temporal resolution. ERPs can provide greater resolution because they are time-locked on a millisecond-by-millisecond basis to brain events during the presentation of a word. For ERP research to advance knowledge about the timing of visual word recognition, however, it must show sensitivity within earlier components. Early, exogenous ERP components have not readily exhibited processing differences arising from

The way forward

EM research, guided by the temporal contours of the reading task, has introduced innovations that use eye contingent display changes – such as the ‘moving window’ [30], ‘boundary’ [7], ‘fast priming’ [31], and ‘disappearing text’ [32] techniques – to explore the intricacies of reading. In ERP research, the increased use of dense-mapping arrays, the application of statistical techniques such as principal component analysis 20, 24, 33 and independent component analysis [34], corroboration from

Acknowledgements

We thank Patrick O'Donnell, Jeffrey Bowers, and the anonymous reviewers for their helpful comments. Preparation of this article was supported by a British Academy grant 30315 to SCS and grants HD17246 and HD26765 to KR.

References (48)

  • S.P. Liversedge et al.

    Saccadic eye movements and cognition

    Trends Cogn. Sci.

    (2000)
  • K. Rayner

    Eye movements in reading and information processing: 20 years of research

    Psychol. Bull.

    (1998)
  • K. Rayner et al.

    Eye movements in reading: psycholinguistic studies

  • H.E. Blanchard

    The acquisition of parafoveal word information in reading

    Percept. Psychophys.

    (1989)
  • S.C. Sereno et al.

    Spelling-sound regularity effects on eye fixations in reading

    Percept. Psychophys.

    (2000)
  • T.W. Picton

    Guidelines for using human event-related potentials to study cognition: recording standards and publication criteria

    Psychophysiology

    (2000)
  • R. Srinivasan

    Estimating the spatial Nyquist of the human EEG

    Behav. Res. Methods Instrum. Comput.

    (1998)
  • M. Kutas et al.

    Psycholinguistics electrified

  • A. Martı́nez

    Involvement of striate and extrastriate visual cortical areas in spatial attention

    Nat. Neurosci.

    (1999)
  • S. Thorpe

    Speed of processing in the human visual system

    Nature

    (1996)
  • J.J. Foxe et al.

    Flow of activation from V1 to frontal cortex in humans: a framework for defining ‘early’ visual processing

    Exp. Brain Res.

    (2002)
  • C.E. Schroeder

    A spatiotemporal profile of visual system activation revealed by current source density analysis in the awake macaque

    Cereb. Cortex

    (1998)
  • H. Buchner

    Fast visual evoked potential input into human area V5

    Neuroreport

    (1997)
  • Cited by (342)

    • Lateralization of early orthographic processing during natural reading is impaired in developmental dyslexia

      2022, NeuroImage
      Citation Excerpt :

      Accordingly, earlier findings obtained for fixed-gaze experiments (Helenius, 1999; Mahé et al., 2013, 2012; Salmelin et al., 1996) seem to translate well to ecologically valid reading, early processing of orthographic information is left-lateralized in control readers also during natural reading and this lateralization is impaired in dyslexics. However, while fixed-gaze adult results indicate weaker lateralization of orthographic processing in dyslexics due to weaker brain activity in the left occipito-temporal region (Brambati et al., 2006; Helenius, 1999; Mahé et al., 2013, 2012; Paulesu et al., 2001; Paz-Alonso et al., 2018; Richlan et al., 2011, 2009; Salmelin et al., 1996), our results suggest impaired lateralization in dyslexics due to bilateral FREA deflection of a magnitude comparable to the left occipito-temporal FREA amplitude of control readers, a finding that is in agreement with fixed-gaze EEG results obtained for children (Fraga González et al., 2016, 2014). Stronger FREA deflection obtained for dyslexics in the right occipito-temporal region might reflect activation of right-hemispheric compensatory mechanisms in subjects with reading difficulties (Cohen et al., 2004, 2003; Liu et al., 2021; Shaywitz and Shaywitz, 2008; Waldie et al., 2013).

    View all citing articles on Scopus
    View full text