Neuropsychologia. 2015 Mar;69:22-30. doi: 10.1016/j.neuropsychologia.2015.01.026.

Attention that covers letters is necessary for the left-lateralization of an early print-tuned ERP in Japanese hiragana.

Okumura Y1, Kasai T2, Murohashi H2.
  • 1Graduate School of Education, Hokkaido University, Kita 11 Nishi 7 Kita-ku, Sapporo 060-0811, Japan; Department of Developmental Disorders, National Institute of Mental Health, National Center of Neurology and Psychiatry, 4-1-1 Ogawahigashi-cho, Kodaira, Tokyo 187-8553, Japan. Electronic address:
  • 2Faculty of Education, Hokkaido University, Kita 11 Nishi 7 Kita-ku, Sapporo 060-0811, Japan.



Extensive experience with reading develops expertise in acquiring information from print, and this is reflected in specific enhancement of the left-lateralized N170 component in event-related potentials. The N170 is generally considered to reflect visual/orthographic processing; while modulations of its left-lateralization related to phonological processes have also been indicated. However, in our previous study, N170-like response to Hiragana strings lacked left-lateralization when the stimuli were completely task-irrelevant in rapid-presentation sequences [Okumura et al. (2014). Early print-tuned ERP response with minimal involvement of linguistic processing in Japanese Hiragana strings. Neuroreport 25, 410-414]. This suggests that, despite the highly transparent character-to-syllable correspondence, the phonological mapping of Hiragana strings requires some kind of attention toward print. To verify this notion, the present study examined ERPs under the same experimental condition as in the previous study, except that the task required attention to a stimulus attribute (i.e., color). As a result, Hiragana words and nonwords elicited left-lateralized negative deflection in the occipito-temporal region during 130-170ms post-stimulus in comparison to symbol strings, but only when the print had a narrow intercharacter spacing. Moreover, we observed the enhancement of very early occipital ERP in response to words during 70-100ms. The present results suggest that visual attention plays a role in early print processing, which may contribute to our understanding of the mechanisms that underlie expert as well as impaired reading.

KEYWORDS: Event-related potentials; Left-lateralization; N170; Print processing; Visual attention

PMID: 25613647



Background    Reading is an ordinary activity that many of adults can perform without apparent efforts. When words come into our sight, we instantly get to their sound and meaning as if we recognize them “automatically.” But to what extent the visual word processing is automatic? Does merely seeing words, or mere projection of those onto a retina, are sufficient to trigger reading-related processes without any intention? The automaticity of visual word processing has been a hot topic in studies of reading, which have provided wide variety of evidences to support the above notion.

As a possible neural manifestation of automatic reading-related processes, the present study focused on print-tuned N170 component in event-related brain potentials (ERPs). It is a visual-evoked posterior negativity that peaks at around 140-200 ms poststimulus, which shows specific enhancement to visual words and word-like stimuli (i.e., prints) against non-linguistic visual stimuli (e.g., symbols and figures) predominantly over the left occipito-temporal regions (1). Notably, the specificity to print does not require participants’ to read them; more specifically, the print-tuned N170 can be observed under non-linguistic tasks that involve only simple physical features of print like size and color (1, 2). Furthermore, this component has been shown to appear in association with reading development (3). Therefore, the print-tuned N170 indicates that the identification of print input and the onset of left-lateralized reading processes, perhaps early letter-to-sound mapping (phonological mapping; 4), could occur automatically in skilled readers.

Findings Prior to the Present Study   Prior to the present study, we have conducted a study to verify if the processes reflected in the print-tuned N170 are truly automatic (5). While many studies have observed it under non-linguistic tasks, none have examined whether print stimuli that are completely task-irrelevant could also elicit it. In addition, the study used a presentation rate that was much faster than any other studies to further limit a chance for prints to be processed intentionally. Participants were presented with rapid stream (i.e., 2-3 stimuli per second) of Hiragana words, nonwords and alphanumeric symbol strings and asked to detect occasional changes in the color of fixation cross. As a result, while the print-specific enhancement of N170 was successfully replicated, it was distributed equally over the left and right hemispheres. Since left-lateralized N170 has been observed for Hiragana strings in other studies (6, 7), it was concluded that the lack of left-lateralization may be due to the ignorance of the print stimuli, or not directing attention toward them. In other words, our previous study raised a possibility that print-tuned N170 may involve reading-related process(es) that are not necessary automatic.

The Present Findings  To confirm this possibility, the present study examined print-tuned N170 for rapidly-presented and task-relevant Hiragana strings. Participants were asked to detect occasional presentation of Hiragana of symbols strings in blue fonts, so now they must direct their attention toward prints. As a result, Hiragana words and nonwords elicited left-lateralized N170, which confirmed that at least a part of early reading-related processes require attention toward prints. Interestingly, the left-lateralization effect was observed only for Hiragana strings with regular interletter spacing, which suggested that prints must fall within the spotlight of, or  “covered” by, attention in order to trigger those processes. Furthermore, we surprisingly identified a novel word-specific enhancement of an occipital negativity before 100 ms poststimulus, which was not observed in our previous study. All these results indicated that at least some of the reading-related process in early stages of print processing depends on attention.

Significance and Clinical Implications  The present study indicated that visual word recognition should not be fully automatic, even in skilled readers. Although relatively lower-order processes (e.g., identification of print input) were shown to be accomplished automatically without attention, it was suggested to play a critical role in higher-order processes such as phonological mapping. Based on this notion, it can be predicted that difficulty in reading could arise from deficits in attentional functions. Indeed, visual attention deficit has been indicated as one possible cause of developmental dyslexia (8), a specific learning disorder that is characterized by unexpectedly low reading and spelling abilities despite the average intelligence and adequate learning environment. Therefore, further clarification of the roles that attention play in visual word processing should contribute to better understanding of mechanisms that underlie reading impairments. For instance, high prevalence of dyslexia in attention deficit hyperactivity disorder, another common developmental disorder, may be accounted for by clarifying the nature of their attentional impairments and specific influence of those on their reading. To conclude, seeking for the “non-automatic” aspects of reading should lead to further insights into this complex activity, which in turn would improve ways to deal with deficits in reading.




Figure 1. Schematic illustrations of the relationship between attention to print and N170 distribution for Japanese Hiragana Strings. A) Print-tuned N170 lacked left-lateralization when participants were paying attention to the fixation cross (for details, see 5). B) Attending to the stimulus resulted in the emergence of left-lateralized print-tuned N170 only when letters were normally spaced (present study). Illustrations of spotlights indicate expected focus of attention.



1         Bentin S, Mouchetant-Rostaing Y, Giard MH, Echallier JF, Pernier J 1999 ERP manifestations of processing printed words at different psycholinguistic levels: Time course and scalp distribution. J Cogn Neurosci 11: 235-60.

2         Lin SE, Chen HC, Zhao J, Li S, He S, Weng XC 2011 Left-lateralized N170 response to unpronounceable pseudo but not false Chinese characters-the key role of orthography. Neuroscience 190:200-206.

3         Maurer U, Brem S, Kranz F, Bucher K, Benz R, Halder P, Steinhausen H-C, Brandeis D. 2006 Coarse neural tuning for print peaks when children learn to read. Neuroimage 33:749-758.

4         Maurer U, McCandliss B 2007 The development of visual expertise for words: The contribution of electrophysiology. In: Grigorenko EL, Naples AJ, editors. Single-word Read. Biol. Behav. Perspect., Mahwah, NJ: Lawrence Erlbaum Associates; p. 43–64.

5         Okumura Y, Kasai T, Murohashi H 2014 Early print-tuned ERP response with minimal involvement of linguistic processing in Japanese Hiragana strings. Neuroreport 25:410-414.

6         Maurer U, Zevin JD, McCandliss BD 2008 Left-lateralized N170 effects of visual expertise in reading: evidence from Japanese syllabic and logographic scripts. J Cogn Neurosci 20:1878–91.

7         Horie S, Yamasaki T, Okamoto T, Nakashima T, Ogata K, Tobimatsu S 2012 Differential roles of spatial frequency on reading processes for ideograms and phonograms : A high-density ERP study. Neurosci Res 72:68–78.

8         Vidyasagar TR, Pammer K 2010 Dyslexia: a deficit in visuo-spatial attention, not in phonological processing. Trends Cogn Sci 14:57–63.




Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann