September 2016: Visual Word Identification and Attention

Overview

Besner et al. (2016) offer their view of the role that attention plays in reading behavior, using examples from many of their past individual studies with adults. The authors argue that reading is not an automatic process and that three distinct types of attention (spatial, central, executive) are important for word identification.

Background

This article by Besner et al. (2016) focuses on the role of three cognitive aspects of attention in the reading process through the lens of their experimental studies with typical adult readers. Therefore, by way of background, the information below helps frame the narrow focus of this view in a context and literature that is potentially more familiar to educators.

At the behavioral level, it has long been known that difficulties with reading (e.g., reading disability or dyslexia) and difficulties with attention (e.g., attention deficit hyperactivity disorder) frequently co-occur. Many studies have evaluated the extent of this overlap in diagnoses, the correlation between reading and attentional factors, or the skills that help differentiate difficulties in one area from difficulties in another (and from comorbid difficulties). In many of the above studies, attention is frequently operationalized as a behavioral difficulty noted by parents and/or by teachers. This kind of behavioral attention is viewed differently from (though hypothesized to be related to) attention as a cognitive construct, as in Besner et al. (2016).

Skilled readers appear to glide through text. Given that adults read text or individual words, especially those that are familiar, with little perceived effort, it might be presumed that this process occurs automatically. Although it may be true that some consider reading to be automatic, individuals who have watched children develop reading skills, particularly those who struggle, can attest to the fact that learning to read is not automatic. Thus, in some ways, the argument that reading is not automatic may strike some educators as obvious. However, it is also true that as skills improve, the reading process does indeed become less effortful. This idea is supported by neuroimaging findings, in which activity in the ventral occipital-temporal areas is associated with reading skill, which is thought to reflect whole-word recognition of familiar words in contrast to the “sounding out” of unfamiliar words or nonwords.

In the above context, then, this review highlights three specific ways in which attention, when elaborated as a cognitive construct, can influence the reading process, even against a background of seemingly automatic recognition. Specifically, the authors review the roles that spatial attention, central attention, and executive attention play in word reading.

Key Findings

Spatial Attention

Spatial attention with regard to reading refers to allocating mental resources, either explicitly or implicitly, between or within words. If spatial attention were unimportant, manipulating spatial focus would not get in the way of identifying individual words (for better or worse).

The impact of spatial attention can be shown through priming effects—for example, when a shape appears in a spatial location where a future word will be displayed, identification of the word is faster than when the shape appears in a location different from the word. This effect occurs for both high- and low-frequency words, suggesting that this type of attention is allocated prior to the word being processed semantically. Moreover, when a word reliably appears in a given location, it is possible to ignore words that appear in distractor locations. In both cases, spatial attention must be invoked prior to explicit word reading. Even in an implicit task (e.g., a word appears in one space and a color patch appears in another, and the participant is told to attend to the color patch), the word can be ignored so long as the target task demands enough spatial attention.

In addition, the Stroop effect occurs when a color word (e.g., blue) is written in an inconsistent color (e.g., red). In this case, when told to name the ink color, participants are slow to do so because the word is so familiar. However, if a single letter is spatially cued, this effect is reduced. Such results imply that spatial manipulation interferes with a word identification process, further revealing the role of spatial attention in reading.

Central Attention

Central attention with regard to reading refers to allocating mental resources to multiple tasks. If central attention disrupts reading processes, it would indicate that word identification is not automatic.

Some studies investigating this idea are based on the dual-route theory of reading, in which there are two routes to reading: a lexical route (there is a direct link from lexical/semantic representation to phonology) and a sublexical route (pronunciations are based on spelling-to-sound correspondences). According to the theory, reading regular words (e.g., save) can be accomplished via either route, reading irregular words (e.g., have) must be accomplished via the lexical route, and reading nonwords (e.g., zave) must be accomplished via the sublexical route.

Experimental studies indicate that, at least in skilled readers, processing words sublexically (i.e., sounding them out) requires central attention whereas processing words lexically does not. However, when the letter units are orthographically identified, semantic processing does indeed require central attention, even via the lexical route; this can be demonstrated by manipulating the Stroop effect discussed earlier.

Executive Attention

Executive attention with regard to reading refers to altering the allocation of mental resources to situational demands. If executive attention were at play in the reading processes, manipulating situational demands would interfere with reading and would indicate that word identification is not automatic.

To test this effect, participants are required to switch between reading words with different characteristics (e.g., regular, irregular, nonwords). Recall that irregular words are processed via the lexical route, nonwords are processed via the sublexical route, and regular words can be processed via either route. It turns out that a “switch cost” is evident when routes need to be alternated (e.g., reading an irregular word followed by a nonword), but there is no switch cost when routes do not need to be alternated (e.g., when reading a regular word followed by an irregular word or a nonword). These and other results indicate that because situational factors influence reading performance, executive attention is needed to manage these contextual factors.

Recommendations

This type of review article complements a variety of work that addresses the relationship between reading and attention. In contrast to many studies on the topic, this article highlights a different way of conceptualizing this relationship by focusing on experimental paradigms to tease apart the specific cognitive contributions of different types of attention. Although it is not highlighted above, Besner et al. (2016) believe future studies should also focus on the extent to which behavioral and cognitive accounts correspond to structural and functional neuroimaging results (see below suggestion for further reading). Integrating knowledge across behavioral, cognitive, and imaging paradigms is essential; doing so will likely yield new insights into how attention operates for reading and thus whether it may be manipulated to improve reading.

This study focuses on the role of attention in reading for adults. In doing so, it highlights a balance between the automatic and the attentional-demanding aspects of reading. It is important to consider the degree of automaticity in reading and the degree to which different attentional processes break down during the process. Evaluating such processes in children, particularly those struggling to read, would be an obvious extension to this study and one that may be exploited to better assist those who struggle. Besner et al. (2016) also note that attending to reading skill level would be an important direction to explore in future studies.

For Further Reading

This chapter focuses on the role of attentional control for word identification:

Balota, D. A., & Yap, M. J. (2006). Attentional control and flexible lexical processing: Explorations of the magic moment of word recognition. In S. Andrews (Ed.), From inkmarks to ideas: Current issues in lexical processing (pp. 229–258). New York, NY: Psychology Press.

This work provides a framework for understanding the extent to which various cognitive models can be expressed in terms of their neuroanatomical correlates:

Taylor, J. S. H., Rastle, K., & Davis, M. H. (2013). Can cognitive models explain brain activation during word and psuedoword reading? A meta-analysis of 36 neuroimaging studies. Psychological Bulletin, 139(4), 766–791. doi:10.1037/a0030266

This work differentiates attentional control processes, behaviorally and via imaging, that operate at different points in the reading process and that are consistent with cognitive models of reading:

Inhen, S. K. Z., Petersen, S. E., & Schlaggar, B. L. (2015). Separable roles for attentional control sub-systems in reading tasks: A combined behavioral and fMRI study. Cerebral Cortex, 25, 1198–1218. doi:10.1093/cercor/bht313

This review focuses on the role of spatial attention for word identification:

Montani, V., Facoetti, A., & Zorzi, M. (2014). Spatial attention in written word perception. Frontiers in Human Neuroscience, 8, 42. doi:10.3389/fnhum.2014.00042