Developmental dyslexia is defined as an unexpected difficulty in reading in children and adults who otherwise possess the intelligence and motivation considered necessary for accurate and fluent reading and who also have had reasonable reading instruction. Although dyslexia was first described more than a century ago, only within the last two decades have neuroscientists been able to determine the neural systems influencing reading and reading disability through brain-imaging studies.

These advances in understanding the neurobiological underpinnings of reading and dyslexia have been informed and facilitated to a large degree by progress in understanding the cognitive basis of reading and dyslexia. Although a number of dyslexia theories have been proposed, a strong consensus now supports the phonological theory, which recognizes that speech is natural, while reading is acquired and must be taught. In order to read, a child must acquire the "alphabetic principle"—the insight that spoken words can be pulled apart into the elemental particles of speech and that the letters in a written word represent these sounds. Results from large and well-studied populations confirm that a deficit in phonology represents the most robust and specific correlate of dyslexia and form the basis for the most successful and evidence-based interventions designed to improve reading [summarized in (S Shaywitz, 2003)].

How Does the Brain Read?

Functional brain-imaging studies have mapped the neural circuitry for reading, converging on three left-hemisphere neural systems for reading (reviewed in Price & Mechelli, 2005; S.E. Shaywitz & Shaywitz, 2005), as shown in Figure 1 below.

Brain-imaging investigations also demonstrate differences in activation patterns between good and struggling readers at all ages. Studies indicate that younger children without dyslexia demonstrate significantly greater activation in the three left-hemisphere neural systems than do children with dyslexia. There appears to be compensation in anterior regions so that the difference between older children without an impairment and children with dyslexia are confined to two posterior regions, the parieto-temporal and occipito-temporal systems (B. Shaywitz et al., 2002).

These data along with research from other investigators show that children with dyslexia exhibit a failure of the left-hemisphere posterior brain systems to function properly during reading. One of these systems, the left occipito-temporal region, has been of most interest to reading researchers. Considerable research in the last five years indicates a critical role for this region in influencing skilled, fluent reading, with neurons in the occipito-temporal area coding for words and letter strings. Not surprisingly, this region has been labeled the "word-form" area (for discussion, see Dehaene, Cohen, Sigman, & Vinckier, 2005). Disruption or significant underactivation of posterior neural systems in dyslexia, especially disruption of the word form area, has been termed the "neural signature" of dyslexia, as shown in Figure 2 below.

As good readers grow older, the "word-form" area develops, while readers with dyslexia develop a region more posterior and medial. High-quality evidence suggests that readers with dyslexia learn to read by using memory-based systems. For example, a recent study found that the neural system that develops in good readers is in about the same anatomic region as that used by readers of the sound-based Japanese Kana writing system.

In contrast, the system developing in dyslexic readers is located in the region used by readers of the memory-based Japanese Kanji writing system (B. Shaywitz et al., 2007; S.E. Shaywitz et al., 2003). Thus, there is converging evidence indicating that over time, dyslexic readers come to rely on a memory-based system—so that rather than learning to sound out words, they are dependent on memorizing printed words.

Functional brain-imaging has been helpful in examining whether the neural systems for reading are malleable and whether the underactivation of these neural networks in struggling readers can be modified by an effective reading intervention. Compared with struggling readers who received other types of intervention, children who received an experimental intervention (focusing on evidence-based application of the alphabetic principle) not only improved their reading but, compared to pre-intervention brain-imaging, demonstrated increased activation in the neural systems for reading (B. Shaywitz et al., 2004). Other investigators also have found that an effective reading intervention influences neural systems in the brain (reviewed in SE Shaywitz & Shaywitz, 2005). These data have important implications for public policy related to teaching children to read: the provision of an evidence-based reading intervention at an early age improves reading accuracy and facilitates the development of those neural systems that underlie skilled reading.

From Research to Policy

Findings from laboratories around the world indicate a neural signature for dyslexia in every language —a disruption of posterior reading systems, primarily systems serving skilled, automatic (fluent) reading—and have implications for the acceptance of dyslexia as a valid disorder. For the first time, research provides irrefutable evidence that dyslexia is "real."

This demonstration has implications not only for reading instruction, but also for the provision of accommodations, a critical component of management for young adults attending postsecondary and graduate programs. Such findings should make policy-makers more willing to allow children and adolescents with dyslexia to receive high-stakes tests accommodations—such as extra time—that would allow readers with dyslexia who have a disruption in the word-form area influencing skilled, fluent reading to be on a level playing field with their peers who do not have a reading disability. The use of advances in neuroscience to inform educational policy and practices provides an exciting example of translational science being applied for the public good.

The work described in this review was supported by grants from the National Institute of Child Health and Human Development (P50 HD25802; RO1 HD046171 and R01 HD057655) to Sally Shaywitz and Bennett Shaywitz.

Portions of this article appeared in SE Shaywitz, Gruen, & Shaywitz, 2007; SE Shaywitz, Mody, & Shaywitz, 2006; SE Shaywitz, Morris, & Shaywitz, in press, and are used with permission.