How the Brain Reads

If you are a skilled reader, you can comprehend this sentence as soon as you read it. Yet the process of reading is incredibly complex, and to do it your brain must integrate two normally distinct functions: visual processing and language comprehension.

Not surprisingly, the first step in reading involves visual processing, which occurs in the occipital lobes at the back of the brain (Figure 1). Once visual processing is complete, information about letters and letter strings is transmitted to the left temporal lobe, where most language processing occurs. The left temporal lobe identifies words from letter strings, and transmits this information to other parts of the brain, including the frontal lobe, to glean meaning from the text.

The entire process of reading can be divided into four steps: (1) processing of visual information, (2) identifying letters, letter groups, and whole words, (3) deciphering the meaning of individual words, and (4) extracting meaning from groups of words. Each of these steps is described in more detail below.

Processing of visual information

Images are projected onto the retina upside-down and backwards,  such that the the left side of an image is projected onto the right side of the retina, and is transmitted to the visual processing center in the right hemisphere of the brain. The right side of an image is projected onto the left side of the retina, and is transmitted to the visual processing center in the left hemisphere of the brain. In both cases, the visual processing center is located toward the rear of the occipital lobe, which is located at the back of the brain (Figure 2). 

The visual processing centers transmit signals to different parts of the brain, depending on image type. For example, an image of a face is sent to one region in the left occipital lobe, while an image of written text is sent to another region of this lobe. The region that processes written text, which is the same for all readers of all languages, is called the visual word form area. This area plays such an essential role in reading that neurologist Stanislas Dehaene calls it the letterbox region. The letterbox region is located at the border of the occipital and temporal lobes (Figure 1). 

Damage to the letterbox region caused by brain injury or stroke results in the loss of the ability to read, with no effect on the ability to speak or understand spoken language, a condition known as pure alexia. Persons with damage to the letterbox region can even write, although they lack the ability to read what they just wrote! Visual processing occurs in both hemispheres, but the letterbox region is only located in the left hemisphere (Figure 1).

Identifying letters and letter strings

The letterbox region is able to identify a particular letter in a specific region of the visual field. Interestingly, this region is not only able to identify a particular letter, it is able to recognize all variants of that letter, regardless of size, font, or whether it is upper or lowercase. The ability of the letterbox region to recognize different variants of the same letter MaKeS It EaSy To ReAd TeXt ThAt Is WrItTeN iN a MiXtUrE oF uPpEr AnD lOwErCaSe LeTtErS. 

Once individual letters are identified, the letterbox region zooms out to analyze strings of letters in the order that they appear on the page. In this way, words containing the same letters in a different order, such as AND and DNA, can be distinguished. The letterbox region responds more strongly to letter strings that adhere spelling rules, such at “wh” and “ing”, than to strings that violate these rules, such as “qnf”. 

By analyzing letter strings, the letterbox region is able to identify parts of words and ultimately, whole words. Unfamiliar and nonsense words can also be identified, as long as they follow the conventions of spelling for the language of the reader. The steps described so far all occur in the visual processing centers of the brain. The next steps occur in the language processing center, which deciphers meaning.

Deciphering the meaning of words

Brain imaging studies indicate that the letterbox region transmits information to the language processing center by two different routes: a phonological route that converts letters to sounds, and a semantic route that uses a mental dictionary to identify words. 

Each route seems to play a particular role in identifying words. For example, common words, such as ‘house’, which are presumably entered into the reader’s mental dictionary, activate the semantic route. Nonsense words that are not entered into this mental dictionary, such as ‘houts’, activate the phonological route, as do unknown words. 

Often, both routes must be activated to correctly identify a word. For example, misspelled words, such as ‘wimen’ and uncommon words, such as ‘sextant’, initially activate the phonological route. But once the sounds are identified, the semantic route is activated. Both the phonological route and the semantic route are described in more detail below.

The phonological route involves the left upper temporal lobe (Figure 1), which is known to play a key role in speech sound analysis. A subregion of this part of the brain, called the planum temporale, is stimulated when a reader sees a letter and hears the sound it represents. For example, this region is stimulated when a reader sees the letter ‘o’ and hears the /o/ sound. This region is not stimulated if the reader sees the letter ‘o’ and hears a different sound, such as ‘a’. 

The planum temporale seems to be able to identify the sound represented by a single letter. But processing of sounds represented by multiple letters is more complex, and likely involves verbal working memory. Verbal working memory is what we use to remember a phone number as we dial. Verbal working memory may also help us to process syllables, allowing us to decipher long words, such as acetaminophen.  

Verbal working memory involves several areas of the brain, including Broca’s area, the parietal lobe just above the planum temporale, the cerebellum, and the basal ganglia. Broca’s area, which is located in the left prefrontal cortex, plays an important role in speech production. The cerebellum, which is located in the lower brain, and the basal ganglia, which are located in the middle of the brain, have long been known to play a role in motor control. It is only recently that their role in verbal working memory has been elucidated. 

The semantic route involves the left middle temporal region, which seems to be able to identify synonyms. For example, this region responds more strongly if two words with different meaning, such as ‘honey’ and ‘couch’, are flashed in rapid succession than if two synonyms, such as ‘sofa’ and ‘couch’, are likewise flashed. Presumably, the synonyms ’sofa’ and ‘couch’ require less processing by the middle temporal region as they are recognized as having the same meaning. These results suggest that the middle temporal region sorts words into categories based on meaning. However, to truly grasp the concept that a word invokes, other parts of the brain must be activated.

Certain areas of the brain are activated when we think about a particular concept. For example, when we think about a hammer, the part of the brain that stores information about tools is activated. Other parts of the brain are activated when we think about plants, animals, people, beliefs, or body movements. Activation of concept-specific regions is the same regardless of whether the trigger for a thought is something we see or spoken or written words.

Many areas of the brain that store information have overlapping functions. For example, the part of the brain that stores information about what it means to bite is also activated when we bite into something. Thus, the word bite invokes an almost visceral response: when we read the words ‘I bite into an apple’, we can almost feel like we are performing this action. The words ‘I am afraid the dog will bite me’ invoke a very different response, as reading not only activates areas of the brain that store meaning, it also activates areas of the brain that regulate emotion. Reading is not about defining a particular word; it is about comprehending what words mean when they are grouped together into phrases and sentences.

As mentioned previously, the middle part of the left temporal lobe appears to be involved in recognizing individual words. Processing of whole sentences occurs further forward in this lobe and in the lower left frontal lobe (Figure 1). Neuroscientist Antonio Damasio believes that all of these areas act as convergence zones, accepting information and then deciding which areas of the brain must be accessed to comprehend it.

Extracting meaning from groups of words

Whole sentence processing is used to analyze the syntax, or structure of a sentence. Based on syntax, we can tell that the sentence “The dog bit the man” has a different meaning than “The man bit the dog”, despite the fact that both sentences contain the same words.

Whole sentence processing also enables us to find the most likely meaning of an ambiguous sentence, such as “Include your children when baking cookies”.  To process this sentence, we must consider whether we are more likely to bake our children or teach them how to bake. Fortunately, we are able to consider both possibilities simultaneously, through a method called parallel processing.

Parallel processing allows us to quickly choose the correct meaning of a word based on context. Consider the following sentence, in which the word ‘wind’ appears twice: “It is difficult to wind the kite string in this strong wind.” Through parallel processing, our brains are able to consider all definitions and pronunciations of the word ‘wind’ simultaneously before settling on the one that makes the most sense in each instance.

Conclusion

Reading is such as complex process that it unsurprising that so many people struggle to do it. Fortunately, the science of how the brain reads gives us insight into how to best teach reading. For example, since we know that the brain of a fluent reader has ability to recognize letters and to match these letters to the sounds they represent, we can surmise that a beginner reader must learn to do the same. At the same time, the beginner reader must be exposed to as many different words and concepts as possible to build their mental dictionary. How best to do this will be the topic of a future post.

References

Center for Alexia Research and Rehabilitation (n.d.). What is Alexia? CARR. Retrieved November 23, 2024, from https://www.aphasiaresearchcenter.org/alexia

Dehaene, S. (2009). Reading in the brain. New York

Dehaene, S., Jobert, A., Naccache, L., Ciuciu, P., Poline, J. B., Le Bihan, D., & Cohen, L. (2004). Letter binding and invariant recognition of masked words: behavioral and neuroimaging evidence. Psychological science, 15(5), 307-313.

Devlin, J. T., Jamison, H. L., Matthews, P. M., & Gonnerman, L. M. (2004). Morphology and the internal structure of words. Proceedings of the National Academy of Sciences, 101(41), 14984-14988.

Emch, M., Von Bastian, C. C., & Koch, K. (2019). Neural correlates of verbal working memory: An fMRI meta-analysis. Frontiers in human neuroscience, 13, 180.

Erskine, L., & Herrera, E. (2014). Connecting the retina to the brain. ASN neuro, 6(6), 1759091414562107.

Kweldju, S. (2015). Neurobiology Research Findings: How the Brain Works during Reading. Pasaa: Journal of language teaching and learning in Thailand, 50, 125-142.

Landi, N., Frost, S. J., Mencl, W. E., Sandak, R., & Pugh, K. R. (2013). Neurobiological bases of reading comprehension: Insights from neuroimaging studies of word-level and text-level processing in skilled and impaired readers. Reading & Writing Quarterly, 29(2), 145-167.

Van Atteveldt, N., Formisano, E., Goebel, R., & Blomert, L. (2004). Integration of letters and speech sounds in the human brain. Neuron, 43(2), 271-282.

Vinckier, F., Dehaene, S., Jobert, A., Dubus, J. P., Sigman, M., & Cohen, L. (2007). Hierarchical coding of letter strings in the ventral stream: dissecting the inner organization of the visual word-form system. Neuron, 55(1), 143-156.