Science behind fluent reading

You must surely have come across someone in our circles who would read story books or novels in no time. Have you ever wondered why they can read so fast and why some are slow-readers?

Now a study had found that along with how our brain is wired, this is a result of how our formative years are shaped by the environment we grow up. For grown-ups, it may not matter much, but for children, reading fluently is an important milestone. Understanding reading fluency in children has posed many challenges to educators and scientists.

Now, researchers at the Indian Institute of Science (IISc), Bengaluru and the University of Oxford have come up with visual predictors for assessing reading fluency in children. The study was published in the Journal of Experimental Psychology: General by the American Psychological Association (APA) earlier this year.

The authors were Aakash Agarwal, K V S Hari and S P Arun from IISc and Sonali Nag from the University of Oxford and the Promise Foundation. This was funded by the DBT/WT India Alliance and the University of Oxford grants to one of the study authors.

Prof S P Arun, one of the researchers, notes: “Learning to read is a major milestone for children but we don't fully understand the individual variations.” In an earlier study, the same researchers had identified the changes in the brain that help in visually processing words and help people read efficiently.

“From previous work, we know that children who rapidly name objects are also fluent readers. This makes sense since rapid naming captures the ability to convert visual images into sound, a process integral to reading. But do fast readers differ in their visual processing too? This was not known, and that's where we come in,” says Arun.

Measuring speed and fluency

The individual variations for alphabetic languages are explained by two cognitive tasks—phoneme awareness (PA) and rapid automatised naming (RAN). PA measures the ability to manipulate phonemes in a word. RAN measures the speed of naming visually presented letters or objects.

In these cases, they use a pair of letters, syllables or words written consecutively or adjacently, called bigram. For example, AB, AN, AT etc with only two letters make a bigram.

The researchers investigated the relationship between reading fluency and visual processing by testing two specific hypotheses. In the first hypothesis, they asked whether learning to read results in the formation of specialised bigram detectors.

The second hypothesis was that reading fluency variations across children would be predicted by upright bigram processing during visual search, over and above the variation predicted by RAN tasks.

The researchers tested children on a reading task to measure their reading fluency, a naming task and then a simple visual task to find the odd item among many identical ones. Further, they also got the children to do three types of searches involving either single letters, pairs of upright letters (or bigrams) and inverted bigrams. They wanted to test whether fluent readers solved these tasks differently.

The tests were administered on children in grades 3 to 5, in the age group of 7 to 11 years, over two time periods, 10 months apart. The tests were administered to each student for two standardised measures of reading fluency—word and paragraph reading.

Each child was tested on a visual search task that included both single letters along with upright and inverted bigrams to measure the visual processing.

“We chose visual search because it is a natural, intuitive task for children—they have to simply search for an odd-one-out. Yet it has an objective measure—that is, correctly identifying the target," says Aakash Agarwal. He explains that measuring the time taken for visual search can yield many insights into the underlying representations of visual features, including printed letters.

The results in the first part revealed good model fits. They found that letter-letter interactions were weaker for upright bigrams compared to inverted bigrams. In other words, children were quick to spot specific visual processing differences in upright bigrams compared to inverted bigrams.

This prompted them to then go on to test the second hypothesis. They tested if children are better at upright bigrams due to reduced letter interactions, can this help predict their reading fluency?

“It turns out, this was a highly specific effect: It worked only for upright bigrams but not any other measure,” exclaims Prof S P Arun. According to the researchers, another interesting aspect was that it worked for upright bigrams with normal spacing between letters and not for upright bigrams with large spacing.

"Using our insights, we aim to develop smartphone-based games that would solve the literacy crisis in India and also help children with reading difficulties (dyslexia)," says Agarwal.

The study notes that all these results show that there is a highly specific change in visual processing that explains reading fluency in children, which was also predicted by language–based measures. Thus, the use of visual predictors for assessing reading fluency in children makes this study stand out.

In any case, as schools have reopened, amidst coping with the learning loss, the outcomes of this study can be used to assess reading fluency. This could then help in identifying specific interventions to ensure reading fluency is enhanced, ultimately aiding the child’s development needs.

A recent Annual Status of Education Report (ASER) on the status of learning during the pandemic had some startling revelations on reading ability in children across three states—Karnataka, Chhattisgarh and West Bengal. In Karnataka, the percentage of children in Class I who can read decreased from 53.8 % in 2014 to 43.2% in 2021. A similar trend has been observed in other states and in children in higher grades (classes II, III and V). This also highlighted a ‘learning loss’ of more than one year as a consequence of the pandemic.