Background noise – like the chatter in a coffee shop or the drone of passing traffic – might slow our reading speed, but according to a study of Russian readers, it doesn’t affect how our brain comprehends written text.
The study looked at the effects of auditory noise and visual distractions such as typos or poor formatting. Curiously, the researchers from the National Research University Higher School of Economics in Russia found dealing with a clutter of words increased reading speed, possibly because we find the process more irritating and want to finish reading quickly.
For example, if you’re wondering whether you should be listening to podcasts or music while working, the study has some interesting points to make. In particular, it examined how we might change our reading style to compensate for auditory or visual noise.
“Overall, previous studies reported a detrimental effect of both auditory and visual noise on reading fluency and comprehension, though their results varied,” write linguistics researcher Nina Zdorova and colleagues.
“So far, none of the studies exploring the influence of noise evaluated it in the framework of the language processing theories.”
One of the language processing theories examined was the noisy channel model, which posits that our brain deals with noise by looking at the meaning of individual words more and at entire sentences less. We then use a bit of smart guesswork to infer the overall meaning and relationships between words.
The second theory is the good enough model; that’s when our brains aren’t analyzing every single detail of a text but instead only grabbing enough words for a ‘good enough’ understanding. By focusing less on the precise syntax, our brains preserve some cognitive resources to deal with noise.
To see how reading was affected by noise in regard to these models, the researchers ran two experiments: one on auditory noise (71 participants) and one on visual noise (70 participants). Eye-tracking devices were used to study reading fluency, with follow-up tests used to judge comprehension.
Crucially, some of the target sentences given to the volunteers were tweaked to be more semantically implausible: that’s where wonky and confusing grammar or punctuation means you need two or three takes to actually understand what a sentence means.
When it came to the auditory noise test, background chatter from overlapping podcasts caused people to spend longer looking at the key section of sentences before completing their reading. This extra time could compensate for the noise, meaning sentence comprehension isn’t affected.
In the visual noise test – created by putting other short words and phrases alongside the sentences to be read – comprehension remained the same while reading speed increased. That’s a bit puzzling considering previous studies, but the researchers think people just wanted to finish the task quickly, with the visual noise an uncomfortable distraction.
“In both experiments, we observed that longer total reading time was associated with an accuracy increase for implausible sentences,” write the researchers.
“This is predicted by the good-enough processing model and indicates that good enough, semantically-based processing is faster than syntactically-based algorithmic processing.”
There’s a lot going on in this study, but overall it’s a bigger win for the good-enough language processing theory – and an indication that auditory and visual noise doesn’t make us rely any more or less on this particular comprehension method while we’re reading.
As for how background noise affects reading, this study matches earlier ones: most of the time, no differences in comprehension are observed, although some types of noise (such as music we don’t usually listen to) can be distracting in this context.
With so many variables to measure in terms of what’s being read and what the accompanying noise is, further study is required to learn more – especially about reading in different languages since syntax varies. However, potential distractions may not interrupt your reading as much as you think.
“We could not confirm the predictions that noise increases reliance on semantics,” write the researchers. “This does not invalidate the noisy channel and good enough processing models, and rather requires further research on this topic.”
The research has been published in PLOS ONE.