Evidence for prosody in silent reading

Gross, J., Millett, A.L., Bartek, B., Bredell, K.H., Winegard, B. (2014). Evidence for prosody in silent reading. Reading Research Quarterly, 49(2), 189-208.


What goes on in our heads when we are reading? The mysterious depths of how the reading process occurs in the mind are explored in this challenging but fascinating article on prosody in silent reading.

Prosody is defined here as “the rhythm and melody of language” (p.189). It is easy to see how prosody can be a factor in oral reading and language; anyone who has listened to a text rendered in a monotone voice, and then a text rendered by a skillful speaker, reader, or actor, can readily understand how prosody can make a difference in how one comprehends oral texts. We don’t usually think of prosody in connection with silent reading, but the evidence from the two research studies reported here suggests that prosody is probably going on in readers’ minds as they read silently, and that prosodic skill levels in silent reading might affect how well a reader comprehends what he or she reads.

The connection between speech and print is shown here as strong, and very important to the reading process. The authors write about an “expressive inner voice” (p. 190) that is in the minds of skilled readers, and helps them make sense of texts. This inner voice might help a reader determine what words (or perhaps larger text chunks?) contain important information and should be emphasized. In oral texts, prosody can include such things as the length of an utterance, the tone, the pitch, the stress, and the intonation (p. 190). Written texts contain very few of these kinds of cues. There is some help from punctuation, but there is much more to prosody than that. How do silent readers determine what should be emphasized? For that matter, how do oral readers decide how to use prosodic cues? Some people are better at this than others; musicians, actors, speakers, and all kinds of performers seem to have a natural talent. How much of this is deliberate, and how much is so speedily and automatically done that it seems effortless? Are these performers aware of what they are doing? The authors here propose that the same kind of processing that occurs when a person is reading orally is going on during skilled silent reading. Could it be surmised, then, that skill with oral reading prosody would transfer to skill with silent reading prosody, and vice versa? How exactly does prosody, either in oral or silent reading, help the reader comprehend?

The authors attempted to capture this whole process by providing emphasis (words in all caps) for certain words within a text read by the participants (who were university students). Participants were asked to decide whether that emphasis was helpful (congruous) in understanding the text or not helpful (incongruous). Examples given were of both types. In the second study, the speed with which participants made their decisions was also measured. In general, the researchers’ prediction that the readers would be better able to read and rate the congruous examples was supported by the evidence gathered. This points to a sensitivity to prosody that carried over into oral reading. I wonder how this would look for younger readers, or even for adults older than the typical college students.

The authors provide extensive examples of the stimuli used in their study in an appendix. I spent some time examining those, and was struck by how difficult it must have been to generate the examples. A few examples could be ambiguous; in a couple of cases I wondered if an example considered “incongruous” might be perceived as “congruous” by someone with a creative and less conventional viewpoint. I was reminded of some of my experiences with gifted students who could explain their supposedly “wrong” answers in perfectly logical ways. I once had an autistic student whose perceptions of what “made sense” were unique and certainly different from the norm, but made sense in their own way when the student shared his logic. So, probably the notion of congruous vs. incongruous as a dichotomous concept is oversimplified. Still, I doubt if I could come up with better examples than the ones shared here, and I appreciate that the authors did share them.

Long after I read this article, my mind was abuzz with the possibilities. I know I will return to this article to further plumb the depths of what this research is telling us. I will be following up and checking out some of the previous research the authors cite. I was particularly fascinated by the studies cited on page 191 where readers’ reading rates slowed down or speeded up after they had heard the voices of the authors who wrote the texts they were reading. If the author had a slow voice, the readers’ silent reading rate slowed down, too, and the rates increased when the author had a more rapid delivery. I want to look more closely at that.

I’m also fascinated by the ideas near the end of the article for explicitly sensitizing young children to the connection between speech and print and the rhythms of language. As with other emergent literacy skills, there will be some children who learn about prosody and connect speech and print naturally through experiences at home, but others may need some explicit help. I found the “tapping game” (p. 199) the authors propose for teaching about stressed and unstressed syllables as well as the rhythm and meter in poetry quite interesting. Could parents and caregivers as well as teachers do these things with children? Would this be a way to simulate some of the things that happen naturally in the homes of children who seem to “naturally” learn about prosody?

Though I recommend this article, I do have some cautions. Though fascinating, the article is challenging reading, and has vocabulary that not all readers, even academics, may be confident of. Even as a scholar and literacy specialist, I had to check on several word meanings and make sure I was clear of their meanings. If I were using the article in a graduate course, I’d probably use it as an opportunity to model how to choose words for preteaching, deciding when that is necessary, and how to do the preteaching. Here, many of the words I wasn’t sure of turned out to be important to constructing the meaning of the article, so preteaching is warranted. My vocabulary list here included these words: nascent, disambiguation, suprasegmental, homographs (as opposed to homophones), salient, and of course, prosody. Your list may be different, depending on your background (linguists may not need a vocabulary review).

Do take a look at this study in spite of the challenges, and don’t let that daunt you. Even with the vocabulary issues, it is a readable and engaging report on a topic that those of us who are fascinated with the miracle of reading will be riveted to.