Reading text out loud has been shown to improve memory recall compared to its silent counterpart. But does this enhancement extend to understanding the material at a deeper level? A recent study published in the journal Memory & Cognition sought to answer this very question.
Previous research on study strategies has explored various methods to enhance learning effectiveness, focusing on how different approaches impact memory retention and comprehension. Notably, strategies like self-quizzing, spaced repetition, and elaborate self-explanation have been shown to improve learning outcomes. However, these methods often require a significant investment of time.
Among simpler techniques, reading aloud has emerged as a potentially efficient alternative. This interest in reading aloud as a study strategy dates back to early 20th-century research, which suggested that vocalization could aid in memorizing material, a phenomenon later termed the “production effect.”
Yet, while the production effect’s influence on memory has been well-documented, its impact on deeper comprehension remains less clear, highlighting a gap in our understanding of how vocalization influences learning beyond mere recall.
“We wanted to determine whether the production effect, a well-known memory improvement technique, could extend to deeper comprehension of written text beyond the typical word lists that researchers use in memory studies,” explained study author Brady R. T. Roberts, a postdoctoral scholar at the University of Chicago, who conducted the research as a PhD student at the University of Waterloo.
“There had been work in the educational literature that showed the related ‘read aloud technique’ improved comprehension, but that research tended to define certain types of ‘comprehension’ much in the same way we would define rote memorization in the field of psychology.”
To investigate this, Roberts and his colleagues conducted a series of four experiments.
Experiment 1 included 47 university students from the University of Waterloo, who engaged with 10 short passages from the Nelson-Denny Reading Test, reading some passages aloud and others silently in a randomized order to control for potential order effects. This within-subject design ensured each participant acted as their own control. After reading, participants answered multiple-choice questions designed to assess memory and comprehension.
Memory-focused questions tested the participants’ ability to remember specific details mentioned in the text. Comprehension-focused questions, on the other hand, required participants to engage with elements such as the theme or tone of the passage, its gist, or the inferences that could be derived from the text.
Experiment 2 expanded on the initial findings by including an additional condition: reading text silently in an unusual font, specifically Sans Forgetica, hypothesized to create a “desirable difficulty” and thus potentially enhance memory and comprehension. This experiment was conducted online due to the COVID-19 pandemic, broadening the participant pool to include 114 individuals recruited through the MTurk platform and an online research participation system at Flinders University.
After filtering for various criteria, the final sample consisted of 64 participants. The methodology mirrored Experiment 1 in terms of the reading and testing process but included the Sans Forgetica condition to test whether visual degradation of text could similarly enhance learning outcomes.
Experiment 3 aimed to further validate the findings of the previous experiments with a larger and more diverse sample. This experiment was also conducted online, recruiting 167 participants through the Prolific online recruitment system. The study maintained the original design of reading passages aloud or silently but enhanced data quality controls, including more stringent criteria for participation and data inclusion.
A total of 151 participants’ data were analyzed, with the study seeking to replicate the memory benefit observed in earlier experiments and further examine the effect on comprehension. Experiment 3 also explored participants’ intuitions about the efficacy of reading aloud versus silently.
Experiment 4 sought to generalize the findings to different materials and examine the comprehension effect more closely. This experiment used new reading passages obtained from the Test Prep Review website, chosen for their educational relevance and the inclusion of comprehension-focused questions.
Conducted online with 167 participants recruited via Prolific, the methodology was similar to Experiment 3, with adaptations to ensure the new materials were comparable in difficulty and content to those used in previous experiments. The final analysis included data from 131 participants, focusing solely on comprehension to pinpoint whether vocalization could enhance understanding beyond memory recall.
Across the experiments, the researchers consistently found that reading aloud significantly improved memory for the material over reading silently. These results align with the concept of the production effect, which suggests that the act of vocalization enhances the memorability of the material.
However, when it came to comprehension, the study’s findings painted a different picture. Despite the clear benefits for memory recall, reading aloud did not confer any significant advantage for comprehension. The comprehension-focused questions yielded similar accuracy rates regardless of whether the passages were read aloud or silently.
This outcome indicates that while vocalization makes specific details of the text more memorable, it does not inherently improve the reader’s capacity to grasp underlying concepts or draw connections between different pieces of information.
“While reading aloud can indeed improve your memory, it cannot aid in your deeper comprehension of text,” Roberts told PsyPost. “So, read your grocery lists aloud to remember them better, but don’t bother reading your textbook chapters aloud.”
Conducted primarily in a controlled environment, the study leaves room for exploring how these findings translate to real-world learning scenarios, including classrooms and self-study sessions. Future research could expand on these results, perhaps examining how different types of material or diverse subject matter might interact with the production effect.
“The major caveat is that we are relying on Bayesian evidence for a null effect of reading aloud in the case of comprehension, consistent as that finding is,” Roberts said. “Future studies will need to confirm this null effect in more experiments, especially with different measures of comprehension to ensure that our test materials were not the reason for null findings.”
“I think our study reveals a more general point that memory researchers would be wise to consider: Even though memory is presumably required before comprehension can occur, improving memory is not always enough to cause significant improvements in deeper understanding,” Roberts added.
The study, “Reading text aloud benefts memory but not comprehension,” was authored by Brady R. T. Roberts, Zoey S. Hu, Eloise Curtis, Glen E. Bodner, David McLean, and Colin M. MacLeod.
