Kathryn R. Pearson and Dr. Dillon Inouye, Instructional Psychology and Technology

Speed reading experts deem subvocalization as a significant obstacle to achieving optimal reading rates. Perhaps one of the most telling factors in assessing an individual’s capacity for speed reading is the presence or absence of subvocalization. Repressing this “phoenemic recoding” is a key step in many reading enhancement programs. However, this valuable step in maximizing reading ability and comprehension cannot be measured in most college-level readers.

The degree of freedom from subvocalization in the reading process is one of the principle factors affecting reading perception. The superiority of the silent reading process over the customary vocalization technique is generally known. As children of normal mental capacity and ability progress, they find that reading silently helps them to read faster. “Reading is a process of direct association between perceptual stimulus (what we see) and meaning, without any intervening subvocalization…of thinking the meaning rather than saying the words” (Buswell, 1947).

I agree with the statement, “readers will not develop flexibility of rate without training” (Bergquist, 1984). Students must be aware of regressive habits like subvocalization. Theoretically, then, stopping this phonemic recoding, or subvocalization, opens the floodgate of reading speed. A major limitation in assessing subvocalization is the difficulty in physically differentiating between advanced readers who subvocalize and those who do not. My first pilot test was performed under the direction of Dr. Christopher Dromey of the Department of Audiology and Speech-Language Pathology. In this test, I used a speech pathology instrument to test advanced readers for subconscious physiological signs of movement during reading. My subject pool included 10 undergraduate students from varying majors of study at Brigham Young University, Provo. Seven were men, three were women, all from similar socioeconomic backgrounds. All were native speakers of English who were self-proclaimed “proficient” rapid readers. The subjects were randomly assigned to an identical sequence of pre-approved literature and magnetic placement testing. A repeated measure design was used with each subject as his/her own control, using identical reading passages.

Subjects were tested individually, with no contact with other subjects before or after the testing. They were assigned to a test sequence of pre-approved literature while a tiny magnet was placed on the tip of their tongue with dental adhesive. Any movements of the tongue or jaw were recorded while reading and resting between reading passages. In the first phase, subjects simply engaged themselves as they would in a normal, relaxed reading atmosphere while in the second phase, they engaged in concentrated rapid reading. After introducing the students to the instruments in the sound booth, they spent one minute relaxing and adjusting to the testing environment. During this time we observed the frequency of the subject’s swallowing and unconscious tongue movements. They were then told they were to read several passages silently to themselves at both normal and rapid speeds, depending on what we asked. All passages were of three varying difficulties obtained from the Reading/Writing Center at BYU. We recorded results with a sensitive microphone placed above their head and with the magnetic headset that detected any movement of the magnet on the tip of the tongue. We did the same procedure with the second and third passages, informing the subjects that there would be a brief comprehension quiz at the end of the reading. A post-test was given to measure recall and understanding of the passages. All subjects were tested for signs of movement, mumblings, and any elevation, protraction, or retraction of the tongue. All notes were collected for analysis.

Despite our expectations of significant quantitative data, extremely little motion was detected. What few numbers were obtained were not enough to work with. My hypothesis was pending on an expected trace of movement in the more difficult passages to show that when presented with harder learning materials, and knowing of an upcoming test on those materials, there is a pattern of regression in reading. None of the subjects exhibited any physiological evidence that they were discreetly reforming the words as they read. Every one showed average or high comprehension in the post-test quiz for both speeds of reading. How can the goal of systematically eliminating subvocalization be accomplished when there are absolutely no physiological signs of subvocalizing at this reading level? From the negative results of this pilot test, I saw many things I could have done more effectively in the testing sequence and selection. I was particularly concerned with the likely possibility of instrumental interference. The headset, though designed to be as unobtrusive as possible, was still awkward and it was obvious to the subjects we were observing the movements of their mouths. This may have contributed to a more guarded reading, and therefore a less accurate result.

In subsequent talks with Dr. Inouye, the discussion quickly shifted from measuring regressive reading habits of university students to more theoretical, and more essential, thoughts. My current speed reading research is based on the idea of dichotic listening. This is perhaps the core of the whole phenomena—the ability for humans to perceive and understand multiple lines at once. This has never been tested before and we believe it will be a key step towards understanding and defending speed reading as both a valuable skill and a testable physiological fact.

During the Winter semester, I completed the graduate course in Empirical Inquiry taught by Dr. Inouye which delved further into the problems and of research design. I also participated in and used the IP&T 515R Rapid Reading course as a source of further theory, inspiration and instruction. I devoted many hours to increasing my own reading speed as well as researching multiple areas of this effective phenomenon.

The results of this first study indicate that using this type of physical detection does not contribute extensively to the significant increase teachers and students are looking for in rapid reading. In fields where the reward for retention and understanding is high, finding ways to help many achieve those goals is essential. I have learned so much about proper testing, validity concerns, subject selection, and statistics through this experience in collaborating with professors, students, and controversial theories both in the class and in the field. This project has been a valuable step towards future success in my career in instructional research. The results of my current project will be evaluated in the IP&T research methods/statistics course.