As I was reading in preparation for writing a post about “the science of reading,” I found that Mark Seidenberg had posted a note on that very topic on 23 March 2022. As many readers will know, Professor Seidenberg is an eminent psychologist who studies language, cognition, and neuroscience…and reading. In his post he made several points that I hoped to make, so this post will turn out to be focused on my analysis of his message.

In his post that has the catchy title, “Don’t Stop Believin’,” Seidenberg (2022), responded to the question of whether he “believe[s] in the science of reading.” He explains that he thinks scientific study of reading is important and that there is much a thoughtful consumer of that research can learn by examining that research.

Do I support the “science of reading movement,” the efforts of a loosely-defined cohort of individuals and organizations to link reading research and educational practices? Yes and no.
Yes, I support efforts to make the linkage and participate in them myself. No, I do not approve of all activities that are being undertaken under the rubric of the “science of reading”. My investment in using research to improve literacy outcomes is such that I will do what I can to identify policies and projects that seem ill-advised or ill-informed, given what I know about reading and the educational context. And do what I can to help devise better approaches based on what researchers have found, taking into account the conditions that govern learning in the real world. That is certainly the more difficult task.

He elaborated, though, by expressing his concern regarding “over-reliance on teaching children rules.” In his post, Professor Seidenberg explained that he promotes a balanced approach to learning to be literate, but he doesn’t mean “balanced” in the sense that word is often used in discussions about literacy. He’s not taking an anti-phonics view, as many advocates of “balanced literacy” do.

If I’m understanding his argument correctly, he objected to reliance on teaching students “rules.” I share this concern, but only to a degree. Let me explain.

What are rules?

For those who know a bit about printing, a “rule” is a solid line composed of “em” dashes. That meaning is off base. Although the meaning is closer to what we need to discuss, I am also not talking about rules as in the specifications for playing a sport of game. In the case of literacy, I think “rules” should be taken to mean verbal statements of widely applicable principles in reading and spelling words.

Here are a couple of familiar rules:

1. “I before e except after c, or when sounded as ‘a’ as in neighbor and weigh.”

2. “If it has an e on the end, the vowel will be ‘long.’”

3. “To make it plural, just add an s.”

Most such rules for spelling and reading are inaccurate. They are rife with exceptions. For example, as a clever page on the Miriam-Webster (2022) site notes, the rule about “i before e, except after c…” is simply mistaken; the rule would need to have many more subrules (than ‘a’) to cover all the spellings where e precedes i. Readers will have little trouble identifying exceptions to #s 2 and 3.

In other publications, including one (Seidenberg et al., 2020; SET pal Devin Kearns is a co-author), Professor Seidenberg (note that his name has an ‘e’ before the ‘i’; I wager his name is not pronounced “Say-den-berg!”) discussed problems with rules. He and his colleagues go beyond exceptions in objecting to teaching reading and spelling rules. For example, they tie their objections to the idea that rule-governed behavior is not automatic, but decoding needs to be automatic.

I agree that over-reliance on rules can induce inadequate learning, especially in reading. As Brown et al. (1981) argued, following rules “blindly” does not lead to broad learning. They argued that reading instruction needs to move to self-controled behavior, a point that seems similar to me to the ideas Seidenberg has argued.

I get the impression, however, that Professor Seidenberg is aiming his concern at his perception of educators. “Don’t just teach rules,” if I can state his argument as a rule. I wish he would allow for some exceptions. Not all educators teach decoding by having children verbalize rules.

In fact, in his focus on the psychology of reading, I fear that Professor Seidenberg may have missed some pretty powerful research about teaching, especially teaching reading. There are many studies in the educational literature that directly address teaching of reading, and that I hope Professor Seidenberg will find the time to review. Indeed, in the same issue of Reading Research Quarterly where Seidenberg et al. (2020) appeared, there is a paper by Ehri (2020) that discusses her research on decoding instruction. And there’s lots more.

Reading instruction research

I’m just going to catalog a few studies to provide starting places for folks who want to read some research about teaching reading. My sadness is that some folks who talk about “the science of reading” do not seem to know that this knowledge exists.

1. Carnine (1976):

   The instructional procedure of separating the similar sounds e and i from each other during instruction was evaluated in two experiments. In Experiment 1,42 first graders were assigned to a group in which e and i were introduced together or were separated by four other letters. Children in the similar-separated group made more correct responses to the two target letters during training, but the posttest scores were low and did not differ for the two groups. By measuring trials to criterion, Experiment II investigated the efficiency of separating similar sounds and of cumulatively introducing each sound. In the cumulative introduction procedure, children were brought to criterion on each group of sounds before a new sound was introduced. Thirty-five pre-schoolers completed training in one of three groups: similar-separated with cumulative introduction, similar-together with cumulative introduction, and similar-separated with simultaneous introduction. Children from the similar-separated group with cumulatively introduced sounds reached criterion significantly more quickly than the similar-together group and the simultaneous group. Posttest scores for all the children were substantially higher than in Experiment I and were significantly higher for the two cumulative introduction groups than for the simultaneous group. [Abstract]

2. Carnine (1977):
   Carnine examined the effects of teaching children to read words using sounds versus teaching them to read the words as wholes (i.e., “whole word method”). The children taught to sound out the set of words learned them faster than those taught the words a entities. They also had higher scores on a set of transfer words that were not regularly pronounced (i.e., didn’t follow the rules). [JohnL’s summary]

3. Carnine (1980):
   The procedure of sequencing visual discriminations (the letters b, d, p, and q) in an easy to difficult progression was investigated. In a low-confusion-alternatives first sequence only one choice in an inital match-to-sample task was similar to the target letter. In a high-confusion-alternatives first sequence all the choices in the initial task were similar to the target letter. Twenty preschoolers were randomly assigned to either a low-confusion-alternatives first or a high-confusion-alternatives first group. The preschoolers who began with the low-confusions alternatives reached criterion on a subsequent successive match-to-sample task in significantly fewer trials, M = 31.0 versus 69.1. [Abstract]

4. Carnine et al. (1984):
   Two studies were conducted to evaluate students' ability to utilize contextual information in learning the meaning of unfamiliar words. A descriptive study involving fourth-, fifth-, and sixth-grade children was designed to examine the differential effects of form and proximity of contextual information on students' learning of unfamiliar words. An experimental study involving the same-aged students was conducted to examine the differential effects of three intervention strategies designed to facilitate the use of contextual information in learning the meanings of unfamiliar words. The results of the descriptive study suggest that students were better able to determine the meaning of unfamiliar words when (a) contextual clues were provided, (b) students were older, (c) the clues were in synonym rather than inference form, and (d) contextual clues were closer to the unfamiliar word. In the experimental study, rule-plus-systematic-practice and systematic-practice-only conditions produced higher transfer scores than a no-intervention condition. [Absract]

5. Patching et al. (1986):
   The present study compared a control and two experimental treatments, a direct-instruction approach and a workbook-with-corrective-feedback approach, in training three critical reading skills. The skills were the ability to detect instances of (a) faulty generalization, (b) false causality, and (c) invalid testimonial. Thirty-nine fifth-grade students were randomly assigned to one of three groups. When the 3-day instructional intervention was completed, students were given three specially designed tests: one main measure, a domain-referenced test geared to the material taught, and two supplementary measures. Significant differences were found on performance on the main posttest between the direct-instruction sample and both the workbook-with-corrective-feedback and no-intervention samples. Results from the two supplementary tests were either nonsignificant or favored direct instruction.[Absract]

Now, readers may have observed that I drew these examples from only one group of researchers (Carnine and colleagues). Please do not form a misrule that (a) the list shows all of that group’s sudies (I just picked a few that I thought were important in the context of this post) or (b) that this group is the only one doing pedagogical research on reading. Indeed, I limited myself to people who probably identify as “educators” rather than “psychologists,” “doctors,” “speach-language pathologists,” and the like—even though researchers in those fields could do such research. Readers with whom I’ve interacted will know that I admire reading research by others such as (but not limited to) Stephanie Al Otaiba, Benita Blachman, Roxanne Hudson, Holly Lane, Rollanda O’Connor, Paige Pullen, Sharon Vaughn, and many others.

One could create quite a seminar by just just following this rule: Select three or four papers by each of these folks and cluster them topically.

Sources

Brown, A. L., Campione, J. C., & Day, J. D. (1981). Learning to learn: On training students to learn from texts. Educational Researcher, 10(2), 14-21. https://journals.sagepub.com/doi/pdf/10.3102/0013189x010002014

Carnine, D. W. (1976). Similar sound separation and cumulative introduction in learning letter-sound correspondences. Journal of Educational Research, 69(10), 368-372.

Carnine, D. W. (1977). Phonics versus look-say: Transfer to new words. The Reading Teacher, 30(6), 636-640. https://www.jstor.org/stable/pdf/20194348.pdf

Carnine, D. W. (1980). Two letter discrimination sequences: High-confusion-alternatives first versus low-confusion-alternatives first. Journal of Reading Behavior, 12(1), 41-47. https://journals.sagepub.com/doi/pdf/10.1080/10862968009547350

Carnine, D., Kameenui, E. J., & Coyle, G. (1984). Utilization of contextual information in determining the meaning of unfamiliar words. Reading Research Quarterly, 19(2), 188-204. https://doi.org/10.2307/747362

Ehri, L. C. (2020). The science of learning to read words: A case for systematic phonics instruction. Reading Research Quarterly, 55, S45-S60. https://ila.onlinelibrary.wiley.com/doi/full/10.1002/rrq.334

Miriam-Webster. (2022). I before E Except after C: The famous rhyme is wrong. Here's why. Retrived 24 March 2022. https://www.merriam-webster.com/words-at-play/i-before-e-except-after-c

Patching, W., Kameenui, E., Carnine, D., Gersten, R., & Colvin, G. (1983). Direct instruction in critical reading skills. Reading Research Quarterly, 18(4), 406-418. https://doi.org/10.2307/747376

Seidenberg, M. S. (2013). The science of reading and its educational implications. Language Learning and Development, 9, 331-360. https://www.tandfonline.com/doi/full/10.1080/15475441.2013.812017

Seidenberg, M. S. (2022, March 23). Don’t stop believin’. https://seidenbergreading.net/2022/03/23/i-get-questions/

Seidenberg, M. S., & Borkenhagen, M. C. (2020). Reading science and educational practice: Some tenets for teachers. The Reading League Journal, 1(1), 7-11.

Seidenberg, M. S., Borkenhagen, M. C., & Kearns, D. M. (2020). Lost in translation? Challenges in connecting reading science and educational practice. Reading Research Quarterly, 55, S119-S130. https://ila.onlinelibrary.wiley.com/doi/full/10.1002/rrq.341