The Devil’s in the g-Tails

_ઠ^σ

A counter with an ear, a link, and a loop
comprise the typographical troop
of ligaturing accoutrements
which, to display online, is a task
Aussi insurmontable que Olympus Mons.

There are two lowercase (or miniscule) letters in the English language which have more than one form. The letters /a/ and /g/ both have a single-story and a double-story form, the former seeming to have derived from the miniscule letter /c/ for both, but the latter having decidedly more exotic roots. Nearly all native English speakers can read and identify both versions of /a/. However, despite the ubiquitous nature of the double-story /g/ (approximately 97% of all adult fiction and non-fiction print with a font employing the double-story /g/, as does 74% of children's books and 89% of chapter books for older children), adults have a difficult time identifying and creating the loop-tailed variant, despite perfect reading literacy. Cognitive science students at John's Hopkins University recently conducted an experiment on the phenomenon of literacy without the ability to recreate on command, and their results were so significant it may drive change in the manner by which penmanship is taught.

"We think that if we look at something enough, especially
if we have to pay attention to its shape as we do during
reading, then we would know what it looks like. But our
results suggest that's not always the case."

Michael McCloskey, Cognitive scientist and senior author on the paper

Title: The Devil’s in the g-Tails: Deficient letter-shape knowledge and awareness despite massive visual experience
Authors: Kimberly Wong (first author; current Junior year undergraduate), Frempongma Wadee (co-author; Bachelor’s degree Graduate), Gali Ellenblum (co-author; current graduate student in Cognitive Sciences), & Michael McCloskey (senior author)
Argument: Understanding letter shapes is a critical component of reading. However, the assumption that knowing letter recognition and knowing letter creation are correlated activities is weighed and found wanting.
Social Impact: Experiment findings may change how letter recognition and letter writing is taught.

The cognitive science team of Wong, et al, devised three experiments to test the nature of alternate-form letter recognition among literate readers. The three experiments were conducted with different sets of volunteers, and all validated a hypothesis that there is abysmal alternate-form letter recognition for one of the english alphabet's most ubiquitous and gnarly typeface forms.

• Experiment One: Research team asked 38 adults to identify all miniscule letters with two forms
• Experiment Two: Research team asked 16 adults to silently read a passage and verbalize words containing /g/
• Experiment Three: Research team asked 25 adults to pick the correctly drawn /g/ out of a lineup

Experiment One Results While most volunteers readily identified /a/ as having two forms, only 2 out of 38 volunteers named /g/. When asked to draw the second form, only 1 drew it correctly (and the team was grading generously at that). Even with prompting, the participants simply could not recollect the alternate-form of /g/. "We would say: 'There're two forms of g. Can you write them?' And people would look at us and just stare for a moment, because they had no idea," said first author Kimberly Wong. "Once you really nudged them on, insisting there are two types of g, some would still insist there is no second g." Remember: this particular alternate form, the loop-tail /g/, is present in 97% of adult fiction and non fiction. Only 3% of the survey could identify and draw it from memory.

Experiment Two Results All volunteers were asked to read a passage silently, but to speak aloud every word containing a /g/. There were 14 words containing a /g/, all presented as the loop-tailed form. All volunteers spoke all 14 words aloud. 100%!!! Alas, the research team was tricky. After asking for such hyper focus on /g/, they simply asked each participant to draw the form of the /g/ they had read.

Half drew an open-loop /g/. A form which appeared nowhere in the passage they had just read. 50% completely failed to store the form they had just read, and created the nearest analog in their working memory.

Of the half who did recognize they were reading loop-tailed /g/'s, again only 1 individual managed to draw it correctly. A lowly 6%, and who knows if they were luckier for having a smaller sample size...

Experiment Three Results In an effort to make it as easy as possible, and to provide a bias-correcting element, the final set of volunteers was asked to pick the correct form of a double story /g/ from a multiple choice diagram. Volunteers also were asked to pick the correct form of /a/ with the same manipulations of form. By switching which side the ear and link of the /g/ were present on (analgous to the hook and tail of the /a/), there were four choices to chose from. Participants succeeded 24% of the time, failing to beat pure random selection (25% for all four choices). The results are even worse when you consider a whopping 56% selected the mirror image of the correct form. A clear sign from their subconscious that they need to reflect more upon the content they read.