Language
Novel Spellings Task
Also called: orthographic learning task, self-teaching paradigm, orthographic learning paradigm, homophone choice task
Measure how readers learn the spellings of new words. Participants read short stories with made-up words in one of two spellings that sound the same, then pick and type the spelling they saw.
9-12 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 98
- Evidence grade
- D
- Registered clinical studies
- 0
People on the coast call it the kleef. A kleef has gray wings.
What it measures
Measures how readers learn the spellings of new words while reading. Participants read eight short stories, each using a made-up word six times in one of two spellings that sound the same, and answer a question about each story. A surprise test then asks them to pick the spelling they saw from four look-alikes and to type each word. Scores how often the seen spelling is chosen and typed, against the other spelling that sounds the same.
Constructs
- Orthographic learning
- Spelling knowledge
- Reading for meaning
Research uses
- Reading research
- Spelling research
- Literacy research
How the paradigm works
Readers learn a new word's spelling from a few encounters while reading for meaning, the self-teaching that builds up the spellings of words. In this orthographic learning task, participants read eight short stories silently and answer one question about each story's events. Each story uses one made-up word six times, in one of two spellings that sound the same, such as kleef or kleaf; each participant sees four words in one spelling and four in the other, a different split for each seed. The instructions mention only the stories. A surprise test then shows each word's two spellings with a look-alike of each that changes one letter, and asks which was in the story; participants also type each word as it was spelled. A setting puts the typing first. The stories and spellings are written for each of the eight languages with that language's own sound-alike spellings, such as another way to write a vowel, a silent letter or, in Japanese, a long vowel written two ways in katakana; Chinese words swap a character for another with the same syllable and tone.
Scores
- Spellings Chosen%
- Percentage of answered choices on which the spelling seen in the story was chosen.
- Higher means more spellings learned
- Sound-Alike Spellings Chosen%
- Percentage of answered choices on which the other spelling that sounds the same was chosen.
- Lower means the spelling, not only the sound, was learned
- Spelling Advantagepoints
- Spellings chosen minus sound-alike spellings chosen, in percentage points.
- Above 0 means the seen spelling was learned
- Spellings Typed%
- Percentage of typed answers that match the spelling seen in the story.
- Higher means more spellings recalled
- Sound-Alike Spellings Typed%
- Percentage of typed answers that match the other spelling that sounds the same.
- Informational
- Story Questions Correct%
- Percentage of the story questions answered correctly.
- Under 50 is flagged: the stories may not have been read for meaning
- Median Reading Timeseconds
- Median time a story stayed on screen.
- Informational; under 6 seconds is flagged
- Answers Under 100 mscount
- Choices and story answers made within 100 ms of appearing.
- 0 expected
- Spelling Questions Answeredcount
- Choices and typed spellings in the most complete run.
- 16 standard
- Spelling Questions Plannedcount
- Choices and typed spellings planned: two for each of the 8 words.
- 16 standard
Use and evidence
98 papers from 2021 to 2025 name this task in their title or abstract, up 5% compared with 2016 to 2020. That ranks it 260 of 632 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 29 |
| 2024 | 18 |
| 2023 | 16 |
| 2022 | 15 |
| 2021 | 20 |
| 2020 | 12 |
| 2019 | 18 |
| 2018 | 25 |
| 2017 | 19 |
| 2016 | 19 |
| 2015 | 14 |
| 2014 | 17 |
| 2013 | 16 |
| 2012 | 11 |
| 2011 | 13 |
| 2010 | 5 |
| 2009 | 6 |
| 2008 | 17 |
| 2007 | 4 |
| 2006 | 5 |
| 2005 | 3 |
| 2004 | 3 |
| 2003 | 1 |
| 2002 | 1 |
| 2001 | 1 |
| 2000 | 2 |
Use grade D
30 to 99 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 6
- Registered clinical studies using it as an outcome
- 0
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.44.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Stories
- 8 original stories in each of 8 languages, read silently at the participant's own pace, each followed by 1 question about its events.
- Made-up words
- 1 in each story, shown 6 times in one of 2 spellings that sound the same. Each participant sees 4 words in each spelling, a different split for each seed.
- Spellings
- Each language uses its own sound-alike spellings, such as ee and ea in English, ai and ei in German or b and v in Spanish. Japanese writes a long vowel or a v sound two ways in katakana; Chinese swaps a character for another with the same syllable and tone.
- Surprise test
- The instructions mention only the stories. Choice: the spelling seen, the other spelling and a look-alike of each that changes 1 letter. Typing: each word as it was spelled.
- Test order
- Choice first, then typing, as published. A setting puts typing first.
- Practice
- 1 short text and question with feedback before the stories.
- Scores
- Percentage of choices that picked the spelling seen and the percentage that picked the sound-alike spelling, with the same for typed answers. Answers under 100 ms count as anticipations.
The stories and spellings are our own. Reading is silent, and nothing is recorded from a microphone. The test reveals its words, so each person takes it once. Scores support research and do not diagnose. No percentile output.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: novel_spellings_choice from the most complete run, each checked against the canonical spellings and look-alikes for its word and language, with the spelling shown rebuilt from the seed; one of four options per word. Accuracy significantly below chance (one in four) invalidates and accuracy not above chance is flagged. Answers under 200 ms on at least half of the choices invalidate. Fewer than half of the planned choices and typed spellings invalidates and any shortfall is flagged. Fewer than half of the story questions right, or a median under 6 seconds on a story, is flagged.
Why these rules
Accuracy significantly below chance (one-sided exact test, p < .05) cannot come from guessing; it means the response mapping was reversed or the task was not followed (Slick, Sherman and Iverson 1999 treat below-chance forced-choice performance as definite response bias). Chance-level accuracy can be a real result in impaired participants, so it is only flagged. The spelling seen is in each position equally often. An answer under 200 ms is faster than a simple visual reaction, so it cannot reflect the item. Following response-time effort screening for untimed tests (Wise and Kong 2005), the session is invalid only when at least half the answers are that fast. The story questions check that the stories were read for meaning, which the self-teaching paradigm needs, so missing them is flagged rather than invalidating. The words are made up, so nothing can be looked up in another tab.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Test order | Choice, then typing (standard) | Choice, then typing (standard), Typing, then choice |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Untimed test: runs on desktops, laptops, tablets and phones. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- No percentile. Scores and quality flags only. How norms work
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Novel Spellings Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Share DL, 1999; Share DL, 1995), with the standard settings (test order: Choice, then typing). The main score was spellings chosen (percentage of answered choices on which the spelling seen in the story was chosen).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Share, D. L. (1995). Phonological recoding and self-teaching: sine qua non of reading acquisition. Cognition, 55(2), 151–218. https://doi.org/10.1016/0010-0277(94)00645-2The self-teaching hypothesis: decoding a new word while reading gives a chance to learn its spelling.
- Share, D. L. (1999). Phonological recoding and orthographic learning: A direct test of the self-teaching hypothesis. Journal of Experimental Child Psychology, 72(2), 95–129. https://doi.org/10.1006/jecp.1998.2481Children read short texts with made-up words, and three days later chose, named and spelled the spellings they had read better than sound-alike spellings.
- Chrabaszcz, A., Gebremedhen, N. I., Alvarez, T. A., Durisko, C., & Fiez, J. A. (2023). Orthographic learning in adults through overt and covert reading. Acta Psychologica, 241, 104061. https://doi.org/10.1016/j.actpsy.2023.104061Adults read stories that used each made-up word six times in one of two sound-alike spellings, aloud or silently; then a choice among the word, its sound-alike and a look-alike of each, and spelling.
Source papers
- Share (1999); Chrabaszcz et al. (2023): orthographic learning from reading.Paradigm source, Adults who read the session language
Questions
How long does the Novel Spellings Task take?
About 9 to 12 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops, tablets and phones. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Spellings Chosen, Sound-Alike Spellings Chosen, Spelling Advantage, Spellings Typed, Sound-Alike Spellings Typed, Story Questions Correct, Median Reading Time, Answers Under 100 ms, Spelling Questions Answered and Spelling Questions Planned. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Novel Spellings Task used in research?
98 papers from 2021 to 2025 name it in their title or abstract, up 5% compared with 2016 to 2020. That ranks it 260 of 632 test families in the ConductCognition test census (use grade D). Source: ConductCognition test census 0.44.0, counted 12 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Novel Spellings Task have?
In the published research, 0 papers report its test-retest reliability and 0 registered clinical studies use it as an outcome (evidence grade D in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Test order (Choice, then typing (standard), Typing, then choice). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese, with its words in the same language. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.