Language
Novel Word Learning Task
Also called: new word learning task, pseudoword learning task, nonword-picture learning
Measure word learning: participants learn made-up names for new shapes over three rounds, then match names to shapes again after a break.
9-11 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Preview any test free after you sign up.
- Papers, 2021 to 2025
- 304
- Evidence grade
- C
- Registered clinical studies
- 7
What it measures
Measures how well people learn new words. Participants learn 12 made-up words, each the name of a new shape, then pick each word's shape from four. The list and the test come three times and the test once more after a short break. The main score is the number of words matched to their shapes over the three rounds.
Constructs
- Word learning
- Associative memory
- Learning across trials
- Retention
Research uses
- Language development and developmental language disorder research
- Word learning research in adults
- Memory research at the group level
How the paradigm works
The Novel Word Learning Task measures how well people learn new words: a word form they have never met and the thing it names. In the method of novel word learning studies, people learn made-up words as the names of objects over repeated exposures and are tested on the pairings, for example by choosing the object a word names (Storkel, Armbrüster and Hogan, 2006; Coutanche and Thompson-Schill, 2014). Learning across exposures and retention after a delay are measured apart, because a word picked out right away can be lost minutes later (Horst and Samuelson, 2008). Novel word learning is a core method of research on language development and developmental language disorder, where children learn fewer new words than their age peers (Kan and Windsor, 2010), and of research on word learning in adults (McGregor and colleagues, 2017).
More about the paradigm
This version teaches 12 made-up words of three syllables, each the name of a shape drawn by ConductScience's code, so that no name the participant already knows fits it. The words are written: in Latin letters in seven languages and in katakana in Japanese, so the test needs no sound. In each round, every shape is shown with its name for 4 seconds (2 to 8 in the settings); then each name is shown with four shapes, and participants click or tap the one it names, with no feedback and no time limit. The three wrong shapes are shapes of other names, and the right shape takes each place equally often. The list and the test come three times, each time in a new order. A 2 minute break of true-or-false arithmetic follows (60 to 600 seconds in the settings), then a delayed test of every name with no study. A practice with three names and feedback comes first, and which name goes with which shape is drawn for each session.
The main score is the number of names matched to their shapes over the three rounds, from 0 to 36; chance is 9. The task also reports the score of each round and of the delayed test, the gain from the first round to the third, the delayed score as a percent of the third round, the median choice time, choices under 100 ms (anticipations, never counted as correct) and the break arithmetic. Scores support research and do not diagnose.
Scores
- Total Names Matchedcount
- Names matched to their shapes over the three rounds, 0 to 36 at the standard settings; chance is 9.
- Higher is better
- Matched, Round 1count
- Names matched to their shapes in the first round.
- Higher is better
- Matched, Round 2count
- Names matched to their shapes in the second round.
- Higher is better
- Matched, Round 3count
- Names matched to their shapes in the third round.
- Higher is better
- Matched, Delayed Testcount
- Names matched to their shapes after the break, with no study.
- Higher is better
- Learning Gaincount
- Names matched in the third round minus the first.
- Positive values show learning across rounds
- Percent Retained%
- Names matched in the delayed test as a percent of the third round.
- Higher is better; can exceed 100
- Median Choice Timems
- Median time of the choices, leaving out those under 100 ms.
- Lower is faster
- Anticipationscount
- Choices under 100 ms, never counted as correct.
- Quality check
- Break Accuracy%
- Share of the true-or-false arithmetic answered correctly during the break.
- Shows the break task was done
- Break Answerscount
- Arithmetic problems answered during the break.
- Shows the break task was done
- Choices Recordedcount
- Choices recorded in the run.
- Fewer than planned invalidates the session
- Choices Plannedcount
- Choices the session planned: every name in the three rounds and the delayed test.
- Context for the scores
Use and evidence
304 papers from 2021 to 2025 name this task in their title or abstract, up 16% compared with 2016 to 2020. That ranks it 85 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 76 |
| 2024 | 70 |
| 2023 | 62 |
| 2022 | 48 |
| 2021 | 48 |
| 2020 | 46 |
| 2019 | 68 |
| 2018 | 52 |
| 2017 | 39 |
| 2016 | 57 |
| 2015 | 46 |
| 2014 | 33 |
| 2013 | 42 |
| 2012 | 30 |
| 2011 | 35 |
| 2010 | 32 |
| 2009 | 17 |
| 2008 | 13 |
| 2007 | 19 |
| 2006 | 17 |
| 2005 | 9 |
| 2004 | 16 |
| 2003 | 5 |
| 2002 | 10 |
| 2001 | 3 |
| 2000 | 4 |
Use grade B
300 to 999 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 7
- Papers on reliability or validity
- 35
- Registered clinical studies using it as an outcome
- 7
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 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Words
- 12 made-up words of three syllables, written in Latin letters, and in katakana in Japanese sessions. None is a word, a reading or a pinyin reading in any of the eight languages.
- Shapes
- 12 shapes drawn by code in 6 colors, striped or dotted, with no name a participant already knows. Which word names which shape is drawn for each session.
- Study
- Every shape with its name once a round, 4 seconds each (standard; 2 to 8) with a 500 ms gap, in a new order each round.
- Test
- After each study list, each name with 4 shapes; the participant clicks or taps the one it names, with no time limit and no feedback. The 3 wrong shapes belong to other names: each shape is a wrong option 3 times a test, and the right shape takes each place 3 times.
- Rounds
- 3 rounds of study and test, then 120 seconds (standard; 60 to 600) of true-or-false arithmetic, then a delayed test of every name with no study.
- Practice
- 3 names studied and tested with 3 shapes each, each answer followed by the right shape for 2,000 ms.
- Anticipations
- A choice under 100 ms is an anticipation and never counts as correct.
The method follows novel word learning studies that teach made-up names for new objects and test them after each exposure and again after a delay (Storkel and colleagues, 2006; Horst and Samuelson, 2008). The words and shapes are ConductScience's own, and the words are written, so the test needs no sound. Scores support research and do not diagnose.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: novel_word_learning_choice from the scored run, one choice among four shapes for every word in the three rounds and the delayed test. Accuracy significantly below chance (one in four) invalidates and accuracy not above chance is flagged. Choices under 200 ms on at least half of the words invalidate. The same place on 90% of at least 12 choices, with accuracy not above chance, invalidates. Fewer choices recorded than planned invalidate. No arithmetic answered during the break, or break arithmetic at or below chance over at least 10 answers, is flagged for review. A hidden tab withholds the percentile.
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. One option on at least 90% of answered trials, when the correct options are balanced and accuracy is not above chance, means the answers did not depend on the items (the task counterpart of long-string responding in surveys; Curran 2016). The right shape takes each of the four places equally often, so one place chosen for nearly every word cannot follow the words. 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 scores are counts over every test, so a session with missing choices cannot be compared with complete ones. The break keeps the words out of rehearsal before the delayed test; no arithmetic answered, or true-or-false answers at or below chance, means the break may not have been used for the arithmetic. A hidden tab lets the study, delay or response timers run without the participant, so a percentile is withheld after one.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Study time per word (seconds) | 4 | 2 to 8 |
| Break before the delayed test (seconds) | 120 | 60 to 600 |
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.
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Novel Word Learning Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Storkel, Armbrüster & Hogan, 2006; Horst & Samuelson, 2008), with the standard settings (study time per word (seconds): 4; break before the delayed test (seconds): 120). The main score was total names matched (names matched to their shapes over the three rounds, 0 to 36 at the standard settings; chance is 9).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Storkel, H. L., Armbrüster, J., & Hogan, T. P. (2006). Differentiating phonotactic probability and neighborhood density in adult word learning. Journal of Speech, Language, and Hearing Research, 49(6), 1175-1192. https://doi.org/10.1044/1092-4388(2006/085)Made-up names for new objects, tested after set numbers of exposures.
- Horst, J. S., & Samuelson, L. K. (2008). Fast mapping but poor retention by 24-month-old infants. Infancy, 13(2), 128-157. https://doi.org/10.1080/15250000701795598A retention test minutes after learning.
- Coutanche, M. N., & Thompson-Schill, S. L. (2014). Fast mapping rapidly integrates information into existing memory networks. Journal of Experimental Psychology: General, 143(6), 2296-2303. https://doi.org/10.1037/xge0000020Written names for unknown objects, tested by choosing among pictures 10 minutes later.
- Angwin, A. J., Wilson, W. J., Arnott, W. L., Signorini, A., Barry, R. J., & Copland, D. A. (2017). White noise enhances new-word learning in healthy adults. Scientific Reports, 7, 13045. https://doi.org/10.1038/s41598-017-13383-3Study rounds of written names with pictures, each followed by a test.
- Kan, P. F., & Windsor, J. (2010). Word learning in children with primary language impairment: A meta-analysis. Journal of Speech, Language, and Hearing Research, 53(3), 739-756. https://doi.org/10.1044/1092-4388(2009/08-0248)Review of novel word learning in children with language impairment.
- McGregor, K. K., Gordon, K., Eden, N., Arbisi-Kelm, T., & Oleson, J. (2017). Encoding deficits impede word learning and memory in adults with developmental language disorders. Journal of Speech, Language, and Hearing Research, 60(10), 2891-2905. https://doi.org/10.1044/2017_JSLHR-L-17-0031Novel word learning and retention in adults with developmental language disorder.
Source papers
- Storkel HL, Armbrüster J, Hogan TP (2006). Differentiating phonotactic probability and neighborhood density in adult word learning. Journal of Speech, Language, and Hearing Research, 49(6):1175-1192.Paradigm source, Adults
- ConductCognition platform normsParadigm source, N/A
Questions
How long does the Novel Word Learning Task take?
About 9 to 11 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?
Total Names Matched, Matched, Round 1, Matched, Round 2, Matched, Round 3, Matched, Delayed Test, Learning Gain, Percent Retained, Median Choice Time, Anticipations, Break Accuracy, Break Answers, Choices Recorded and Choices Planned. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Novel Word Learning Task used in research?
304 papers from 2021 to 2025 name it in their title or abstract, up 16% compared with 2016 to 2020. That ranks it 85 of 395 test families in the ConductCognition test census (use grade B). Source: ConductCognition test census 0.14.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Novel Word Learning Task have?
In the published research, 7 papers report its test-retest reliability and 7 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Which settings can a researcher change?
Study time per word (seconds) (2 to 8) and break before the delayed test (seconds) (60 to 600). 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. Its words are made up and the same in every language, written in Latin letters, or in katakana in Japanese sessions. 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.