A lexical decision task is a research method in which someone decides whether a presented sequence of letters or sounds is a word in a specified language. A screen might show garden, followed by a plausible but invented sequence such as nustle. The participant presses one key for a real word and another for a nonword. Researchers record both the decision and, when the experiment is designed for it, the time taken to respond. This apparently simple choice can help investigate how written or spoken word forms are recognised.
Why does it matter for vocabulary? A familiar word can often be identified quickly, while a less familiar word or a convincing pseudoword may require more processing. Context, spelling patterns, word frequency, visual presentation, response rules and the participant’s language experience can influence the result. That makes the lexical decision task a tool for studying how vocabulary becomes accessible to a reader. It does not mean that a single fast button press measures comprehension, intelligence, effort or permanent word mastery.
This guide explains words and nonwords, lexicality, response time, accuracy, semantic priming, orthographic similarity and experiment design. It distinguishes a research instrument from an ordinary classroom vocabulary exercise and gives original examples for Primary and Secondary English. The technical background comes from peer-reviewed work on lexical decision, naming and recognition. It also explains common confounds and ethical limits so students understand what the experiment shows—and what it cannot establish.
Editorial note: All example test items, timing values and learner responses are invented demonstrations. The lexical decision method is a research tool and this article does not validate a clinical or standardised assessment.
Contents: word recognition, decisions and what experiments actually measure
Explore 67 chapters, tests and worked examples
- The basic experiment: word or not a word?
- A standard research description
- What does lexical mean?
- What is a nonword?
- Pseudowords and unpronounceable strings differ
- Why a real word may feel familiar
- Frequency can affect response times
- Response time is measured from an event
- Accuracy and speed are different outcomes
- A speed–accuracy trade-off
- Visual versus auditory lexical decision
- What a trial looks like
- Balanced word and nonword trials
- Response bias can matter
- The need for a consistent response mapping
- Practice trials are not the same as test trials
- The problem of guessing from appearance
- Orthographic neighbours can compete
- Phonological similarity matters in auditory tasks
- Pronounceability can be controlled
- The difference from a spelling test
- The difference from a definition quiz
- The difference from reading aloud
- The difference from nonword repetition
- The difference from phonotactic legality
- Semantic priming introduces a second question
- The classic Meyer–Schvaneveldt contribution
- Relatedness and association differ
- The prime must not give away the response
- Word frequency and priming can interact
- Word length can be a confound
- Orthographic regularity can matter
- The effect of task order
- The role of attention
- The role of visual presentation
- Language experience affects what counts as known
- Regional spelling variants matter
- Proper names need a rule
- Neologisms reveal a boundary
- A classroom version should not pretend to be clinical
- A simple reading activity
- A meaning follow-up can reveal a gap
- A spelling follow-up answers another question
- A morphology follow-up
- A word-sense follow-up
- A pseudoword can sound like a familiar word
- A phonotactic follow-up
- A research design with clear independent variables
- A research design with clear dependent variables
- Extreme reaction times need a policy
- A toy dataset with four trials
- The mean can hide individual differences
- A fast response can be a guess
- A slow response can be thoughtful
- An auditory task includes accent variation
- A practical teacher–student discussion
- One invented exercise about word families
- One invented exercise about specialist terminology
- A well-designed control item
- Why replication matters
- The ethics of testing children
- Accessibility and sensory differences
- A five-column study worksheet
- What a result cannot prove
- Ten diagnostic questions
- Frequently asked questions
- Sources and the next experiment
1. The basic experiment: word or not a word?
Imagine a fictional participant looking at a screen displaying GARDEN. They choose Word. A second display shows NUSTLE, deliberately invented for the exercise, and they choose Not a word.
The experimenter can count correct decisions under the stipulated vocabulary and language context. If responses are timed accurately, they can also examine response latency.
The task is defined by the decision about lexical status, not simply by showing words quickly or asking someone to repeat a vocabulary list.
2. A standard research description
A peer-reviewed study on what lexical decision and naming reveal about reading describes a laboratory procedure in which a participant sees a letter string and presses a key to indicate whether it is a word or a pseudoword.
The article distinguishes lexical decision from word naming and records speed and accuracy as separate outcomes. It also considers how task performance relates to reading measures.
A classroom can illustrate the same idea with printed cards, but such an exercise is not automatically equivalent to the standardised laboratory procedure.
3. What does lexical mean?
Lexical concerns the words and stored word-related knowledge used in a language. A lexical decision asks whether the stimulus corresponds to a recognised word under the task’s criteria.
This is not the same as answering a question about the word’s full meaning. Someone may recognise that a specialist term exists without being able to explain it accurately.
Separate lexical familiarity, actual semantic knowledge and sentence-level use when interpreting a learner’s performance.
4. What is a nonword?
A nonword is a presented form that is not treated as an established word under the experiment’s defined language and stimulus set.
Researchers may create pronounceable pseudowords, or less wordlike strings, to test different processing questions. What counts as a real word can change with dialect, specialist terminology and recent coinages.
Check the stimulus list against suitable lexical sources rather than inventing a string and announcing that no English speaker could ever use it.
5. Pseudowords and unpronounceable strings differ
A pseudoword can resemble ordinary language patterns while lacking an established lexical entry under the test. A string that violates familiar sound or spelling constraints may be easier to reject for a different reason.
If the word trials contain common everyday words and the nonword trials contain only impossible-looking sequences, the experiment may reveal an easy orthographic distinction rather than subtle lexical recognition.
Stimulus construction matters because the difficulty of the nonwords can change the measured performance.
6. Why a real word may feel familiar
A participant has encountered the word garden many times in reading and conversation. That experience can make recognition easier under particular conditions.
A rare technical term might still be a legitimate word but less familiar to the participant. A slow response therefore need not indicate that the word is invalid.
Researchers distinguish a word’s lexical status from its frequency, context and the individual’s experience.
7. Frequency can affect response times
Word-frequency effects are studied by comparing lexical items with different occurrence rates in appropriate language corpora, while controlling other relevant properties.
A familiar high-frequency word may be recognised more efficiently than a low-frequency one in many conditions, but the specific result depends on task design.
Do not assume a single quick response proves a word was personally encountered a precise number of times.
8. Response time is measured from an event
In a timed visual lexical decision trial, the clock may start when the letter string appears and stop when the participant presses a response key.
Research must define the presentation time, response device and accepted trial window. An ordinary teacher’s stopwatch will generally be less precise than a controlled computer recording.
Report the method before comparing reaction times. A number measured under a different procedure may not be directly comparable.
9. Accuracy and speed are different outcomes
A participant may answer more quickly at the cost of making more mistakes. Another may pause and achieve higher accuracy.
Neither performance can be understood completely from reaction time alone. Researchers often examine both correct-trial latency and error rates.
Teaching pupils to press as fast as possible without attending to accuracy can change the strategy rather than demonstrate improved vocabulary.
10. A speed–accuracy trade-off
If a fictional learner initially waits to check every word and later begins guessing, their average response time may decrease while their error rate rises.
Calling the later session an improvement would be premature. The student may have adopted a different threshold for responding.
Compare both outcomes and instructions. A faster choice is meaningful only within the intended decision standard.
11. Visual versus auditory lexical decision
A visual task presents written letter strings. An auditory task presents spoken items and asks whether they are words in the tested language.
The two tasks involve overlapping lexical knowledge but different sensory and processing demands. Hearing speech in noise is not the same as recognising printed letters.
Use the task modality that matches the research question and avoid generalising a result across reading and listening without evidence.
12. What a trial looks like
A typical computer experiment may begin with a fixation or ready cue, then display the stimulus and wait for a response before a brief pause.
The published reading investigation describes letter-string trials with defined presentation timing, keyboard decisions and timed termination.
A classroom imitation can simplify these details, but should not claim professional timing precision unless its hardware and protocol support it.
13. Balanced word and nonword trials
A study may present a mix of real words and constructed nonwords so the participant must evaluate the stimulus instead of always making the same response.
The ratio and ordering can influence expectations. If nearly everything is a word, a participant might develop a bias toward the Word key.
A useful study reports the balance and randomisation method rather than hide them behind a headline about speed.
14. Response bias can matter
Research on instructions and word bias in lexical decision investigates whether the emphasis given to participants changes word-versus-nonword responding.
The authors report differences between lexical categories and consider the contribution of response strategies. A second study of instructions found that changing emphasis affected some reaction-time patterns.
The lesson is not one universal bias value. It is that task instructions can change how observed decisions should be interpreted.
15. The need for a consistent response mapping
An experiment might assign the left key to Word and the right key to Not a word. Other designs can reverse the mapping or counterbalance it across participants.
Participants need practice so the experiment does not simply measure confusion about which button to press.
Record the mapping and consider whether motor preference or familiarity with the keyboard could influence the timing.
16. Practice trials are not the same as test trials
A short practice set helps participants understand the rule and the device. Feedback may be given during practice but withheld during measured trials, depending on the experiment.
If learning continues throughout a long test, later responses may reflect changing familiarity with the task as well as lexical processing.
Report what practice occurred and how trial order was managed before interpreting results.
17. The problem of guessing from appearance
A string such as XQZPL may look obviously unlike English spelling. The participant can reject it using orthographic familiarity without necessarily attempting lexical identification.
A plausible pseudoword such as NUSTLE is more challenging because it fits familiar-looking patterns, though its actual status should be checked for the intended language and variety.
Researchers select nonword properties deliberately. An easy Not a word response can reveal more about stimulus design than about depth of vocabulary.
18. Orthographic neighbours can compete
A written word can resemble other words differing by one letter. A pseudoword that looks similar to common words may be harder to reject than an unusual sequence.
The strength and direction of neighbourhood effects depend on stimulus properties, frequency and the task’s processing dynamics.
Do not assert that every visually similar word always delays recognition. The experiment has to test that particular hypothesis.
19. Phonological similarity matters in auditory tasks
Two spoken forms can be close in sound but differ in one phoneme. An auditory nonword can resemble a familiar real word.
Misrecognition may depend on accent, acoustic context and competition among candidates. A visual spelling test does not directly measure every aspect of that process.
The Minimal Pair guide helps distinguish meaningful sound contrasts from mere spelling differences.
20. Pronounceability can be controlled
A study comparing words and pseudowords may try to make the nonwords plausible under the tested language’s phonotactic patterns.
This can reduce a shortcut based on obviously impossible sound sequences, though constructing matched stimuli has practical and theoretical challenges.
Phonotactic legality is not the same as lexicality: a pronounceable form can still be absent from the established lexicon.
21. The difference from a spelling test
A spelling test asks someone to produce or select the correct written form. A lexical decision task asks whether a presented form counts as a word.
An individual may recognise a word yet misspell it when writing from memory, or spell a rare word correctly after deliberate learning but respond slowly in a speeded task.
These outcomes involve related knowledge without being interchangeable measures.
22. The difference from a definition quiz
A learner can press Word for a familiar technical noun without explaining its meaning. Another might know a concept through listening but fail to recognise its unusual written spelling quickly.
A definition quiz tests semantic explanation more directly. A lexical decision task primarily measures performance on word/nonword categorisation under the stated conditions.
Use the right task for the learning question instead of treating every exercise involving words as a measure of the same ability.
23. The difference from reading aloud
A naming task may ask a participant to pronounce a printed string. Lexical decision requires a categorical word status response.
Both have been used in psycholinguistic research, and they can show different patterns because their response demands differ.
The lexical decision and naming study explicitly investigates these tasks as distinct indicators of printed word processing.
24. The difference from nonword repetition
Nonword repetition asks a participant to hear and repeat an unfamiliar spoken sequence. Lexical decision asks for a word/nonword judgement.
One task involves immediate encoding and production of a sound form; the other involves categorisation under lexical criteria. A learner may perform differently in them for several reasons.
Do not convert one result into a diagnosis about memory, pronunciation or reading without the relevant assessments.
25. The difference from phonotactic legality
A word can be phonotactically well formed while not being an established word, and a loanword or unusual name may depart from familiar patterns while still being a real lexical item in context.
Lexicality and phonotactic acceptability should be defined separately when choosing stimuli.
The companion guide on phonotactic constraints explains how the tested language permits some sound sequences more readily than others.
26. Semantic priming introduces a second question
In semantic priming, a participant encounters a prime word before a target word, and researchers examine how the relationship between them affects recognition.
For instance, a related prime such as DOCTOR may change responses to NURSE compared with an unrelated prime under appropriate controls.
A priming effect can inform theories of word recognition or semantic processing, but it does not by itself show what a particular pupil consciously thought.
27. The classic Meyer–Schvaneveldt contribution
David E. Meyer and Roger W. Schvaneveldt’s 1971 paper Facilitation in Recognizing Pairs of Words reported experiments with pairs of letter strings and found faster responses for related word pairs in specified task conditions.
The study helped establish the significance of semantic relationships in recognition performance. It used particular pair-based task designs rather than every modern single-word lexical decision procedure.
Do not attribute a universal one-cause explanation to the result without considering the task and later theoretical developments.
28. Relatedness and association differ
Two words can be semantically similar, such as couch and sofa, or associated through a common situation, such as doctor and hospital.
A priming design should state which kind of relation it tests. Shared scenes, categorical similarity and learned association can lead to different interpretive questions.
An association is not proof that the words are synonyms or that their referents are the same entity.
29. The prime must not give away the response
If every word target follows one kind of prime and every nonword follows another, participants may learn the pattern and respond strategically.
A credible experiment balances or counterbalances conditions so the observed difference is not simply an unintended answer cue.
Task structure matters as much as interesting vocabulary examples.
30. Word frequency and priming can interact
More familiar target words can differ from less familiar targets in processing speed. If the related-prime condition happens to contain only easy words, a faster average cannot be attributed confidently to semantic priming.
Match or statistically account for relevant lexical properties before claiming a specific effect.
The general research principle is to isolate the variable being investigated instead of decorating an uncontrolled comparison with precise milliseconds.
31. Word length can be a confound
Long strings may require different visual processing from short ones, while word length can correlate with other lexical properties.
If one condition includes mostly three-letter common words and another includes lengthy unfamiliar strings, the interpretation is unclear.
Control or report length alongside frequency, similarity and nonword structure when the research question requires those distinctions.
32. Orthographic regularity can matter
English spellings differ in how predictably they correspond to pronunciations. Word recognition and naming can interact with these properties.
A lexical decision effect attributed to familiarity might actually reflect differences in spelling-to-sound regularity or related stimulus characteristics.
Select conditions deliberately and avoid treating every English word as equally regular in form.
33. The effect of task order
A participant may become faster with experience using the response keys or tired after a long series of trials.
Randomised or counterbalanced orders and appropriate breaks can help reduce these influences, although no design eliminates every confound.
Report whether a difference appears within comparable trial blocks rather than compare a person’s first trial with their hundredth as though only vocabulary changed.
34. The role of attention
A participant who is distracted by noise or another task may respond slowly even to a very familiar word.
Task performance combines lexical processing with attention, perception, motor response and decision strategy. The reaction time is not a direct reading of one invisible mental event.
Keep the conditions appropriate and avoid negative personal labels based on a few slow responses.
35. The role of visual presentation
Font clarity, display size, colour contrast and exposure duration can affect recognition. A blurred letter string creates a different challenge from a clear one.
An accessible experiment should consider participant needs and the medium used for presentation.
Do not compare a paper card task with a high-refresh-rate computer task as if their millisecond measurements were automatically equivalent.
36. Language experience affects what counts as known
A bilingual participant may recognise a legitimate word from one language that the experiment treats as a nonword in another.
The task must define its target language and handle cross-language overlap appropriately. It should not describe multilingual knowledge as an error in general ability.
An unfamiliar dialect word or proper name can produce the same kind of categorisation problem if the stimulus policy is vague.
37. Regional spelling variants matter
English contains conventional differences such as colour and color. Both can be legitimate words in appropriate varieties.
A research instrument using one spelling convention should define how accepted variants are treated, and avoid confusing dialect differences with a participant’s vocabulary deficiency.
Clear stimulus criteria protect interpretation and fairness.
38. Proper names need a rule
A string might be a person’s name, a place name or a brand, but not a common dictionary word. The experiment must decide how such items are classified.
Participants should receive instructions aligned with that policy. Otherwise two reasonable responders could disagree for principled reasons rather than because one processed the letters incorrectly.
The same issue arises with specialist technical words and newly coined terms.
39. Neologisms reveal a boundary
A newly coined expression may be understandable and already used by a small community without appearing in a general learner dictionary.
Whether a lexical decision experiment counts it as a word depends on the defined population, register and stimulus source.
Do not infer that a form cannot exist because it was marked Not a word in one study’s intentionally limited lexicon.
40. A classroom version should not pretend to be clinical
A teacher can show ten familiar words and ten invented plausible nonwords and ask pupils to classify them. This illustrates lexical decision.
But the exercise lacks the calibration, normative sampling and validity required for a clinical or standardised assessment. Results should not be used to label a child with a disorder or predict intelligence.
Describe the activity as a demonstration of how recognition and decision rules work.
41. A simple reading activity
Prepare a balanced set of recognisable words, checked nonwords and one clearly explained language policy. Show the stimuli individually and let pupils choose Word or Not a word.
Afterwards ask what made some choices difficult. Was the item unfamiliar, similar to a known word or unusual in spelling?
Discussion can reveal vocabulary questions, but a pupil’s explanation is different evidence from the timed recognition decision.
42. A meaning follow-up can reveal a gap
A pupil classifies PREDICTABLE as a word but cannot explain it in a sentence. The teacher has evidence of form recognition, but not sufficient evidence of understanding.
A follow-up asks the pupil to identify the meaning, use the word in context and compare it with a close alternative.
The two tasks together give a richer instructional picture without pretending that the original lexical decision measured everything.
43. A spelling follow-up answers another question
A learner recognises GARDEN but later writes a misspelling from memory. That suggests a distinction between recognition of a presented form and independent written production.
Use a spelling activity if accurate written recall is the teaching goal, not additional rounds of word/nonword button presses alone.
Vocabulary knowledge involves interconnected yet distinct skills.
44. A morphology follow-up
A pupil identifies happiness as a word and notices its relationship to happy and the suffix -ness.
A teacher can ask how the suffix changes grammatical class and whether the word fits a new sentence.
The Morphological Productivity guide examines word formation rather than speeded recognition.
45. A word-sense follow-up
A learner recognises BANK as a word but the lexical decision gives no direct information about whether they know its financial and geographical meanings.
A short contextual sense-selection task can investigate that separate ability.
The Word-Sense Inventory guide explains how one form can represent several meanings.
46. A pseudoword can sound like a familiar word
Invented strings are often designed to resemble genuine vocabulary. A participant may momentarily consider several candidate real words before rejecting the form.
The difficulty is part of what researchers can study, provided the stimuli are properly constructed and the analysis does not overstate the evidence.
Do not assume a hesitant response means the participant cannot read.
47. A phonotactic follow-up
A plausible spoken nonword may follow familiar English sound-sequence patterns, while another may contain a sequence generally disallowed in the tested variety.
Comparing these can raise questions about language-specific phonotactic knowledge, though the experiment must control other differences.
The Macquarie University phonotactics guide explains that an allowable sequence need not be an established word.
48. A research design with clear independent variables
Suppose a study asks whether word frequency influences lexical decision latency. Frequency class could be an independent variable, while mean correct-trial response time is one outcome.
Appropriate controls might include length, orthographic neighbourhood and nonword properties. The experiment should state how those features were operationalised.
Do not claim causation from a casual classroom comparison whose words were chosen after seeing which results were fastest.
49. A research design with clear dependent variables
Accuracy rate and response latency are two distinct possible dependent measures. Some studies may also model item-level or participant-level differences.
Each outcome requires clear recording rules: which trials count, how timeouts are handled and whether error trials are excluded from latency summaries.
A numeric result without these rules can conceal important methodological choices.
50. Extreme reaction times need a policy
One trial may have an unusually long response because the participant looked away or asked a question. The analysis should define how unusual observations are handled.
Researchers may use prespecified exclusion rules or statistical models, with appropriate justification. Deleting slow trials only because they weaken a desired finding is not responsible.
Record the decision and sensitivity of the conclusion to such exclusions when conducting real research.
51. A toy dataset with four trials
Imagine fictional correct-response latencies of 620, 710, 650 and 820 milliseconds for four word trials. The arithmetic mean is 700 milliseconds.
That average describes only the invented observations. It is not a norm for any age group, reading level or language.
A meaningful study would need more trials, participants and careful design before interpreting differences between conditions.
52. The mean can hide individual differences
Two groups can have similar average response times but different spread, error rates or item patterns.
A single mean should therefore be accompanied by a description of the sample and appropriate variability measures in real research.
Do not equate one pupil’s slower mean with poorer overall vocabulary when the task demands and their experiences may differ.
53. A fast response can be a guess
A pupil presses Word quickly for every item. The result may include some correct answers by chance, but the decision strategy does not demonstrate reliable recognition.
Examining error rates and the response pattern can reveal the issue. A task must balance speed against accuracy rather than reward clicking alone.
Use the outcome to improve the procedure before diagnosing the learner.
54. A slow response can be thoughtful
A pupil encounters an unfamiliar legitimate word and pauses to compare it with remembered forms. They eventually classify it correctly.
The extra time may reflect careful uncertainty rather than incompetence. The underlying process cannot be determined from the latency alone.
Teaching should respond to the child’s actual understanding of the word, not only the time they took to press a key.
55. An auditory task includes accent variation
Listeners can recognise spoken words across accents, but unfamiliar phonological patterns may increase task difficulty under some conditions.
That difficulty does not establish a deficit in intelligence or lexical knowledge. The selected recordings, target variety and participant’s exposure matter.
Test meaningful comprehension separately when the instructional goal is understanding spoken English in real contexts.
56. A practical teacher–student discussion
Teacher: ‘You marked this item Not a word. What made you choose that?’ Pupil: ‘I haven’t seen it before.’
The teacher explains that unfamiliarity does not prove invalidity; a rare word may be legitimate. They check a reliable source together.
The discussion turns the experiment into a lesson about evidence, lexical status and the limits of personal experience.
57. One invented exercise about word families
Show the words happy and happiness alongside a carefully checked nonword. Ask pupils to identify recognised forms and explain the relationship between the two real words.
The lexical decision addresses form status; the morphology follow-up addresses how meaning and grammar relate across the family.
Mark the tasks separately instead of pretending the button decision establishes morphological analysis.
58. One invented exercise about specialist terminology
Show an established Science word unfamiliar to some pupils. After they decide, provide a clear definition and contextual example.
A previously unknown legitimate word can become part of the student’s vocabulary through instruction. The first response does not define the child’s potential.
Do not describe the item as a pseudoword simply because the class has not encountered it before.
59. A well-designed control item
A control item should address a specific source of confusion. To examine spelling familiarity, the experiment might include nonwords matched on length and plausible letter patterns.
The matching choices need research justification. An arbitrarily difficult control can make another condition look artificially easy.
Explain what the control is controlling and what additional differences remain.
60. Why replication matters
One experimental result can be affected by stimulus choices, sampling, noise and analysis decisions. A replicated pattern across appropriate studies provides stronger evidence.
Research on lexical decision has developed over decades with several theoretical models and methodological debates.
Do not turn a single small result into a universal claim about how every reader’s brain stores words.
61. The ethics of testing children
A classroom demonstration should be age-appropriate, accessible and free from humiliating labels. Research involving children requires appropriate consent and institutional safeguards.
Individual reaction times can be sensitive educational data in some contexts and should not be published casually.
Use fictional or anonymised examples for teaching the concept unless proper authorisation and privacy protections are in place.
62. Accessibility and sensory differences
Visual or auditory impairments, motor differences and assistive input devices can affect a speeded task independent of lexical knowledge.
A fair procedure must consider whether the task and equipment match the participant’s access needs.
Do not make high-stakes judgements from a speeded test that measures barriers in the interface as much as recognition.
63. A five-column study worksheet
Record Stimulus, Word/Nonword Status Under the Rules, Participant Response, Accuracy and Response Time if measured adequately.
Add notes on word frequency, plausibility or confounds when investigating a real hypothesis. A classroom card exercise may leave the timing column blank.
Ask students to distinguish observed performance from the explanation proposed for that performance.
64. What a result cannot prove
A correct fast response does not prove deep comprehension, permanent memory or fluent composition. A slow response does not prove an absence of vocabulary knowledge.
Lexical decision performance depends on many factors, and the research question must match the interpretation.
Use additional tasks for meanings, speaking, spelling, reading comprehension and transfer when those are the educational outcomes of interest.
65. Ten diagnostic questions
1. Is lexical decision a word/nonword categorisation task? Yes. 2. Does it measure reaction time in some designs? Yes. 3. Is a pseudoword automatically an impossible sound sequence? No.
4. Does a recognised word guarantee understanding? No. 5. Are naming and lexical decision identical? No. 6. Can instructions affect performance? Yes. 7. Can frequency matter? Yes.
8. Are speed and accuracy interchangeable? No. 9. Does one slow response diagnose a learning disorder? No. 10. What is the core purpose? Investigating recognition decisions under defined conditions.
66. Frequently asked questions
Is this a vocabulary test? It can investigate aspects of word recognition, but a complete vocabulary assessment requires other measures. Do all legitimate words appear in every dictionary? No; language variety, names, technical terms and coinages complicate the boundary.
Can children try it? A carefully designed low-stakes demonstration can teach recognition principles. Can reaction time identify intelligence? Not responsibly from a simple word/nonword exercise.
Why is it a vocabulary technology? It turns the distinction between recognising a lexical form and understanding its meaning into a concrete research question.
67. Sources and the next experiment
What Lexical Decision and Naming Tell Us About Reading provides a described laboratory lexical-decision procedure and analysis. Piercey’s study on instructions and word bias examines response strategy, while a later related study examines the effects of emphasising nonword responses. Meyer and Schvaneveldt’s 1971 word-pair experiment supplies historical context for semantic priming.
All stimulus examples, individual responses and numerical times here are fictional. No classroom result is presented as a standardised norm or a causal brain measurement.
Try a low-stakes word/nonword demonstration with verified stimuli, then ask a separate question about the meaning of a recognised word. Compare what each activity reveals, what it leaves unknown and what further evidence would be necessary. That is the central lesson of lexical decision.