Definition:

An experimental method in which participants see (or hear) a stimulus and decide as quickly as possible whether it is a real word. Reaction time and accuracy are the primary dependent variables used to infer properties of word storage and retrieval.

In-Depth Explanation

The lexical decision task (LDT) is the workhorse of word recognition research. A participant sits in front of a screen, a stimulus appears (e.g., DOCTOR, FLURM, NURSE), and the participant presses a “yes” or “no” key. The time from stimulus onset to keypress is recorded to the millisecond.

What makes the task informative is that the pattern — what slows responses, what speeds them, what produces errors — reveals the architecture of the mental lexicon:

Frequency effects: High-frequency words (e.g., the, house) are recognized faster than low-frequency words (e.g., oboe, tacit). This is one of the most replicated findings in psycholinguistics and forms the basis for models of lexical access.

Priming effects: If BREAD appears briefly before BUTTER, participants respond faster to BUTTER than if BREAD was preceded by an unrelated word. This is semantic priming, and it reveals the networked structure of lexical representations.

Neighborhood effects: Words with many orthographic neighbors (words differing by one letter, like hate, gate, late, rate) show complex facilitation and inhibition patterns that inform models of visual word recognition.

Morphological effects: Morphologically complex words like teacher show whether they are processed via their stem (teach) or stored as whole forms — a debate between full-listing and decompositional models.

In second-language research, LDT is used to measure L2 vocabulary depth, automaticity of L2 lexical access, and the integration of L2 words into the mental lexicon over time.

The auditory lexical decision task uses spoken stimuli and examines speech perception and spoken-word recognition.

History

Meyer and Schvaneveldt (1971) introduced the semantic priming version of the task, which revealed that recognizing NURSE is faster after DOCTOR than after BREAD, providing direct evidence for associative spreading activation in the lexicon. Coltheart et al. (1977) developed the neighborhood size measure that became central to models of visual word recognition. The task was progressively refined to control for non-word foils, spelling regularity, and stimulus onset asynchrony (SOA) in priming designs.

The task became the primary data source for several major computational models including the parallel activation and competition (PAC) model, the interactive activation model (McClelland & Rumelhart, 1981), and the Bayesian reader model (Norris, 2006).

Common Misconceptions

“LDT tests knowledge of a word.” LDT measures lexical access speed, not necessarily whether someone knows a word’s meaning. A participant could respond correctly to tacit without being able to define it.

“Non-words are just made-up random strings.” Non-word foils in well-designed LDT studies are pseudowords — letter strings that conform to the phonotactics of the language (e.g., FLURP, not XQTZK) — to ensure participants genuinely have to search the lexicon.

“Faster is always better.” Response speed must be interpreted alongside error rates and in relation to trial type. A participant who responds quickly but inaccurately may be using a liberal threshold, not showing superior lexical processing.

Criticisms

• LDT response times reflect not just lexical access but also post-access decision processes. What participants are measuring when they say “yes” or “no” is disputed.
• The task is artificial: recognizing isolated words is different from word recognition in continuous reading or speech.
• Frequency norms used in study designs are often outdated or corpora-specific, affecting how “frequency effects” are interpreted.

Social Media Sentiment

The term appears in academic and language-researcher communities but rarely in general language-learning discussion. Occasionally cited in second-language vocabulary research discussions on Twitter/X, particularly in debates about measuring receptive vs. productive vocabulary knowledge in learners.

Related Terms

• Priming — the mechanism revealed by LDT priming designs
• Syntactic Priming — structural (rather than lexical) priming
• Working Memory — often studied in relation to task performance
• Formulaic Sequences — whole-form storage predicts LDT advantages for frequent phrases

See Also

• Coltheart et al. (1977) — neighborhood size and LDT
• Norris (2006) — Bayesian reader model

Research

• Meyer, D. E., & Schvaneveldt, R. W. (1971). Facilitation in recognizing pairs of words: Evidence of a dependence between retrieval operations. Journal of Experimental Psychology, 90(2), 227–234.
• McClelland, J. L., & Rumelhart, D. E. (1981). An interactive activation model of context effects in letter perception. Psychological Review, 88(5), 375–407.
• Coltheart, M., Davelaar, E., Jonasson, J. T., & Besner, D. (1977). Access to the internal lexicon. In S. Dornic (Ed.), Attention and Performance VI (pp. 535–555). Academic Press.
• Norris, D. (2006). The Bayesian reader: Explaining word recognition as an optimal Bayesian decision process. Psychological Review, 113(2), 327–357.