Distinguishing common and task-specific processes in word identification: A matter of some moment?

Andrews, Sally; Heathcote, Andrew · 2001 · Journal of Experimental Psychology Learning Memory and Cognition

DOI: 10.1037/0278-7393.27.2.514

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the cognitive processes underlying visual word recognition, specifically addressing the debate over whether word frequency effects reflect a common lexical access mechanism or are influenced by task-specific decision and production processes. Traditional research has relied on mean reaction times (RTs), which can obscure complex distributional changes. The authors argue that analyzing the shape of RT distributions provides a more refined evaluation of underlying processes than average speed alone. The central question is whether the larger frequency effect observed in lexical decision tasks (LDT) compared to naming tasks arises from familiarity-based decision processes specific to the LDT, nonlexical pronunciation routes in naming, or a combination of both. To address this, the authors conducted two experiments using a factorial design of word frequency (high vs. low) and animacy (animate vs. inanimate). In Experiment 1, 74 participants performed five tasks: lexical decision, semantic categorization (animacy judgment), standard word naming, word/nonword naming, and lexically contingent naming (naming only words). Experiment 2 included delayed naming. The researchers estimated RT distributions using ex-Gaussian parameters (mu, sigma, tau) and Vincentile averages. The parameter tau specifically indexes the skew of the distribution, reflecting the proportion of slow responses, while mu reflects the location of the main body of the distribution. The results revealed that low-frequency words produced more skewed RT distributions (higher tau) than high-frequency words in all tasks except delayed naming. This differential skew was most pronounced in tasks requiring lexical discrimination, such as the LDT. In contrast, the semantic categorization task yielded highly skewed distributions for all words, but the frequency effect was driven by shifts in the location of the distribution (mu) rather than changes in skew. The findings indicate that the word frequency effect is not uniform across tasks; it involves both a common process affecting the speed of lexical access and task-specific processes that influence the likelihood of slow, additional processing stages. The significance of these findings lies in challenging the "magic moment" view of lexical access, which assumes a single, discrete point of retrieval common to all tasks. Instead, the data support a model where performance is determined by the interaction of common lexical retrieval processes and task-specific mechanisms. For instance, the LDT appears to allow for fast, familiarity-based decisions that are less reliable for low-frequency words, leading to increased skew. Conversely, naming tasks may involve nonlexical routes that mitigate the frequency effect. By demonstrating that frequency affects different parameters of the RT distribution depending on the task, the study provides evidence that word identification is not a unitary process but involves distinct contributions from lexical access, semantic retrieval, and motor production.

Key finding

Low-frequency words yielded more skewed reaction time distributions than high-frequency words in most word identification tasks, with the exception of delayed naming and semantic categorization, indicating that task-specific processes modulate the effects of lexical access.

Methodology

lab_experiment

Sample size: 74

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via author_sweep_intake on 2026-05-28.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• temporal