Cognitive control during selection and repair in word production

ABSTRACT

Word production entails selection of lexical items and their relevant segments among competitors, as well as monitoring and repair processes. In two experiments, we studied the control processes involved in selection (selection control) and covert-error- interception (post-monitoring control). Selection control was studied by manipulating the overlap (contextual similarity) either in semantics or in segments between two objects that participants repeatedly named. Post-monitoring control was examined by asking participants to switch the name of the two objects, thus suppressing a prepotent response (reversal). Results showed robust costs of contextual similarity (for both semantic and segmental overlap) and reversal, but the two did not interact. Moreover, we found no reliable correlation between the cost of contextual similarity at the semantic and segmental levels, suggesting stage-specific selection control processes. The reversal cost, however, was reliably correlated between semantically and segmentally related pairs, implying a control process that is shared by both stages of production.

KEYWORDS:

• Spoken and written word production
• executive function
• cognitive control
• semantic interference
• monitoring and repair

Acknowledgments

We would like to thank Bob Slevc for his useful comments on the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Nazbanou Nozari http://orcid.org/0000-0002-2215-8533

Notes

1. Note that this is a finding from picture–picture interference paradigms, and not from the picture–word interference paradigms in which a semantically related word distractor often creates interference.

2. We did not include degree of similarity, because the critical predictions of Experiment 2 do not depend on that variable. To ensure that this exclusion did not significantly change the model fit, we tested the reported model against a full model including degree-of-similarity and all its interactions. The change in fit was not significant (χ² = 11.7, p = .16).

3. To assess the correlation without the influence of potential outliers, we also calculated the non-parametric Spearman's rank-ordered correlation index, which was −0.08, p = .56.

4. The non-parametric Spearman's rank-ordered correlation index for this correlation was 0.52, p = .002.

Additional information

Funding

This work was supported by the Therapeutic Cognitive Neuroscience Fund awarded to B. Gordon, and the NIH [grant DC012283] to B. Rapp.