ABSTRACT
Background: A number of studies with healthy participants have shown that the retrieval of a word results in faster subsequent retrieval of the same word and slower retrieval of a semantic coordinate. Similarly, studies have shown that naming in people with aphasia benefits from prior repetition of the same word. However, it remains unclear how previous retrieval of a semantically coordinated and/or associated word affects subsequent word retrieval in aphasia.
Aims: To determine whether production of a semantic coordinate and/or associated word facilitates word retrieval in a case series of people with aphasia.
Methods & Procedures: Two experiments were conducted. Experiment 1 involved 12 people with chronic aphasia and impaired picture naming. It examined the effect of repeating an associated coordinate (e.g., car) in the presence of a picture (e.g., bus) affected later picture naming of the same picture (bus), when compared to repetition of identity primes (bus), unrelated primes (e.g., sea) and unprimed items. Experiment 2 used the same paradigm with nine people with aphasia to examine the effect of primes that were either only associates (e.g., conductor) or coordinates (e.g., ferry) of the target (bus).
Outcomes & Results: Both experiments showed significant improvement in accuracy and facilitation of naming latencies in the identity condition. In Experiment 1, significant interference was detected at the group level for the associated coordinates, with increased reaction times. However, no significant effects were observed from association and coordination independently in Experiment 2.
Conclusions: Given that repetition of associated coordinates impairs subsequent naming but there was no identifiable effect of items that were only either associated or coordinated, it seems likely that semantic interference is most prevalent when primes are more closely semantically related to the target. The finding of interference from these closely related associated coordinates (e.g., car-bus) has implications for treatment paradigms that include such items.
Acknowledgments
First and foremost, we would like to thank to all the participants who kindly agreed to take part in this research. We also would like to thank to Nick Riches and Matthew Miller for their help in auditory recording of the stimuli.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Podraza and Darley do not provide statistics at the single subject level, however, we have used the rule of thumb that in a McNemar’s test at least 6 items difference between conditions is required for a significant result.
2. Note that Howard et al. (Citation1985) called this condition “associate” but clearly state that these items were “the most closely related member of the [target’s] category of which we had a picture” (p61).
3. In contrast to analysis in Experiment 1, syllables and phonemes were not included as random slopes, because the model failed to converge.