Repetition priming across distinct contexts: Effects of lexical status, word frequency, and retrieval test

Abstract

Repetition priming, the facilitation observed when a target is preceded by an identity prime, is a robust phenomenon that occurs across a variety of conditions. Oliphant (1983), however, failed to observe repetition priming for targets embedded in the instructions to an experiment in a subsequent lexical decision task. In the present experiments, we examined the roles of priming context (list or instructions), target lexicality, and target frequency in both lexical decision and episodic recognition performance. Initial encoding context did not modulate priming in lexical decision or recognition memory for low-frequency targets or nonwords, whereas context strongly modulated episodic recognition for high-frequency targets. The results indicate that priming across contexts is sensitive to the distinctiveness of the trace and the reliance on episodic retrieval mechanisms. These results also shed light on the influence of event boundaries, such that priming occurs across different events for relatively distinct (low-frequency) items.

Keywords:

• Repetition priming
• Recognition
• Context effects

The research reported was completed as partial fulfilment of the requirements for the doctoral degree for the first author, who was supported by a Dean's Dissertation Fellowship awarded by Washington University in Saint Louis. Additional funding was provided by the National Institute on Aging (NIA) Training Grant AG00030. Special thanks to Roddy Roediger, Larry Jacoby, and Mark McDaniel who served on the dissertation committee and to Kara Chin, Jessie Hu, and Myra Blake for assistance in data collection.

Notes

¹ Analyses on standardized RTs revealed highly similar patterns of results in both experiments but are not reported for the sake of brevity.

² Reported degrees of freedom are corrected for sphericity (Greenhouse–Geisser corrections are reported).

³ Although the absence of a lexicality effect between HF targets and nonwords is unusual, a potential explanation is that the nonwords were slightly shorter than the words. Thus, the length confound might have favoured faster responses to nonwords (Chumbley & Balota, 1984).

⁴ Because of substantial differences in baseline RTs (i.e., RTs to control items) between LF and HF targets, we also conducted an analysis on proportional measures of priming—that is, (control RT – repeated RT)/control RT (see Schnyer et al., 2007)—and also z scored transformed analyses (see Faust, Balota, Spieler, & Ferraro, 1999). Only an effect of frequency emerged, F(2, 118) = 9.33, p < .001, partial η²  = .14. Significant priming was found for LF items (M = .05), but no priming for HF words or nonwords (M = –.01 and M = –.03, respectively). The effect of encoding condition and the interaction were not reliable, both Fs < 1.0, ps > .40. Thus, the priming effect as a function of frequency was not due to base-rate differences in RTs.

⁵ One slightly surprising finding in the recognition test was the absence of any order effects for targets in the study list condition. Because of the well-documented effects of interference in memory (see M. C. Anderson & Neely, 1996, for a review), one might have expected differences in memory performance depending on the order in which the study list was presented. One possibility is that there were both effects of retroactive interference on the list when it was studied before the instructions and effects of proactive interference when it was studied second, basically equating the two conditions. The relative primacy and recency benefits might have further equalized performance.

⁶ We thank Glen Bodner for pointing this out.

⁷ Analyses on proportional RTs revealed a nonsignificant effect of word frequency, F(1, 29) = 2.63, p = .12, partial η² = .08. Overall, for LF targets the priming effect was .06; for HF words it was .02. The effect of encoding condition was not significant, F < 1, nor was the interaction, F < 1. However, follow-up t tests revealed significant priming only for LF targets in both the instructions and study list conditions, t(29) = 2.8, p = .009, and t(29) = 2.8, p = .008, respectively. Priming for HF targets was not significant, both ts < 1. ps > .35.

⁸ There are other possible mechanisms of priming to consider. For example, TAP accounts of priming attribute facilitation in repetition priming to the overlap in component processes (e.g., conceptual, perceptual processing) between the priming event and the transfer task, with maximal priming occurring when the two tasks are identical (Morris et al., 1977). In the present study, there was no clear evidence that differences in processing (i.e., conceptual vs. perceptual) modulated priming in LDT for LF or nonword targets, suggesting that for certain classes of items some other factor might be more critical. Perhaps these classes of items capture sufficient attention that they received more perceptual-level processing than HF targets. The attention to orthographic features of the LF targets would have facilitated performance on the LDT whereas HF words might be more likely to result in automatic processing and less sensitive to task-dependent processes (see Franks et al., 2000). Thus, it seems that a simple TAP account cannot fully accommodate the present data, without positing some additional mechanism or component process that is sensitive to target frequency. Another account of repetition priming in LDT is some variant of stimulus response learning (see Horner & Henson, 2009; Logan, 1990). Specifically, for positive transfer to be observed in LDT, one would expect that at some level participants code the stimuli in the text as words or nonwords. Because the instructions to Experiment 1 included nonwords, and several examples were given, it is possible that some covert classification of items did occur. In Experiment 2, nonwords were not included in the instructions or in the study list, yet the pattern of results was quite similar, suggesting that stimulus/response learning cannot fully accommodate these data.