Highly skilled participants and failures to redirect attention: Two plausible reasons for failing to replicate Paap and Noel's effect.

Abstract

The current authors argue that Pexman and Lupker's failures to replicate can be attributed to 2 factors: (a) their memory-load task did not redirect participants' attention away from the naming task, and (b) their participant population appears to have consisted of homogeneous and highly skilled readers. Pexman and Lupker also claimed that our results may have been a Type 1 error caused by bad luck and idiosyncratic differences in item difficulty. The authors show that the probability of this misfortune is extremely small. Three simulations explore the use of S. E. Bernstein and T. H. Carr's (1996) criteria for selecting individuals with different reading styles, and new evidence is presented that shows that individual differences in reading skill assessed a priori can be used to predict performance in Paap and Noel's task.We thank Penny Pexman, Steve Lupker, Stuart Bernstein, and Tom Carr for their interest and insights on the task of naming words under a memory load. Despite clear differences of opinion, they worked hard to make sure that no stone was left unturned in bringing forth our view of this complicated story. We also thank Eamon Strain, Karalyn Patterson, and Max Coltheart for thencontributions to the paradigm and to our article.

Word naming and memory load: Still searching for an individual differences explanation.

“…Only in Becker's (1976) experiment, when both lexical decisions and tone responses were required sequentially on the same trial, was there evidence that a concurrent task (in this case, an auditory tone-detection task) had a larger effect on low-frequency words than on high-frequency words (the regularity of the words was not reported). When all of this research is considered, it seems clear that only Becker's (1976) data and Paap and Noel's (1991) data provide any support for Paap and Herdman's (1998) claim that on the basis of previous research, one would expect "that a concurrent memory task should interfere more with low-frequency words than with those of high frequency" (p. 846). Further, given that Becker used a tone-detection task rather than a memory task, the support that those data can provide for this claim are meager at best.…”

“…More importantly, there was no evidence for the other Paap and Noel's (1991) pattern, an increased frequency effect for regular words. 4 Of the four points that we have made in the immediately preceding paragraphs regarding the question of the size of the memory-load effect, Paap and Herdman (1998) have specifically addressed one of them-our claim that the high-frequency word data demonstrate that our participants were being affected by the memory load. Paap and Herdman are unconvinced by these data and, further, go on to suggest that our inability to obtain larger load effects for lowfrequency regular words than for high-frequency words is a problem in and of itself, independent of the issues surrounding Paap and Noel's (1991) theorizing.…”

Paap and Noel revisited: Individual differences in independent word recognition tasks in relation to the dual-task paradigm

“…Using a priori selection tasks of no load or low digit recall conditions, Pexman and Lupker (1998) failed to find release from competition, observing a weak effect only when participants were selected on the basis of their naming performance within the Paap and Noel dual task itself. It appears that the Bernstein and Carr criteria are unreliable for selecting the same sub-set of participants across tasks (Paap & Herdman, 1998;Pexman & Lupker, 1998). Paap and Herdman (1998) argue that Pexman and Lupker's (1998) consistent failure to find release from competition may have occurred because they failed to orient participants towards the digit load task.…”

“…Although not reporting an item analysis, Herdman and Beckett (1996) provide the only clear replication in their Experiment 1. Herdman, Thompson, Greenham, andPiazze (1994, cited in Paap &Herdman, 1998) report a study where individuals who were self-identified poor readers, supported by performance on a standardized comprehension test, exhibited release from competition. However, in this study, low-frequency regular words were also facilitated by the higher digit load.…”

“…Only the participants exhibiting these criteria showed release from competition and a three-way interaction, although this failed to reach significance by items. Critics of the Bernstein and Carr study argue that they used criteria for allocating participants to groups that were likely to produce release from competition as a result of regression to the mean, rather than any real differences in word recognition architecture (cf., Bernstein et al, 1998;Paap & Herdman, 1998;Pexman & Lupker, 1998). This concept has been rigorously investigated by Pexman and Lupker (1998).…”