Working Memory Capacity as a Determinant of Proactive Interference and Auditory Distraction.

Individual differences in working memory capacity are related to performance on a range of elemental and higher order cognitive tasks. The current experiment tests the assumptions of two theoretical approaches to working memory capacity: working memory as executive attention and working memory as temporary binding. These approaches are examined using a short-term updating task where proactive interference is manipulated, such that old responses have to be suppressed in favour of new responses. A second source of distraction is introduced by way of irrelevant, to-be-ignored background speech that accompanies presentation of the list items. This speech reinforces either the to-be-remembered item on the current list, or the to-be-suppressed item. Working memory capacity was significantly related to overall level of correct performance on the short-term task, and to the degree of proactive interference experienced. However, there was no evidence for individual differences in the ability to suppress the interfering foil, nor in priming effects associated with the irrelevant speech. The results provided little support for the working memory capacity as executive attention perspective, some evidence for the binding perspective, but also evidence supporting the fact that some effects of distraction are not under voluntary control.

Redintegration, task difficulty, and immediate serial recall tasks.

While current theoretical models remain somewhat inconclusive in their explanation of short-term memory (STM), many theories suggest at least a contribution of long-term memory (LTM) to the short-term system. A number of researchers refer to this process as redintegration (e.g., Schweickert, 1993). Under short-term recall conditions, the current study investigated the effects of redintegration and task difficulty in order to extend research conducted by Neale and Tehan (2007). Thirty participants in Experiment 1 and 26 participants in Experiment 2 completed a serial recall task in which retention interval, presentation rate, and articulatory suppression were used to modify task difficulty. Redintegration was examined by manipulating the characteristics of the to-be-remembered items; lexicality in Experiment 1 and wordlikeness in Experiment 2. Responses were scored based on correct-in-position recall, item scoring, and order accuracy scoring. In line with the Neale and Tehan results, as the difficulty of the task increased so did the effects of redintegration. This was evident in that the advantage for words in Experiment 1 and wordlikeness in Experiment 2 decreased as task difficulty increased. This relationship was observed for item but not order memory, and findings were discussed in relation to the theory of redintegration.

Phonological effects in forward and backward serial recall: qualitative and quantitative differences.

Forward serial recall is affected by a diverse range of phonological factors that are readily replicated and relatively well understood. In contrast with backward recall, these phonological effects are not consistently replicable in that some studies show that the effects are present and some show the effects are absent or severely attenuated. Moreover at the theoretical level there is no consensus about how participants perform backward recall. The current research was aimed at understanding the differences between forward and backward recall by using meta-analytic techniques on 16 previously published experiments that examined the effects of benchmark phonological factors on both forward and backward recall. In each of the studies, recall was decomposed into 2 components, the first 2 items output and the remaining later responses. A consistent pattern emerged in the data. Each effect was present in both the early and late output positions in forward recall. The effects were present in the late output positions in backward recall, but the effects were weaker than in forward recall. The phonological variables had little impact on early output in backward recall (with the exceptions of articulatory suppression). The presence of qualitative differences between forward and backward recall and quantitative differences between studies have implications for the theoretical understanding of direction of recall in immediate memory tasks.

Backward recall and benchmark effects of working memory.

Working memory was designed to explain four benchmark memory effects: the word length effect, the irrelevant speech effect, the acoustic confusion effect, and the concurrent articulation effect. However, almost all research thus far has used tests that emphasize forward recall. In four experiments, we examine whether each effect is observable when the items are recalled in reverse order. Subjects did not know which recall direction would be required until the time of test, ensuring that encoding processes would be identical for both recall directions. Contrary to predictions of both the primacy model and the feature model, the benchmark memory effect was either absent or greatly attenuated with backward recall, despite being present with forward recall. Direction of recall had no effect on the more difficult conditions (e.g., long words, similar-sounding items, items presented with irrelevant speech, and items studied with concurrent articulation). Several factors not considered by the primacy and feature models are noted, and a possible explanation within the framework of the SIMPLE model is briefly presented.

Is spoken duration a sufficient explanation of the word length effect?

The word length effect is one of the cornerstones of trace decay plus rehearsal models (TDR) of memory. Words of long spoken duration take longer to rehearse than words of short spoken duration and as such suffer more decay and are thus less well recalled. The current experiment manipulates both syllable length and spoken duration within words of fixed syllable length in an aim to test the assumptions of the TDR model. Our procedures produced robust effects of both syllable length and spoken duration in four measures of the time it takes to pronounce the different types of words. Serial recall for the same materials produced robust syllable effects, but no duration effects.

The role of context in producing item interactions and false memories.

Cued recall with an extralist cue poses a challenge for contemporary memory theory in that there is a need to explain how episodic and semantic information are combined. A parallel activation and intersection approach proposes one such means by assuming that an experimental cue will elicit its preexisting semantic network and a context cue will elicit a list memory. These 2 sources of information are then combined by focusing on information that is common to the 2 sources. Two key predictions of that approach are examined: (a) Combining semantic and episodic information can lead to item interactions and false memories, and (b) these effects are limited to memory tasks that involve an episodic context cue. Five experiments demonstrate such item interactions and false memories in cued recall but not in free association. Links are drawn between the use of context in this setting and in other settings.