A series of five experiments explore the influence of articulatory suppression on immediate memory for auditorily presented items with a view to testing the revised concept of an articulatory loop. Experiments 1, 2 and 3 demonstrate that the phonological similarity effect is not abolished by articulatory suppression, whether this occurs only at input or at both input and recall. Experiments 4 and 5 show that the tendency for long words to be less well remembered than short is abolished by articulatory suppression, even when presentation is auditory, provided suppression occurs during both input and recall. These results are consistent with the concept of a loop comprising a phonological store, which is responsible for the phonological similarity effect, coupled with an articulatory rehearsal process that gives rise to the word length effect.

A series of nine experiments studied the disrupting effect of a secondary task on retrieval from long-term memory. The first five experiments studied the influence of concurrent card sorting or digit span on free recall or paired-associate learning of word lists. When performance was measured in terms of recall probability, the concurrent tasks produced a clear impairment of learning but a much smaller and less consistent impairment of retrieval. This result contrasts with earlier studies by other investigators who used performance on the secondary task as their measure. These tended to show greater disruption at retrieval. Such studies typically use latency measures, and Experiment 6 explored this by using a recognition paradigm in which both accuracy and latency could be measured simultaneously. Concurrent digit load had no significant effect on recognition probability but did produce a clear increase in response latency.Experiment 7 explored the latency effect further, using a semantic memory paradigm. Subjects were required to verify sentences while retaining sequences ranging from zero to nine random digits. Decision latency increased monotonically with digit load, whereas accuracy did not decline significantly until the load exceeded six digits. Experiment 8 required subjects to make semantic category judgments while retaining sequences of six digits. Response latency was increased by concurrent digit load, but the magnitude of this effect was substantially the same for easy and difficult decisions. Concurrent load had no statistically reliable effect on errors. Finally, Experiment 9 examined the effect of concurrent digit load on the rate of generating items from semantic categories. There was a clear and consistent effect.Conclusions are as follows: (a) A demanding concurrent task does not substantially reduce the probability of retrieving an item from either episodic or semantic memory. This suggests the possibility that the process of memory search may be largely automatic, (b) In contrast, concurrent load during learning has a substantial effect on recall performance. This is interpreted in terms of the total time hypothesis; a task that distracts attention from learning will reduce the amount learned, (c) A concurrent task during retrieval does have a clear effect on latency. It is suggested that this reflects response competition at output, (d) The contrast between the pattern shown by errors and that shown by latencies suggests a general methodological conclusion: Attempts to estimate the attentional demands of any task should be interpreted with considerable caution when based on a single measure, be it performance errors, performance latency, or response to a probe reaction time signal. General conclusions demand converging operations, not single paradigms.518

Brown ( I 976) has provided an analysis of the effect of the memorability of an item on the confidence with which it is accepted or rejected in a test of recognition or recall. When the subject has no clear recollection of the inclusion of an item in an input list, he is assumed to evaluate its memorability in the context of the experiment before he decides whether to accept or reject it. If the judged memorability is high, the absence of a clear recollection is stronger evidence against the item than if it is low. A specific prediction is that memorable distractors in a recognition test will be more confidently rejected than nonmemorable ones. This prediction was tested and confirmed in three experiments in which recognition was tested by +category rating.Except in Experiment I, items memorable to individual subjects were identified by administering a questionnaire. For example, in Experiment I11 forenames of immediate family were assumed to have high memorability. This experiment also included word frequency as a variable. Low-frequency distractors were rejected significantly more firmly than high-frequency distractors : extraction of memorable names enhanced this effect. The relationship of memorability to word frequency is discussed.

No abstract

A series of experiments explored the role of subvocalisation in fluent reading. Experiment I showed that when subjects were required to suppress articulation while reading, their ability to detect anomalous words or errors of word order in prose was markedly impaired although speed of reading was unaffected. Experiment II showed that this decrement was not a general effect due to performing a secondary task, since a concurrent tapping task did not impair detection accuracy. A third study explored the role of acoustic interference in reading by requiring subjects to detect errors in prose while attempting to ignore irrelevant speech, with or without articulatory suppression. Once again articulatory suppression led to a clear decrement in the subject's ability to detect errors, while unattended speech had no effect on performance. None of the manipulations influenced the speed with which the subjects performed the reading task. It is concluded that subvocalisation allows the creation of a supplementary articulatory code which is produced and utilised in parallel with other aspects of reading. Such a code seems particularly suitable for monitoring order information.