City Research Online

Abstract

The main objective of the thesis was to investigate how an act of recall is represented in memory and monitored. The thesis attempted to understand the effects recall might have on the underlying representation of an item in terms of (i) episodic theory, and (ii) the pre-theoretical assumption that a memory trace constitutes a collection of attributes. A further aim of the thesis was to trace the development of the ability to assess previous recall performances.

These aims were pursued in three sets of studies. In the first it was proposed that performance features derived from the articulatory and sensory feedback of a response during recall become established as part of the attribute ensemble defining an item in memory. It was supposed that these changes would at least partly mediate memory for past recall and also influence learning in a multi-trial free recall situation. These proposals were supported in that both memory for remembered events and the rate of learning varied directly as a function of the number of performance features available during recall.

The development of the ability to assess past recall was investigated in a second set of studies. Contrary to recent findings, young children were found to be relatively poor in assessing past recall. This was true even where the memory task had been otherwise modified to suit the vocabulary and estimated memory spans of the specific age groups involved.

The notion that recall might result in the formation of a unique trace in episodic memory was tested in a third set of studies, by means of the paradigm developed by Murdock and Anderson (1975). Little support for this hypothesis or for Murdock and Anderson's model was found, however. A second experiment undertook to investigate their ‘conveyor belt’ model of memory further but again the results were not completely definitive.

The question of whether an interpolated recall task influences recognition of items which have been presented but not recalled, was pursued in a final study. Whilst initial testing made no difference to overall recognition scores a closer analysis seemed to show that recognition of non-recalled items was worse following recall, than recognition of comparable items was where no recall trial had intervened.

It was concluded that recall has effects upon the underlying representation of an item which are different to those of presentation and constitute more than a simple increment in trace strength. It was also suggested that this influence was unlikely to be uniform and constant but will vary, for instance, with trials in multi-trial recall. It remains uncertain as to whether or not the influence of recall extends to the formation of a trace separate to that formed at presentation. Young children's relative inability to appraise their recall was speculated to perhaps underlie their well documented failure to make spontaneous use of mnemonic strategies.

Publication Type:	Thesis (Doctoral)
Subjects:	B Philosophy. Psychology. Religion > BF Psychology
Departments:	Doctoral Theses Doctoral Theses > School of Arts and Social Sciences Doctoral Theses

Export

Downloads