Bizarreness and recall

Abstract

Although memory experts have long advocated the use of bizarre imagery to aid memory, most research has failed to find bizarreness to be effective. Recently, bizarreness has been shown to be effective, but those experiments involved mixed lists in which each subject had both bizarre and nonbizarre materials. The goal of the present research was to use the same materials but in an unmixed-list design in which each subject had only bizarre or only nonbizarre sentences.Contrary to previous research, there were no significant differences between bizarre and nonbizarre conditions in terms of the number of individual words recalled, and bizarreness produced significantly lower recall of complete sentences. It was concluded that bizarreness is effective only when bizarre and nonbizarre materials are provided by the experimenter in a mixed-list situation.

Tests of a Cognitive-Resource-Allocation Account of the Bizarreness Effect

“…First, processing time may not be the best measure of attention. Although Cox and Wollen (1981) and Wollen and Cox (1981) equated processing time with attention in their explanation of the bizarreness effect, earlier models that included reference to attentional mechanisms suggested that attention could be measured in a different way. For example, Merry (1980) and Merry and Graham (1978) described attention not in terms of processing time but in terms of cognitive-resource allocation-that is, they suggested (a) that comprehension of bizarre information requires more cognitive resources (not merely more time) than does comprehension of common information and (b) that the increased effort explains the memory advantage for bizarre items relative to common items.…”

Assessing a retrieval account of the generation and perceptual-interference effects

“…In the mixed-list condition, the retrieval set encompasses an assortment of usual and unusual items, conferring relative distinctiveness onto the unusual condition, which in turn produces the typical effect in recall. This possibility has been raised with several variables that are subject to design effects, such as generation and bizarreness (Begg et al, 1989;Cox & Wollen, 1981;McDaniel et al, 2000;McDaniel et al, 2005;Schmidt, in press). …”

Further constraints on the bizarreness effect: Elaboration at encoding Bridget Robinson-Riegler

“…Another important factor in determining the bizarreness effect is list structure (mixed lists, in which both bizarre and common sentences are presented, vs. unmixed lists, in which all sentences presented are bizarre or common). The majority of studies done with unmixed lists (Collyer, Jonides, & Bevan, 1972;Cox & Wollen, 1981;Marshall, Nau, & Chandler, 1980) have not resulted in superior retention for bizarre as opposed to common items (see Einstein, McDaniel, & Lackey, 1989, for an important exception). In contrast, the majority of studies done with mixed lists (McDaniel & Einstein, 1986;O'Brien & Wolford, 1982;Pra Baldi et al, 1985) have resulted in enhanced free recall for bizarre as opposed to common items.…”