Imagery encoding and false recognition errors: Examining the role of imagery process and imagery content on source misattributions

Abstract

Imagery encoding effects on source-monitoring errors were explored using the Deese-Roediger-McDermott paradigm in two experiments. While viewing thematically related lists embedded in mixed picture/word presentations, participants were asked to generate images of objects or words (Experiment 1) or to simply name the items (Experiment 2). An encoding task intended to induce spontaneous images served as a control for the explicit imagery instruction conditions (Experiment 1). On the picture/word source-monitoring tests, participants were much more likely to report "seeing" a picture of an item presented as a word than the converse particularly when images were induced spontaneously. However, this picture misattribution error was reversed after generating images of words (Experiment 1) and was eliminated after simply labelling the items (Experiment 2). Thus source misattributions were sensitive to the processes giving rise to imagery experiences (spontaneous vs deliberate), the kinds of images generated (object vs word images), and the ways in which materials were presented (as pictures vs words).

No imagination effect on boundary extension

“…Unlike memory for actions (Thomas et al 2003), there was not an inflation of memory errors following more elaborate encoding for BE. We also found no reduction of BE, as had been observed for imagined objects (Foley et al 2010), suggesting that memory was not improved by the act of imagining additional details.…”

“…Regardless of whether we asked them to imagine additional sounds, odors, tastes, or even visually what more of the photograph would look like, elaboration did not increase BE. Elaboration also did not reduce BE, which might have occurred if the act of generating additional information improved accuracy, as Foley et al (2010) had found within the Deese/Roediger-McDermott paradigm for word and picture lists.…”

“…However, the act of imagining isolated objects improves memory, reducing source memory errors (Foley et al 2010). Being specifically asked to elaborate either cross-modally (Exps.…”

“…Using the Deese-Roediger-McDermott paradigm with participants asked to remember lists of pictures or words, Foley et al (2010) found reduced errors when participants were instructed to deliberately generate images of pictures or words. They suggested that the act of generating the images distinguished these items from the actual list items, and thus reduced source memory errors.…”

“…Foley (2012) highlighted a distinction between spontaneous imagery and intentional self-generated imagery; the present research included both, and we will return to this distinction in the General Discussion. To the extent that scene representation, as measure by BE, is related to imagining isolated objects (the task in Foley et al 2010), additional imagery instructions might reduce BE.…”