Toward a complete decision model of item and source recognition

Abstract

In a recognition memory test, subjects may be asked to decide whether a test item is old or new (item recognition) or to decide among alternative sources from which it might have been drawn for study (source recognition). Confidence-rating-based receiver operating characteristic (ROC) curves for these superficially similar tasks are quite different, leading to the inference of correspondingly different decision processes. A complete account of source and item recognition will require a single model that can be fit to the entire data set. We postulated a detection-theoretic decision space whose dimensions, in the style of Banks (2000), are item strength and the relative strengths of the two sources. A model that assumes decision boundaries of constant likelihood ratios, source guessing for unrecognized items, and nonoptimal allocation of attention can account for data from three canonical data sets without assuming any processes specifically devoted to recollection. Observed and predicted ROCs for one of these data sets are given in the article, and ROCs for the other two may be downloaded from the Psychonomic Society's Archive of Norms, Stimuli, and Data, www.psychonomic.org/archive.

Recognition memory zROC slopes for items with correct versus incorrect source decisions discriminate the dual process and unequal variance signal detection models.

“…If source discrimination relies primarily on the high-threshold recollection process, then the rec ognition zROCs should have a shallower slope for source-correct than source-incorrect items and the refined source zROC must be u-shaped (see Figure ,3, Panel 1). If familiarity drives source 3 The BBG model is similar, but not identical, to the proposal that participants make random source guesses for any item they fail to recog nize (Hautus et al, 2008;Slotnick & Dodson, 2005). One critical differ ence is that the point where the bounded distributions converge is not determined by a participant's criterion for making an "old" response, so in the BBG model, it is possible for "new" responses to have above chance source discrimination if the recognition criterion is sufficiently conserva tive (as observed by Stams, Hicks, Brown, & Martin, 2008).…”

“…For example, the model could predict that if one selects out targets that received a high confidence "new" response, then the Source A items should be consistently attributed to Source B, and vice versa (Hautus et al, 2008). Of course, this pattern is not observed in empirical data.…”

Receiver operating characteristic analysis of eyewitness memory: Comparing the diagnostic accuracy of simultaneous versus sequential lineups.

“…Indeed, ROC analysis is widely used to differentiate between competing theories of recognition memory in the experimental psychology literature (e.g., Hautus, Macmillan, & Rotello, 2008;Heathcote, 2003;Mickes, Johnson, & Wixted, 2010). However, as Clark et al (2011) observed, theoretical accounts of eyewitness memory are in short supply.…”

Paying Attention to Attention in Recognition Memory

“…However, recent UV SDT extensions acknowledge the importance of encoding-related factors (Hautus, Macmillan, & Rotello, 2008). Additionally, assessments of the competing views of recollection have often focused on either neuroimaging or cognitive modeling (see Malmberg, 2008 for review).…”