Shades of the mirror effect: Recognition of faces with and without sunglasses

Abstract

A mirror effect was found for a stimulus manipulation introduced at test. When subjects studied a set of normal faces and then were tested with new and old faces that were normal or wearing sunglasses, the hit rate was higher and the false alarm rate was lower for normal faces. Hit rate differences were reflected in remember and sure recognition responses, whereas differences in false alarm rates were largely seen in know and unsure judgments. In contrast, when subjects studied faces wearing sunglasses, the hit rate was greater for test faces with sunglasses than for normal faces, but there was no difference in false alarm rates. These findings are problematic for single-factor theories of the mirror effect, but can be accommodated within a two-factor account.The mirror effect refers to the regularity of recognition memory in which classes of stimuli that are relatively easy to identify as old when old are also relatively easy to classify as new when new. In other words, the difference in discriminability is observed in both hit and false alarm rates rather than in only one of these measures. Glanzer and Adams (1985), in a review of80 recognition experiments, showed that the mirror effect can be seen for a variety of stimulus manipulations (e.g., natural language word frequency, concreteness, meaningfulness, and pictures vs. words) and that it holds for both yes/no and forced-choice recognition test procedures.The mirror effect has been extended to tests of associative recognition (Greene, 1996;Hockley, 1994) as well as discriminations of order (Greene, 1996) and frequency (Greene & Thapar, 1994); it is also reflected in measures ofresponse latency (Hockley, 1994). Hintzman, Caulton, and Curran (1994) have further demonstrated that dividing attention at test, or using a response-signal procedure to control processing time at retrieval, does not eliminate or attenuate the mirror effect. Hintzman et al. concluded that the mirror effect "may best be attributed to the inherent nature of the retrieval and judgment processes that underlie recognition memory" (p. 286). Hintzman et al. (1994) further noted that the mirror effect has proven to be difficult to account for in the context of most current models of recognition memory. This is because most models view recognition memory in terms

Effect of retention interval on the simultaneous cognate-noncognate and remember-know mirror effects

“…When MEMORY, 2006, 14 (1), 79±86 recollection is reduced at testing, the hit advantage is attenuated or reversed, leaving the false alarms untouched. This reduction is also present when the overall results are partitioned as a function of confidence or r-k response categorisation (Hockley, Hemsworth, & Consoli, 1999;Joordens & Hockley, 2000). When pseudowords are not recollectable, due to their novelty, they do not show a mirror pattern.…”

Matching Face Images Taken on the Same Day or Months Apart: the Limitations of Photo ID

“…However, wearing glasses produced fewer matching errors than the no disguise condition (but see Hockley, Hemsworth & Consoli, 1999;Terry, 1994), and these two conditions did not differ from the hat condition (but see Henderson et al, 2001;.…”

Familiarity and recollection in item and associative recognition

“…The remember/know response procedure was also used in this experiment. (This study is reported as Experiment 1 by Hockley, Hemsworth, & Consoli, 1999.) The remainder of the interval was occupied in conversation with the experimenter.…”