Switch and bait: probing the discriminative basis of odor identification via recognition memory.

When people misidentify everyday odors, as they often do, their errors may conceivably lie in faulty perceptions or in faulty access to the names. Discussions of the matter usually focus on the latter, as if people had no problems with perceptual accuracy. (The problem of faulty access may get attention because its high subjective impact makes it particularly memorable, when it does occur.) However, studies have demonstrated breakdowns in ability to discriminate quality, from which it follows that people will misidentify items through perceptual confusions. Furthermore, misidentifications often contain considerable information about the identities of items, as if people simply did not perceive the items accurately, but perhaps fuzzily or with some perceptual bias. Recognition memory, with a 2-day interval between inspection and test, provided a vehicle to address two questions on this topic: (i) Would people notice that we had switched items and had presented for recognition items that matched their misidentifications rather than the original items inspected? (ii) Would people not only fall for the false bait, but actually identify the switched items correctly, and thereby imply that they were 'tuned' to perceive those odors? People commonly failed to notice the switches, i.e. took the bait and commonly identified the switched items with veridical names. Although subject to further study, the outcome suggests that when people give such names as garlic for vinegar, orange for lime, soy sauce for molasses and many others, the errors often lie largely at a perceptual stage of processing, i.e. at input rather than output.