Thi: paper describe s research whose goal is to determine the implications 01' verbal classificatory judgrnents for recognition rneruorv and recall. TOII'ard this end, Ss were required to answer 100 queries 01' attribution and superordlnaticn Ils a TW[NGI:: sudden? Is SP1~ACH ecstatic? Is a CORKSCREW an opener Is a DCNGEO;-'; 1 scholar") befare being tested unexpectedly on their ability to rernernber either the uppercase "keywords" or the lowercase "descriptors." l.exical mernorv did not depend on whether a ward had been part 01' an attribu tiv, or a superordinate query. But words from "incongruous" queries almost invariably were more poorly remembered-under conditions 01' free recall, cued recall. and recognition mernorv-than word s from "congruous" querie s. Ccngruous cues, but not incongruous ones, greatly facilitated recall. with keyword s being morc effective cues than descriptors. Recognition mernory 01' keywords was uniforrnly superior to that 01' descriptors. It is argued that the large and pervasive mernorial advantages 01' congruity arise because a congruous query, unlike an incongruous one, fosters a relational encoding 01' keyword and descriptor.

The method of free response refers to the following listening situation. Against a background of noise, a weak signal is presented several times in a long (2-min) observation interval. The temporal intervals between the presentations of the tones are randomly distributed; consequently, the listener does not know when a tone will occur, and he does not know how many tones will be presented. From one series of observation intervals to the next, the listener is instructed to adopt various criteria and to press the single response-key each time he “hears a tone.” The problem consists in the determination of a procedure that allows the total number of yes responses to be partitioned meaningfully between “hits” and “false alarms.” A model is developed in which the measurable quantity, rate of response, is related to the “hit rate” and to the “false-alarm rate.” Although the criterion adopted by the listener cannot be directly evaluated, the use of a wide range of criteria makes it possible to estimate the detectability ds of the signal. Two experiments are described, and the results support the model.

No abstract

A word list was designed so that half its words would denote targets when any of a number of target classes were defined. After scanning this list for targets, subjects were unexpectcdly testcd on their ability to recognize the words they had scanned. Recognition memory depended on how targets had been defined: a search for members of a semantically defined target class resulted in better recognition than a search for targets structurally defined. As a group, targets were better recognized than non-targets after a search for words that (a) contained the letter A, (b) denoted living things, or ( c ) denoted geographical locations. Recognition memory declined by as much as 75 per cent in d' from the &st quarter of testing to the last. Items prior to a word's input as well as decisions prior to its testing diminished its recognizability, whereas items following its input exerted comparatively little deleterious effect.

Word length and rarityOperating characteristics were used to assess the importance of word-frequency and word-length in recognition memory. It was much easier for Ss to distinguish between old and new rare words than between old and new common words. Rare (but not common) words were easier to recognize when they were polysyllabic than when they were not.