Academic journal article Memory & Cognition

The Role of Recollection and Familiarity in the Context Variability Mirror Effect

Academic journal article Memory & Cognition

The Role of Recollection and Familiarity in the Context Variability Mirror Effect

Article excerpt

Context variability refers to the number of preexperimental contexts that are associated with concepts. In four experiments, we investigated the basis for increased recognition memory for low context variability words. Low context variability was associated with greater recollection in the hit rates, and high context variability was associated with greater familiarity in the false alarms. Shortening the study time reduced recollection, but low context variability still influenced recollection in the hit rates. A modality change from study to test also reduced recollection but preserved recollective differences for low versus high context variability items. One interpretation of the results suggests that low context variability evokes more specific and, perhaps, idiosyncratic recollective associations during learning and that these associations support better recognition in the nit rates. By contrast, activating the larger number of associations for high context variability items may be mistaken for familiarity in the false alarm rates.

A mirror effect occurs in recognition memory tests when a particular class of items produces a higher hit rate and a lower false alarm rate, as compared with another class of items. For example, low-frequency words usually yield a greater hit rate than do high-frequency words, but false alarms are made less often to low-frequency words than to high-frequency words (see, e.g., Glanzer & Adams, 1985; Hirshman & Arndt, 1997; Hockley, 1994). A variety of stimulus comparisons produce mirror effects, such as manipulations of meaningfulness, frequency, concreteness, pictures versus words, and so forth. Steyvers and Malmberg (2003) added to the list of variables producing a mirror effect by reporting that hit rates were mirrored in false alarm rates on the basis of a dichotomy of high versus low normative context variability. They defined context variability as the number of preexperimental contexts in which a word is encountered (cf. Dennis & Humphreys, 2001). A word such as summit is a low context variability word, because it is encountered mainly in conversations and texts having to do with earth formations or mountain climbing. By contrast, a word such as meal is encountered in many more preexperimental contexts in everyday life and, therefore, is a high context variability word. As with word frequency, low context variability yielded higher hit rates and lower false alarm rates, as compared with high context variability items. The goal of the present study was to explore the basis of the context variability effect. To do so, we used the remember-know procedure to obtain approximations of recollection and familiarity, in order to understand which component of recognition contributes to the hit rate and false alarm rate advantages (e.g., Jacoby, Yonelinas, & Jennings, 1997; Kishiyama & Yonelinas, 2003).

One working hypothesis is that the context variability mirror effect is similar in kind to the word frequency effect. If this is so, context variability should behave in a manner consistent with, say, Joordens and Hockley's (2000) account of the word frequency mirror effect. They argued elegantly, from both data and theory, that the hit rate advantage enjoyed by low-frequency words is due to greater recollection from the study episode (cf. Guttentag & Carroll, 1997; Reder et al., 2000). When recollection is reduced by longer retention intervals or through very brief encoding, the hit rate advantage for low-frequency words is attenuated or eliminated, but the word frequency differences in the false alarm rates remain unchanged (see also Malmberg & Nelson, 2003). According to their account, preexperimental familiarity drives the differences in the false alarm rates, and therefore, high-frequency words will virtually always yield higher false alarm rates than will low-frequency words. Joordens and Hockley, as well as Gardiner and Java (1990), have found greater proportions of remember (R) responses for low- than for high-frequency words for hits. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.