Large-scale databases of proper names

Abstract

Few tools for research in proper names have been available-specifically, there is no large-scale corpus of proper names. Two corpora of proper names were constructed, one based on U.S. phone book listings, the other derived from a database of Usenet text. Name frequencies from both corpora were compared with human subjects' reaction times (RTs) to the proper names in a naming task. Regression analysis showed that the Usenet frequencies contributed to predictions of human RT, whereas phone book frequencies did not. In addition, semantic neighborhood density measures derived from the HAL corpus were compared with the subjects' RTs and found to be a better predictor of RT than was frequency in either corpus. These new corpora are freely available on line for download. Potentials for these corpora range from using the names as stimuli in experiments to using the corpus data in software applications.Proper names present an interesting puzzle to psychology. Although very important for adequate social performance, for instance, memory for proper names is inferior to memory for other types of information (Cohen & Burke, 1993). McWeeny, Young, Hay, and Ellis (1987) found that last names that connote some occupation (e.g., "Joe Baker went to the store") are more difficult to recall than if the actual occupation had been used ("The baker went to the store"). Burgess and Conley (1998) uncovered other interesting aspects of memory for proper names. They demonstrated, in the hyperspace analogue to language (HAL) semantic model of memory, that proper names cluster tightly in high-dimensional memory space, that ambiguity in proper names draws a name out of this pure name space toward words with more semantic depth, and that proper name semantic neighborhoods can encode information such as gender, It seems clear that proper names involve the storage and retrieval of important cognitive information; however, it has only lately been an area of broader investigation. Three of the 11 chapters in the recent volume Corpus Processingfor Lexical Acquisition deal with the issues involved in the representation of proper names (Boguraev & Pustejovsky, 1996). Cohen and Burke edited a special issue of Memory that contained papers on the experimental, neuropsychological, and theoretical aspects ofproper names. These examples provide evidence that research in proper names in all areas of psychology could be a rich field of inquiry. This research was supported by a NSF Presidential Faculty Fellow Award SBR-9453406 to C.B. We thank Maureen Keeney and Janet Eakins for their work in data collection. More information about research at the Computational Cognition Lab, a HAL demo, and reprint information can be found at http://HAL.ucr.edu. Correspondence concerning this article should be addressed to P. Conley, Psychology Department, 1419 Life Sciences Bldg., University of California, Riverside, CA 92521-0426 (e-mail: patrick@cassandra.ucr.edu).Unfortunately, the tools necessary to perform research in proper names are somewhat lacking. Specifically, ...

Charlie Brown Versus Snow White: The Effects of Descriptiveness on Young and Older Adults' Retrieval of Proper Names

“…For example, names are typically lower in overall frequency of use than other sorts of words (e.g., even James, a fairly high-frequency surname, is less commonly encountered than the biographical information is a student or the common noun house). Under this account, low word frequency, not status as a name per se, causes the increased difficulty in name memory (see Conley, Burgess, & Hage, 1999 for a discussion of name frequency). In addition, there is a greater ''set size'' of possible phonological sequences for names than there is for other types of words, and this wider variety of potential sound combinations might contribute to names' difficulty (e.g., Brennen, 1993).…”

Cognition, language, and aging

“…While Rastle and Burke (1996) found advantages for prior processing of phonological information in terms of reducing TOTs, they found that asking participants to first rate how pleasant each of the words were (semantic processing), prior to retrieving them from definition questions, did not provide any extra advantage in reducing TOTs above and beyond simply reading the phonology of the TOT word. Cross and Burke (2004) showed that the lack of facilitation from prior processing of semantic information on TOT incidence also occurred for proper names. In this experiment, young and older adults (mean age 72) were first asked a question about a famous character (e.g., Eliza Doolittle), were subsequently shown a picture of the person depicting that famous character (e.g., a picture of Audrey Hepburn portraying Eliza Doolittle), and were asked to produce the name of the portrayer (e.g., Audrey Hepburn).…”

“…When more phonological components are needed to retrieve the entire name phrase, there are correspondingly more opportunities for retrieval failures to occur, which explains why individuals tend to have more TOTs for targets who are known by three names (e.g., actor Sarah Jessica Parker) as opposed to just two (e.g., actor Julia Roberts; Hanley & Chapman, 2008;Stevenage & Lewis, 2005). Proper names are also generally accessed less frequently than nouns; even the most common names, like Smith, are accessed less often than other types of words, which makes the connections among conceptual, lexical, and phonological information within the proper name hierarchy comparatively weak and more prone to retrieval failures (e.g., Cohen & Burke, 1993;Conley, Burgess, & Hage, 1999).…”

Cognition, Language and Aging

“…When more phonological components are needed to retrieve the entire name phrase, there are correspondingly more opportunities for retrieval failures to occur, which explains why individuals tend to have more TOTs for targets who are known by three names (e.g., actor Sarah Jessica Parker) as opposed to just two (e.g., actor Julia Roberts; Hanley & Chapman, 2008;Stevenage & Lewis, 2005). Proper names are also generally accessed less frequently than nouns; even the most common names, like Smith, are accessed less often than other types of words, which makes the connections among conceptual, lexical, and phonological information within the proper name hierarchy comparatively weak and more prone to retrieval failures (e.g., Cohen & Burke, 1993;Conley, Burgess, & Hage, 1999).…”