| Abstract: | A review of the literature indicates that although event frequency is known to have many behavioral consequences, the intervening representation process(es) are not well understood. Storage of frequency information poses a number of theoretical and methodological problems. 4 hypotheses regarding frequency representation are distinguished (trace-strength, multiple-trace, numerical inference, and multiple-process), and evidence from previous experiments using verbal, nonverbal, and nonstimulus events is evaluated. None of the proposed models can be ruled out entirely, although the multiple-trace hypothesis offers the best explanation for data obtained from "verbal" studies. Unfortunately, nearly all the data collected to date involve verbal stimuli and traditional verbal-learning paradigms, raising the question of representativeness. If frequency can be stored other than by trace multiplexing, it would most likely show up in other event domains. Acquisition of frequency information is typically by incidental means, making such factors as set, task characteristics, and event composition (i.e., anything that could control attention during acquisition) doubly important. It is argued that any real understanding of basic frequency mechanisms must await an expanded scope of investigation. (54 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
