Frequency Judgment: The Relationship Between Judged and Actual Frequency

Abstract

The exact nature of the relationship between judged and actual frequency of stimuli encountered outside experimental control has been an area of longstanding but limited interest for several decades. A growing consensus among researchers who examined this relationship is that it is nonlinear and best represented by a negatively accelerating function. However, researchers who examined the relationship sometimes did not compare tested nonlinear functions to what had already been reported in the literature or even a linear function. What is unclear is if a single nonlinear function, among those reported in the literature, can be used to represent the relationship between judged and actual frequency across a variety of stimuli encountered outside of experimental control. Two longstanding theories of frequency judgment have been shown to hold under various contexts and study methods. Whether the direct-coding theory or the memory strength theory of frequency judgment is supported when examining the response times across Likert-type response options has yet to be determined. The direct-coding theory seems to be supported by frequency judgment tasks where frequency information is controlled experimentally. The memory strength theory seems to be supported by frequency judgment tasks where frequency information is not experimentally controlled and accumulated over time. The direct-coding theory would be supported if no difference is found in response times between response options. The memory strength theory would be supported if the response times decreased monotonically from almost never to almost always. The current study showed that a negatively accelerating function best represented the relationship between judged and actual word frequency, judged and actual celebrity frequency, and judged and actual supermarket sales frequency. Further, a single nonlinear model (i.e., the Weber-Fechner law) best represented this relationship across all stimulus sets. The current study also showed that direct-coding theory was not supported by the evidence presented, but the evidence in favor of the memory strength theory was mixed.