ABUNDANT PRIMACY EFFECTS IN MAGNITUDE ESTIMATION OF LOUDNESS WITH INVARIANT STIMULUS RANGE

  • Alexander N. Sokolov
  • Pietro Guardini
  • Paul Enck
  • Marina Pavlova

Abstract

Since the seminal work by Stevens (1956, 1975) on establishing the psychophysical law, sensory attributes of stimuli such as loudness of tones, have commonly been assessed by magnitude estimation, in which participants directly assign a number to their current sensation (Gescheider, 1997). Magnitude estimation has been argued to largely resist the context effects, although the stimulus range was shown to modulate the loudness estimation of tones presented within the context of a low or a high intensity range (Gescheider & Hughson, 1991). Here, we examine if magnitude estimation of loudness varies as a function of the frequency of occurrence (base rate) and serial order of tones taken from the same intensity range. Fifty six healthy adults assigned to seven separate groups estimated - without a modulus - a set of five tones (60, 65, 70, 75, and 80 dB SPL, sound pressure level) that were equally frequent (base rate, 10-10-10-10-10, soft tones come on the left) and randomized (i) in a standard computer-assisted way or with a bias such that either mainly (ii) soft or (iii) loud tones occurred on the initial trials, i.e., at the series outset. With different-frequent tones, the four randomization conditions were as follows: (iv, v) on overall frequent soft tones (20- 14-8-4-4) and (vi, vii) on overall frequent loud tones (4-4-8-14-20) with either mainly soft or loud tones presented at the series outset. The results indicate abundant primacy effects on magnitude estimation of loudness: Regardless of the overall base rate, higher estimations occurred with mainly soft rather than loud tones presented at the series outset. For the first time, we show the context effects in absolute magnitude estimation of loudness without varying the intensity range. The outcome questions the context invariance of psychophysical scales derived from absolute magnitude estimation. Further research is required to determine if the primacy effects observed are response-bias or sensory dependent.

Section
Full Articles