ASA 129th Meeting - Washington, DC - 1995 May 30 .. Jun 06
5pPP4. Precedence and plausibility.
William A. Yost
Sandra J. Guzman
Parmly Hear. Inst., Loyola Univ. Chicago, 6525 N. Sheridan Rd., Chicago,
IL 60626
The ``Clifton effect'' [R. K. Clifton, 1834--1835 (1987)] was
studied in a sound-deadened room with seven loudspeakers. One loud speaker
produced a source click while other loudspeakers produced delayed copies
simulating echoes (delays: 2--30 ms). Each combination of source and echoes is
one click event and was presented as a train of click events (1--20 click
events). A train was presented to listeners who made two judgments for the LAST
click event presented: (1) The number of loudspeakers which produced sounds for
the last click event, and (2) the loudspeaker location for each perceived
source. ``Catch trials'' were introduced to make sure listeners used all
possible responses and were able to locate the loudspeaker sources. When more
than 10 click events were presented, a switch in conditions was introduced
between the 10th and 11th click event. If the switch was plausible for a
natural source and its echoes, responses indicated that listeners processed
delayed clicks as echoes. If the change was implausible, then responses after
the switch changed indicating listeners processed all clicks as if they were
sources rather than echoes. [Work supported by NIH and AFOSR.]