People often misremember complex scenes by blending what they actually perceived with what they inferred, leading to the extrapolation of complex scenes in memory. Such memory distortions are commonly linked to reality-monitoring errors (RMEs) and have been well-documented in both the visual and haptic modalities. In the auditory domain, however, research using a similar methodology has instead observed a contraction in memory for the temporal content of sounds (Hutchison et al., Journal of Experimental Psychology: Learning, Memory, and Cognition, 38, 1469–1489, 2012). The present study investigated whether scene extrapolation in memory also occurs in the auditory modality by testing participants’ recall of sound sources simultaneously present within a scene. Using an auditory recognition task derived from that used by Hutchison et al. Journal of Experimental Psychology: Learning, Memory, and Cognition, 38, 1469–1489, (2012), participants memorized target scenes composed of three or four sound sources, varying in semantic consistency (i.e., sources from the same category or not). Memory was tested using a trial-by-trial design, in which probe scenes were presented at the same or different size (i.e., with more or fewer sound sources). Results showed that participants tended to remember targets as containing fewer sound sources than were presented, suggesting a memory contraction rather than the extrapolation observed in other modalities. This contraction was attenuated when scenes were semantically consistent. The findings are discussed in relation to plausible (a)modal processes that could underlie scene distortions in memory.
