Informational masking in infancy.

摘要

Detection thresholds for a 300-ms, 1-kHz tone were measured for infants and untrained adults in one of three masker conditions: (1) a random-frequency multitone masker, (2) a constant-frequency multitone masker, or (3) a noise band. The random-frequency masker was expected to generate high stimulus uncertainty whereas the constant-frequency masker was expected to generate low stimulus uncertainty. The noise masker was included as a primarily energetic masking comparison. Maskers repeated at 300-ms intervals throughout testing at an overall level of 60 dB SPL. The signal was played synchronously with one interval of the repeating masker. An observer-based testing method was used. Following training to 80%-correct criterion, detection thresholds were determined adaptively. In all conditions, infants' thresholds were higher than adults'. For both age groups, thresholds were highest in the presence of the random-frequency multitone masker, ranging from 47 to 72 dB SPL for the adults and from 76 to 84 dB SPL for the infants. Infants' thresholds were also considerably elevated in the presence of the constant-frequency multitone masker. Although adults also produced elevated thresholds in this condition, the large age difference suggests that infants are particularly susceptible to the effects of informational masking, whether the masker is uncertain or not.; In order to test infants, procedural modifications were made to existing adult procedures. These modifications included reduced training, an undefined listening interval, a masker that repeated throughout testing, and a yes/no, single-interval paradigm. In the second experiment, the effects of these modifications on adults' performance were examined. Detection improved with training. In addition, subjects received benefit from a defined listening interval. The effects of masker type (repeating or gated) and procedure (yes/no or 2IFC) were less clear. However, most subjects appeared to perform better in the presence of the repeating versus the gated masker when a yes/no procedure was used. Given the procedural effects observed here, it is likely that adults participating in the first experiment were influenced by the modified procedure. However, there is no evidence to suggest that the infants would be influenced by the procedural changes differently than the adults.