Listener judgments of perceived age

When listening to a talker, listeners not only hear their message but also make judgments about the indexical properties of the talker, such as their gender or mood. The present experiment involved listening to pairs of speech stimuli produced by a single talker and judging in which sample the talker was older. The speech stimuli were a subset of samples used by Hunter et al. (2015), who used recordings of a single talker taken over 48 years. The paired comparisons method was the same as that used by Ozdogan (2023), who had recordings of two talkers taken over 20 years. Thirteen young adult listeners with self-reported normal hearing completed the paired comparisons task. After completing this task, the listeners were asked to identify which acoustic features of the talker's voice caused them to perceive the talker as older versus younger. Contrary to Ozdogan (2023), where listeners performed at chance levels, the present listeners were able to correctly identify the older sample about 70% of the time. This improved performance may be due to the larger time period covered in the current samples (48 vs 20 years) and/or to the previous talkers being professional speakers while the current talker was not.