The impact of response latency on perceived level of intelligence and likability of autistic speakers

This study examines how acoustic differences in latency (including extended pauses) impact listeners’ perceptions of intelligence and likability. Forty neurotypical adults listened to a series of response recordings from eight autistic speakers. The original recordings for each task response were made using a Blue Yeti USB microphone. The recordings were saved as .wav files at a sampling rate of 44.1 kHz and a quantization of 24 bits. Using Adobe Audition editing software, response recordings were truncated to be between 25–30 s, with full utterances and several turn-taking cycles between the interviewer and the autistic individual. Recordings were artificially modified to reflect response latency times of 350 ms (the average response latency of a neurotypical speaker), 700 ms, 2 s, and 3 s. Listeners were instructed to rate the intelligence and likability of the speaker on a visual analog scale with descriptive points of reference (e.g., “Very Poor” to “Very Good”). Differences in listener ratings as a function of neurological status and pause duration will be discussed. It is hoped that the results of this study will lead to more effective speech therapy interventions for autistic young adults, therefore improving their quality of life.