Behavioral Study On Perception of Emotional Speech in Individuals with Autism Spectrum Disorders

Background:  Smooth communication requires not only understanding of linguistic meaning, but also interpretation of emotions based on nonverbal information such as facial expressions and prosody. Individuals with autism spectrum disorders (ASD) have difficulty holding smooth conversations due to their inability to perceive others’ emotions during communication. Previous studies on emotion perception in ASD have primarily been conducted from a visual perspective such as perception of expressions, and few studies have been conducted from an auditory perspective with a focus on the ability to perceive emotions conveyed in speech.

Objectives: To elucidate the characteristics of speech perception in ASD from a behavioral perspective.

Methods: Subjects were 12 individuals with ASD not receiving drug therapy and 12 typical development aged between 10 and 15 years. Diagnosis was made in person using the DSM-IV. The sex ratio, age, and IQ did not significantly differ between the two groups. The experiment was conducted using Presentation (Neurobehavioral Systems, Inc., USA). Speech stimuli consisting of words and SVO (Subject + Verb + Object) sentences conveying three types of emotions (happy, anger, normal voice) were provided to subjects, who were given the tasks of semantic perception, which involved identifying the semantic content of the speech stimulus, and emotion perception, which involved identifying the emotion associated with the speech. On each task, subjects were asked to respond by pressing one of three buttons, and the correct response rate and response time were compared between the groups.

Results: On the emotion perception task, no intergroup differences were seen in the correct response rates for both the word and SVO sentence stimuli, but the response time was significantly longer in the ASD group (mean, approximately 200 ms; P<0.05). On the semantic perception task, the ASD group had a significantly lower correct response rate (0.95±0.04 vs. 0.99±0.02; P<0.005) as well as a longer response time when an SVO sentence stimulus for anger was presented.

Conclusions: I-M Eigsti (2011) suggested that sites for perception of emotional speech may differ between ASD and TD. In the present study, individuals with ASD took longer than TD to process—in other words, to perceive and output—the emotions expressed by the speaker. This finding suggests that individuals with ASD and TD have different perception processes. The results also indicate that individuals with ASD may have difficulty correctly perceiving the speaker’s intentions when they are expressed with an unpleasant emotion. In order to enable individuals with ASD to improve their perception ability, it may be important for those providing support to individuals with ASD to pay careful attention to the emotions conveyed in verbal speech during conversations.