Auditory emotion recognition impairments in schizophrenia: relationship to acoustic features and cognition

Abstract

Objective: Schizophrenia is associated with deficits in the ability to perceive emotion based on tone of voice. The basis for this deficit remains unclear, however, and relevant assessment batteries remain limited. The authors evaluated performance in schizophrenia on a novel voice emotion recognition battery with well-characterized physical features, relative to impairments in more general emotional and cognitive functioning.

Method: The authors studied a primary sample of 92 patients and 73 comparison subjects. Stimuli were characterized according to both intended emotion and acoustic features (e.g., pitch, intensity) that contributed to the emotional percept. Parallel measures of visual emotion recognition, pitch perception, general cognition, and overall outcome were obtained. More limited measures were obtained in an independent replication sample of 36 patients, 31 age-matched comparison subjects, and 188 general comparison subjects.

Results: Patients showed statistically significant large-effect-size deficits in voice emotion recognition (d=1.1) and were preferentially impaired in recognition of emotion based on pitch features but not intensity features. Emotion recognition deficits were significantly correlated with pitch perception impairments both across (r=0.56) and within (r=0.47) groups. Path analysis showed both sensory-specific and general cognitive contributions to auditory emotion recognition deficits in schizophrenia. Similar patterns of results were observed in the replication sample.

Conclusions: The results demonstrate that patients with schizophrenia show a significant deficit in the ability to recognize emotion based on tone of voice and that this deficit is related to impairment in detecting the underlying acoustic features, such as change in pitch, required for auditory emotion recognition. This study provides tools for, and highlights the need for, greater attention to physical features of stimuli used in studying social cognition in neuropsychiatric disorders.

Figure 1. Map of stimuli included in the full battery on mean base pitch (F0M) and pitch variability (F0SD) (A) and mean intensity-(VIntM) and intensity variability (VIntSD) (B)

Variability in feature by pitch-based stimuli was determined by one-way ANOVAs across emotions. Among stimuli that were considered to be pitch-based, there was significantly variability in mean base pitch (F0M, p<.0001) and pitch variability (F0SD, p<.0001) but not mean voice intensity (VIntM, p=.13) or intensity variability (VIntSD, p=.4) (shown). Other variables (not shown) that showed significant variability across emotions were the floor frequency of the base pitch (F0floor, HF500 (p=.01), mean pitch of p=.04), maximum frequency of the base pitch (F0max, p<.0001), pitch contour (F0contour, p=.02) and mean pitch of the first formant (F1M, p=.006). A discriminant function analysis with pairwise comparison demonstrated significant contribution of several pitch variables, including F0SD, F0Max, maximum frequency of the first formant (F1Max), and F0Contour differentiation of emotional stimuli. Neither VIntM or VIntSD contributed significantly to this discriminant function. When intensity-based stimuli as a group were compared to pitch-based stimuli, VIntM (p=.001), VIntSD (p=.004) (shown), and HF500 (p<.0001), and mean bandwidth of the first formant (F1BW) (p=.011) (not shown) were significantly different across stimuli. In contrast, pitch-based measures including F0M (p=.008) amd F0SD (p=.27) were not different. A discriminant function showed significant contribution only of VIntM to differentiation of intensity- vs. pitch-based stimuli, with no further contribution of other intensity- or pitch-based variables.

Figure 2

Relative between-group performance (mean ± sem) on the full version of the auditory emotion recognition (AER) task. **p<.01 vs. control. *** p<.001 vs. control

Figure 3

Relative between-group performance (mean ± sem) to pitch- vs. intensity-based stimuli extracted from the full auditory emotion recognition battery (left) and from a brief replication battery (right), showing deficits to pitch- vs. intensity-based emotion recognition ***p<.001 vs. control

Figure 4

Relative between-group performance (mean ± sem) in tone matching test (TMT), face emotion recognition (ER40) and WAIS-III Processing Speed Index (PSI) *** p<.001

Figure 5

A. Correlation between tone matching and auditory emotion recognition (AER) performance across patients (pink) and controls (blue). Correlation was significant both across groups (r=.56, n=98, p<.0001) and within patients (r=.42, n=66, p<.0001) and controls (r=.49, n=32, p=.004) alone. Furthermore, correlations in both patients (p=.03) and controls (p=.002) remained significant even following co-variation for general cognitive dysfunction (PSI). B: Path analysis demonstrating both sensory-specific (TMT) and general cognitive (PSI) contributions to impaired auditory emotion recognition in schizophrenia. Numbers represent standardized regression weights between indicated variables. Model fit parameters including residual Chi-square over degrees of freedom (CMIN/DF)=.91, Root mean square error of approximation (RMSEA)=0 and Hoelter statistic (.05) =560, suggest strong statistical model. Additional paths did not lead to further statistical improvement of the model fit.