Abstract
Stress alters vocal production, particularly by affecting laryngeal muscle function. Despite several studies on voice acoustics under stress, no systematic synthesis exists. This study aimed to conduct a systematic review and meta-analysis of experimental studies assessing the impact of stress on vocal fundamental frequency (F0). Following PRISMA guidelines, a systematic search was conducted in PubMed and Scopus databases for studies published between January 2010 and September 2024. Eligible studies included adult participants exposed to experimental or naturalistic stressors, with pre- and post-stress voice recordings analysed using objective acoustic measures. Data were extracted regarding F0, and random-effects meta-analyses were performed. Subgroup analyses were conducted based on gender, speech type, and stress induction method. Ten studies met the inclusion criteria, encompassing 1148 observations. The meta-analysis showed a significant increase in F0 after stress (SMD = 0.5504, 95% CI [0.3014, 0.7995], p < 0.001) with substantial heterogeneity (I(2) = 69.2%). However, evidence of publication bias was observed; trim-and-fill attenuated the pooled estimate to a nonsignificant effect (SMD = 0.1710; 95% CI: -0.1472 to 0.4891; p = 0.2923). Subgroup analyses revealed stronger effects in women (SMD = 0.7128, 95% CI [0.3763, 1.0492], p < 0.001) and during spontaneous speech (SMD = 0.7911, 95% CI [0.4492, 1.1331], p < 0.001), with nonsignificant results for men and standardized speech. Effects tended to be larger for naturalistic stressors than validated laboratory procedures, though the between-group comparison was not statistically significant. F0 shows promise for stress assessment, but publication bias and heterogeneity warrant caution. F0 should be considered a noninvasive candidate biomarker requiring validation in large, prospective studies using standardized protocols and representative, gender-stratified cohorts.