Did the AI-voice newscast EEG study also measure behavioural credibility, and what did it find?

Problem. Whether synthetic voice lowers credibility of narrated news. Conditions. Frontiers in Psychology 2023 study: a pre-experiment session plus a formal EEG/ERP session (30 participants). Method. The pre-experiment was a within-subject design replicating earlier studies on news attributed to machine versus human newscasters. Result. Human voices were perceived as more credible and fluent than AI voices (credibility t(9) = -4.000, p = 0.003; comprehensibility not significant). Correction. The earlier finding should say the paper itself reports a small behavioural credibility result (df = 9, so about 10 raters; not stated separately here), consistent in direction with the EEG result, rather than implying EEG is the only credibility evidence and contrasting it with behavioural ratings in other work. Version/date. Accepted 27 Oct 2023. Curator qualifications: All cited documentation excerpts located on the live pages by curator GET; no video sources. Correction of earlier record(s) C3-B622-11 from a discovery audit; the curator re-located the cited excerpts by timed GET. Limitations: Pre-experiment sample size is inferred from df = 9 (the excerpted text does not state n); AI voice and stimuli details of the pre-experiment were not checked beyond the quoted passages. Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.

{
  "citations": [
    {
      "url": "https://www.ebi.ac.uk/europepmc/webservices/rest/PMC10701901/fullTextXML",
      "note": "Pre-experimental results; Materials and methods: Experiment design"
    }
  ]
}