The investigators recruited 68 healthy young adults (mean age 18.94 years, SD = 1.46; 85% women). They used the DASS-21 anxiety subscale as a self-report index of sustained physiological hyperarousal, and the TAS-20 to measure alexithymia. The authors reported institutional ethical approval from the Universitat de Valencia ethics committee (Reference: 2023-PSILOG-2558999). All other demographic or clinical inclusion criteria were not detailed in the source beyond the age and sex distribution provided.
Participants completed a 64-trial Mixed Gambles Task to elicit risky choices. Individual choice behavior was summarized by three Prospect Theory parameters: loss aversion, value sensitivity, and choice consistency. The study team estimated these parameters using hierarchical Bayesian analysis. For moderation testing, the authors used Hayes' PROCESS macro and applied Johnson–Neyman probing to locate the TAS-20 values at which the conditional effects became significant. The preprint did not report additional task timing, payoff magnitudes, or model priors in the abstract; those methodological specifics were not provided in the source text.
Data indicated that higher self-reported hyperarousal predicted two specific changes in decision metrics. First, hyperarousal was associated with reduced value sensitivity—participants reporting greater physiological arousal showed degraded sensitivity to magnitudes of gains and losses in the Prospect Theory model. Second, hyperarousal correlated with lower choice consistency, meaning these participants made less predictable choices given the modeled subjective utilities. The authors reported no association between hyperarousal and loss aversion. The hierarchical Bayesian parameter estimates and statistical effect sizes were reported in the full manuscript, but the abstract provides only directional relationships and significance for these outcomes.
The study team tested whether alexithymia moderated the associations between self-reported hyperarousal and each Prospect Theory parameter. They reported two distinct conditional patterns. For the hyperarousal–choice consistency link, moderation analysis showed significance only below a TAS-20 threshold of 51.05; that threshold applied to 67.65% of the sample. The investigators described this pattern as monotonic buffering—lower TAS-20 scores (fewer alexithymic traits) were linked to an observable negative association between hyperarousal and choice consistency. For the hyperarousal–value sensitivity link, the conditional effect appeared within a bounded TAS-20 range (45.31–66.08), which the authors interpreted as a zone in which intermediate alexithymia may confer maximal vulnerability. The source explicitly states that the latter model only reached trend significance and should be treated as exploratory.
Across models, neither self-reported hyperarousal nor its interaction with TAS-20 scores predicted loss aversion. The authors reported null findings for this parameter in all moderation analyses summarized in the abstract. The source did not provide alternate loss-aversion metrics or subgroup analyses that might explain the null result; therefore, only the reported absence of association can be stated.
This report is a preprint and has not undergone peer review; the authors explicitly note that the findings should not guide clinical practice. The sample was young and predominantly female, which the abstract reports but does not allow broad generalization; any claims beyond the sampled population would exceed the source data. The authors declared no competing interests and confirmed IRB oversight. The source states that all data produced are available upon reasonable request to the authors. Specifics omitted from the abstract—such as detailed task parameters, full model specifications, effect sizes, confidence intervals, and any sensitivity analyses—were not provided in the source text and therefore are not asserted here.
Study questions remaining in the manuscript include whether observed conditional effects replicate in larger and more diverse samples, whether objective physiological measures of arousal would align with self-report, and how task or modeling choices might influence the pattern of moderation by alexithymia. The authors treated one interaction as exploratory and flagged it accordingly, and no regulatory milestones or clinical recommendations were reported in the source.