Individuals with Williams syndrome show intact behavioral feedback learning but no loss-driven pupil dilation, a dissociation from typically developing controls¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r? · q3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? TD individuals showed larger pupil dilation following losses than wins (b = 0.37), while WS individuals showed no reliable difference between valences (b = 0.03). → Hallman 2026 q3 i? There was moderate evidence for a syndrome-by-feedback interaction for WS (b = 0.34), indicating feedback valence influenced pupil responses differently in WS than TD; this was not evident for FXS or DS. → Hallman 2026 (2)

Evidence¶

Hallman 2026¶

Hallman, A. E. Z., Willfors, C., Frick, M. A., Nordgren, A. & Kleberg, J. L. (2026). Comparing feedback learning and arousal responses in Down, Fragile X, and Williams syndromes. npj Science of Learning. https://doi.org/10.1038/s41539-026-00438-2

q3 · i? · causal · r?

Follow-up pairwise comparisons of pupil dilation (Fig. 2) in the Results section of the Bayesian LMM. The article concludes "feedback valence may haveno reliable effect on pupil dilation" in WS, whereas TD showed loss-driven dilation.

"For the TD group, pupil dilation were larger following negative feedback (losses) compared to positive feedback (wins) (b = 0.37, 95% HPD [0.23, 0.5]). In contrast, the WS group showed no reliable difference in pupil response between negative and positive feedback (b = 0.03, 95% HPD [−0.19, 0.24])."

Hallman 2026 (2)¶

Hallman, A. E. Z., Willfors, C., Frick, M. A., Nordgren, A. & Kleberg, J. L. (2026). Comparing feedback learning and arousal responses in Down, Fragile X, and Williams syndromes. npj Science of Learning. https://doi.org/10.1038/s41539-026-00438-2

q3 · i? · causal · r?

The syndrome-by-feedback interaction in the pupil LMM (Table 4), Results section. The interaction was significant for WS but not for the FXS or DS groups, for which the evidence remained inconclusive.

"For WS, there was moderate evidence for an interaction effect between syndrome and feedback (b = 0.34, 95% CI [0.09, 0.59], BF = 6.66, pd =1 ) , indicating that feedback valence influenced pupil responses differently in WS compared to TD. This effect was notevident for the FXS or DS groups."

Discussion¶

Related Claims¶

• All groups looked longer at the previously rewarded option than at the loss-associated option, with no group differences in gaze allocation — related
• On loss trials, smaller pupil dilation predicted response change in TD participants while larger dilation predicted it in FXS participants, with no robust effect in DS or WS — related
• All groups, including individuals with Down, Fragile X, and Williams syndromes, adapted their choices based on deterministic feedback, staying with rewarded options and switching after losses — a broader claim this one bears on