Social feedback enhances learning in Williams syndrome

Abstract

Williams syndrome (WS) is a rare genetic condition characterized by high social interest and approach motivation as well as intellectual disability and anxiety. Despite the fact that social stimuli are believed to have an increased intrinsic reward value in WS, it is not known whether this translates to learning and decision making. Genes homozygously deleted in WS are linked to sociability in the general population, making it a potential model condition for understanding the social brain. Probabilistic reinforcement learning was studied with either social or non-social rewards for correct choices. Social feedback improved learning in individuals with Williams syndrome but not in typically developing controls or individuals with other intellectual disabilities. Computational modeling indicated that these effects on social feedback were mediated by a shift towards higher weight given to rewards relative to punishments and increased choice consistency. We conclude that reward learning in WS is characterized by high volatility and a tendency to learn how to avoid punishment rather than how to gain rewards. Social feedback can partly normalize this pattern and promote adaptive reward learning.

Reduced effects of social feedback on learning in Turner syndrome

“…The control group was recruited in the context of a larger study which also includes children with intellectual disability and other rare genetic disorders linked to intellectual disability as well as typically developing males. The control group is therefore partly overlapping with the group reported in Kleberg et al (2023) 44 . Participants who did not explore both options of the task (> 90% choices of one stimulus), (control: n = 2, Turner syndrome: n = 2) were excluded.…”

“…The reinforcement learning task (described in Fig. 1 ) was adapted from Kleberg et al (2023) 44 . Participants completed two versions of the tasks (henceforth the social and non-social conditions) using either a tablet or computer.…”

“…A simulation-based power analysis conducted using the simr package in R 58 based on the observed random effects structure in Kleberg et al (2023) indicated that the study had > 80% power to detect within-group effects of condition corresponding to a 7% difference in the proportion of correct responses, which was considered a meaningful effect 44 .…”

“…Social reinforcement learning is currently studied in a wide range of neurodevelopmental conditions 25 , 31 , 32 , and alterations have been found in conditions such as ASD 39 , 40 , depression 41 – 43 , and genetic conditions associated with altered social interaction 44 . No studies have been conducted in Turner syndrome.…”

“…In computational modeling, mathematical models are fitted to the data in order to extract latent variables representing underlying cognitive processes 47 which, for example, has been instrumental in understanding the neural basis of reinforcement learning 48 , 49 . Computational modelling methods are also increasingly used to characterize phenotypes and examine the potential underlying mechanisms of psychiatric symptoms 31 , 43 , 44 , 50 , 51 . Furthermore, parameters from computational modeling of reinforcement learning data have been hypothesized as a promising biomarker for mental health conditions affecting reward-driven behavior in theoretical frameworks such as the The National Institute of Mental Health's Research Domain Criteria (RDoC) Initiative 52 .…”