@article{WaltmannSchlagenhaufDeserno2022,
  author    = {Waltmann, Maria and Schlagenhauf, Florian and Deserno, Lorenz},
  title     = {Sufficient reliability of the behavioral and computational readouts of a probabilistic reversal learning task},
  series = {Behavior Research Methods},
  volume    = {54},
  journal   = {Behavior Research Methods},
  number    = {6},
  issn      = {1554-3528},
  doi       = {10.3758/s13428-021-01739-7},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-324246},
  pages     = {2993-3014},
  year      = {2022},
  abstract  = {Task-based measures that capture neurocognitive processes can help bridge the gap between brain and behavior. To transfer tasks to clinical application, reliability is a crucial benchmark because it imposes an upper bound to potential correlations with other variables (e.g., symptom or brain data). However, the reliability of many task readouts is low. In this study, we scrutinized the retest reliability of a probabilistic reversal learning task (PRLT) that is frequently used to characterize cognitive flexibility in psychiatric populations. We analyzed data from N = 40 healthy subjects, who completed the PRLT twice. We focused on how individual metrics are derived, i.e., whether data were partially pooled across participants and whether priors were used to inform estimates. We compared the reliability of the resulting indices across sessions, as well as the internal consistency of a selection of indices. We found good to excellent reliability for behavioral indices as derived from mixed-effects models that included data from both sessions. The internal consistency was good to excellent. For indices derived from computational modeling, we found excellent reliability when using hierarchical estimation with empirical priors and including data from both sessions. Our results indicate that the PRLT is well equipped to measure individual differences in cognitive flexibility in reinforcement learning. However, this depends heavily on hierarchical modeling of the longitudinal data (whether sessions are modeled separately or jointly), on estimation methods, and on the combination of parameters included in computational models. We discuss implications for the applicability of PRLT indices in psychiatric research and as diagnostic tools.},
  language  = {en}
}
@phdthesis{Waltmann2024,
  author    = {Waltmann, Maria},
  title     = {Neurocognitive mechanisms of loss of control in Binge Eating Disorder},
  doi       = {10.25972/OPUS-36430},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-364300},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2024},
  abstract  = {Binge Eating Disorder (BED) is a common, early-onset mental health condition characterised by uncontrollable episodes of overeating followed by negative emotions such as guilt and shame. An improved understanding of the neurocognitive mechanisms underlying BED is central to the development of more targeted and effective treatments. This thesis comprises a systematic review and three empirical studies contributing to this endeavour. BED can be thought of as a disorder of cognitive-behavioural control. Indeed, self-report evidence points towards enhanced impulsivity and compulsivity in BED. However, retrospective self-reports do not capture the mechanisms underlying impulsive and compulsive lapses of control in the moment. The systematic review therefore focussed on the experimental literature on impulsivity and compulsivity in BED. The evidence was very mixed, although there was some indication of altered goal-directed control and behavioural flexibility in BED. We highlight poor reliability of experimental paradigms and the failure to properly account for weight status as potential reasons for inconsistencies between studies. Moreover, we propose that impulsivity and/or compulsivity may be selectively enhanced in negative mood states in BED and may therefore not be consistently detected in lab-based studies. In the empirical studies, we explored the role of behavioural flexibility in BED using experimental and neuroimaging methods in concert with computational modelling. In the first empirical study, we assessed the reliability of a common measure of behavioural flexibility, the Probabilistic Reversal Learning Task (PRLT). We demonstrate that the behavioural and computational metrics of the PRLT have sufficient reliability to justify past and future applications if calculated using hierarchical modelling. This substantially improves reliability by reducing error variance. The results support the use of the PRLT in the second and third empirical studies on development and BED. Because a majority of patients develop BED as adolescents or young adults, we speculated that it may emerge as a consequence of disrupted or deficient maturation of behavioural flexibility. Little is known about typical development in this domain. We therefore investigated normative development of reversal learning from adolescence to adulthood in the second empirical study. Typically- developing adolescents exhibited less adaptive and more erratic and explorative behaviour than adults. This behaviour was accounted for by reduced sensitivity to positive feedback in a reinforcement learning model, and partially mediated by reduced activation reflecting uncertainty in the medial prefrontal cortex, a region known to mature substantially during adolescence. In the third empirical study, we investigated reversal learning in BED, paying special attention to potential biases associated with learning from wins vs learning from losses. We speculated that negative urgency could make it more difficult for BED patients to learn and make decisions under pressure to avoid losses. To dissociate between effects of excess weight and BED, we collected data from obese individuals with and without BED as well as normal-weight controls. As hypothesised, there were subtle neurocognitive differences between obese participants with and without BED with regard to learning to obtain rewards and to avoid losses. Obese individuals showed relatively impaired learning to obtain rewards, while BED patients showed relatively impaired learning to avoid losses. This was reflected in differential learning signals in the brain and associated with BED symptom severity. In sum, this thesis shows that the evidence on impulsivity and compulsivity in BED is inconsistent and offers potential explanations for this inconsistency. It highlights the need for reliability in interindividual difference research and indicates ways to improve it. Further, it charts the typical development of reversal learning from adolescence to adulthood and underscores the relevance of exploration in the context of learning and decision-making in adolescence. Finally, it demonstrates qualitative differences between BED and obesity, hinting at a pivotal role of aversive states in loss of control in BED.},
  subject      = {Binge-eating Disorder},
  language  = {en}
}