Sequential learning models for the Wisconsin card sort task: Assessing processes in substance dependent individuals

Abstract

The Wisconsin Card Sort Task (WCST) is a commonly used neuropsychological test of executive or frontal lobe functioning. Traditional behavioral measures from the task (e.g., perseverative errors) distinguish healthy controls from clinical populations, but such measures can be difficult to interpret. In an attempt to supplement traditional measures, we developed and tested a family of sequential learning models that allowed for estimation of processes at the individual subject level in the WCST. Testing the model with substance dependent individuals and healthy controls, the model parameters significantly predicted group membership even when controlling for traditional behavioral measures from the task. Substance dependence was associated with a) slower attention shifting following punished trials and b) reduced decision consistency. Results suggest that model parameters may offer both incremental content validity and incremental predictive validity.

Toward a computational cognitive neuropsychology of Wisconsin card sorts: a showcase study in Parkinson’s disease

“…Compared to PD patients off dopaminergic medication, PD patients on dopaminergic medication show a reversed pattern in learning from feedback: Learning from positive outcomes was facilitated, but learning from negative outcomes was impaired. The cited evidence for a dopaminemodulated learning from positive and negative feedback in PD comes from a probabilistic reinforcement learning study that was conducted by Frank et al (2004) and replicated in other studies (Bódi et al 2009;Cools et al 2006;Moustafa et al 2008;Palminteri et al 2009;Voon et al 2010; see also Robbins and Cools 2014).The current study addresses two questions: First, we examine whether the assumption of distinct learning rates for positive BRepeat^feedback and for negative BShift^feedback is supported on a computerized variant of the WCST (cWCST), as originally conjectured by Bishara et al (2010) …”

“…Furthermore, ALM-2 was used for analyses of therapeutic effects in schizophrenia (Cella et al 2013). However, differences in model performance between ALM-1 and ALM-2 were small (Bishara et al 2010). Importantly, the ALM-1 and the ALM-2 result in different hypotheses about WCST performance.…”

“…In a recent study, integration errors have been proposed as an additional error measure that can provide incremental information about participants' WCST performance (Lange et al 2016a). Figure 1 gives a trial-by-trial example of the WCST and the operational definition of perseverative errors, integration errors, and set-loss errors (for further information about the WCST, see Bishara et al 2010;Eling et al 2008;Nyhus and Barceló 2009). While these three types of WCST errors (perseverative error, integration error, set-loss error) are clearly distinguishable on an operational level, little is known about the latent cognitive processes that lie behind their appearance.…”

“…Second, model-based cognitive neuropsychology offers a method to formalize assumptions about cognitive processes, which are assumed to explain patterns of observed behavior in neurological patients (e.g., Botvinick and Plaut 2004;Cooper and Shallice 2006;Farrell and Lewandowsky 2015;Frank and O'Reilly 2006;Frank et al 2004;Mavritsaki et al 2010;Mavritsaki et al 2009;Plaut et al 1996). The modeling approach has repeatedly been applied to the WCST (Amos 2000;Berdia and Metz 1998;Bishara et al 2010;Cella et al 2013;Dehaene and Changeux 1991;Farreny et al 2016;Kaplan et al 2006;Kimberg and Farah 1993;Levine and Prueitt 1989;Monchi et al 2000). Our study is based on the model that was proposed by Bishara et al (2010).…”

“…The modeling approach has repeatedly been applied to the WCST (Amos 2000;Berdia and Metz 1998;Bishara et al 2010;Cella et al 2013;Dehaene and Changeux 1991;Farreny et al 2016;Kaplan et al 2006;Kimberg and Farah 1993;Levine and Prueitt 1989;Monchi et al 2000). Our study is based on the model that was proposed by Bishara et al (2010). (Barceló 2003;Kopp and Lange 2013;Lange et al 2016a;Lange et al 2015;Lange et al 2016b).…”