19 February 2018

Sensitivity and specificity of the Intra-Extra Dimensional Set-Shift task

Using the Intra-Extra Dimensional Set Shift task to distinguish within-and between-diagnostic categories.

The Intra-Extra Dimensional Set Shift (IED) task assesses the processes involved in categorising stimuli into sets (visual discrimination of shapes vs. lines), and responding flexibly (shifting attention) to changes in stimuli. 

The IED task requires participants to learn the rule and select the correct icon (a specific shape or line). The task builds in complexity as distractors are added and the rule changes. The rule changes are both intra-dimensional (e.g. shapes are still the relevant set, but a different shape is now correct) and extra-dimensional (e.g. shapes are no longer the relevant set, instead one of the line stimuli is now correct).

Watch the video below for a clip of the IED task in action.

The task allows the maintenance and flexibility of attention to be assessed. These insights into attentional processes have been applied to the identification of selective impairments both within- and between- diagnostic categories.

Performance on IED distinguishes the freezing gait subtype in Parkinson’s disease  

Freezing of gait (FOG) refers to when a patient initiates walking to find that (temporarily) they cannot take a step (Nutt et al., 2011). FOG is potentially the most severe symptom of Parkinson’s disease (PD) (Beck, Ehgoetz Martens, & Almeida, 2015), and affects around half of patients at the advanced stages of the disease (Giladi et al., 2000). Adults with FOG are at very high risk of falls, which can come at a cost to their independence (Nutt et al., 2011). FOG may be triggered when a patient has to handle cognitive demands whilst walking (Stefanova et al., 2014), and thus embodies the way in which Parkinson’s impacts both psychological and physical functioning.

To explore which executive functions are implicated in FOG, Stefanova and colleagues (2014) compared performance on the IED and SST CANTAB tasks for 66 patients with PD, of which 30 did (FOG+), and 36 did not (FOG-) have FOG symptoms. The study found that performance on the IED task was specifically sensitive to the presence of FOG. Of the FOG+ patients, 32% showed such poor performance that they did not progress from the rule learning phase of the IED task to the intra (then extra) dimensional set-shift (Stefanova et al., 2014). In contrast, only 4% of FOG- patients failed to progress and 0% of 22 healthy controls (Stefanova et al., 2014). This finding would suggest the FOG+ PD patients have specific deficits in rule learning, and ignoring distractors (Stefanova et al., 2014).

Children with autism, schizotypal disorder and co-morbid autism-schizotypal disorder all perform differently on the IED task

Autism spectrum disorder is typically characterised by difficulties in communicating and socialising, as well as restricted interests and restricted or repetitive behaviours (American Psychiatric Association, 2013). Schizotypal disorder also presents with discomfort in social interactions, but with the addition of perceptual distortions and abnormal beliefs (American Psychiatric Association, 2013). Although the disorders are diagnostically distinct (American Psychiatric Association, 2013), their phenotypic similarities mean that they can co-occur at high-rates. In children, the impact of this co-morbidity (autism spectrum and schizotypal disorder) on executive function is little understood.

To understand the different cognitive profiles associated with autism, schizotypal disorder and the co-morbidity, performance on the IED task was compared between the groups (Abu-Akel et al., 2017).  The study tested 67 children (aged 6-12 years), of which 32 were typically developing, 15 had autism, 8 had schizotypal disorder, and 12 had the co-morbidity. The study found that, compared to typically developing children, the children with autism specifically struggled with extra-dimensional set-shifting (Abu-Akel et al., 2017). In contrast, the children with schizotypal disorder showed deficits in intra-dimensional set-shifting, compared to the children with autism and the typically developing controls.

The authors had hypothesised that the co-morbidity of autism and schizotypal disorder would manifest in exacerbated deficits in set-shifting, compared to the children who present with one or other disorder (Abu-Akel et al., 2017). However, the authors found that the overall performance on the IED task was comparable between the typically developing and co-morbid groups (Abu-Akel et al., 2017). This composite score is potentially misleading as although the children with the co-morbidity showed a comparable number of errors on the extra-dimensional phase of the task to the typically developing children, they actually made more errors at the intra-dimensional phase than both the typically developing children and the children with autism. Nevertheless, this clear distinction in performance between the groups suggests that attentive mechanisms are differentially impaired (and spared) as part of the different disorders, and their convergence.

Summary: IED distinguishes within- and between-diagnostic categories

In summary, IED appears to be sensitive to selective impairments in rule learning and reversal. Sensitivity, and specificity, to attentional impairments both within- (Parkinson’s patients with and without FOG) and between-diagnostic categories (autism, schizotypal disorder, and co-morbid autism-schizotypal disorder), means that IED may be used in future research to unpick the neurological underpinnings associated with cognitive flexibility.

You may also be interested in:

Scroll to Top