Ecological validity of a simplified version of the multiple errands shopping test

Abstract

Shallice and Burgess (1991) reported the utility of the Multiple Errands Test (MET) in discriminating executive deficits in three frontal lobe patients with preserved high IQ, who were otherwise unimpaired on tests of executive function. The aim of this study was to ascertain the value of a simplified version of the MET (MET–SV) for use with the range of people more routinely encountered in clinical practice. Main findings were as follows: 1) The test discriminated well between neurological patients and controls, and the group effects remained when the difference in current general cognitive functions (WAIS–R FSIQ) was taken into account. 2) The best predictors of performance in the healthy control group (n = 46) were age and the number of times participants asked for help (with more requests associated with poorer performance). 3) In the neurological group, two clear patterns of failure emerged, with performance either characterized by rule breaking or failure to achieve tasks. These two patterns were associated with different dysexecutive symptoms in everyday life. 4) The patients not only made more errors than controls, but also different ones. A scoring method that took this into account markedly increased test sensitivity. 5) Many patients passed traditional tests of executive frontal lobe function but still failed the MET–SV. This pattern was strongly associated with observed dysexecutive symptoms in everyday life. The results demonstrate the clinical utility of the test, and suggest that there are two common and independent sources of failure on multitasking tests in a general neurological population: memory dysfunction, and initiation problems. (JINS, 2003, 9, 31–44.)