Cognitive estimation: Performance of patients with focal frontal and posterior lesions

Abstract

The Cognitive Estimation Test (CET) is a widely used test to investigate estimation abilities requiring complex processes such as reasoning, the development and application of appropriate strategies, response plausibility checking as well as general knowledge and numeracy (e.g., Shallice and Evans, 1978; MacPherson et al., 2014). Thus far, it remains unknown whether the CET is both sensitive and specific to frontal lobe dysfunction. Neuroimaging techniques may not represent a useful methodology for answering this question since the complex processes involved are likely to be associated with a large network of brain regions, some of which are not functionally necessary to successfully carry out the CET. Instead, neuropsychological studies may represent a more promising investigation tool for identifying the brain areas necessary for CET performance. We recently developed two new versions of the CET (CET-A and CET-B; MacPherson et al., 2014). We investigated the overall performance and conducted an error analysis on CET-A in patients with focal, unilateral, frontal (n = 38) or posterior (n = 22) lesions and healthy controls (n = 39). We found that frontal patients' performance was impaired compared to healthy controls on CET. We also found that frontal patients generated significantly poorer estimates than posterior patients on CET-A. This could not be explained by impairments in fluid intelligence. The error analyses suggested that for CET-A, extreme and very extreme responses are impaired following frontal lobe damage. However, only very extreme responses are significantly more impaired following frontal lobe than posterior damage and so represent a measure restricted to frontal "executive" impairment, in addition to overall CET performance.

Keywords: Cognitive estimation test; Executive functions; Fluid intelligence; Prefrontal cortex.

Fig. 1

Boxplot displaying performance on CET-A Adjusted scores of Frontal, Posterior and HC.

Fig. 2

Axial slice renders of the lesion distributions. The underlay is the mean of all MR volumes and the overlay is the sum of the thresholded lesion masks, all transformed into MNI space, resliced to 1.5 mm³ isotropic resolution, and displayed in neurological convention. The vertical dimension is given above each slice. Visualized with Micron (http://people.cas.sc.edu/rorden/mricron/index.html).

Fig. 3

a. Scatter Plot displaying correlation between Measure I –Quite Extreme responses and fluid intelligence (Ravens) in Frontal patients. b. Scatter Plot displaying correlation between Measure II –extreme and very extreme responses and fluid intelligence (Ravens) in Frontal patients. c. Scatter Plot displaying correlation between Measure III –very extreme responses and fluid intelligence (Ravens) in Frontal patients.