Frontal lobe deficits after head injury: Unity and diversity of function

Deficits of ''executive function'' are commonly described following either focal frontal lobe lesions or the diffuse pathology of closed head injury. Here, correlational methods were used to investigate common elements in different tests conventionally used to measure executive deficits. Conventional executive (CONVEX) tests, sometimes adapted to give a variety of performance scores, were administered to a sample of 90 head-injured patients, along with a range of other clinical and neuropsychological measures. Following on from previous reports showing deficits in general intelligence or Spearman's g after frontal lobe lesions, tests relating to g were administered in a more detailed follow-up of 24 patients. The data showed uniformly low correlations between CONVEX tests, no higher indeed than correlations with others tests in the battery. Extracting more detailed scores in search of executive subfunctions such as switching or impulse control simply produced yet lower correlations. Though correlations between CONVEX tests were weak, the common element that these tests did share was closely related to g and to a phenomenon we call goal neglect, or disregard of a known task requirement. In this head-injured sample, goal neglect and g were related more to generalised atrophy than to focal lesions. We suggest that g reflects a process of forming an effective task plan by activation of multiple goals or action constraints. Most tests, including CONVEX tests, are to some extent sensitive to this process. Beyond this very general component, individual CONVEX tests may measure little of broad significance.