Neuropsychological Findings in Eating Disorders
Christoph J. Lauer
Neuropsychological research in eating disorders initially was driven by the observation that anorexic patients with evidence for perinatal brain injury showed a poor treatment outcome; particularly, a smaller weight gain was found to be associated with more frequent complications in pregnancy and delivery. Such complications were suggested to increase the susceptibility to subtle cerebral damage, which then may lead to a poor clinical prognosis. Hereby, neuropsychological testing was thought to comprise the advantage of favouring identification of cortical dysfunctions caused by these subtle brain damages below anatomically or physiologically detectable levels. The general hypothesis formulated proposed that the pattern of neurocognitive deficits might be suggestive of dysfunction of relative specific brain systems (Braun and Chouinard, 1992). In following these assumptions, most of the early neuropsychological investigations applied formal test batteries such as the 'Luria-Nebraska Neuropsychological Battery' or the 'Wechsler Intelligence Scales' (WAIS). However, while such test batteries are useful in making general conclusions about the presence of impairments in individuals with a given brain injury, they are not flexible enough to allow the investigator to assess the variety of functions that may underlie a performance deficit on a complex cognitive task in patients with, for example, an eating disorder. Poor performance may be the result of a broad range of possibilities, including damage to one of several areas, the accumulative effect of mild deficits in multiple areas, or factors unrelated to specific brain dysfunctions (see: Keefe, 1995). Therefore, neuropsychology should provide more than a window into the everyday mental processes of psychiatric patients; it should provide an objective description of what areas of behaviour and cognition are likely to be a problem for the patient and — more important—what areas are not.
This paper is divided into five sections. First, a short overview of neurocognitive observations in non-eating disordered subjects (e.g. dieters) will be provided; thereafter, the findings obtained during the acute state of an eating disorders will be summarized, followed by the presentation of neurocognitive changes after treatment, the attempt to identify neuropsychological predictors of the treatment outcome and a short section aiming at the neurocognitive-based identification of patients dropping out of therapy. Finally, the significance of various factors mediating the neurocognitive deficits in eating disorder patients will be discussed.
Because little scientific effort has so far been undertaken to evaluate the neurocognitive function in overweight and obese subjects not suffering from an additional disease known to affect neuropsychological task performance (such as the Prader-Willy syndrome, the sleep apnoea syndrome, the Down syndrome), these types of eating disorders will not be considered in this paper.
PRISONERS OF WAR AND IN NON-EATING
DISORDERED, NORMAL WEIGHT DIETERS
In the following the suggestion that severe starvation and weight loss may cause sustained impairments in brain morphology and cognitive functioning, Sutker and co-workers (1987, 1990) evaluated the neuropsychological task performances in former prisoners of war (POW) who had experienced severe weight loss during captivity due to biological and psychological stress. Compared to combat veterans not captured and interned, the high weight-loss POW (loss of more than 35% of preconfinement weight) were deficient on the 'Attention-Concentration' Factor of the WAIS-R and on measures of immediate and delayed memory as assessed by the Wechsler-Memory-Scale (WMS); on the other hand, no significant impairment was obvious on the WAIS-R subtests 'digit symbol' (sustained attention), 'block design' (visuospatial construction) and 'digit span' (span of attention, working memory). The authors concluded that their findings of deficits on measures of immediate recall combined with lowered 'Attention-Concentration' factor performances raises the possibility that memory problems cited frequently in POW self-reports may be more attributable to deficiencies in attention, concentration and perhaps organizing functions than to memory storage or retrieval processes.
In healthy normal weight non-dieting females, various degrees of 'acute' food deprivation (miss one meal, miss two meals and miss all food for 24 hours) did not result in obvious effects on neurocognitive functions such as sustained attention, attentional focus, simple reaction time or immediate memory, except for a significant slower psychomotor speed performance (finger tapping) after 24-hour of food deprivation (Green, Elliman and Rogers, 1995). However, the same group reported that normal weight female dieters displayed poorer vigilance performance, slower reaction times and poorer immediate recall of words when they were dieting compared to a period of normal food intake. Performance on psychomotor speed was not impaired (Green and Rogers, 1995). Because self-reported dietary restraint, but not anxiety and depression, was increased during dieting, the authors related their findings to an association between dieting behaviour and high levels of distractibility; they proposed that impaired cognitive performance is closely related to dieting or the perceived need to diet per se. An increased distractibility (reduced capacity to inhibit responding to task irrelevant information and to facilitate responses that are goaldirected; assessed by the Stroop Colour Naming Task) was also reported in restrained eaters after a 'high-caloric' preload (single Twix bar) compared to a 'low-caloric' preload (cream cracker) and to unrestrained eaters (Odgen and Greville, 1993); the restrained females took longer to colour name active state words, food words and body size words after the high-caloric preload than unrestrained eating females. Interestingly, not only do acute food deprivation and prolonged periods of dieting affect some facets of neurocognitive