In the Preface to the first Journal of Head Trauma Rehabilitation issue on impaired self-awareness (ISA), published in 1998, it was noted that the topic was "emerging from decades of relative neglect."1 Less than a decade later, impaired awareness of deficit following traumatic brain injury (TBI) has attracted more research attention but remains mysterious as to causal mechanisms and appropriate treatment strategies. The 6 articles in this issue provide more food for thought as to what factors contribute to ISA after TBI, an essential step before the design and application of empirically based treatments.

The first article, by Sherer and colleagues, seeks to address the basic question of the neural substrate of ISA. This is a difficult question in a population characterized by diffuse neuropathology, but the study capitalizes on a large, prospective sample and quantitative methods applied to computed tomography (CT). Surprisingly, perhaps, the results suggest that it is the number, but not the location, of brain lesions on CT that is important in predicting ISA, as is the duration of posttraumatic amnesia. Both the latter variable and the number of lesions may indicate disruption in diffuse cerebral networks that are also critical for an integrated, temporally updated picture of the "self."

The next article by Sawchyn and colleagues also implicates the overall severity of injury as predictive of ISA. Additionally, these researchers examine the impact of emotional adjustment on one's ability to detect and acknowledge deficit areas. They found that adjustment was more important than severity in predicting ISA, although the overall level of ISA was lower than anticipated in their sample. The Sawchyn et al study used a sample of persons at several years postinjury on average. This highlights that we must consider the adaptations made over time after a TBI and the variability across people in studies of postacute ISA, which may be a less relevant clinical concern over time.

The 2 succeeding articles, by Ownsworth and Fleming and Hart et al, examine the relationship of executive function to ISA. In addition, like Sawchyn et al, Ownsworth and Fleming examine the relationship of emotional function to ISA. Ownsworth and Fleming also compare the contributions of both to prediction of outcome. They conclude by noting a number of important interrelationships in a complex model of psychosocial function and emphasize that both depression and metacognitive skills, of the type needed to prevent ISA, are related to better outcomes. The Hart et al study uses a simpler model to examine the relationship between executive function (measured by a composite score) and self-awareness of "real-world" attention and behavior deficits. While some evidence for a relationship was found, the authors conclude that ISA is likely to be related to multiple cognitive factors and to specific subcomponents of executive function.

Another theme reflected in the Hart et al article and in the following one, by Garmoe and colleagues, is the assessment of self-awareness in persons without neurologic injury. Recent research in social psychology has reminded us that normal self-awareness is far from perfect and may even be inversely related to the level of actual skill.2 The Hart et al study avoids the ceiling effects that may apply to measures of ISA based upon basic activities of daily living by employing a measure designed to detect "normal" everyday cognitive lapses. Even with this measure, control participants agreed closely with their proxy respondents and no relationship was found between executive skills and ISA. The article by Garmoe and colleagues also examines the accuracy of self-report in the absence of neurologic injury. Having previously reported a substantial discrepancy between self-ratings and therapist ratings of functional competence in a sample of rehabilitation inpatients with TBI, they undertook the same comparison in a group of hospitalized persons without brain injury (orthopedic injury patients). Interestingly, therapists rated the orthopedic group slightly but significantly lower than how those patients rated themselves. This finding suggests that the ISA observed in TBI, when it is based on comparing patient and therapist ratings, may include a "rater effect" induced by the critical eye of the therapist, who is vigilant about finding areas to correct.

The last article, by Prigatano and colleagues, examines the association between self-rated distress in family members of persons with TBI and their ratings of the degree of ISA shown by the person with TBI. These investigators found that self-rated distress for family members was associated with the degree of impairment in patient's self-awareness. Similar associations were found in comparison groups of patients with dementia and those with self-reported memory complaints. While these correlational results do not establish a causal link, these findings may indicate that patient ISA is a significant factor in determining the degree of caregiver distress. This result is consistent with earlier findings indicating that personality change after TBI is more distressing to family members than is physical disability.

While the investigations described in this special issue make a significant contribution to our understanding of ISA after TBI, much remains to be learned. We are just beginning to understand how emotional factors, brain areas affected, specific patterns of cognitive deficits, and overall injury severity interact to cause ISA. Because of different measurement strategies used in various studies with only limited comparison of these methods to date, controversy remains regarding the best approach for measurement of ISA. We have also only limited evidence regarding the degree of ISA that is necessary to have an impact on long-term outcome. Finally, rigorous studies of treatment approaches for ISA are almost nonexistent. Perhaps, some of these research needs will be resolved in time for a third special issue on ISA before another decade has passed.

Tessa Hart, PhD

Issue Editor, Moss Rehabilitation Research Institute, Philadelphia, Pa; Department of Rehabilitation Medicine, Jefferson Medical College, Thomas Jefferson University; Philadelphia, Pa

Mark Sherer, PhD

Issue Editor, Methodist Rehabilitation Center, Jackson, Miss, University of Mississippi Medical Center, Jackson, Miss

REFERENCES