Remediation of attention deficits in head injury.
Head injury is associated with psychological sequelae which impair the patient's psychosocial functioning. Information processing, attention and memory deficits are seen in head injuries of all severity. We attempted to improve deficits of focused, sustained and divided attention. The principle of overlapping sources of attention resource pools was utilised in devising the remediation programme. Tasks used simple inexpensive materials. Four head injured young adult males with post concussion syndrome underwent the retraining program for one month. The patients had deficits of focused, sustained and divided attention parallel processing, serial processing, visual scanning, verbal learning and memory and working memory. After the retraining programme the deficits of attention improved in the four patients. Serial processing improved in two patients. Parallel processing and neuropsychological deficits did not improve in any patient. The symptom intensity reduced markedly and behavioural functioning improved in three of the four patients. The results supported an association between improving attention and reduction of symptom intensity. Attention remediation shows promise as a cost effective, time efficient and simple technique to improve the psychological and psychosocial functioning of the head injured patient.
Head injuries are a common cause of disability in the productive age group of persons upto the the age of 40 years. There are physical, psychological and social sequelae of head injury. Psychological sequelae of cognitive deficits and emotional problems contribute significantly to the disability in the patient. Attention deficits are a common sequelae in injuries of all severity., The deficits are in the areas of selective attention or an inability to withstand distraction, visual scanning or complex attention, and inability to divide attention between two tasks.
Attention being a basic cognitive function, improving attention is associated with improvement of the speed of information processing which is aetiological in post concussion syndrome. Skills during rehabilitation are facilitated by improving the patient's attention. Programme to improve attention have used tasks involving symbol detection, auditory target detection, reaction time, visual search, time estimation and rhythm detection. Severely head injured patients have been trained from 3 weeks to 3 months in daily sessions. Improvements have been observed on psychometric and behavioural measures.,,
We attempted to improve deficits of focused, sustained and divided attention in four head injury patients. The multiple resource model of attention was the framework for the assessment of attention deficits as well as for the remediation of the same. Attention is viewed as a resource which is drawn from separate pools. The two cerebral hemispheres are postulated to be reservoirs of attention resources which are independent of each other. Stimulus modality, mode of processing and the nature of response would define the attention resource pool for each task. Simultaneous performance on two tasks or the performance of a single task in the presence of a distractor would become difficult to the extent that the tasks overlap in terms of the attention resource pool. We used this principle to device remediation tasks to improve focused and divided attention. The difficulty level of tasks increased with increase in the overlap of the attention resource pools.
Sample : Four patients with closed head injury who complained of somatic/cognitive/emotional symptoms following head injury underwent remediation [Table I]. The patients had deficits on the attention tests described below.
Assessment : Focused, sustained and divided attention were tested. Neuropsychological functioning was assessed by the NIMHANS neuropsychological battery and information processing by the NIMHANS information processing battery. Symptom intensity was rated on a 9 point scale with 0 representing absence of the symptom and 9 representing maximum severity of symtom. Behaviour was assessed by the neurobehavioural rating scale. Assessments occurred prior to and after the remediation programme. Three of the four patients were also assessed one month after termination of the remediation.
Attention tests : The digit symbol substitution test of the Wecshler adult intelligence scale was given with pre recorded crowd noises as the distracter in the focused attention test. In the task, numbers 1 to 9 were paired with simple line figures. The patient filled 90 numbers with the corresponding line figures for a duration of two minutes. Score was the number of correct substitutions. Sustained attention was measured by a visual target detection task. Similar to the task of Weintraub and Mesulam, 15 arrays of 8 figures each were spread over a page. Each array contained eight figures randomly arranged. One of the figures was a target which the subject had to strike out. There were 3 pages with a total of 135 target figures. A time of five minutes was given, with the score being the number of correctly identified targets. Divided attention was measured by combining the triads task with the rhythm detection task. Word triads were printed on 120 cards. In each triad two words belonged to a category while the third did not. The cards were manually presented for 2-3 seconds and the subject had to call out the odd words. The rhythm detection task consisted of pre recorded rhythms presented through ear phones. Rhythm changes occured every 7-10 seconds. The task lasted 3 minutes with 20 rhythm changes. Patients divided their attention equally between between the two tasks. Efficiency of the division was measured by the efficiency index (EI). The score on each task was computed into a percentage (X and Y) of the maximum score on the task and used in the formula EI=X2
The NIMHANS information processing battery measured simple and choice reaction time, serial processing using the recognition threshold task with and without distraction and parallel processing using the span of apprehension task. Visually presented digits were the stimuli. Deficits were recorded by comparing with norms developed on thirty normal persons. The stimulus presentation and recording of responses were computer controlled. In the serial processing tasks, responses for the stimulus duration of 20 milliseconds was scored for this study. Both the above batteries have been described elsewhere. Symptom rating and neurobehavioural rating were recorded through clinical interview of the patient and a close relative.
Remediation : The goal of remediation was to improve strategies for allocation of attention resources. Difficulty of resource allocation increased in proportion to the degree of overlap in functional cerebral space of the sources of attention. It was assumed that by increasing the task difficulty in this manner, the patient would develop resource allocation skills which would reduce the attention deficits. Remediation lasted for one month with daily 45 minute sessions. Focused, sustained and divided attention tasks were given daily. Each level of resource allocation was a phase. The task difficulty increased over the phases. In each aspect of attention, the phases changed when the performance was at 70% accuracy compared to the maximum possible score on that remediation task and when this level of performance was maintained for 3 days. These criteria were used to ensure that at each phase the efficiency of attention allocation had reached a high level and had stabilised. The shift in phases for each aspect of attention was independent of performance on other aspects of attention. At each phase shift, parallel forms of the criterion attention tests were administered.
Remediation of Focused Attention : A symbol-symbol substitution task was used with different distracters. The combination of symbol pairs were changed on each day to avoid practice. Three minutes each were allotted to the task with and without the distracter. The number of correct pairing formed the score. Distracters changed across the phases. The degree of overlap of the distracter with the task in terms of the source of resource allocation was an index of complexity. The source of the attention of the distracter was initially from a different hemisphere, followed by the same hemisphere as the index task. In the subsequent phases the distracter drew attentional resource from the same modality and then from the same process [Table II].
Sustained Attention : Sorting tasks were used for training sustained attention. Task complexity increased in terms of the time allotted for sorting, the ratio of target stimuli to be sorted and the similarity in the stimuli that were sorted. The number of target stimuli correctly sorted formed the score. There were four phases based on these parameters [Table III].
Divided Attention : Two tasks were administered simultaneously. There were two phases. In `phase I' the two tasks drew attention resources from different hemispheres. Category fluency (left hemisphere) was combined with a task requiring the detection of form and colour. The detection of form and colour consisted of 150 cards each with one of 6 basic forms (circle, hexagon, rectangle, square, triangle, diamond) in one of different colours (pink, green, blue, red, orange, brown, black, yellow). In each session a form or colour was identified as the target and there were 25 targets which the patient identified. The cards were presented at the rate of one every 10 seconds. In `phase II' figural fluency (right hemisphere) was combined with comparison forms task (right hemisphere). Common objects were placed on a tray and the patient paired them by touching with left hand and without looking at the objects. Number of objects correctly paired was the score. In the fluency tasks, the number of items generated were the scores. In each phase the tasks were performed singly and in the dual task condition. The patients did not receive any other form of intervention during the period of the study. The first three patients came for a follow up assessment one month after the post remediation assessment.
The sucess of the remediation program was evaluated on three levels. Level  assessed improvement on the criterion attention tests which indicated an improvement in the basic parameters of focused, sustained and divided attention. Level  assessed generalisation to tasks which measured other cognitive functions such as information processing, memory, visuo spatial abilities. Level  assessed generalisation of the improvement to every day functioning indexed by reduction of symptoms and improvement in behaviour.
[Figure 1], [Figure 2] and [Figure 3] indicate that each of the four patients improved in focused, sustained and divided attention as measured by the criterion attention tasks. The patients progressed across the phases in each of the parameters. Cases 1, 2 and 3 completed all the four phases of the remediation program. These patients have been able to perform at increasing overlap between the index task and the distracting stimulus. The performance on the test of focused attention test has also increased over the phases in cases 1, 3 and 4. The rate of improvement is not similar among the four patients. In cases 1 and 3 deficits were severe while case 4 had moderate deficits at the pre assessment. At the end of remediation the scores exceed the normal mean in case 3 and 4. Case 4 had improved, as there was no deficit at the end of retraining.
Case 2 had no deficits to start with and quickly reached scores which were higher than the normal mean. Possibly because of the absence of deficits, this patient reached the four phases within 15 sessions, while the other three patients who had deficits took ,,,, sessions [Figure 1]. Sustained attention was improved in three phases for all the four cases. Patients could sustain attention for  minutes [Table III]. Cases 1, 3 and 4 had severe to moderate deficits in the pre training assessment. Sustained attention improved to the normal level in these three patients by the end of remediation. In fact cases 3 and 4 had scores which were higher than the nomal mean. Case 2 had no deficits and obtained scores which were higher than the normal mean throughout [Figure 2]. Divided attention improved in all the four patients as they reached the second phase of the two tasks tapping the same hemisphere. Cases 1, 2 and 4 had moderate deficits which improved to normal levels of performance by the end of the phase II. Case 3 did not have deficits initially and scored higher than the normal mean at the end of the remediation [Figure 3]. Cases 1, 3, 4 had moderate to severe deficits in focused, sustained and divided attention at the pre assessment. At the end of retraining all three aspects of attention were normal. The rate of improvement differed among the patients and between the task [Figure 1] [Figure 2] [Figure 3]. Case 2 had deficits only in divided attention which became normal at the end of retraining. The remediation had improved the parameters of attention in all the four cases.
Level 2 generalisation was assessed in terms of improvement in cognitive functions apart from attention. None of the four patients had deficits in simple and choice reaction time at the pre assessment which was maintained even in the subsequent two assessments [Table IV]. Parallel processing was deficient in all the four cases before the remediation. It continued to be deficient even after remediation indicating that improving attention did not improve parallel processing. Serial processing with and without distraction were deficient in cases 1, 2 and 3 in the pre assessment. It had become normal in cases 1 and 2 in the post assessment. Case 3 did not improve. The neuropsychological deficits were in the areas of visual scanning, working memory and verbal learning and memory which did not improve with retraining. Improving attention had improved serial processing in two patients. Parallel processing, working memory, verbal learning and memory and visual scanning had not improved.
Level 3 generalisation referred to improvement in every day functioning as seen in changes in ratings on the neurobehavioral scale and symptom ratings. Mean symptom intensity as rated by the patient and as rated by the significant other have decreased markedly in cases 1, 2 and 3. The neurobehavioural rating scale also indicates a reduction in behaviour problems in these cases. Case 4 did not report any reduction of symptom intensity. A slight reduction of behaviour problems was also noted [Table V]. Case 4 also had deficits in focused, sustained and divided attention at the pre assessment which improved to normal levels at the end of retraining. He did not have any deficits on the neuropsychological battery. The only deficit present was in parallel processing on the information processing battery. Improving attention had reduced the average symptom intensity and behaviour problems in three of the four cases.
The quantum of improvement in attention tests, behaviour functioning and reduction in symptom intensity were calculated by computing the percentage of change with reference to the pre assessment score. Divided attention and behavioural functioning had improved to the highest degree, i.e. 158% and 148% respectively, followed by improvement of 109% in focused attention. Sustained attention has improved to a small degree (27%) perhaps due to a ceiling effect in the pre assessment itself. Symptom intensity as rated by the patient has reduced by more than 67% while that rated by the significant other has reduced by 80%. A significant improvement occurred in symptom intensity.
Remediation trained the capacity to draw attention resources from overlapping resource pools. Each of the four patients could do increasingly complex tasks as the remediation progressed. In each aspect of attention, i.e. focused, sustained and divided attention, the patients could progress to increasingly complex phases. This capacity to allocate attention resources was reflected as improvement in the attention tests in each of the patients. The program has shown that it is possible to remedy attention deficits in head injured patients in a span of 25 sessions. The clinical utility of this improvement is indicated by the significant reduction in symptom intensity. Maintenance of the improvement at one month follow up further augments the clinical utility of the program. Patients suffered from cognitive, somatic and emotional symptoms. The improvement in patients' behavioural, cognitive and emotional functioning indicates that there is an association between attention deficits and post concussion syndrome. We had observed focused attention deficits and divided attention deficits in head injured patients with post concussion syndrome.6,7
The reduction of attention deficits did not significantly improve deficits in parallel processing, verbal learning and memory, working memory and visual scanning, which indicates that improvement on the attention tests were not due to practice alone. If practice alone were to be the cause of improvement in the post assessment, improvement would have occurred in the other cognitive tests as well. In the absence of a control group it is difficult to tease out the effect of non specific factors in improving attention. A further study is required with a larger sample to weed out the effects of practice, daily therapist interaction and hospital visits. The present study has found that attention remediation is a promising method to improve the behavioural functioning of head injured patients. Remediation of attention deficits has shown promise as a treatment for post concussion syndrome.