Introduction: Many patients with brain tumours experience dysfunction in several cognitive domains. Given the limited survival times of the majority of patients with brain tumours, maintenance or improvement of quality of life is as important as increasing survival time. Impaired cognition has a negative impact on quality of life and as such, cognitive function is becoming an increasingly important endpoint in clinical trials in neuro-oncology. However, measuring cognition in patients with brain tumours is problematic for a number of reasons. Most intracranial tumours are initially treated with surgery and studies of neurosurgical morbidity often evaluate physical as opposed to cognitive domains, yet the latter can have a greater negative impact on the patient’s quality of life. This thesis therefore details cognition in brain tumour patients at the time of presentation (pre-operatively) and examines the effects of surgical intervention on cognitive function. Of particular interest is the potential utility of inspection time, a computer-based measure of the brain’s information processing efficiency, as a measure of brain slowing as a result of the tumour and as an indicator of response to surgical intervention. Methods: The study is based on a cohort of 118 newly-presenting patients with a supratentorial brain tumour who were to have surgery (biopsy or resection). Each patient was administered a comprehensive battery of cognitive tests prior to surgery (baseline). The battery comprised inspection time testing, other standardised cognitive measures and assessment of mood, quality of life and functional status. Post-operatively, each patient repeated the inspection time test in addition to a selected number of the other tests administered at baseline. For comparison, a group of patients admitted for elective spinal surgery (n = 85) were also tested pre- and post-operatively. A group of healthy volunteers provided a second control group by being tested twice (n = 80). Results: The brain tumour cohort were significantly impaired by comparison with both control groups at baseline (pre-operatively) on the majority of the cognitive measures, including inspection time. Baseline inspection time scores were significantly related to some scores on the EORTC Quality of Life Questionnaire in the brain tumour group, but not in the spinal surgery group. There was no significant difference between the brain tumour and spinal surgery groups in term of the levels of pre-operative anxiety and depression. The brain tumour cohort showed significantly greater relative deterioration on inspection time following surgery by comparison with both control groups. The brain tumour cohort also deteriorated significantly on several other measures postoperatively by comparison with the healthy control group. Detailed analyses were carried out to determine the differential effects of tumour type, location, and type of surgery (biopsy or resection) on inspection time and other functions in the brain tumour group. Conclusions: Tumour-related cognitive impairment appears to be common in a heterogeneous group of brain tumour patients with a variety of different tumours located throughout the brain. Surgical intervention has a negative impact on function in brain tumour patients, although this deterioration may be transient. General slowing of visual information processing appears to be common to brain tumour patients and the inspection time task provides a feasible and useful method of assessment in brain tumour patients. The task is sensitive to tumour-related brain slowing and can provide a reliable assessment of response to surgery. Given the task’s advantages over more commonly-used cognitive measures, it could be usefully incorporated into cognitive tests batteries in neuro-oncology.

Current status data are usually obtained with a failure time variable T which is diffcult observed but can be determined to lie below or above a random monitoring time or inspection time t. In this work we consider bivariate current status data ${t,delta_1,delta_2}$ and assume we have some prior information of the bivariate failure time variables T1 and T2. Our main goal is to find an optimal inspection time for testing the relationship between T1 and T2.

Due to various shortcomings of the visual system, some visual stimuli can only be identified with 100% accuracy if they are shown for a certain amount of time. This time can be measured using the Inspection Time (IT) paradigm. In an IT task, a “pi” figure with differing leg lengths is typically presented briefly (e.g., 20-200 ms) and is then immediately masked to prevent retinal afterimages. Participants are subsequently required to choose which of the two legs was longer. The objective of this task is to determine the shortest amount of time the pi figure needs to be shown for it to be perceived with 80% accuracy. Given that visual processing has been shown to be altered during and /or prior to a movement, the present experiment sought to test how the requirement to perform a motor task affected IT. Twenty-eight participants took part in the experiment, which was comprised of three conditions: no-movement (NM), peak velocity (PV), and foreperiod (FP). In the NM condition, participants grasped a manipulandum and engaged in the IT paradigm. At the end of every trial, participants verbally stated which leg they believed was longest. In the PV condition participants made a rapid movement to a target, and the IT stimulus was presented when their limb reached peak velocity. Finally in the FP condition the IT stimulus was presented during foreperiod (FP). In all three conditions the IT stimulus was randomly presented from between 15-105 ms (in 15 ms increments) and masked for 400 ms. Results showed no significant differences on the IT task between the NM and FP conditions, suggesting no visual upregulation during foreperiod. However, IT performance was significantly poorer in the PV condition in comparison to both the NM and FP condition, suggesting a visual downregulation at that particular movement kinematic.