Device-Embedded Cameras for Eye Tracking-Based Cognitive Assessment: Implications for Teleneuropsychology.

Introduction:Widespread screening for cognitive decline is an important challenge to address as the aging population grows, but there is currently a shortage of clinical infrastructure to meet the demand for in-person evaluation. Remotely delivered assessments that utilize eye-tracking data from webcams, such as visual paired comparison (VPC) tasks, could increase access to remote, asynchronous neuropsychological screening for cognitive decline but further validation against clinical-grade eye trackers is required.Methods:To demonstrate equivalence between a novel automated scoring system for eye-tracking metrics acquired through a laptop-embedded camera and a gold-standard eye tracker, we analyzed VPC data from 18 subjects aged 50+ with normal cognitive function across three visits. The eye tracker data were scored by the manufacturer's software, and the webcam data were scored by a novel algorithm.Results:Automated scoring of webcam-based VPC data revealed strong correlations with the clinical-grade eye-tracking camera. Correlation of mean VPC performance across all time points was robust: r = 0.95 (T1 r = 0.97; T2 r = 0.88; T3 r = 0.97; p's < 0.001). Correlation of per-trial performance across time points was also robust: r = 0.88 (T1 r = 0.85; T2 r = 0.89; T3 r = 0.92; p's < 0.001). Mean differences between performance data acquired by each device were 0.00.Conclusion:These results suggest that device-embedded cameras are a valid and scalable alternative to traditional laboratory-based equipment for gaze-based tasks measuring cognitive function. The validation of this technique represents an important technical advance for the field of teleneuropsychology.

Progressive cognitive deficit, motor impairment and striatal pathology in a transgenic Huntington disease monkey model from infancy to adulthood.

One of the roadblocks to developing effective therapeutics for Huntington disease (HD) is the lack of animal models that develop progressive clinical traits comparable to those seen in patients. Here we report a longitudinal study that encompasses cognitive and motor assessment, and neuroimaging of a group of transgenic HD and control monkeys from infancy to adulthood. Along with progressive cognitive and motor impairment, neuroimaging revealed a progressive reduction in striatal volume. Magnetic resonance spectroscopy at 48 months of age revealed a decrease of N-acetylaspartate (NAA), further suggesting neuronal damage/loss in the striatum. Postmortem neuropathological analyses revealed significant neuronal loss in the striatum. Our results indicate that HD monkeys share similar disease patterns with HD patients, making them potentially suitable as a preclinical HD animal model.

A two years longitudinal study of a transgenic Huntington disease monkey.

BACKGROUND: A two-year longitudinal study composed of morphometric MRI measures and cognitive behavioral evaluation was performed on a transgenic Huntington's disease (HD) monkey. rHD1, a transgenic HD monkey expressing exon 1 of the human gene encoding huntingtin (HTT) with 29 CAG repeats regulated by a human polyubiquitin C promoter was used together with four age-matched wild-type control monkeys. This is the first study on a primate model of human HD based on longitudinal clinical measurements. RESULTS: Changes in striatal and hippocampal volumes in rHD1 were observed with progressive impairment in motor functions and cognitive decline, including deficits in learning stimulus-reward associations, recognition memory and spatial memory. The results demonstrate a progressive cognitive decline and morphometric changes in the striatum and hippocampus in a transgenic HD monkey. CONCLUSIONS: This is the first study on a primate model of human HD based on longitudinal clinical measurements. While this study is based a single HD monkey, an ongoing longitudinal study with additional HD monkeys will be important for the confirmation of our findings. A nonhuman primate model of HD could complement other animal models of HD to better understand the pathogenesis of HD and future development of diagnostics and therapeutics through longitudinal assessment.

A behavioral task predicts conversion to mild cognitive impairment and Alzheimer's disease.

BACKGROUND/RATIONALE: Currently, we cannot reliably differentiate individuals at risk of cognitive decline, for example, mild cognitive impairment (MCI), Alzheimer's disease (AD), from those individuals who are not at risk. METHODS: A total of 32 participants with MCI and 60 control (CON) participants were tested on an innovative, sensitive behavioral assay, the visual paired comparison (VPC) task using infrared eye tracking. The participants were followed for 3 years after testing. RESULTS: Scores on the VPC task predicted, up to 3 years prior to a change in clinical diagnosis, those patients with MCI who would and who would not progress to AD and CON participants who would and would not progress to MCI. CONCLUSIONS: The present findings show that the VPC task can predict impending cognitive decline. To our knowledge, this is the first behavioral task that can identify CON participants who will develop MCI or patients with MCI who will develop AD within the next few years.

Detecting cognitive impairment by eye movement analysis using automatic classification algorithms.

The Visual Paired Comparison (VPC) task is a recognition memory test that has shown promise for the detection of memory impairments associated with mild cognitive impairment (MCI). Because patients with MCI often progress to Alzheimer's Disease (AD), the VPC may be useful in predicting the onset of AD. VPC uses noninvasive eye tracking to identify how subjects view novel and repeated visual stimuli. Healthy control subjects demonstrate memory for the repeated stimuli by spending more time looking at the novel images, i.e., novelty preference. Here, we report an application of machine learning methods from computer science to improve the accuracy of detecting MCI by modeling eye movement characteristics such as fixations, saccades, and re-fixations during the VPC task. These characteristics are represented as features provided to automatic classification algorithms such as Support Vector Machines (SVMs). Using the SVM classification algorithm, in tandem with modeling the patterns of fixations, saccade orientation, and regression patterns, our algorithm was able to automatically distinguish age-matched normal control subjects from MCI subjects with 87% accuracy, 97% sensitivity and 77% specificity, compared to the best available classification performance of 67% accuracy, 60% sensitivity, and 73% specificity when using only the novelty preference information. These results demonstrate the effectiveness of applying machine-learning techniques to the detection of MCI, and suggest a promising approach for detection of cognitive impairments associated with other disorders.

Eye tracking during a visual paired comparison task as a predictor of early dementia.

The authors present findings from a behavioral task (visual paired comparison) using infrared eye-tracking that could potentially be useful in predicting the onset of Alzheimer's disease. Delay intervals of 2 seconds and 2 minutes were used between the initial viewing of a picture and when the picture was displayed alongside a novel picture. Eye-tracking revealed that at the 2-second delay, 6 patients with mild cognitive impairment, 15 matched control participants (normal control), and 4 neurological control participants with Parkinson's disease performed comparably, viewing the novel picture greater than 71% of the time. When the delay increased to 2 minutes, patients with mild cognitive impairment viewed the novel picture only 53% of the time (P < .05), while control participants and participants with Parkinson's disease remained above 70%. These findings demonstrate the usefulness of this task for assessing normal as well as impaired memory function.

Towards a transgenic model of Huntington's disease in a non-human primate.

Non-human primates are valuable for modelling human disorders and for developing therapeutic strategies; however, little work has been reported in establishing transgenic non-human primate models of human diseases. Huntington's disease (HD) is an autosomal dominant neurodegenerative disorder characterized by motor impairment, cognitive deterioration and psychiatric disturbances followed by death within 10-15 years of the onset of the symptoms. HD is caused by the expansion of cytosine-adenine-guanine (CAG, translated into glutamine) trinucleotide repeats in the first exon of the human huntingtin (HTT) gene. Mutant HTT with expanded polyglutamine (polyQ) is widely expressed in the brain and peripheral tissues, but causes selective neurodegeneration that is most prominent in the striatum and cortex of the brain. Although rodent models of HD have been developed, these models do not satisfactorily parallel the brain changes and behavioural features observed in HD patients. Because of the close physiological, neurological and genetic similarities between humans and higher primates, monkeys can serve as very useful models for understanding human physiology and diseases. Here we report our progress in developing a transgenic model of HD in a rhesus macaque that expresses polyglutamine-expanded HTT. Hallmark features of HD, including nuclear inclusions and neuropil aggregates, were observed in the brains of the HD transgenic monkeys. Additionally, the transgenic monkeys showed important clinical features of HD, including dystonia and chorea. A transgenic HD monkey model may open the way to understanding the underlying biology of HD better, and to the development of potential therapies. Moreover, our data suggest that it will be feasible to generate valuable non-human primate models of HD and possibly other human genetic diseases.

Neural correlates of knowledge: stable representation of stimulus associations across variations in behavioral performance.

Behavioral responses to a sensory stimulus are often guided by associative memories. These associations remain intact even when other factors determine behavior. The substrates of associative memory should therefore be identifiable by neuronal responses that are independent of behavioral choices. We tested this hypothesis using a paired-associates task in which monkeys learned arbitrary associations between pairs of visual stimuli. We examined the activity of neurons in inferior temporal cortex as the animals prepared to choose a remembered stimulus from a visual display. The activity of some neurons (22%) depended on the monkey's behavioral choice; but for a novel class of neurons (54%), activity reflected the stimulus that the monkey was instructed to choose, regardless of the behavioral response. These neurons appear to represent memorized stimulus associations that are stable across variations in behavioral performance. In addition, many neurons (74%) were modulated by the spatial arrangement of the stimuli in the display.

Selective neurotoxic amygdala lesions in monkeys disrupt reactivity to food and object stimuli and have limited effects on memory.

Monkeys with bilateral neurotoxic amygdala lesions and normal monkeys were administered tests of emotional reactivity, recognition memory, and reward association memory. There were 3 main findings. First, monkeys with amygdala lesions performed differently than normal monkeys on initial administrations of the emotional reactivity tests and on retests that were given 21-23 months after surgery. Second, they performed like normal monkeys on tests of recognition memory. Third, they were initially impaired on a test of reward association memory, but they were not impaired on a retest that was given 16 months after surgery. These findings underscore the role of the amygdala in aspects of emotional reactivity and reward association memory, but not in recognition memory. In addition, at least some of the behavioral effects of amygdala damage can be long lasting.

Anterograde amnesia and temporally graded retrograde amnesia for a nonspatial memory task after lesions of hippocampus and subiculum.

We studied the importance of the hippocampus and subiculum for anterograde and retrograde memory in the rat using social transmission of food preference, a nonspatial memory task. Experiment 1 asked how long an acquired food preference could be remembered. In experiment 2, we determined the anterograde amnesic effects of large lesions of the hippocampus that included the subiculum. In experiment 3, large lesions of the hippocampus that included the subiculum were made 1, 10, or 30 d after learning to determine the nature and extent of retrograde amnesia. Normal rats exhibited memory of the acquired food preference for at least 3 months after learning. Hippocampal lesions that included the subiculum produced marked anterograde amnesia and a 1-30 d temporally graded retrograde amnesia. The results show the importance of the hippocampus and related structures for nonspatial memory and also demonstrate the temporary role of these structures in long-term memory.