The questions that motivate our research all revolve around visual cognition and mostly around visual memory. Specifically, we are interested in how do we maintain visual information that is no longer present? How does the brain represent visual memories across eye movements and over time? What are the neural mechanisms underlying forgetting in healthy people and patients with brain disorders?
We are engaged in investigating these topics using various research methods including neuroimaging and behavioral experiments, in healthy individuals and patients with neurological disorders. We also develop a new computational framework designed to capture novel, key aspects of visual memory.
Visual Working Memory
Develop more sensitive neurological diagnostic procedures
We aim at developing new Neurological and Neuropsychological assessment tools for cognitive deficits. We believe that the conventional methods (mostly based on pen and pencil tests) could be improved by designing patient friendly experiments based on recent knowledge gathered in the cognitive neuroscience field of research.
For example: Alzheimer’s project: Long- term episodic memory deficits in Alzheimer’s disease are part of the disease definition. However, due to the ideosycratic nature of episodic memory, assessing abnormal function is time consuming and requires subjective decisions. Recent studies suggest that patients with Alzheimer’s disease have a specific deficit in remembering bound features in working memory (over a few seconds) compared with unbound or isolated features. We have tested eight individuals with familial Alzheimer’s disease caused by pathological mutations in presenilin 1 and amyloid precursor protein genes, 12 asymptomatic mutation carriers and 62 healthy controls were assessed using a recently developed object localization task with an analogue measure of report. Strikingly. asymptomatic mutation carriers (which had normal scores in standard cognitive tests) had a specific impairment recalling the location of a target item when there were multiple items present. Mislocalization errors were due to participants’ increased tendency to localize the target items at the specific locations of other items in the memory array, thus providing evidence of object- location binding failure. This contrasted with a more extensive pattern of impairment in individuals with established familial Alzheimer’s disease involving identification failures, imprecise localization and object-location binding deficit. This suggests that the fidelity with which object identity and its location are bound in working memory may be a sensitive early cognitive marker of Alzheimer’s disease. Our novel behavioral paradigm may provide a more practical and sensitive means for detecting medial temporal lobe dysfunction in Alzheimer’s disease than conventional neuropsychology tests.
A collaboration with elminda and Sheba hospital for the developing a predictive tool for the efficiency of therapy for clinical depression. This study aims to develop a new tool using several bio-markers and machine learning tools.