My research is into aspects of human memory and attention, including changes in these functions across the adult life span. In terms of theoretical approaches, I take the view that memory processes are activities of the mind and brain as opposed to structural 'things in the head'. By this view, the processes of acquiring information ('encoding') and using or re-experiencing that information later ('retrieving') are mental and neural activities, much as is the case with perceiving and thinking. In an early paper with Robert Lockhart we proposed a 'levels of processing' framework for memory research, stressing this active approach and making the case that 'deeper' semantic processes are the ones associated with best subsequent memory performance. I have conducted a number of experimental studies over the past 30 years to illustrate and confirm this point of view. We have shown for example that memory performance can vary as much as 400% (a fourfold increase) depending simply on the type of processing that is carried out at the time of initial acquisition. Results from other labs have also shown that a second major factor is the similarity between encoding operations and subsequent retrieval operations -- greater similarity means better performance.More recently I have been involved in work exploring the brain correlates of these effects using the functional neuroimaging facilities at the Rotman Institute. Several studies have now shown that the benefits to memory found with deep semantic processing are associated with neural activity in the ventral areas of the left prefrontal cortex.My research into memory changes with aging follows the same principles. Age-related memory problems are real, but the degree of impairment is quite variable depending on how memory is tested. The ability to recall events and facts with few aids and out of context is typically poor in older adults, whereas the ability to recognise verbal or pictorial material can be virtually at the level of a younger person. It thus seems that the ability to 'self-initiate' appropriate mental processes is difficult for older people (possibly because they involve frontal lobe functions) and that by the same token older adults profit greatly from the 'environmental support' provided by reinstating the context of original learning or by the test materials themselves, as with recognition tests. We are currently applying these ideas to the development of techniques to counteract memory failures in older people. One interesting problem relates to the nature of the benefits associated with physical activity and with long-lasting mental activities such as solving puzzles and lifelong bilingualism. These effects contribute to 'cognitive reserve' and one focus of current work is to document these effects systematically and to gain some insight into their neural correlates.This research is supported by the Natural Sciences and Engineering Research Council of Canada, the Canadian Institutes of Health Research, and by the Alzheimer Society of Canada.