A Multi-species Investigation of Memory: from Human Recollection to Rodent Navigation

Robert Gardner

Advisor: Giorgio Ascoli, PhD, Department of Psychology

Committee Members: Theodore C. Dumas, Linda D Chrosniak

Krasnow Building, #222
December 04, 2014, 02:45 PM to 11:45 AM

Abstract:

This dissertation takes a multi-species perspective to investigate memory across several types and focuses of retrieval. The first set of experiments quantified two important, yet relatively unexplored, dimensions of human recollection: its content and frequency. These studies primarily focused on memory of personally-experienced events (autobiographical memory: AM). In particular, participants reported the number of details (among specified content categories, e.g., People, Places, and Things) that comprised word-cued AMs dated to various life periods (see http://cramtest.info). Application of this methodology to subjects from numerous age groups revealed several age-related effects on memory content. Notably, the amount of detail associated with typical AMs increased with the age of the participant, and decreased with the age of the memory. To estimate the occurrence of recollection in natural settings, an experience sampling approach was utilized. In addition to targeting AM, this design probed recollections of to-be-experienced events (termed here, prospective memory: PM). Younger subjects experienced AM and PM equally often. In contrast, while older adults engaged in AM as often as younger subjects, they experienced PM about twice as frequently. In addition, AM and PM occurrence rates were positively correlated, most strongly among younger individuals. Altogether, these experiments on human memory demonstrate differences between younger and older adults both in how personal events are remembered and in the temporal focus of typical recollection.

The second set of experiments investigated the interplay between different decision-making strategies (which rely upon distinct types of memory) during the execution of spatial navigation tasks designed for rodents. A place strategy is attentive, depends on memory of the spatial arrangement of landmarks, and is used to flexibly locate the position of a goal. In contrast, a response strategy is generally thought to be automatic, and engages the expression of a fixed well-learned motor sequence. Replicating prior work, rats increased their reliance on response navigation with repeated training on a dual-solution choice (i.e., one that can be solved using either strategy), suggesting an experience-dependent shift from attentive to automatic performance. However, during serial navigation, when a subsequent (secondary) choice required attentive spatial memory, the strategy transition on the dual-solution (primary) choice was blocked. Control experiments specified this effect to the cognitive aspect of the secondary task. Moreover, this effect was only present when the attentive secondary task was initiated at the onset of training; secondary training did not increase the use of place strategies on the primary choice once response navigation was established. Taken together, these results demonstrate the sensitivity of strategy reliance to the cognitive demand of subsequent behaviors — an influence from which previously formed habits may be protected. In a related experiment, neural activation, based on activity-dependent gene expression, was estimated in structures implicated in place (e.g., hippocampus) and response (i.e., dorsolateral striatum) navigation among rats assigned either to a relatively brief or to a protracted training schedule on a dual-solution task. Results suggest that the ratio of activation across neural structures may be a good predictor of the experience-dependent emergence of response navigation and/or the dominant strategy. Implications and limitations of all experiments are discussed.