Affiliations: [a] Faculty of Life Sciences, University of Manchester, Manchester, UK | [b] Department of Psychology, University of Durham, Durham, UK
Correspondence:
[*]
Correspondence to: Dr. John Gigg, Faculty of Life Sciences, University of Manchester, 3.614 Stopford Building, Oxford Road, Manchester M139PT, UK. Tel.: +44 161 306 3876; E-mail: j.gigg@manchester.ac.uk.
Abstract: Episodic memory loss is a defining feature of early-stage Alzheimer's disease (AD). A test of episodic-like memory for the rat, the What-Where-Which occasion task (WWWhich), requires the association of object, location, and contextual information to form an integrated memory for an event. The WWWhich task cannot be solved by use of non-episodic information such as object familiarity and is dependent on hippocampal integrity. Thus, it provides an ideal tool with which to test capacity for episodic-like memory in the 3xTg murine model for AD. As this model captures much of the human AD phenotype, we hypothesized that these mice would show a deficit in the WWWhich episodic-like memory task. To test the specificity of any episodic-like deficit, we also examined whether mice could perform components of the WWWhich task that do not require episodic-like memory. These included object (Novel Object Recognition), location (Object Location Task, What-Where task), and contextual (What-Which) memory, as well as another three-component task that can be solved without reliance on episodic recall (What-Where-When; WWWhen). The results demonstrate for the first time that control 129sv/c57bl6 mice could form WWWhich episodic-like memories, whereas, 3xTgAD mice at 6 months of age were impaired. Importantly, while 3xTgAD mice showed some deficit on spatial component tasks, they were unimpaired in the more complex WWWhen combination task (which includes a spatial component and is open to non-episodic solutions). These results strongly suggest that AD pathology centered on the hippocampal formation mediates a specific deficit for WWWhich episodic-like memory in the 3xTgAD model.
