Alzheimers disease (AD) is a chronic neurodegenerative disease characterized by the accumulation of extracellular beta-amyloid (Aß) plaques and intracellular neurofibrillary tangles in the brain and a profound memory loss with increasing age. Transgenic mouse models have been developed which partially mimic similar aspects observed in human AD. The amyloid cascade hypothesis postulates an imbalance in Aß production and clearance to initiate AD pathology. Soluble Aß is generally accepted as the putative culprit furthermore initiating tau hyperphosphorylation and other pathological events. Targeting Aß production, aggregation and clearance have been suggested as possible avenues to ameliorate AD pathology and improve cognitive functions. In this dissertation, we approached and investigated AD pathology from a behavioral neuroscientific perspective with cognitive performance as our primary readout. Behavioral readouts were supplemented with immunohistochemical readouts where possible.We introduced a multimetric behavioral test battery which measures different memory processes such as spatial, extinction and fear learning. Cognitive performance was evaluated in two AD mouse models, each representing one side of AD pathology namely cerebral amyloidosis (APPPS1-21) and tauopathy (THY-Tau22). We hypothesized that different memory processes might be affected in these models based on the assumption that amyloid and tau pathology follow different but predictable expansion patterns. 9 mo old THY-Tau22 mice showed deficits in spatial learning (hippocampus-dependent), whereas similar aged APPPS1-21 mice showed deficits in extinction and fear learning in addition (frontal cortex- and amygdala-dependent). Since the prefrontal cortex is the starting point and end point for amyloid and tau pathology, respectively, a minor additional hypothesis was that cognitive functions controlled by the frontal cortex would contribute most to the differentiation of APPPS1-21 transgenic mice and wild type mice. This minor hypothesis was not confirmed with discriminant function analysis.We next explored the potential benefits of a novel apparatus to measure learning and memory in mice. Group-housed APPPS1-21 are tested in the IntelliCage® for an extended time period with little human interference. Preliminary data show that APPPS1-21 transgenic mice performed equally well to wild type mice. Social interaction with wild type mice during IntelliCage® testing might have restored cognitive deficits in APPPS1-21 mice. Studies confirmed environmental enrichment to be a potential factor in improved learning. We are currently investigating the effects of separate genotype housing on learning in APPPS1-21 mice.With our multimetric test battery, we evaluated the efficacy of several therapeutic compounds to intervene with AD-induced cognitive deficits and biochemistry. We hypothesized that TUDCA would mediate gamma-secretase activity through reduction of CTGF expression. TUDCA-supplemented APPPS1-21 mice displayed improved memory and reduced amyloid deposition compared to control APPPS1-21 mice. TUDCA supplementation did not cause adverse effects such as weight loss or changes in grooming activity. Decreased CTGF expression and other biochemical changes were provided in other published manuscripts.SEN1500 was recently introduced as a small molecule compound that interferes with soluble Aß toxicity. SEN1500 treatment resulted in a dose-dependent improvement of memory in APPPS1-21 mice, but did not change soluble or insoluble Aß levels. Possibly synaptic changes occurred without changing Aß levels. This was suggested in our in vitro synaptophysin assay.In a recent report, bexarotene rapidly cleared amyloid deposition and reduced cognitive deficits in several AD mouse models. We failed to replicate these findings in three independent experiments: Swiss CD1 mice, Beagle dogs and APPPS1-21 mice. No changes in Aß levels in APPPS1-21 mice were measured even though bexarotene did reach the central nervous system. Moreover, bexarotene treatment resulted in adverse effects in APPPS1-21 mice such as weight loss, skin rash and breathing problems.We observed improved spatial learning, spatial retention memory and fear retention memory after SSP-002392 treatment, a 5-HT4 receptor agonist, in scopolamine-treated mice. Moreover, 5-HT4 treatment did not change anxiety levels. 5-HT4 receptor stimulation is recently considered a potential symptomatic treatment for AD.We will discuss potential limitations of rodent AD models and challenges in developing novel mouse tasks. There is a definite need to continue research on new and better protocols and apparatus for reliable and valid assessment of cognitive performance and other aspects of mouse behavior. We recommend a more careful attitude in the interpretation of preclinical results and their translation to the human condition. The behavioral neuroscientist/psychologist, armed with a full repertoire of sensitive tools to measure cognitive and behavioral changes in preclinical models, is a crucial player in the search for novel and clinically active disease-modifying therapeutics, and ultimately in the war against AD. Multidisciplinary research should focus on more nuanced and prudent analysis of experimental therapeutics to limit the amount of too much good news presently confusing the field. Most of all, we share the hope of millions of AD patients and their caretakers for curative AD therapy that would reverse AD neuropathology as well as restore cognitive functioning with minimal adverse effects.