Discoveries of long-term potentiation and immediate early gene in the central nervous system have enabled new developments in experiments on learning and memory. These experiments are conducted in many kinds of animals with different procedures, physiology, chemistry and pharmacology. However, there is still some confusion when these various procedures are discussed. Memory is defined as information storage of an animal's previous experiences. The memory induces changes in behavioral performance. This means that memory must be observed in whole animals, and one question that can occur is how does long-term potentiation, for example, correlate with memory. Furthermore, memory has been divided into two major classifications, declarative and non-declarative, from the comparison of amnesias observed in humans and animals. The declarative memory can be observed in human subjects, but not in animals. This article presents a neuronal circuit concerning memory formation and some results obtained from benzodiazepines, and it discusses some problems encountered executing when experiments on learning and memory. In addition, the discussion speculates over the possibility for an “antidementia drug”.
This report deals with three different maze methods using spontaneous learning behavior. Fortyeight mice housed in an apparatus with a multiple maze mastered the maze task on the 5th day after the start of housing. Then they were divided into four groups, and two of the groups were treated with AF64A (3.5 nmol, i.c.v.) or trimethyltin (TMT, 3 mg/kg, p.o.), and the other two groups were treated with the vehicle as the respective control. Seven days after the treatment, their memory retrieval was tested. Subsequently, the same mice were housed in the apparatus with a T-maze. After the finish of the experiment using the T-maze, they were housed in the apparatus with an eight arm radial maze. The control groups mastered the T-maze task with a 3-sec delay on the 4th day after the start of housing and the radial maze task on the 10th day after the start of housing. Both the treatments lowered the performance in all maze tasks. These results show that the mice housed in the apparatus with maze learned to negotiate the maze spontaneously, and the apparatuses are useful for estimating memory in mice with little effort.