Of two runs. During the first (sample) run (upper panel) the mouse is forced to choose one of the two target arms by blocking the other arm. In the second (choice) run (lower panel), it is rewarded if it chooses the previously unvisited arm. () Wild-type ( = 9) but not TgH1aV.Fb ( = 12) learned the task. TgH1aV.Fb mice, however, showed the natural preference of rodents to visit the previously unvisited arm on this task, as shown by significantly better than chance level (50%) of spontaneous alternation during the first training block. () The success rate of wild-type increased over the course of the training. TgH1aV.Fb mice, however, started and finished the training at the same success rate without any significant learning throughout the training. () To question if the working memory deficit can be recovered upon reversal of the transgenic expression in the adult animal, we tested the wild-type ( = 7) and TgH1aV.Fb mice ( = 7) after giving dox for 4 weeks, starting 4 months after birth, and tested them on the T-maze. After five blocks of training, wild-type but not TgH1aV.Fb mice learned the task. () To study if the Homer1a expression in hippocampus alone could cause the working memory deficit, we virally expressed H1aV in hippocampi of adult wild-type mice for 4 weeks before testing the mice on the T-maze. Mice injected with an empty viral construct ( = 6) learned the task just like the uninfected wild-types (gray bars). Mice expressing H1aV in the hippocampus ( = 7, black bars), however, failed to learn the task, even after five blocks of training.Copyright information:Taken from "Select Overexpression of Homer1a in Dorsal Hippocampus Impairs Spatial Working Memory"Frontiers in Neuroscience 2007;1(1):97-110.Published online Jan 2007PMCID:PMC2518050.