Individual choices shape life course trajectories of brain structure and function beyond genes and environment. We hypothesized that individual task engagement in response to a learning program results in individualized learning biographies and connectomics. Genetically identical female mice living in one large shared enclosure freely engaged in self-paced, automatically administered and monitored learning tasks. We discovered growing and increasingly stable inter-individual differences in learning trajectories. Adult hippocampal neurogenesis and connectivity as assessed by a high-density multi-electrode array positively correlated with the variation in exploration and learning efficiency. During some tasks, divergence transiently collapsed, highlighting the sustained significance of context for individualization. Thus, equal environments and equal genes do not result in equal learning biographies because life confronts individuals with choices that lead to divergent paths.

This dataset contains behavioral data from a study examining how individual choices shape brain structure and function in mice, beyond genetic and environmental factors. Using the IntelliCage (IC) system, a fully automated home-cage apparatus from TSE Systems, we tracked the exploratory, learning, and social behaviors of 16 genetically identical female mice over two months. The apparatus is equipped with four operant conditioning corners, each of which contains two water bottles with sweetened water as reward, two nosepoke holes, and doors allowing or denying access to the reward. The system, controlled by software, allowed us to create specific learning protocols where access to rewards was based on predefined combinations of factors, such as corner visits, nosepoke patterns, and individual mouse identity. The IC system recorded events such as corner visits, nosepokes, and licking behavior through RFID antennas, nosepoke sensors, and lickometers. Mice engaged in 10 self-paced, pre-programmed learning tasks, with each animal's behavior automatically recorded and time-stamped. The dataset includes raw event logs, detailing the time, duration, mouse ID, corner ID, and corresponding rewards for each behavioral event, providing a comprehensive look into individual learning trajectories.

During the 10-day adaptation phase, mice were familiarized with the enclosure, presence of rewards at the corners, the requirement to nosepoke for a reward, and the time restrictions on reward access. The learning phase consisted of five main tasks with ten variations: Place Learning with 3 Correct Corners (PL3CC), Place Learning with 1 Correct Corner (PL1CC), Patrolling (with correct corners shifting either clockwise or anticlockwise), Reverse Patrolling (where the correct corner's direction was reversed), and the Serial Reversal Task, in which two diagonally opposite corners were rewarded and the assignment of correct corners was reversed every four days. Behavioral data was recorded automatically by the IntelliCage system, capturing detailed information about each mouse's visits to operant corners. Exploratory behavior during the adaptation phase was assessed through roaming entropy (RE). This metric was derived from the duration of visits to different corners and then converted into the probability of the mouse being located in a specific corner on a given day using Shannon entropy. In the learning phase, the percentage of correct visits was used to track learning, where a correct visit was defined as a visit to a rewarded corner followed by more than one nosepoke, indicating reward-seeking behavior. Flexibility in adapting to new tasks was assessed by calculating how often mice visited previously correct corners after a task reversal.