Meta Questions
In the figure above, note that the learning functions serve to organize the research projects and personnel into implementation units. These units in turn, feed back into the meta-questions. This symbolic feedback allows for more focused research projects and conversations within and across these units, while at the same time the needs of these learning functions feed back into our cross-function initiatives and influence the design of our meta-questions.
The purpose of CELEST is our enduring vision to study the brain as a learning system. The evaluation of learning by considering a systems level approach has been a guiding concept since the start of CELEST. For the first five years of CELEST, our scientific initiatives were organized by research thrusts that represented traditional boundaries of neuroscience (e.g. audition, vision, motor control, etc), which over time limited the growth of our research across these traditional boundaries. After Year 5, we self-critically evaluated our research, identifying the synergies already ongoing and those that could evolve, and the strengths of our researchers and faculty. With advice from the distinguished members of our External Advisory and Scientific Review Board (EASRB), we defined our four cross-function initiatives and learning functions.
As we begin Year 7, these cross-function initiatives are the drivers of our science, while the learning functions have emerged as natural categories for implementation units that feed back into our initiatives. By posing scientific questions driven by the cross-function initiatives, we can constrain our research to allow us to advance the neuroscience of learning by evaluating the brain as a system. As a result, the top of the pyramid in the figure above represents the most stable and enduring guiding principle in our center and the bottom represents the fluid and dynamic nature of our research projects as we build upon our research results and advance our understanding of learning.
The direct impact of our Center is felt with the development of technologies: from bench to models to applications. Our technological impact focuses on three key areas: neuromorphic computing with an emphasis on autonomous robotics, brain-machine interfaces, and medical applications. Figure 2 shows how the integration of experiments, modeling, and technology across our cross-function initiatives and learning functions creates our center added value.
CELEST facilitates cross-talk between experimentalists, modelers, and engineers and supports transition of research up the columns (as illustrated in red) of the figure, i.e. from bench to models to applications. Note that it is not possible for CELEST projects to “fill” all 48 cells in the figure, but that CELEST supports projects that have the best potential to develop outgrowths in a vertical direction of this figure.
The graduate students at CELEST are extremely energetic and motivating to our Center. They work together to build interesting collaborations and motivate research directions. The students generate a strong bottom-up push by their enthusiasm and interest in CELEST research.