The study of learning has engaged scientists, educators, and philosophers for centuries, underscoring that multiple approaches and levels of analysis provide important, convergent insight into its mechanisms. Two Psychology Professors, Sarah London and Marc Berman, have joined their complementary expertise to tackle novel questions about the relationship between development, experience, and neural networks in the context of learning.

The London lab employs a combination of molecular, genomic, and behavioral strategies to identify neural properties that promote and limit the ability to learn in a non-human animal, the zebra finch songbird. The Berman lab combines non-invasive functional Magnetic Resonance Imaging (fMRI) and network-level computational tools to determine the effects of the physical environment on properties of neural networks and cognitive outcomes in humans. Over the course of months, London and Berman had several casual conversations about how interesting it would be to fuse the advantages of their expertise. Notably, individual zebra finch songbirds can be raised in controlled environments that manipulate brain functioning. With fMRI, one can quantify whole brain network properties from multiple timepoints within the same individual to assess how environmental manipulations over time affect brain functioning. From these conversations, they began a collaboration focused on a novel question: Can the ability to learn be predicted by features of a brain network?

Together, the labs devised methods to acquire anatomical and functional scans from zebra finches across development, and to compare neural networks from birds who had been reared in environments known to differentially affect the ability to learn. Analysis of the datasets are ongoing, but with extensive work from their PhD student, Elliot Layden, several intriguing results are emerging about how brains work. First, the mammalian brain has tremendous functional symmetry as measured by the correlation between brain activity of homologous pairs of brain regions (e.g., the left superior parietal lobule and the right superior parietal lobule). Much of this symmetry has been attributed to the Corpus Collosum (CC), a dense bundle of axonal fibers that connect the two hemispheres.   Interestingly, zebra finches also exhibit a high degree of functional symmetry that matches that of mammals, but, importantly, zebra finches do not have a CC.  This research demonstrates that functional symmetry may have evolved much earlier than when the CC developed suggesting that the CC may accentuate functional symmetry, but is not the cause of it. Second, intriguing differences in functional symmetry occur when zebra finch begin to learn to sing compared to when they are kept from hearing bird song. In addition, it appears that increased symmetry is related to the potential to learn and decreases as learning progresses.

These projects have revealed how brains with different structures “solve” the same cognitive problems to support learning of complex and meaningful behaviors. From them emerge new insight into neural networks for learning, which opens new avenues for future inquiry. London and Berman appreciate that these successes could only have resulted from collaborative efforts across psychological disciplines, and are looking forward to continuing to work together.