Embedded Files
Byte Sized Letter Ten
Byte Sized Letter Ten
On This Week's Newsletter - Tuesday, 11/19 to Monday, 11/25
On This Week's Newsletter - Tuesday, 11/19 to Monday, 11/25
Click on any of the titles to read the full piece!
This UC Davis study investigated how our brains help us plan and navigate through different environments to reach specific goals. The study had participants solve mazes in an fMRI scanner and examine how their hippocampal place cells encoded the maze. Researchers found that participants who had the same goal had more similar representations of the world than those with different goals. Researchers also found that the hippocampus looks ahead and can identify critical points.
Prior to the publication of this paper, it was formalized that, through place cells and grid cells, the hippocampus played a role in constructing our internal representation of the world around us, a cognitive map. This paper (Stachenfeld et al. 2017) proposed that the hippocampus may encode something more akin to a predictive map.
This work by Piray and Daw is intimately related to the Successor Representation model of a cognitive map. The researchers propose the Default Representation, DR, which improves upon the SR by removing the SR's fundamental biases while being equally representative of the hippocampus. The DR is also better at planning and replanning and can better explain various behaviors (such as habits).
This Oxford study reveals how the hippocampus works like a Lego set, combining pre-existing mental building blocks to create new understanding. Rather than learning everything from scratch, the brain reuses and recombines familiar elements to quickly adapt to new situations: it uses a process called "replay," allowing the brain to rehearse and update mental maps, allowing for efficient learning and navigation in both familiar and new environments.
This study by Whittington et al. proposes a model that unifies two views of the hippocampus: spatial (Tolman) and non-spatial relational memory (Eichenbaum). The two views can be unified under the common language of constructing useful generalizations. The model was shown to be able to learn abstract concepts, like families, after only being told about certain familial relationships between certain individuals.
This UCL study looked at how the brain applies knowledge across different situations by having participants learn relationships between images arranged in graph structures of differing complexities. A brain region called the entorhinal cortex could recognize and apply hexagonal patterns across different-sized networks with different images, similar to how we navigate spaces. However, this ability didn't work for more complex patterns.
This paper by George et al. proposes a probabilistic sequence model that can learn cognitive maps from aliased (ambiguous) sensory observations. Essentially, they propose that the mental representation of space, our “cognitive map,” is a latent property of higher-order sequence learning. This allows the model to explain diverse hippocampal phenomena through a single-sequence learning mechanism.
Want to submit a piece? Or trying to write a piece and struggling? Check out the guides here!
Thank you for reading. Reminder: Byte Sized is open to everyone! Feel free to submit your piece. Please read the guides first though.
All submissions to berkan@usc.edu with the header “Byte Sized Submission” in Word Doc format please. Thank you!
Page updated
Google Sites
Report abuse