Exploring hippocampal learning and retrieval in the T-maze

Observing the behaviour of the virtual rat during learning and retrieval

• The virtual rat is learning during the time-span from t=0 to t=15000 ms of the simulation. After that, the virtual rat navigates by retrieving episodic memories of paths through its hippocampal network.
• Open the camera object to see the virtual rat move during the simulation.
  
• Open the SpikePositionRecorder to see the place fields that develop during the learning phase of the simulation.
  

Looking at patterns of activity in CA3, ECIII and CA1

• Use the vector recorders attached to the CA3, ECIII and CA1 populations, and the central vector recorder that receives input from all three populations. Insure that the "recordFromStart" box is on.
• To see line graphs, select "plot: lines". Note that you may need to increase the number of points that are displayed when the simulation is run to 20000 ms. To do so, open the relevant vector recorder, click on the rectangular button at the top-right within the graph window and select "Helper". Then set "maxNPoints" to a larger value, e.g 60000. To zoom into specific sections of a line graph, press the left mouse button (to zoom only along one axis, press when the mouse is located either below the x-axis or to the left of the y-axis). The right mouse button zooms out. You can choose to see or omit the graphs of individual signals by toggling the switches under "displayedData". Note that the first 3 signals have been included to provide insight into the theta modulation of synaptic transmission and plasticity in the three populations.
• To see the strengthening of weights, open the synapseRecorder attached to the recurrent fibres of a population.
  

Observing the breakdown of function without theta

• Edit the normalized-theta population (smaller cyan block to the left of the STM-buffer population).
  
• Select the vectorRenormalizer within the normalized-theta population.
• Modify the theta modulation of synaptic transmission by changing the values for parameters "Y0" and "Y1". In the working model, a potential of -80 mV is converted to the modulation ratio 0 (in "Y0"), and a potential of -60 mV is converted to the modulation ratio 1 (in "Y1"). By setting the value for "Y1" to a more positive value, e.g. -40 mV, the maximum input of -60 mV will be converted to a lower ratio (to 0.5 if -40 mV was set). Similarly, if the value in "Y0" is made more negative, e.g. -100 mV, then the minimum input of -80 mV will be converted to a higher ratio (to 0.5 if -100 mV was set). If both are altered, the theta modulation of synaptic transmission can be limited to a smaller range. E.g. Setting "Y0" to -100 mV and "Y1" to -40 mV will restrict the ratios that are produced to the range from 0.33 to 0.67. Such a range will provide significantly less modulation, and therefore less distinction between operational phases, such as encoding and retrieval modes.

Changing the location of the food reward

• Use the left mouse button to zoom into the T-maze. Locate the circular insertion point of the food reward (the object with the "fruit-loop" image) in the left arm of the T-maze.
  
• Use the mouse to drag this insertion point from the left arm of the T-maze to the end of the right arm.
• Run the simulation again to observe that the virtual rat is able to learn and recall this new path to the food reward goal.

Changing the location of the retrieval start location

• Zoom into the T-maze. Locate the circular insertion point at the bottom of the T-maze stem.
  
• Drag this insertion point some distance up the stem of the T-maze. At t=16000 ms, the virtual rat will be placed there.
• Run the simulation again to observe that the virtual rat can recall a path that leads to the goal anywhere along the known trajectories.