Loading…
CNS*2020 Online has ended
Welcome to the Sched instance for CNS*2020 Online! Please read the instruction document on detailed information on CNS*2020.
Back To Schedule
Sunday, July 19 • 8:00pm - 9:00pm
P16: A computational model for time cell-based learning of interval timing

Log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
Sorinel Oprisan, Tristan Aft, Michael Cox, Mona Buhusi, Catalin Buhusi

Zoom meeting link:  https://cofc.zoom.us/j/4096287484
P108 from 2:00-2:30 PM
P16 from 2:30-3:00 PM

Lesion and pharmacological studies found that interval timing is the emergent property of an extensive neural network that includes the prefrontal cortex (PFC), the basal ganglia (BG), and the hippocampus (HIP). We used our Striatal Beat Frequency (SBF) model with a large number of PFC oscillators to produce beats from the coincidence detection performed by BG [1,2]. The response of the PFC-BG neural network provides an output that (1) accurately identifies the criterion time, i.e., the time at which the reinforcement was presented during reinforced trails, and (2) is scalar, i.e., the prediction error is proportional to the criterion time. We found that, although the PFC-BG can create beats, the accuracy of the timing depends on the number of PRC oscillators and the frequency range they cover [4].

The ability to discriminate between multiple durations requires a metric space in which durations can be compared. We hypothesized that time cells, which were recently discovered in the hippocampus and ramp-up their firing when the subject is at a specific temporal marker in a behavioral test, can offer a time base for interval timing. We expanded the SBF model by incorporating the HIP time cells that (1) provide a natural time base, and (2) could be the cellular root of the scalar property of interval timing observed in all behavioral experiments (bit see [5]). Our model of interval timing learning assumes that there are two stages of this process. First, during the reinforced trials, the subject learns the boundaries of the temporal duration. This process is similar to the HIP space cell activity that first forms an accurate spatial map of the edges of the environment. Subsequently, the time cells are recruited to cover the entire to-be-timed duration uniformly. Without any learning rule, i.e., without any feedback from the PFC-BG network, the population of time cells simply produces a uniform average time field. In our computational model, the learning rule requires the HIP time cell to adjust their activity to mirror the output of the PFC-BG network. A plausible mechanism for the modulation of HIP time cell activity could involve dopamine released during the reinforced trials. We tested numerically different learning rules and found that one of the most efficient in terms of the number of trails required until convergence is a the diffusion-like, or nearest- neighbor, algorithm.

References

[1] Oprisan SA, Aft T, Buhusi M, and Buhusi CV, Scalar timing in memory: A temporal map in the hippocampus, J. Theor. Biol. 2018, 438:133 – 142.

[2] Oprisan SA, Buhusi M, and Buhusi CV, A Population-Based Model of the Temporal Memory in the Hippocampus, Front. Neurosci. 2018, 12:521.

[3] Buhusi CV, Oprisan SA, Buhusi M. Clocks within Clocks: Timing by Coincidence Detection. Curr Opin Behav Sci. 2016, 8: 207-213.

[4] Buhusi CV, Reyes MB, Gathers CA, Oprisan SA, Buhusi M. Inactivation of the Medial-Prefrontal Cortex Impairs Interval Timing Precision, but Not Timing Accuracy or Scalar Timing in a Peak-Interval Procedure in Rats. Front Integr Neurosci. 2018, 12:20.

[5] Oprisan SA, Buhusi CV. What is all the noise about in interval timing? Philos Trans R Soc Lond B Biol Sci. 2014, 369(1637): 20120459.

Speakers
avatar for Sorinel Oprisan

Sorinel Oprisan

Professor, Department of Physics and Astronomy, College of Charleston


timing pdf

Sunday July 19, 2020 8:00pm - 9:00pm CEST
Slot 17