Movies

For a general description of the research done in the Shadlen lab, please see the lab HomePage. I have done movies for many of the researchers that have worked in the Shadlen lab. Below are movies made representing some of the earliest experiments done in the lab.

Some more recent experiments were done by Tianming Yang. Those experiments are described in this article. The movies that I made for Tianming are described and linked to from this page.

My boss was on the Charlie Rose show, and showed some movies I made. More information about that is on my Blog.

Jamie Roitman (now at University of Chicago, this article describes her most recent research) did some experiments in the lab that are described in this PDF (Roitman JD, Shadlen MN (2002), Response of neurons in the lateral intraparietal area during a combined visual discrimination reaction time task. J Neurosci 22:9475-9489). These movies are a re-creation of a couple of trials from those experiments. The movies show what the monkey saw, and super-imposed on this is where the monkey was looking during the experiment (yellow trace). During the trial, the monkey saw random dot motion, in which a proportion of the dots were moving for limited lifetimes in one of two directions. The monkey decides which direction the dots are moving, and then when the center dot (white, called the fixation point, because during the trial the monkey is required to fixate on this dot) turns off, the monkey moves his/her eyes to the red target that corresponds to the direction the monkey believes the dots were traveling towards. During the experiment, she recorded from neurons in the lateral intraparietal (LIP) area of the cortex, which we believe are involved in the decision-making process for this task. The reason we believe this can be seen in the following movies. Both of these movies are from trials taken from an experiment in which she recorded from the same neuron. This neuron fires when the monkey chooses the target in the upper right. Everything below the bottom red dot is a re-creation of the neuron firing, to show the pattern of the firing and how it corresponds to what the monkey is doing/seeing, and not something that the monkey sees during the experiment. The yellow line at the bottom of the movie is a timeline that shows when the random motion came on (jumps to a high value while motion is on), and above that, white vertical lines that correspond to the neuron spiking (also called firing). These present themselves in real-time during the movie; the ticks that you hear also correspond to the neuron spiking.

Neuron n585f, trial 47

In this first movie, you will notice how the amount of spikes increases dramatically when the motion comes on. This is because this neuron in LIP has evaluated the information coming in from visual area MT (middle temporal) and its change in spike rate correlates to the motion being toward the top-right. We believe these LIP neurons are not simply reporting which direction the motion is going, but are instead involved in the decision process. While the stimulus is on, the firing rate appears to increase until it reaches a threshold value. This can be seen more clearly in the figures from Jamie's paper. During the delay portion of the trial (after the motion has turned off, but before the fixation point has turned off), MT neurons stop firing, since there is no stimulus, but you will notice that the LIP neuron we are recording from continues to fire. The neuron seems to "remember" which direction the motion is going. The small triangle above the yellow line indicates when the eye left the fixation point to report the decision, which follows the fixation point going out, so it is easy to see the delay period in this trial. Additionally, in cases where the monkey chose the wrong target, the firing of the neurons we record from in LIP represent the the wrong target, although usually the response of MT cells is consistent with the direction of motion.

Now let's look at a movie where the direction of motion is away from the preferred direction of motion for this cell. In this trial, the firing of the cell dramatically decreases after the motion comes on, and continues to fire at a minimum throughout the rest of the trial. Sure enough, this time, the monkey looks towards the target on the bottom left.

Neuron n585f, trial 43

These movies were made by taking data recorded during the experiment, using Matlab to create the sound file using the spike data, and then converting the recorded data format to XML, and importing it and the sound file to Flash. The movies were then exported as Quicktime movies.

  LinkedIn