Functional Imaging with Reinforcement, Eyetracking, and Physiological Monitoring

This tutorial shows a method for studying the mechanism of reinforcement using functional brain imaging or FMRI. In this protocol, subjects undergo functional imaging while performing a visual decision making task, which is reinforced by stimuli that a rewarding juice aversive, air puff, or neutral auditory tone through scanner compatible goggles. Subjects must decide whether an array of dots are traveling fast or whether they are moving slow.

Eye movements and physiological responses such as breathing and heart rate are also monitored. Hi, I'm Jack Grin Band from the Laboratory of Vince Ferrera and Joy Hirsch in the Department of Neuroscience at Columbia University, and I'm Franco Past. Also from the lab here at Columbia University.

Today we'll show you a procedure for functional imaging with behavioral reinforcement, physiological monitoring, and eye tracking. We use this procedure in our laboratory to study decision making with a positive and adverse reinforcement. So let's get started.

The first steps in running an FMR experiment involve setting up and checking the equipment. These steps can be done in any order during our experiments, subjects are rewarded for making correct behavioral responses. The rewarding stimulus is the subject's favorite beverage delivered by a juice dispenser that controls the amount of liquid delivered to the subject's mouth.

The juice dispenser consists of a reservoir, a computer controlled solenoid valve, and a long tube that delivers juice to the subject. All the electronic components of this dispenser are kept outside the scanner room, so they don't introduce any artifacts in the MR signal. The juice dispenser is rinsed and filled with the subject's preferred beverage.

Before scanning, the system is tested to make sure that the juice is flowing. Subjects are punished for incorrect responses by a 50 millisecond puff of air to the eye. A pressure regulator delivers a controlled air puff, which the subject will rate as aversive but not traumatic.

The pressure regulator is initially set to 30 PSI and is connected to a source of compressed air, which could be an air tank or house air. The solenoid valve is controlled by a computer signal. The air is delivered to the subject through a piece of one 16th inch tigon tubing, which is kept shorter than about eight feet.

In order to avoid a delay between the time the valve opens and the time that the compressed air reaches the subject, hence the pressure regulator sits inside the scanner room because any motion such as a reaction to the air puffs or head movement during swallowing may introduce artifacts into the FMRI image. A bite bar is used to minimize head movement. The bite bar is attached to the RF head coil and has a custom fitted mouthpiece made from thermoplastic material tubing for juice delivery and respiratory gas monitoring is incorporated into the mouthpiece.

The subject should simply maintain good contact between their upper teeth and the mouthpiece. It is important to tell them not to bite down hard on the mouthpiece. This will cause their jaw muscles to tire heartbeat is monitored during the experiment in order to later remove any artifacts during data analysis that might arise from increased blood oxygenation in the brain, not due to neuronal activity.

To measure changes in oxygenation related to the cardiac cycle, we use a pulse oximeter to measure blood ox oxygenation in the fingertip using an infrared sensor. The pulse oximeter is MR compatible and its output signal is fed through a filter panel to the control room where it is digitized and stored on computer. We also measure respiration with a respiratory gas monitor or RGM that measures expired carbon dioxide levels.

Since respiration affects blood oxygenation, the RGM is MR compatible and its output signals are sent to the control room where they are digitized and stored on computer. Visual stimulation is provided by a pair of scanner compatible goggles, which stimulate both eyes independently. The goggles include a miniature infrared camera for measuring eye movements in order to tell where the subject is looking, we track eye movements using infrared video O iconography.

Movements of the eye affect the position of visual stimuli on the retina, which affects visual responses in the brain. Eye movements can also be used by the subject to indicate VA aural responses. In this method, an infrared camera is used to track movements of the pupil.

The infrared emitter and camera are built into a pair of specially designed goggles. These are the same goggles that provide visual stimulation. The data from the goggles is processed by a dedicated computer that converts images of the eye into analog signals for horizontal and vertical eye position.

The scanner is located in an electrically and magnetically shielded room. All electrical signals going from the control room to the scanner room pass through a filter panel, remove any frequencies that could create an artifact in the MR image. Before scanning, each subject must go through a safety and consent process.

The study is explained, all risks are discussed and the subject gives their consent. This process is designed to protect subjects from coercion and to protect both their privacy and health. The subject is screened for metal anywhere inside or outside the body.

Prior to experimentation, subjects are made thirsty by voluntarily restricting their fluid intake for six hours. This way, juice received during the experiment becomes extremely rewarding. When ready to be scanned, the subject must put in earplugs to protect the ears from the scanner noise.

The subject must also put on MR compatible headphones to communicate with the researchers and to hear instructions. During the scanning session, we use T one weighted structural images that give a clear definition of the morphology of the subject's brain. The subject lies passively and remains as still as possible.

During this phase, which lasts about 10 minutes, we use T two weighted functional sequences to show changes in blood oxygenation that are correlated with neural activity. It is during this phase that the subject performs a perceptual decision making task with visual stimuli. The subject views a pattern of moving dots and makes a perceptual judgment about their direction or speed of motion.

The subject indicates their response by pressing buttons. These responses are reinforced by juice of correct and air puffs. If incorrect, auditory stimuli are used as secondary reinforcers.

Eye movements are continuously monitored during this phase. Each experimental run lasts 11 minutes and there are four runs in a scanning session. We use diffusion weighted images, DWI to determine structural connectivity between brain regions.

During this phase. The subject lies motionless for about 12 minutes. We've just shown you a functional imaging experiment with behavioral reinforcement, physiological monitoring, and eye tracking.

We have demonstrated how to use this procedure to study decision making with aversive and petitive reinforcement. So that's it. Thanks for watching and good luck with your experiments.