Reinforcement Familiarization

Top  Previous  Next

About the task

 

Purpose: to train the subject to take rewards, and optionally associate them with a sound. The task presents reward at random intervals

 

Configuring the task

 

What constitutes "a reinforcer" is defined in the General Parameters. This allows you to set the reward sound and define what sound is associated with the presentation of a reward.

 

MonkeyCantab_ReinforcementFamiliarizationParams

 

Special options for licker training. These options are only applicable if you have a pump delivering liquid reinforcement and a lick sensor. If you wish, you can proceed through two initial stages before the main schedule starts. These extra options do NOT use the reward parameters specified in the General Parameters - for example, no sounds are played. If you select one or more of these options, they are triggered in the order they are listed in the parameters dialogue box: (1) pump switched on... (2) lick-contingent juice on FR1 schedule... (3) FT/RT schedule.
Pump switched on at start of session. If selected, the pump is activated without a pause at the start of the session, until the subject has made the number of licks specified in the licks box. At this point, the pump is switched off. If you specify zero licks here, the pump will run for the entire session and the program will never proceed to another phase.
Lick-contingent juice on FR1 schedule for rest of session. This option allows you to reward licking for the rest of the session. Each lick activates the pumps for a fixed time (specified in the Each lick triggers... box).
Schedule parameter (s). Now begins the proper schedule; all rewards delivered here follow the settings specified in the General Parameters.
Fixed-time schedule. Reward is delivered at regular intervals. Reward is delivered; the program waits for the end of reinforcement; this parameter specified how long it will then wait before delivering the next reward. (Note that this is not quite the standard parameter of an FT schedule: the standard way is to specify the time between the start of reward 1 and the start of reward 2; here, this parameter specifies the time between the end of reward 1 and the start of reward 2. This is an easy way to prevent the second reinforcer being scheduled while the first is still being delivered! You can think of this parameter as the inter-reinforcement interval.)
Random-time schedule. An imaginary clock ticks once a second. Every clock tick, the program flips a coin that will turn up heads with a probability of 1/parameter. If the coin turns up heads, a reinforcer is delivered. Thus, if the parameter is 30, reinforcement is delivered with a probability of 1/30 every second, meaning that on average reinforcement is delivered once every 30 seconds. (However, for as long as it takes to deliver the reinforcer or the reinforcer-associated sound, the clock stops ticking in order to prevent another reinforcer being delivered at the same time - if you need a precise RT schedule, you must take this time into account too.)
Maximum number of rewards. Once this number of rewards has been delivered, the task stops. (Specify 0 for no limit.) This number includes rewards earned on the FR1 licking part of the task, but not on the "free juice" part.
Maximum time permitted during task. Once this time limit has been reached, the task stops. (Specify 0 for no limit.)

 

You must specify either a maximum number of rewards, or a maximum time, or both.

 

The houselight is on during the task.