Talos (item #01010) is an integrated platform for automated training and analysis of animal’s behavior in decision-making tasks. It features precise, moment-to-moment experimental control, and seamless integration of behavior tracking with recording and manipulating neural activities. Its comprehensive capabilities expedite the experimental timeline significantly while markedly improving data reproducibility by minimizing the need for human intervention.

It is crafted with lever pressing and automated food and water delivery to training and testing mice in both the free-consumption choice task and the two-alternative forced choice (2AFC) instrumental task. Its primary aim is to uncover the neural circuits that govern goal-directed decision-making related to hunger and thirst. (Click to access the publication in Nature Neuroscience)

Price inquiry: info@behaviorsync.com

Key features

  • Compatible with tethered animals.
  • Realtime tracking of animal’s motion, level pressing and licking behavior.
  • An automatic reward delivery system that ensures meticulous control of food and water disbursement (6ul).
  • Utilization of four motorized slides to enable automated extension and retraction of two levers, food dispenser, and waterspout.
  • High-precision integration of optical stimulation and electrophysiology recording.
  • Customized field programmable gate array (FPGA) circuitry for processing camera and sensor data, while initiating reward delivery and manipulation of brain activities within nanoseconds.
  • Software runs on a windows host PC to send commands to and collect data from the behavioral apparatus via a USB cable.


Contact us to customize the dimensions at info@behaviorsync.com

  • Constructed from high-quality acrylic, the mouse cage boasts a front wall on hinges for easy mouse loading, which is securely fastened with a magnetic latch. It also features a stainless-steel mesh floor accompanied by a sleek drawer underneath, designed to gather mouse droppings for convenient cleaning and maintenance.
  • Speakers to deliver auditory stimuli.
  • Multi-camera recording to track animal’s path and capture detailed decision-making behaviors.
  • Precise control over the animal’s behavior sequence through the implementation of four motorized slides that can be extended into or retracted from the behavior cage. Two of the slides hold levers with limit switches that can be pressed by the animal, whereas the other two slides hold tubes for dispensing food or water rewards.
  • Food and water are released using solenoid pinch valves. Water is supplied through gravity from a syringe reservoir, while food mixture is dispensed from a syringe operated by air pressure to move the plunger. 
  • Meticulous control of food and water disbursement at 6ul to ensure animals consistently engaged in all the trials without reaching a state of satiety.
  • The entire apparatus is orchestrated by a customized field programmable gate array (FPGA) circuitry that coordinates slide motor control, logging and management of video and sensor data, pulse sequence for laser, video display, and dispense control.
  • A control PC runs software to set experimental parameters.
  • The time stamps (transistor-transistor logic pulses) of trial start/end, photo stimulation and water/food delivery were recorded and integrated with electrophysiology, allowing events synchronization with spike timing.



Apparatus dimensions: 20 × 9 × 11 in (51 × 23 × 28 cm) (L × W × H)

Cage dimensions: 8 × 8 × 8.5 in (20 × 20 × 22 cm) (L × W × H)

Cage material: Cast acrylic

Cage color/wall pattern: Customizable


Cage material: Cast acrylic

All apparatus dimensions, cage dimensions, cage color, and wall patterns are customizable.


  • Operant Conditioning.
  • Decision Making tasks.
  • Free consumption choice task with lick-triggered food or water delivery (Eiselt et al., 2021).

A lick-triggered consumption task was used to assess preference for selecting between water and food. Mice were provided access to the water and food spouts for 15 seconds. Every tenth lick at either spout led to the automatic dispensing of a food or water reward. Following the 15-second period, both spouts were retracted, and this was followed by an inter-trial interval lasting 4 to 30 seconds. Each session was comprised of 50 trials. 

The onset of each trial was indicated by a 1-second tone at 12 kHz. Following a 1-second delay, both levers were extended and accessible for the animal during a 5-second response window. Pressing one lever resulted in the delivery of a food reward (6 μl), while the other lever press led to the dispensing of water (6 μl). The lever–reward associations were randomized across animals by the software. Once a lever press was detected, both levers retracted, and the corresponding reward spout was extended. Animals had a 5-second consumption period with access to the reward spout before it was withdrawn. This was followed by a variable inter-trial interval ranging from 4 to 30 seconds. If no lever press occurred within the response window, the levers were retracted, the inter-trial interval began, and the absence of a behavioral response was recorded as a "miss." Animals completed 100 trials during each session. A session was deemed successful if the animal pressed at least 68% of the time for water during periods of thirst or for food when experiencing hunger. Data collection and analysis were automated.