For our olfactometer design, the main goal was to expand the existing design for three odors.

Our first draft (Fig. 1.1) was based on last year's design (available on their Tiki Wiki website) but with an additional odor channel including the valves necessary to control the airflow. We decided to move away from the non-return valves proposed by the previous group, and instead included valves at the start of each pipetrack so that when one odor is chosen, no air flows into the odor vials for the other odors (instead of air flowing into those vials but having no exit). We then decided to extend our design to include potentially 6 odors, should future expansion be required. While we initially designed and planned to 3D print the box, we experienced difficulties with the functionality of the 3D printers during this research. As a result, we replicated the 3D design with a sturdy plastic sheet which we cut and superglued together to manufacture our final olfactometer design.

Designing the head-fixing apparatus was based on extensive research which compared many commercially available setups. Since we are aware of the responsibility we hold by designing a system with which the mouse will be head-fixed followed by a surgical procedure, we wanted to make sure that we choose the most harmless and convenient way of fixing the mouse possible.

In the design stage, we tried to mimic the designs of commercially available devices such as the Neurotar head-fixing system (Fig 2.1). This setup included crocodile clamps to fix the head-bar and screws to hold it in place. The next draft was based on a design by Komiyama et al. (2010). In this design, the mouse would be placed in a tube. We decided not to choose this design because of the effort of putting the mouse in the tube. The next draft (Fig. 2.3) was based on a paper by Slocum, A. H. (1992). In their paper, the authors describe the benefits of including a three-grooved kinematic coupling. We elaborate on this topic in our final head-fixing design section. In our draft however, we mimicked the design described in the paper published by Kim, S. J., Slocum, A. H., & Scott, B. B. (2022) with a triangular shaped head-bar. This opened new challenges in fixing the head bar, which is why we switched to a rectangular shaped head-bar design. Lastly, we found that in the Neurotar magnetic clamps included a magnetic ring to decrease the effort of guiding the mouse in the correct position to be fixed in their . We found it challenging to incorporate a ring shaped magnet, which is why we chose round magnets in the end.

Designing the lick tracking unit was a particular challenge as existing technology was too expensive. One of the main challenges was to spatially include the water spout into the setup, so that the mouse can reach it whilst being head-fixed with the odor port over the mouth and nose. This encouraged us to redesign the odor port (Fig. 1.2). We chose to build a wide cone to not disturb the mouse's whiskers and would still allow for the mouse to access the water spout.

The next challenge was to design a setup to convert the analogue licking signal into a digital one which could be tracked during experimentation. Before our final design, we designed a setup with a metal sock at the end of a water tube (Fig. 1.3). This sock was intended to conduct the electric signal (essentially "closing" the circuit with the tongue) as the mouse stands on a tin-foil base. However, with our final design, we do not need this additional conduction, and so the metal sock was removed.

Another challenge was to regulate the water flow depending on the odor released. Our initial idea was to curve a tube and therefore restrict the water from flowing out when the water valve is closed. However, opted to use a water valve which controls water flow, and so the tube curving was no longer necessary.