Cues must be strictly controlled in behavioral neurophysiology experiments.  Crucial information is available in the onset latency and duration of neural firing patterns, so it is crucial that the experimenter be able to precisely control cue onset and offset.  Moreover firing selectivity for cues would be compromised if cues were allowed to mix together.  These requirements are easily met with visual, auditory or tactile cues.  Onset, offset and separation of different items is simplified because the light, sound and touch are essentially instantaneous compared to the speed of neural transmission.  Odors on the other hand present a challenge to the experimenter.  Unlike cues in these other modalities, odors must travel via diffusion through the air and then into the nasal mucosa before being detected by the olfactory chemoreceptors.  This diffusion requires time and makes control of onset and offset difficult, increasing the likelihood that one cue could mix with another.

To prevent or minimize these possibilities, it is important to exercise rigorous control of the odor cues.  This control is accomplished via a system of air and vacuum lines, solenoids and flowmeters.  Central to this system is a removable odor cartridge that isolates the odor cues from each other (Figure 1).

---

Figure 1

Removable 16-odor cartridge.  The solenoid valves allow independent delivery of any one of the 16 different odors.  Quick release connections at manifolds upstream and downstream from the odor dilutions chambers allow different cartridges to be used in the behavioral setup, so that many different odors can be used in a learning experiment.  Odors are isolated into separate lines that are connected only at the manifolds.  Backflow is prevented by the solenoids and also by checkvalves located at both manifolds.

---

This cartridge is machined from 2 pieces of  ½” polycarbonate.  The bottom piece contains depressions to hold each of 16 different odor dilution chambers made from glass sample tubes (Fischer Scientific, 0333922M).  Each chamber contains a unique odor (International Flavors and Fragrances) diluted in propylene glycol.  The top piece is machined with 16 circular depressions.  Each circular cutout fits an o-ring, which seals the opening of the glass tube (Figure 2) when the two polycarbonate pieces are bolted together.  Two holes are drilled in the polycarbonate inside each circular cutout (Figure 2).  These openings are tapped with 10-32 threads to accept barbed fittings (Minuteman Controls, F312085).  Air entry to each chamber is controlled by a solenoid valve (McMaster-Carr, 4916K13, 4916K81) attached to the polycarbonate top next to the openings to the odor dilution chamber (Figure 2).  This valve is activated via computer-controlled relays (Newark, 81F9285, 75F004).  The entire cartridge, containing 16 different odors, may be rapidly attached to the behavioral system via quick release connections/manifolds (Minuteman Controls, 2BH-19 and Small Parts, 01003312) located at either end of the lines (Figure 1).  Odor mixing at these manifolds is prevented by the presence of check valves (Minuteman Controls, F2804403B85).  Connections are made using 1/8” ID polyurethane tubing (Minuteman Controls, Special Order Item), which is color-coded for ease of use.

---

Figure 2

Close-up view of odor-dilution chamber.  This view shows how each odor is isolated.  Polyurethane tubing brings air from the manifold to the solenoid valve.  Opening of the valve allows a one-way flow of air through barbed fittings in the polycarbonate top piece into the odor chamber, displacing odor-saturated air out the second barbed fitting and on to the downstream manifold.  Note the circular depression fitted with an o-ring sealing the top of the glass odor chamber to the polycarbonate.

---

Note that all lines and equipment, such as the control solenoids, checkvalves, and the downstream manifold, are dedicated to a particular odor on the cartridge.  Additional components, such as the flowmeters or upstream lines, that control the odors but do not come in contact with the odorized air flow are located off the cartridge.  The only component not on the cartridge that comes into contact with multiple odors is the final delivery line into the odor delivery chamber on the behavioral box.  This line is common to all odors and may be easily changed as necessary.

---

Suppliers

International Flavors and Fragrances, 521 West 57th Street, New York, NY  10019, 212-765-5500
Fischer Scientific, 585 Alpha Drive, Pittsburgh, PA  15238, 800-766-7000
McMaster-Carr Supply Company, 6100 Fulton Industrial Boulevard, Atlanta, GA  30336, 404-346-7000
Minuteman Controls, 7 Foster Street, PO Box 1559, Wakefield, MA  01880, 800-343-8633

Web Content from:  Schoenbaum, G.  Olfactory Learning and the Neurophysiological Study of Rat Prefrontal Function.  In: CRC Series: Methods and Frontiers in Neuroscience.  Edited by S.A. Simon and M.A.L. Nicolelis, CRC Press, NY, 2000.