Thursday, November 11, 2010

DIY Psychophysics


I used to love building models when I was a kid: fighter planes, space capsules, cars of all sorts. My better projects, like the Saturn launch vehicle with a collapsible gantry, stayed on the shelf in my bedroom. The rest—like the crappy Avanti Sports Coupe and a Mustang fastback—were blown up with firecrackers in carefully orchestrated backyard demolitions (timer fuse is a great thing). Model cement used to give me nosebleeds, but that was just an annoyance. 

Later, in junior high, I made large balsa wood gliders from kits. They rarely flew more than a few seconds before crashing, but I loved the building process—hours of cutting out pieces, assembling wing sections, and doping the rice paper that covered the whole plane. The sweet smell of that amber doping compound comes back to me as I think about it. Is that stuff even sold any more?

For some reason, my juvenile hobbies didn’t translate into grownup scientific skills: I was never a “hands” guy in the lab. I envied pals who built their own neurophysiology rigs and customized signal processing electronics. Instead of computer-controlled, constant airflow olfactometers tweaked out with mass flow controllers and pressure-compensated scent injection lines, I made do with squeeze bottles and sniff jars. In the corporate world I watched as big-time R&D money was poured into elaborate olfactometers of dubious practical use. I soured on the Big Metal approach and became an advocate of low tech psychophysics.

Still, there are some applications that demand precisely timed delivery of odor at controlled concentrations—fMRI brain imaging research on smell, for example, or experiments that match smells to sounds and visual images. A big barrier to entry is the pain-in-the-ass factor of designing and building a scent delivery device or olfactometer. This has kept a lot of otherwise smart and creative people from getting involved in smell research.

Well, that may change thanks to the altruistic work of Johan Lundström and his colleagues at the Monell Chemical Senses Center in Philadelphia. They’ve published a design for a practical, computer-controlled, general-purpose, and—most importantly—inexpensive olfactometer. It’s a dream come true for the scientific DIY crowd, complete with parts list, prices, and a how-to assemble guide with photos.

An olfactometer basically consists of an air compressor, tubing, valves, and odor reservoirs. It injects short bursts of odor into an airstream delivered to the individual nostrils via plastic nose pieces, or via a nose mask. The Monell device is compact—it fits in a 25-inch tall cabinet with a 20 x 20 inch footprint. The cabinet, which can be purchased, is the most expensive item on the parts list at $1,000. A custom three-way valve manifold also costs $1,000. The are twelve other parts, some required in multiples. Total cost of the unit: $5,284. Which is a bargain, I might add.

Are your DIY skills up to the task? Lundström et al. think so:
. . . we argue that an adult with enough technical knowledge to put up a shelf using both screws and plugs is capable of building the olfactometer described here.
They compare the technical challenge to assembling a piece of IKEA furniture. (Attention, academic nerd-balls, you’ve been called out!) Accordingly, I think Monell’s DIY olfactometer should be christened “The Lundström”. It’ll look great in your lab along with the Ikvar toiletries cabinet, Skrogval audio rack, and Smölstad futon.

Now get to work!

2 comments:

Olfacta said...

We put together a big desk from Office Depot once. We got the drawers backwards, and realized (much too late) that the desk was too big to get it out of the room.

So, no, no homemade olfactometers for us any time soon!

EdC said...

Sounds like a great idea. Unfortunately neither I nor my library have a subscrition to the Int J Psychophysiol., so I can't see more than the abstract. Do you know if the full text will be available?

Do you know how they solve the problem of controlling the vapor concentration? E.g., different liquids have different vapor pressures, evaporate at different rates, and the rates depend on air flow over the liquid surface, ...