Let'Swind !

What are the limits of sensitive perception of both humans and wind sensors  ?

We are the Let’Swind team from the FdV bachelor program, hosted by the CRI (Center for Research and Interdisciplinary). During this second week of Biosensors, an interdisciplinary seminar, we had to develop and design a scientific research project connected with the notion of force ! But then you may ask : which forces do you mean ? And this would be the one million dollars question !

Various kinds of forces are present in our environment : vibration, gravity, osmotic pressure and so on. It’s well-known that diverse electronic sensors are commonly used to measure those forces but we often forget the huge influence they have on biological life. Tactile perception is an example of an exquisitely important sensitive aptitude essential to many organisms (for instance, it allows birds to optimize their flight and accomplish their seasonal migration).

In this way we decided to devote our one week research project to the perception of wind in humans. We know humans to be really complex organisms with multiple senses consistently interacting with each other. It is recognized that blind people’s other functional senses are more developed. According to this fact, and in the light of an extensive scientific literature about senses interactions (like Sound enhances touch perception wrote by Tony Ro and published in 2009), we decided to measure the ability of both humans and wind sensors to detect the origin of a source of wind (the “from where it comes from”) when this perception is disturbed by a sound (which consists, as the wind, of a type of vibration).

The primary goal of the Biosensors is for us to compare a biological sensor to an electronic one. Human biological sensors were tested among FDV students. Indeed, their huge network of nerve-endings and touch receptors in the skin make humans relevant study subjects. In the other hand, we tested a wind sensor, a thermal anemometer extremely sensitive since it was primarily designed for human breath detection.

Figure 1: Experimental set-up

Our experimental set-up was designed for the following purpose: comparing the sensors precision with an increasing angle (θ) between an almost noiseless fan (our source of airstream)  and a speaker diffusing the sound of a fan.

To do so, we conceived a huge protractor that we fixed on the floor and in the center of which we disposed our sensors. During the experiment we modified the angle between the fan and the speaker (see figure 1). We tested several angles : 0°, 30°, 60° and 180°.

To encourage our “human guinea pigs” to focus on the wind source, we blindfolded their eyes and covered their ears by means of an headset (playing no sound).

As we can see on the video, the biological sensors were disposed in the center of the protractor and were told to show, with their arm, the direction where they felt the wind with the highest intensity.

We could demonstrate that, in contrast to humans who are multisensory organisms, thermal wind sensors can only measure the wind’s intensity : a variation of the sound can not disturb the values measured by the sensor. Despite of this, we still tried to observe if a correlation could be made between the wind’s intensity and its direction as it hits the sensor : we tested different angles between the fan and the orientation of the wind sensor (every 10° from 0 to 180°) and we observed that wind sensor was more sensitive to crosswind than headwind.

Figure 2

This gauges (see figure 2) represent how far people were from the real fan position  for four speaker positions (where 0 would be a perfect answer). As you can see, when the fan and the speaker are in the same position (first gauge), humans are quite talented to identify the direction the wind is coming from: they give a quite accurate answer (12° wrong) in around 15 seconds. But then, as the speaker is moved away from the fan, they need an increasing answer time, sometimes reaching 30 seconds !  It highlights people’s confusion due to the contradictory informations their senses are giving them: they feel wind from one direction but hear the fan from another …

They had to deal with this contradictory perceptual informations to give an answer. In this way, our results suggest that our guinea pigs’s sense of touch is effectively disturbed by sound, as people still use and rely on their sense of hearing.  

By the end of the week, we had the great opportunity to meet Daniel Pressnitzer, researcher at Perceptual Systems Laboratory, who helped us to identify the usual bias affecting experiments with humans. For example, even if we gave  the very same instructions to every tested students, we know for a fact that each person would have developed its own strategy to find the origin of the airstream, and some people would focus on what they hear to find the position of the fan despite of our instruction. Moreover, in all cognitive experiment, a procedural learning phase usually occurs before the start of an experiment. We did not have time to include it in our experimental protocol. Finally, that kind of study must be done on a larger number of people and repeated lots of time for each to engender relevant conclusions, from 30 to 100 for each condition according to the researcher!

Xander, Yulian, Cécile & Adèle