The face of a rat is peppered with stiff, conical hairs called vibrissae, more commonly known as whiskers. As nocturnal animals that live in dingy, cramped hovels, rats rely on tactile information from their whiskers to get through the night. Using just their whiskers, they can run through complicated mazes and recognize complex objects – tasks that humans typically rely on vision to accomplish.

These feats of tactile perception begin with mechanosensory neurons, whose dendrites reside at the base of each whisker. A longstanding goal has been to understand what information these mechanosensory neurons are extracting from the whiskers. Specifically, do mechanosensory neurons encode object position and whisker movement during touch, or do they simply encode the mechanical forces acting at the base of the whisker?

To distinguish between these possibilities, Nicholas Bush, Christopher Schroeder and colleagues from Mitra Hartmann's lab at Northwestern University, USA, used single-unit extracellular recordings to measure the spiking activity of mechanosensory neurons that innervate whisker follicles. They investigated two types of whisker wiggling: active movements in awake, behaving rats and manual movements, in which they deflected the whiskers of anesthetized rats. By tracking the position of the whiskers with video, the authors were able to precisely measure kinematic variables of whisker movement, such as the angular position, velocity and distance to an object. They also used a biomechanical model of whisker bending to estimate the mechanical forces and moments at the base of the whisker.

The authors then examined how the spike trains of mechanosensory neurons encode tactile stimuli. They employed a framework called a generalized linear model that allowed them to estimate which input variables were best at predicting the firing rate of each mechanosensory neuron. Their models revealed that different neurons encode different combinations of kinematics and mechanics. Overall, however, the activity of most mechanosensory neurons was better predicted by mechanical rather than kinematic variables.

This finding alters our view of tactile processing in the whisker system. Previous studies have shown that rodents can use their whiskers to precisely localize the position of objects and that object position can be decoded from neurons in the somatosensory cortex. Thus, it was thought that sensory neurons might directly encode object position. However, these results indicate that additional processing may be needed to transform the code for mechanical force into a code for position and movement.

Where might this transformation take place? Each whisker is innervated by hundreds of mechanosensory neurons that fall into at least eight different classes, based on their anatomy and physiology. These sensory neurons project into the trigeminal nucleus of the brainstem. From here, central neurons route whisker information to multiple different brain regions, including the somatosensory cortex by way of the thalamus. Perhaps as early as the trigeminal nucleus, integration of diverse whisker signals could underlie a transformation from a mechanical to a kinematic code. Cracking this problem will require a detailed understanding of different mechanosensory neuron classes and the logic of their integration within circuits of the central nervous system.

Another possibility is that the brain does not need to transform the neural code from mechanics to kinematics, because these two codes are tightly correlated in a rat's normal life. In this paper, the authors compared manual and active whisking in order to sample a broad range of whisker movements. In awake, behaving rats, whisker movements are far more constrained. Perhaps the limited range of natural whisking could allow downstream circuits to decode object location directly from a code based on mechanics. Testing this idea might require teaching rats to whisk outside of their comfort zone, a noble but unenviable task for which volunteers may not be forthcoming.

Bush
,
N. E.
,
Schroeder
,
C. L.
,
Hobbs
,
J. A.
,
Yang
,
A. E.
,
Huet
,
L. A.
,
Solla
,
S. A.
and
Hartmann
,
M. J.
(
2016
).
Decoupling kinematics and mechanics reveals coding properties of trigeminal ganglion neurons in the rat vibrissal system
.
Elife
e13969
.