Two Coding Schemes in Gustatory Cortex
Max L. Fletcher, M. Cameron Ogg, Lianyi Lu, Robert J. Ogg, and John D. Boughter, Jr.
(see pages 7595–7605)
A longstanding controversy in neuroscience regards how tastes are represented in the nervous system. Some researchers have argued for a labeled-line scheme, in which taste-receptor cells and all downstream neurons are dedicated to representing a specific taste category (sweet, salty, etc.). Others, however, have argued that taste is represented by a combinatorial code, in which most cells respond to multiple taste categories, and the representation of taste depends on the activation of specific subsets of these cells. As with most scientific controversies, the truth is likely to be between these two extremes. Indeed, most studies have found that throughout the gustatory pathway, some neurons are narrowly tuned, responding best to a single taste category, while other neurons are more broadly tuned, responding to multiple categories. A notable exception is a study in which calcium imaging in the gustatory insular cortex revealed discrete clusters (“hot spots”) of neurons that responded to single taste categories, but found no evidence of neurons responsive to multiple tastes (Chen et al. 2011 Science 333: 1262). This study provided strong support for the labeled-line hypothesis of gustatory coding.
Fletcher et al. now bring new life to the combinatorial code hypothesis. They used a recently developed calcium sensor (GCaMP6s), which is more sensitive than that used by Chen et al., to record responses of neurons in mouse gustatory cortex, including a broad region between hot spots where Chen et al. found no taste-responsive neurons. Consistent with the previous work, Fletcher et al. found many neurons that responded selectively to sweet, salty, and bitter tastes. Unlike the previous study, however, they also found neurons that responded to sour taste, and they found no evidence of spatial clustering of neurons having similar response properties. Most importantly, they found that ∼45% of recorded neurons responded to multiple tastes.
These results support the hypothesis that the CNS uses both labeled-line and combinatorial codes to represent taste qualities. To bolster this conclusion, responses to a broader range of tastants—different bitter- or sweet-tasting compounds, for example—should be examined. Future work should also investigate the extent to which the two coding schemes interact and whether they serve the same or different functions.
Texture Discrimination via Passive Whisker Stimulation
Pauline Kerekes, Aurélie Daret, Daniel E. Shulz, and Valérie Ego-Stengel
(see pages 7567–7579)
Rodents use their whiskers for many things, including navigation, judging distances, and discerning object shape and texture. Rats continually explore their environment by rhythmically sweeping their whiskers (whisking). Thus, whisking is a type of active sensing, analogous to humans running their fingers over an object. Information obtained by whisking is thought to be processed in the primary somatosensory (barrel) cortex.
Although active whisking is unnecessary for navigation in rats, whether it is needed for fine texture discrimination has been unclear. Moreover, the necessity of barrel cortex in passive whisker-dependent sensation has been debated. To address these questions, Kerekes et al. trained rats to run through a maze in which the direction of reward was indicated by the presence or pattern of vertical bars on the corridor walls. Because rats were motivated to run quickly to obtain the reward, their whiskers contacted the tactile cue for only a brief period—less time than is needed to complete a whisking cycle.
Learning this task proved difficult for rats. The authors initially had to repeat the cue at multiple locations along the passage and then remove the cues one by one before rats could reliably use a single cue to determine the correct turn. Nonetheless, rats eventually learned to discriminate the presence of bars versus a smooth surface and subsequently, to discriminate regularly and irregularly spaced bars that were presented simultaneously on opposite sides of the snout. Performance of the task was impaired by trimming the whiskers or by inhibiting barrel cortex, indicating that both structures were involved in the task. Remarkably, however, rats eventually learned to navigate the maze even without whisker-derived cues, despite the researchers' attempts to remove any olfactory, visual, or auditory cues.
These results demonstrate that rats can use whiskers to discriminate textures even without actively whisking, and that this ability depends on barrel cortex activity. The authors suggest that the rats can also discriminate tactile cues with the skin of their snout and trunk when whiskers are removed. Future work can build on these results to investigate neural mechanisms underlying passive versus active tactile sensing.
Footnotes
This Week in The Journal was written by Teresa Esch, Ph.D.