Main Goals

With the long-term goal of understanding neural mechanisms that underlie adaptive perceptual processing, our research focuses on the tactile system of the rat. Rats, like some other rodents, possess a specific system for active touch that uses the long facial whiskers (vibrissae) to gather information about the immediate environment. The main effort of our laboratory is aimed at deciphering the neuronal mechanisms that underlie vibrissal touch. Additional efforts in our laboratory are dedicated to studying active touch and active vision in humans. The latter are guided by detailed neuronal knowledge accumulated in the rat, with the eventual goal of developing efficient tactile substitutions for the blind.

Last Update: January 2007

Previous Work Processing of Passive Touch We have previously found that sensory information in the vibrissal system is not processed the same way along the two main afferent pathways of the vibrissal system (the lemniscal and paralemniscal). Whereas spatial information is processed primarily along the lemniscal pathway, temporal information is processed primarily along the paralemniscal pathway. (Ahissar et al., 2000; Ahissar et al., 2001; Sosnik et al., 2001). These results were obtained by recording neuronal activity in anesthetized rats while applying mechanical stimuli to passive whiskers. Natural active touch in awake rats involves motor-driven sensory acquisition and rats actively move their whiskers to sample the world near their snout. To study active whisking with controlled stimulus application, we developed a technique of artificial whisking in anesthetized rats.
Encoding of Active Touch Using this technique, we found that the signals sent to the brain by the whiskers during active touch differ from those transmitted during passive touch. Specifically, we found that active touch is reported to the brain by different classes of signals that either describe the movement of the whiskers, contact, or combinations of whisking and contact. (Szwed et al., 2003).
Parallel Pathways More recently, we discovered that these three classes of signals are carried to the thalamus by three separate afferent pathways: the paralemniscal, a newly discovered "extralemniscal" pathway (discovered by the group of Martin Deschênes), and the lemniscal, respectively. These three classes of signals begin to integrate in the barrels (SI) cortex, although a segregation between whisking and contact signals is still evident between layers 5a and 4 (barrels), respectively. (Derdikman et al., 2006; Yu et al., 2006).
Encoding of Location Using the same technique we studied the neural codes used by vibrissal receptors to encode the coordinates of object location in three dimensions. We found that the most efficient neural code for each of the three spatial dimensions is different: temporal for the horizontal axis (along whisker rows), spatial for the vertical (along whisker arcs) and rate for the radial axis (from the face out). (Szwed et al., 2003; Szwed et al., 2006).
Behavior We examined the ability of rats to resolve horizontal object location and its dependence on whisker configuration and whisking parameters. We found that rats could discriminate offsets in horizontal (posterior-anterior) location as small as 0.24 mm (~1 deg). Rats could learn the task with all the whisker configurations tested (all, one row, or one arc intact on each cheek), except that of a single intact whisker (C2) on each cheek. However, rats initially trained with multiple whiskers typically improved their performance when re-tested later with a single whisker intact on each cheek. In general, lower thresholds were obtained with fewer intact whiskers, and the lowest thresholds were significantly less than the typical inter-vibrissal spacing. Rats typically whisked when performing this task, and performance dropped to chance level when whisking was prevented by cutting the facial motor nerve. Furthermore, performance levels for all whisker-array configurations tested, except that of a single pair of whiskers from onset of training, correlated positively with the net whisking spectral power at 5-25 Hz. From these experiments, we conclude that object localization in the rat vibrissal system is an active process: whisker movements are both required and beneficiary, in a graded fashion, for making accurate positional judgments. (Knutsen et al., 2005; Knutsen et al., 2006).
Current Work We are now addressing the potential involvement of these pathways in three nested motor-sensory loops, each of which controls whisker movement according to specific information: movement kinematics, contact, and object identity, respectively. We are also addressing central mechanisms underlying the decoding and processing of active touch, focusing on object localization. We record in the thalamus and cortex of anesthetized rats while inducing artificial whisking, and examine the transformations occurring along parallel sensory pathways. We interpret the results in light of working models inspired by the peripheral encoding of object location. Also, we study the mechanical transformations underlying the peripheral encoding we observed, and the neuronal mechanisms responsible for the decoding of the acquired tactile information. In nature, active touch is under behavioral control and we have established an electrophysiological setup for behaving rats. Thus, we can test the effects of behavior on motor and sensory aspects of active touch, via behavioral control of neuronal processing along the various motor-sensory-motor loops. Current work focuses on whisking behavior, task-dependent behavioral control of whisking kinematics, and touch-related control of whisking. Future experiments will focus on recording of individual neurons from various stations along the parallel motor-sensory loops. In parallel to our quest for neuronal mechanisms underlying active touch in rats, we are studying active sensing of vision and touch in humans with the goal of developing efficient tactile substitutions for the blind. We are using non-invasive methods to track the operation of motor-sensory loops during active sensation using high resolution eye and finger tracking and electroencephalogram (EEG) recordings.     Future Plans Our general plan is to obtain a detailed understanding of active touch in an experimental model (rats in this instance), construct a comprehensive model for active touch in humans, and understand principles of active vision in humans. The anticipated findings, beyond their contribution to understanding the brain, should facilitate development of efficient tactile substitutions for the blind, optimization of sensory acquisition in robots, and sensory throughput in normal subjects. Understanding active sensation requires understanding operation of the neuronal micro-circuitry during behavior. Currently, studies at the micro-circuitry level in behaving animals cannot be performed with the accuracy required for elucidating their mechanisms. In behaving animals, motor-sensory-motor loops, and other top-down processes, control sensation by continuously adjusting motor variables. The result is an ever changing process that does not allow step-by-step systematic investigation of neuronal mechanisms. Moreover, with chronic recordings in behaving animals, recording sites cannot be verified for each recorded cell.  In contrast, most of our findings with anesthetized rats were only possible because we could precisely determine the recording site for each cell. Thus, we will continue our strategy of studying anesthetized and behaving rats in parallel for addressing each specific motor-sensory question. In general, the neuronal circuitry of temporal and spatial processing of vibrissal information will be studied with artificial whisking in anesthetized rats. The behavioral relevance of these neuronal processes, their operation in natural-like conditions, and their role in determining strategies of active sensing, will be studied during the performance and learning of perceptual tasks. Our working models and their predictions will be continuously updated. When a thorough understanding of the mechanisms underlying active sensing will be approached, their plasticity will be addressed at the neuronal, algorithmic, and behavioral levels. Across these levels, we will link mechanisms of plasticity related to individual neurons, motor-sensory loops, strategies of active sensing, task performance, and sensory substitutions. Specific future projects include: object localization via closed loop operation in the vibrissal system - parallel closed loop operation (algorithms, interactions, competition, hierarchy) - motor strategies of active touch - sensory (neuronal, morphological and mechanical) processing of object location - thalamocortical computations neuronal plasticity and behavior - functional links between neuronal plasticity and motor behavior - functional links between neuronal plasticity and perceptual changes theoretical, engineering-based, models of the vibrissal system studies of human touch and vision (experiments, theory, and sensory substitution).

References 
Ahissar E, Sosnik R, Haidarliu S (2000) Transformation from temporal to rate coding in a somatosensory thalamocortical pathway. Nature 406:302-306.

Ahissar E, Sosnik R, Bagdasarian K, Haidarliu S (2001) Temporal frequency of whisker movement. II. Laminar organization of cortical representations. J Neurophysiol 86:354-367.

Derdikman D, Yu C, Haidarliu S, Bagdasarian K, Arieli A, Ahissar E (2006) Layer-specific touch-dependent facilitation and depression in the somatosensory cortex during active whisking. J Neurosci 26:9538-9547.

Knutsen PM, Derdikman D, Ahissar E (2005) Tracking whisker and head movements in unrestrained behaving rodents. J Neurophysiol 93:2294-2301.

Knutsen PM, Pietr M, Ahissar E (2006) Haptic object localization in the vibrissal system: behavior and performance. J Neurosci 26:8451-8464.

Sosnik R, Haidarliu S, Ahissar E (2001) Temporal frequency of whisker movement. I. Representations in brain stem and thalamus. J Neurophysiol 86:339-353.

Szwed M, Bagdasarian K, Ahissar E (2003) Encoding of vibrissal active touch. Neuron 40:621-630.

Szwed M, Bagdasarian K, Blumenfeld B, Barak O, Derdikman D, Ahissar E (2006) Responses of trigeminal ganglion neurons to the radial distance of contact during active vibrissal touch. J Neurophysiol 95:791-802.

Yu C, Derdikman D, Haidarliu S, Ahissar E (2006) Parallel Thalamic Pathways for Whisking and Touch Signals in the Rat. PLoS Biol 4:e124.

Last Update: January 2007
top