Sensory deprivation during early development causes an increased exploratory behavior in a whisker-dependent decision task

Abstract

Stimulation of sensory pathways is important for the normal development of cortical sensory areas, and impairments in the normal development can have long-lasting effect on animal's behavior. In particular, disturbances that occur early in development can cause permanent changes in brain structure and function. The behavioral effect of early sensory deprivation was studied in the mouse whisker system using a protocol to induce a 1-week sensory deprivation immediately after birth. Only two rows of whiskers were spared (C and D rows), and the rest were deprived, to create a situation where an unbalanced sensory input, rather than a complete loss of input, causes a reorganization of the sensory map. Sensory deprivation increased the barrel size ratio of the spared CD rows compared with the deprived AB rows; thus, the map reorganization is likely due, at least in part, to a rewiring of thalamocortical projections. The behavioral effect of such a map reorganization was investigated in the gap-crossing task, where the animals used a whisker that was spared during the sensory deprivation. Animals that had been sensory deprived performed equally well with the control animals in the gap-crossing task, but were more active in exploring the gap area and consequently made more approaches to the gap - approaches that on average were of shorter duration. A restricted sensory deprivation of only some whiskers, although it does not seem to affect the overall performance of the animals, does have an effect on their behavioral strategy on executing the gap-crossing task.

Keywords: Barrel cortex; development; gap-cross; sensory deprivation; whisker tracking.

Figure 1

Sensory deprivation causes structural changes in the barrel size. (A) In the sensory deprivation protocol used, the C- and D-row whiskers were spared during different periods of development. (B) Barrels at the level of layer 4 were stained with cytochrome oxidase. The barrels A1–A4, B1–B4, C1–C4, and D1–D4 were traced to calculate barrel area. The dotted circles show schematically the tracing of A1, B1, C1, and D1. Scale bar 1 mm. (C) The ratio of the total adjusted area of the spared (C+D rows)/deprived (A+B rows) in the left hemisphere in control and P0 animals. The ratio was larger for the P0 group compared with control (*P < 0.05).

Figure 2

Sensory deprivation did not affect the performance in the gap-crossing task. (A) In the gap-crossing task, the animal is placed on a platform (home platform) and uses its whiskers to judge the distance to the other platform (target platform). Each platform is surrounded by high walls so the only exit is toward the gap separating the platforms. Motions sensors (MS) are used to track the position of the animal on the platform and used to measure how often the animal approaches the gap and how long time it spends exploring the gap. (Not drawn to scale.) (B) The average maximum distance achieved during the 7-day test period was similar in the control and P0 group. On each day only animals that made at least one crossing were included.

Figure 3

Sensory exploration strategy is affected by sensory deprivation. (A) P0 animals made more attempts to jump over the gap in comparison with control animals. The differences are significant at a gap distance of 5.5 cm, which is the distance where the animals can only rely on their whiskers to contact the target platform. (B) The duration the animal spends exploring the gap is shorter in P0 animals compared with control animals. *P < 0.05. Error bars show mean ± SEM.

Figure 4

The spatiotemporal profile during exploration of the gap. (A) The animals' spatiotemporal profile was calculated by tracking the nose position in the gap space. For each tracked frame, the x and y coordinates of the nose were extracted and the profiles were created from all tracked frames. The pseudocolor coding corresponds to the probability of having the nose positioned at the specific area for a given frame. In comparison with control, the spatiotemporal profile for the P0 animals is more homogenously distributed in the gap space. Color scale bar 0–1.5 × 10⁻⁴.(B) To evaluate the probability of the animal being in a given position (Probability of Presence) along the y-axis (which corresponds to distance from the target platform), the two-dimensional data were collapsed in one-dimensional (1D) by averaging along the x-axis (upper graph). The cumulative distribution function (F[x]) of the 1D data is shown on the lower graph. The more homogenous spatiotemporal profile for the P0 animals compared with control is evident in the cumulative distribution plot as there is less curvature for the P0 group (a random distribution gives a straight line). A Kolmogorov–Smirnov test showed significant difference between the P0 and control distributions (P = 0.0475).

Figure 5

Whisker kinematics as a function of distance from the target platform. (A) For control animals, the whisking amplitude decreases when the animal is within reach of the target platform with its whiskers (at ∼13 mm). (B) In the P0 animals, similarly to that observed for the control animals, the whisking amplitude decreases when the animals touch the target platform, but in addition, the P0 animals are also more actively exploring other areas of the gap; thus, there are more low-amplitude whisks (see text for details) when the animal is exploring the surrounding close to the home platform.