Characterization of phase transition in the thalamocortical system during anesthesia-induced loss of consciousness

Abstract

The thalamocortical system plays a key role in the breakdown or emergence of consciousness, providing bottom-up information delivery from sensory afferents and integrating top-down intracortical and thalamocortical reciprocal signaling. A fundamental and so far unanswered question for cognitive neuroscience remains whether the thalamocortical switch for consciousness works in a discontinuous manner or not. To unveil the nature of thalamocortical system phase transition in conjunction with consciousness transition, ketamine/xylazine was administered unobtrusively to ten mice under a forced working test with motion tracker, and field potentials in the sensory and motor-related cortex and thalamic nuclei were concomitantly collected. Sensory and motor-related thalamocortical networks were found to behave continuously at anesthesia induction and emergence, as evidenced by a sigmoidal response function with respect to anesthetic concentration. Hyperpolarizing and depolarizing susceptibility diverged, and a non-discrete change of transitional probability occurred at transitional regimes, which are hallmarks of continuous phase transition. The hyperpolarization curve as a function of anesthetic concentration demonstrated a hysteresis loop, with a significantly higher anesthetic level for transition to the down state compared to transition to the up state. Together, our findings concerning the nature of phase transition in the thalamocortical system during consciousness transition further elucidate the underlying basis for the ambiguous borderlines between conscious and unconscious brains. Moreover, our novel analysis method can be applied to systematic and quantitative handling of subjective concepts in cognitive neuroscience.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Verification of electrodes location.

Electrode placements in ventral posteromedial (A) and ventral lateral (B) nuclei were histologically verified with light microscopy. Arrow point indicates the tip of electrode. fi, fimbria of the hippocampus; st, stria terminalis; ic, internal capsule; cp, cerebral peduncle (basal part); VPM, ventral posteromedial nucleus; VPL, central posterolateral nucleus; ZI, zona incerta; STh, subthalamic nucleus; Po, posterior thalamic nuclear group; VM, ventromedial thalamic nucleus; DG, dentate gyrus; CM, central medial nucleus; VL, ventral lateral nucleus; VP, ventral posteromedial and posterolateral nuclei; Rt, reticular thalamic nucleus; LV, lateral ventricle. Scale bar is 1 mm.

Figure 2. Dynamics of brain state in temporal domain.

A. Representative time series for an electroencephalogram (EEG) from primary motor cortex (M1), local field potentials from ventral lateral (VL) thalamus, and motion signal around the time of drug administration (ADM), loss of motion (LOM), deep anesthesia, recovery of motion (ROM), and resumption of walking. For the periods of anesthesia, recovery of motion and resumption of walking, the motion signal is exaggerated by five times for clearance. Dashed lines in the panels indicate corresponding events. B. Representative plot of δ band (1–4 Hz) phase synchronization for thalamocortical motor pair (M1 and VL) and sensory pair (primary somatosensory cortex and ventral posteromedial nucleus). Dashed lines indicate drug administration (tADM), loss of motion (tLOM), and recovery of motion (tROM), from left to right. C. Ensemble average (N = 10) of the order parameter (black solid) and its standard error mean (gray errorbars) as a function of rescaled time. The rescaled time is obtained by normalizing measurement time with respect to the blackout period. The moments of tLOM, and tROM correspond to 0 and 1, respectively. D. Ensemble-averaged absolute value of time derivative of order parameter (black solid) and its standard error mean (gray errorbars). The peak amplitude during anesthetic induction period was statistically significantly larger than the value at recovery (t-test, P<0.05). E. Probability of transition between up and down states at 10-s time intervals. Gray errorbars indicate standard error mean of the transition probability. For a given 10-s observational period, the number of epochs was approximately twenty during the transitional periods and fifteen during the deep anesthesia period.

Figure 3. Path-dependent change of brain state and metastable fluctuation during transition.

A. Anesthetic concentration estimated with the three-compartment is plotted with respect to rescaled time, where the moments of tLOM, and tROM correspond to 0 and 1, respectively. Black solid line indicates averaged anesthetic concentration among ten subjects and gray errorbars indicate standard error mean. B. Ensemble average (N = 10) of the order parameter (greenish gray solid) and sigmoidal curve fitting as a function of anesthetic concentration c (black solid). Gray errorbars indicate standard error mean of the order parameter. Arrowheads show the passage of time. C. Ensemble-averaged time courses of hyperpolarizing (solid) and depolarizing (dashed) susceptibilities. Gray errorbars indicate standard error mean of the susceptibilities. The distributions of concentration for maximal susceptibility are shown as the embedded boxplots at the corresponding peaks. D. Representative plot of the recurrence probability of the order parameter for each bin of anesthetic concentration (bin size = 0.1/0.05 for the induction and emergence periods, respectively). Traces of the order parameter at the transitional moments at approximately 0.8 (during induction of anesthesia) and 0.2 (during recovery) are provided as inset figures. The anesthetic concentration of 0.0 at the left side of the axis indicates the moment of drug administration (ADM).