TY - JOUR
T1 - Auditory predictive coding across awareness states under anesthesia
T2 - An intracranial electrophysiology study
AU - Nourski, Kirill V.
AU - Steinschneider, Mitchell
AU - Rhone, Ariane E.
AU - Kawasaki, Hiroto
AU - Howard, Matthew A.
AU - Banks, Matthew I.
N1 - Funding Information:
This work was supported by the National Institutes of Health Grants R01-DC04290, R01-GM109086, and UL1-RR024979, National Science Foundation Grant CRCNS-IIS-1515678, and the Hoover Fund. We thank Haiming Chen, Phillip Gander, Bradley Hindman, Christopher Kovach, Bryan Krause, Rashmi Mueller, Caitlin Murphy, Melissa Paul-
Publisher Copyright:
© 2018 the authors.
PY - 2018/9/26
Y1 - 2018/9/26
N2 - The systems-level mechanisms underlying loss of consciousness (LOC) under anesthesia remain unclear. General anesthetics suppress sensory responses within higher-order cortex and feedback connections, both critical elements of predictive coding hypotheses of conscious perception. Responses to auditory novelty may offer promise as biomarkers for consciousness. This study examined anesthesia-induced changes in auditory novelty responses over short (local deviant [LD]) and long (global deviant [GD]) time scales, envisioned to engage preattentive and conscious levels of processing, respectively. Electrocorticographic recordings were obtained in human neurosurgical patients (3 male, 3 female) from four hierarchical processing levels: core auditory cortex, non-core auditory cortex, auditory-related, and PFC. Stimuli were vowel patterns incorporating deviants within and across stimuli (LD and GD). Subjects were presented with stimuli while awake, and during sedation (responsive) and following LOC (unresponsive) under propofol anesthesia. LD and GD effects were assayed as the averaged evoked potential and high gamma (70–150 Hz) activity. In the awake state, LD and GD effects were present in all recorded regions, with averaged evoked potential effects more broadly distributed than high gamma activity. Under sedation, LD effects were preserved in all regions, except PFC. LOC was accompanied by loss of LD effects outside of auditory cortex. By contrast, GD effects were markedly suppressed under sedation in all regions and were absent following LOC. Thus, although the presence of GD effects is indicative of being awake, its absence is not indicative of LOC. Loss of LD effects in higher-order cortical areas may constitute an alternative biomarker of LOC.
AB - The systems-level mechanisms underlying loss of consciousness (LOC) under anesthesia remain unclear. General anesthetics suppress sensory responses within higher-order cortex and feedback connections, both critical elements of predictive coding hypotheses of conscious perception. Responses to auditory novelty may offer promise as biomarkers for consciousness. This study examined anesthesia-induced changes in auditory novelty responses over short (local deviant [LD]) and long (global deviant [GD]) time scales, envisioned to engage preattentive and conscious levels of processing, respectively. Electrocorticographic recordings were obtained in human neurosurgical patients (3 male, 3 female) from four hierarchical processing levels: core auditory cortex, non-core auditory cortex, auditory-related, and PFC. Stimuli were vowel patterns incorporating deviants within and across stimuli (LD and GD). Subjects were presented with stimuli while awake, and during sedation (responsive) and following LOC (unresponsive) under propofol anesthesia. LD and GD effects were assayed as the averaged evoked potential and high gamma (70–150 Hz) activity. In the awake state, LD and GD effects were present in all recorded regions, with averaged evoked potential effects more broadly distributed than high gamma activity. Under sedation, LD effects were preserved in all regions, except PFC. LOC was accompanied by loss of LD effects outside of auditory cortex. By contrast, GD effects were markedly suppressed under sedation in all regions and were absent following LOC. Thus, although the presence of GD effects is indicative of being awake, its absence is not indicative of LOC. Loss of LD effects in higher-order cortical areas may constitute an alternative biomarker of LOC.
KW - Averaged evoked potential
KW - Consciousness
KW - Electrocorticography
KW - High gamma
KW - Human auditory cortex
KW - Propofol
UR - http://www.scopus.com/inward/record.url?scp=85052133278&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85052133278&partnerID=8YFLogxK
U2 - 10.1523/JNEUROSCI.0967-18.2018
DO - 10.1523/JNEUROSCI.0967-18.2018
M3 - Article
C2 - 30126970
AN - SCOPUS:85052133278
SN - 0270-6474
VL - 38
SP - 8441
EP - 8452
JO - Journal of Neuroscience
JF - Journal of Neuroscience
IS - 39
ER -