oscillations and connectivity

Cortical activity can be decomposed into individual frequency components; due to the ability to reveal also oscillatory transients, wavelet frequency decomposition is a popular technique. Two principal responses are differentiated: evoked responses that are phase-locked to the stimulus and are well represented in the averaged LFP waveforms, and induced responses that jitter in timing and phase depending on other cortical inputs. These are lost in the standard averaging techniques. Induced responses are interesting, since they reflect the influence of other cortical inputs (corticocortical inputs) on sensory responses. Top-down effects are also involved in the generation of induced responses.

We used this approach to investigate the induced responses in animals with no hearing experience. We compared time-frequency representations in both evoked and induced domain in hearing, acoustically stimulated cats, hearing, electrically stimulated cats, and congenitally deaf cats, all in adult age. There was a near complete loss of induced responses in primary and secondary auditory areas in congenitally deaf cats (Yusuf et al., 2017). It involved α, β and γ frequency range. Since induced responses are considered the substrate of corticocortical coupling, this further supports the concept of reduced corticicortical interactions in congenital deafness.  The area involved in cross-modal reorganization (PAF) showed reduced ongoing α activity. Increase in α activity correlates with suppression in cortical areas when processing is directed away from these, e.g. to another modality. The results supports the role of α activity in such functional suppression in heteromodal stimulation. To allow the involvement of the area in heteromodal processing, cross-modal reorganization must lead to reduced ongoing α activity.

Our lab aims to find functional measures of higher order cortical processing beyond auditory features. The reported outcome indicates that in congenital deafness, a sensory stimulus is not integrated with ongoing activity. Ongoing activity carries information on what the cortex has been processing before and during the sensory stimulus. Integration of ongoing and sensory-evoked activity is key for integration of the sensory input into the present internal model of the environment and the self in an awake human subject. It is also a fingerprint of functional connectivity (Yusuf et al., 2021, see also >>>).

Furthermore, connectivity analysis between primary and secondary auditory areas supported predicive coding: in acoustically-stimulated hearing cats strong top-down signaling and weak bottom-up signaling was observed, in electrically-stimulated hearing cats the top-down signaling was weak and the bottom-up signaling much stronger. This is consistent with prediction being represented by top-down interactions, and prediction error by bottom-up interactions. In congenitally deaf animals, the top-down interaction disappeared, supporting our working hypothesis that experience is required for top-down interactions (Yusuf et al., 2021).

Cortical oscillations

Analysis of oscillatory activity in the primary auditory cortex (A1) and a secondary area (PAF) generated in response to cochlear stimulation, showing total, evoked and induced activity. Phase-locking factor (PLF) is additionally shown to support the computation of evoked responses. From Yusuf et al., 2017.

Time representation of oscillatory responses of one example recording site, individual trials (stimulus repetitions) shown in different colors, „stim.“ indicate the time of stimulus presentation. The evoked response is phase-locked and well reproducible across trials, whereas the induced response jitters in time and phase.

Functional connectivity between the primary auditory cortex (field A1) and secondary auditory field (field PAF). Time relative to onset of a cochlear implant stimulus. In an „early window“ within 50 ms post stimulus, connectivity in gamma range is observed, whereas in the „late window“ (200 ms after the stimulus) synchronization in the alpha range appeared. In deaf animals there was loss of the late connectivity. Granger causality analysis demonstrated that this connectivity is in majority due to top-down information flow.

The method used (pairwise phase consistency) is not dependent on signal power, and thus this result is not a mere reflection of the reduced induced activity in the same window. Rather, the loss of induced activity is a fingerprint of the lost top-down interactions in congenital deafness. For connectivity analysis, see Yusuf et al. (2021).

A1 infra - PAF
A1 supra  - PAF