Imagined speech can be decoded from low- and cross-frequency intracranial EEG features

Reconstructing intended speech from neural activity using brain-computer interfaces holds great promises for people with severe speech production deficits. While decoding overt speech has progressed, decoding imagined speech has met limited success, mainly because the associated neural signals are weak and variable compared to overt speech, hence difficult to decode by learning algorithms. We obtained three electrocorticography datasets from 13 patients, with electrodes implanted for epilepsy evaluation, who performed overt and imagined speech production tasks. Based on recent theories of speech neural processing, we extracted consistent and specific neural features usable for future brain computer interfaces, and assessed their performance to discriminate speech items in articulatory, phonetic, and vocalic representation spaces. While high-frequency activity provided the best signal for overt speech, both low- and higher-frequency power and local cross-frequency contributed to imagined speech decoding, in particular in phonetic and vocalic, i.e. perceptual, spaces. These findings show that low-frequency power and cross-frequency dynamics contain key information for imagined speech decoding.

Relationship between high-frequency activity in the sensory and the motor hand areas, and regional myelin content

BackgroundThe human primary sensory (S1) and primary motor (M1) hand areas feature high-frequency neuronal responses. Electrical nerve stimulation evokes high-frequency oscillations (HFO) at around 650 Hz in the contralateral S1. Likewise, paired-pulse transcranial magnetic stimulation of M1 produces short interval intracortical facilitation (SICF) of motor evoked potentials in contralateral hand muscles. SICF features several peaks of facilitation which are separated by inter-peak intervals resembling HFO rhythmicity.HypothesisIn this study, we tested the hypothesis that the individual expressions of HFO and SICF are tightly related to each other and to the regional myelin content in the sensorimotor cortex.MethodsIn 24 healthy volunteers, we recorded HFO and SICF, and, in a subgroup of 20 participants, we mapped the cortical myelin content using the ratio between the T1- and T2-weighted MRI signal as read-out.ResultsThe individual frequencies and magnitudes of HFO and SICF were tightly correlated: the intervals between the first and second peak of cortical HFO and SICF showed a positive linear relationship (r= 0.703, p< 0.001), while their amplitudes were inversely related (r= −0.613, p= 0.001). The rhythmicity, but not the magnitude of the high-frequency responses, was related to the cortical myelin content: the higher the cortical myelin content, the shorter the inter-peak intervals of HFO and SICF.ConclusionThe results confirm a tight functional relationship between high-frequency responses in S1 (i.e., HFO) and M1 (i.e., SICF). They also establish a link between the degree of regional cortical myelination and the expression of high-frequency responses in the human cortex, giving further the opportunity to infer their possible generators.

Neuronal activity in the human amygdala and hippocampus enhances emotional memory encoding

Emotional events are often easier to recall, and comprise our most valuable memories. Here, as subjects performed a memory task in which they recalled emotional stimuli more readily than neutral stimuli, we used direct brain recording and stimulation in the hippocampus and amygdala to identify how the brain prioritizes emotional information for memory encoding. High-frequency activity (HFA), a correlate of local neuronal spiking, increased in both hippocampus and amygdala when subjects successfully encoded emotionally arousing stimuli. Direct electrical stimulation applied to these regions during encoding decreased HFA and selectively impaired retrieval for emotional stimuli. Finally, depressed subjects' memory was biased more by valence than arousal, and they exhibited a congruent increase in HFA as a function of valence. Our findings thus provide evidence that emotional stimuli up-regulate activity in the amygdala--hippocampus circuit to enhance memory for emotional information, and suggest that targeted modulation of this circuit alters emotional memory processes.

Ripples reflect a spectrum of synchronous spiking activity in human anterior temporal lobe

Direct brain recordings have provided important insights into how high frequency activity captured through intracranial EEG (iEEG) supports human memory retrieval. The extent to which such activity is comprised of transient fluctuations that reflect the dynamic coordination of underlying neurons, however, remains unclear. Here, we simultaneously record iEEG, local field potential (LFP), and single unit activity in the human temporal cortex. We demonstrate that fast oscillations within the previously identified 80-120 Hz ripple band contribute to 70-200 Hz high frequency activity in the human cortex. These ripple oscillations exhibit a spectrum of amplitudes and durations related to the amount of underlying neuronal spiking. Ripples in the macro-scale iEEG are related to the number and synchrony of ripples in the micro-scale LFP, which in turn are related to the synchrony of neuronal spiking. Our data suggest that neural activity in the human temporal lobe is organized into transient bouts of ripple oscillations that reflect underlying bursts of spiking activity.

Long Time Constant May Endorses Sharp Waves and Spikes Than Sharp Transients in Scalp Electroencephalography: A Comparison of Both After-Slow Among Different Time Constant and High-Frequency Activity Analysis

Objective: To clarify whether long time constant (TC) is useful for detecting the after-slow activity of epileptiform discharges (EDs): sharp waves and spikes and for differentiating EDs from sharp transients (Sts).Methods: We employed 68 after-slow activities preceded by 32 EDs (26 sharp waves and six spikes) and 36 Sts from 52 patients with partial and generalized epilepsy (22 men, 30 women; mean age 39.08 ± 13.13 years) defined by visual inspection. High-frequency activity (HFA) associated with the apical component of EDs and Sts was also investigated to endorse two groups. After separating nine Sts that were labeled by visual inspection but did not fulfill the amplitude criteria for after-slow of Sts, 59 activities (32 EDs and 27 Sts) were analyzed about the total area of after-slow under three TCs (long: 2 s; conventional: 0.3 s; and short: 0.1 s).Results: Compared to Sts, HFA was found significantly more with the apical component of EDs (p &lt; 0.05). The total area of after-slow in all 32 EDs under TC 2 s was significantly larger than those under TC 0.3 s and 0.1 s (p &lt; 0.001). Conversely, no significant differences were observed in the same parameter of 27 Sts among the three different TCs. Regarding separated nine Sts, the total area of after-slow showed a similar tendency to that of 27 Sts under three different TCs.Significance: These results suggest that long TC could be useful for selectively endorsing after-slow of EDs and differentiating EDs from Sts. These findings are concordant with the results of the HFA analysis. Visual inspection is also equally good as the total area of after-slow analysis.

Direct brain recordings reveal continuous encoding of structure in random stimuli

The brain excels at processing sensory input, even in rich or chaotic environments. Mounting evidence attributes this to the creation of sophisticated internal models of the environment that draw on statistical structures in the unfolding sensory input. Understanding how and where this modeling takes place is a core question in statistical learning. It is unknown how this modeling applies to random sensory signals. Here, we identify conditional relations, through transitional probabilities, as an implicit structure supporting the encoding of a random auditory stream. We evaluate this representation using intracranial electroencephalography recordings by applying information-theoretical principles to high-frequency activity (75-145 Hz). We demonstrate how the brain continuously encodes conditional relations between random stimuli in a network outside of the auditory system following a hierarchical organization including temporal, frontal and hippocampal regions. Our results highlight that hierarchically organized brain areas continuously attempt to order incoming information by maintaining a probabilistic representation of the sensory input, even under random stimuli presentation.

Moment-by-moment tracking of naturalistic learning and its underlying hippocampo-cortical interactions

Humans form lasting memories of stimuli that were only encountered once. This naturally occurs when listening to a story, however it remains unclear how and when memories are stored and retrieved during story-listening. Here, we first confirm in behavioral experiments that participants can learn about the structure of a story after a single exposure and are able to recall upcoming words when the story is presented again. We then track mnemonic information in high frequency activity (70–200 Hz) as patients undergoing electrocorticographic recordings listen twice to the same story. We demonstrate predictive recall of upcoming information through neural responses in auditory processing regions. This neural measure correlates with behavioral measures of event segmentation and learning. Event boundaries are linked to information flow from cortex to hippocampus. When listening for a second time, information flow from hippocampus to cortex precedes moments of predictive recall. These results provide insight on a fine-grained temporal scale into how episodic memory encoding and retrieval work under naturalistic conditions.

Medial temporal lobe high-frequency activity at feedback signals error and predicts subsequent memory during spatial learning

The ability to incorporate information about feedback is critical for associative learning. The medial temporal lobe (MTL) and prefrontal cortex (PFC) are thought to be involved in processing feedback as new associations are learned. However, the relative contributions of these regions to feedback processing and subsequent memory performance in humans are poorly understood. To address this question, we tested pre-surgical epilepsy patients with depth electrodes implanted in the MTL and PFC using a spatial memory task in which subjects learned object-location associations over time. We found increased high-frequency activity (HFA; 40-100 Hz), thought to reflect local excitatory activity, in the MTL and dorsolateral PFC (dlPFC) at feedback for high error trials. In the MTL, this HFA error signal predicted greater trial-by-trial decreases in error from one training block to the next indicating that these signals are involved in updating memory representations or modifying incorrect associations during learning. The opposite pattern of activity was observed during retrieval, with greater MTL and dlPFC HFA predicting lower error, replicating previous results from our group. Overall, these data suggest putative mechanisms for the learning of object-location associations.