Beta range responses in thalamic CM/Pf complex indicate a role in speech cue signaling

Speech processing often requires cognitive control to focus on certain cues and initiate adequate responses to the speech input. The centromedian-parafascicular complex (CM/Pf) of the thalamus has been described as part of a cognitive control loop and may focus cortical and basal ganglia processing on task relevant information. It may also have an important function in the cognitive control of speech processing. Postoperative recordings of local field potentials (LFPs) in patients implanted with electrodes for deep brain stimulation (DBS) allow to obtain direct neurophysiological information with high spatial and temporal accuracy from the CM/Pf in humans. Here, we characterize the neural responses in the CM/Pf in speech processing and processing of action cues.
We used a multi-speaker paradigm with task relevant auditory and visual cues. In each trial, an auditory cue word was uttered in one of two concurrent speech streams that signaled which speaker would provide task relevant information later on in the speech stream. After speech offset a visual cue indicated that the participant should press buttons that corresponded to the task relevant information the speaker with the cue word had used. LFPs were obtained from two patients with chronic neuropathic pain (one woman, one man, aged 57 and 55 years, both right handed) implanted with quadripolar DBS electrodes in the right CM/Pf. Task performance was high (94% and 95% correct) indicating that participants paid attention to the cue and switched attention to the task-relevant speech stream. Average time-frequency responses showed transient increases in the beta range (20-30 Hz) after the cue word in the speech stream and in the theta band (4-8 Hz) after the visual cue indicating the beginning of the response interval. Single trial latencies of the beta-band responses correlated with the cue word onsets and latencies of the theta-band responses correlated with the visual cue onsets. Beta responses after the cue word were mostly greater than after the task relevant words.
Our results support the notion that the CM/Pf is part of a cognitive control loop involved in the initiation of goal relevant selection based on speech and visual information. Importantly, these signals are not only related to an externally triggered orienting response as they are elicited by both, the semantics of speech as well as the visual cue in the task context. Furthermore, they do not reflect the attentional orienting to a speech stream because the CM/Pf responses are transient after the cue.