Current Biology 24, 1–7, September 22, 2014 ª2014 The Authors http://dx.doi.org/10.1016/j.cub.2014.08.016 Report Inducing Task-Relevant Responses to Speech in the Sleeping Brain Sid Kouider, 1, * Thomas Andrillon, 1,2 Leonardo S. Barbosa, 1,2 Louise Goupil, 1,2 and Tristan A. Bekinschtein 3,4 1 Laboratoire de Sciences Cognitives et Psycholinguistique, CNRS/EHESS/DEC-ENS, 29 Rue d’Ulm, 75005 Paris, France 2 Ecole Doctorale Cerveau-Cognition-Comportement, Universite ´ Pierre et Marie Curie, 9 Quai Saint Bernard, 75005 Paris, France 3 Cognition and Brain Sciences Unit, Medical Research Council, 15 Chaucer Road, Cambridge CB2 7EF, UK 4 Department of Psychology, University of Cambridge, Downing Street, Cambridge CB2 3EB, UK Summary Falling asleep leads to a loss of sensory awareness and to the inability to interact with the environment [1]. While this was traditionally thought as a consequence of the brain shutting down to external inputs, it is now acknowledged that incoming stimuli can still be processed, at least to some extent, during sleep [2]. For instance, sleeping partic- ipants can create novel sensory associations between tones and odors [3] or reactivate existing semantic associations, as evidenced by event-related potentials [4–7]. Yet, the extent to which the brain continues to process external stim- uli remains largely unknown. In particular, it remains unclear whether sensory information can be processed in a flexible and task-dependent manner by the sleeping brain, all the way up to the preparation of relevant actions. Here, using se- mantic categorization and lexical decision tasks, we studied task-relevant responses triggered by spoken stimuli in the sleeping brain. Awake participants classified words as either animals or objects (experiment 1) or as either words or pseu- dowords (experiment 2) by pressing a button with their right or left hand, while transitioning toward sleep. The lateralized readiness potential (LRP), an electrophysiological index of response preparation, revealed that task-specific prepara- tory responses are preserved during sleep. These findings demonstrate that despite the absence of awareness and behavioral responsiveness, sleepers can still extract task- relevant information from external stimuli and covertly pre- pare for appropriate motor responses. Results We studied whether the categorization of spoken words can still trigger task-relevant motor plans during early sleep stages. One main difficulty in addressing this issue consists in instructing a new task to sleeping subjects, arguably because prefrontal regions dealing with executive functions are then particularly suppressed in comparison to other cortical regions [8, 9]. One potential solution is to rely on the induction approach commonly used by studies on implicit perception in awake participants. This research reveals that the processing stream involved in making a semantic clas- sification can, through explicit practice, be automatized and bypass prefrontal regions. Under those conditions, the cate- gorization of visual words and numbers can lead to the covert activation of motor cortex even when those stimuli are masked and presented below the threshold of consciousness [10, 11]. In the current study, we extend this task induction strategy to track the ability of sleepers in extracting task-relevant informa- tion from speech and preparing for the appropriate motor plan. LRPs Reveal Semantic Classification and Response Preparation before and after Falling Asleep We recorded the electroencephalogram (EEG) of human participants while they were awake and instructed them to classify spoken words as animals or objects (Figure 1). This procedure allowed us to compute lateralized readiness poten- tials (LRPs)—a neural marker of response selection and prep- aration [12]—by mapping each specific semantic category to a specific motor plan (e.g., animals with the right hand and ob- jects with the left hand, counterbalanced across participants). This design allows for the assessment of lateralized response preparation toward the side associated with the appropriate semantic category. Thus, it allows for testing of whether sen- sory signals are processed beyond semantic levels by probing how the meaning extracted from external words can lead to the covert selection and preparation of context-dependent actions. Testing conditions encouraged the transition toward sleep while remaining engaged with the same task set: sub- jects received explicit allowance to fall asleep and were sitting in a dark room, eyes closed, in a reclining chair, listening to several repetitions of the same list of stimuli with a long inter- trial interval of 6–9 s. Crucially, participants received an entirely new list of words (n = 48) during sleep to ensure that their responses were based on the extraction of word meaning rather than a mere reactivation of stimulus-response associa- tions established during the wake stage. Sleep onset was assessed online both behaviorally, by ensuring the absence of overt responses for at least 2 min of stimulation, and electrophysiologically, through sleep markers (i.e., disappearance of low-amplitude alpha/beta rhythms and development of high-amplitude delta/theta rhythms [see Fig- ure S1 available online], presence of slow eye movements and other sleep graphoelements such as vertex sharp waves, and regular spontaneous and evoked K complexes or sleep spin- dles) before and after the presentation of each word. Partici- pants underwent the transition from full wakefulness to light sleep and then oscillated primarily between the non-rapid eye movement 1 (NREM1) and NREM2 stages. Note that trials were only considered as NREM1 when there was a complete lack of alpha rhythm accompanied by sleep markers. In order to discard epochs comprising brief awakenings and microar- ousals (i.e., reappearance of a wake-like EEG activity for less than 3 s; percentage of trials: mean = 11.6, SD = 8) and to ensure that each trial included in the sleep conditions genuinely re- flected a state of sleep, we performed an offline and conserva- tive evaluation of sleep stages relying on strict criteria. Scoring was performed here by two trained neurophysiologists blind to experimental conditions who additionally verified that partici- pants remained asleep after stimuli onset by tracking any *Correspondence: sid.kouider@ens.fr This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/3.0/). Please cite this article in press as: Kouider et al., Inducing Task-Relevant Responses to Speech in the Sleeping Brain, Current Biology (2014), http://dx.doi.org/10.1016/j.cub.2014.08.016