The “stream of consciousness” we experience is an internal and purely subjective phenomenon. For this reason, we can only infer that others have conscious experience based on their overt behaviour. Particularly, the ability to process stimuli from the environment (connectedness) and to respond in a meaningful and goal-directed way (responsiveness) are considered indicative of covert awareness. Yet, a growing body of evidence suggests that subjective awareness, connectedness, and responsiveness can, in principle, be dissociated. Firstly, although we are unresponsive and disconnected form the external environment during REM sleep, we often experience vivid dreams, which are phenomenally similar to waking consciousness. Secondly, rich hallucinations are also reported after recovery from ketamine-induced unresponsiveness. Thirdly, apparently a significant group of patients considered to be in a vegetative state due to persistent behavioural unresponsiveness, might actually remain conscious. In my presentation I will discuss the dissociation between inner awareness, connectedness, and responsiveness in the context of neuroimaging studies seeking neuronal correlates of consciousness. Further, I will present a study conducted to investigate brain spontaneous functional connectivity as a mechanism of subjective inner awareness. We used two independent data sets and analysed brain activity in a range of states, including responsive and conscious (wakeful rest), unresponsive but conscious (ketamine sedation, REM sleep), and unresponsive and unconscious (propofol sedation, NREM sleep). First, electrocorticography (ECoG) data from an anesthesia study performed on macaque monkeys, where various anaesthetics were used across experimental sessions (ketamine, medetomidine, and propofol). Second, local filed potentials acquired from 10 epilepsy patients implanted with depths electrodes, from whom brain activity was recorded during wakefulness, REM, and N-REM sleep. We observed hyper-correlated brain activity (i.e. more correlations and less anti-correlations) during propofol-induced loss of consciousness, but not during wakefulness and ketamine sedation. In the sleep experiment hyper-correlated brain activity was revealed during slow-wave sleep, but not during REM sleep and wakefulness. Thus, the subjective “stream of consciousness” is maintained when brain activations are relatively independent from each other, but when brain signals become hyper-correlated the inner awareness is lost.