The neural correlates of human vibrotactile working memory: Converging evidence from functional magnetic resonance imaging, electroencephalographic and behavioral studies

The present dissertation aimed to shed more light on the psychological mechanisms and the neural basis of tactile working memory (WM). For this purpose, a vibrotactile delayed discrimination task was studied using the methods of functional magnetic resonance imaging (fMRI), electroencephalography (EEG), concurrent subliminal electrical stimulation and psychophysics. The fMRI study (Study I) showed that a broad network of brain regions much broader than known from previous studies in non-human primates supports the performance of a vibrotactile delayed discrimination task in the different task periods: encoding, maintenance, decision making. The analysis of oscillatory activity over the somatosensory cortex in the EEG study (Study II) and the experiment using subliminal electrical stimulation to locally inhibit the primary somatosensory cortex (S1) (Study III) suggest that S1 does not contribute to the active maintenance of the vibrotactile memory trace. The level of activity in S1 during the early delay period depends on the efficiency with which subjects encode the vibrotactile stimulus. Study II and III also showed that the activation level of S1 in the pre-trial period plays an important role. Study III suggests that, in this task period, S1 activity is up-regulated under WM demands probably reflecting the operation of top-down attentional control. Study III indicates that increasing local inhibition of S1 in the pre-trial period improves performance by facilitating bottom-up processing. Importantly, long-term memory representations of the average frequency of the stimulus set strongly influence performance giving rise to the time-order effect (Study IV). Additionally, the fMRI data (Study V) showed that a somatosensory network integrates information about the current vibrotactile stimulus and the representation of the average vibration frequency during stimulus encoding and maintenance.