Absence Seizures (AS), a feature of many idiopathic generalised epilepsies, are characterised by a loss of consciousness and concomitant bilateral, synchronous 2.5 – 4Hz ‘spike and wave’ discharges (SWD) in the EEG. It is well established that AS expression relies on the integrity of both cortical and thalamic structures, given that 1) cortical initiation sites have been identified in both patients (frontal cortex) and animal models (perioral somatosensory cortex); 2) cortical and thalamic cells fire synchronously during AS and 3) spike and wave activity can be induced by electrical stimulation of the thalamus. However, the precise contributions of distinct neuronal populations in thalamic and cortical territories to the generation, maintenance and termination of AS remain unknown. This study investigated whether AS can be suppressed or induced by modulating the firing of discrete cell populations within the thalamocortical network, using a well-established genetic model of AS, the Genetic Absence Epilepsy Rats from Strasbourg (GAERS). Adult male GAERS (n=4, 250-350g) were fully anaesthetised (2.5% isoflurane inhalation), injected with 1µl pAAV-CaMKIIα-hChR2(H134R)-mCherry (1.14 x 109 GC/µl) into the ventrobasal thalamus and implanted with fiber optic-tetrode microdrives targeted to the same region, in accordance with UK legislation and local ethical guidelines. A minimum of 3 weeks post-injection, 473nm laser pulses were delivered to freely moving animals upon detection of SWD using a closed-loop system, or during seizure-free periods of active wakefulness (AW), quiet wakefulness (QW) and slow wave sleep (SWS). Driving excitation of ChR2+ thalamocortical neurons, using light stimulations as brief as 100ms, was able to block spontaneously occurring AS (mean 3-20Hz power ± SEM. Sham trials 356.4 ± 26.2 AU, versus laser trials 178.6 ± 27.4 AU, ANOVA p<0.001). Interestingly, single, brief stimulations (5-200ms) of the same population of thalamocortical neurons were also able to induce AS that were indistinguishable from natural AS in their EEG signature (~7Hz), multi-unit firing and concomitant behavioural arrest. Induction of AS was highly dependent on arousal state, whereby seizures could be induced during QW (54.2 ± 20.8%, mean induction rate for 5ms pulses ± SEM), but less so during AW (14.1 ± 1.6%, ANOVA p>0.05). During SWS, neither single nor patterned stimulation of thalamocortical neurons was able to induce AS (0 ± 0%, ANOVA p<0.05). The occurrence of both spontaneously occurring and light-induced AS during QW (5ms pulses: 79.3 ± 5.7%) was markedly reduced by administration of ethosuximide (100mg/kg i.p; 49.4 ± 6.1%, T-test p<0.01). This work provides further insight into the precise role of thalamocortical neurons in the generation, maintenance and termination of AS.