The state-dependency of transcranial magnetic stimulation (TMS) can be used to investigate the neural properties of subregions of the stimulated region. The objective of the present study was to determine whether state-dependency can reveal letter selectivity in the left posterior parietal cortex (PPC), a region known to contain letter-selective neurons. In two experiments, we used visual priming and adaptation to modulate the initial activation state of the left PPC prior to application of TMS. In the priming experiment, TMS was applied over the left PPC during the delay between the prime and the target stimulus on each experimental trial. Left PPC TMS reversed the effects of priming by facilitating the detection of non-primed letters, whereas detection of primed letters was unaffected. As neurons tuned to non-primed letters were less active at the time of TMS application than neurons tuned to the primed letters, this finding demonstrates that TMS preferentially facilitates the detection of attributes encoded by the less active neural populations. A similar facilitation of the less active neural populations was observed when adaptation was used to suppress letter-selective neurons prior to application of TMS. Our study demonstrates that TMS-priming and TMS-adaptation paradigms can reveal letter selectivity in the left PPC and thus be useful in the study of language processes. Our results also show that the state-dependent TMS effects obtained with visual priming are similar to those found with TMS adaptation: in both cases, attributes encoded by the less active neural populations are preferentially facilitated.