Many philosophers claim that the subjective character of conscious experience creates a fundamental deficit in third-person (extrospective) access to first-person experience. This paper argues that extrospective measurement methods suffer from no such deficit when compared with measurement techniques in the natural sciences. After rejecting principled objections, a historical comparison shows extrospective methods are still in an early stage of development but can be improved via a bootstrapping strategy similar to that used in physical measurement. Multiple sources of evidence should allow substantial advances. New developments in pain measurement demonstrate this bootstrapping strategy is already bearing fruit.
The stream of conscious experience, even when it changes rapidly and spontaneously, can be predicted from brain activity alone. Using binocular rivalry to induce frequent, involuntary shifts in perception without altering what the eyes see, researchers measured brain activity with fMRI. Multivariate pattern analysis of signals from visual cortex accurately tracked participants' reported conscious experience as it fluctuated. Prediction in primary visual cortex relied on eye-based signals, while higher visual areas reflected the color of the percept. Accurate prediction was also possible using signals recorded during stable viewing, showing the method generalizes across conditions and does not depend on motor responses.