The flash-lag effect and motion perception timing: a framework grounded in a proposed biophysical mechanism for perception as a first-person property.

The flash-lag effect (FLE), a dissociation between physical events and their perceived time and location, has an unresolved neural basis. Here, the FLE is re-examined within the semblance hypothesis, in which perception, a first-person property, arises from the rapid formation and reversal of inter-postsynaptic functional LINKs (IPLs). A novel, unpredictable stimulus such as a flash is hypothesized to require de novo IPL formation, with perceptual latency Tpercept(flash) ≈ 70-100 ms. Predictable, continuously tracked motion is instead hypothesized to reactivate IPLs formed at motion onset, with updates within Tupdate ≈ 41-46 ms. This latency difference is proposed to account for the baseline FLE, FLE ≈ v · (Tpercept(flash) - Tupdate), a temporal offset of 24-59 ms. IPL reactivation is assumed to support both ongoing updates and the percept's forward extrapolation, with the projection horizon τproj set equal to Tupdate ≈ 41-46 ms as a modeling assumption. The tight constraints of a direction-reversal experiment may be explained, in an interconnected manner, by τproj and IPL dissociation and their effects on successive stages of percept formation. The percept is proposed to be projected to its computed location without an additional processing delay, with neural latency compensated by a time-to-space transformation (Δt→Δx), which is a modeling postulate. These proposals are used to formulate testable predictions.