It's right there in the abstract: > We ran a subjective flicker marking experiment to fit the model parameters The article goes on to describe how they perfor

It's right there in the abstract: > We ran a subjective flicker marking experiment to fit the model parameters The article goes on to describe how they performed cross-validation on the experimental data to validate how accurate the predictions are. Or am I misunderstanding your complaint?