Under what conditions can a true "science of mental life" arise from psychological investigations? Can psychology formulate scientific laws of a general nature, comparable in soundness to the laws of physics? I argue that the search for such laws must return to the forefront of psychological and developmental research, an enterprise that requires extensive collaboration between psychologists, neuroscientists, physicists, and mathematicians. Psychological laws may arise from at least 3 sources: the anchoring of thought processes in the biophysics of the brain, the computational constraints on possible mental algorithms, and the internalization of physical or statistical laws into our brains during evolution or development. I consider as an illustration the domain of numerical cognition, where a few solid psychophysical and decision-making laws have been established and related in part to their evolutionary precursors and neural bases. From this platform, I tentatively outline a few promising research directions in the domains of infant development, reading acquisition, executive control of multiple tasks, access to conscious report, and the spontaneous flow of conscious thoughts.